Analysis of material efficiency aspects of personal computers

Tecchio P., Ardente F., Marwede M.,

Clemm C., Dimitrova G. Mathieux F.

Analysis of material efficiency aspects of

personal computers product group

January 2018

EUR 28394 EN

This publication is a technical report by the Joint Research Centre (JRC), the European Commission’s science and

knowledge service. It aims to provide evidence-based scientific support to the European policy-making process.

The scientific output expressed does not imply a policy position of the European Commission. Neither the European

Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of

this publication.

Contact information

Name: Fabrice Mathieux

Address: Joint Research Centre, Via E. Fermi 2749, 21027 Ispra, ITALY

Email: fabrice.[email protected]uropa.eu

Tel. +39 332789238

JRC Science Hub

https://ec.europa.eu/jrc

EUR 28394 EN

PDF

ISBN 978-92-79-64943-1

ISSN 1831-9424

doi:10.2788/89220

ISBN 978-92-79-64944-8

ISSN 1018-5593

doi:10.2788/679788

Luxembourg: Publications Office of the European Union, 2018

Reuse is authorised provided the source is acknowledged. The reuse policy of European Commission documents is

regulated by Decision 2011/833/EU (OJ L 330, 14.12.2011, p. 39).

For any use or reproduction of photos or other material that is not under the EU copyright, permission must be

sought directly from the copyright holders.

How to cite: Tecchio, P., Ardente, F., Marwede, M., Christian, C., Dimitrova, G. and Mathieux, F., Analysis of

material efficiency aspects of personal computers product group, EUR 28394 EN, Publications Office of the European

Union, Luxembourg, 2018, ISBN 978-92-79-64943-1, doi:10.2788/89220, JRC105156.

All images © European Union 2018, except:

Front page, © BillionPhotos.com, image #102613073 Source: [Fotolia.com]

 

 
Contents 
List of figures ........................................................................................................ 5 
List of tables ......................................................................................................... 7 
Acknowledgements ................................................................................................ 9 
Executive summary ............................................................................................. 10 
Abbreviations ..................................................................................................... 14 
List of definitions ................................................................................................. 16 
Introduction .................................................................................................. 17 
Background information ................................................................................. 18 
1  Market data ............................................................................................. 18 
2  Expected lifetime ..................................................................................... 19 
2.1  External power-supply lifetime ........................................................... 22 
2.2  Battery lifetime ................................................................................. 22 
3  Bill of materials ........................................................................................ 25 
3.1  Desktop computers ........................................................................... 25 
3.2  Notebook computers ......................................................................... 26 
3.3  Tablets ............................................................................................ 30 
3.4  External power supplies ..................................................................... 31 
3.5  Batteries .......................................................................................... 32 
4  Content of precious and critical raw materials .............................................. 35 
4.1  Content of cobalt in batteries ............................................................. 37 
4.2  Content of rare earths in HDDs ........................................................... 37 
5  Interoperable external power supplies ........................................................ 38 
5.1  Common external power supplies for mobile phones ............................. 38 
5.2  Common EPS for personal computers .................................................. 39 
5.3  USB cables and connectors ................................................................ 41 
6  Environmental impacts ............................................................................. 42 
6.1  Standards for environmental assessment of ICT products ...................... 44 
Analysis of end-of-life practices for the product group ........................................ 46 
1  Analysis of recycling/recovery practices ...................................................... 46 
1.1  Recycling/recovery of desktop computers (without integrated display) .... 46 
1.2  Recycling/recovery of integrated desktop computers ............................. 51 
1.3  Recycling/recovery of notebooks ......................................................... 54 
1.4  Recycling/recovery of tablets .............................................................. 56 
1.5  Focus on recycling/recovery of electronic PCBs ..................................... 57 
1.6  Focus on recycling/recovery of batteries contained in the product group .. 58 
1.7  Future recycling scenario for notebooks ............................................... 63 
2  Analysis of repair/reuse practices ............................................................... 64 

 

 
2.1  Reuse and repair of notebooks ........................................................... 64 
2.2  Reuse and repair of tablets ................................................................ 72 
Discussion and identification of hot spots and of improvement opportunities ......... 76 
1  EU Ecolabel and EU green public procurement criteria ................................... 78 
Possible actions to improve waste prevention .................................................... 79 
1  Battery durability ..................................................................................... 79 
1.1  Rationale ......................................................................................... 79 
1.2  Possible improvements ...................................................................... 81 
1.3  Initial assessments of benefits/impacts (battery lifetime optimisation) ..... 83 
1.4  Other potential benefits (information about battery-cycle life) ................ 85 
2  Decoupling external power supplies from personal computers ........................ 87 
2.1  Rationale ......................................................................................... 87 
2.2  Possible improvements ...................................................................... 88 
2.3  Initial assessments of benefits/impacts ................................................ 88 
3  Durability testing for personal computers .................................................... 95 
3.1  Rationale ......................................................................................... 95 
3.2  Possible improvements ...................................................................... 99 
3.3  Future  improvements:  development  of  additional  standards  on  endurance 
testing  99 
Possible actions to enhance repair/reuse ........................................................ 101 
1  Disassemblability  of  key  components  for  personal  computers  to  enhance 
repairability ................................................................................................... 101 
1.1  Rationale ....................................................................................... 101 
1.2  Possible improvements .................................................................... 101 
1.3  Initial assessments of benefits/impacts .............................................. 106 
1.4  Potential benefits for other product categories .................................... 109 
2  Secure data deletion for personal computers ............................................. 110 
2.1  Rationale ....................................................................................... 110 
2.2  Possible improvements .................................................................... 110 
2.3  Potential benefits for the product group ............................................. 111 
Possible actions to enhance recyclability ......................................................... 113 
1  Dismantlability of key components for personal computers .......................... 113 
1.1  Rationale ....................................................................................... 113 
1.2  Possible improvements .................................................................... 114 
2  Marking of plastic components ................................................................. 116 
2.1  Rationale ....................................................................................... 116 
2.2  Possible improvements .................................................................... 117 
3  Declaration of flame-retardant content ..................................................... 118 
3.1  Rationale ....................................................................................... 118 

 

 
3.2  Possible improvements .................................................................... 119 
4  Identifiability of batteries ........................................................................ 120 
4.1  Rationale ....................................................................................... 120 
4.2  Possible improvements .................................................................... 121 
5  Provision of information on the content of critical raw materials ................... 122 
5.1  Rationale ....................................................................................... 122 
5.1.1  Recovery of cobalt .................................................................... 122 
5.1.2  Recovery of rare earth elements ................................................. 122 
5.2  Possible improvements .................................................................... 123 
6  Initial assessments of benefits/impacts ..................................................... 125 
6.1  Potential benefits for other product categories .................................... 130 
Conclusions ................................................................................................ 131 
References ....................................................................................................... 135 
   

 
5 
 
List of figures 
Figure  1  —  Shipments  of  hard-  and  solid-state  disk  drives  (HDD/SSD)  (in  millions) 
worldwide in computers from 2012 to 2017 (Statista, 2016) and projections for 2018-2020 
(own elaboration). ............................................................................................... 19 
Figure 2  — Average first-use time of notebooks in Germany (n=2 268 in 2012, lowest 
value n=244 in 2004; 2008, 2009: not specified) (Prakash et al., 2016b). ................. 21 
Figure  3  —  Calendar  ageing  of  nickel-manganese-cobalt-oxide  (NMC)  cells  over  time 
depending  on  SoC  at  an  elevated  temperature  (50 °C).  Capacity  fade  (ratio  between 
current and initial capacity) under varying SoC over time (in days) is shown in diagram 
(a) and the corresponding increase in internal resistance (ratio between current and initial 
resistance) in the cells in (b) (Schmalstieg et al., 2014) ........................................... 24 
Figure 4 — Types of batteries used and expected to be used by notebooks and tablets 
(Berger, 2012) .................................................................................................... 32 
Figure  5  —  Volumes  of  the  different  lithium-ion  battery  subchemistries  in  notebooks 
(Laptop NiMH, Laptop LiNMC and Laptop LiCoO2) and tablets (Tablet LiNMC) put on market 
(top) and the estimated volumes of such waste batteries generated (bottom) in the EU 
(Chancerel et al., 2016). ...................................................................................... 33 
Figure 6 — Cobalt potential in kg and distribution over the products sold in Germany in 
2007 (blue line) and 2012 (red line) (Chancerel et al., 2015) ................................... 36 
Figure 7 — CRMs and other relevant materials in notebooks (Chancerel et al., 2015) .. 36 
Figure 8 — Detail of the content of cobalt for different batteries (Sommer et al., 2015) 37 
Figure 9 — Graphical representation of an EPS with micro-B connector, detachable cable 
with USB type-A connector and USB type-A receiver (image credits: ©Pugetbill 2011) 39 
Figure 10 — USB type-C cable and connectors (image credits: ©USB Implementers Forum 
2014) ................................................................................................................ 42 
Figure 11 — Disassembly sequence of a mini desktop computer (HP, 2016) ............... 48 
Figure 12 — Detail of the interior of a mini desktop computer containing a battery (). . 49 
Figure 13 — Desktop computers EoL scenarios. Scenario 1 (Manual dismantling, shredding 
and mechanical sorting); Scenario 2 (Shredding and mechanical sorting) ................... 51 
Figure 14 — Example of an integrated desktop computer. ........................................ 52 
Figure 15 — Disassembly sequence for an integrated desktop computer () ................. 53 
Figure 16 — Mechanical crushing, separation and sorting (Scenario 1). Cu = copper. Al = 
Aluminum. ......................................................................................................... 55 
Figure 17 — Manual medium-depth recycling scenario for notebooks (Scenario 2). Cu = 
copper. Al = Aluminum. ....................................................................................... 55 
Figure 18 — Magnesium frames in tablets (Schischke et al., 2014) ............................ 57 
Figure 19 — Current legislative battery marks (source: BAJ) .................................... 59 
Figure 20 — Examples of battery marks in current practice (unreleased data from Slates 
D4R) ................................................................................................................. 59 
Figure 21 — Battery-recycle mark, developed by the BAJ and promoted to be used as an 
international standard, which indicates the four different battery types by colour and in 
text. .................................................................................................................. 60 
Figure 22 — The two-digit code, developed and recommended for use by BAJ, which is 
added  to  the  logo for  LIB  to  identify:  the  metal  with  the  highest mass  in  the  positive 
electrode (first digit); and the presence of a metal which hinders recycling (second digit).
......................................................................................................................... 60 

 
6 
 
Figure 23 — Application of the battery-recycle mark on a notebook battery (left) and a 
tablet battery (right) (sources: newbatteryshop.com and ifixit.com). ......................... 61 
Figure  24  —  Battery  markings  developed  by  IEC  as  published  in  the  draft  standard 
circulated in March 2017. ..................................................................................... 61 
Figure 25 — Most common components in notebooks that suffered damage or breakage 
(IDC, 2016). ....................................................................................................... 65 
Figure 26 — Types of accidents causing notebook, tablet and handheld device damage, 
according to the IDC (2016). ................................................................................ 65 
Figure 27 — Types of accidents causing notebook damage, according to the IDC (2010).
......................................................................................................................... 66 
Figure 28 — Reasons for replacing a notebook after first use (Prakash et al., 2016b) ... 70 
Figure 29 — Most common components in tablets that suffered damage or breakage (IDC, 
2016). ............................................................................................................... 72 
Figure 30 — Failed attempted to separate the front glass from the display panel (Schischke 
et al., 2014) ....................................................................................................... 74 
Figure 31 — Potential material saving (t/year) in 2020, 2025 and 2030, divided by product 
categories: notebooks and tablets. ........................................................................ 91 
Figure 32 — Picture of a drop test applied to a notebook (Westpak, 2013). ................ 96 
Figure 33 — Plastic marking according to ISO11469 (adapted from Bombardier (2010)).
....................................................................................................................... 118 
Figure 34 — References scenarios for the calculation of the benefits related to material-
efficiency actions to enhance the recyclability of notebooks. ................................... 127 
   

 
7 
 
List of tables 
Table 1 — Estimated annual sales (2012-2030) for product categories in the EU market 
(Viegand Maagøe and VITO, 2017). Values in millions of units. ................................. 18 
Table 2 — Typical lifetime of computers and related products according to Viegand Maagøe 
and VITO (2017). ................................................................................................ 20 
Table 3 — Use time and desired lifetime of products (Wieser and Tröger, 2016). n= number 
of respondents (population interviewed: 1 009 Austrian residents). ........................... 21 
Table 4 — Full-charge capacity projections after 1 year of use (HP Inc., 2016)............ 24 
Table  5  —  Desktop-computer  bill  of  materials (BoM)  according  to  Song  et  al.  (2013). 
Packaging included. ............................................................................................. 25 
Table 6 — Desktop-computer BoM according to Teehan and Kandlikar (2013). Packaging 
excluded. ........................................................................................................... 26 
Table 7  — BoM for notebooks, modified from  Talens Peiró et al. (2016b) with mass of 
battery as in Clemm et al. (2016) ......................................................................... 27 
Table 8 — BoM of HDDs and ODDs (Talens Peiró and Ardente, 2015)......................... 28 
Table 9 — Composition of SSDs (Seagate, 2016) .................................................... 28 
Table 10 — Summary BoMs considering Table 7 and Table 8 .................................... 29 
Table 11 — Average composition of PCBs in notebooks (Chancerel and Marwede, 2016)
......................................................................................................................... 29 
Table 12 — BoMs of 20 tablets, tablets with aluminium housing, and tablets with plastic 
housing (all averages) (Schischke et al., 2014) ....................................................... 30 
Table 13 — BoMs for EPS. Different sources. .......................................................... 31 
Table 14 — Mass of components of the notebook battery (Clemm et al., 2016). .......... 34 
Table 15 — BoM of an LCO notebook-battery cell from one of the largest cell manufacturers 
worldwide (Clemm et al., 2016). ........................................................................... 34 
Table 16 — Consumption of energy and auxiliaries during the production of one cell as 
detailed in Table 15 (Clemm et al., 2016). ............................................................. 35 
Table 17 — Average composition of LCO, NMC and NCA batteries for notebooks ......... 35 
Table 18 — Recommended categories of EPS (ITU-T L.1002, 2016) ........................... 41 
Table 19 — Relevant sources in the scientific literature ............................................ 44 
Table 20 — Recycling rate of materials in PCB of notebook properly treated (Chancerel and 
Marwede, 2016) .................................................................................................. 58 
Table 21 — Li-ion battery-recycling plants processes ............................................... 63 
Table 22 — Summary of ‘hot spots’ of computer sub-product groups (as identified during 
the analysis of Sections 2 and 3) ........................................................................... 77 
Table 23 — Material savings (batteries in million units/year, materials in t/year) achievable 
when a battery-optimisation software is implemented in notebooks. .......................... 85 
Table 24 — Material savings [t/year] divided by category (plastics, ferrous metals, non-
ferrous metals, electronics). Notebooks................................................................ 92 
Table 25 — Material savings [t/year] divided by category (plastics, ferrous metals, non-
ferrous metals, electronics). Tablets. .................................................................... 92 
Table 26 — Material savings divided by substance [t/year]. Notebooks (only the mass of 
electronics was considered for this assessment). ..................................................... 93 

 
8 
 
Table 27  — Material savings divided by substance [t/year].  Tablets (only the mass of 
electronics was considered for this assessment). ..................................................... 94 
Table 28 — Testing procedures of the EN 60068 series. ........................................... 96 
Table 29 — Examples of IEC 60529 test levels and short descriptions. The level number 
specifies the second digit of the IP code (BS EN 60529:1192+A2:2013). ................... 98 
Table 30 — Durability testing for notebook computers proposed by Dodd et al. (2015).
....................................................................................................................... 100 
Table 31 — Proposed logos and correlation with the ease of disassembly of batteries. 105 
Table 32 — Notebook failure rates due to hardware malfunction and accident, according 
to two sources of data: the IDC (2010, 2016) and SquareTrade (2009). .................. 107 
Table 33 — Computers expected to report failures (failure rate of 18.5 % for notebooks 
and  14.5 %  for tablets).  Estimations  based  on  the  sales expected  for 2020,  2025  and 
2030. .............................................................................................................. 107 
Table 34 — Notebooks and tablets expected to be discarded as WEEE (repair rate 80 % 
for mobile computers in the first 2 years of use, with warranty plans; repair rate 20 % for 
mobile computers older than 2 years of use, with no warranty plans; average failure rate 
of 18.5 % for notebooks and 14.5 % for tablets)................................................... 108 
Table 35 — Products (million units/year) and material (t/year) savings thanks to enhanced 
reparability (repair rate 81-84 % for computers in the first 2 years of use, with warranty 
plans, average failure rate of 18.5 % for notebooks and 14.5 % for tablets). ............ 108 
Table 36 — Products (million units/year) and material (t/year) savings thanks to enhanced 
reparability  (repair  rate  24-36 %  for  computers  older  than  2 years,  with  no  warranty 
plans, average failure rate of 18.5 % for notebooks and 14.5 % for tablets). ............ 109 
Table 37 — Methods for secure data deletion (Talens Peiró and Ardente, 2015) ........ 111 
Table 38 — Table for the calculation of the index on ‘Flame retardant in plastic components’ 
for computers (modified from (Ardente et al., 2016)). ........................................... 120 
Table 39 — Average recycling rates of different materials from PCBs separated for recycling 
(source: Chancerel and Marwede, 2016) .............................................................. 126 
Table 40 — Estimated benefits thanks to enhanced recyclability of notebooks. .......... 128 
Table 41 — Revised benefits due to the potential strategies to enhance the recyclability of 
notebooks (based on a reduced amount of waste properly collected). ...................... 129 
   

Acknowledgements

This report is part of the project ‘Technical support for environmental footprinting, material

efficiency in product policy and the European Platform on Life Cycle Assessment’ (LCA)

(2013-2017) funded by the Directorate-General for Environment.

The authors would like to thank Paolo Tosoratti, from the Directorate-General for Energy,

Unit ENER.C.3 for all the support, guidance and suggestions provided during the project.

The authors would like to also thank Dr Larisa Maya-Drysdale and colleagues from Viegand

Maagøe, for input data and suggestions provided for this work.

Finally, the authors of this report thank all of the stakeholders who contributed to this

work with constructive comments and suggestions for all the information and suggestions

provided, which were essential for the analysis of end-of-life practices for the product

group.

Executive summary

This report has been developed within the project ‘Technical support for environmental

footprinting, material efficiency in product policy and the European Platform on Life Cycle

Assessment’ (LCA) (2013-2017) funded by the Directorate-General for Environment.

The report summarises the findings of the analysis of material-efficiency aspects of the

personal-computer (PC) product group, namely durability, reusability, reparability and

recyclability. It also aims to identify material-efficiency aspects which can be relevant for

the current revision of the Ecodesign Regulation (EU) No 617/2013. Special focus was

given to the content of EU critical raw materials (CRMs) (

) in computers and computer

components, and how to increase the efficient use of these materials, including material

savings thanks to reuse and repair and recovery of the products at end of life. The analysis

has been based mainly on the REAPro method (

) developed by the Joint Research Centre

for the material-efficiency assessment of products.

This work has been carried out in the period June 2016-September 2017, in parallel with

the development of The preparatory study on the review of Regulation 617/2013 (Lot 3)

— computers and computer servers led by Viegand Maagøe and Vlaamse Instelling voor

Technologisch Onderzoek NV (VITO) (2017) (

). During this period, close communication

was maintained with the authors of the preparatory study. This allowed ensuring

consistency between input data and assumptions of the two studies. Moreover, outcomes

of the present research were used as scientific basis for the preparatory study for the

analysis of material-efficiency aspects for computers. The research has been differentiated

as far as possible for different types of computers (i.e. tablet, notebooks and desktop

computers).

The report starts with the analysis of the technical and scientific background relevant for

material-efficiency aspects of computers, such as market sales, expected lifetime, bill of

materials, and a focus on the content of CRMs (especially cobalt in batteries, rare earths

including neodymium in hard disk drives and palladium in printed circuit boards).

Successively the report analyses the current practices for repair, reuse and recycling of

computers.

Based on results available from the literature, material efficiency of the product group has

the potential to be improved, in particular the lifetime extension. The residence time (

) of

IT equipment put on the market in 2000 versus 2010 generally declined by approximately

10 % (Huisman et al., 2012), while consumers expressed their preference for durable

goods, lasting considerably longer than they are typically used (Wieser and Tröger, 2016).

Design barriers (such as difficulties for the disassembly of certain components or for their

processing for data sanitisation) can hinder the repair and the reuse of products.

Malfunction and accident rates are not negligible (IDC, 2016, 2010; SquareTrade, 2009)

and difficulties in repair may bring damaged products to be discarded even if still

functioning.

Once a computer reaches the end of its useful life, it is addressed to ‘waste of electrical

and electronic equipment’ (WEEE) recycling plants. Recycling of computers is usually

based on a combination of manual dismantling of certain components (mainly components

containing hazardous substances or valuable materials, e.g. batteries, printed circuit

boards, display panels, data-storage components), followed by mechanical processing

including shredding. The recycling of traditional desktop computers is perceived as non-

(

) Critical raw materials (https://ec.europa.eu/growth/sectors/raw-materials/specific-interest/critical_it).

(

) Ardente and Mathieux (2014). ‘Identification and assessment of product’s measures to improve resource

efficiency: the case-study of an energy using product’, Journal of cleaner production

(http://doi.org/10.1016/j.jclepro.2014.07.058).

(

) Preparatory study on the review of Regulation 617/2013 (Lot 3) — computers and computer servers (draft

report). Viegand Maagøe and VITO (2016). Prepared for European Commission DG Energy C.3

(https://computerregulationreview.eu/).

(

) The time of non-functioning or unused appliances in stock is included.

problematic by recyclers, with the exception of some miniaturised new models (i.e. mini

desktop computers), which still are not found in recycling plants and which could present

some difficulties for the extraction of printed circuit boards and batteries (if present). The

design of notebooks and tablets can originate some difficulties for the dismantling of

batteries, especially for computers with compact design.

Recycling of plastics from computers of all types is generally challenging due to the large

use of different plastics with additives, such as flame retardants. According to all the

interviewed recyclers, recycling of WEEE plastics with flame retardant is very poor or null

with current technologies.

Building on this analysis, the report then focuses on possible actions to improve material

efficiency in computers, namely measures to improve (a) waste prevention, (b) repair and

reuse and (c) design for recycling. The possible actions identified are listed hereinafter.

(a) Waste prevention

a.1 Implementation of dedicated functionality (

) for the optimisation of the lifetime of

batteries in notebooks: the lifetime of batteries could be extended by systematically

implementing a preinstalled functionality on notebooks, which makes it possible to

optimise the state of charge (SoC) of the battery when the device is used in grid

operation (stationary). By preventing the battery remaining at full load when the

notebook is in grid operation, the lifetime of batteries can be potentially extended by

up to 50 %. Users could be informed about the existence and characteristics of such a

functionality and the potential benefits related to its use.

a.2 Decoupling external power supplies (EPS) from personal computers: the provision

of information on the EPS specifications and the presence/absence of the EPS in the

packaging of notebooks and tablets could facilitate the reuse by the consumer of

already-available EPS with suitable characteristics. Such a measure could promote the

use of common EPS across different devices, as well as the reuse of already-owned

EPS. This would result in a reduction in material consumption for the production of

unnecessary power supplies (and related packaging and transport) and overall a

reduction of treatment of electronic waste. The International Electrotechnical

Commission (IEC) technical specification (TS) 62700, the Standard Institute of

Electrical and Electronics Engineers (IEEE) 1823 and Recommendation ITU-T L.1002

can be used to develop standards for the correct definition of connectors and power

specifications.

a.3 Provision of information about the durability of batteries: the analysis identified the

existence of endurance tests suitable for the assessment of the durability of batteries

in computers according to existing standards (e.g. EN 61960). The availability of

information about these endurance tests could help users to get an indication on the

residual capacity of the battery after a predefined number of charge/discharge cycles.

Moreover, such information would allow for comparison between different products and

potentially push the market towards longer-lasting batteries.

a.4 Provision of information about the ‘liquid ingress protection (IP) class’ for personal

computers: this can be assessed for a notebook or tablet by performing specific tests,

developed according to existing standards (e.g. IEC 60529). Users can be informed

about the level of protection of the computer against the ingress of liquids (e.g. dripping

water or spraying water or water jets) and in this way prevent one of the most common

causes of computer failure.

The yearly rate of estimated material saving if dedicated functionality for the optimisation

of the lifetime of batteries (a.1) were used ranges from around 2 360 to 5 400 tonnes (t)

(

) E.g. a dedicated tool or software.

of different materials per year. About 450 t of cobalt, 100 t of lithium, 210 t of nickel and

730 t of copper could be saved every year.

The estimated potential savings of materials when EPS are decoupled from notebooks and

tablets (a.2) are in the range 2 300-4 600 t/year (80 % related to the notebook category,

and 20 % to tablets). These values can be obtained when 10-20 % of notebooks and

tablets are sold without an EPS, as users can reuse already-owned and compatible EPS.

Under these conditions, for example, about 190-370 t of copper can be saved every year.

This estimate may increase when the same EPS can be used for both notebooks and tablets

(at the moment the assessment is based on the assumption that the two product types

were kept separated).

Further work is needed to assess the potential improvements thanks to the provision of

information about the durability of batteries (a.3), and about the ‘liquid-IP class’ (a.4).

The former option (a.3) has the potential to boost competition among battery

manufacturers, resulting in more durable products. The latter option (a.4) has the

potential to reduce computer damage due to liquid spillage, ranked among the most

recurrent failure modes.

(b) Repair/reuse

b.1 and b.2 Provision of information to facilitate computer disassembly: the disassembly

of relevant components (such as the display panel, keyboard, data storage, batteries,

memory and internal power-supply units) plays a key role to enhance repair and reuse

of personal computers. Some actions have therefore been discussed (b.1) to provide

professional repair operators with documentation about the sequence of disassembly,

extraction, replacement and reassembly operations needed for each relevant

component of personal computers, and (b.2) to provide end-users with specific

information about the disassembly and replacement of batteries in notebooks and

tablets.

b.3 Secure data deletion for personal computers: this is the process of deliberately,

permanently and irreversibly erasing all traces of existing data from storage media,

overwriting the data completely in such a way that access to the original data, or parts

of them, becomes infeasible for a given level of effort. Secure data deletion is essential

for the security of personal data and to allow the reuse of computers by a different user.

Secure data deletion for personal computers can be ensured by means of built-in

functionality. A number of existing national standards (HMG IS Standard No 5 (the

United Kingdom), DIN 66399 (Germany), NIST 800-88r1 (the United States (US)) can

be used as a basis to start standardisation activities on secure data deletion.

The estimated potential savings of materials due to the provision of information and tools

to facilitate computer disassembly were quantified in the range of 150-620 t/year for

mobile computers (notebooks and tablets) within the first 2 years of use, and in the range

of 610-2 460 t/year for mobile computers older than 2 years.

Secure data deletion of personal computers, instead, is considered a necessary

prerequisite to enhance reuse. The need to take action on this is related to policies on

privacy and protection of personal data, as the General Data Protection

Regulation (EU) 2016/679 and in particular its Article 25 on ‘data protection by design and

by default’. Future work is needed to strengthen the analysis, however it was estimated

that secure data deletion has the potential to double volume of desktop, notebook and

tablet computers reused after the first useful lifetime.

c.1 Provision of information to facilitate computer dismantling: computers could be

designed so that crucial components for material aspects (e.g. content of hazardous

substances and/or valuable materials) can be easily identified and extracted in order to

be processed by means of specific recycling treatments. Design for dismantling can

focus on components listed in Annex VII of the WEEE directive (

). The ‘ease of

dismantling’ can be supported by the provision of relevant information (such as a

diagram of the product showing the location of the components, the content of

hazardous substances, instructions on the sequence of operations needed to remove

these components, including type and number of fastening techniques to be unlocked,

and tool(s) required).

c.2 Marking of plastic components: although all plastics are theoretically recyclable, in

practice the recyclability of plastics in computers is generally low, mainly due to the

large amount of different plastic components with flame retardants (FRs) and other

additives. Marking of plastic components according to existing standards (e.g.

ISO 11469 and ISO 1043 series) can facilitate identification and sorting of plastic

components during the manual dismantling steps of the recycling.

c.3 FR content: according to all the recyclers interviewed, FRs are a major barrier to

plastics recycling. Current mechanical-sorting processes of shredded plastics are

characterised by low efficiency, while innovative sorting systems are still at the pilot

stage and have been shown to be effective only in certain cases. Therefore, the

provision of information on the content of FRs in plastic components is a first step to

contribute to the improvement of plastics recycling. Plastics marking (as discussed

above) can contribute to the separation of plastics with FRs during the manual

dismantling, allowing for their recycling at higher rates (in line with the prescription of

IEC/TR 62635, 2015). However, detailed information about FRs content could be given

in a more systematised way, for example through the development of specific indexes.

These indexes could support recyclers in checking the use of FRs in computers and in

developing future processes and technologies suitable for plastics recycling. Moreover,

these indexes could support policymakers in monitoring the use of FRs in the products

and, in the medium-long term, to promote products that use smaller quantities of FRs.

An example of a FR content index is provided in this report.

c.4 Battery marks: the identification of the chemistry type of batteries in computers is

necessary in order to have efficient identification and sorting, and thus to improve the

material efficiency during the recycling. It is proposed to start standardisation activities

to establish standard marking symbols for batteries. The examples of the ‘battery-

recycle mark’, developed by the Battery Association of Japan (BAJ), and the current

standardisation activities for the IEC 62902 (standard marking symbols for batteries

with a volume higher than 900 cm

) may be used as references to develop ad hoc

standards.

The benefits of actions for the design for recycling can be relevant. In particular, the

proposed actions should contribute to increase the amounts of materials that will be

recycled (6 350-8 900 t/year), in particular plastics (5 950-7 960 t/year of additional

plastics), but also metals such as cobalt (55-110 t), copper (240-610 t), rare earths as

neodymium and dysprosium (2-7 t) and various precious metals (gold (0.1-0.4 t),

palladium (0.1-0.4 t) and silver (2-7 t)). Compared to the amount of materials recycled in

the EU (2012 data), these values would represent a recycling increase of 1-2 % for cobalt,

2-5 % for palladium, and 13-50 % for rare earths.

(

) ‘Design for dismantling’ is also in line with the principles of the WEEE directive, which in Article 4 states

that appropriate measures should be encouraged ‘so that the ecodesign requirements facilitating reuse and

treatment of WEEE established in the framework of Directive 2009/125/EC are applied’.

Abbreviations

ABS acrylonitrile butadiene styrene

AC alternating current

ATA advanced technology attachment

BAJ Battery Association of Japan

BaU business as usual

BFR brominated flame retardants

BOA bill of attributes

BoM bill of materials

CAS CESG Assured Service

CD-ROM compact disc — read-only memory

CESG Communications-Electronics Security Group

CPU central processing unit

CRMs critical raw materials

DC direct current

DG Directorate-General

EEE electrical and electronic equipment

EERA European Electronics Recycling Association

EMI electromagnetic interference

eMMC embedded multimedia card

EoL end of life

EPS external power supply

EuRIC European Recycling Industries’ Confederation

FR(s) flame retardant(s)

GF glass fibre

GfK Growth from Knowledge

GHG Greenhouse gas

GPP green public procurement

GPU graphics processing unit

GSM Global System for Mobile Communications (Groupe Spéciale Mobile)

GSMA GSM Association

HDD hard disk drive

HP Hewlett-Packard

IC integrated circuit

ICT information and communications technology

IDC International Data Corporation

IEC International Electrotechnical Commission

IEEE Institute of Electrical and Electronics Engineers

IP ingress protection

IT information technology

ITU International Telecommunication Union

IZM Institut für Zuverlässigkeit und Mikrointegration

JRC Joint Research Centre

LAN local-area network

LCA life-cycle assessment

LCD liquid-crystal display

LCO lithium-cobalt-oxide

LED light-emitting diodes

LFP lithium-iron-phosphate

LIB lithium-ion batteries

LMO lithium-manganese-oxide

MFA material flow analysis

MOST Maynard operation sequence technique

MoU memorandum of understanding

NCA nickel-cobalt-aluminium

NiMH nickel-metal-hydride

NIR near infra-red

NMC nickel-manganese-cobalt-oxide

ODD optical disk drive

OEM original equipment manufacturer

OLED organic light-emitting diodes

PATA Parallel ATA

PBB polybrominated biphenyls

PBDEs polybrominated diphenyl ethers

PC personal computer

PCB printed circuit board

PMMA poly(methyl methacrylate)

PWD password

QR quick response

RAL Reichs-Ausschuss für Lieferbedingungen

RAM random-access memory

REE rare earth elements

RoHS restriction of hazardous substances

ROM read-only memory

SATA Serial ATA

SBS-IF Smart battery systems implementers forum

SD secure digital

SIM subscriber identification module

SoC state of charge

SoH state of health

SSD solid-state drive

TS technical specification

UK United Kingdom

US United States

USB universal serial bus

WEEE waste of electrical and electronic equipment

WLAN wireless local-area network

XML extensible markup language

XRF x-ray fluorescence

ZIF zero insertion force

List of definitions

Built-in functionality: a functionality provided by the product that does not rely on

components which are not already included in the said product.

Component: constituent part of a device which cannot be physically divided into smaller

parts without losing its particular function (EN 50625-1:2014).

Disassembly: non-destructive taking apart of an assembled product into constituent

materials and/or components (from Standard BS 8887-2:2009).

Dismantling: taking apart of an assembled product into constituent materials and/or

components (based on the definition of disassembly from Standard BS 8887-2:2009).

Display panel: electronic display assembly (e.g. liquid-crystal display or other

technologies) together with their casing where appropriate (revised from Directive

2012/19/EU).

Firmware: system, hardware, component, or peripheral programming provided with the

product to provide basic instructions for hardware to function inclusive of all applicable

programming and hardware updates.

Secure data deletion: the effective erasure of all traces of existing data from storage

media, overwriting the data completely in such a way that access to the original data, or

parts of them, becomes infeasible for a given level of effort.

State of charge (SoC): the ‘[…] remaining battery capacity expressed as percentage of

full-charge capacity’ (SBS-IF, 1998) and hence the ‘fuel gauge’ indicating the currently

available battery charge. The SoC may also be defined as the remaining battery capacity

expressed as a percentage of the design capacity (also ‘rated capacity’, as stated by the

manufacturer).

State of health (SoH): the ratio between a battery’s full-charge capacity over the initial

(design) capacity expressed in percentage. The SoH indicates how much of its (initially

theoretically available) capacity a battery has retained at a given time.

Technical documentation: documentation made available by manufacturers on websites,

concerning repair/recycling of products, kept available for a specified number of years

after the last product has been placed on the market.

User documentation: documentation made available by manufacturers for end-users, on

websites and user manuals, kept available for a specified number of years after the

products have first been placed on the market.

We consider resource efficiency as a combination of energy efficiency and material

efficiency. Thus, material efficiency does not directly regard resources used to produce

energy, nor energy used during the lifecycle of products (Tecchio et al, 2017).

Definitions concerning personal computer product categories and parts used in this report

are listed in the preparatory study on the review of Regulation 617/2013 (Lot 3) —

computers and computer servers led by Viegand Maagøe and VITO (2017)

(https://computerregulationreview.eu/).

1 Introduction

This report has been developed within the project ‘Technical support for environmental

footprinting, material efficiency in product policy and the European Platform on LCA’

(2013-2017) funded by the Directorate-General for Environment. The report aimed to

analyse the material efficiency of the personal-computer product group, and to identify

relevant and workable criteria on material efficiency that could be used for the revision of

the Ecodesign Regulation (EU) No 617/2013, which is currently underway.

Nowadays, most modern industrial operations are based on a linear model in which

materials are extracted and processed, products are made and are eventually disposed of

at the end of their lifespans. As evidenced by growing material scarcity around the globe,

this linear ‘take, make, dispose’ model is inherently unsustainable (Ellen MacArthur

Foundation, 2016). The European Commission is committed to a sustainable, low-carbon,

material-efficient and competitive circular economy (European Commission, 2015a), a

strategy that includes the shifting of the concept from ‘waste’ to ‘resources’, boosting the

market for secondary raw materials and taking a series of actions to encourage recovery

of CRMs. In particular, the analysis herein presented responds to the commitment of giving

emphasis to circular economy aspects in future product requirements under the Ecodesign

directive.

The analysis is based on the REAPro (

) method developed by the Joint Research

Centre (JRC) for the resource-efficiency assessment of products (Ardente and Mathieux,

2014), and it followed the results of previous assessments of specific product groups (e.g.

electronic displays, washing machines, dishwashers, enterprise servers, computers,

vacuum cleaners) in the context of the Ecodesign directive or of EU Ecolabel regulation.

We consider resource efficiency as a combination of energy efficiency and material

efficiency. Thus, material efficiency does not directly regard resources used to produce

energy, nor energy used during the lifecycle of products (Tecchio et al, 2017).

The present report begins with an analysis of the current situation for the personal-

computer product group, including: a presentation of background information including

market data, bill of materials and the environmental performance of this product group

(Section 2); an analysis of recycling, repair/reuse practices for this product group (Section

3). Based on this analysis, a series of material-efficiency ‘hot spots’ for computers is

identified (Section 4). Successively the report introduces some requirements that could be

potentially applied to this product group in the context of the Ecodesign directive,

addressing material saving (Section 5), repair/reuse (Section 6) and recycling (Section 7).

Benefits associated to these requirements are formalised and, when possible, quantified.

(

) Resource-efficiency assessment of products.

2 Background information

2.1 Market data

As reported by Viegand Maagøe and VITO (2017) on the Review of

Regulation (EU) 617/2013 (Lot 3) — computers and computer servers (draft report), sales

and stock data of personal computers

(

)

within the EU market can be derived by analysing

past sales and market trends. Database platforms such as Statista (2016) were consulted

by which market analysts estimate the worldwide shipment of desktop computers,

notebooks and tablets.

Table 1 provides the projections of estimated sales for different product categories in the

European market, focusing on 2020, 2025 and 2030. Market projections are confirmed by

recent studies, for example the work published by Risk & Policy Analysts Limited (2014).

According to that study, the European market accounted for around 20 % of global tablet

sales in 2010, but market analysts are expecting this proportion to decrease as the

European market becomes saturated; for the same reasons, market analysts are expecting

the global sales of notebooks to decrease in the coming years. Again according to

elaborations made by Risk & Policy Analysts Limited (2014) and values reported by

Statista, the share of the EU market can be estimated to be in the range of 34-37 % for

global notebook sales.

Table 1 — Estimated annual sales (2012-2030) for product categories in the EU market

(Viegand Maagøe and VITO, 2017). Values in millions of units.

Product categories

2012

2013

2014

2015

2020

2025

2030

million units/year

Notebook

50.66

47.21

46.79

42.40

41.66

41.55

41.74

Desktop computer

19.13

15.77

14.84

12.74

12.05

13.47

13.60

Integrated desktop

0.77

0.63

0.59

0.51

0.48

0.54

Thin client

1.35

1.43

1.31

1.37

Integrated thin client

0.13

0.14

0.13

0.14

Tablet/slate

28.46

44.74

45.21

40.79

38.38

38.47

38.56

Portable all-in-one

0.21

0.19

0.17

Workstation

0.71

0.75

0.80

0.79

0.82

0.85

Small-scale server

0.17

0.18

0.19

0.20

0.21

0.22

0.23

Total computers

101.59

110.95

110.19

99.05

95.24

96.73

97.20

From Table 1 it is possible to note that main shares of the market sector will be represented

by notebook computers, tablet/slate computers and desktop computers (almost 97 % of

the total number of computers in 2030). The shipments of tablets grew significantly until

(

)

Definitions of product categories are available in the preparatory study on the review of Regulation 617/2013,

prepared by Viegand Maagøe and VITO (2017).

2014, when the European market became saturated. The shipments of desktop computers

and notebooks are already gradually decreasing.

A more-detailed market analysis was focused on the two main typologies of storage that

can be used for personal computers: hard disk drives (HDDs) and solid-state drives

(SSDs). The two typologies exist for both internal and external data storage.

HDDs are traditional spinning hard drives consisting of a metal platter with a magnetic

coating on which a read/write head gets access to the data while the platter is spinning.

The more-recent SSDs, instead, store data on interconnected flash memory chips. SSDs

are generally more expensive but also faster than HDDs. According to pcmag.com (

SSDs also have better durability. SSDs have no mechanical parts in motion, even though

they do wear out over time. Thanks to a command technology that dynamically optimises

read/write cycles, however, the likelihood of encountering read/write errors in SSDs in the

first 6 years of use is very low.

According to Statista (2016), it is possible to estimate the shipments (and therefore the

production) of computers with HDDs and computers with SSDs from 2012 to 2017. The

source estimates that shipments of HDDs will decline in the future while SSD shipments

will show an increase. Projections to 2020 were developed (Figure 1). The year 2020 could

then be identified as the most probable break-even point between the two trends.

Figure 1 — Shipments of hard- and solid-state disk drives (HDD/SSD) (in millions)

worldwide in computers from 2012 to 2017 (Statista, 2016) and projections for 2018-2020

(own elaboration).

Although it is not forecast that SSDs will significantly reduce the usage of HDDs, a

technological breakthrough could cause the price of SSDs to drop significantly, which in

turn would drive replacement of HDDs by SSDs (Sprecher et al., 2014a).

2.2 Expected lifetime

Expected lifetime of products is key information to estimate potential end-of-life (EoL)

flows, and several figures can be found in literature for the personal-computer product

group. Hennies and Stamminger (2016) discussed types of obsolescence, describing

‘functional obsolescence’, which is induced by innovations, new features and new

(

) PC Magazine digital edition, provides lab-tested reviews, how-to guides and news about the latest tech

trends.

100

150

200

250

300

350

400

450

500

2012 2013 2014 2015 2016 2017 2018 2019 2020

Quantity of hard

- and solid

-state disks

(in millions)

year

HDD SSD

interfaces, and ‘desire obsolescence’, which is brought about through the desire for trends,

designs and lifestyles that makes products old-fashioned. These two types of obsolescence

are playing a role also for the personal-computer product group, and are not negligible

reasons for their relatively short (compared to other electric and electronic equipment,

such as household appliances for instance) lifetime.

As previously stated, several lifetime figures can be found in literature, especially for

notebook computers. A survey conducted by the Öko-Institut (Germany) shows that the

average duration of the first use of a notebook ranges from 5 to 6 years (Prakash et al.,

2016a). Hennies and Stamminger (2016) reported an average lifetime of 5 years before

notebooks are discarded. A Dutch study reported a lifetime of 7 years (Wang et al., 2013),

while a recent survey conducted among Austrian residents reported 4.1 years, as a useful

lifetime (namely the time until a replacement is bought).

Viegand Maagøe and VITO (2017) reported a typical lifetime of 5 years for notebooks,

6 years for desktop computers, and 3 years for tablets; the authors relied on literature

findings, expert assumptions and industry inputs (see Table 2).

Table 2 — Typical lifetime of computers and related products according to Viegand Maagøe

and VITO (2017).

Product category

Typical lifetime, years

Notebook

Desktop

Integrated desktop

Thin client

Integrated thin client

Tablet/slate

Portable all-in-one

Workstation

Small-scale servers

Hennies and Stamminger (2016) also reported the fate of notebooks after the first use:

most of them are not discarded, even when they are defective. Some 41 % are set aside

(50 % in the case of defective notebooks), only 23 % are disposed of and 33 % are passed

on.

Huisman et al. (2012) took into consideration this behaviour and analysed the residence

time, (average lifetime including the time of non-functioning or unused appliances in

stock). Different types of equipment put on the market for all years between 1990 and

2010 were considered by the authors. From their analysis, basically all appliances show

decreasing residence times. For IT equipment put on the market in 2000 versus 2010, for

instance, the residence time declined by approximately 10 %. The shortening trend for IT

product lifespan was identified also in less recent studies, in China (Yang et al., 2008) and

India (Dwivedy and Mittal, 2010).

The Austrian resident survey, mentioned before, also compared the first-use time

(products in use, time until a replacement is bought) of notebooks (4.1 years) with the

desired lifetime (the time consumers desire the product to be functioning). The latter was,

on average, 7 years (Wieser and Tröger, 2016). From these results, Wieser and Tröger

(2016) highlighted that consumers want durable goods to last considerably longer than

they are used. If asked to spontaneously name a product for which the expected lifetime

(the time people expect the product to work) is shorter than expected, notebook

computers were ranked in the sixth place (after mobile phones, TVs, washing machines,

coffee machines and dishwashers).

Table 3 — Use time and desired lifetime of products (Wieser and Tröger, 2016). n= number

of respondents (population interviewed: 1 009 Austrian residents).

First use time (n=574)

Desired lifetime (n=996)

Notebooks

4.1 years

7.0 years

Another survey, this time conducted between 2014 and 2016 among Swiss consumers

highlighted that nowadays notebook users expect longer lifetimes, compared to the past.

The survey showed how the desired lifetime of notebooks nowadays is 20 % higher than

the desired lifetime of the product type they used in the past (Thiébaud-Müller et al.,

2017).

Other main findings from Thiébaud-Müller et al. (2017) are as follows.

— The median service lifetime of notebooks (at the moment the notebook is not in

use anymore, therefore stored or disposed of) is reported to be 5 years, while the

median intended (desired) first service time is reported to be 6 years.

— The median ‘second service lifetime’ (the active use of a second-hand notebook) is

2 years.

— The median ‘storage time’ (the time between the active use of a new device and

its final disposal or its transfer to a different user) is 1 year.

— About 60 % of the notebooks go to storage after first use, about 10 % to second

use and 20 % to collection schemes (

— A large share of the devices stored go to second use. This means that in total nearly

30 % of all notebooks in Switzerland go to second use.

— About 20 % of the devices in second use go to third use. The analysis of the change

over time of the service lifetime (histogram) show no significant trend for the

temporal change of the service lifetime for notebooks.

Finally, the Growth from knowledge (GfK) consumer panel collected the average first-use

time in Germany, 2004-2007 and 2010-2012. The first-use time seems to have a peak in

2005/2006 with 6 years and declines to 5.1 years in 2012. The reasons for replacement

and whether the notebooks have a second life were not determined (Prakash et al.,

2016b).

Figure 2 — Average first-use time of notebooks in Germany (n=2 268 in 2012, lowest

value n=244 in 2004; 2008, 2009: not specified) (Prakash et al., 2016b).

(

) Remaining 10 % go to donation, municipal waste, or unknown.

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Average first use time (years)

Year

As a final remark, consumer behaviour can also be listed as a reason for the present

problem of increasing amounts of electronic waste (the ‘desire obsolescence’). As prices

fall, consumers can receive incentives to buy new appliances instead of continuing to use

their current ones (Aladeojebi, 2013).

2.2.1 External power-supply lifetime

The active lifetime of EPS used for portable devices is estimated to be 5 years, aligned

with the expected lifetime of notebooks (Table 2). Accordingly, when the lifetime of the

device is shorter (e.g. 3 years for tablets, see Table 2), the active life of EPS is reduced

as well. The overall lifetime, instead, is largely determined by the lifetime of electrolytic

capacitors an EPS is made of (IEC/TS 62700, 2014). The overall lifetime (i.e. the age of

discarded appliances) is often significantly longer than the active lifetime, as the stocking

in a drawer phenomena is common in the case of small electric devices (Bio Intelligence

Service, 2007). Risk & Policy Analysts Limited (2014) estimated that only 5 % of

consumers dispose of their old mobile phone when they purchase a new one and a typical

consumer keeps an old handset for 2.37 years before it enters the waste stream. The

same delay can be assumed to their related EPS. Furthermore, according to IEC/TS 62700

(2014), if one EPS is used for several computers simultaneously, the lifetime may become

shorter (each manufacturer decides on the lifetime of electrolytic capacitors considering

how many years the computer is used (IEC/TS 62700, 2014)).

2.2.2 Battery lifetime

Battery durability is a key feature for users. In a survey conducted by the IDC (2010) (

68 % of respondents confirmed that the battery lifetime on their notebook computers was

not sufficient for their business needs, and over half stated that battery failures caused

problems for their business. The most common problem was lost productivity, cited by

45 % of respondents, followed by lost/delayed sales (22 %) and loss of critical company

data (17 %).

The durability of batteries potentially limits the lifetime of the device it is powering, if

battery replacement is economically not feasible, or technically not possible. This may lead

to early disposal of devices and thus contradicts the overall objective of material efficiency.

This is especially important concerning lithium-ion batteries (

) (LIB), not only do LIB

contain a high amount of critical materials such as cobalt (see Section 2.4.1), they also

involve substantial environmental impacts during their manufacturing (Section 2.6).

Battery durability is determined by a battery’s specific cycle life and calendar life. Cycle

life is usually described by the number of charge/discharge cycles a battery can withstand

before losing a certain portion of its initial capacity. A cycle is defined as ‘an amount of

discharge approximately equal to the value of design capacity’ (SBS-IF, 1998), with design

capacity referring to the theoretical capacity of a new battery (pack) (also: ‘rated capacity’

during a 5-hour discharge, as declared by the manufacturer). Today’s LIB inevitably lose

a minor amount of their capacity with each charging cycle due to a number of physical

and chemical processes (Broussely et al., 2005; Sarre et al., 2004; Schmalstieg et al.,

2014; Vetter et al., 2005). A battery’s cycle life is determined by many factors, such as

the quality of the manufacturing processes, the temperature while charging and

discharging and the cycle depth, among others (Vetter et al., 2005). Calendar life is

described by the portion of capacity a LIB inevitably loses over time, even though it is not

in use, for example while in storage. The rate at which an LIB loses capacity over time is

also determined by a number of factors, such as the surrounding temperature and its SoC

(Vetter et al., 2005).

It has been found that one major factor determining both the cycle life and calendar life

of LIB is the SoC. The SoC is the ‘[…] remaining battery capacity expressed as a percentage

(

) International Data Corporation, Framingham, Massachusetts (MA), US (www.idc.com).

(

) See Section 2.3.5 for detail on other types of batteries.

of full-charge capacity’ (SBS-IF, 1998) and hence the ‘fuel gauge’ indicating the currently

available battery charge. Studies have shown that the cycling of a battery around a very

high SoC (between 90 and 100 %) is particularly damaging and cell capacity fades

comparatively quickly. However, when cycled around an average SoC of 50 % (between

45 and 55 % SoC), the cycle life increases dramatically (Schmalstieg et al., 2014).

Similarly, calendar life increases with lower levels of SoC: as can be seen in Figure 3 (part

a), the capacity of battery cells with higher SoC fades considerably quicker compared to

those with lower SoC when in storage. For example, after 300 days of storage at 50 °C, a

battery with 90 % SoC has lost more than 20 % of its capacity, while a battery with 10 %

SoC has lost only around 5 %. Hence, it can be concluded that a high SoC during use and

storage of a notebook battery can be expected to shorten its useful life considerably.

Ideal conditions for storing a battery over a longer period is said to be at around 50 %

SoC. This avoids the damaging effects of a high SoC on one hand and, on the other hand,

avoids running into very low SoC through self-discharge, where battery cells may be

damaged irreversibly via deep discharge (e.g. Apple, 2016).

Usually battery life is stated in charge/discharge cycles before the original capacity

degrades to 80 %: for consumer products being between 300 and 500 cycles (Battery

University, 2016a) and up to 1 000 cycles (Apple, 2016). For heavy users who charge their

notebooks or tablets every day, this would amount to a total lifetime of the battery to up

to 1.4 years (500 cycles, 1 cycle per day) or 2.8 years (1 000 cycles, 1 cycle per day),

respectively. Of course, batteries can continue to be used even below 80 % capacity,

although the runtime of the device will be decreased.

However, as discussed above, the number of charging cycles alone is not sufficient to

predict the lifetime of LIB. This is also indicated in a study examining the durability of

notebooks used for several years in office environments (administration) in Germany. It

was found that the cycle frequency was quite low, with around 50 % of the notebook

batteries only accumulating 30 cycles or less per year. Hence, it was assumed that the

notebooks had mostly been used stationary (possibly with docking stations). However,

despite the low cycle count, the capacity had decreased dramatically in many cases

(Clemm et al., 2016). This indicates how the cycle count alone is not a good indicator to

project battery durability and that factors such as the surrounding temperature and

average SoC, among others, need to be taken into account.

Data from industry show how the SoH (the ratio between a battery’s full-charge capacity

over the initial (design) capacity) is projected under varying use patterns (Table 4). It is

shown that the capacity is expected to fade quicker in a notebook used stationary in a

docking station and charged only once a week, compared to a notebook cycled daily. While

no difference is expected under low power loads (word processing, email), the effect is

pronounced under moderate and high power load. Hence, the factors increasing capacity

fade are high temperatures and high discharge rate, rather than the cycle count.

Figure 3 — Calendar ageing of nickel-manganese-cobalt-oxide (NMC) cells over time

depending on SoC at an elevated temperature (50 °C). Capacity fade (ratio between

current and initial capacity) under varying SoC over time (in days) is shown in diagram

(a) and the corresponding increase in internal resistance (ratio between current and initial

resistance) in the cells in (b) (Schmalstieg et al., 2014)

Table 4 — Full-charge capacity projections after 1 year of use (HP Inc., 2016)

Power load

(applications)

Mobile computer battery

cycled daily

(25 ºC (77 ºF))

Stationary computer

(with docking station)

Battery cycled weekly

(> 35ºC (95 ºF))

Low (word processing,

internet, email) >

80 %

Moderate (wireless,

spreadsheets, database

management)

80 %

70 %

High (computer-aided

design, 3D games, DVDs,

high LCD brightness)

60 %

50 %

2.3 Bill of materials

The present section illustrates a number of available studies investigating the composition

of computers and computer components. These references have been used to estimate

the average bills of materials (BoMs) for desktop computers, notebooks, tablets, EPS and

batteries. Reference values were also retrieved by using information directly collected by

the authors of the present report, or public declarations by manufacturers.

2.3.1 Desktop computers

Several BoMs are available for desktop computers, even if not always directly comparable.

A material flow analysis (MFA) at the level of specific materials was conducted by Van

Eygen et al. (2016), who provided the average materials composition of desktop

computers: ferrous metals (37 %), aluminium (5 %), copper (4 %), precious metals

(0.01 %), other non-ferrous metals (1 %), plastics (19 %), minerals and others (34 %).

Other studies provided more-specific BoMs. Among the most recent, the studies conducted

by Song et al. (2013) and Teehan and Kandlikar (2013) can be cited.

Song et al. (2013) analysed a Dell desktop-computer unit (Table 5), but did not disclose

which model. Teehan and Kandlikar (2013), on the other hand, worked on a specific Dell

Optiplex 780 Minitower desktop (Table 6). According to Dodd et al. (2016, 2015), within

the production phase of desktop computers, specific components can be identified as

environmental ‘hot spots’ such as the motherboard (often referred also as ‘mainboard’)

and other printed circuit boards (PCBs), the CD-ROM, the HDD and the power supply.

CRMs and precious metals, such as silver, gold and palladium, contained in the

motherboard and other PCBs, can be relevant for various environmental impact categories.

Table 5 — Desktop-computer bill of materials (BoM) according to Song et al. (2013).

Packaging included.

Category

Voltage [V]

Current [A]

Power [W]

Small 1

1.5 to 3

7.5 to 15

Small 2

2.5

Small 3

2.25

Medium

Big

4.25

2.5.3 USB cables and connectors

Micro-USB (

) connectors may represent a way to grant compatibility and interoperability

of common EPS.

As mentioned in Section 2.5.1, a voluntary commitment to harmonise EPS in the EU was

set among mobile-phone manufacturers and, as a result, it is now requiring EPS

compatibility through micro-USB connectors. According to Risk & Policy Analysts Limited

(2014), the market share of tablets with micro-USB charging solutions has increased over

the period 2009-2013. For notebooks, however, very few micro-USB charging solutions

are adopted and proprietary charging solutions are dominant.

— A model of the tablet market suggests that between 2008 and 2013, 69 % of

models were supplied with a proprietary EPS but the micro-USB charger has

become more commonplace, rising from 17 % of sales in 2011 to 47 % in 2013.

— The power requirements of notebooks can vary greatly depending on the size and

internal components, with most charging in the range 40 W to 90 W, although this

can be as low as 15 W and as high as 240 W. Therefore, micro-USB connectors are

not suitable yet for charging many notebooks, having a limit at 60 W (3 A of current

and 20 V of voltage) (USB Implementers Forum, 2016).

As found by Risk & Policy Analysts Limited (2014), 2013 was a turning point in terms of

EPS, as the use of micro-USB EPS noticeably increased while proprietary EPS decreased.

By looking at the tablets sampled by the authors (top 20 Amazon best sellers (

), sampled

in September 2016), the use of micro-USB charging resulted the most common technology

adopted.

Power delivery for portable information devices is continuously evolving and new USB

technologies tend to combine data transport with high power delivery (IEC/TS 62700,

2014). Indeed, new USB technologies seem to target a higher range of power delivered

to the device. However, this may not be sufficient to cover the specification of the Standard

IEEE 1823 (2015), which specifies a power range of 10-240 W delivered to the device.

One opportunity may be represented by the technology USB power delivery (USB PD),

which supports up to 100 W of bidirectional power (sink/source) and up to 5 Gbit/s of data

transport over USB (IEC/TS 62700, 2014), which became 10 Gbit/s with the release of the

USB 3.1 Gen 2 specification (USB Implementers Forum, 2016). This technology is not

covering the entire range proposed by the Standard IEEE 1823, but is a suitable solution

for devices requiring 60-100 W (notebooks included).

While the common EPS described in the Standard IEC 62684:2011 (for mobile phones)

adopted a micro-USB connector (as in Figure 9), another new USB specification for a small

24-pin reversible-plug connector was developed and named USB type C (Figure 10). USB

type-C cables and connectors were developed to supply mobile devices, including

notebooks and tablets, building on the new USB 3.1 Gen 2 standard for power and speed

(

) Universal serial bus.

(

) Nine different manufacturers were present in the tablet sample.

performance, which supports up to 100 W of bidirectional power (USB Implementers

Forum, 2016). Among the main features it is possible to identify the reversible-plug

orientation, the reversible cable direction and the scalable power. The Recommendation

ITU-T L.1002 suggests implementing the USB type-C connector for the interface of EPS,

in order to support broad reusability and interoperability (USB type-C receptacles as

specified in IEC 63002, IEC 62680-1-2 and IEC 62680-1-3).

Figure 10 — USB type-C cable and connectors (image credits: ©USB Implementers Forum

2014)

2.6 Environmental impacts

Several studies focused on the environmental assessment of ICT devices are available in

the literature. Several LCA studies have shown that the manufacturing phase of ICT

products has a proportionally increasing environmental impacts as compared to the use

phase (Prakash et al., 2016c). However, a main issue is represented by the

representativeness of the product or system under analysis: as the technological progress

in this sector is very fast, environmental results can be considered outdated after few

years. Therefore, relying on dated studies might divert from the current state of play, and

deep knowledge on the system under analysis is needed.

In this regard, Deng and Williams (2011) studied the question of how to measure

technological progress at the product level, by using a 1997 desktop computer as a base

case and a 2007 desktop computer as a replacing product. Generally, the production phase

(which involves material extraction/processing and the manufacturing phase) and use

phase have the highest impact in the life cycle of ICT products. The use phase seems to

be predominant in energy consumption and global warming for some ICT products but for

others, especially energy-efficient and low-weight products, manufacturing may dominate

(Arushanyan et al., 2014). In the desktop-computer case study, on one hand the

technological progress reduces the energy consumption of electronic devices (which was

measured by the authors as the energy required by the transistors), on the other hand it

might induce demand for more sophisticated components and for more powerful chips

(which contain many more transistors), and therefore shift the highest impact to the other

life-cycle phases (Deng and Williams, 2011).

Many studies focused on the initial stage of the ICT-device life cycle, using a ‘cradle-to-

gate’

approach, focusing mainly on the production phase. Teehan and Kandlikar (2013)

estimated and compared the embodied CO

emissions of 11 ICT products, including

desktop computers, notebook computers, a thin client device, an LCD monitor, small

mobile devices (e.g. tablets and e-readers), a rack server and a network switch. The

authors deviated from the conclusions drawn by Deng and Williams (2011), as they

claimed that embodied CO

emissions for newer products are 50-60 % lower than

corresponding older products with similar functionality, largely due to decreased material

usage, especially reductions in integrated circuit content. Furthermore, the embodied CO

impact identified in the study was found to be linear with respect to mass, with a range of

27-39 kg CO

equivalent (eq) per kg of ICT product. This range, however, was judged to

be very general and not totally appropriate for comparisons (Andrae, 2016). Also

Malmodin et al. (2014) estimated the embodied impact of desktop computers (200-800 kg

eq/device) and notebook computers (100-400 kg CO

eq/device), with very wide

ranges.

For the EoL phase, treatments have to be properly managed, as electronic waste may

contain hazardous constituents that may negatively impact the environment and affect

human health if not properly managed (Nnorom and Osibanjo, 2008). Also in this case it

is possible to obtain data from the literature, but impacts of EoL processing deeply depends

on the approach used to model impacts. Andrae (2016) recently published a review of

methodological approaches used to conduct LCAs of consumer electronics. The author

stated that LCA is the primary tool to study reuse, recycling and remanufacturing for an

electronic product. However, external comparisons among LCAs are only meaningful if

models from different companies (including functional unit and system boundaries) are

technically comparable (Andrae and Vaija, 2014). Van Eygen et al. (2016) conducted an

MFA devoted to the natural-material consumption of the recycling-chain system. Desktop

and notebooks were the product groups analysed by the authors: collection, primary

treatment and end-processing were the three phases of the EoL. The biggest impact for

the primary-treatment step is caused by the use of chemicals, while the production of

secondary metals represented the first cause of impacts for the end-processes. The

treatment of the notebook batteries was found to be responsible for around 16 % of the

impact.

Many studies have assessed the environmental impacts of LIB of electric vehicles in recent

years, however, data on batteries powering mobile devices such as notebooks, tablets and

smartphones is scarce. Clemm et al. (2016) calculated the environmental impact of the

manufacturing phase (based on primary industry data), distribution and EoL phase of a

notebook LIB. The consumption of energy during the use phase is not assessed, as it

serves the purpose of powering the notebook and thus must be allocated to the notebook

itself. It was found that the production phase dominates the life cycle in all impact

categories (e.g. 95 % in terms of global warming potential (GWP)), while the recovery of

cobalt and nickel lowers the overall impact by a few percentage points (e.g. 4.3 % GWP).

A contribution analysis found that the cells have the largest impact (87 % GWP), followed

by the battery-management system (12 % GWP) and finally the housing. As a

subcomponent of the cells, the cathode material (LCO) has the highest impact, followed

by the anode (graphite) and the electrolyte (LiPF

). The consumption of auxiliaries also

contributes up to 24 % of the impacts (acidification potential) to cell production.

The material composition of the notebook battery is comparable to batteries powering

tablets and smartphones and thus can be used for those devices as well.

Finally, focusing on the external power supply, Cucchietti et al. (2011) calculated the

environmental impact of the production phase of chargers for mobile phones, by using a

‘cradle-to-gate approach. The considered impact categories were climate change

(measured by means of the GWP indicator) and energy demand (measured by means of

the gross energy requirement indicator) and for both of them the main contributor to

results was represented by electronic components (75-80 % of the embodied impacts).

The authors concluded that also in the case of the other impact categories, the contribution

to the results of electronics remains higher than 70 %.

Table 19 — Relevant sources in the scientific literature

Authors and year

Type of product

Scope — impacts

Andrae & Andersen, 2010

LCA of desktop computers

and notebook computers

Literature review — GWP,

Energy demand

Benoit et al., 2012

Generic notebook

computer

Cradle to grave — social

impacts

Choi et al., 2006

Personal computer

Cradle to grave — several

indicators

Ciroth & Franze, 2011

Specific notebook

computer

Cradle to grave — several

indicators + social impacts

Duan et al., 2009

Desktop computer

Cradle to grave — several

indicators

Ekener-Petersen and

Finnveden, 2013

Generic notebook

computer

Cradle to grave — social

impacts

Eugster et al., 2007

Desktop computer

Cradle to grave

Manhart and

Griesshammer, 2006

Notebook computer

Cradle to gate — social

impacts

James and Hopkinson,

2007

Personal computer

several indicators

PE International, 2008

Notebook computer

Global warming

Prakash et al., 2016

LCA of desktop computers,

notebook computers and

mini desktop computers

Cradle to grave — GWP

Teehan & Kandlikar, 2012

Desktop computer

Literature review — GWP,

Energy demand

Williams, 2004

Personal computer

Use phase — energy use

2.6.1 Standards for environmental assessment of ICT products

The European Framework Initiative for Energy & Environmental Efficiency in the ICT Sector

grouped a series of standards focused on the environmental analysis of ICT products.

Narrowing the analysis to goods only, it is possible to list the following.

— ETSI 203 199/ITU 1410: Methodology for environmental LCA of ICT goods,

networks and services.

— IEC 62921: Quantification methodology for greenhouse gas emissions for

computers and monitors.

— IEC 62725: Analysis of quantification methodologies of greenhouse gas emissions

for electrical and electronic products and systems.

— IEC 50600-4: Design of data centre facilities and infrastructures.

— Greenhouse gas (GHG) protocol ICT Hardware: Product life-cycle accounting and

reporting standard ICT Sector Guidance (Chapter 6). Guide for assessing GHG

emissions of hardware.

— GHG protocol ICT Software: Product life-cycle accounting and reporting standard

ICT sector guidance (Chapter 7). Guide for assessing GHG emissions related to

Software.

— GreenGrid Carbon-usage effectiveness: A green grid data centre sustainability

metric (carbon-usage effectiveness (CUE)).

— EU energy star: Labelling energy-efficient office equipment.

— EPEAT: Electronic product environmental assessment tool, by US EPA.

3 Analysis of end-of-life practices for the product group

For the analysis of the material-efficiency aspects of computers the REAPro (resource-

efficiency assessment of products) method (Ardente and Mathieux, 2014) was applied.

The method starts from the analysis of current EoL practices and processes, to identify

product-resource-efficiency ‘hot spots’. ‘Hot spots’ are those product’s aspects that are

relevant for some observed EoL processes, as for example: components that are relevant

for the product durability, content of hazardous substances or relevant materials (e.g.

precious or CRMs), parts that difficult to be treated and recycled, etc.

As highlighted in Section 2, the greatest part of the market for the computer product group

is represented by notebook computers, tablet/slate computers and desktop computers

(97 % of the total number of computers, in 2030) therefore the next sections have a

particular focus on these three subcategories.

Section 3.1 illustrates processes currently performed for the recycling and recovery of

computers, representing the so-called EoL scenarios (Ardente and Mathieux, 2014).

Section 3.2 discusses some potential alternative processes to the scrapping of a product,

such as prolonging its lifetime through repair and reuse.

3.1 Analysis of recycling/recovery practices

The recycling of computers is regulated by the European WEEE directive (European Union,

2012). According to this directive, components in computers that require selective

treatments include:

— batteries;

— PCB larger than 0.1 dm

;

— LCDs panels larger than 1 dm

;

— external electrical cables;

— mercury-containing components, such as switches or backlighting lamps;

— plastic-containing brominated FRs;

— electrolyte capacitors containing substances of concern (height > 25 mm, diameter

> 25 mm or proportionately similar volume).

However, the treatment of these components (and, in general, treatment of computers)

are variable according to their type. Peculiarities of computer recycling will be discussed

in the following sections.

3.1.1 Recycling/recovery of desktop computers (without integrated

display)

The recycling of desktop computers currently follows two main scenarios, combining

optional and (to a certain degree) accurate dismantling and depollution of the computer,

with the following shredding and material recovery.

The first scenario (manual dismantling as initial treatment) has the benefit of separating

the components with high integrity and purity, which allows a higher recovery rate in their

following recycling processes. The main disadvantage is related to the labour cost and to

the higher level of time required (with consequently a lower amount of waste treated per

hour). Inversely, the second scenario (shredding as initial treatment) presents better

economic performance (in terms of costs of treatment per tonne), while the efficiency of

the sorting of materials is lower. The balance between the two scenarios is therefore

represented by the potential economic gain from the additional recovery of certain

precious metals (e.g. palladium, gold and silver) and valuable materials (e.g. copper) due

to the dedicated-manual dismantling compared to the labour costs. It is also highlighted

that the content of steel and aluminium does not represent a discriminating factor between

the two EoL scenarios, since these materials are generally recovered at high rates with

mechanical treatments. On the other side, the separation of plastics does not create an

economic gain, since they have low recyclability (due to the content of several additives

as FRs and fillers) and limited value. Shredded plastics from computers are generally

contaminated by various other fractions and suitable for the manufacture of lower quality

products (downcycling) or incinerated.

Based on direct observations of the authors at some European recycling plants (in Belgium,

Germany, Spain, Italy and France) within the current and previous projects (Ardente et

al., 2013; Ardente and Mathieux, 2012 a; Talens Peiró et al., 2016; Talens Peiró and

Ardente, 2015), manual disassembly is generally largely implemented for desktop

computers, mainly because of the modular design and low efforts for the dismantling (all

components are generally fastened with standard screws and full disassembly takes

around 2-4 minutes) compared to the additional gains from the sorting and the dedicated

recycling of certain components (e.g. SSD, HDD, ODD and PCBs — including the

motherboard, CPU, RAM modules and graphic cards). However, this analysis was carried

out on waste desktop computers at the recycling plants, concerning mainly devices

produced some years before. This implies that future products could pose some

dismantling problems especially for new devices which are of very small dimensions,

sometimes commercially referred as ‘mini’ desktop computer. These computers are

characterised by a very compact structure (similar to that of games consoles). Based on

recent studies by Prakash et al., they have a lifetime of 5 years and lower-life-cycle GWP

(959 kg CO

eq.) compared to notebooks and normal desktop computers. However,

Prakash et al. did not provide specific detail on the recycling of ‘mini’ desktops, limiting to

report the amount of greenhouse gas emissions for their disposal (estimated at 4.8 kg CO

eq. in comparison to the 9.6 kg CO

eq. of a normal desktop) (Prakash et al., 2016b).

From interviews with recyclers it was not possible to collect much information about the

EoL of ‘mini’ desktop, since this type of computer had not yet reached the recycling

facilities. Even the scientific literature is lacking for such information.

Only one manufacturer was found to be providing public information concerning EoL

disassembly instructions for some mini desktop computers (

)

(

), identifying the

components that require selective treatments (such as PCBs, batteries, plastics containing

brominated FRs, electrolytic capacitors (contained in the power supply) and external

cables) (HP, 2016). This document also includes a detailed sequence of pictures that

illustrate all the procedures and steps needed to disassemble all the main components of

the product, including the frames, various PCBs (motherboard, memory, wireless LAN

card), the storage systems (HDD and SSD), the fan and thermal units) (Figure 11).

(

)

http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/disassem

bly_deskto_201512319023191.pdf

(

)

http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/disassem

bly_deskto_2014516234519169.pdf

Figure 11 — Disassembly sequence of a mini desktop computer (HP, 2016)

Videos on the web are another available source of information concerning the disassembly

of mini desktop, although they do not relate specifically to recycling activities but mainly

to repair or illustrative purposes (

)

(

) (

)

(

). According to these videos, screws and

snap fits are mostly used for the fastening. The disassembly appears generally

straightforward, although some difficulties could arise because of the use of tiny screws,

screws covered by footpads or rubber, and the position of snap fits and screws which are

not always easily accessible. In one case the presence of a 1 000 mAh battery was also

observed inside the mini desktop (

) (Figure 12).

(

) https://www.youtube.com/watch?v=18Q_-23f8Mw

(

) https://www.youtube.com/watch?v=PTKKWTau-Pc

(

) https://www.youtube.com/watch?v=O-CJxAF2XFc

(

) https://www.youtube.com/watch?v=PrT2ycNjotI

(

) Larger capacity batteries are intended for certain models, making them more similar to notebooks

(http://www.digitaltrends.com/computing/the-voyo-v2-mini-pc-is-a-desktop-that-thinks-its-a-laptop/) or

power bank (http://www.komu.it/prodotti/mini-pc/).

1. Remove

bottom screws

2. Remove

bottom case

3. Remove HDD

5. Remove

motherboard and

system battery

4. Disassembly of

the HDD

6. Remove memory

card

7. Remove WLAN

card

8. Remove SSD

9. Remove fan and

thermal module

Figure 12 — Detail of the interior of a mini desktop computer containing a battery (

The compact structure of mini desktops could make their recycling similar to that of games

consoles (mainly based on the manual disassembly to sort out PCBs and successive

shredding of the remaining parts (

). With increasing miniaturisation of electronics such

as ‘system on chip’, there is increased scope for reducing the size and weight of the

devices, with associated reductions in material and transport. In addition, it is possible to

design products for recycling, such as reducing the number of different types of plastics

and simplifying disassembly. However, very compact products may be harder to

disassemble and recycle, and lighter products may be weaker which could increase the

amount of packaging required (AEA, 2010). Small dimension can also cause the product

to be improperly disposed of, for example, into waste bins (Huisman et al., 2015).

Available EoL information on mini desktop is, anyway, still limited and will be

complemented with experience from recyclers once this type of waste will reach the WEEE

facilities.

Undifferentiated shredding of desktop computers was not directly observed, but it is

discussed in the scientific literature especially concerning the potential losses of precious

and valuable substances contained in PCBs (Chancerel et al., 2009; Rahimifard et al.,

2009). EoL scenarios for desktop computers are following described in detail (Figure 13).

— Scenario 1: Manual dismantling, shredding and mechanical sorting.

 The desktop computer is loaded on the dismantling table. Cables are

extracted (when present) (

 After opening the casing (metal lid, plastic frames), the operator dismantles

all internal components, starting from cables and connectors, various PCBs

(such as motherboard and graphic cards) and power supply. The

dismantling proceeds with all the relevant components, including storage

system (SSD or HDD), and ODD (when present).

 Additional dismantling is undertaken on key components, especially on the

motherboard, to separate the memory RAM, the CPUs (after the preventive

extraction of the heat exchanger), and button-cell batteries, (when

present). These components are separated because of their high values in

terms of precious-metal content.

(

) From: https://www.youtube.com/watch?v=JjddUXpkZf4

(

) https://www.youtube.com/watch?v=Zs2SpbNYA0Y

(

) Waste desktops generally arrive at the recycling plant already deprived of their external cables.

 Components separated are sorted out into differentiated boxes and

successively addressed to external dedicated recycling plants (

) to

optimise material recovery.

 Unsorted components are shredded and successively metal and plastic

components are sorted via mechanical treatments (e.g. magnetic and eddy-

current separators, densimetric separators).

Examples of video on this scenario are available on the web (

)

(

— Scenario 2: Shredding and mechanical sorting.

 The desktop computer is loaded onto a conveyor belt after extracting

external cables (

 Metals and plastic components are sorted by mechanical treatments (e.g.

magnetic and eddy-current separators, densimetric separators).

Examples of videos on this scenario are available on the web (

)

(

) A large variability of the dedicated recycling processes was observed for components dismantled at the

recycling plants. For example, PCBs are sent to metal smelters, other electronics to companies specialised

in the concentration of valuable fractions, and large plastic parts to companies specialised in polymer

sorting. Material fractions which are high concentrated thanks to deep manual dismantling or high-

technology plants have generally a higher value but also, higher costs. The selection of the further

processing of sorted components depends, therefore, on the balance between the costs for processing and

their potential residual value.

(

) https://www.youtube.com/watch?v=5BkgSEBIFjw (accessed September 2016).

(

) https://www.youtube.com/watch?v=uSvfun8FC-c (accessed September 2016).

(

) In some cases a pre-shredding phase is possible, where the desktops are partially shredded and opened,

and afterwards recycling operators hand-pick and sort some parts for separate recycling (such as HDDs or

pieces of PCBs).

(

) https://www.youtube.com/watch?v=gDtj_Skhffg (accessed September 2016).

(

) https://www.youtube.com/watch?v=RXZMM6_TRrE (accessed September 2016).

Figure 13 — Desktop computers EoL scenarios. Scenario 1 (Manual dismantling, shredding

and mechanical sorting); Scenario 2 (Shredding and mechanical sorting)

3.1.2 Recycling/recovery of integrated desktop computers

Integrated desktop computers are a particular case of desktop computers that use an

integrated display. A few examples have been found concerning the EoL disassembly of

certain integrated desktops (

) (

), and suggest the disassembly of the following

components for selective treatments: various PCBs, batteries, liquid-crystal displays

(LCDs) and external cables. An example of the disassembly sequence for an integrated

desktop computer is illustrated in Figure 15. Disassembly can be made with standard tools

(screwdriver for screws type ‘TORX T8’). No relevant differences are identified for the

recycling processes for integrated desktop with or without touch screens.

Since this type of computer has only recently been introduced to the market, their market

share is still little (Viegand Maagøe and VITO, 2017), therefore, the number of these

products reaching EoL is still limited. Based on direct interviews of the authors of this

present report with recyclers, it resulted that currently very few samples of integrated

desktops have been treated so far, and this waste is generally recycled in the process line

for the treatment of electronic displays (monitors and televisions). It is also recognised

that recycling operators cannot easily distinguish integrated desktop computers from

simple computer monitors based on a superficial look at the exterior.

Even the disassembly information suggested by the manufacturer for integrated desktops

(as detailed in Figure 15) is very similar to that recommended for electronic displays, as

can be observed by a comparison with similar EoL information provided for monitors (

(

)

http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/disassem

bly_deskto_2016524202927.pdf

(

)

http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/disassem

bly_deskto_201093204739.pdf

(

) Various similarities can be observed between the disassembly of integrated desktops and the disassembly

of monitors, as for example in:

Scenario 1)

Scenario 2)

Waste desktop

computer

Manual

dismantling

- Cables / connectors

- Power supply

- Storage system

- ODD

to dedicated

recycling processes

PCBs

- Memory

- CPU

- Button batteries

- Remaining parts of PCB

to dedicated

recycling processes

Additional

dismantling

Residual parts (metal

and plastic casings)

Shredding

Material

recovery

Waste desktop

computer

Shredding

Material

recovery

Some minor differences are observed in the disassembly of certain electronic components

(such as SSD, HDD or ODD) in use in the integrated desktop while generally missing in

electronic displays.

Evidences currently collected (from recyclers and the little available documentation from

manufacturers) suggest that the recycling of integrated desktops is analogous to that of

standard monitors and other electronic displays. Therefore, potential material-efficiency

requirements for integrated displays could be built analogously to those of electronic

displays (promoting the ease of dismantling and the provision of information for recyclers).

Figure 14 — Example of an integrated desktop computer.

http://h22235.www2.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/

disassembly_monito_200983185211.pdf

Figure 15 — Disassembly sequence for an integrated desktop computer (

)

(

)

http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_MultiCountry/disassem

bly_deskto_201093204739.pdf

1. Release 9 screws for motherboard shielding,

then remove motherboard shielding assay

2. Remove RAM from motherboard

3. Release screws for converter board,

and then remove converter board

4. Release 4 screws for power & ODD & I/O board,

then remove them.

5. Release 4 screws for power & ODD & I/O

board, then remove them

Detail of the LCD panel disassembly process

a. Remove 3 screws and the PCB cover film from

the module

b. Release the hooks around the module for

remove the front metal frame

c. Take off panel assembly

d. Release lamp wire from housing and remove

tape

e. Remove plastic hosing

f. Remove plastic hosing

e. Remove light guide panels, lamps and films

Integrated desktop disassembly

3.1.3 Recycling/recovery of notebooks

Recycling of notebooks assumes, after the extraction of the battery and the display panel

(which is usually done manually), further manual sorting (e.g. removal of circuit boards,

and capacitors (

), storage systems or optical drives) and/or mechanical liberation

(shredding). Thereafter, further separation and sorting generates fractions, which are then

forwarded to final treatment (

The display panel is usually further dismantled manually or semi-automatically (

) into

fractions and components, e.g. iron and plastic fractions, and LCD panel and circuit board

fractions (

). At present, LCD panels are either stored for future treatment or treated with

technologies that are still in an early development stage or under development (usually,

the polarisation foils are removed from the LCD panel, the LCD panel is mechanically

broken down (e.g. crushed) and indium is mobilised through hydrometallurgical treatment

(Rasenack and Goldmann, 2014; Rotter et al., 2012)). Other fractions are forwarded to

be further processed using interim and final treatment technologies.

In principle, treatment operators combine different mechanical and manual dismantling

and separation methods, depending on which components they target and whether they

have acceptors for special parts that are difficult to process such as HDDs. In the following,

two scenarios are presented: mechanical treatment after depollution (scenario 1), or

medium-depth manual dismantling of the notebook (scenario 2).

— Scenario 1: Mechanical crushing and sorting. After the removal of the battery and

display panel, the entire device is treated in a medium shredder for further

separation of the different fractions (see Figure 16).

— Scenario 2: Manual medium-depth dismantling. After the removal of the battery

and display panel, certain high value components are manually recovered from the

device (Figure 17), such as;

 the mainboard (including CPU and RAM) and other PCBs, directly forwarded to

the copper smelter;

 HDDs and ODDs, to be forwarded to a medium shredder for further separation

of iron, aluminium, magnets and circuit board fractions.

The rest of the notebook’s body goes then to a medium shredder for further separation of

fractions (Gabriel, 2015; Vannieuwenhuyse, 2016).

(

) Currently large electrolyte capacitors containing substances of concern have to be separately removed, as

described in Section 3.1. However, newer notebooks do not contain such capacitors, unless integrated into

PCBs.

(

) The final treatment aims to produce secondary raw materials, reuse appliances and components, and to

treat fractions by incineration and dispose of them e.g. at landfill sites.

(

) E.g. http://www.mrtsystem.com/products/flat-panel-processor/

(

) In line with requirements of WEEE Directive, display panels greater than 100 square centimetres are

removed for depollution at the recycling plant, with the removal of the mercury-containing CCFL

backlighting (when present). However, as it is assumed that more-recent devices feature LED backlighting,

the mercury-containing fractions are not considered here.

Figure 16 — Mechanical crushing, separation and sorting (Scenario 1). Cu = copper. Al =

Aluminum.

Figure 17 — Manual medium-depth recycling scenario for notebooks (Scenario 2). Cu =

copper. Al = Aluminum.

Dismantling/removal

of battery

Dismantling and

sorting of display

unit

Technologies/markets

under development

Mechanical crushing,

separation and sorting

Battery

Display

unit

Further manual

dismantling or

mechanical crushing,

separation and sorting

LCD panel

Remaining

parts

Other metal smelters

Steel mill traditional

Cu smelter

traditional

Cu smelter special

Plastic Recycling

Production of other

products of/with

plastics

Al smelter

Waste incineration

Battery sorting/

separation/recycling

Further

separation

, sorting

and

conditioning

Landfill

Dismantling

Technologies/markets

under development

Medium shredder

Battery

Display unit

Further manual

dismantling or

mechanical separation

and sorting

LCD panel

Remaining

parts

Other metal smelters

Steel mill traditional

Cu smelter

traditional

Cu smelter special

Plastic Recycling

Production of other

products of/with

plastics

Al smelter

Waste incineration

Battery sorting/

separation/recycling

Landfill

PCBs

Medium shredder

Drives

Further

separation

, sorting

and

conditioning

3.1.4 Recycling/recovery of tablets

Currently, the number of discarded tablet computers reaching recycling facilities is still

extremely limited; therefore EoL scenarios for these products can be only tentatively

assessed. A first evaluation has been based on information provided by three recyclers of

WEEE in Germany (ELPRO, Braunschweig; Adamec Recycling GmbH, Fürth and EGR

Elektro-Geräte Recycling, Herten) (Schischke et al., 2014). Further interviews with some

recyclers were made in 2016, but no additional conclusions can be drawn.

After battery removal, three possible pre-processing approaches are identified, depending

on the facility and taking into account economic considerations.

— Scenario 1: shredding of the whole device via cross-flow shredder.

— Scenario 2: deep-level manual dismantling of the subassemblies (such as aluminium

or plastic housing, mainboard, LCD, magnesium frame if present), using predominantly

screw drivers (battery powered and hydraulic).

— Scenario 3: direct treatment in copper smelter after removal of the battery.

The representativeness of the second scenario is limited, since the likelihood that the

labour cost for manual dismantling is not covered by the value of material disassembled

for recycling is very high.

Such treatment chains are similar to the ones described in scenarios 1 and 2 for the

notebook section.

Based on interviews and consultations with recycling operators about the deep-level

manual dismantling scenario, the following materials and components were identified as

potentially relevant:

Plastics

In general, plastics can be separated according to their colour: white (including light grey),

black, and mixed colours. White plastics have a significantly higher value compared to

black plastics. Black plastics contain carbon black, which complicates the proper

identification and subsequent separation.

Aluminium

Aluminium housing is of high interest for material recycling and it can justify a slightly

increased disassembly effort. Magnets (or other metal parts such as copper) attached to

the aluminium housing can reduce the recovery value via mechanical processing.

Magnesium

Magnesium frames are found in most of the dismantled tablets with plastic back-covers.

Currently, magnesium frames are not dismantled into separate fractions, but are rather

processed together with the aluminium fraction. For high-quality magnesium recycling, it

is necessary to achieve a high purity magnesium fraction, which is difficult via mechanical

separation due to the similar physical properties of Al and Mg.

Figure 18 — Magnesium frames in tablets (Schischke et al., 2014)

Magnesium was also identified as a CRM for the EU and it could potentially become an

interesting material for recyclers and justify manual-disassembly costs, taking into

account increased flows of EoL tablets in the future.

Display panels

Display panels contain minuscule quantities of indium and REE as well as gold in minor

amounts, which is used for interconnects and connectors of light-emitting diodes (LEDs)

and for controlling ICs. However, recycling systems are not yet adjusted to recover them

efficiently (Nissen et al., 2013). Evidence from interviews with recyclers indicates that

currently only a limited number of tablets reach recycling facilities. Therefore, recyclers’

statements are based on their experience with other electronic devices containing display

panels. The time needed to separate the display panel from the rest of the device is critical.

According to the recyclers, the display panel would be separated manually under the

condition that it is easily accessed and removed. If the front glass is not fused to the rest

of the LCD unit, it would be separated. However, as tablets do not contain mercury-

containing backlights, separation of the display panels has a lower priority compared to

e.g. the pre-processing of display panels from older notebooks and electronic displays.

PCB

Tablet mainboards are considered high-grade. After tablet opening, the PCB can be easily

removed and sorted. No removal of electromagnetic interference (EMI) shields from PCBs

is provided for as the amount of material is not worth the effort.

3.1.5 Focus on recycling/recovery of electronic PCBs

Due to the weight fraction of PCBs in notebooks (around 15 %), their recyclability has a

strong influence on the overall recyclability rate for computers. It is therefore crucial to

estimate the detail of the recyclability of elements (element groups) contained in PCBs

because of the relevance of precious and trace metals from an economic, environmental

and/or criticality perspective (Chancerel and Marwede, 2016).

A recent study by Chancerel and Marwede (2016) analysed the recyclability of PCBs,

following the compiling of an exhaustive chemical composition based on several sources,

and identifying the specific recycling rates per elements. Table 20 provides detail for the

main recycled materials. It is noticed that 22-25 % of the mass of the PCB is recycled (and

over 90 % of the material value), while around 60 % is recovered as other material, 16 %

recovered as energy and 10-30 ppm is disposed of (Chancerel and Marwede, 2016).

Table 20 — Recycling rate of materials in PCB of notebook properly treated (Chancerel and

Marwede, 2016)

Material in PCB

Recycling rate [ %]

95 %

80 %

95 %

90 %

80 %

95 %

75 %

50 %

80 %

3.1.6 Focus on recycling/recovery of batteries contained in the product

group

After collection, batteries are usually sorted according to their chemistries (lead acid,

alkaline, NiCd, NiMH, Li-ion, etc.) before being conducted to recycling treatments (Accurec,

2016). The sorting of batteries is currently mostly done manually. Those who do the

sorting attempt to identify the battery chemistry primarily via the labels on

packaging/casing of the batteries. However, in practice labels are sometimes missing,

making identification and sorting difficult. According to the interviews carried out with the

German battery-recycling company Accurec, dismantling centres remove EoL batteries

from the WEEE stream, nevertheless during removal, batteries are regularly damaged or

the cells are removed from the battery pack. Because of the absence of a label at a cell

level, cell batteries are classified as not identifiable and are sent to dedicated landfills,

thus they are lost for appropriate recycling. This is due to a lack of requirements for battery

labelling.

Currently, there are three different marks required by law worldwide which aim to highlight

the presence of Ni, Cd or Pb (Figure 19). No labelling is mandatory to comprehensively

identify the battery chemistry in Europe. However, according to Accurec, incorrect sorting

of Pb or NiCd batteries with LIB complicates the recycling processes and potentially poses

risks for the workers and for the environment. For example, in the case of missorting of

NiCd batteries into LIB, the toxic Cd metal can be released in the off-gas because the

treatment of LIB does not intend to treat Cd (Accurec, 2016). So, to avoid environmental

pollution, a more expensive off-gas cleaning system must be applied.

Europe

Japan, US

Taiwan

Figure 19 — Current legislative battery marks (source: BAJ)

In practice, manufacturers usually apply battery marks according to their chemistries,

however not in a consistent manner (see examples in Figure 20).

Figure 20 — Examples of battery marks in current practice (unreleased data from Slates

D4R)

In addition, the different chemistries of LIB (LCO, NMC, LFP, etc.) are currently not

indicated on battery packs or cells, leading to economic and material losses. Depending

on the Li-ion chemistry, the content of cobalt varies from 0 to 15 % by weight. However,

usually all Li-ion battery subtypes are co-processed, making the subsequent separation

and extraction of metals more difficult and expensive. For example, in the extraction

process of cobalt from high cobalt concentrates (LCO-type LIB), the iron and phosphorous

from the mixed processing of LFP batteries become polluting elements and need to be

removed. Such removal increases the cost of the process. Therefore, a batch-wise

treatment allows for better concentration of the target metals than a diluted mixture and

is more feasible from both a technical and economical point of view. In order to implement

more precise sorting and dedicated treatment of batteries according to their sub-

chemistry, a more-detailed indication on battery packs as well as at the cell level would

be in need (Accurec, 2016).

In Japan, rechargeable batteries are often marked with the ‘battery-recycle mark’ (Figure

21), developed by the BAJ (

). Currently, battery manufacturers in Japan are required to

display the Moebius loop mark on batteries (Figure 19) under the law for promotion on

effective utilisation of resources (

). However, as there is currently no international

standard for battery marking available, the BAJ promotes the internationally standardised

use of one battery mark as shown in Figure 21. The logos identify four different types of

battery chemistries by colour and abbreviation: NiCd, Ni-MH, Li-ion and Pb.

Figure 21 — Battery-recycle mark, developed by the BAJ and promoted to be used as an

international standard, which indicates the four different battery types by colour and in

text.

The BAJ points out the advantages of the use of an internationally standardised battery-

recycle mark as follows.

— Meeting all the various marking standards globally on each battery can be

challenging for manufacturers from a space and design perspective.

— The production of different labels, separate production runs for each destination,

and separate inventories increases costs for manufacturers.

— The use of various marks on each battery leads to lower recognition by users,

impeding efforts to raise the collection rate.

The BAJ believes that the international use of the battery-recycle mark will increase

recognition, hence contributing to improved portable-battery recycling globally, as well as

saving costs for battery manufacturers.

Figure 22 — The two-digit code, developed and recommended for use by BAJ, which is

added to the logo for LIB to identify: the metal with the highest mass in the positive

electrode (first digit); and the presence of a metal which hinders recycling (second digit).

Battery recyclers have requested that the marking should include an additional mark to

identify Li-ion batteries containing over a certain amount of tin and phosphorous. Following

the request, the BAJ recommends the industry also add a two-digit code to the logo for

LIB to specify with the first digit the metal (such as Co, Mn, Ni, or Fe) predominantly found

(by mass) in the cathode, and whether tin or phosphorous exceeding a specified threshold

are contained in the battery (Figure 22). The battery-recycle mark is currently applied to

battery packs by several manufacturers. Examples are shown in Figure 23.

(

) http://www.baj.or.jp/e/

(

) Global Environment Centre Foundation: Law for promotion on effective utilization of resources, 2016,

http://nett21.gec.jp/Ecotowns/

Figure 23 — Application of the battery-recycle mark on a notebook battery (left) and a

tablet battery (right) (sources: newbatteryshop.com and ifixit.com).

Similar to the processes in Japan, battery recyclers have requested that the IEC develop

standard for battery marking to improve the recognition of battery chemistry. The reasons

are provided as follows:

Many recycling processes are chemistry specific, undesired events may occur when a

battery which is not of the appropriate chemistry enters a given recycling process.

Therefore, in order to ensure safe handling during sorting and recycling processes, it is

necessary to mark the battery so as to identify its chemistry. (IEC 62902 draft, 2017).

A draft standard was circulated March-June 2017, which specified the appearance, colour

and size of the marking (Figure 24). The Mobius loop is to be included if not yet placed on

the battery in a different position. It should be noted that the scope of the draft standard

is currently limited to batteries with a volume of more than 900 cm

and hence does not

apply to portable batteries for ICT devices. An expansion of the scope to also include

portable batteries, as well as a process to create a European standard, is conceivable.

Figure 24 — Battery markings developed by IEC as published in the draft standard

circulated in March 2017.

In interviews with several actors of the LIB-value chain, the benefits and potential

problems of a uniform battery-marking system were investigated. While the approach was

welcomed by some actors, the interviews pointed out contradictory views, in particular

with regard to the information content and colour of such a marking. It was stated that in

order to identify and verify the most relevant marking parameters, additional

standardisation work has to be carried out. The interviewees recommended the

consideration of the following aspects.

— The marking should lead to a positive value for the recyclers and the environment

(by increasing the recycling rate) and be able to adapt towards the changing

material composition of batteries due to novel battery technologies.

— For safety reasons, the marking should contain additional information, e.g.

indicating the type of electrolyte used, together with an indication of whether or

not it is flammable.

— Requirements need to be uniform on a global scale. For instance, the Mobius loop

symbol indicates that a product is recyclable, however in the US there is no

dedicated recycling of LIB in place.

— Having a colourful marking could complicate or mislead battery sorting, e.g.

according to the US NFPA 704 Hazard Identification System (

), the blue colour

indicates the level of health hazard. However, this hurdle seems to have been

overcome in the efforts by the IEC mentioned above, as battery-recycling facilities

strongly supported the colour coding.

— The marking should indicate the process compatibility (e.g. indicate the presence

or absence of phosphorous).

— The letter size indicating the battery type needs to be standardised.

Concern was raised regarding the indication of the cobalt content in batteries. While this

could lead to improved, batch-wise treatment of batteries according to their cobalt

content, there is a risk of ‘cherry picking’ of LIB with high cobalt content, potentially

leading to a lower economic incentive to recycle remaining batteries with little to no cobalt

content, as cobalt typically remains the dominant economic driver for LIB recycling.

Additional remarks

— A certification system for battery recycling is needed ensuring that recycling on a

global level follows the same standards and efficiency.

— In the future, technologies for LIB recycling have to adapt to the changing material

composition. In the longer term, recyclers will have to develop recycling

technologies according to the LIB subchemistries. In this respect, marking of LIB

subchemistries seems useful.

The current main treatment processes for the recycling of batteries include thermal pre-

treatment, mechanical treatment, pyrometallurgy and hydrometallurgy (Accurec, 2015).

Table 21 summarizes a list of LIB recycling plants and the applied recycling processes.

(

) http://www.nfpa.org/codes-and-standards/all-codes-and-standards/list-of-codes-and-

standards?mode=code&code=704

Table 21 — Li-ion battery-recycling plants processes

Company

Recovered

Elements

Process

Country

Glencore

Ni, Co, Cu

Thermal pre-

treatment + pyrometallurgy

+ hydrometallurgy

Canada

Umicore

Ni, Co, Cu, Fe

Pyrometallurgy + hydrometall

urgy

Belgium

Accurec

Ni, Co, Cu

Thermal pre-

treatment + pyrometallurgy

+ hydrometallurgy

Germany

Kyoei Seiko

Ni, Co, Fe

Pyrometallurgy

Japan

JX Nippon

Ni, Co, Cu, Fe, Mn, Li

Thermal pre-

treatment + mechanical

treatment + hydrometallurgy

Japan

Dowa

Ni, Co, Cu

Thermal pre-

treatment + pyrometallurgy

+ hydrometallurgy

Japan

GEM

Ni, Co, Cu, Fe, Mn, Li

Mechanical

treatment + hydrometallurgy

China

Brunp

Ni, Co, Cu, Fe, Mn

Mechanical

treatment + hydrometallurgy

China

Telerecycle

Ni, Co, Cu, Fe, Li

Mechanical

treatment + hydrometallurgy

China

Kobar

Ni, Co, Cu, Fe

Mechanical

treatment + hydrometallurgy

Korea

Recupyl

Ni, Co, Cu, Fe, Li

mechanical

treatment + hydrometallurgy

France

Retriev

Ni, Co, Cu, Fe

Aqueous + mechanical

treatment + hydrometallurgy

United States

SNAM

Ni, Co, Cu

Thermal pre-treatment

France

AkkuSer

Ni, Co, Cu, Fe

Mechanical treatment

Finland

EDI

Ni, Co, Cu

Mechanical treatment

France

Batrec

Ni, Co, Cu

Mechanical treatment

Switzerland

3.1.7 Future recycling scenario for notebooks

As the design of notebooks changes, the treatment processes can change in the near

future. The compact, highly integrated design of (sub-)notebooks and desktop computers

with integrated displays, the progressively miniaturisation of devices (as for mini desktop

computers), the increased use of SSDs instead of HDDs and the reduction of the content

of precious metals in PCBs (i.e. the declining of the economic gain potentially achievable

from the recycling) (Bangs et al., 2016), together with the currently low commodity prices,

will probably make it technically and economically less feasible to go for a medium-deep

manual dismantling of valuable components. Therefore, operators could focus more and

more on full shredding processes in which, after the removal of the battery, the whole

device (including the display panel, when present) goes to a shredder. This trend for future

EoL scenarios would, however, lead to less-pure separated fractions and higher losses of

valuable materials (Vannieuwenhuyse, 2016).

Currently, the WEEE directive requests that ‘liquid-crystal displays (together with their

casing where appropriate) of a surface greater than 100 square centimetres and all those

back-lighted with gas discharge lamps’ have to be removed and treated selectively

(European Union, 2012). The separate treatment of the display panels with LED

backlighting (or OLED displays (

)) does not necessarily require separate treatment for

depollution. However, treatment operators might still separate the LCD from the rest in

the future because liquid crystals contaminate the plastic fraction after shredding and

sorting (Vannieuwenhuyse, 2016).

3.2 Analysis of repair/reuse practices

Even if manufacturers claim to design products to minimise the need for repair, by means

of the selection of high-quality materials and components, as well as a durable, reliable

structural design (Digitaleurope, 2017a), it was observed that failures in computers occur

quite commonly. Digitaleurope (2014) stated that, every year, about 118 000 t of IT

equipment and spare parts are worldwide shipped for original equipment manufacturer

repair and remanufacturing, of which roughly 28 000 t in Europe. Nearly 59 % of the

shipments take place under warranty, and about 6 % of the products shipped for repair

turn out to be unrepairable.

A Eurobarometer survey observed that, when a main failure occurs, 77 % of EU citizens

would rather repair their goods than buy new ones, but ultimately have to replace or

discard them because they are discouraged by the cost of repairs and the level of service

provided (European Commission, 2014b).

The following sections summarise information about current practices in terms of the

repair and reuse of personal computers, notebooks and tablets in particular. Main failures,

practices in design, user preferences regarding repair or upgrade and recommendations

for ecodesign are provided.

3.2.1 Reuse and repair of notebooks

Main failures

A recent IDC study among 800 United States organisations showed that the average

annual failure rate for notebooks is 18 % (average of company notebooks requiring repair

of some kind, during a year). The rate of failure increases each year a device is in use,

ranging from 11 % failing the first year to more than 20 % failing by year five. Moreover,

by the end of year five, 61 % of notebooks had had a failure that required repair (IDC,

2016).

The IDC also reported the components most often damaged in notebooks, such as the

screen, followed by the keyboard, then the data-storage drive (HDD or SSD) and the

battery (Figure 25). Among the top ways end-users damage devices in their company, the

overwhelming top reason across categories was simply dropping the device while carrying

it. The number 2 issue was spilling liquid on the device, and the number 3 issue was the

device falling off a desk. On average, workers lost about 5.8 working hours for notebook

repairs (Figure 26).

(

) Organic light-emitting diode.

Figure 25 — Most common components in notebooks that suffered damage or breakage

(IDC, 2016).

Figure 26 — Types of accidents causing notebook, tablet and handheld device damage,

according to the IDC (2016).

Interviews with repair-and-reuse operators of professional business notebooks confirm

these findings. According to them, the main frequent failures in notebooks involve:

displays, keyboards, hard drives, batteries, EPS, memories, fans, connectors (USB,

network) and optical plastic elements such as small covers and outer frames (Private

communications, 2017).

These outcomes are also confirmed by a previous IDC study (2010) which shows that

nearly 20 % of notebooks have to be repaired due to a physical failure (14.2 %) or due to

damage from accidents (9.5 %) every year (IDC, 2010). In that survey the majority of

respondents with damaged notebooks reported that the notebook had suffered a damaged

keyboard, followed by notebooks which had suffered damage to the display screen. Non-

exposed parts which are the most prone to damage include batteries and HDDs, both cited

by over half of the respondents (IDC, 2010). The greatest source of damage was human

error (Figure 27). When respondents were asked how their notebooks broke, the majority

responded that the devices were dropped while being carried, followed by two other

reasons, liquid spilled onto the devices, and the device fell off a desk or a table (IDC,

2010).

Figure 27 — Types of accidents causing notebook damage, according to the IDC (2010).

Other sources are available in literature. The insurance company SquareTrade, for

instance, analysed failure rates for over 30 000 new notebook computers covered by

notebook-warranty plans in 2009. Looking at the first 3 years of ownership, 31 % of

notebook owners reported a failure to SquareTrade. Two thirds of these failures (20.4 %)

came from hardware malfunctions, and one third (10.6 %) was reported as accidental

damage (SquareTrade, 2009).

Repair-and-reuse operators report additional reasons for failures (Private communications,

2017).

— Handling through transport and storage.

— Wear and tear.

— Poor design or not enough testing creates inherent defects: palm rests cracking,

hinges breaking, LCD screen pressure marks due to key caps sitting too proud on

the keyboard etc.

Apart from this, repair-and-reuse operators identified other issues which hamper

refurbishment.

— BIOS passwords or registration systems which hinder access to the device (costs

to reset BIOS Passwords are EUR 30-120 which means that this is usually not

economically viable or BIOS password (PWD) reset is not at all possible). Other

systems such as the iCloud (

) or the device enrolment programme from Apple

cannot be reset without the original PWD from the user. Therefore devices cannot

be repaired if there was no factory reset by the user.

— BIOS settings hinder the replacement of components which are not the originals or

the installation of a newer operating system. Repair-and-reuse operators argue

that there is no technical reason not to allow for replacement with other parts.

— There are no standardised mechanical connections to fix hard disks, drives and

other components to the chassis.

— There is no standardised layout of keyboards. Because keyboards are not available,

repair-and-reuse operators imprint blank keys. A standardised layout of keyboards

would facilitate (cross-European) sales of devices.

In a non-representative survey about the failure probability of components in consumer

and business notebooks, four out of four repair services state that batteries and HDDs fail

(very) frequently (Prakash et al., 2016a). The display, display-cover (including frame

joints) and the casing of consumer notebooks break frequently according to three repair

service companies for consumer notebooks, whereas for business notebooks this does not

seem to be the case (Prakash et al., 2016 a; Private communications, 2016). Consumer

products typically look more fashionable but are built to less-robust specifications than

commercial-focused products. Conversely, commercial-focused products have traditionally

been more staid in design, but they are built to endure slightly more robust use (IDC,

2016).

If the mainboard fails repair does not pay off as the cost for a new mainboard usually

exceeds the price of a new notebook (or the residual value of the device). Similarly, the

repair of tablets (of any component) for resale usually does not pay off, as the residual

value of the tablet is too low. Such a repair reportedly makes sense only for high value

brands (Krüger, 2016; Private communications, 2017).

The criteria to decide whether a device is to be refurbished is a quick cost-benefit analysis,

i.e. the potential resale value versus the time (and cost) invested to refurbish the device.

The cost-benefit analysis is usually done through a quick outer inspection and defect

analysis of the device. The effort (time) to refurbish the devices is estimated based on

experience.

The following components are frequently replaced in business devices by repair-and-reuse

operators:

— batteries

— memories

— HDDs

— ODDs

— fan and cooling fins,

— keyboards and keys

— displays

(

) One reuse operator mentioned, that 50 % of iPhones and iPads are discarded because of this issue, although

they are fully functional.

— plastic covers.

Business-device-repair-and-reuse operators report that about one in five batteries in the

devices are replaced. The experience is that the used batteries have 40-80 % of their

original capacity left after 3 years’ run time. Batteries are usually tested for run time

(30 minutes as a benchmark, 60 % of original runtime). If the benchmark is not achieved,

batteries are replaced. Other repair-and-reuse operators report the use of software tools

(to access data on battery SoH stored by the battery-management system) to determine

the remaining capacity of notebooks batteries. It is not the experience of repair-and-reuse

operators that batteries outlast the lifetime of the device (Private communications, 2017).

Replacement parts come from the ‘plundering’ of other devices (stored for spare parts) or

are bought from third parties, B-brands or other repair-and-reuse operators. The

availability of affordable original spare parts (from original manufacturers) is cited as a

problem. Especially small (plastic) parts, wear-and-tear parts, and parts of the chassis are

not available on the market and have to be extracted from other devices. In the future,

3D printing might help to solve this problem. Grade A and A+ batteries are sold to original

manufacturers, which re-sell them for three to five times the original price. Therefore,

repair-and-reuse operators tend to buy grade A- to B batteries from battery

manufacturers. EPS units are available and are usually used across different generations.

Mass-storage-unit (HDD and SSD) data need to be erased to facilitate reuse. Commercial

data-erasure software is usually used to perform this process. Repair-and-reuse operators

stated that this process does not always work satisfactorily. In these cases, the mass-

storage units need to be physically destroyed in order to safeguard the client-data-security

needs (Private communications, 2017).

The following information would facilitate repair and help the repair-and-reuse operators.

— Open access to data sheets with lists of components.

— Information about the ease of battery replacement (made available for

consumers).

— Information about type of tools.

— Exploded diagrams.

The repair-and-reuse operators interviewed have divergent opinions on whether this

information and spare parts should be available for the end-consumer. Some opt for the

‘right to repair’. Others prefer a registration process for authorised repair-and-reuse

operators (as long as this is not bound to the requirement to sell new devices) to ensure

the quality and safety of repair (Private communications, 2017).

Currently, business devices are still easy to repair. However, consumer notebooks are not

refurbished due to their design and low price. No current issues with snap fits in business

notebooks were identified. However, there is no standardisation of the type of screws used

and every manufacturer is using different screws.

According to repair services and repair-and-reuse operators, the following changes in the

design of new notebook products (e.g. sub-notebooks and Ultrabooks™ (

)) compared to

‘regular’ notebooks could limit the lifetime of the devices, i.e. limit the ability to repair or

upgrade the devices (Bölling, 2016; Prakash et al., 2016a) (Private communications,

2017):

— built-in batteries;

(

) Subnotebooks are a very thin and light version of a traditional notebook. They are generally less than

18-21 mm thick and 1.8 kg (Electronics Takeback Coalition 2012). Most types use SSD instead of HDD.

Subnotebooks use low-power processors and feature fast boot times which return the device from standby

mode in a few seconds. They use prismatic battery packs lasting from 5 to 11 hours. ODD and Ethernet

ports are generally omitted due to their limited size. The Ultrabook™ was created as a specification and

trademarked brand by Intel for a class of high-end subnotebooks designed to feature reduced bulk without

compromising battery life [Wikipedia, https://en.wikipedia.org/wiki/Ultrabook accessed on 16.03.2017].

— soldered-in memory (RAM);

— built-in mass storage;

— built-in (wireless) network interface card (NIC);

— small connectors.

Waterproofing of devices is seen as rather critical for repair-and-reuse operators because

the housing of devices is often glued to achieve this effect. Usually, business devices

already have protecting design features for water spillage that work well (Private

communications, 2017).

In some cases, neither HDD/SSD nor RAM are exchangeable against new components in

sub-notebooks (Ultrabooks™); either Ultrabooks™ are secured with special screws or the

RAM, and sometimes SSD flash storage, is soldered directly onto the mainboard (

). The

slimmer and lighter the devices become, the more integrated they are and thus the harder

it is to replace components (Private communications, 2016). It is expected that mass

storage will not fail as frequently in the future, as HDDs are substituted by SSDs, which

have no moving parts (Private communications, 2016). However, there is evidence that

SSDs also degrade over time (Private communications, 2017).

From the industry point of view, the benefits of integrating components are; a more rigid

and uniform design of their devices, slimmer form factor and lower manufacturing costs.

For users, this means that components cannot be easily exchanged or upgraded by

themselves or by professional repair services (for reasonable costs). This may lead a

certain share of users to invest their money in a new device.

User preferences

Regarding end-users, the main inconveniences caused by a defective device are the loss

of productivity and the loss data (IDC, 2016). Defects and failures are also the main

motivations for buying a new notebook (Prakash et al., 2016b). A German-based internet

survey assessed the main reasons for replacing a notebook after first use. It differentiated

three main reasons:

 the old device was defective;

 the old device was malfunctioning or unreliable;

 the old device was still functioning, but the user wanted a better one.

Figure 28 shows that the percentage of notebooks being replaced because they were

defective or malfunctioning increased over time whereas the percentage of still-functioning

notebooks being replaced with a better one decreased (Prakash et al., 2016b).

(

) https://www.heise.de/newsticker/meldung/Oeko-Logo-EPEAT-winkt-Ultrabooks-durch-1729666.html;

http://www.com-magazin.de/praxis/hardware/20-fakten-zu-ultrabooks-7388.html

Figure 28 — Reasons for replacing a notebook after first use (Prakash et al., 2016b)

In a survey by Forsa (2013) in Germany half of the respondents judged it to be important

that old computers can be upgraded with components with higher energy efficiency or

with higher performance. In the same survey 61 % of the people interviewed stated that

they would continue to use a notebook or tablet with a built-in battery even if the battery

was to break or lose capacity as long as they could bring it to an electronic shop and have

the battery replaced there directly on-site. Just 10 % of the survey participants would

send it to a repair service to exchange the battery and then continue to use the device

(Forsa, 2013).

Finally, Prakash et al. (2016b) surveyed the influence on the purchase decision of the

information provided by the original manufacturer (i.e. the availability of spare parts,

repair services, exchangeable parts and their lifetimes). Information about the lifetime

was rated to be important or very important by 45 % of the interviewees. Consumers

would even be willing to pay an additional price for a higher-quality product in the sense

of extending the product life. If no lifetime-related product information is available, the

majority of the interviewees chose the cheapest model (Prakash et al., 2016b). The survey

shows also that the information on the exchangeability of the battery leads to a clear shift

to buy a more expensive notebook, but also the ability to replace/repair the HDDs,

graphics processing unit (GPU) or mainboard has an effect on the buying decision (i.e.

shifting the decision from buying a cheaper notebook without exchangeable parts to

buying a more expensive notebooks with exchangeable parts). The survey finally shows

that the participants would prefer to buy a notebook with an exchangeable battery over a

notebook with built-in batteries.

Recommendations to improve reparability

Talens Peiró et al. (2016) discussed some possible criteria on reparability for computers:

‘computers should be designed such a way that key components (such as HDD/SSD,

memory, screen assembly and LCD backlight, keyboard and track pad, and cooling fan)

where used, are easily accessible and exchangeable by the use of universal tools’ (

)

(Talens Peiró et al., 2016). Suggestions provided by Dodd et al. (2014b) for Ecolabel

requirements were aligned to the proposals by Talens Peiró et al. (2016b) as: ‘All major

repairable/replaceable components of computers, if applicable, such as hard drive,

CD/DVD and Blue-ray drive, printed circuit board, memory, screen assembly, LCD

(

) Defined by the study as: widely used commercially available tools such as screwdrivers, spatulas, pliers, or

tweezers (Talens Peiró et al., 2016).

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2004 2007 2012/2013

Functioning but replaced with better one Defect Malfunctioning, unreliable

backlight, keyboard, track pad, rechargeable battery, cooling fan, catches and hinges shall

be easily accessible and exchangeable by the use of universal tools (i.e. widely used

commercially available tools)’. In order to prove compliance with the criteria above,

manufacturers should provide clear disassembly and repair instructions (e.g. hard or

electronic copy, video) and make them publicly available, to enable a non-destructive

disassembly of products for the purpose of replacing key components or parts for upgrades

or repairs (Talens Peiró et al., 2016). Additionally, a diagram showing the location of the

abovementioned components and how these can be accessed and replaced can be made

available in preinstalled user instructions and via the manufacturer website.

An additional suggestion could regard the distinction between components that can be

repaired and those which need to be replaced (e.g. for upgrade) (Dodd et al., 2016, 2015);

moreover repairs that can be carried out by consumers should be identified and clearly

distinguished by those that necessitate professional work for safety reasons and in respect

of warranty conditions (Dodd et al., 2016, 2015).

Initiatives

The Empowering Repair Co.Project (a collaboration between eBay, HP, and iFixit) aims to

populate a portal with product information to enable more efficient disassembly and

recycling of IT products (Ellen MacArthur Foundation, 2016). Other platforms are already

available with guidelines on how to repair EEE or reviews of products and scores

concerning their ease of repair. A method based on reparability scores has been developed

(e.g. by iFixit) based on the difficulty of opening the device, the types of fasteners found

inside and the complexity involved in replacing major components. Points are awarded for

upgradability, the use of non-proprietary tools for servicing, and component modularity

(Greenpeace, 2017).

Some manufacturers, such as HP, are committed to helping end-users extend the useful

lifespan of products, by freely sharing service manuals and providing a wide range of

service options and product warranties that enable people to repair their devices and

maintain product quality (Ellen MacArthur Foundation, 2016). From a study conducted by

Greenpeace (2017), 3 out of 17 brands assessed make the provision of spare parts and

repair manuals easy to access.

Availability of spare parts is also crucial to allowing the reparability of computers.

According to the Empowering Repair Co.Project, manufacturers should make spare parts

available to all interested parties for a period after manufacturing that reflects the potential

product life and for a price that reasonably reflects the part-production cost.

Summary

Surveys in Austria and Switzerland showed that the first-use time (products in use, time

until a replacement is bought) of notebooks is significantly shorter than the desired lifetime

(the time consumers desire the product to be functioning) (Thiébaud-Müller et al., 2017;

Wieser and Tröger, 2016). The most important reason for buying a new notebook is due

to defects or malfunctions — even more important than technical innovations or a lack of

performance. The most frequent failures in notebooks are keyboards, displays, batteries

and data storage. If those can be repaired or replaced easily many people would continue

to use the same notebook. Increased reparability has the potential to bring more

notebooks into a second or third life.

3.2.2 Reuse and repair of tablets

Main failures

According to repair inquiries on a German website (

) which compares repair prices and

services (repeated inquiries possible) visitors mainly inquire about repairs for tablet-

display defects. Display defects can include the following: an unresponsive touch screen,

a black screen, pixel errors, broken glass or touchscreen (not in order of number reported).

Other failures depend on the specific device (e.g. home button repair for some Apple

tablets or repair of subscriber identification module (SIM) card reader for some Samsung

tablets. Complementary findings can be retrieved from Knack (2016): the main common

quality issues for tablets are: broken pixels (dots or lines on the display), dust or other

dirt behind the screens, dust inside the camera, unresponsive touch screen and

overheating batteries. Samsung for example recalled 2.5 million devices after reports of

batteries exploding while charging (The New York Times, 2016).

Finally, as reported by the IDC (2016), the average annual failure rate observed in US

organisations for tablets is 15.7 %. The most damaged component was the screen,

followed by ports or connectors, the outer chassis and then the battery. On average,

workers lost about 4.2 working hours for tablet repairs (Figure 29).

Figure 29 — Most common components in tablets that suffered damage or breakage (IDC,

2016).

Design approaches and reparability

A non-destructive step-by-step disassembly analysis performed by Fraunhofer IZM

showed that among 21 analysed tablet computers, huge differences in the design

approaches of the various manufacturers can be found. These differences lead to a

significant variety of process steps required for opening the device, dealing with the types

of connections used, and the removal of the main components (such as the battery,

mainboard and display panel) as well as subassemblies (Nissen et al., 2013; Schischke et

al., 2014, 2013).

Tablets feature four main mechanisms to fasten the device, components and

subassemblies. The choice of fastener has significant influence on the time and difficulty

of the disassembly, component removal and reassembly, in particular those listed below.

(

) https://www.handyreparaturvergleich.de

— Screws are a good option in terms of opening, removal of parts and reversibility.

However, the number of screws and the variety of screws (different sizes and

heads) influence the disassembly time. Evidence from the disassembly trials show

that axially accessible screws are typically easier to remove than the one radially

accessible.

— Clips are considered a good option. However, the strength and the accessibility

influence the easy-and-damage-free opening of the device.

— Connectors are used for the electrical connections, e.g. to connect the display or

the battery to the mainboard. Connectors should be detachable as sometimes

disconnecting small connectors can be delicate work and lead to damage.

— Adhesives are a suboptimal solution for fastening the device housing as well as for

the main parts, in particular to attach the battery.

With respect to device opening, tablets feature three main principles — clips, screws, and

adhesives. Evidence from the interviews with two repair companies (iFixit, San Luis

Obispo, California and w-Support, Hartmannsdorf, Germany) indicates that screws are the

preferred option in terms of damage-free opening. If clips are used, the reversibility

depends mainly on their construction, robustness and their ability to disengage

simultaneously. The use of adhesives is suboptimal as it requires heating tools, which

could cause damage to temperature-sensitive components, e.g. the battery and some ICs.

In order to open the device without damage, multiple covers need to be separated such

as camera or speaker covers. In most cases device opening starts from the backside, thus

to reach the components on the opposite side — mainboard, battery and the display (last),

the disassembly has to proceed through the whole device (Nissen et al., 2013).

Two main design principles are used to fasten the battery. In the first case, the battery is

placed in a metal or plastic tray, which is attached (with an average of four screws) to the

device. In the second case, the battery is directly glued into the device. The disassembly

analysis of 21 tablets has shown that it is only after the tablet opening that the battery

can be located. In 17 out of 21 tablets a combination of screws and adhesives were used

to fix the battery in place. In 3 out of 21 tablets the battery wires were even soldered onto

the mainboard. In order to remove the battery, multiple disassembly steps are required

(between 3 and 10) (Nissen et al., 2013; Schischke et al., 2013).

With respect to repair, the glued option is suboptimal in terms of reversibility and safety

as it requires a careful approach in order to not damage the battery cells. Access to the

battery without the need to remove the mainboard is advantageous in terms of repair; it

increases the reversibility and speeds up the process of battery replacement. Batteries

with a connector cable to the mainboard are easier to replace than those with soldered

wires (Nissen et al., 2013).

With regard to mainboard removal, the disassembly study indicates that the use of

connectors allows for non-destructive removal of the mainboard. Easy access to

connectors (on the upper side of the boards) and screws (not hidden under tapes, access

from above) facilitate repair practice.

Moreover, tablets are particularly prone to being dropped, which makes the display a

particularly sensitive part. According to iFixit, breakage of the display is one of the most

frequent reasons for tablet repair (Schischke et al., 2014). An interesting finding is that

the damage can affect only the front glass and not the whole unit. Therefore, non-fusion

of the front glass with the display panel is a precondition for repair.

Therefore, easy access, dismantling and replacement of the display is of particular

relevance. The considerable amount of steps required to access the display complicates

the non-destructive part removal. In addition, working through the entire device increases

the risk of damaging other device subassemblies.

Figure 30 — Failed attempted to separate the front glass from the display panel (Schischke

et al., 2014)

A study of the literature formulated some general recommendations to facilitate the repair

of tablets, based on the analysis of product disassembly and interviews with iFixit and w-

Support (Schischke et al., 2014).

— Easy-to-open and reversible closing mechanism, optimal via several screws. Clips

might be used under the condition that they are robust and easy to disengage.

— A modular design, allowing for an easy-and-damage-free removal, as well as

substitution of subassemblies, especially the ones that are prone to accidental

damage. In general, all broken parts could be repaired under the condition that

they are easily disassembled from each other.

— Colour-coded screws and labelled cables inside the device.

— Non-fusion of front glass with the LCD unit.

— Absence of proprietary screws or fasteners.

— Application of zero-insertion-force (ZIF) connectors for the connection of the

battery and display with the mainboard; ribbon cables are also a possible

alternative.

— Mainboard fixing to the housing via three to a maximum of seven screws.

iFixit also reveals that, due to absence of available information about tablet opening,

usually the very first repair trial of a device just launched on the market causes

unnecessary damage, degrading the product’s value (Schischke et al., 2014). Thus,

information concerning tablet opening and repair is considered highly relevant (e.g. as the

repair Standard IEEE 18741 or oManual

– an open XML-based standard for semantic,

multimedia-rich procedural manuals). In addition, repair makes no sense if spare parts

are not made available from the manufacturers.

The same study identified that the following design features in tablets are suboptimal for

repair (Schischke et al., 2014).

— Attachment of numerous subcomponents to a damage-prone part: in that case, all

of the subcomponents have to be removed before the replacement of the respective

part.

— Adhering of housing, battery, mainboard or display.

See section 6.2.1 for further details.

— Multi-adhering of components, for instance when front glass, backlights and

digitisers are fused together. Breaking any of these parts will require the

replacement of the entire display panel.

 
76 
 
4  Discussion  and  identification  of  hot  spots  and  of 
improvement opportunities 
Sections 2 and 3 aimed to introduce the technical background and current situation of the 
product group, concerning, in particular, market data, BoMs and also an analysis of current 
and potential future practices concerning recycling and repair/reuse. 
During this analysis, several hot spots, i.e. aspects of the product group that are relevant 
from a material-efficiency perspective, were identified. Hot spots consist of, for example, 
problems potentially encountered by EoL operators during the treatment of computers, or 
design features of the products that facilitate (or hamper) disassembly or repair. Identified 
hot spots are summarised in Table 22 for each product sub-group and organised among 
three main material-efficiency aspects: recyclability, repairability/reusability and material 
savings. 
Some  hot  spots  (for  example  the  unknown  content  of  indium  in  notebook  and  tablet 
displays) or potential obstacles for the dismantling of integrated desktop computers, were 
identified during the analysis of similar product groups (e.g. ‘electronic display’). With this 
purpose a draft regulation for ecodesign implementing measures for electronic displays is 
currently  in  preparation  and  it  addresses  some  the  aspects  identified  in  Table  22. 
Therefore, these aspects have been not elaborated further in the present study. Aspects 
detailed  in  Table  22  are  judged  as  relevant  and  these  will  be  further  explored  in  the 
following sections. The table also clarifies which aspects are relevant for the product sub-
groups. 
The following three sections are as follows. 
—  Proposals of actions to improve waste prevention (Section 5) 
—  Proposals of actions to enhance repair/reuse (Section 6) 
—  Proposals of actions to enhance recyclability (Section 7) 
Each proposal is supported by a short introduction on the state of play and motivations 
(Rationale),  and  is  then  discussed  in  terms  of  the  feasibility  of the  action,  focusing  in 
particular on the availability of standardised procedures or on the need for standardisation 
work. Some of the authors observed that the absence of appropriate metrics and standards 
has been a key barrier to material efficiency, and that specific standardisation needs can 
be systematically identified, developing adequate metrics for performance measurements, 
reliable and repeatable tests, and calculation procedures (Tecchio et al., 2017). 
Finally an initial assessment of benefits/impacts for specific actions is presented. The initial 
assessment has been developed considering the potential material savings and additional 
flows of recycled materials obtainable thanks to material-efficiency measures that aim to 
facilitate dematerialisation and a circular economy. We recall that material efficiency does 
not directly regard resources used to produce energy, nor energy used during the lifecycle 
of products (Tecchio et al, 2017). 
Raw materials are crucial to Europe’s economy and essential to maintaining and improving 
our quality of life. Securing reliable and unhindered access to certain raw materials is a 
growing concern within the EU and across the globe. Examples of critical raw materials 
include  REE,  indium  and  cobalt  (Directorate-General  for  Internal  Market,  Industry, 
Entrepreneurship and SMEs, 2016). 
 

Table 22 — Summary of ‘hot spots’ of computer sub-product groups (as identified during the analysis of Sections 2 and 3)

Material-efficiency aspects and

sub-aspects

Identified hot spots

(divided by sub-product groups)

Non-mobile personal computers

Mobile personal computers

Desktop computers and

Integrated desktop computers

Notebooks and

Tablets

Waste

prevention

Lifetime of product

and components.

Limited lifetime of batteries.

Failure problems observed on computer components (e.g.

display panel, keyboard, data storage, battery, ports or

connectors, outer chassis).

Prevention of

material

consumption.

Presence of unnecessary external power supplies (EPS) in new

computers sold.

Repairability/reusability

Need for detailed information about the disassembly of certain

components, such as batteries, memories, data storage, display

panels.

Lack of information crucial for repair

Privacy issues due to data content (and needs for secure data deletion).

Recyclability

Ability to remove

components

containing

valuable and

hazardous

materials

For integrated desktop: Difficulties for the

dismantling of the display panel.

For compact desktop (mini desktops): possible

dismantling problems for the batteries (when

present).

Difficulties for the dismantling of certain components, such as

batteries, display panels, PCBs in various parts.

Recyclability of

plastics

High diversity of plastic with additives (including flame retardants (FRs)) in the whole product group makes difficult to

sort and recycle polymer at a sufficient quality.

Recyclability of

batteries

Identification and composition of batteries is not always clear by

recyclers hence creating inefficiency in the treatments and

material losses.

CRM content

Additional information on the content and location of certain CRMs could incentivise research and investments of new

technologies and process for their recycling.

4.1 EU Ecolabel and EU green public procurement criteria

Reports by Dodd et al. (2016, 2015) and Talens Peiró et al. (2016) were consulted to

understand which options were suggested for the revision of the EU Ecolabel and the EU

green public procurement (GPP) criteria of the product group. Talens Peiró et al. (2016),

in particular, developed specific criteria for the reparability and design for dismantling of

computers and electronic displays. Criteria for the EU Ecolabel were released officially

through the Commission Decision (EU) 2016/1371 of 10 August 2016, establishing the

ecological criteria for the award of the EU Ecolabel for personal, notebook and tablet

computers (European Commission, 2016).

This further analysis was done to understand which options could potentially be adopted

to address the hot spots of Table 22, even though mandatory requirements related to the

Ecodesign directive typically take into consideration verification methods which are

different from the ones applied for voluntary schemes.

A summary of the interested criteria of the Commission Decision (EU) 2016/1371 is given

hereinafter. These criteria, as well as findings from Dodd et al. (2016, 2015) and Talens

Peiró et al. (2016) were used to develop possible improvements under the ecodesign

directive, in Sections 5, 6 and 7.

Lifetime extension: durability testing of personal computers

Notebook computers must pass certain durability tests for the award of the EU Ecolabel.

Each device must be verified to function as specified and to meet the stipulated

performance requirements after performing some mandatory tests (specified in

IEC 60068, e.g. resistance to shock tests, resistance to vibration, drop tests) and

additional durability tests (temperature stress, screen resilience, water-spill ingress, etc.).

Lifetime extension: upgradability and reparability

Key components of personal computers (memories, mass-storage systems, screen

assembly and LCD backlight, keyboards, track pads and cooling fans) should be easily

accessible and exchangeable with the use of universal tools. Furthermore, rechargeable

batteries in mobile computers should be manually extractable, with no need of tools, and

information on how to separate the battery pack should be marked on the base cover of

the chassis. Exceptions were proposed for tablets, for which the extraction of the battery

would be allowed with a maximum of four steps (three for sub-notebooks and

Ultrabooks™) and the use of commercially available tools. The applicant is also asked to

provide clear disassembly and repair instructions.

End-of-life management: plastic components

Plastic components with a mass greater than 25 g for tablets and 100 g for all other

computers should be properly marked, including the presence of FRs. The polymer

composition can be identified by means of ISO 11469 and ISO 1043 markings.

End-of-life management: design for disassembly

Furthermore, key components (PCBs, internal power supplies, HDDs, batteries, etc.)

should be easily extracted from the product. A disassembly test must be carried out

according to the test procedure detailed in the Appendix of the Commission Decision (EU)

2016/1371. The test must record the number of steps required and the associated tools

and actions required to extract the key components.

5 Possible actions to improve waste prevention

According to the waste hierarchy set out by the European Commission (Directive

2008/98/EC, (European Union, 2008)), waste prevention has the first priority, before

preparing for reuse or recycling. Among the strategies to close (and slow) material loops,

design for reliability and durability and design for standardisation and compatibility are

key aspects (Bocken et al., 2016). As such, opportunities to eliminate or postpone factors

potentially leading to premature obsolescence of personal computers have a high priority

in legislation implementation.

5.1 Battery durability

5.1.1 Rationale

End-users want durable goods, such as notebooks, to last considerably longer than they

are currently used (Wieser and Tröger, 2016). Prakash et al. (2016b) surveyed the

influence of information provided by original manufacturers about the availability of spare

parts, repair services, exchangeable parts and lifetime on the purchase decision.

Information about the lifetime was rated as important or very important by 45 % of the

interviewees. Battery durability is considered by users to be a key feature: in a survey

conducted by the IDC (2010), 68 % of respondents confirmed that the battery lifetime on

their notebook computer was not sufficient for their business needs. Respondents also

indicated that 22 % of notebook computers required the purchase of a replacement

battery during its lifetime.

Therefore, increased battery durability becomes important considering the current trend

towards more-integrated devices, leading manufacturers to integrate batteries within

devices and abandoning the previously widespread slide-lock removal mechanisms.

Examples for this development were provided in Section 3.2 of this report. Manufacturers

design integrated batteries to improve the robustness of the whole device and to make

devices thinner, however end-users may face potential difficulties in replacing an

exhausted battery by themselves. Hence, battery durability is a more meaningful factor

than ever.

Information about battery durability

LIB inevitably lose a fraction of their full-charge capacity with every charge/discharge cycle

they go through (see Section 2.2.2). It has been shown that the capacity of some batteries

fades quicker than others (Clemm et al., 2016). To guarantee a minimum level of durability

and hence prevent premature waste generation, battery-cycle tests may be used to

determine the number of charging cycles a battery can withstand before its capacity fades

to a certain threshold.

Current legislation requires manufacturers of notebooks to provide data on the expected

cycle life of batteries in notebooks (Commission Regulation (EU) No 617/2013) (

). In a

non-exhaustive survey of the websites of notebook manufacturers it was found that only

two manufacturers provided such information (Apple (

) and HP (

)), only one of which

refers to specific notebook models. Furthermore, without a set of complementing

information on the methodology applied to determine the minimum number of charging

(

) 7.1 From 1 July 2014

‘7.1.1 Manufacturers shall provide in the technical documentation and make publicly available on free-

access websites the following information:

[…] (o) the minimum number of loading cycles that the batteries can withstand (applies only to notebook

computers)’ (Commission Regulation (EU) No 617/2013).

(

) https://support.apple.com/en-us/HT201585

(

) http://support.hp.com/us-en/document/c00596784

cycles, data cannot be considered meaningful. At the very least the following information

is required to put the number of charging cycles provided by the manufacturer in context:

— the definition of a charging cycle;

— the capacity threshold at which the battery is considered exhausted;

— the measurement methodology (e.g. a testing standard).

A cycle is defined as ‘an amount of discharge approximately equal to the value of design

capacity’ (SBS-IF, 1998), with design capacity referring to the theoretical capacity of a

new battery (pack) (also: ‘rated capacity’ during a 5-hour discharge, as declared by the

manufacturer). In practice, a charging cycle is often defined as discharging (possibly in

several partial-discharge events) and consequently recharging it to 100 % (e.g. Apple).

Information on the methodology and capacity threshold would allow for transparency as

well as a certain degree of comparability between the different cycle numbers

manufacturers provide for their devices. Ideally, a standardised methodology would be

stipulated to allow for greater transparency and comparability.

Battery durability in stationary use

A common use pattern for notebooks is stationary use, in particular in office environments.

Stationary use means non-mobile use, e.g. on a desk, and in grid operation, i.e. directly

plugged into a power outlet or using a docking station. As the battery is constantly

connected to the grid, the battery SoC is permanently close to 100 %. High SoC is known

to accelerate the ageing of LIB (Section 2.2.2). A study on the lifetime of notebook

batteries in the field found that 50 % of the notebook batteries in the offices of companies

or public administrations were cycled up to 30 times per year. Despite the low charging

frequency, a large share of the batteries had lost significant portions of their initial capacity

(Clemm et al., 2016). This is partly attributed to the high SoC during notebook use in grid

operation as well as other factors, such as increased temperatures when working in grid

operation and using a docking station in particular, among other factors. In conclusion,

the user should have the means to increase the durability of device batteries by preventing

a constantly high SoC when using their notebook in grid operation.

It is technically feasible to limit the SoC to which a notebooks battery is charged when

plugged into a power outlet via software tools. Several of the large notebook

manufacturers ship their devices with such software preinstalled (

). One of the features

of this software is the option to use battery-optimising modes. A software button (on/off

switch) allows the user to enable and disable a mode in which the battery is charged up

to a pre-defined or user-defined state of charge, commonly in the range of 50-70 % SoC.

Thus, a high SoC is prevented while using the notebook in grid operation, potentially

increasing battery durability at relatively low cost to the manufacturer.

When battery-optimising mode is not enabled, the software tool of one manufacturer will

recommend the user (via a pop-up message) enable battery-conservation mode, if the

device is used in grid operation (and 100 % SoC) for a predefined period (e.g. 2 hours).

The user can switch off battery-conservation mode and fully charge the battery if needed,

e.g. before using the device in mobile, battery-powered mode. The disabling of battery-

conservation mode can further be triggered at a certain time as defined by the user (e.g.

with a timer coupled to a calendar application). Battery-conservation mode is further

recommended when the device will not be used for a period of time, to decrease calendar

ageing of the battery.

(

) Examples are the Lenovo battery conservation mode, Dell battery meter, Sony battery care.

5.1.2 Possible improvements

Information about battery-cycle life

A minimum level of battery durability could be potentially established when information

about battery-cycle life are made available to end-users (e.g. in the user documentation),

with common testing rules. In a durability test for batteries, two main parameters consist

in the number of charge/discharge cycles and the remaining full-charge capacity compared

to the initial-charge capacity (SoH). So, the two possible ways to identify battery durability

are as follows.

─ Declaring the number of charging cycles device batteries can withstand before the

capacity fades to a set threshold.

─ Declaring the SoH of the battery (the remaining full-charge capacity compared to

the initial-charge capacity) after a predefined number of charging cycles.

The second option seems more practical, as the first option would disadvantage products

with higher durability, as more charging cycles are needed to reach the desired threshold.

The availability of information on battery-cycle life would help end-users to get an

indication on how long the battery in a specific device may last. This piece of information

may potentially be complemented with the manufacturing date of the battery. Moreover,

such a declaration on the cycle stability of the battery allows the comparability among

products of different manufacturers, and potentially pushing the market towards higher-

quality battery cells.

Therefore, batteries could be tested in accordance with the most recent version of the

Standard EN 61960 (

), and results could be communicated as the remaining full-charge

capacity of the battery compared to the initial-charge capacity, after a predefined number

of charge/discharge cycles (e.g. 300 and/or 500 cycles).

Standard EN 61960 defines secondary (

) (rechargeable) batteries (battery packs) and

cells as follows.

— Secondary lithium battery: ‘unit which incorporates one or more secondary lithium

cells and which is ready for use. It incorporates adequate housing and a terminal

arrangement and may have electronic-control devices’.

— Secondary lithium cell: ‘secondary single cell whose electrical energy is derived

from the oxidation and the reduction of lithium. It is not ready for use in an

application because it is not yet fitted in its final housing, terminal arrangement

and electronic-control device’.

According to Section 2.2.2, a remaining charge capacity of 80 % of the initial charge is

typically reached between 300-500 charge/discharge cycles, for consumer products

(Battery University, 2016a). Taking into consideration this evidence and the technological

progress (declarations of batteries that can be considered consumed after 1 000 cycles

are available (

)), it is reasonable to consider 300 and 500 charge/discharge cycles for

the declaration of the remaining charge capacity according to EN 61960 tests, and the

remaining charge capacity after 500 cycles as a possible parameter to be directly

communicated to users. Voluntary declarations about the SoH after a higher number of

cycles (e.g. 750 and/or 1 000 cycles) can be encouraged as well.

Battery manufacturers have a number of possible tests to evaluate battery-cycle life

following the Standard EN 61960. The test on battery life can be applied either at the

battery-cell level or at battery-pack level. Furthermore, non-accelerated or accelerated-

(

) IEC 61960:2011 Secondary cells and batteries containing alkaline or other non-acid electrolytes —

Secondary lithium cells and batteries for portable applications.

(

) IEC definition of a primary cell: ‘cell which is not designed to be electrically recharged’; IEC definition of a

secondary cell: ‘cell which is designed to be electrically recharged’ by way of a reversible chemical reaction

(IEC, 2017).

(

) https://support.apple.com/en-us/HT201585

test procedures are available. Specifically, Section 7.6.3. ‘Endurance in cycles at a rate of

0.5 I

A (accelerated-test procedure)’ is pointed out in order to reduce the burden of the

test requirements (as compared to the regular test procedure set out in Section 7.6.2.);

however, with this approach, batteries are subject to overstressed conditions and capacity

may fade quicker.

Tests conducted at the battery-pack level are closer to reality, considering that notebook

batteries are often composed of four or more cells. However, manufacturers may use the

same battery cells in different pack combinations, so testing a specific cell would give a

good indication of how all packs incorporating that cell behave. It is therefore

recommended to refer to the test for cells rather than for battery packs since single-cell

design may be used in multiple battery-pack designs.

Using the accelerated-test procedure, and assuming that battery charging takes 3.5 hours,

the test procedure for 500 cycles is estimated to result in the following time investments:

— charging: 3.5 hours

— idle time: 0.5 hours

— discharge: 2 hours

— time investment per cycle (sum): 6 hours

— time investment for 500 cycles: 125 days.

However, the non-accelerated-testing procedure can more realistically reproduce use

patterns of notebook and tablet computers, as the prescribed discharge rate of 0.2 C

(discharge within 5 hours) is much closer to the power consumption of such devices

compared to the discharge rate of 0.5 C in the accelerated-testing procedure.

Furthermore, private communications with manufacturers confirmed that non-

accelerated-testing procedures are commonly applied to batteries at the manufacturing

plant.

Under the assumption that battery charging takes 3.5 hours, the test procedure for 500

cycles is estimated to result in the following time investments:

— charging: 3.5 hours

— idle time: 0.5 hours

— discharge: 5 hours

— time investment per cycle (sum): 9 hours

— time investment for 500 cycles: 188 days.

It can be assumed that cell testing of cycle life takes place at the cell manufacturer rather

than the device manufacturer. It can further be assumed that cell manufacturers test their

cells before mass production, in part to provide specifications to their customers. Hence,

it can be assumed that certain testing data on cell cycle life and the applied methodology

is already available to the cell manufacturer and the additional burdens of a legislative

requirement in this context would be limited.

The information about battery durability, to be provided in the user documentation, can

be complemented by the following features of the battery:

— design capacity;

— voltage;

— date of manufacture;

— the capacity threshold at which the battery is considered exhausted;

— the definition of charge/discharge cycle and the measurement methodology used

for testing;

— explanation on how ambient temperature and battery SoC can impact the battery

lifetime;

Coulomb: it is the electric charge transported by a constant current of one ampere in one second.

— battery manufacturer.

Finally, this possible improvement was also based on and compared to what was proposed

for the revision of the EU Ecolabel criteria and EU GPP criteria for personal computers. The

authors of the background studies proposed that points be awarded for durable batteries.

Points were supposed to be achieved at different thresholds (including the here-proposed

threshold of 500 charge/discharge cycles), depending on when the battery reaches 80 %

of the initial full-charge capacity. For verification, the applicant must provide a test report

for the battery cells or packs showing compliance according to the EN 61960 (Dodd et al.,

2016). For the scope of the Ecodesign directive, instead, such a test report could be done

by a third-party laboratory to verify the compliance with what was declared by the

manufacturers.

Future developments may also consider establishing efficiency classes for the SoH of

batteries, using an ‘A-G grading system’, or similar. The collection of more extensive data

about batteries in the market (e.g. through a dedicated database) could help in building

such a grading system.

Battery lifetime optimisation

The durability of notebook batteries could be further improved by implementing a

preinstalled functionality which prevents battery-cell capacity to fade because of being

kept for a long time at a high SoC (see Section 2.2.2: between 90 and 100 %). This may

occur when the device is used stationary (i.e. in grid operation).

Manufacturers can develop in-house solutions (e.g. functionalities, or battery-charging

algorithms) for this problem, even if the literature review highlighted that a practical option

could be to limit the SoC of the battery to a specified value (e.g. 70 % or less compared

to the available full-charge capacity) whenever the device is used stationary (i.e. in grid

operation). In any case, the effectiveness of such a lifetime optimisation function can be

guaranteed if 1) manufacturers take action to inform the users of its existence and the

benefits, 2) end-users can have the option to temporarily disable the limit on SoC (

), but

cannot have the option of setting limits on the SoC that potentially reduce the battery

lifetime, decided by manufacturers.

Such a function could also be used to provide information about the battery features to

end-users, such as:

— current SoC;

— current SoH (as the current full-charge capacity compared to the design capacity);

— number of charge/discharge cycles the battery has already gone through;

— battery temperature;

— battery chemistry;

— other features of the battery (see the list of complementary information to be

provided in the user documentation for battery durability).

5.1.3 Initial assessments of benefits/impacts (battery lifetime

optimisation)

Increased battery durability potentially increases the time the battery can be used in a

notebook (or tablet) before losing as much capacity as to be considered as having reached

its EoL. Thus, either the replacement of the device battery is delayed, or even the disposal

of the entire device is delayed. This may prevent the waste-and-environmental impacts

associated with recycling as well as the manufacturing of a replacement battery.

(

) For a limited amount of time (e.g. until the next restart of the computer).

 
84 
 
As can be derived from Figure 3, the lower the SoC of a battery when in storage (without 
cycling), the higher its durability will be. Depending on the SoC, the capacity of the tested 
cells fades to 80 % full-charge capacity after varying times: 
—  SoC of 95 % after less than 300 days, 
—  SoC of 80 % after around 300 days, 
—  SoC of 70 % after around 400 days, 
—  SoC of 50 % after more than 500 days. 
The testing by Schmalstieg et al. (2014) has been carried out under conditions of elevated 
temperature (50 °C) to speed up the observable effects of calendar ageing at varying SoC. 
However, by deriving a factor which describes how much capacity fade is prevented by 
capping the SoC,  the actual number of  days in  the  accelerated  test  can  be ignored to 
establish the following simplified scenarios. 
Scenario A: A notebook computer is permanently used in grid operation. Assuming that 
a software to limit the SoC in grid operation limits the SoC, the battery durability may be 
increased as follows. 
—  SoC limit at 70 % may increase battery durability by factor [400 days/300 days =] 
1.34. 
—  SoC limit at 50 % may increase battery durability by factor [500 days/300 days =] 
1.67. 
Scenario B: A notebook computer is used 75 % of the time in grid operation and 25 % 
mobile on battery power. The effect of SoC limit would only account for the share of grid 
operation. Hence, if the SoC is capped at 70 % during grid operation, the durability is 
increased by factor 1.26 (calculation below). If SoC is capped at 50 %, the factor is 1.50. 
—  SoC limit at 70 % may increase battery durability by factor [(0.75 * 1.34) + (0.25 
* 1) =] 1.26 
—  SoC limit at 50 % may increase battery durability by factor [(0.75 * 1.67) + (0.25 
* 1) =] 1.50 
Scenario C: A notebook computer is used 50 % in grid operation and 50 % mobile on 
battery power. The effect of SoC limit will only account for the share of grid operation. 
Hence, if the SoC is capped at 70 % during grid operation, the durability is increased by 
factor 1.17 (calculation below). If SoC is capped at 50 %, the factor is 1.34. 
—  SoC limit at 70 % may increase battery durability by factor [(0.5 * 1.34) + (0.5 * 
1) =] 1.17 
—  SoC limit at 50 % may increase battery durability by factor [(0.5 * 1.67) + (0.5 * 
1) =] 1.34 
Prakash et al. (2016c) investigated the effect of extending the lifetime of notebooks used 
in  public  administration  for  a  total  useful  life  of  6 years  instead  of  3 years.  In  their 
assumptions,  the  authors  estimated  that  battery  replacement  is  necessary  in  50 %  of 
notebook computers to allow such a lifetime extension. This assumption can be converted 
into a value of 1.5 batteries/mobile computer, and was here adopted to build a base-case 
scenario  in  which  the  lifetime  of  notebooks,  according  to  Table  2,  is  considered  to  be 
5 years on average. The average mass of a notebook battery was assumed to be 259.6 g 
(according to Table 14) while its average composition was derived from Table 17. 
In our assessment, we considered the projection of shipments of notebooks in 2020 (about 
41.7  million  of  products),  and  that  the  average  mass  and  the  average  composition  of 
batteries are kept constant over time. The market projection for notebooks was gathered 
from Table 1. 
We estimated the benefit of battery-optimisation software with the following parameters. 
—  Base-case scenario: no use of battery-optimisation software and need for batteries 
for each notebook computer set to 1.5 batteries. 

— Scenario B: notebooks working in grid operation 75 % of the time, SoC limits 70 %

and 50 %.

— Scenario C: notebooks working in grid operation 50 % of the time, SoC limits 70 %

and 50 %.

Scenario A (notebooks working in grid operation 100 % of the time) was not considered

realistic. With these assumptions, it was possible to estimate the material saving

achievable thanks to the adoption of scenarios B and C instead of the base-case scenario,

where no battery-conservation software is used. Results are presented in Table 23. Table

23 provides the total mass of materials that can potentially be saved, as well as specific

savings of cobalt (Co), lithium (Li), nickel (Ni) and copper (Cu), compared to the base-

case scenario.

Table 23 — Material savings (batteries in million units/year, materials in t/year)

achievable when a battery-optimisation software is implemented in notebooks.

Future scenarios

Scenario B

Scenario C

cap 70 %

cap 50 %

cap 70 %

cap 50 %

Notebook batteries (million

units/year)

12.9

20.8

9.1

15.9

Cobalt, Co (t/year)

281

454

198

346

Lithium, Li (t/year)

Nickel, Ni (t/year)

131

211

161

Copper, Cu (t/year)

452

730

318

556

Other (t/year)

2 424

3 915

1 707

2 980

Total (t/year)

3 347

5 407

2 357

4 116

The yearly rate of material saving (

) estimable with these hypotheses ranges between

2 357 t and 5 407 t. In the best conditions (Scenario B and SoC limit set to 50 %), about

454 t of cobalt, 97 t of lithium, 211 t of nickel and 730 t of copper can be saved every

year.

A SoC limit set to 50 % would bring the highest material savings, however would also limit

the autonomy of the notebook battery once it is not used in grid operation. A SoC limit set

to 70 %, instead, would allow a higher autonomy in case the notebook battery is not used

in grid operation, but implying reduced material savings. Manufacturers could actively

develop more developed in-house functionalities to prevent the battery to remain at full

charge when the notebook is in grid operation, and to enable one or more limits on the

battery state of charge (SoC) when the notebooks is in grid operation.

5.1.4 Other potential benefits (information about battery-cycle life)

The assessment of the potential benefits related to the communication of the remaining

full-charge capacity of the battery compared to the initial-charge capacity, after a

predefined number of charge/discharge cycles, is characterised by higher uncertainty, for

both notebooks and tablets. As reported in Section 2.2.2, a battery is considered

exhausted when its capacity reaches 80 % of original capacity. However, batteries can

continue to be used even below 80 % capacity, although the runtime of the device will be

decreased. This can be reached after 1 000 charge/discharge cycles, for high-quality

products, but typically it is reached between 300-500 charge/discharge cycles. It is

(

) Scenarios do not consider the amount of materials lost through refining, processing and manufacturing.

assumed that declaring the performance of the battery would boost the competition among

battery manufacturers, resulting in more durable products.

An average value of 1.5 batteries/mobile computer (notebooks and tablets) was again

assumed as base case, and two hypothetical scenarios in which the value is decreased to

1.4 and 1.3 batteries/mobile computers. With these hypotheses, and considering

notebooks and tablets sold in 2020, 2025 and 2030, as in Table 1, the replacement of 4.1-

8.1 million batteries/year for notebooks and 3.8-7.5 million batteries/year for tablets can

potentially be avoided. The total saving of materials can be estimated in the range 1 560-

3 050 t/year. Nonetheless, we highlight that further work is needed to estimate the

potential improvement in battery durability.

5.2 Decoupling external power supplies from personal computers

5.2.1 Rationale

EPS represent a very significant percentage of the whole weight and materials used for

ICT (estimated to be in the range 10-20 %), thus it is important to set specifications for

materials and EoL compatibility, in order to minimise their impact on the environment

(ITU-T L.1002, 2016).

The rationale for this section is to promote the reuse of EPS by means of the following.

— The adoption of interoperable and common EPS (as described in Section 2.5),

which makes the service life of an EPS independent from the product’s useful life.

— The progressive decoupling of products and EPS, which intends to promote the

reuse of EPS already available by the end-users.

Material savings can be achieved thanks to the reduced production and delivery of new

EPS and the consequent reduction in electronic waste. As reported in Section 2.3.4, an

EPS supplies power to notebooks, tablets, and other devices that do not contain internal

components to derive the required voltage and power directly from the grid. The function

carried out by EPS is to transfer power to the device by converting voltage and current

characteristics from the grid to the desired load levels. Therefore, reuse of EPS is possible,

when the power output and other main features (e.g. interfaces, connectors, energy

efficiency, no-load power or safety) are compatible with multiple ICT devices.

An example of EPS reusability thanks to compatible specifications is represented by the

mobile-phone sector. A memorandum of understanding (MoU) was recently signed

between the European Commission and 14 electronics manufacturers. The agreement is

for the harmonisation for all EPS for data-enabled telephones and hence enables the

reusability of the EPS. As reported by Cucchietti et al. (2011), a common EPS would bring

benefits to manufacturers, vendors and customers; the latter category, in particular, would

be able to share just one charger for more than one device. Manufacturers and vendors

would be able to ship and sell their devices without the charger in the package, with

potential material savings due to the reduced use of materials and impacts for transport

and distribution (about 90 % of EPS are manufactured in Asia (Risk & Policy Analysts

Limited, 2014)) and the box containing a new mobile phone can be around 25 % lighter

when an EPS is not included.

Back to the computer product group, as little as 10 years ago, it was observed that efficient

EPS are getting smaller, lighter in weight, and more convenient to store and transport (Bio

Intelligence Service, 2007). PCBs used in EPS were characterised as low-grade

(< 200 ppm gold), the classification used for low mass of valuable materials (Dimitrova,

2012; Goosey and Kellner, 2002). Nowadays, efficient EPS operate at cooler temperatures,

contain fewer parts, and are likely to result in greater product reliability (Bio Intelligence

Service, 2007); it is also possible to find EPS for notebooks with a weight of 85 g, and an

output power of 65 W (FINsix®, 2016) on the market. Moreover, new EPS would not have

large transformers or capacitors (EPS based on the switching-mode technology do not

require such components), and would be characterised by smaller size and weight, thanks

to technological innovation and to more integrated and miniaturised components

(Dimitrova, 2012). The PCBs of EPS could potentially be processed by dedicated recycling

processes to optimise the recycling output, but due to the complex dismantling required

and the small quantity of valuable materials, this becomes not viable economically (Sarkis,

2001).

With these preconditions, it seems reasonable to promote the reuse of EPS for personal

computers in order to extend the lifetime and therefore to enhance material savings.

According to the Risk & Policy Analysts Limited (2014) study, the harmonisation of EPS for

portable electronic devices would affect manufacturers in different ways.

— There would not be significant costs on the manufacturers of portable electronic

devices.

— Significant impacts on competition, competitiveness, trade and investment flows

are not expected.

— Harmonisation might slow down innovation, according to some stakeholders

consulted by the authors.

— Manufacturers of chargers and cables could potentially benefit from the use of more

expensive components, but are also likely to incur revenue losses due to increased

decoupling.

5.2.2 Possible improvements

The provision of information on the EPS specifications and the presence/absence of the

EPS in the packaging of notebooks and tablets could potentially enhance the reuse of

available EPS, and hence result in a significant reduction in material consumption for the

production of unnecessary power supplies and for the treatment of electronic waste. This

information provided to end-users could promote the use of common EPS across different

devices. Material savings can potentially be achieved thanks to reductions in production,

packaging, transport and distribution.

Such information can be conveyed to end-users through the user documentation and a

logo (e.g. on the packaging). For personal computers that use an EPS the information

could include the required power-supply specifications, namely voltage, current and rated

output power.

The main goal of the logo could be to indicate the presence or absence of the EPS within

the packaging. If it is present the end-user can be informed through the user

documentation about the possibility to use the contained EPS with other devices and

compatibly with the EPS specifications. Vice versa; if the EPS is absent the user

documentation can notify the user about the possibility to use an alternative suitable EPS

which meets the device specifications. The user documentation should also inform about

the type of connector required to interface the EPS with the device.

Labelling schemes can be based on standards and recommendations. In particular, the

specifications of the common EPS should include the following.

— The recommended types of device that can be connected (e.g. notebooks, tablets).

— Input voltage type, input voltage range, frequency range and maximum input

current.

— Output voltage, current and power ranges, with efficiency of power conversion.

Standards can be used and can be further developed to illustrate the interoperability of

common EPS for use with notebooks and/or tablets.

IEC/TS 62700 (2014) (DC power supply for notebook computers), Standard

IEEE Std 1823 (2015) (Standard for universal power adapter for mobile devices) and

Recommendation ITU-T L.1002 (External universal power adapter solutions for portable

ICT devices) represent relevant sources to illustrate the common charging capability and

interface requirements for the EPS, as in the case of IEC 62684 (2011) developed for data-

enabled mobile telephones.

5.2.3 Initial assessments of benefits/impacts

The initial assessment was based on future scenarios in which electronic devices

(notebooks and tablets) and EPS are gradually decoupled, meaning that a certain

percentage of products put on the market will not include an EPS in the packaging. Future

scenarios were estimated by the authors of this report.

The same EPS may be used by different types of products which are compatible with the

power-supply specifications (e.g. notebooks and tablets, but also smartphones and other

electronic devices). In order to assess potential benefits related to the decoupling of EPS

from personal computers, two scenarios have been developed, one related to EPS in

notebooks and another for EPS in tablets. In this context, these scenarios do not take into

account that the harmonisation of power specifications could bring about the common use

of the same EPS for tablets and notebooks. These devices used to have different power

requirements (lower for tablets and higher for notebooks), and assessment concerning

these products have also been considered separately in previous research (Risk & Policy

Analysts Limited, 2014). Nowadays however, many notebooks can work with a power

requirement of less than 100 W, and the USB type-C specifications allow scalable power

up to 100 W (

). However, due to the lack of input data, it was not possible to estimate

the number of notebooks and tablets that can potentially share the same EPS. Thus, this

section does not consider the possibility of using the same common EPS for both notebooks

and tablets and assumes that the technology of the common EPS would be based on micro-

USB connectors, although notebooks and tablets could also rely on standard-USB

connectors.

Risk & Policy Analysts Limited (2014) focused on the harmonisation of EPS for portable

electronic devices and found that micro-USB was barely used for charging devices, as it

would be subject to power output limitations. Harmonisation would be more feasible given

the release of IEC/TS 62700 (2014), the Standard IEEE 1823 (IEEE, 2015) and the

Recommendation ITU-T L.1002 (2016).

Scenarios used in previous studies for mobile phones

According to the Risk & Policy Analysts Limited (2014) report, two scenarios were drawn

in order to study the product group of mobile phones, differentiating between possible

action taken by the European Commission. In the first one (namely ‘Scenario 2 %’), the

European Commission was supposed only to encourage discussions among manufacturers

of the relevant devices, with the goal of facilitating a consensus on the use of a common

EPS; as a result, 2 % of devices were supposed to be sold without a charger, based on an

extrapolation of the current decoupling trend (device and EPS) for mobile phones. In the

second one (namely ‘Scenario 50 %’), the European Commission was supposed to propose

legislation requiring that devices use the common EPS; in this case 50 % of mobile phones

were supposed to be sold without a charger. 50 % represents the highest possible

decoupling rate, basing the estimates on the current levels of ownership of mobile phones

and expected charging behaviour of consumers.

Scenarios developed for notebooks and tablets

The assessment of potential benefits related to the proposed action was built considering

the European market projections for notebooks and tablets of Table 1, Section 2.1.

Average masses of 114 g and 440 g were considered for EPS used by tablets and

notebooks respectively, according to Risk & Policy Analysts Limited (2014). The

composition of the two power supplies was retrieved from Table 13, taking into account

the ‘60 W notebook’ for the tablet EPS composition, and the ‘90 W notebook’ for the

notebook EPS composition (Bio Intelligence Service, 2007; Dimitrova, 2012). The average

composition of electronics in EPS for tablets and notebooks has been retrieved from the

values of Table 11.

The assessment was developed on different scenarios for years 2020, 2025 and 2030.

Compared to the study of Risk & Policy Analysts Limited (2014), where only the two

‘Scenario 2 %’ and ‘Scenario 50 %’ were considered as extreme situations (the first

representative of the state of play and the second representative of the best result

(

) The USB power delivery is capable of delivering up to 100 W with the standard USB connector and up to 60

W with the micro-USB connector.

 
90 
 
obtainable thanks to direct legislative action from the European Commission), the current 
assessment also developed some intermediate scenarios. 
—  Scenario 10 % and Scenario 20 %: expected range of decoupling of products and 
EPS potentially obtainable thanks to EPS labelling. 
—  Scenario  30 %  and  Scenario  40 %:  optimistic  range  of  decoupling  obtainable 
thanks to a more stringent actions. 
For each decoupling percentage, material saving is calculated, as the total mass of EPS 
not  produced  and  not  shipped  with  the  given  volume  of  notebooks  and  tablets. 
Assessments do not take into consideration the environmental impact/benefit of transport, 
packaging, use phase and EoL. Inefficiencies occurring during the production phase (e.g. 
generated scraps) were not taken into consideration. 
Results are shown in Figure 31. Histograms were built for the 3 years considered (2020, 
2025 and 2030), representing the estimation of the amount of materials saved (annual 
reduction) thanks to the decoupling of devices (tablets and notebooks) and EPS. When 
considering  the  expected  range  of  decoupling  potentially  obtainable  thanks  to  the 
proposed labelling (Scenarios 10 % and 20 %), about 2 300-4 500 t of materials could be 
saved every year, considering the estimated sales of tablets and notebooks. We highlight 
a slight difference in values in 2020, 2025, 2030, as according to the source of data for 
market projections, shipment and sales will be stable for the two product categories over 
the considered time horizon (Viegand Maagøe and VITO, 2017). Indeed, looking at the 
chart for the year 2030, and considering Scenarios 10 % and 20 %, the estimated potential 
savings  of  materials  are  in  the  range  of  2 295-4 591 t/year  (80 %  allocated  to  the 
notebook product group, 20 % allocated to the tablet product group). This result is 4-6 
times higher than the associated reduction in the consumption of raw materials calculated 
by Risk & Policy Analysts Limited (2014) for the decoupling of mobile phones from their 
chargers, in the EU market from 2011 to 2013. 
Table  24  and  Table  25  provide  an  overview  of  the  specific  material  saving  divided  by 
material  category  (we  identified  four  main  material  categories,  according  to  the  study 
conducted by Bio Intelligence Service, namely plastics, ferrous metals, non-ferrous metals 
and electronics). Table 26 and Table 27 were built using the information from Table 11 
(Section 2.3.2)  and  the  results  of  Table  24  and  Table  25,  in  which  only  the  category 
‘electronics’ was considered. Taking into consideration these assumptions, it was possible 
to  estimate  the  specific  material  saving  due  to  the  avoided  production  of  electronics, 
highlighting  some  precious  metals  and  CRMs  (e.g.  silver,  gold,  beryllium,  cobalt, 
chromium, copper, gallium, palladium and antimony). 
 

Figure 31 — Potential material saving (t/year) in 2020, 2025 and 2030, divided by product

categories: notebooks and tablets.

367

1,833

3,666

5,499

7,332

9,165

457

913

1,370

1,827

2,284

2000

4000

6000

8000

10000

2% 10% 20% 30% 40% 50%

Material saving (tonnes/year)

Decoupling rate

Potential resource saving - 2020

Notebooks Tablets

366

1,828

3,656

5,485

7,313

9,141

458

916

1,373

1,831

2,289

2000

4000

6000

8000

10000

2% 10% 20% 30% 40% 50%

Material saving (tonnes/year)

Decoupling rate

Potential resource saving - 2025

Notebooks Tablets

367

1,837

3,673

5,510

7,346

9,183

459

918

1,377

1,835

2,294

2000

4000

6000

8000

10000

2% 10% 20% 30% 40% 50%

Material saving (tonnes/year)

Decoupling rate

Potential resource saving - 2030

Notebooks Tablets

Table 24 — Material savings [t/year] divided by category (plastics, ferrous metals, non-

ferrous metals, electronics). Notebooks.

Notebooks — Material savings divided by category [t/year]

Decoupling scenario

2 %

10 %

20 %

30 %

40 %

50 %

2020

Plastics

115

574

1 148

1 722

2 296

2 870

Ferrous metals

Non-ferrous metals

430

861

1 291

1 722

2 152

Electronics

163

815

1 629

2 444

3 258

4 073

2025

Plastics

114

572

1 145

1 717

2 290

2 862

Ferrous metals

Non-ferrous metals

429

859

1 288

1 717

2 146

Electronics

162

812

1 625

2 437

3 250

4 062

2030

Plastics

115

575

1 150

1 725

2 300

2 875

Ferrous metals

Non-ferrous metals

431

862

1 294

1 725

2 156

Electronics

163

816

1 632

2 449

3 265

4 081

Table 25 — Material savings [t/year] divided by category (plastics, ferrous metals, non-

ferrous metals, electronics). Tablets.

Tablets — Material savings divided by category [t/year]

Decoupling scenario

2 %

10 %

20 %

30 %

40 %

50 %

2020

Plastics

148

295

443

590

738

Ferrous metals

Non-ferrous metals

139

278

417

556

695

Electronics

167

333

500

667

834

2025

Plastics

148

296

444

592

740

Ferrous metals

Non-ferrous metals

139

279

418

557

696

Electronics

167

334

501

668

835

2030

Plastics

148

297

445

593

742

Ferrous metals

Non-ferrous metals

140

279

419

558

698

Electronics

167

335

502

670

837

Table 26 — Material savings divided by substance [t/year]. Notebooks (only the mass of

electronics was considered for this assessment).

Notebook electronics — Material savings divided by substance [t/year]

Decoupling scenario

2 %

10 %

20 %

30 %

40 %

50 %

2020

0.18

0.90

1.79

2.69

3.58

4.48

0.03

0.16

0.33

0.49

0.65

0.81

0.02

0.08

0.16

0.24

0.33

0.41

0.02

0.08

0.16

0.24

0.33

0.41

0.57

2.85

5.70

8.55

11.40

14.26

30.95

154.77

309.55

464.32

619.10

773.87

0.00

0.01

0.02

0.03

0.04

0.03

0.16

0.33

0.49

0.65

0.81

0.49

2.44

4.89

7.33

9.78

12.22

2025

0.18

0.89

1.79

2.68

3.57

4.47

0.03

0.16

0.32

0.49

0.65

0.81

0.02

0.08

0.16

0.24

0.32

0.41

0.02

0.08

0.16

0.24

0.32

0.41

0.57

2.84

5.69

8.53

11.37

14.22

30.87

154.37

308.73

463.10

617.46

771.83

0.00

0.01

0.02

0.03

0.04

0.03

0.16

0.32

0.49

0.65

0.81

0.49

2.44

4.87

7.31

9.75

12.19

2030

0.18

0.90

1.80

2.69

3.59

4.49

0.03

0.16

0.33

0.49

0.65

0.82

0.02

0.08

0.16

0.24

0.33

0.41

0.02

0.08

0.16

0.24

0.33

0.41

0.57

2.86

5.71

8.57

11.43

14.28

31.01

155.07

310.14

465.22

620.29

775.36

0.00

0.01

0.02

0.03

0.04

0.03

0.16

0.33

0.49

0.65

0.82

0.49

2.45

4.90

7.35

9.79

12.24

Table 27 — Material savings divided by substance [t/year]. Tablets (only the mass of

electronics was considered for this assessment).

Tablet electronics — Material savings divided by substance [t/year]

Decoupling scenario

2 %

10 %

20 %

30 %

40 %

50 %

2020

0.04

0.18

0.37

0.55

0.73

0.92

0.01

0.03

0.07

0.10

0.13

0.17

0.00

0.02

0.03

0.05

0.07

0.08

0.00

0.02

0.03

0.05

0.07

0.08

0.12

0.58

1.17

1.75

2.33

2.92

6.33

31.67

63.35

95.02

126.69

158.37

0.00

0.01

0.03

0.07

0.10

0.13

0.17

0.10

0.50

1.00

1.50

2.00

2.50

2025

0.04

0.18

0.37

0.55

0.74

0.92

0.01

0.03

0.07

0.10

0.13

0.17

0.00

0.02

0.03

0.05

0.07

0.08

0.00

0.02

0.03

0.05

0.07

0.08

0.12

0.58

1.17

1.75

2.34

2.92

6.35

31.75

63.50

95.24

126.99

158.74

0.00

0.01

0.03

0.07

0.10

0.13

0.17

0.10

0.50

1.00

1.50

2.01

2.51

2030

0.04

0.18

0.37

0.55

0.74

0.92

0.01

0.03

0.07

0.10

0.13

0.17

0.00

0.02

0.03

0.05

0.07

0.08

0.00

0.02

0.03

0.05

0.07

0.08

0.12

0.59

1.17

1.76

2.34

2.93

6.36

31.82

63.64

95.47

127.29

159.11

0.00

0.01

0.03

0.07

0.10

0.13

0.17

0.10

0.50

1.00

1.51

2.01

2.51

5.3 Durability testing for personal computers

5.3.1 Rationale

Broadly, there are often differences between traditional and commercial notebooks,

tablets, and handheld devices. Consumer products typically look more fashionable but are

built to less-robust specifications than commercial-focused products. Conversely,

commercial-focused products have traditionally been more staid in design, but they are

built to endure slightly more robust use (IDC, 2016).

As mentioned in Section 3, the two most recurring accidents for both notebooks and

tablets occur because of the following.

— The computer was dropped while being carried or fell off desk or table while in

use.

— Liquid was spilled on the computer.

Possible options to improve the durability performance of personal computers may be

related to resistance to falls (or other mechanical shocks) and resistance to water. This

section provides examples of testing methods for ruggedness and robustness of notebooks

and standardised methods to test resistance to water.

Resistance to drops/falls and shocks

Notebook manufacturers developed testing methods for ruggedness and robustness of

notebooks partially based on the US Military Standard 810 G (

). The MIL-STD-810 G test

method standard is intended to help organisations in preparing tests to evaluate how well

a particular piece of equipment can perform in the field. The standard outlines dozens of

test methods, each associated with a source of environment stress, such as vibration,

moisture, dust, extreme temperatures, or humidity. While there is no one recommended

(or required) list of tests for device categories, most major computer vendors generally

perform between 5 and 8 tests (HP, 2015).

For the durability of the screen for example, the manufacturers mainly perform the

following tests which partially go beyond or differ from the tests set in the military standard

and for which they have developed their own testing equipment and facilities (ASUS, 2016;

HP, 2015; Lenovo, 2016; Samsung, 2015): (corner) drop test; torsion (twist) test; impact

(weight drop) test; compression test; hinge-cycling test.

Manufacturers use those test routines to improve the design of notebooks and thus make

them more durable. This can include more robust components, better layouts and

improved junctions between components. Furthermore, notebook computers must pass

certain durability tests for the award of the EU Ecolabel (European Commission, 2016),

and similar requirements were also developed in the context of the EU GPP scheme, where

points are awarded for products that have passed durability tests carried out according to

IEC 60068, US MIL810G or equivalent (Dodd et al., 2016).

Another group of standards is the EN 60068 series for testing of environmental stress on

electronic components and products. Table 28 shows some of the testing procedures which

could be used to assess the robustness of computers.

(

) http://everyspec.com/MIL-STD/MIL-STD-0800-0899/MIL-STD-810G_CHG-1_50560/

Table 28 — Testing procedures of the EN 60068 series.

Standard

Title

Test description

EN 60068-

2-6

Environmental testing —

Part 2-6: Tests — Tests

Fc: Vibration (sinusoidal)

(IEC 60068-2-6:2007)

This part of IEC 60068 gives a method of test which

provides a standard procedure to determine the

ability of components, equipment and other articles,

hereinafter referred to as ‘specimens’, to withstand

specified severities of sinusoidal vibration. If an item

is to be tested unpackaged it is referred to as a test

specimen. However if the item is packaged then the

item itself is referred to as a product and the item

and its packaging together are referred to as a test

specimen.

EN 60068-

2-7

Basic environmental

testing procedures — Part

2: Tests; test Ga and

guidance: Acceleration,

steady state (IEC 60068-

2-7:1983 + A1:1986)

To prove the structural suitability and the

satisfactory performance of components, equipment

and other electrotechnical products, hereinafter

referred to as ‘specimens’, when subjected to forces

produced by steady acceleration environments

(other than gravity) such as occur in moving

vehicles, especially flying vehicles, rotating parts

and projectiles, and to provide a test of structural

integrity for certain components.

EN 60068-

2-27

Environmental testing —

Part 2-27: Tests — Test Ea

and guidance: Shock

(IEC 60068-2-27:2008)

This part of IEC 60068 provides a standard

procedure for determining the ability of a specimen

to withstand specified severities of non-repetitive or

repetitive shocks.

EN 60068-

2-31

Environmental testing —

Part 2-31: Tests — Test

Ec: Rough handling

shocks, primarily for

equipment-type specimens

(IEC 60068-2-31:2008)

This part of IEC 60068 deals with a test procedure

for simulating the effects of rough handling shocks,

primarily in equipment-type specimens, the effects

of knocks, jolts and falls, which may be received

during repair work or rough handling in operational

use.

EN 60068-

2-75

Environmental testing —

Part 2-75: Tests — Test

Eh: Hammer tests

(IEC 60068-2-75:2014);

German version EN 60068-

2-75:2014

This part -2-75 of IEC 60068 provides three

standardised and coordinated test methods for

determining the ability of a specimen to withstand

specified severities of impact.

Figure 32 — Picture of a drop test applied to a notebook (Westpak, 2013).

Resistance to water

The keyboard, the display, the display-cover (including frame joints) and the casing of

consumer notebooks are the components most prone to fail due to falls/drops or liquid

spillage.

Regarding damage due to liquid spillage, manufacturers have the possibility to adopt

waterproof solutions for certain personal computers, and test their IP according to the

Standard IEC 60529 — Degrees of protection provided by enclosures (ingress protection

(IP) code). The standard classifies and rates the degree of protection provided against,

dust, water, accidental contact, and intrusion through mechanical casings and electrical

enclosures. A device tested according to IEC 60529 is classified using an IP code, according

to the results obtained. The IP code is typically followed by two digits, indicating the solid-

particle-protection class and the liquid-IP class. As an example, an electronic device

classified as IP-22 is protected against insertion of fingers (Solid-particle protection) and

against vertically or nearly vertically dripping water (liquid IP). When no data are available

to specify one of the two protection ratings, the digit is replaced with the letter X (e.g. IP-

X2).

Thus, the second digit specifies the liquid IP that the enclosure provides against harmful

ingress of water, and ranges from 0 to 9. An overview of the types of test set out by

IEC 60529 is provided in Table 29.

An additional durability test was introduced by the decision on EU Ecolabel criteria for

notebook computers (European Commission, 2016), and is focused on water-spill ingress.

The test has the following characteristics.

— The test must be carried out twice.

— A minimum of 30 ml of liquid should be poured evenly over the keyboard of the

notebook or onto three specific, separated locations, then actively drained away

after a maximum of 5 seconds, and the computer then tested for functionality after

3 minutes.

— The test should be carried for a hot and a cold liquid.

— The notebook should remain switched on during and after the test.

The notebook must then be dismantled and visually inspected so as to ensure it passes

the IEC 60529 acceptance conditions for water ingress.

Table 29 — Examples of IEC 60529 test levels and short descriptions. The level number

specifies the second digit of the IP code (BS EN 60529:1192+A2:2013).

Level

Degree of protection

Short description of the test

None

—

Protected against vertically

falling water drops

Vertically falling drops. Test duration: 10

minutes. Water flow rate 1 mm/min.

Protected against vertically

falling water drops when

enclosure tilted up to 15°

Vertically falling drops and object tilted at an

angle of 15° from its normal position. Test

duration: 2.5 minutes for every direction of tilt

(10 minutes total).

Water flow rate 3 mm/min.

Protected against spraying

water

Water falling as a spray at any angle up to 60°

from the vertical, using either a spray nozzle

or an oscillating fixture.

Spray nozzle. Test duration: 5 minutes

minimum. Water flow rate 10 l/min.

Oscillating tube. Test duration: 10 minutes.

Water flow rate 0.07 l/min per hole.

Protected against splashing

water

Water falling as a spray at any angle up to

180° from the vertical, using either a spray

nozzle or an oscillating fixture.

Spray nozzle. Test duration: 5 minutes

minimum. Water flow rate 10 l/min.

Oscillating tube. Test duration: 10 minutes.

Water flow rate 0.07 l/min per hole.

Protected against water jets

Water projected by a nozzle (6.3 mm

diameter) against enclosure from any

direction.

Test duration: 1 minute per square metre for

at least 3 minutes.

Water flow rate 12.5 l/min.

Protected against powerful

water jets

Water projected in powerful jets (12.5 mm

nozzle diameter) against the enclosure from

any direction. Test duration: 1 minute per

square meter for at least 3 minutes. Water

flow rate 100 l/min.

Protected against the effects

of temporary immersion in

water

The enclosure is immersed in water under

specified conditions of pressure and time (up

to 1 m of submersion). Test duration: 30

minutes.

Protected against the effects

of continuous immersion in

water

The enclosure is immersed in water under

specified conditions of pressure and time

(depth specified by manufacturer). Test

duration: by agreement.

Protected against high

pressure and temperature

water jets

Water projected by a fan jet nozzle against the

enclosure from any direction. Test duration:

30 seconds in each position for a minimum of

3 minutes. Water flow rate 15 l/min. Water

temperature: 80 °C.

5.3.2 Possible improvements

The standards IEC 60529 can be used to test the liquid-IP class for personal computers.

The provision of information on the liquid-IP class for personal computers (in particular

notebooks and tablets) to end-users would inform them about the product characteristics.

The users, according to their needs, could be more conscious about the purchase and

contribute in this way to reducing the amount of personal computers repaired or discarded

because of liquid spillage yearly.

Such information can be reported through the technical documentation and conveyed to

end-users through the user documentation and through dedicated pictograms. The main

goal of the pictograms would be to indicate the level of protection against dripping water,

spraying water and water jets.

5.3.3 Future improvements: development of additional standards on

endurance testing

Although the durability of notebooks and of some of their functions is a relevant material-

efficiency aspect both for consumers and for manufacturers (who do communicate on this)

the setting of specific requirements would require some additional standardisation work.

Although some endurance-testing procedures are available, there are no generally agreed

testing parameters, performance benchmarks, critical values and routines specified for

notebooks (Ripperger, 2016). Furthermore, the number of samples of models to be tested

and the verification specifications (pass or fail criteria) have to be detailed further (Dodd

et al., 2015). Thus, more work is necessary to set requirements for the testing of

notebooks against physical stress in order to improve the robustness and ruggedness of

the devices (both notebooks and tablets).

A good start for this set of tests is the recommendation proposed by Dodd et al. (2015)

for the ‘Durability testing of notebooks for the EU Ecolabel’ (Table 30). The drop/fall test,

for instance, is described according to the MIL-STD-810 G, 516.6, Procedure IV. The final

European Commission decision (2016) establishing the ecological criteria for the award of

the EU Ecolabel for personal, notebook and tablet computers, instead, refers to the

Standard series EC 60068. Specifically, IEC 60068-2-31 describes three possible tests.

(a) Drop and topple (a simple test intended to assess the effects of knocks or jolts

likely to be received primarily by equipment-type specimens during repair work or

rough handling on a table or bench).

(b) Freefall procedure 1 (a simple test to assess the effects of falls likely to be

experienced due to rough handling. It is also suitable to demonstrate a degree of

robustness).

be received by certain component-type specimens, for example connectors in

service).

Among the mandatory durability tests for notebook computers described by the EU

Ecolabel for personal, notebook and tablet computers, it is specified that the notebook

must be dropped from a height of 76 cm onto a non-yielding surface covered with a

minimum of 30 mm of wood. One drop must be made on each of the following: top,

bottom, right, left, front and rear side, as well as each bottom corner. The notebook must

be switched off during the test and must successfully boot up following each drop. The

casing must remain integral and the screen undamaged following each test (test method

IEC 60068 Part 2-31: EC — Freefall, procedure 1). The same procedure is applied for

tablets.

In addition to the tests described in Table 30, the durability of the hinges should be tested

through hinge-cycling tests. For instance, one of the additional tests described by the

Commission decision on EU Ecolabel of notebook computers (European Commission, 2016)

100

specifies that the screen must be fully opened and then closed 20 000 times. The screen

must then be inspected for any loss of stability and hinge integrity.

Table 30 — Durability testing for notebook computers proposed by Dodd et al. (2015).

Durability

Test

Test conditions and performance

benchmark

Reference for

test method

Drop

122 cm drop height onto a 5.0 cm of plywood

surface on concrete, 4-6 drops per sample to a total

of 26 drops covering each face, edge and corner.

The notebook is non-operational during the test but

must function following the test.

MIL-STD-810 G,

516.6, Procedure IV

Shock

40 g for 18 tests each applied to Bottom, Left and

Back side.

The notebook is non-operational during the test but

must function following the test.

MIL-STD-810 G,

516.5, Procedure I

For further review of

equivalence: IEC

60068

Vibration

20-2000 Hz, 1.04 Grms*, 1 hour applied to bottom,

left and back side.

The notebook is to be operational during and after the

test.

* root mean square acceleration

MIL-STD-810 G,

514.6, Category 24

For further review of

equivalence: IEC

60068

Temperature

Three 24 hour exposure cycles for each extreme in a

test chamber – 29 °C and 63 °C.

The test to be repeated for an operational and non-

operational notebook. The notebook must be

checked that it functions following each routine.

MIL-STD-810 G,

501.5, Procedure II

For further review of

equivalence: IEC

60068

Water ingress

0.2 litres of water is to be poured evenly over the

main body of the open keyboard face of the

notebook, drained after 3 seconds, inverted on its

side for 45 seconds and then tested after 2 minutes.

The notebook is to be operational during and after

the test.

MIL-STD-810 G,

506.5, Procedure III

For further review of

equivalence: IEC

60529

Screen pressure

25 kg loading to be applied to the centre of the

screen lid with the notebook placed on a flat surface.

The screen to then be inspected for lines, spots and

cracks.

No formal reference:

stakeholder input

required.

Potential to refer to

panel pressure test

methods.

Keyboard

accelerated live

10 million random keystrokes simulation for (to be

specified) product samples. The keys to then be

inspected for their integrity.

No formal reference:

stakeholder input

required.

101

6 Possible actions to enhance repair/reuse

6.1 Disassemblability of key components for personal computers

to enhance repairability

6.1.1 Rationale

For mobile personal computers, display panels, batteries, keyboards and data storage are

the components most prone to fail or to be damaged (see Section 3.2, statistics by the

IDC (2010, 2016) and interviews with stakeholders). Furthermore, battery performance is

one of the key features for consumer choice (Dodd et al. 2016, 2015) but degrades over

time and can influence the service life of the device. Also, mass storage and memory

significantly determine the performance of both mobile and non-mobile personal

computers (i.e. the used capacity can limit the usability of the device).

Manufacturers’ designs aim to minimise the need for repair, through the selection of high-

quality materials and components, together with a durable, reliable structural design

(Digitaleurope, 2017a). However both the average annual failure rates of computers (IDC

(2016) estimated 18 % for notebooks and 15.7 % for tablets) and the reparability rates

(every year, about 6 % of the products shipped for repair or remanufacturing to

manufacturers turn out to be unrepairable (Digitaleurope, 2014)) are not negligible.

According to a 2014 Eurobarometer survey, 77 % of EU citizens would rather repair their

goods than buy new ones, but ultimately have to replace or discard them because they

are discouraged by the cost of repairs and the level of service provided (European

Commission, 2014b). For end-users, the availability of repair options to fix day-to-day

problems with the devices at reasonable costs is an important factor for a substantial

prolongation of the product lifetime (Dodd et al. 2016, 2015). However, the trend to build

and sell more-integrated devices such as sub-notebooks or tablets (see Section 2.1),

makes an easy repair or upgrade more difficult (see Section 3.2), i.e. components such as

track pads, keyboards, or network interface card (NIC) cannot be easily disassembled,

repaired/replaced and reassembled. Although a repair might be feasible, the difficulty and

the cost may lead a certain share of users to rather purchase a new device.

Overall, the ease of repair, or upgrade, becomes more and more important in order to:

— prolong the operational life of the device (by enhancing repair and refurbishing),

and

— avoid environmental impacts due to the manufacturing of a new device and the

disposal of electronic waste (by enhancing preparation for reuse).

Design for ease of maintenance and repair, design for upgradability and adaptability,

design for standardisation and compatibility, design for disassembly and reassembly are

recognised to be key strategies to improve product life extension (Bocken et al., 2016)

6.1.2 Possible improvements

According to the Ellen MacArthur Foundation (2016), products should be entirely

assembled by reversible means such as screws instead of glue, rivets or non-reversible

snap locks. The use of proprietary fasteners should be avoided. Batteries should be easily

replaceable, preferably without the use of any tools, and should not be glued or soldered

to a product. Components should not be integrated to such a degree as to make individual

replacement of functional components impossible. Finally, manufacturers should make

repair information available: as soon as a product is launched; to all interested parties,

including non-profit repair initiatives; free of charge (Ellen MacArthur Foundation, 2016).

The reversible disassembly of relevant components (such as batteries, internal power-

supply units, displays, mass-storage systems, memories, keyboard, track pad, network-

interface card (NIC), wireless local-area-network (LAN) card and cooling fan assemblies)

plays a key role in enhancing reuse of personal computers.

102

Possible actions to enhance repair and refurbishing, but also preparation for reuse, can be

listed, considering different levels.

— Professional repair operators can be provided with information about the

disassembly, replacement and reassembly operations needed for each relevant

component of computers.

— Users can be provided with clear and easily accessible information about the

disassembly and replacement of the batteries used in computers.

Such documentation could be useful not only to enhance repair and refurbishing of EEE,

but also for preparation for reuse of WEEE. It has been recognised that repair (and

upgrade) of components should not be limited only to the manufacturer’s authorised

service providers during the warranty period, but generally to professional repairers (BIO

by Deloitte, 2015; RREUSE, 2013), in order to reduce safety risks (e.g. due to improper

repairs or incorrect components). End-users or non-professionals should be allowed to

replace components, which are easy exchangeable; in any case where only official repair

services are available, this will limit competition and may not help to reduce repair costs

(Dodd et al., 2015). On the other hand, manufacturers are reluctant to disclose proprietary

information and would prefer to limit the availability of disassembly instructions for

authorised repair services only.

Key components were already identified by Talens Peiró et al. (2016) and proposed by

Dodd et al. (2016) for the revision of the EU GPP criteria of personal computers. As

proposed by these two studies, the verification of this criterion is done with a manual,

provided by the applicant, which includes an exploded diagram of the device illustrating

the parts that can be accessed and replaced.

In the context of the Ecodesign directive, however, documentation on the sequence of

disassembly, replacement and reassembly operations could be provided for key

components (highlighted by literature reviews on frequent failures, damages, and

interviews with repair operators), when present in the product.

— Notebooks (and desktop computers): batteries (including stand-by button cells on

the motherboard), internal power-supply units, display (

), data storage (HDD,

SSD and eMMC), memories, keyboards.

— Tablets: batteries and displays.

Other relevant notebook components cited by repair operators were: network-interface

cards, wireless LAN card, track pads, ports and connectors, cooling fan assemblies, audio

connectors and cameras.

Relevant information for professional repair operators can include: exploded diagrams (

)

of the product (showing the location of components); disassembly sequences; type and

number of fastening technique(s) to be unlocked; tool(s) required; warnings if delicate

disassembly operations are involved (risk of damage). Diagrams, photos or videos showing

the disassembly steps could be used to accompany and better communicate this

information. A comprehensive set of information should also include information about the

safety requirements and risks (if any) related to the disassembly, replacement and

reassembly operations. Such documentation could be available to professional repairers,

and to users (for repair operations that they can safely perform).

The Open Manual Format (oManual) could be used to make the abovementioned

information available. oManual is an open XML-based standard for semantic, multimedia-

rich procedural manuals. It can be used to store and present e.g. service manuals, ‘how

(

) Listed by Greenpeace (2017) as the most problematic component for design for repairability of mobile

devices (notebooks, tablets and mobile phones). The display is often designed in a way that makes

replacement very costly. Two-thirds of the devices (30 out of the 44) that were tested had displays that

were designed to be difficult or costly to replace (Greenpeace, 2017).

(

) Information such as exploded diagrams is especially useful for preparation for reuse and remanufacturing

purposes, where it is economical to consult such information online (EERA, 2016).

103

to’ guides, assembly instruction and user manuals (IEEE 1874, 2013). The oManual

structure is suitable for describing/documenting the steps (disassembly, dismantling) for

specific products. It provides the necessary structure to describe the steps in words and

pictures/videos. Ongoing European standardisation work could elaborate on this

standardised format and could help to more precisely specify the information to be

provided.

Nevertheless, when software (

) or firmware (

) is required by the repaired/reused device

to function, the software or firmware should be used as recommended by the original

equipment manufacturer (prEN 50614, 2016).

Future developments to assess the ease of disassembly could focus on the standardisation

of tools needed to disassemble a device (see Recchioni et al. (2016)), and on the number

and types of disassembly steps (

) needed for certain repairs (Vanegas et al., 2017,

2016). The use of the Maynard operation sequence technique (MOST) is a more elaborate

way to illustrate a step. MOST is based on fundamental activities called standard

sequences, which are a set of basic motions, which include horizontal actions over a

distance, physical move in the vertical direction, the action of gaining control, the action

of placement and the action of loosening. The application of the MOST would require the

establishing of how to describe/list each (dis)assembly step in a consistent and

comprehensive way, for example by using a standardised structure (including the

abovementioned oManual).

A ‘step’ can then be defined as a sequence of certain activities (Vanegas et al., 2016).

Vanegas et al. (2016), for example, identified six basic tasks (sequence of basic motions)

for the disassembly of a household appliance (electronic display): tool change, identifying

connectors, manipulation of the product, positioning, disconnection, and removing. For

each task, they specified a sequence of activities. For repair activities, the reverse tasks

to assemble the product also need to be specified. The method was recently updated by

Peeters et al. (2018) who enlarged the database of disassembly sequences (i.e. adding

new types of connectors, such as cable connectors, cable plugs and glues requiring

wedge/pry and peel actions to be released) and the scope of the analysis (i.e. including

the calculation of reassembly operations). Peeters et al. (2018) tested the updated method

on two case studies represented by notebook computers. The updated method allows now

to evaluate both the ease of disassembly (eDIM

) and the ease of reassembly (eDIM

)

metrics. The sum of the two metrics (eDIM

and eDIM

) estimates the overall effort needed

for disassembling and reassembling one or more components.

Upgradability of personal computers

Especially for tablets and Ultrabooks™, the upgrade of components such as the main

memory (RAM) or data storage is currently technically limited due to the high integration

and the small form factor of the devices. An extension of the mass storage for example is

in some devices feasible (e.g. through extra slots for secure digital (SD) cards), but not

for all models of computers. Technical possibilities and limits of replacement and upgrade

have to be discussed with stakeholders.

In a survey by Forsa (2013), in Germany, half of the respondents judged it to be important

that computers can be upgraded with components with higher energy efficiency or with

higher performance. In the same survey, 61 % of the people interviewed stated that they

would continue to use a notebook or tablet with a built-in battery in a case where the

battery breaks or loses capacity if they can bring it to an electronic shop and the battery

is replaced there directly on-site.

(

) An ordered set of instructions and associated data, which controls and protects the operation of equipment

(prEN 50614, 2016).

(

) A computer program or data that cannot be easily changed by the user (prEN 50614, 2016).

(

) A possible definition of disassembly ‘step’ (or disassembly task) is ‘a basic disassembly action that cannot

be further disaggregated’. A simple definition is to say that one-step finishes with the removal of a part or

a change of a tool.

104

In the context of the revision of the EU GPP criteria, for instance, Dodd et al. (2016)

proposed points that can be awarded to devices with the potential to replace and upgrade

the RAM (socketed design) and the potential to expand the storage by using slots

supporting mass-storage media.

Batteries

As mentioned at the beginning of Sections 5.1.1 and 6.1.1, battery performance

represents one of the key features for consumer choice. However, batteries degrade over

time, and replacements may be necessary to re-establish the initial performance of the

whole product. In a survey conducted by the IDC (2010), over half of respondents stated

that battery failures caused problems for their business. Respondents also indicated that

22 % of notebook computers required the purchase of a replacement battery during their

lifetime. Consumentenbond (2015) reported that 77 % of consumers were able to replace

the battery of their notebooks themselves, in 2012, while this percentage fell to 42 % in

2015 (

). Participants of a recent German survey preferred to buy a notebook with an

exchangeable battery over a notebook with a built-in battery (Prakash et al., 2016b). The

ease of access and replacement of the battery of a personal computer becomes therefore

relevant, especially when this operation has to be done by end-users or by professional

repair operators. Nevertheless, removable batteries instead of built-in batteries are also

welcomed by recyclers, in order to dismantle easier (EuRIC, 2016a). This piece of

information can be provided for end-users before the moment of purchase through the

use of different logos.

For the revision of the EU GPP criteria for personal computers, Dodd et al. (2016) proposed

a comprehensive criterion on ease of replacement for rechargeable batteries. In their

proposal, rechargeable batteries must not be glued or soldered into portable products (as

a core criterion). Furthermore, Talens Peiró et al. (2016) and Dodd et al. (2016) identified

three different requirements for batteries with a performance of less than 800

charge/discharge cycles (when tested according to IEC EN 61960): manually

exchangeable, without tools, for notebooks and portable all-in-one computers;

exchangeable with a screwdriver in a maximum of three steps for sub-notebooks; and

exchangeable with a screwdriver and spudger in a maximum of four steps for tablets and

two-in-one notebooks. We elaborated further on these options, to avoid differentiations

among mobile computer subcategories, and we thought about possible logos to be used

to identify different levels of difficulty in replacing the battery of a personal computer.

— Logo 1: identifies that the batteries of the portable computer can be

disassembled and replaced by the user, with or without the need of tools.

— Logo 2: identifies that the batteries of the portable computer cannot be

disassembled and replaced by the user: this task requires assistance.

The disassembly operations in logo 1 should be performed using manual or power-driven

standard tools. The list of the tools to be considered can be drawn from Annex B of

Recchioni et al. (2016). Assistance is required for disassembly operations of logo 2,

because of the complexity of the disassembly, or because of the use of glues and

adhesives, or because the disassembly operation may damage the product or compromise

the safety of the end-user.

Definitions and possible symbols to identify logo 1 and logo 2 (Table 31) should be targeted

by standardisation activities.

(

) 612 notebook computers were analysed.

105

Table 31 — Proposed logos and correlation with the ease of disassembly of batteries.

Logo 1

Battery can be removed and replaced by

the user, with or without the need of

tools.

Logo 2

Battery replacement requires

assistance (*).

(*) The battery contained in the product

cannot be replaced by the end-user, but

by professionals. The user manual could

provide detail and information about the

customer service to be contacted.

This classification concerning logo 1 could be potentially split into two categories.

— Logo 1.1: identifies that the batteries of the portable computer can be removed

and replaced by the user, without the use of manual or power-driven tools.

— Logo 1.2: identifies that the batteries of the portable computer can be removed

and replaced by the user, with the use of manual or power-driven tools.

As proposed for the revision of the EU GPP criteria, manufacturers could illustrate how the

battery is installed in the product, the steps required to remove and cover markings, and

provide comprehensive instructions in the user manuals.

In the future, more ambitious actions could be proposed about the ease of disassembly of

the product. Indeed, the repair operators interviewed confirmed that firmly glued

computer components may represent an obstacle to the product repair. Nevertheless,

adhesive tapes used to fasten two or more parts may reduce the ease of disassembly of a

product, even though they also seal out water and dust.

The adhesive strength of the adhesive tapes used in electronics is generally specified in

N/cm, representing the amount of force required to peel up a cm-wide strip of tape at a

constant rate. However, the configuration and amount of tape used can vary significantly,

drastically changing the difficulty in separating adhered components. For these reasons,

setting a limit to the adhesive strength of adhesive tapes used in computers would not

help in improving their ease of disassembly. A more verifiable measure could be the pull-

off force required to separate adhered components, measured as a pressure, in N/cm

However, further standardisation work is needed to identify a reliable procedure to test

adhesive strength, environment conditions (temperature) and reference values.

Future actions

Considering the ease of disassembly of computers, a metric/index on disassembly of

products can be developed, where quantitative parameters for disassembly are either

Boolean (e.g. ‘Are only reversible disassembly operations necessary to open the device?’

(yes/no) or integer values (e.g. number of steps X to remove the battery). A threshold on

the number of disassembly steps to access, replace and reassemble a particular

component of a device could be established to ensure fast and safe repairs. ‘Reversible

disassembly’, in this context, means that all of the following are true.

(a) The sequence of disassembly steps can be reversed to assemble the product.

(b) The parts to be (dis)assembled have not broken in any case of professional

handling.

Further standardisation work might be necessary to set out unambiguously what

‘disassembly steps’ are. Moreover, further research would be needed to establish the

target value for the threshold on the number of steps. Standards under the development

106

of European Mandate M/543 for material-efficiency aspects of energy-related products

(European Commission, 2015b) could serve the purpose, as those relate to the

development of methods to assess ability to access or remove components from products

to facilitate the repair, remanufacture or reuse.

Regarding the ease of disassembly of batteries in personal computers, future actions could

consider asking manufacturers to design computers in such a way that the battery can be

always be replaced by end-users, with or without tools, and with instructions provided in

the user documentation. Fixed batteries could still be used in cases where there are safety

issues, documented by manufacturers.

6.1.3 Initial assessments of benefits/impacts

Because notebook and tablet computers are so mobile, they are much more susceptible

than desktop computers to potential risks such as travel wear and tear, airport security,

and everyday accidents such as bumps or spills. Recent IDC studies focused on the US

business sector found nearly 20 % of notebook computers break or require repair at some

point in their lifetime. In their 2010 survey, each year 9.5 % of notebook computers in

business organisations were damaged due to an accident, and 14.2 % of notebooks

reported other kinds of physical problems (i.e. hardware malfunction) (IDC, 2010).

According to the 2016 survey, instead, the percentage of notebook computers in business

organisations damaged due to an accident slightly increased (11 % each year), while the

other reported physical problems decreased (about 12 % each year). Overall, about

18.5 % of notebooks and 14.5 % of tablets needed repair of any kind (IDC, 2016).

Another source, SquareTrade, analysed failure rates for over 30 000 new notebook

computers covered by warranty plans and found that one third of all notebooks will fail

within 3 years. As mentioned in Section 3.2.1, two thirds of these failures came from

hardware malfunctions, and one third was reported as accidental damage (SquareTrade,

2009).

Upgradability is also a key element to extend the lifetime of a computer. According to the

survey conducted by Prakash et al. (2016a), the main reasons for buying a new notebook

was that the old one was faulty (46 %), followed by ‘the old one didn’t have enough

functions’ (25 %). From another survey, half of the respondents judged relevant that old

computers or notebooks can be upgraded with components with higher energy efficiency

or with higher performance (Forsa, 2013).

In our analysis we focused on possible scenarios with improved reparability only, trying to

understand the amount of notebook and tablet computers that risk being discarded

because repair is not feasible. Possible improvements in terms of reparability aim to reduce

this amount and to extend the lifetime of devices that were damaged or which reported

malfunctions.

A summary of failure rates due to malfunctions and accidents is provided in Table 32 for

notebooks. The total failure rate for notebook computers within the first year of use is

7.2 %, 19.7 % in the first 2 years of use and 31 % within the first 3 years, according to

SquareTrade. On the other hand, the most recent average failure rate was reported to be

18.5 % by the IDC and this average value was used to develop a scenario for the

reparability of notebook computers. On the other hand, the average failure rate of 14.5 %

(again reported by the IDC) was used to develop a scenario for the reparability of tablet

computers.

107

Table 32 — Notebook failure rates due to hardware malfunction and accident, according

to two sources of data: the IDC (2010, 2016) and SquareTrade (2009).

Failure rates

(IDC, 2010)

(IDC, 2016)

(SquareTrade, 2009)

First year of

use

2 years of

use

3 years of

use

Hardware

malfunction

14.2 %

12.3 %

4.7 %

12.7 %

20.4 %

Accident

9.5 %

11.0 %

2.5 %

7.0 %

10.6 %

Total

(

) 19.6 %

18.5 %

7.2 %

19.7 %

31.0 %

Considering the estimated annual sales of personal computers (years 2020, 2025 and

2030, Section 2.1), it was possible to estimate the number of mobile computers

(notebooks and tablets) sold in a certain year that are likely to report a failure. However,

it was not possible to estimate the uncertainty associated with these projections, and

alternative scenarios were analysed. Table 33 provides detail about the number (millions

of units per year) and the mass (t per year) of computers that are expected to report a

failure, considering the average yearly failure rate of 18.5 % for notebooks and 14.5 %

for tablets. This evaluation was done on the estimated sales of personal computers, and

is expected to be greater when considering the stock volume.

Table 33 — Computers expected to report failures (failure rate of 18.5 % for notebooks

and 14.5 % for tablets). Estimations based on the sales expected for 2020, 2025 and

2030.

Year

2020

2025

2030

Notebooks

million units/year

7.71

7.69

7.72

Tablets

million units/year

5.57

5.58

5.59

Total

million units/year

13.27

13.26

13.31

Notebooks

t/year

14 875

14 835

14 903

Tablets

t/year

2 942

2 949

2 956

Total

t/year

17 817

17 785

17 859

In our scenario we considered that some of the computers reporting a failure are not

repaired and are discarded as WEEE. The percentage of computers that are repaired,

instead, is represented by the repair rate. As no data on repair rates are available, to the

knowledge of the authors, estimations were developed thanks to expert judgments. Repair

rates depend on several factors, such as the type of failure, the type of repair needed, the

initial cost of the computer, the age of the device or warranty plans. In our hypothesis we

considered two main scenarios.

— Computers in the first 2 years of use, with warranty plans. Repair rate 80 %.

— Computers older than 2 years, with no warranty plans. Repair rate 20 %.

Results are reported in Table 34, as the number (millions of units per year) and the mass

(t per year) of notebooks and tablets that are expected to be discarded as WEEE and

therefore directed to EoL processes.

(

) Notebook computers needing repair of any kind.

108

Table 34 — Notebooks and tablets expected to be discarded as WEEE (repair rate 80 %

for mobile computers in the first 2 years of use, with warranty plans; repair rate 20 % for

mobile computers older than 2 years of use, with no warranty plans; average failure rate

of 18.5 % for notebooks and 14.5 % for tablets).

Year

2020

2025

2030

Age of the

computer

years

≤ 2

> 2

≤ 2

> 2

≤ 2

> 2

Notebooks

million

units/year

1.54

6.17

1.54

6.15

1.54

6.18

Tablets

million

units/year

1.11

4.45

1.12

4.46

1.12

4.47

Total

million

units/year

2.65

10.62

2.65

10.61

2.66

10.65

Notebooks

t/year

2 975

11 900

2 967

11 868

2 981

11 923

Tablets

t/year

588

2 354

590

2 359

591

2 365

Total

t/year

3 563

14 254

3 557

14 228

3 572

14 287

We then observed what could be the situation when the percentages of computers

discarded as WEEE are decreased by 5-20 %, thanks to improved reparability. The two

scenarios in any case of enhanced reparability can be then summarised as follows.

— Notebooks and tablets in the first 2 years of use, with warranty plans. Repair rates

hypothesised to be 81-84 %.

— Notebooks and tablets older than 2 years, with no warranty plans. Repair rates

hypothesised to be 24-36 %.

Results are reported in Table 35 (computers in the first 2 years of use, with warranty

plans) and Table 36 (computers older than 2 years, without warranty plans), expressed

as material savings. Repairs were simulated considering an average repair service with

replacement of the components more prone to fail according to the IDC, namely the

display, the battery, the HDD and the motherboard.

Table 35 — Products (million units/year) and material (t/year) savings thanks to enhanced

reparability (repair rate 81-84 % for computers in the first 2 years of use, with warranty

plans, average failure rate of 18.5 % for notebooks and 14.5 % for tablets).

Year

2020

2025

2030

Notebooks

million units/year

0.077 -

0.308

0.077 -

0.307

0.077 -

0.309

Tablets

million units/year

0.056 -

0.223

0.056 -

0.223

0.056 -

0.224

Total

million units/year

0.133 -

0.531

0.133 -

0.531

0.133 -

0.533

Notebooks

t/year

135 -

542

135 -

540

136 -

543

Tablets

t/year

18 -

Total

t/year

153 -

613

153 -

612

154 -

615

Considering the first 2 years of use, between 0.13 and 0.53 million units of notebooks and

tablets, expected to be discarded as WEEE, are now considered as potentially repaired

devices. With this hypothesis between 150 and 620 t of materials can be saved every year.

When notebooks and tablets older than 2 years and without warranty plans are

considered, repair rates and therefore material savings can be potentially increased, with

material savings ranging from 610 to 2 460 t per year (Table 36).

109

The values of Table 35 and Table 36 take into consideration the amount of resources

required to produce the spare parts necessary for the repair.

Table 36 — Products (million units/year) and material (t/year) savings thanks to enhanced

reparability (repair rate 24-36 % for computers older than 2 years, with no warranty

plans, average failure rate of 18.5 % for notebooks and 14.5 % for tablets).

Year

2020

2025

2030

Notebooks

million units/year

0.308 -

1.233

0.307 -

1.230

0.309 -

1.236

Tablets

million units/year

0.223 -

0.890

0.223 -

0.893

0.224 -

0.895

Total

million units/year

0.531 -

2.124

0.531 -

2.122

0.533 -

2.130

Notebooks

t/year

542 -

2 166

540 -

2 161

543 -

2 170

Tablets

t/year

72 -

287

72 -

288

72 -

288

Total

t/year

613 -

2 453

612 -

2 448

615 -

2 459

6.1.4 Potential benefits for other product categories

The assessment of the potential benefits related to enhanced reparability of desktop

computers is characterised by higher degrees of uncertainty. To the knowledge of the

authors, there is no statistical analysis reporting figures for yearly failure rates of desktop

computers, so a hypothetical failure rate of 16.5 % (as an average of 18.5 % for

notebooks and 14.5 % for tablets) was assumed.

In the case of desktop computers, however, it a very high repair rate of 90 % (the

percentage of devices which reported a failure and were repaired) was assumed,

independent of the age of the desktop computer. It was also assumed that enhanced-

reparability strategies would bring smaller benefits in terms of repair-rate increase (in the

range of 0.5-1 %).

With these hypotheses, and considering estimated market data for years 2020, 2025 and

2030 (as in Table 1) 2-2.24 million desktop computers will report failures, and without

enhanced-reparability strategies 0.20-0.22 million computers will be discarded. Taking

into consideration the hypotheses on enhanced-reparability strategies, 96-216 t of

materials are saved every year instead.

However, we remark that future work is needed to collect data about the repair services

of desktop computers in order to strengthen the estimates.

110

6.2 Secure data deletion for personal computers

6.2.1 Rationale

One major barrier to the reuse, repair and recycling of computers is data-privacy issues.

A recent article by Polverini et al. (2017) investigated the relationships between resource

efficiency aspects in electronics and data protection and cybersecurity issues. This article

identified that users are generally not keen in reusing computer because not ensured about

the properness of data erasure after first use.

Desktop computers, notebooks and tablets regularly store sensitive and confidential data

pertaining to users and organisations, including (but not limited to) documents, photos,

videos, location and contact data, stored on various media such as HDD, SSD, flash, SIM

and memory cards. The major operating systems usually include an option to ‘factory

reset’ the device, bringing the device into its original factory state (

). However, this does

not necessarily guarantee that all the personal data of the user are deleted

comprehensively and permanently. Hence, it is believed that data-privacy issue is one of

the major factors that discourage users from making their obsolete but functional devices

available to the reuse market or to the appropriate recycling paths in the case of

dysfunctional devices.

Data sanitisation is the process of deliberately, permanently and irreversibly removing or

destroying the data stored on a memory device (prEN 50614, 2016). Other techniques of

data eradication do not allow the reuse of the device (e.g. degaussing magnetic media,

drilling HDD platters). Besides data sanitisation, it may be viable to encrypt user data and

so permanently delete the key required for decryption so as to ensure third parties cannot

access the user data thereafter. This means that the data are still physically present on

the storage media, but permanently inaccessible.

It should be noted that depending on the effort invested, it cannot necessarily be fully

guaranteed, that user data cannot be recovered using highly sophisticated technical tools.

Hence, users would mainly benefit from reasonably-safe data sanitisation, without taking

into account data recovery methods that require large amount of temporal and financial

investments.

6.2.2 Possible improvements

Personal computers could have tools available (or preinstalled (

)) to permanently delete

any personal data contained in data-storage systems without compromising the

functionality of the whole device for further reuse. Secure (

) data deletion could be

ensured by means of a dedicated functionality or software. If data deletion cannot be

ensured, personal computers could have tools available to encrypt personal data in storage

systems and to permanently delete the key required for decryption.

A study on computer severs (Talens Peiró and Ardente, 2015) compiled a list of available

standards by country, based on data from Hintermann and Fassnacht (2008) and Fisher

(2015). Talens Peiró and Ardente (2015) also provided a list of possible methods available

for data deletion, as detailed in Table 37.

According to the US Department of Defense Standard 5220.22-M for clearing and

sanitisation for different types of media, data clearing is defined as ‘a method of

(

) At the time of writing this feature is available in some form at least on Windows 10, macOS X, Android and

iOS.

(

) A built-in functionality can be defined as a functionality that does not require the installation or usage of

additional software or hardware components not already present in the provided system.

(

) Secure data deletion means the effective erasure of all traces of existing data from storage media,

overwriting the data completely in such a way that access to the original data, or parts of them, becomes

unfeasible for a given level of effort.

111

sanitisation by applying logical techniques to sanitise data in all user-addressable storage

locations for protection against simple non-invasive data recovery techniques using the

same interface available to the user; typically applied through the standard read and write

commands to the storage device, such as by rewriting with a new value or using a menu

option to reset the device to the factory state (where rewriting is not supported)’.

Hence, standards on data sanitisation are particularly relevant to enable the reuse of

devices. The draft Standard prEN 50614 (2016) provides some examples of nationally

approved data-sanitisation standards, such as HMG IS Standard No 5 (UK), DIN 66399

(Germany), NIST 800-88r1 (US). Other initiatives in support of data deletion are the CPA

security characteristics for data sanitisation — flash based storage, the CPA security

characteristic overwriting tools for magnetic media version 2.1, and the CAS sanitisation

requirements version 2.0 Nov 2014.

While the user-addressable storage in desktop computers can oftentimes be disassembled

with reasonable effort, storage solutions in more-integrated devices, such as notebooks

and tablets, are less easily accessed. This emphasises the importance of tools that allow

the users to delete their data, without having to rely on third parties, before the devices

are passed on for reuse or recycling.

The German environment label Blue Angel has a set of criteria for mobile phone (RAL

GmbH, 2013). Section 3.3.3 of the document describes the requirements in terms of data-

deletion issues:

To allow a second use of a mobile phone the device shall be designed so as to allow the

user to completely and safely delete all personal data on his own without the help of pay

software. This can be achieved by either physically removing the memory card or with the

help of software provided by the manufacturer free of charge. When using a software, the

deletion process shall at least include an ‘overwrite’ of all the data stored with a random

pattern, or, in the case of flash storage with zero values.

Table 37 — Methods for secure data deletion (Talens Peiró and Ardente, 2015)

Data-deletion method

Overwriting

(number)

Description of

overwriting cycles

Bruce-Schneier algorithm

first: zero

second: one

third to seventh:

random character

Peter-Gutmann algorithm

random character

Pfitzner (created by Roy Pfitzner)

random character

Random data

customised

random character

Secure Erase (Parallel ATA (PATA)

and Serial ATA (SATA) based hard

drives)

first: writes a binary

one or zero

Write zero (used by Windows Vista

and following windows versions)

zero

6.2.3 Potential benefits for the product group

The need to take action on secure data deletion is in line with the principles of privacy and

protection of personal data as set by the General Data Protection Regulation

(EU) 2016/679 and in particular its Article 25 on ‘data protection by design and by default’.

If the ambitious objectives concerning reuse contained in the circular economy package

are to be implemented, special care should be taken for the protection of personal data

contained in electronic products and components. Secure data deletion is also becoming

a day-to-day activity of EoL operators (EGG 2016+, 2016).

112

Secure data deletion is therefore considered as a necessary prerequisite to allow reuse of

computers, complying with principles on privacy and protection of personal data.

The assessment of the potential benefits related to secure data deletion is however

characterised by certain degrees of uncertainty. To the knowledge of the authors, there is

no statistical analysis reporting figures for computers discarded (or not repaired/reused)

due to data-deletion issues, nor robust surveys.

It was assumed that currently 5 % of the volume of desktop, notebook and tablet

computers of Table 1 is reused after the first useful lifetime, without repair activities or

component replacement, but only thanks to cleaning and data deletion. It was then

assumed that secure data deletion would bring additional benefits in terms of reuse rates

(6-10 % of the volume of desktop, notebook and tablet computers of Table 1).

With these assumptions, and considering estimated products sold for 2020, 2025 and

2030, as in Table 1, 0.9-4.7 million computers would be reused and therefore a significant

amount of materials would benefit from an extended lifetime. The lifetime extension

depends on a case-by-case basis. Considering these hypotheses, however, it is possible

to estimate a lifetime extension for 2 300-12 200 t of materials every year (59 % for

desktop computers, 33 % for notebooks and 8 % for tablet).

Also in this case we remark that future work is needed to strengthen the estimates.

113

7 Possible actions to enhance recyclability

The design for dismantling of computers is a necessary precondition for their efficient and

swift depollution and recycling. The following sections aim to provide proposals to support

the ease of dismantling.

7.1 Dismantlability of key components for personal computers

7.1.1 Rationale

According to the analysis in Section 3, notebooks and tablets at EoL, after depollution with

the extraction of the battery, can follow two main processing routes: a first one based on

the full mechanical crushing (shredding) and sorting of the waste; and a second one

including some additional pre-treatments (medium-depth manual dismantling) before

subsequent shredding and mechanical sorting.

Article 15 of the WEEE directive (European Union, 2012) also calls for ‘Member States to

take necessary measures to ensure that producers provide information free of charge

about preparation for reuse and treatment in respect of each type of new EEE placed for

the first time on the European Union market within 1 year after the equipment is placed

on the market’. Relevant information about EEE placed on the market is crucial for WEEE

treatment operators. Indeed, the rapid evolution in product design, the miniaturisation of

EEE, components and materials used for their manufacturing some of which are critical

make their repair and recycling increasingly challenging (EuRIC, 2016b). However,

according to the association of reuse-and-recycling industries this article remained so far

largely neither implemented nor enforced (EuRIC, 2016b).

These considerations have also been confirmed by the recyclers interviewed, who

reiterated that, for the safe and efficient recycling of computers, products should be

designed so that the access and removal of batteries (

) (including button batteries

contained in the mainboard), display panels and PCBs (contained in several components,

including motherboard, memory RAM, CPUs, graphic cards, and mass-storage systems) is

facilitated. In particular, there is the risk that certain components of computers (e.g.

batteries and display panels) which are difficult to extract would be shredded together

with other waste, with the consequent dispersion of pollutants and contamination of other

recyclable fractions (DEFRA, 2006), the risk of explosion in the shredders (Hand, 2013;

Powel, 2002), and the irreversible loss of valuable materials (Van Eygen et al., 2016).

Improper battery treatments can be associated with risks in terms of worker and facility

safety (Section 3.1.6), including accidental fires in the WEEE treatment plants.

For the safe and efficient recycling, information on dismantling process and location of

battery and other valuable components is essential (EERA, 2016). Information could

concern the following.

— Extra information on materials that are recyclable if certain technology is used (for

example for certain plastic components containing additives).

— Content of dangerous components/substances used (as a minimum the ones

mentioned in Annex VII of the WEEE directive, see Section 3.1): provision of a

short description and photo, and the location where these are usually found in the

appliance.

(

) Measures to improve the design for disassembly of batteries from computers are also in lines with the

principles of the Batteries Directive 2006/66/EC, which state in article 11 that appliances should be designed

in such a way ‘that waste batteries and accumulators can be readily removed. Appliances into which

batteries and accumulators are incorporated shall be accompanied by instructions showing how they can

be removed safely and, where appropriate, informing the end-user of the type of the incorporated batteries

and accumulators’.

114

— Dismantling instructions: these could include exploded diagrams of the computer

model, indicating the opening mechanism and required tools; in the case of clips,

this should include information related to the direction in which the housing should

be opened.

— How to recognise special models and specific dismantling instructions for them.

— Advice on collection (separate/mixed) and on logistics.

Additional relevant information could also include the following (EERA, 2016).

— Extra information on materials that are recyclable if certain technology is used (e.g.

poly-methyl methacrylate (PMMA) plates from displays to be dismantled manually).

— Information on those batteries which cannot be removed without (advanced) tools

(providing then information on what tools should be used and where to find them).

— Description of the component/substance and its different types, e.g. whether

substance is dangerous.

— Personal-protection equipment needed for handling.

— Risks for workers when the waste is not properly dismantled.

— Advice on possibilities to sort the components or substances (when different

treatment is possible for different types).

— Advice on available treatment techniques.

It is also highlighted that such information, should be provided in a standardised manner

to maximise its operational utility (EuRIC, 2016a).

Apart from all this information (to be provided e.g. via digital platforms), recyclers stressed

the importance of labelling, provided that the information fulfils the following conditions

(EERA, 2016): it is uniform; it is adopted early and by all; and it is visible and easily

recognisable (big logos or letters, colours). The labelling should be applied to (EERA,

2016):

— provide information on hazardous components and substances;

— give instructions for logistics and/or treatment.

It is also recognised that up to one third of total WEEE produced in the EU, including

computers, is not correctly disposed of and treated (Huisman et al., 2015). In particular,

there is a risk that the small dimensions of IT equipment would facilitate the incorrect

sorting by users into the waste bin. Economic incentives for a proper waste collection and

treatment are crucial, as for example, establishing deposit/refund systems for computers,

in order to incentivise users for a proper disposal of the waste and improve material

recovery (Huisman et al., 2015; Zhong and Schiller, 2011).

7.1.2 Possible improvements

Computer designs could be developed so that components that are crucial for material-

efficiency aspects can be easily located, extracted and addressed to specific recycling

treatments. Measures to ease the disassembly have been proposed and analysed for

various EEE (Ardente et al., 2013; Ardente and Mathieux, 2012b; Talens Peiró et al., 2016;

Talens Peiró and Ardente, 2015).

In order to facilitate the ease of dismantling of key components (such us batteries, PCB

assemblies larger than 0.1 dm

, display panels larger than 1 dm

, any mercury-containing

component or capacitors containing electrolyte) specific joining or sealing techniques can

be used. In particular, large number of different fastenings and/or certain types of

fastening which are difficult to be dismantled can represent an obstacle for recyclers for

the efficient recovery of key components. According to EERA (2016) and EuRIC (2016a),

in order to improve the recycling, it is absolutely useful that components of different

material composition (such as plastics and metals or batteries and PCB) are not soldered,

115

glued or permanently fixed together. Recyclers experience many difficulties when these

material groups are fixed together, resulting in higher material losses (

Key components for this and other product groups were also identified by Talens Peiró et

al. (2016) (computer and electronic displays) and Dodd et al. (2016, 2015), who

suggested thresholds on the maximum time required to extract them from the device.

Such an approach, proposed for the revision of the EU Ecolabel and EU GPP criteria,

however, poses some problems for verification, and its application in the context of the

ecodesign directive would require standardised procedures to measure or calculate the

time required for the extraction.

Ease of dismantling can be proved and enhanced thanks to a comprehensive

documentation on the sequence of operations needed to access the key components,

describing the type and number of fastening technique(s) to be unlocked, and tool(s)

required. As for the disassembly, in this case the exploded diagram of the product showing

the location of the components to be dismantled can also be useful. The recycling industry

welcomes making the information available electronically/on digital platform(s), which is

the main format easily accessible by relevant stakeholders (EuRIC, 2016a). The period of

15 years is considered by the recycling industry to be an possible time frame with regard

to the duration for which the document should be kept available (EuRIC, 2017).

It is important to remark that evidence collected so far indicate that the design of desktop

computers is generally not posing dismantling problems during recycling (see Section

3.1.1). The design of the new models of desktops (e.g. ‘mini desktop’) could be the source

of some problems in the future, since their compact structure makes their design similar

to that of games consoles (

). However, a limited number of these computer models have

reached EoL, and therefore EoL processes can only be estimated through analogies with

similar product groups (such as games consoles or notebook computers). Based on the

very limited information available from a manufacturer (see Section 3.1.1), mini desktop

computers should not cause high levels of difficulty for their recycling. However, based on

the comments received from a stakeholder, the compact structure of mini desktops could

hamper the dismantling of these computers and the removal of batteries and other

valuable components as CPU and SSDs.

For desktop computers with integrated displays, their EoL treatments are affected by

problems similar to those of electronic displays (see Section 3.1.2). Furthermore, they

could be specifically labelled in order to allow recycling operators to identify them as

computers already at the early stages of the recycling process.

Some components, such as tablet frames containing magnesium, although shown as

relevant for dismantling in the analysis of recycling practices (see Section 3.1.4), have

been excluded from the list of targeted components. This is due to the fact that frames

can take various form and shapes and that the requirement could hence be difficult to be

verified. It is however argued that ensuring an easier dismantling of the abovementioned

components should also enhance facilitated dismantling of frames.

According to a European recycler association, information relevant for dismantling should

be made accessible to recyclers and market-surveillance authorities, ideally through

dedicated digital platforms (

), as for paper documentation there is the risk that it is static

and becomes outdated if not revised in time (EERA, 2016).

(

) The consequence is that, if this is not regulated in some way, recycling and recovery rates as given in Annex

V of the WEEE directive cannot be met (EERA, 2016).

(

) In the case of games consoles, industries proposed a ‘self-regulatory initiative’, addressing also some EoL

issues (Sony, Microsoft, Nintendo, 2015). This documents states that ‘to improve both recycling and reuse

at end-of-life, maintenance and refurbishment is possible by non-destructive disassembly’ and ‘To improve

recycling at end-of-life, console plastics parts > 25g are marked indicating their material composition (using

ISO conforming marks)’.

(

) There are ongoing projects about how to develop and communicate relevant information for recyclers. For

116

A standardised format for the documentation to support the verification of the requirement

have to be established. For example, the format published by the Austrian ministry of

environment (

) can represent a basis. Moreover, this standardised format should be

based on the horizontal standardisation work under European Mandate M/543 on material-

efficiency aspects of energy-related products (European Commission, 2015b), which

requires ‘documentation and/or marking regarding information relating to material

efficiency of the product taking into account the intended audience (consumers,

professionals or market-surveillance authorities)’ to be developed.

Additional work is also necessary to unambiguously set out what is a high adhesion two-

sided adhesive tape. Future options could improve the design for the dismantling of the

products (based, for example, on the development of metrics to assess the ease of

dismantling (

)). Again, standardisation work under Mandate M/543 (European

Commission, 2015b) could be initiated, for the specific development of methods to assess

the ability to access or remove certain components or assemblies from products, and to

facilitate their extraction at EoL for ease of treatment and recycling.

7.2 Marking of plastic components

7.2.1 Rationale

New EEE use more and more quantities and different types of plastics (WEEE forum, 2017).

The EEE industry accounts for 5-7 % of the total European plastic demand, and the

polymers used are highly engineered with the inclusion of additives. Although in theory

plastics are all perfectly recyclable, in practice the recyclability of plastics is generally very

low (EN TS 16524, 2013). ‘Products consisting mainly of plastic have a very low

recyclability rate in practice and it is all the lower when different plastics are combined in

the same product’ (EN TS 16524, 2013). Moreover, the European Commission in 2013

observed that only a small fraction of plastic waste is at present recycled (European

Commission, 2013). Appropriate measures to enhance the recycling of plastics could also

improve competitiveness and create new economic activities and jobs (European

Commission, 2013).

Plastic recycling poses various problems (Elo et al., 2009; Peeters et al., 2014).

— The lack of process capable of performing plastic sorting and separation.

— Plastic can be recycled roughly a limited number of times; then the plastic is worn

out and of poor quality.

— Complexity of the plastic mix, which makes it both difficult to separate plastics from

each other and generally expensive to recycle.

— Plastics can contain several additives which degrade the virgin plastic.

— Plastic can be reinforced or mixed with metals and other non-plastics, which

degrade the plastic when recycled.

— Most plastics type are only present in relatively small flow amounts, which makes

it difficult to achieve the required economies of scale for advanced recycling

operations.

example, the EU Horizon 2020 project ‘CloseWEEE’ (http://closeweee.eu/) aims to develop processes for

separation and recovery of materials (including plastics, CRMs, and other valuable metals) from WEEE

streams, and to improve the flow of information to recyclers through a dedicated digital platform (named

‘recycler information centre’ — http://www.werecycle.eu/) in order to make recycling procedures quicker

and safer

(

) ‘Leitfaden für die Behandlung von Elektro- und Elektronikgeräte‘ (EERA, 2016).

(

) For examples of metrics to assess the ease of disassembly, see: Vanegas P., Peeters J.R., Cattrysse D.,

Duflou J.R., Tecchio P., Mathieux F., Ardente F., 2016. Study for a method to assess the ease of disassembly

of electrical and electronic equipment — Method development and application in a flat panel display case

study. EUR 27921 EN. doi:10.2788/130925 (Available: https://ec.europa.eu/jrc/en/publication/study-

method-assess-ease-disassembly-electrical-and-electronic-equipment-method-development-and?search)

117

Density sorting of plastic (via sink-float techniques) is currently the easiest and still most

adopted sorting systems for shredded plastics (Peeters et al., 2014). Different plastics are

separated according to their different densities thanks to water or air separators. Some

advanced processes for the separation of plastics are currently under development (e.g.

near infra-red (NIR) analysis spectroscopy, x-ray fluorescence (XRF) spectroscopy, Visible

light optical separation), although their efficiency of separation and their applicability to

the sorting of shredded plastics are still under investigation (Elo et al., 2009; Peeters et

al., 2014). Sorting of different plastics is also performed based on manual dismantling.

This technique can be technically and economically viable for high-quality technical plastics

used in EEE, including computers (Mathieux et al., 2008; Peeters et al., 2014).

The efficiency of manual sorting of plastics is, however, dependent on the properness of

plastic marking, values of recyclates and labour cost. Marking of plastic should follow

standardised approach, as that proposed by ISO 11469 (ISO 11469, 2000), and standards

of the series ISO 1043 (EN ISO 1043-1, 2002; EN ISO 1043-4+A1:2016, 1999).

Nevertheless, EERA (2016) observed that markings on plastics in use nowadays are not

fully reliable in some cases. Moreover, codification of WEEE plastics is often ambiguous

and not harmonised at EU level (WEEE forum, 2017). Tests carried out at the premises of

an EERA member showed that markings on the back-covers of flat panel displays were not

reliable, and the polymer type often did not match with the marking. Recyclers who follow

the markings can therefore end up separating materials incorrectly and this could

potentially lead to them having contaminants (such as brominated flame

retardants (BFRs) in materials where BFRs should not be present) (EERA, 2016). Audits

should be enforced to ensure that the marking and the plastic type match.

7.2.2 Possible improvements

Associations of WEEE recyclers suggested that the proper marking of plastics (and their

additives and FRs) would be beneficial for recycling companies, especially for recyclers

that dismantle plastic parts manually (EERA, 2016). WEEE Forum (2017) recommends

developing best-practice guidelines for plastics sorting, characterisation of plastic fractions

as well as monitoring and tracing of the destination of sorted outputs.

In order to improve the manual separation of valuable plastic parts, the marking of plastic

parts above a certain weight (e.g. 50 g) could be systematically applied.

The marking of plastic parts, as said, should follow a standardised approach (see

Section 7.2.1), with specific exemptions (

) as for example, the following.

— PCB assemblies.

— PMMAs, and other optical plastic parts.

— Wiring and cables.

— Packaging, tape and stretch wraps.

— Labels.

— Electrostatic discharge components and, electromagnetic interference components.

— Acoustics module.

— Plastics where marking is not possible because of the shape or size of the part, or

when the marking would impact on the performance or functionality of the part, or

where marking is technically not possible because of the moulding method.

As reported in Ardente et al. (2016), plastic components containing FRs can be recognised

through standardised symbols. Both the document establishing EU Ecolabel criteria

(European Commission, 2016), and the study supporting the revision of the EU GPP criteria

(

) These exemplary exemptions are derived from comments received from industries related to the marking

of plastics in other products (electronic displays).

118

of personal computers (Dodd et al., 2016) include the marking of plastic components and

plastic components with FRs, by means of ISO 11469 and ISO 1043 markings.

An example of comprehensive plastic marking is shown in Figure 24. Among the different

information in the marking, polymer type and FRs are probably the most relevant for

recyclers.

Figure 33 — Plastic marking according to ISO11469 (adapted from Bombardier (2010)).

It is also highlighted that aspects not specified by the Standard ISO 11469 are (Ardente

et al., 2016):

— components to be marked,

— dimensions of marking,

— font to be used,

— location and visibility,

— additional quality characteristics (e.g. being legible, visible, durable and indelible).

7.3 Declaration of flame-retardant content

7.3.1 Rationale

As mentioned in Section 7.2, recycling of plastics can pose various problems during the

recycling, especially due to the content of additives as FRs. According to all the recyclers

interviewed, FRs (including BFRs) are the major barrier to plastic recycling (WEEE Forum,

2017). Current mechanical-sorting processes of plastics with additives are characterised

by low efficiency, while innovative sorting systems are still at the pilot stage and revealed

to be effective only in specific cases (Ardente et al., 2016).

FRs are chemical additives added to plastics to avoid potential internally and externally

initiated ignitions. FR are used for EEE and, in particular, computers. For example, the

analysis of the bill of material of notebooks revealed the presence of two large plastic parts

(mass around 100 g) in polycarbonate with halogen-free phosphorous compound (code

FR 40, according to ISO 1043-4).

However, FRs can reduce the recyclability of plastic parts. The presence of additives can

reduce the mechanical properties of the materials, requiring additional treatments and

additives to compensate for the degradation of such properties, as well as reduce the value

of the materials in the market, and consequently the economic feasibility of recycling

(Dawson and Landry, 2005). For such a purpose, the (IEC/TR 62635, 2015) suggests in

the annexes that a 0 % recycling rate should be considered for polymers with FRs that are

not properly separated from the other materials before the shredding.

Moreover, some FRs (such as certain BFRs) have high toxicity and for this reason they

have been regulated, for instance by the Directive 2011/65/EU on the restriction of the

use of certain hazardous substances in electrical and electronic products (RoHS). This

directive established that Member States must ensure that new electrical and electronic

equipment put on the market does not contain substances such as polybrominated

biphenyls (PBB) or polybrominated diphenyl ethers (PBDE) (European Union, 2011). In

addition, the directive on WEEE states in Annex VII that plastic-containing BFRs have to

be removed from any separately collected WEEE (European Union, 2012).

119

7.3.2 Possible improvements

The provision of information on the content of FRs in plastic parts is a first step to

contribute to the improvement of plastic recycling. Plastic marking (as discussed in

previous sections) can contribute to the separation of plastics with FRs during the manual

dismantling, allowing their recycling at higher rates (in line with prescription of

IEC/TR 62635, 2015). However, more-detailed information about the composition of the

product (including detail of plastic composition) can be beneficial for recyclers and it is

also in line with the principles of the WEEE directive (

The provision of information on the FRs content could be structured and communicated in

a systematised way through specific indexes. These indexes could support recyclers to

check the use of FRs in computers and to develop processes and technologies suitable for

plastic recycling in future. Moreover, these indexes could allow policymakers to monitor

the use of FRs in the products and, in the medium-long term, to promote products that

use smaller quantities of FRs.

As an example, the ‘Flame retardant in plastic components’ index, as specified by Ardente

et al. (2016), aims to do the following.

— Detail plastic components that contains FRs (including mass and type of plastic

components; mass and type of FRs).

— Provide, in a very synthetic way, an overview of the content of FRs.

To simplify the calculation and communication of this index, the scope of the index could

be restricted to plastic components larger than a certain mass (e.g. larger than 50 g). In

addition, some plastic components could be excluded from this calculation (e.g. PCB

assemblies and cables, which always contain FRs). Exemptions could be also planned for

information that is confidential (e.g. the type of certain FRs). In this case, it could be

sufficient to declare that a certain part contains FRs, without specifying the types of FR.

Exemptions need to be also adequately documented.

An example of a calculation table for the ‘Flame retardant in plastic components’ index is

provided in Table 38. All masses are approximated at the gram.

(

) The WEEE directive, in Article 4, states that appropriate measures should be encouraged ‘so that the

ecodesign requirements facilitating reuse and treatment of WEEE established in the framework of Directive

2009/125/EC are applied’.

120

Table 38 — Table for the calculation of the index on ‘Flame retardant in plastic components’

for computers (modified from (Ardente et al., 2016)).

Brand name and Product family:

A) Total mass of the computer [g]:

i. Plastic part

(with flame retardants)

(*)

ii. Mass

[g]

iii.

Polymer(s)

(**)

iv. Flame

retardant(s)

(***)

Part

(1)

…

Part

(2)

…

Part

(j)

…

B) Total [g]

v. Plastic part

(without flame retardants)

(****)

vi. Mass

[g]

vii.

Polymer(s)

(**)

Part

(1)

…

Part

(2)

…

Part

(k)

…

C) Total [g]

B) Total mass of plastic parts in the computer with flame

retardants [g]

(sum of masses in column ii)

C) Overall mass of plastic parts in the computer without

flame retardants [g]

(sum of masses in column vi)

Index [%]

Ratio (in percentage) of plastic parts containing flame

retardants on the total mass of the computer (

𝐵

𝐴

)

Ratio (in percentage) of plastic parts containing flame

retardants on the total mass of plastic parts (

𝐵

𝐵+𝐶

)

(*)

plastic parts containing flame retardants, larger than 50 g

(**)

standard abbreviated term for the polymer(s) in the plastic part, according to EN ISO 1043 series

(***)

standard code number of the flame retardant(s) in the plastic part, according to EN ISO 1043 series

(****)

plastic parts not containing flame retardants, larger than 50 g

7.4 Identifiability of batteries

7.4.1 Rationale

According to the analysis in Section 2.3.5, the market for rechargeable lithium-ion

batteries is growing rapidly, accelerated through the demand increase in portable

electronics, such as tablet and notebook computers. After collection, batteries at the EoL

mostly appear as mixtures and are subject to manual sorting according to their

chemistries. The identification of the chemistry type is based on the logo placed on the

battery packaging/casing. In practice, however, when the batteries reach the recycling

facility, the logos are sometimes missing, making identification and sorting difficult. In

order to release manual labour force, raise the sorting speed as well as accuracy, better

121

marking/identification with improved readability is required in order to implement efficient

identification and sorting.

According to interviews with German battery recyclers, battery marking will facilitate the

separation of mixed batteries and therefore increase the recycling rates of Li-ion batteries.

Furthermore, interviews revealed that the cobalt content in LIB varies between 0 and 15 %

based on the battery sub-chemistry. A more-detailed logo indicating the sub-chemistry

system will be beneficial for more precise sorting and dedicated batch-wise treatment.

7.4.2 Possible improvements

According to EERA (2016), colouring at the component level is good for recyclers to create

awareness and traceability of these components and/or materials and substances that

need to be removed. This principle can be specifically applied to batteries to identify the

battery chemistry.

Battery packs and cells (including those incorporated into battery packs) can be identified

with the ‘battery-recycle mark’, or a similar marking symbol. Indeed, the ‘battery-recycle

mark’ and the IEC draft standard represent a good basis for colour-based logos, even

though additional standardisation activities should be initiated in order to adapt it to the

EU legislation. The battery logo would reduce the limits of current marking practices if

properly applied (visible, durable, legible and indelible). The identifiability of battery

chemistry would be enhanced by the use of different colours.

Standardisation activities are currently ongoing to approve a draft international standard

titled Secondary batteries: Marking symbols for identification of their chemistry

(IEC 62902 draft, 2017). The draft document specifies methods for the clear identification

of secondary cells, batteries, battery modules and monoblocs according to their chemistry

(electrochemical storage technology), by using the battery-recycle mark. The draft

standard concerns secondary cells, batteries, battery modules and monoblocs with a

volume of more than 900 cm³. The marking is applicable for secondary cell and batteries

of following chemistries only:

— lead acid (Pb) (colour: grey)

— nickel cadmium (NiCd) (colour: green)

— nickel-metal hydride (NiMH) (colour: orange)

— lithium-ion (Li-ion) (colour: blue)

— secondary lithium metal (Li metal) (colour: blue).

The draft standard also specifies the dimensions of the marking symbols (with and without

the recycling symbol), how the markings can be fixed to the battery (either by printing or

labelling) and which procedure can be performed to test durability of marking to chemical

agents.

If approved, this draft document may be the starting point for batteries with a volume of

less than 900 cm³, as is the case for personal computers.

Beside of the content of the draft IEC standard, for lithium-ion batteries, a two-digit code

may be added to indicate the content of specific metals as well as substances hindering

recycling.

To improve automated battery sorting solutions, future schemes could go beyond the

proposed colour-coded ‘battery-recycling mark’. One option suggested by a large German

battery-recycling company is to add a QR (quick response) code to both battery cell and

pack. QR codes were initially designed by the automotive industry for its assembly lines

and would well-suited to the need of treatment operators (EuRIC, 2016a). The QR code

could provide more precise information related to the battery subtype, concentration of

cobalt and other REEs as well as a link to material safety sheets. Access to the information

can be limited only to dedicated treatment operators part of the official compliance

schemes to mitigate concerns over innovations in battery technologies.

122

7.5 Provision of information on the content of critical raw materials

7.5.1 Rationale

Within the ‘raw materials initiative’, the European Commission identified a list of CRMs

that are crucial for the EU economy (

). The criticality associated with these materials is,

in many cases, compounded by low substitutability and low recycling rates. Therefore,

boosting material efficiency and increasing the recyclability of these materials has been

identified as one pillar to reduce the risks associated with their supply.

Several CRMs are contained in computers such as cobalt in the batteries, neodymium and

other REE in the HDD magnets, indium in the display panels, magnesium in some metal

frames, and various CRMs (including palladium, REE, antimony, beryllium, cobalt, gallium,

chromium, silicon) in the PCBs.

The knowledge of computer components containing CRMs (with details on the

composition) would facilitate their identification by operators during EoL processing.

Together with requirements on the design to ease the dismantling, this type of labelling

of CRMs could increase the efficiency in the sorting of relevant components, addressing

them to the proper treatments and, ultimately, increase their recycling rates (Ardente and

Mathieux, 2014).

A detailed analysis of all the CRMs used in computers remains a challenge, because of the

large variety of CRMs present in several components. Manufacturers have difficulties in

collecting this information from suppliers, as typically the chemical composition of a

component or alloy is not specified to suppliers (Digitaleurope, 2017b). However, industry

already demonstrated the willingness to voluntarily provide information about some

CRMs (

) in an aggregated format (Digitaleurope, 2013). The next sections will focus on

two main CRMs that have been identified as relevant during the analysis of composition of

computers and EoL scenarios (Sections 2 and 3).

7.5.1.1 Recovery of cobalt

To implement more precise sorting and dedicated treatment of LIB according to their sub-

chemistry, an indication of the cobalt concentration in batteries is needed. For example,

elements such as iron and phosphorous from LFP batteries represent an obstacle for the

recovery of cobalt from high cobalt concentrates (LCO-type LIB). Thus, such polluting

elements need to be removed. However the removal process for such polluting elements

can increase the cost of the whole process. Therefore, the sorting of batteries according

to their sub-chemistry, as a preliminarily to further treatment, makes it possible to have

batches of waste which are richer in the concentration of the target metals (e.g. cobalt).

Compared to the treatment of diluted mixtures of different battery types, the treatment

of these concentrated batches is more feasible from both a technical and an economical

point of view. As hydrometallurgical processing focuses on selected materials, for instance

either cobalt and copper in the case of LCO batteries or copper and manganese for LMO

batteries, the loss of material groups can be minimised when batch-wise processing per

battery type is facilitated by labelling. The mass concentration allows for a better

assessment of the economic viability of the treatment, i.e. improves the precision of the

estimate of cobalt content and the material matrix (Accurec, 2016).

7.5.1.2 Recovery of rare earth elements

HDDs represent one of the main electronic components for containing certain rare earths.

As detailed in Section 2.4.2, HDDs contains NdFeB magnets which mainly contain

(

) The list of CRM is provided in:

http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52014DC0297&from=EN

(

) Indium volumes by display technology type.

123

neodymium (23-25 % in mass) and a few percentages of other rare earths (such as

dysprosium and praseodymium).

Compared to the total NdFeB production capacity, the recovery potential from HDDs is in

the 1-3 % range, however the recycling could be potentially increased if neodymium could

be traced from mine to material, product, and finally to waste (Sprecher et al., 2014a). It

is estimated that neodymium in magnets will be the largest source of recycled neodymium

until 2025 (München and Veit, 2017). The recycling of rare earths from HDDs is technically

feasible, if these components are properly extracted and sorted from other waste streams.

HDD from different computer types (notebook and desktop) can take the same recycling

route (München and Veit, 2017). In addition, Sprecher et al. (2014a) highlighted that

NdFeB magnets should be treated relatively close to the waste-collection and treatment

points, as shipping and handling large volumes of NdFeB magnets may be problematic

because of the very high magnetic strength.

The separation of HDDs and NdFeB magnets occurs either after waste shredding or manual

dismantling (Sprecher et al., 2014a). The former option, namely recycling through

shredding, results in a very significant (> 90 %) loss of NdFeB, which is mainly lost in the

ferrous fraction; after shredding, the neodymium must be leached out of the material and

then be reprocessed in almost the same manner as virgin material is processed (Sprecher

et al., 2014a). Neodymium liberated through shredding may also contaminate other

recyclable fractions (Sprecher et al., 2014a). This option is therefore regarded as less

efficient in terms of material recovery.

Vice versa, manual dismantling of HDDs proved to be much more efficient and have a

lower environmental impact (Sprecher et al., 2014a). Experimental measurement of the

efficiency of manual extraction of HDD from waste computers under current processes

resulted in around 35 % (Sprecher et al., 2014a). This percentage could be further

increased thanks to improved design of the product for the dismantling of the HDDs (as

in requirement of Section 6.1) or provision of information on the content and location of

neodymium (see e.g. Talens Peiró and Ardente (2015) for the enterprise-servers product

group). Neodymium from magnets can be then further recycled through ‘hydrogen

decrepitation’ process (Zakotnik et al., 2006) or by raising the temperature of the material

above its Curie temperature (312 °C) in order for it to lose its magnetic properties

(Dupont and Binnemans, 2015). The recycling of neodymium following these techniques

can reach values of efficiency of 90 % (Sprecher et al., 2014a), and significantly lower

environmental impacts (from 60 % to 90 % lower) compared to primary production

(Sprecher et al., 2014b).

7.5.2 Possible improvements

Measures to improve the recycling of neodymium and other rare earths from magnets

include the following.

 The declaration of the content of rare earths (e.g. the proposal for ecodesign

requirements for fans (

)).

 The provision of instructions for the dismantling (e.g. the requirement for the

dismantling of magnets in ventilation units (

)).

(

) According to the preparatory study of ventilation fans, it is proposed that manufacturers declare the weight

(if any) of the permanent magnets containing rare earths, in kg with 2 digits (e.g. ‘permanent magnets

2.12 kg’), on the nameplate and in the technical document (VHK, 2015).

(

) ‘The manufacturer’s free access website shall make available detailed instructions, inter alia, identifying the

required tools for the manual disassembly of permanent magnet motors, […] for the purpose of efficient

materials recycling […]’ (European Union, 2014).

124

 The potential labelling/marking of the components (e.g. the proposal for a QR code

on REE content developed by NSF (2015) for the environmental labelling enterprise

servers (

)).

Information about the presence of CRMs in computers could include: the content of cobalt

in batteries; the content and location of components containing rare earths (e.g.

neodymium, dysprosium, praseodymium in HDD magnets); the content and locations of

palladium in PCBs.

This information could be as follows.

 Voluntarily provided by manufacturers in an aggregated format to be agreed with

the recycling industry.

 Provided by manufacturers in the technical documentation needed to support the

ease of dismantling of key components (Section 7.1).

 Provided by manufacturers through specific optical label (e.g. QR codes), to be

developed in the future to report the content of CRMs. These optical labels could

be placed directly onto the component or, alternatively, on the computer back-

cover.

However, it is recognised that, to be effective and easily verifiable, the provision of

information on the presence of CRMs in computers requires a standardised format for such

communication, including for example, dedicated labelling. Standards under the

development within the European Mandate M/543 (European Commission, 2015b) could

serve the purpose, as those related to the ‘use and recyclability of CRMs to the EU’ and

the development of ‘documentation and/or marking regarding information relating to

material efficiency of the product’.

(

) ‘The manufacturer shall indicate the type of actuator/voice coil and spindle magnets in the product’s hard

disk drive on the external enclosure of the hard disk drive by means of a QR code. The QR code shall link

directly to the magnet type and location information on a publicly available database or the manufacturer’s

website in at least English. The QR code shall be printed in black on a white background if one or more of

the magnets contain neodymium. The QR code shall include a non-machine readable chemical symbol (Nd)’

(NSF, 2015).

125

7.6 Initial assessments of benefits/impacts

This section aims to present scenarios and impact assessments developed to enhance the

recyclability of notebooks. This section refers to all of the improvements proposed in this

chapter (Sections 7.1, 7.2, 7.3, 0 and 7.5), since these are produced as a result of

combined action dedicated to promoting more material-efficient treatments and are

therefore assessed together.

The potential benefits have been assessed based on the comparison of some reference

scenarios (Figure 34). In particular, it is assumed that without any specific product

measure on recyclability, waste notebooks will be treated in future analogously to the

current situation and following the processes described in Section 3.1.3 (defined as

‘business as usual — BaU’ scenarios). On the other hand, it is assumed that the proposed

actions will improve the economic viability of treatments which are more focused on

medium-depth manual dismantling compared to treatments based only on ‘mechanical

crushing and sorting’ after depollution. Moreover, a better design for recycling of

notebooks could increase the separation of valuable components and the recycling rates

of materials within these ‘improved scenarios’. These two parameters (i.e. the flow of

waste treated in the different scenarios and the efficiency of the different recycling

treatments) are assumed to be affected by recycling improvements and this is reflected in

the modelling of the following assessment.

In particular, the considered reference scenarios are as follows.

— Business as Usual (BaU) scenario: this reflects the base-case EoL treatments for

notebooks, as described in Section 3.1.3. In particular, it is assumed that these

scenarios are equally representative of EU treatments, with 50 % of the waste flows

processed with depollution and mechanical crushing and sorting (BaU1), while

another 50 % is processed according to “depollution, medium-depth dismantling

and subsequent mechanical crushing and sorting” (BaU2). Compared to BaU1, the

BaU2 scenario is characterised by higher recycling rates of batteries, PCB (

storage systems and ODD, thanks to the more careful manual dismantling during

the depollution and the following dedicated recycling.

— Improved scenario: moderate (named as scenario I.1 in Figure 34). These

scenarios assume that, thanks to the proposed improvement actions, the flow of

waste notebooks processed through “depollution, medium-depth manual

dismantling and mechanical crushing” would increase compared to the BaU

scenario. This can be justified with the reduction in effort to locate and dismantle

relevant components and the consequent reduction in labour costs. The

investigated actions would grant an increased battery separation for waste flow

entering the different treatments (i.e. I.1.1 and I.1.2 in Figure 34), and higher

recycling rates for the waste flows treated by “depollution, medium-depth manual

dismantling and mechanical crushing” (I.1.2 in Figure 34).

— Improved scenario: high (I.2 in Figure 34). This scenario is analogous to the

previous one (I.1) with the difference that material-efficiency actions are expected

to produce higher benefits, in terms both of higher flows of waste treated by

“depollution, medium-depth manual dismantling and mechanical crushing” (I.2.2

in Figure 34) and higher separation and recycling rates of components and

materials. In addition, it is also expected that the waste treated through

“depollution, mechanical crushing & sorting” (I.2.1 in Figure 34) would achieve a

higher extraction rate of the batteries.

For the assessment it is considered that there will be 41.7 million notebook sold in 2020

(Section 2.1) that will reach their EoL in 2025. Assumptions about the composition of the

notebooks are as follows.

— BoM of notebook components as detailed in Section 2.3.2.

(

) Button cell batteries are considered to be further separated from PCBs manually dismantled.

126

— Average composition of batteries as detailed in Section 2.3.5 (Table 17).

— Storage systems are assumed to be constituted 50 % by SSD and 50 % by HDD

(as discussed in Section 2.1, market projections to 2020). The content of

neodymium and other rare earths is 30 % of the magnets in HDD (Prakash et al.,

2014).

Table 39 illustrates the average recycling rates of different metals from notebook PCBs

separated for dedicated treatments (derived from Chancerel and Marwede (2016)). The

same recycling rates of Table 39 are assumed for SSD (

The assumed recycling rates of metals from batteries extracted and separately treated

are: cobalt 90 % (Chancerel and Marwede, 2016); nickel 62 % and copper 90 % (Wang

et al., 2014); lithium 50 % (

). Recycling rate of rare earths (neodymium and dysprosium)

from HDD magnets extracted and separately treated is assumed 90 % (Sprecher et al.,

2014a). ODDs that are separated are assumed to be dismantled to extract the PCB, while

the remaining parts are treated by mechanical crushing and sorting. Recycling rates of

other components materials are derived from IEC (2012). All the assumptions on waste

flows and recycling rates are summarised in Figure 34.

Table 39 — Average recycling rates of different materials from PCBs separated for recycling

(source: Chancerel and Marwede, 2016)

Materials in PCB

Recycling rate

Materials in PCB

Recycling rate

95 %

80 %

0 %

95 %

0 %

75 %

95 %

0 %

50 %

80 %

SiO

0 %

CaO

0 %

MgO

0 %

95 %

NaO

0 %

50 %

0 %

80 %

90 %

(

) SSD have a structure similar to that of PCB and are assumed to be collected together with PCBs and recycled

in the same facilities.

(

) The recycling of lithium, although technically feasible with high efficiency (50-90 %) (Kushnir, 2015; Wang

et al., 2014), is still not largely developed in the EU. Currently, a plant for the recycling of lithium has been

established in France. Similar plants could be set in the EU, especially assuming in the next future a large

growth of the amount of waste batteries sorted for recycling.

127

Figure 34 — References scenarios for the calculation of the benefits related to material-

efficiency actions to enhance the recyclability of notebooks.

The results of the assessment are illustrated in Table 40. In particular, benefits have been

estimated in terms of additional recycled materials obtained by moving from the BaU

scenarios through the moderately improved scenario “I.1” and the highly improved

scenario “I.2” respectively.

Recycling scenarios of Notebooks

BaU1) Depollution, mechanical

crushing & sorting

BaU2) Depollution, dismantling,

mechanical crushing & sorting

• Battery separation for recycling : 70%

• PCB separation for recycling: 50%

• Storage systems and ODD separation

for recycling: 10%

(Detail of recycling rates of materials in PCB

and batteries provided in Tables)

Recycling rates of main materials:

• Copper : 85%

• Steel: 94%

• Aluminium: 91%

• Rare earth elements: 90%

Metals in batteries:

• Cobalt: 90%

• Lithium: 50%

• Nickel: 62%

• Copper: 90%

Plastics:

• PP / PE (without additives): 90%

• HIPS (without additives): 83%

• ABS (without additives): 74%

• Other plastics: 0%

Business as usual

I.1.1) Depollution, mechanical

crushing & sorting

I.1.2) Depollution, dismantling,

mechanical crushing & sorting

Battery separation for

recycling : 80%

Recycling of materials and

components: (as BaU1)

Battery separation for

recycling : 95%

PCB separation: 85%

Storage systems and ODD

separation: 50%

Recycling rates of plastics

parts (>100g) of PP, PE,

ABS, HIPS, PC, PMMA (with

or without additives): 94%

Recycling of other

materials and components

(as BaU2)

I.1) Improved scenario (moderate)

I.2.1) Depollution, mechanical

crushing & sorting

I.2.2) Depollution, dismantling,

mechanical crushing & sorting

I.2) Improved scenario (high)

Battery separation for

recycling : 90%

Recycling of other

materials and

components: (as BaU1)

Battery separation for

recycling: 99%

PCB separation for

recycling: 95%

Storage systems and ODD

separation: 85%

Recycling rates of plastics

parts (>100g) of PP, PE, ABS,

HIPS, PC, PMMA (with or

without additives): 94%

Recycling of other materials

and components (as BaU2)

• Battery separation for recycling: 90%

• PCB separation for recycling: 75%

• Storage systems and ODD separation

for recycling: 40%

(Detail of recycling rates of materials in

PCB and batteries provided in Tables)

Recycling rates of main materials:

• Copper: 95%

• Steel, aluminium, neodymium and

metals in batteries (as BaU1)

• Plastics: (as BaU1)

128

Table 40 — Estimated benefits thanks to enhanced recyclability of notebooks.

Materials

Amount of additional recycled

materials [tonne]

Moderate

improved

scenario (I.1)

High

improved

scenario (I.2)

Plastics (from various components)

8 066.9

10 755.9

Copper (from various components)

318.3

763.0

Silver (from PCBs)

2.6

8.5

Gold (from PCBs)

0.2

0.5

Bismuth (from PCBs)

0.1

0.3

Nickel (from PCBs)

9.7

22.2

Lead (from PCBs)

8.0

21.1

Palladium (from PCBs)

0.2

0.5

Tin (from PCBs)

12.3

32.3

Zinc (from PCBs)

8.5

22.6

Bromine (from PCBs)

18.0

47.2

Antimony (from PCBs)

2.5

6.5

Neodymium (from HDDs magnets)

1.9

7.0

Cobalt (from batteries)

74.8

144.5

Lithium (from batteries)

8.7

16.8

Nickel (from batteries)

23.8

46.0

Comparing the estimated benefits in Table 40 with current amounts of recycling materials

it is observed that:

— In 2012, 10.9 t of palladium was recycled in the EU28 (BIO by Deloitte, 2015). The

analysed scenario would grant an additional recycling of about 0.2 t of palladium

(scenario I.1) in the future, equivalent to 1.8 % of the current recycling amount,

and achieve up to 0.5 t of additional recycled palladium (scenario I.2) equivalent

to 4.7 % of the current recycling amount.

— In 2012, 6.3 kt of cobalt was recycled in the EU28 (BIO by Deloitte, 2015). The

analysed scenario would grant an additional recycling of 74.8 t of cobalt (scenario

I.1) in the future, equivalent to 1.2 % of the current recycling, and achieve up to

144.5 t of additional recycled cobalt (scenario I.2) equivalent to 2.3 % of the

current recycling amount.

— In 2013, 14 t of neodymium was recycled in the EU28 (BIO by Deloitte, 2015). This

implies that a large share of neodymium in WEEE is currently lost. The analysed

scenario would grant an additional recycling of 1.9 t of neodymium (scenario I.1)

in the future, equivalent to 13.5 % of the current recycling, and to achieve

optimistically up to 7 t of additional recycled neodymium (scenario I.2), equivalent

to about 49.7 % of the currently recycled amount.

— Lithium from batteries is largely not recycled. According to BIO by Deloitte (2015),

the lithium currently recycled amounts to 16 t. The analysed scenario would

improve the battery extraction in the future, and would promote the recovery of

lithium as well. The amount of additional lithium potentially recycled ranges from

129

8.7 t (scenario I.1) up to 16.8 t (scenario I.2). These amounts are equivalent to

around 50-100 % of the current recycled masses.

— Compared to the current recycling of antimony in the EU (9.7 kt (BIO by Deloitte,

2015)) the improvements would also allow moderate minor benefit in terms of

additional antimony recycled (up to 6.5 t, equivalent to 0.1 % of the current

recycling).

Finally, the proposed requirements will also contribute to increase the amounts of recycled

plastics (8-10 kt of additional plastics), copper (318-763 t) and precious metals (0.2-0.5 t

of gold; 0.2-0.5 t of palladium; 2.6-8.5 t of silver).

The previous estimated benefits are based on the assumption that all the waste notebooks

will be properly collected and treated in the EU at EoL. However, there is evidence of large

amounts of waste electronics that are illegally exported or improperly collected and treated

(e.g. disposed of into trash bins) (Huisman et al., 2015). Assuming a loss of 26 % of the

flow of waste notebooks, potential reduced benefits have been estimated (Table 41).

Table 41 — Revised benefits due to the potential strategies to enhance the recyclability of

notebooks (based on a reduced amount of waste properly collected).

Materials

Amount of additional recycled

materials [t]

Moderate

improved

scenario (I.1)

High

improved

scenario (I.2)

Plastics (from various components)

5 969.5

7 959.4

Copper (from various components)

239.7

609.3

Silver (from PCBs)

2.0

6.6

Gold (from PCBs)

0.1

0.4

Bismuth (from PCBs)

0.1

0.2

Nickel (from PCBs)

7.3

17.7

Lead (from PCBs)

6.1

17.5

Palladium (from PCBs)

0.1

0.4

Tin (from PCBs)

9.4

26.8

Zinc (from PCBs)

6.5

18.6

Bromine (from PCBs)

13.7

39.1

Antimony (from PCBs)

1.9

5.4

Neodymium (from HDDs magnets)

1.4

5.1

Cobalt (from batteries)

55.6

107.8

Lithium (from batteries)

6.5

12.5

Nickel (from batteries)

17.7

34.3

130

7.6.1 Potential benefits for other product categories

The assessment of the potential benefits related to design for dismantling strategies is

more difficult and uncertain for computers other than notebooks.

In the case of tablets, a small amount of waste currently reaches recycling facilities. Still

the dismantling process for tablets is under development and refinement by recyclers.

However, Section 3.1.4 analysed some criticalities during the processing of waste tablets,

mainly related to the extraction of the batteries and PCBs. Design for recycling strategies

as proposed in this section could contribute to simplifying the pre-processing of tablets

and increase the material efficiency of the recycling processes overall, in terms of higher

quantity/quality of materials separated for recycling.

For the assessment of the benefits of design for dismantling strategies for tablets it is

roughly assumed to achieve similar improvements as discussed for notebooks. This implies

that the efficiency of sorting and processing of PCBs and batteries could increase by around

10-20 %. Considering the average BoM of tablet (as in Table 12), the average composition

of batteries (as in Table 11) and the average composition of batteries as for notebooks (as

in Table 17), and assuming the same recycling rates as in Figure 32, it is roughly estimated

that the additional amounts of recycled materials are: 30-60 t of cobalt, 4-7 t of lithium,

80-170 t of copper and 0.2-0.6 of various precious metals.

The process of dismantling and depolluting traditional desktop computers is instead well

established and no criticalities have been identified in our analysis. However, the market

of traditional desktop computers is estimated to continue declining, while the market

shares of new types of desktops (e.g. mini desktops) are expected to grow in the next

future. These new desktops can pose some problems during recycling, especially due to

the very compact structure and the difficulties in extracting PCBs and batteries potentially

contained in the computer. Designing for recycling strategies as proposed in this section

could contribute to keeping the attention of manufacturers focused on the EoL aspects of

these desktops and to promote designing computers for recycling solutions that facilitate

their processing. However, due to the lack of information about the flows of mini desktops

sold and their BoM, it is not possible to quantify such benefits.

131

8 Conclusions

The research findings of this technical report can be grouped into two levels. We firstly

identified the so-called hot spots, namely the aspects of the personal-computer product

group that are relevant from a material-efficiency perspective. Hot spots include the

problems currently encountered by users (e.g. design features of the products that may

hinder reuse, disassembly or repair) and by EoL operators (e.g. during the collection and

treatment of computers). We then provided an analysis of potential actions aimed at

overcoming the current hot spots and at improving the material efficiency of the product

group, in particular enhancing durability, reusability, reparability and recyclability. Such

analysis aimed to identify material-efficiency aspects which can be relevant for the current

revision of the Ecodesign Regulation (EU) No 617/2013. This work has been carried out in

the period June 2016-September 2017, in parallel with the development of The

preparatory study on the review of Regulation No 617/2013.

Possible actions to improve material efficiency of personal computers were classified into

three levels and also according to the waste hierarchy set out by the European Commission

(Directive 2008/98/EC), in which waste prevention has the first priority, before preparing

for reuse and finally recycling. Among the strategies to close material loops, durability and

reusability are key material-efficiency aspects. As such, opportunities to eliminate or

reduce factors potentially leading to breakages or loss of performance of personal

computers have a high priority. A special focus was given to the content of raw materials

(including EU CRMs) in computers and how to increase the efficient use of these materials,

including material savings thanks to waste prevention and increased reuse and repair, and

strategies to improve material recycling at the EoL of the products.

Possible actions to improve waste prevention

Among the possible actions to improve waste prevention we discussed two strategies to

improve battery durability for mobile personal computers, one strategy for reducing the

need for unnecessary EPS, and one strategy to raise awareness about the durability of

computers, in particular the resistance to liquid ingress.

Battery durability resulted to be a key feature for users. A key indicator is represented by

the remaining charge capacity of the battery compared to the initial-charge capacity,

measured after a predefined number of charge/discharge cycles (e.g. 300 and 500 cycles).

This information could be provided to users to raise awareness of an important aspect

influencing the performance of the product. The assessment of this indicator can be based

on the existing procedures set out in Standard EN 61960. This strategy could also

represent a potential trigger for competition, as the performance of the battery could be

evaluated with common tests and conditions. It will also help to gather data for future

improvements (e.g. drawing up performance classes for grading battery performance).

For battery durability, it was also proposed to promote the use of algorithms to manage

the SoC of batteries while notebooks are in grid operation, a condition that may recur in

office environments, for example. This would prevent the battery remaining at a SoC of

100 % for long time, a condition that accelerates the ageing of the battery. In this case,

standards to support the development, implementation and verification of such algorithms

have to be developed.

EPS were addressed as they represent a significant percentage of the whole weight and

materials used for ICT (10-20 %), thus it is important to minimise the need for materials

and the impact on the environment. The action proposed is to declare, in a standardised

way, the presence/absence of the EPS within the packaging of the product, and the

compatibility with other devices. Such a proposal has the potential to trigger the

decoupling of EPS and devices and to inform users about how and when an EPS can be

reused. IEC TS 62700, Standard IEEE 1823 and Recommendation ITU-T L.1002 can be

used to develop standards for connectors and power specifications.

132

The last strategy aims to address the problem of products damaged by drops/falls or liquid

spillage. Components such as the keyboard, the display, display-cover (including frame

joints) and the casing of consumer notebooks are the components most prone to fail due

to drops/falls or liquid spillage. Manufacturers have the possibility to test waterproof

solutions for certain personal computers, and test their ingress protection (IP) according

to the Standard IEC 60529 — Degrees of protection provided by enclosures (IP code). The

action proposed, also in this case, is to declare in a standardised way the IP class.

Standardisation activities may be needed to further define existing (but not standardised)

tests, specifically relevant for the personal-computer product group. As an example, the

EU Ecolabel criteria for computers include a test on durability against water-spill

ingress (

100

). Such a test relies on the Standard IEC 60529 to establish the acceptance

conditions only and, therefore, standardisation activities are needed to univocally set out

testing conditions.

Possible actions to enhance repair and reuse

Among the possible actions to enhance repair and reuse we identified two strategies

related to the ease of disassembly of personal computers, and one option related to the

deletion of personal data in mass-storage systems.

Ease of disassembly is a crucial feature to improve the reusability of products, but also to

extend their lifetime. EU citizens would rather repair their goods than buy new ones, but

ultimately have to replace or discard their goods because they are discouraged by the cost

of repairs and the level of service provided. We proposed that ad hoc documentation could

be prepared not only to provide useful information to enhance repair and refurbishing of

EEE, but also for preparation for reuse of WEEE. Two levels of information were considered.

The first level involves professional repair operators, who can be provided with information

about the disassembly, replacement and reassembly operations needed for a set of

relevant components of personal computers (batteries, internal power-supply units,

display, data-storage memories, keyboards, etc.). The second level involves users, who

can be provided with clear and easy accessible information about the ease of disassembly

and replacement of batteries used in personal computers. In particular, users can be

informed on whether the replacement of the battery can be done by them, or by a

professional repair operator. In both cases, standards have to be developed to ensure that

the information is provided using a common format.

One major barrier to the reuse, repair and recycling of computers is data-privacy issues.

Desktop computers, notebooks and tablets regularly store sensitive and confidential data

for users and organisations. As a number of operating systems already include the option

to ‘factory reset’ the device, bringing the device into its original factory state, we propose

to preinstall a built-in functionality to ensure secure data deletion. Secure data deletion

means the effective erasure of all traces of existing data from storage media, overwriting

the data completely in such a way that access to the original data, or parts of them,

becomes unfeasible for a given level of effort. A number of existing national standards

(HMG IS Standard No 5 (UK), DIN 66399 (Germany), NIST 800-88r1 (US)) can be used

as basis to develop international standards on secure data deletion. The protection of

personal data contained in electronic product and components should be treated with

special care, and the need to take action on secure data deletion is also in line with the

principles of privacy and protection of personal data as set by the General Data Protection

Regulation (EU) 2016/679.

(

100

) The test shall be carried out two times. A minimum of 30 ml of liquid shall be poured evenly over the

keyboard of the notebook or onto three specific, separated locations, then actively drained away after a

maximum of 5 seconds, and the computer then tested for functionality after 3 minutes. The test shall be

carried for a hot and a cold liquid (European Commission, 2016).

133

Possible actions to improve recyclability

Finally, among the possible actions to enhance recyclability, we discussed five possible

strategies to improve the ease of dismantling of personal computers which are aiming to

facilitate the extraction and the identification of different components and materials.

The design for the dismantling of computers is a necessary condition to efficiently depollute

and recycle them. As such, computer components that are crucial for material-efficiency

aspects should be easily located and extracted, in order to be properly and efficiently

addressed in specific recycling treatments. Thus, the first strategy proposed targets for

the ease of dismantling of products, which can be proved and enhanced thanks to

comprehensive documentation provided by manufacturers. This documentation may

include the sequence of operations needed to access the key components, namely the

ones listed in the Annex VII of the WEEE directive. Such documentation would serve as

guidance for recyclers and as proof that key components can be accessed and dismantled.

At the time being, as stated for the actions to enhance repair and reuse, standards have

to be developed to ensure that the information is provided using a common format.

Another related strategy to promote recyclability of computers is to mark plastic parts in

order to recognise the type of plastic used and sort them correctly. Marking of plastic

should follow a standardised approach, such as that proposed by ISO 11469, and

standards of the series ISO 1043. Plastic marking can contribute to separation of plastics

with FRs during the manual dismantling, allowing their recycling at higher rates, however,

more-detailed information about the composition of the product (including detail of plastic

composition) can be beneficial for recyclers and it is also in line with the principles of the

WEEE directive. On such purpose, a third strategy is proposed, as regards the declaration

of flame-retardant content in plastics. The provision of information on the content of FRs

in plastic parts is a first step to contribute to the improvement of plastic recycling. Detailed

information about FRs content could be given in a more systematised way, for example

through the development of specific indexes. These indexes could support recyclers in

checking the use of FRs in computers and in developing future processes and technologies

suitable for plastics recycling. Moreover, these indexes could support policymakers in

monitoring the use of FRs in the products and, in the medium-long term, to promote

products that use smaller quantities of FRs.

In order to properly sort batteries by battery chemistry, a further action proposed is to

adopt a battery-recycle mark, which reduces the limits of current marking practices if

properly applied (visible, durable, legible and indelible). The identifiability of battery

chemistry would be enhanced by the use of different colours. Standardisation activities

are needed, even though they are currently ongoing for batteries of big dimensions. The

draft international Standard IEC 62902, titled ‘Secondary batteries: Marking symbols for

identification of their chemistry’, is currently being discussed. If approved, this draft

document may be the starting point for batteries with a volume of less than 900 cm³, as

in the case of personal computers.

Finally, we also discussed which type of information on the content of CRMs could be

helpful to improve the recycling efficiency. The provision of information could include: the

content of cobalt in batteries, the content and location of components containing rare

earths (e.g. neodymium, dysprosium, praseodymium in HDD magnets), and the content

and locations of palladium in PCBs. Having the knowledge of computer components

containing CRMs (with detail of the composition) would facilitate their identification by

operators during EoL processing. The provision of information on the content of CRMs

could increase the efficiency in sorting relevant components, addressing them with proper

recycling treatments and, ultimately, increase their recycling rates. It was recognised that,

to be effective and easily verifiable, this provision of information requires a standardised

format for such communication. Standards under development within the European

Mandate M/543 could serve the purpose, such as the ones related to the ‘use and

recyclability of CRMs to the EU’ and to the development of ‘documentation and/or marking

regarding information relating to material efficiency of the product’.

134

Concluding remark

Overall, this analysis identified precise possible actions for improving material-efficiency

performances of the product group. Furthermore, it also aimed at further stimulating the

discussions between industry and policymakers about industrial practices that will enable

a circular economy, focusing in particular on waste prevention, durability, reuse, repair

and recycling. Furthermore, already-available or under-development international

standards relevant for material-efficiency practices were identified and discussed. Further

discussions will set out whether and how these material-efficiency strategies can be

implemented in practice, also considering their verifiability by third parties (e.g. market-

surveillance authorities).

135

References

Accurec, 2016. Accurec — Interviews.

Accurec, 2015. ‘Comparative study on li-ion battery recycling technologies’, H2020

CloseWEEE project, unpublished.

AEA, 2010. Building on the eco-design directive, EuP group analysis (I) ENTR Lot 3 Sound

and imaging equipment. Task 1-7 report, AEA/ED45386/Issue 1.

Aladeojebi, T. K., 2013. ‘Planned obsolescence’, International Journal of Scientific &

Engineering Research, Vol. 4, pp. 1504-1508.

Andrae, A. S. G., 2016. ‘Life-cycle assessment of consumer electronics: A review of

methodological approaches’, IEEE consumer electronics magazine, Vol. 5, pp. 51-60.

(doi:10.1109/MCE.2015.2484639)

Andrae, A. S. G., Vaija, M. S., 2014. ‘To which degree does sector specific standardization

make life cycle assessments comparable? — The case of global warming potential of

smartphones’, Challenges, Vol. 5(2), pp. 409-429. (doi:10.3390/challe5020409)

Apple, 2016. Maximizing battery life and lifespan

(http://www.apple.com/batteries/maximizing-performance/).

Ardente, F., Mathieux, F., 2014. ‘Identification and assessment of product’s measures to

improve resource efficiency: the case-study of an energy using product’, Journal of cleaner

production, Vol. 83, pp. 126-141. (doi:10.1016/j.jclepro.2014.07.058)

Ardente, F., Mathieux, F., 2012a. Integration of resource efficiency and waste

management criteria in European product policies - Second phase. Report n° 2: Application

of the project’s methods to three product groups. (doi:10.2788/75910)

Ardente, F., Mathieux, F., 2012b. Integration of resource efficiency and waste

management criteria in European product policies — Second phase. Report n° 1. Analysis

of durability. (doi:10.2788/72577)

Ardente, F., Mathieux, F., Talens Peiró, L., 2016. Revision of methods to assess material

efficiency of energy related products and potential requirements. (doi:10.2788/517101)

Ardente, F., Mathieux, F., Talens Peirò, L., 2013. Environmental footprint and material

efficiency support for product policy. Report on benefits and impacts/costs of options for

different potential material efficiency requirements for electronic displays.

(doi:10.2788/28569)

Arushanyan, Y., Ekener-Petersen, E., Finnveden, G., 2014. ‘Lessons learned — Review of

LCAs for ICT products and services’. Computers in industry, Vol. 65, pp. 211-234.

(doi:10.1016/j.compind.2013.10.003)

ASUS, 2016. Progressive from every angle, touch and user experience

(https://www.asus.com/Microsite/commercial/ASUSPRO/progressive.html).

Bangs, C., Meskers, C., Van Kerckhoven, T., 2016. ‘Trends in electronic products — the

canary in the urban mine?’, Electronics Goes Green 2016+, IEEE.

Battery University, 2016 a. How to prolong lithium-based batteries

(http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries).

Battery University, 2016b. BU-205: Types of lithium-ion

(http://batteryuniversity.com/learn/article/types_of_lithium_ion).

Berger, R., 2012. ‘The lithium-ion battery value chain’ International conference on energy

and automotive technologies, Istanbul.

BIO by Deloitte, ICF-GHK and SERI 2015. Study on socioeconomic impacts of increased

reparability.

Bio Intelligence Service, 2007. Preparatory studies for eco-design requirements of EuPs.

136

Bocken, N. M. P., de Pauw, I., Bakker, C., Grinten, B. van der, 2016. ‘Product design and

business model strategies for a circular economy’, Journal of industrial and production

engineering, Vol. 33, Iss. 5. (doi:10.1080/21681015.2016.1172124)

Bölling, K., 2016. Personal communication on 06.11.2016.

Bombardier, 2010. Marking of plastics, rubbers and thermoplastic elastomers.

Broussely, M., Biensan, P., Bonhomme, F., Blanchard, P., Herreyre, S., Nechev, K.,

Staniewicz, R.J., 2005. ‘Main aging mechanisms in Li ion batteries’, Journal of power

sources, Vol. 146, Iss. 1-2, pp. 90-96. (doi:10.1016/j.jpowsour.2005.03.172)

Chancerel, P., Mahlitz, P., Chanson, C., Binnemans, P., Huisman, J., Guzman Brechu, M.,

Rotter, V.S., Nissen, N.F., Lang, K.-D., 2016. ‘Stocks and flows of critical materials in

batteries: data collection and data uses’, Electronics Goes Green 2016+, IEEE.

Chancerel, P., Marwede, M., 2016. Feasibility study for setting-up reference values to

support the calculation of recyclability/recoverability rates of electr(on)ic products.

(doi:10.2788/901715)

Chancerel, P., Marwede, M., Nissen, N.F ., Lang, K. -D., 2015. ‘Estimating the quantities

of critical metals embedded in ICT and consumer equipment’, Resources, conservation and

recycling, Vol. 98, pp. 9-18.

Chancerel, P., Meskers, C. E. M., Hagelüken, C., Rotter, V. S., 2009. ‘Assessment of

precious metal flows during preprocessing of waste electrical and electronic equipment’,

Journal of industrial ecology, Vol. 13, pp. 791-810. (doi:10.1111/j.1530-

9290.2009.00171.x)

Chancerel, P., Rotter, V. S., Ueberschaar, M., Marwede, M., Nissen, N. F., Lang, K. -D.,

2013. ‘Data availability and the need for research to localize, quantify and recycle critical

metals in information technology, telecommunication and consumer equipment’, Waste

management and research, Vol. 31, pp. 3-16. (doi:10.1177/0734242X13499814)

Clemm, C., Mählitz, P., Schlösser, A., Rotter, V. S., Lang, K. -D., 2016. ‘Umweltwirkungen

von wiederaufladbaren Lithium-Batterien für den Einsatz in mobilen Endgeräten der

Informations- und Kommunikationstechnik (IKT)’, UBA, Texte 52/2016.

Consumentenbond, 2015. Laptopaccu’s steeds lastiger zelf te vervangen

(https://www.consumentenbond.nl/laptop/laptopaccu-steeds-lastiger-zelf-te-vervangen)

(accessed 19.04.17).

Cucchietti, F., Giacomello, L., Griffa, G., Vaccarone, P., Tecchio, P., Bolla, R., Bruschi, R.,

D’Agostino, L., 2011. ‘Environmental benefits of a universal mobile charger and energy-

aware survey on current products’ 2011 IEEE 33rd International telecommunications

energy conference (INTELEC). IEEE, pp. 1-9. (doi:10.1109/INTLEC.2011.6099888)

Daoud, D., 2010. The business case for ruggedized PCs.

Dawson, R. B., Landry, S. D., 2005. ‘Recyclability of flame retardant HIPSs, PC/ABS, and

PPO/HIPS used in electrical and electronic equipment’, Proceedings of the 2005 IEEE

international symposium on electronics and the environment, 2005. IEEE, pp. 77-82.

(doi:10.1109/ISEE.2005.1436998)

Defra, 2006. Battery waste management life cycle assessment, Department for

Environment, Food and Rural affairs (Defra) (UK).

Deng, L., Williams, E.D., 2011. ‘Functionality versus ‘typical product’ measures of

technological progress’, Journal of industrial ecology, Vol. 15, pp. 108-121.

(doi:10.1111/j.1530-9290.2010.00306.x)

Directorate-General for Internal Market, Industry, Entrepreneurship and SMEs, 2016.

Critical raw materials (https://ec.europa.eu/growth/sectors/raw-materials/specific-

interest/critical_en) (accessed 18.10.16).

137

Digitaleurope, 2017 a. The contribution of the digital industry to repair, refurbishment and

remanufacturing in a circular economy.

Digitaleurope, 2017b. Industry reaction on CRM information requirements following the

Dutch and German comments on the consultation forum on servers (LOT 9).

Digitaleurope, 2014. Trans-boundary movements of waste vs used goods 4.

Digitaleurope, 2013. Position paper on dismantling, indium declaration and recyclability

for TVs and displays.

Dimitrova, G., 2012. Impact of innovations in electronic equipment and components on

their reuse and recycling. University of Natural Resources and Life Sciences, Vienna.

Dodd, N., Vidal-Abarca Garrido, C., Gama Caldas, M., Graulich, K., Bunke, D., Groß, R.,

Liu, R., Manhart, A., Prakash, S., 2016. Revision of the EU green public procurement (GPP)

criteria for computers and monitors — technical report — final criteria.

(doi:10.2791/027791)

Dodd, N., Vidal-Abarca Garrido, C., Wolf, O., Graulich, K., Bunke, D., Groß, R., Liu, R.,

Manhart, A., Prakash, S., 2015. Revision of the European ecolabel criteria for personal,

notebook and tablet computers. (doi:10.2791/780423)

Dodd, N., Wolf, O., Graulich, K., Groß, R., Liu, R., Manhart, A., Prakash, S., 2014.

Development of European ecolabel and green public procurement criteria for personal and

notebook computers, technical report task 1, scope and definitions.

Dupont, D., Binnemans, K., 2015. ‘Recycling of rare earths from NdFeB magnets using a

combined leaching/extraction system based on the acidity and thermomorphism of the

ionic liquid [Hbet][Tf2N]’, Green chemistry, Vol. 17, pp. 2150-2163.

doi:10.1039/C5GC00155B

Dwivedy, M., Mittal, R.K., 2010. ‘Future trends in computer waste generation in India’,

Waste Management, Vol. 30, pp. 2265-2277. (doi:10.1016/j.wasman.2010.06.025)

Ecoinvent, 2013. Ecoinvent database (http://www.ecoinvent.ch/) (accessed 23.4.15).

EERA, 2016. EERA position paper. Netherlands.

EGG 2016+, 2016. Electronics Goes Green 2016+ Closing session.

Ellen MacArthur Foundation, 2016. Empowering repair.

Elo, K., Karlsson, J., Lydebrant, K., Sundin, E., 2009. Automation of plastic recycling — A

case study. Sapporo (Japan).

EN ISO 1043-1, 2002. Plastics. Symbols and abbreviated terms. Basic polymers and their

special characteristics.

EN ISO 1043-4+A1:2016, 1999. Plastics. Symbols and abbreviated terms. Flame

retardants.

EN TS 16524, 2013. Mechanical products. Methodology for reduction of environmental

impacts in product design and development.

EuRIC, 2017. Position on eco-design for displays: resource efficiency requirements.

EuRIC, 2016 a. EuRIC concrete proposals ecodesign for WEEE.

EuRIC, 2016b. Call for concrete actions to implement Article 15 of the WEEE directive on

information for treatment facilities.

European Commission, 2016. Commission Decision (EU) 2016/1371 of 10 August 2016

establishing the ecological criteria for the award of the EU Ecolabel for personal, notebook

and tablet computers.

European Commission, 2015 a. COM(2015) 614 final. Communication from the

Commission to the European Parliament, the Council, The European Economic and Social

138

Committee and the Committee of the Regions. Closing the loop — An EU action plan for

the circular economy.

European Commission, 2015b. M/543 C(2015) 9096 Commission implementing decision

of 17.12.2015 on a standardisation request to the European standardisation organisations

as regards ecodesign requirements on material-efficiency aspects for energy-related

products in support of the imp. doi:10.1017/CBO9781107415324.004

European Commission, 2014 a. Critical raw materials profiles. Report of the ad hoc working

group on defining critical raw materials. http://bookshop.europa.eu/ No 388.

doi:10.2779/1 482

European Commission, 2013. COM(2013) 123 final — Green paper on a European strategy

on plastic waste in the environment. doi:COM(2013) 123 final.

European Commission, 2009. Press release IP/09/1049: Commission welcomes industry’s

commitment to provide a common charger for mobile phones

(http://europa.eu/rapid/press-release_IP-09-1049_en.htm?locale=en) (accessed

29.07.16).

European Union, 2014. Commission Regulation (EU) No 1253/2014 of 7 July 2014

implementing Directive 2009/125/EC of the European Parliament and of the Council with

regard to ecodesign requirements for ventilation units. Official Journal of the European

Union.

European Union, 2012. Directive 2012/19/EU of the European Parliament and of the

Council of 4 July 2012 on waste electrical and electronic equipment (WEEE).

European Union, 2011. Directive 2011/65/EU of the European Parliament and of the

Council of 8 June 2011 — RoHS. Official Journal of the European Union.

(doi:10.1017/CBO9781107415324.004)

European Union, 2008. Directive 2008/98/EC of the European Parliament and of the

Council of 19 November 2008 on waste and repealing certain directives. LexUriServ.

(doi:2008/98/EC.; 32008L0098)

FINsix®, 2016. DARTTM The world’s smallest laptop charger®

(https://finsix.com/shop/dart/) (accessed 9.12.16).

Fisher, T., 2015. Data sanitization methods. A list of software based data sanitization

methods.

Forsa, 2013. Meinungen zu Umweltaspekten bei Computern.

Gabriel, R., 2015. End-of-life scenarios created in WF-RepTool.

Goosey, M., Kellner, R., 2002. A scoping study end-of-life printed circuit boards.

Greenpeace, 2017. How repairable is your mobile device? A product guide to best-selling

smartphones, tablets and laptops.

Grzesik-Wojtysiak, K., Kukliński, G., 2013. ‘Screening life cycle assessment of a laptop

used in Poland’. Environment protection engineering, Vol. 39 (3).

GSMA, 2009. Mobile’s green manifesto.

Hand, C., 2013. Dealing with waste lithium batteries | Croner-i

(https://app.croneri.co.uk/feature-articles/dealing-waste-lithium-batteries-0) (accessed

13.9.16).

Hennies, L., Stamminger, R., 2016. An empirical survey on the obsolescence of appliances

in German households. Resources, conservation and recycling, Vol. 112, pp. 73-82.

(doi:10.1016/j.resconrec.2016.04.013)

Hintermann, R., Fassnacht, C., 2008. Leitfaden zum Sicheren Datenlöschen.

139

Hischier, R., Wäger, P. A., 2015. ‘The Transition from desktop computers to tablets: a

model for increasing resource efficiency?’ ICT Innovations for Sustainability.

Houlihan, K., 2013. A streamlined life cycle assessment of Mobile computing

(laptop/slate). Report 20/11/2013.

HP, 2016. Product end-of-life disassembly instructions. HP ProDesk 600 G1 Desktop Mini

PC.

(http://www.hp.com/hpinfo/globalcitizenship/environment/productdata/Countries/_Multi

Country/disassembly_deskto_2014516234519169.pdf)

HP, 2015. Technical white paper testing the business ruggedness and reliability of HP

business PCs.

HP Inc., n.d. HP notebook PCs — understanding lithium-ion and smart battery technology

| HP® customer support (http://support.hp.com/us-

en/document/c00596784#c00596784_warranty) (accessed 13.9.16).

Huisman, J., Botezatu, I., Herreras, L., Liddane, M., Hintsa, J., Luda di Cortemiglia, V.,

Leroy, P., Vermeersch, E., Mohanty, S., van den Brink, S., Ghenciu, B., Dimitrova, D.,

Nash, E., Shryane, T., Wieting, M., Kehoe, J., Baldé, C. P., Magalini, F., Zanasi, A., Ruini,

F., Bonzio, A., 2015. Countering WEEE illegal trade summary report, market assessment,

legal analysis, crime analysis and recommendations roadmap, Unu. Lyon, France.

(doi:978-92-808-4560-0)

Huisman, J., van der Maesen, M., Eijsbouts, R. J. J., Baldé, C. P., Wielenga, C. A., 2012.

The Dutch WEEE flows. Bonn, Germany.

IDC, 2016. Pay Now, save Later: The business case for rugged devices.

IDC, 2010. White paper — The business case for ruggedised PCs.

IEC, 2017. Area: 482: Primary and secondary cells and batteries

(http://www.electropedia.org/iev/iev.nsf/index?openform&part=482).

IEC/TR 62635, 2015. IEC/TR 62635:2015 Guidelines for end-of-life information provided

by manufacturers and recyclers and for recyclability rate calculation of electrical and

electronic equipment.

IEC/TS 62700, 2014. IEC/TS 62700:2014 DC power supply for notebook computers.

IEC 62902 draft, 2017. International Electrotechnical Commission (2017) — 62902 Ed.1.0:

Secondary batteries: Marking symbols for identification of their chemistry.

IEEE 1874, 2013. IEEE Standard for documentation schema for repair and assembly of

electronic devices. (doi:10.1109/IEEESTD.2014.6712032)

IEEE Std 1823, 2015. IEEE Standard for universal power adapter for mobile devices.

Investing News, 2016. 6 types of lithium-ion batteries

(http://investingnews.com/daily/resource-investing/energy-investing/lithium-

investing/6-types-of-lithium-ion-batteries/)

ISO 11469, 2000. Plastics — Generic identification and marking of plastics products.

ITU-T L.1002, 2016. Recommendation ITU-T L.1002 External universal power adapter

solutions for portable information and communication technology devices.

Kahhat, R., Poduri, S., Williams, E., 2011. Bill of Attributes (BOA) in life cycle modeling of

laptop computers. White Paper #103.

Kasulaitis, B. V., Babbitt, C. W., Kahhat, R., Williams, E., Ryen, E. G., 2015. ‘Evolving

materials, attributes, and functionality in consumer electronics: Case study of laptop

computers’, Resources, conservation and recycling, Vol. 100, pp. 1-10.

(doi:10.1016/j.resconrec.2015.03.014)

140

Knack, O., 2016. 5 common tablet PC defects and how to address them

(http://www.globalsources.com/NEWS/SIC-5-common-tablet-pc-defects-and-how-to-

address-them.htm) (accessed 16.9.16).

Krüger, A., 2016. Personal communication on 14.11.2016.

Kushnir, D., 2015. Lithium-ion battery recycling technology 2015 current state and future

prospects. Gothenburg, Sweden.

Lenovo, 2016. Beyond mil-spec [WWW Document].

Malmodin, J., Lundén, D., Moberg, Å., Andersson, G., Nilsson, M., 2014. ‘Life cycle

assessment of ICT’, Journal of industrial ecology, Vol. 18, pp. 829-845.

(doi:10.1111/jiec.12145)

Mathieux, F., Froelich, D., Moszkowicz, P., 2008. ‘ReSICLED: a new recovery-conscious

design method for complex products based on a multicriteria assessment of the

recoverability’, Journal of cleaner production, Vol. 16, pp. 277-298.

doi:10.1016/j.jclepro.2006.07.026

München, D.D., Veit, H.M., 2017. ‘Neodymium as the main feature of permanent magnets

from hard disk drives (HDDs)’, Waste management, Vol. 61, pp. 372-376.

(doi:10.1016/j.wasman.2017.01.032)

Nissen, N. F., Stobbe, L., Oerter, M., Scheiber, S., Schlösser, A., Schischke, K., Lang, K.-

D., 2013. ‘Design features of current tablet computers to facilitate disassembly and repair’,

Proceedings of the 8th Ecodesign 2013, Jeju, Korea.

Nnorom, I. C., Osibanjo, O., 2008. ‘Overview of electronic waste (e-waste) management

practices and legislations, and their poor applications in the developing countries’,

Resources, conservation and recycling, Vol. 52, pp. 843-858.

(doi:10.1016/j.resconrec.2008.01.004)

Peeters, J. R., Vanegas, P., Tange, L., Van Houwelingen, J., Duflou, J. R., 2014. ‘Closed

loop recycling of plastics containing flame retardants’. Resources, conservation and

recycling, Vol. 84, pp. 35-43. (doi:10.1016/j.resconrec.2013.12.006)

Peeters J. R., Tecchio P., Ardente F., Vanegas P., Coughlan D., Duflou J. 2018. eDIM:

further development of the method to assess the ease of disassembly and reassembly of

products - Application to notebook computers; EUR 28758 EN; doi:10.2760/864982

Polverini D., Ardente F., Sanchez I., Mathieux F., Tecchio P., Beslay L. Resource efficiency,

privacy and security by design: a first experience on enterprise servers and data storage

products triggered by a policy process. Computer and Security, 2017 (doi:

https://doi.org/10.1016/j.cose.2017.12.001)

Powel, J., 2002. Large volumes of electronics scrap are shredded before recycling, and

this booming trend has many new twists and turns.

Prakash, S., Dehoust, G., Gsell, M., Schleicher, T., Stamminger, R., 2016 a. Einfluss der

Nutzungsdauer von Produkten auf ihre Umweltwirkung: Schaffung einer

Informationsgrundlage und Entwicklung von Strategien gegen „Obsoleszenz’, Öko-Institut

eV. (doi:10.1017/CBO9781107415324.004)

Prakash, S., Dehoust, G., Gsell, M., Schleicher, T., Stamminger, R., 2016b. Einfluss der

Nutzungsdauer von Produkten auf ihre Umweltwirkung: Schaffung einer

Informationsgrundlage und Entwicklung von Strategien gegen „Obsoleszenz’ (unpublished

annex).

Prakash, S., Köhler, A., Liu, R., Stobbe, L., Proske, M., Schischke, K., 2016c. ‘Paradigm

shift in green IT — Extending the life — Times of computers in the public authorities in

Germany’, Electronics Goes Green 2016+, pp. 1-7.

141

Prakash, V., Sun, Z.H.., Sietsma, J., Yang, Y., 2014. ‘Electrochemical recovery of rare

earth elements from magnet scraps ‐ a theoretical analysis’, ERES2014: 1st European

Rare Earth Resources Conference|Milos|04‐ 07/09/2014.

prEN 50614, 2016. Draft standard for comments (general, techncial, editorial) — draft

developed by CLC/TC 111X-WK07 — Requirements for the preparation for reuse of waste

electrical and electronic equipment.

Private communications, 2017. Private communications and interviews with repair

operators, refurbishers and stakeholders.

Private communications, 2016. Private communications with stakeholders.

Rahimifard, S., Abu Bakar, M.S., Williams, D.J., 2009. ‘Recycling process planning for the

end-of-life management of waste from electrical and electronic equipment’, CIRP Annals

— Manufacturing technology. pp. 5-8. (doi:10.1016/j.cirp.2009.03.080)

RAL GmbH, 2013. Basic criteria for award of the environmental label, mobile phones, RAL-

UZ 106.

Rasenack, K., Goldmann, D., 2014. ‘Herausforderungen des Indium-Recyclings aus LCD-

Bildschirmen und Lösungsansätze’, Thomé-Kozmiensky, K.J., Goldmann, D. (Eds.),

Recycling Und Rohstoffe, Band 7. VIVIS Verlag.

Recchioni, M., Ardente, F., Mathieux, F., 2016. Environmental footprint and material-

efficiency support for product policy. Feasibility study for a standardised method to

measure the time taken to extract certain parts from electrical and electronic equipment.

European Commission, JRC, Institute for Environment and Sustainability, Ispra.

(doi:10.2788/29866)

Ripperger, J., 2016. Personal communication on 28.10.2016.

Risk & Policy Analysts Limited, 2014. Study on the impact of the memorandum of

understanding (mou) on harmonisation of chargers for mobile telephones and to assess

possible future options.

Rotter, V. S., Flamme, S., Götze, R., Ueberschaar, M., 2012. ‘Thermodynamische

Herausforderung bei Recycling von Nebenmetallen’, Thomé-Kozmiensky, K. J., Goldmann,

D. (Eds.), Recycling Und Rohstoffe, Band 5. TK Verlag, Neuruppin.

RREUSE, 2013. Investigation into the repairability of domestic washing machines,

dishwashers and fridges.

Samsung, 2015. Light, but Durable: Reliability Testing of Samsung Notebooks [WWW

Document].

Sarkis, J., 2001. Greener manufacturing and operations. Greenleaf Publishing.

Sarre, G., Blanchard, P., Broussely, M., 2004. ‘Ageing of lithium-ion batteries’, Journal of

power sources, Vol. 127, pp. 65-71. (doi:10.1016/j.jpowsour.2003.09.008)

SBS-IF, 1998. ‘Smart battery specifications implementers forum: smart battery system

specifications’, Smart battery data specifications revision 1.1.

Schischke, K., Stobbe, L., Dimitrova, G., Scheiber, S., Oerter, M., Nowak, T., Schlösser,

A., Riedel, H., Nissen, N. F., 2014. Disassembly analysis of slates: design for repair and

recycling evaluation. Berlin, Germany.

Schischke, K., Stobbe, L., Scheiber, S., Oerter, M., Nowak, T., Schlösser, A., Riedel, H.,

Nissen, N. F., 2013. Disassembly analysis of slates: design for repair and recycling

evaluation. Berlin, Germany.

Schmalstieg, J., Käbitz, S., Ecker, M., Sauer, D. U., 2014. ‘A holistic ageing model for

Li(NiMnCo)O2 based 18 650 lithium-ion batteries’, Journal of power sources, Vol. 257, pp.

325-334. (doi:10.1016/j.jpowsour.2014.02.012)

142

Seagate, 2016. Seagate life cycle analyses & corporate sustainability

(http://www.seagate.com/gb/en/global-citizenship/life-cycle-assessment/) (accessed

8.16.16).

Sommer, P., Rotter, V. S., Ueberschaar, M., 2015. ‘Battery related cobalt and REE flows

in WEEE treatment’, Waste management. (doi:10.1016/j.wasman.2015.05.009)

Song, Q., Wang, Z., Li, J., Yuan, W., 2013. ‘Life cycle assessment of desktop PCs in Macau’,

International journal of life cycle assessment, Vol. 18, pp. 553-566. (doi:10.1007/s11367-

012-0515-7)

Sprecher, B., Kleijn, R., Kramer, G.J., 2014 a. ‘Recycling potential of neodymium: the case

of computer hard disk drives’, Environmental science and technology, Vol. 48, pp. 9506-

9513. (doi:10.1021/es501572z)

Sprecher, B., Xiao, Y., Walton, A., Speight, J., Harris, R., Kleijn, R., Visser, G., Kramer,

G.J., 2014b. ‘Life cycle inventory of the production of rare earths and the subsequent

production of NdFeB rare earth permanent magnets’, Environmental science and

technology, Vol. 48, pp. 3951-3958. (doi:10.1021/es404596q)

SquareTrade, 2009. 1 in 3 Laptops fail over 3 years: Netbooks fail 20 % more than laptops;

ASUS & Toshiba the most reliable [WWW Document].

Statista, 2016. Shipment forecast of laptops, desktop PCs and tablets worldwide from

2010 to 2019 (in million units) (http://www.statista.com/statistics/272595/global-

shipments-forecast-for-tablets-laptops-and-desktop-pcs/) (accessed 16.8.16).

Stuhlpfarrer, P., Luidold, S., Antrekowitsch, H., 2015. ‘Recycling of Nd2Fe14B-magnets’,

Proceedings of EMC 2015.

Talens Peiró, L., Ardente, F., 2015. Environmental footprint and material-efficiency

support for product policy — Analysis of material-efficiency requirements of enterprise

servers. (doi:10.2788/409022)

Talens Peiró, L., Ardente, F., Mathieux, F., 2016. Analysis of material-efficiency aspects of

Energy-related Product for the development of EU Ecolabel criteria — Analysis of product

groups: personal computers and electronic displays. (doi:10.2788/642541)

Tecchio, P., McAlister, C., Mathieux, F., Ardente, F., 2017. In search of standards to

support circularity in product policies: a systematic approach. Journal of cleaner

production, doi:http://dx.doi.org/10.1016/j.jclepro.2017.05.198

Teehan, P., Kandlikar, M., 2013. Comparing Embodied Greenhouse Gas Emissions of

Modern Computing and Electronics Products. Environ. Sci. Technol. 47, 3997-4003.

doi:10.1021/es303012r

The New York Times, 2016. Samsung to recall 2.5 million Galaxy Note 7 s over battery

fires (http://www.nytimes.com/2016/09/03/business/samsung-galaxy-note-

battery.html?_r=0)

Thiébaud -Müller, E., Hilty, L. M., Schluep, M., Widmer, R., Faulstich, M., 2017. ‘Service

lifetime, storage time, and disposal pathways of electronic equipment: a Swiss case study’,

Journal of industrial ecology. (doi:10.1111/jiec.12551)

USB Implementers Forum, 2016. Universal Serial Bus — USB type-CTM Cable and

Connector Specification [WWW Document]. URL

http://www.usb.org/developers/usbtypec/ (accessed 7.29.16).

Van Eygen, E., De Meester, S., Tran, H. P., Dewulf, J., 2016. ‘Resource savings by urban

mining: The case of desktop and laptop computers in Belgium’, Resources, conservation

and recycling, Vol. 107, pp. 53-64. (doi:10.1016/j.resconrec.2015.10.032)

Vanegas, P., Peeters, J., Cattrysse, D., Duflou, J., Tecchio, P., Mathieux, F., Ardente, F.,

2016. Study for a method to assess the ease of disassembly of electrical and electronic

143

equipment. Method development and application to a flat panel display case study. EUR

27921 EN. doi:10.2788/130925

Vanegas, P., Peeters, J.R., Cattrysse, D., Tecchio, P., Ardente, F., Mathieux, F., Dewulf,

W., Duflou, J.R., 2017. ‘Method to assess ease of disassembly for ecodesign and treatment

cost evaluation’, Resources, conservation and recycling.

(doi:https://doi.org/10.1016/j.resconrec.2017.06.022)

Vannieuwenhuyse, T., 2016. Oral communication on 15.9.2016.

Vetter, J., Novák, P., Wagner, M. R., Veit, C., Möller, K.-C., Besenhard, J. O., Winter, M.,

Wohlfahrt-Mehrens, M., Vogler, C., Hammouche, A., 2005. ‘Ageing mechanisms in lithium-

ion batteries’, Journal of power sources, Vol. 147, pp. 269-281.

(doi:10.1016/j.jpowsour.2005.01.006)

Viegand Maagøe and VITO, 2017. Preparatory study on the Review of Regulation 617/2013

(Lot 3) — Computers and Computer Servers (draft report).

von Geibler, J., Ritthoff, M., Kuhndt, M., 2003. The environmental impacts of mobile

computing — A case study with HP — Final report.

Wang, F., Huisman, J., Stevels, A., Baldé, C. P., 2013. ‘Enhancing e-waste estimates:

Improving data quality by multivariate input-output Analysis’, Waste management Vol.

33, pp. 2397-2407. (doi:10.1016/j.wasman.2013.07.005)

Wang, X., Gaustad, G., Babbitt, C.W., Richa, K., 2014. ‘Economies of scale for future

lithium-ion battery recycling infrastructure’, Resources, conservation and recycling Vol.

83, pp. 53-62. (doi:10.1016/j.resconrec.2013.11.009)

WEEE Forum, 2017. WEEE plastics and chemical, product and waste legislation. 5 July

2017 Draft.

Westpak, I., 2013. Slides used during a webinar on ‘Product Reliability Testing:

Environmental and Mechanical Requirements’;

Wieser, H., Tröger, N., 2016. The use time and obsolescence of durable goods in the age

of acceleration. An empirical investigation among Austrian households. AK Wien.

Yang, J., Lu, B., Xu, C., 2008. ‘WEEE flow and mitigating measures in China’, Waste

management, Vol. 28, pp. 1589-1597. (doi:10.1016/j.wasman.2007.08.019)

Zakotnik, M., Devlin, E., Harris, I.R., Williams, A.J., 2006. ‘Hydrogen decrepitation and

recycling of NdFeB-type sintered magnets’, Journal of iron and steel research,

international, Vol. 13, pp. 289-295. (doi:10.1016/S1006-706X(08)60197-1)

Zhong, H., Schiller, S., 2011. ‘Design of the expense allocation mechanism in e-waste

recycling deposit system under EPR framework’, ICSSSM11, pp. 1-6, IEEE.

(doi:10.1109/ICSSSM.2011.5959352)

GETTING IN TOUCH WITH THE EU

In person

All over the European Union there are hundreds of Europe Direct information centres. You can find the

address of the centre nearest you at: http://europa.eu/contact

On the phone or by email

Europe Direct is a service that answers your questions about the European Union. You can contact this

service:

— by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),

— at the following standard number: +32 22999696, or

— by electronic mail via: http://europa.eu/contact

FINDING INFORMATION ABOUT THE EU

Online

Information about the European Union in all the official languages of the EU is available on the Europa

website at: http://europa.eu

EU publications

You can download or order free and priced EU publications from EU Bookshop at:

http://bookshop.europa.eu. Multiple copies of free publications may be obtained by contacting Europe

Direct or your local information centre (see http://europa.eu/contact).

XX-NA-xxxxx-EN-N

doi:10.2788/89220

ISBN 978-92-79-64943-1

LB-NA-28394-EN-N