Environmental Impact of Refurbished Devices

ASSESSMENT OF THE
ENVIRONMENTAL
IMPACT OF A SET OF
REFURBISHED
PRODUCTS
FINAL REPORT
Assessment of the environmental impact of a set of refurbished products | 1 |

THANKS TO
Christian Brunot (Largo)
Emma Carre (Recommerce)
Marie Castelli (Back Market)
Sophie Chavignon (Itancia)
Lucylle Dennequin (Dupli)
Rachel Dethier (Sirrmiet)
Jean Christophe Estoudre (Sofi Groupe)
Erwann Fangeat (ADEME)
Frédéric Gandon (Largo)
Jean-Lionel Laccoureye (Bak2)
Etienne Lees Perasso (Tide Environnement)
Michel Maggi (Sofigroupe)
Camille Richard (Back Market)
Regis Robin (Itancia)
François Perabo (Sofi Groupe)
Veronique Vallejo (Le GSM)
Benoit Varin (Recommerce/Rcube)
Hugo Verger (Dupli)
Nicolas Flipo (Recyclea)
Nathalie Isach (Lm Eco Production)
Didier Moiroud (Ingram Micro Services)
Garik Maureemootoo (Emmaus Connect)
Camille Bardou (Emmaus Connect)
Sarah Maisonneuve (Ateliers Du Bocage)
Miryam Nachat (Atoutek)
Alban Regnier (Okamac)
Thomas Bardou (ATF)
Valérie Brouard (ATF)
Assessment of the environmental impact of a set of refurbished products | 2|

CITING THIS REPORT
Authors: Erwann Fangeat, ADEME, Laurent Eskenazi, Eric Fourboul, Hubblo, Julie Orgelet-Delmas,
DDemain, Etienne Lees Perasso, Firmin Domon, LCIE Bureau Veritas
2022. Assessment of the environmental impact of a set of refurbished products
final report - 180 pages
This report is available online at https://librairie.ademe.fr/
Any depiction or reproduction of this document and any translations hereof, whether in full or in part, made without
the consent of the author or their successors in title is prohibited under the French Intellectual Property Code (Art. L
122-4) and constitutes a counterfeit punishable under the French Penal Code. Only copies or reproductions reserved
strictly for the copyist’s private use and not for public use, as well as analyses and brief quotations justified by the
critical, educational or informative nature of the work in which they are included, are authorised (Art. 122-5), subject
nonetheless to the provisions of Articles L 122-10 to L 122-12 of the Intellectual Property Code, concerning
photocopying.
This document is distributed by ADEME
ADEME
20, avenue du Grésillé
BP 90 406 | 49004 Angers Cedex 01
Contract number: 2021MA000057
Study carried out on behalf of ADEME by: DDemain, Hubblo, LCIE Bureau Veritas
Technical coordination: ADEME: Erwann Fangeat
Department: DECD / SER

SUMMARY ...................................................................................................................................................... 4
GLOSSARY .................................................................................................................................................... 10
1. CONTEXT AND AIMS OF THE PROJECT ......................................................................................... 12
1.1. Context ...................................................................................................................................................... 12
1.2. Aims of the study ...................................................................................................................................... 13
1.3. Methodology ............................................................................................................................................. 14
Selection, classification and characterisation of impacts ...............................................18

Standardisation & weighting ................................................................................................18
1.4. Study procedure ....................................................................................................................................... 19

1.5. Critical review ........................................................................................................................................... 19
1.6. Project stakeholders.................................................................................................................................21
1.7. Validity of the results ...............................................................................................................................21
2. SCOPE OF THE STUDY ...................................................................................................................... 22
2.1. Scope of the study .................................................................................................................................. 22
2.2. Managing the comparison between new and refurbished ................................................................ 26

3. ENVIRONMENTAL IMPACT INDICATORS AND DATA .................................................................. 31
3.1. Type and source of data ..........................................................................................................................31
3.2. Data on physical characteristics ........................................................................................................... 32
3.3. Data on incoming and outgoing flows involved in the refurbishment process ............................. 33
Electrical and electronic equipment and subassembly .................................................. 43

Industrial and domestic waste ............................................................................................ 44
Values of the R2 & R3 factors ............................................................................................. 45
3.4. AILC methodology and types of impact .............................................................................................. 45
3.5. LCA modelling tool .................................................................................................................................. 49
4. SMARTPHONES AND TABLETS FAMILY .......................................................................................... 50
4.1. Overview ................................................................................................................................................... 50
4.2. Defining reference models ..................................................................................................................... 50
4.3. Hypotheses and entry data.....................................................................................................................51
4.4. Results ....................................................................................................................................................... 55

Focus on screen type............................................................................................................ 66
Focus on batteries ................................................................................................................ 67
Focus on means of upstream transport ............................................................................ 67
Focus on means of distribution by the private individual .............................................. 68
Presentation of results for the reference scenarios..........................................................71

Presentation of results – Impacts avoided .........................................................................71
Variants and ranges of variation ......................................................................................... 72
4.5. Extrapolation to tablets.......................................................................................................................... 76
5. COMPUTER FAMILY........................................................................................................................... 81
5.1. Laptop computers ....................................................................................................................................81
Profile of refurbishers ........................................................................................................... 82

Default characteristics ......................................................................................................... 83
Collection data per FU ......................................................................................................... 83
Default characteristics ......................................................................................................... 85
Total results at a functional unit level (1 year of use) ...................................................... 88

Results broken down by life cycle impact ........................................................................ 88
Results broken down by stage of refurbishment ............................................................. 89
Range of variation by type of refurbishment ................................................................... 90
Range of variation of impact depending on the refurbishers .........................................91
Range of variation by supply and refurbishment location ............................................. 96
Range of variation based on the lifespan .......................................................................... 96
Additional sensitivity analyses ............................................................................................ 97
Standardised results – world inhabitant approach ....................................................... 100

Results scaled up to planetary boundaries ...................................................................... 101
Comparison of the impacts of new and refurbished laptops - Substitution approach

102
5.1.4.4.1.1. Presentation of results for the reference models .......................................................... 102
5.1.4.4.1.2. Presentation of results – Impacts avoided ...................................................................... 102
5.1.4.4.1.3. Variants and ranges of variation ....................................................................................... 103
Comparison of the impacts of new and refurbished laptops - Depreciation approach
105
Comparison – Analysis of behavioural scenarios ........................................................... 106
5.2. Desktop computers ............................................................................................................................... 108
5.2.1. ............................................................................ Overview

109

Profile of refurbishers ......................................................................................................... 109
Collection data per FU ........................................................................................................ 110
Default characteristics ........................................................................................................ 111
Total results at a functional unit level (1 year of use) ..................................................... 114

Results broken down by life cycle impact ....................................................................... 114
Results broken down by stage of refurbishment ............................................................ 115
Range of variation by type of refurbishment .................................................................. 116
Range of variation of impact depending on the refurbishers ....................................... 117
Range of variation by supply and refurbishment location
121
Range of variation based on the lifespan ......................................................................... 121
Additional sensitivity analyses .......................................................................................... 122
5.2.4.3.8.1. Focus on RAM ...................................................................................................................... 122
5.2.4.3.8.2. Focus on hard disks ............................................................................................................ 123
Standardised results per world inhabitant...................................................................... 124

Results scaled to planetary boundaries ........................................................................... 125
Comparison of the impacts of new and refurbished laptops - Substitution approach

126
5.2.4.5.1.1. Presentation of results for the reference models at a
lifespan level ............................................................................................................................................... 126
5.2.4.5.1.2. Presentation of results – Impacts avoided ...................................................................... 126
5.2.4.5.1.3. Variants and ranges of variation ....................................................................................... 127
Comparison of the impacts of new and refurbished desktop computers
- Depreciation approach .......................................................................................................................... 128
6. LIMITS ................................................................................................................................................ 131

7. CONCLUSIONS ................................................................................................................................ 133
8. GENERAL RECOMMENDATIONS ................................................................................................... 136
9. APPENDICES .................................................................................................................................... 138
9.1. Critical review report ............................................................................................................................ 138
9.2. Collection file example – synopsis ...................................................................................................... 139
9.3. Data used for modelling ....................................................................................................................... 140
9.4. Impact data – new equipment ............................................................................................................. 151
9.5. Additional quantified results for the main product families .......................................................... 154
9.6. Results extrapolated to other product families ............................................................................... 166
BIBLIOGRAPHICAL REFERENCES .............................................................................................................. 171

10. INDEX OF TABLES AND FIGURES .................................................................................................. 172
TABLES 172

FIGURES ................................................................................................................................................................. 172
11. ABBREVIATIONS AND ACRONYMS .............................................................................................. 174

ABSTRACT
The purpose of this study is to assess the multi-criteria environmental impact of the
refurbishment of digital devices destined for the general public (smartphones, tablets, desktop
computers and laptop computers) and to identify the difference between using refurbished
devices and purchasing new devices.
To do this, we implemented a 2-stage approach:
 Creating reference models based on data collected from refurbishers
 Conceptualising these models to:
 Assess the impact of refurbishing practices
 Assess the impact of user behaviours
It seems that refurbishment can have a significant effect on indicators such as the depletion of
fossil and mineral resources, climate change and the production of WEEE (electronic waste).
For each category of products, we were able to:
 Identify the main contributing factors to the impact of refurbishment (supply and
distribution logistics, practices associated with part replacements, cleaning
practices)
 Identify which conditions were favourable to the development of an

environmentally friendly refurbished goods sector (short supply chains and slow
means of transport, using second-hand parts, refurbishing products that have
already had a long lifespan, optimising packaging and removing accessories)
 Identify which consumer behaviours should be encouraged (keep devices for longer,
avoid buying the latest generation of devices and having all their parts upgraded)
Refurbishment has a beneficial impact on the environment in the reference scenarios. However,
the varied nature of these practices and behaviours (lifespan) can result in an increased impact
on the environment, as highlighted by the depreciation approach and the behavioural analysis.
Therefore:
 Consumers should: keep their devices for longer and buy, from local refurbishers,
equipment that has had a proper first life (> 2 years for smartphones, > 3 years for
laptop computers)
 Refurbishers should: optimise their processes and carry out informed refurbishment
(with regard to the number and origin of the parts), prioritising the available pool of
devices that have already had a first life
 Legislators should: provide information on the concept of a ‘first life’, encouraging

any stakeholders adopting an approach of recovering ‘old’ products

Glossary
LCA-A, attributional (or analysis by attributes)1: An LCA where the system being examined is composed
of basic processes connected by flows from the technosphere that are directly attributable to the system.
The system is deemed to be established (in permanent operation). The consequences triggered by the
alternatives compared do not significantly call into question the supply chains.
LCA-C, consequential (or analysis by consequences)1: An LCA where the system being examined is
composed of basic processes connected by economic flows but also by processes indirectly affected by
the establishment or change of the life cycle for the product under study.
Method (of environmental assessment)2: a set of rules and calculation steps that can be used to generate
an assessment of the environmental impact of a system, with a view to measuring and analysing its effects
on the environment in order to prevent any harmful consequences to the environment.
Life cycle analysis (LCA) method3: compilation and assessment of the potential inflows, outflows and
environmental impacts of a system of products over its lifespan.
Reference guide: structured set of recommendations (whether normative or not) and best practices,
used for implementing a method in a given context, for a given product category, or for a specific
objective.
ICT - Information and communication technologies4: a diverse set of technological tools and resources
used to transmit, store, create, share or exchange information. These technological tools and resources
include computers (laptop or desktop, workstations, etc.), the Internet (websites, software, blogs and
emails), live broadcasting technologies (data centres, servers, etc.) and devices (radio, television and
webcasting), recorded broadcasting technologies (podcasting, audio and video players, and storage
devices) and telephony (fixed or mobile, satellite, video-conferencing, etc.).
Refurbished product: A second-hand product or spare part may be described as a ‘refurbished product’
or be described using the term ‘refurbished’ if the following conditions are met:
“1. The product or part has undergone tests covering all its functions in order to establish that it meets
legal safety requirements and performs in a way that the consumer is legitimately entitled to expect;
“2. Where applicable, the product or spare part has undergone one or more procedures designed to
restore its functions. These procedures include the deletion of any data recorded or stored and pertaining
to previous use or a previous user, before the product or part changed hands. [Decree 2022-190 of 17
February 2022 on the conditions for using the terms ‘refurbished’ and ‘refurbished product’].
1
European Commission Joint Research Center (JRC), ILCD handbook – The International Reference Life Cycle Data
System, 2012.
2
ADEME. (2020, 7 October). ). L’évaluation environnementale dans l’industrie et les services. Outils et
méthodes.(Environmental assessment in industry and services. Tools and methods.)
3
International Organization for Standardization (ISO). (2006). ISO 14040:2006 - Environmental management — Life
cycle assessment — Principles and framework.
4
UNESCO. (n.d). Information and communication technologies (ICT).

1. Context and aims of the project
1.1. Context
Now that the environmental crisis is becoming ever increasingly palpable, new models of sustainable
consumption must be found. To preserve the quality of our ecosystem, we need to conserve its resources
and save them for targeted usages that facilitate the ecological transition.
However, it would appear that we are moving from a reliance on fossil fuels to a reliance on abiotic
resources, and that an untenable competition is arising between companies in the renewable energy and
the digital sector.5The digitalisation of society is a rampant phenomenon that is contributing significantly
to this antagonism. The digital sector has permeated every aspect of our daily life, both personal and
professional. We use smartphones to communicate, computers to play/work, tablets to learn, consoles
to have fun, etc. The average number of connected devices owned by each person in France – 15 – is
considerably higher than the global average of 8. However, digital devices are among the most complex
to manufacture and the least durable.6 As an example, the average usage period for a smartphone is
between 23 and 37 months7, although it takes over 50 different materials and the extraction of over 200
kg of materials to manufacture just one phone8 (MIPS indicators - Material Input per Service Unit).
In the digital sector at a European level, user terminals generate the highest impact, at between 59% and
88% of the total impact. It is therefore time to act, and promoting refurbishment seems to be one of the
most suitable avenues.
Refurbishment, which did not have an official legal definition until 2022, is on the rise, in both homes and
in organisations. A refurbished product is a second-hand product or spare part that has been subjected
to tests covering all its functions and, where applicable, has been the subject of one or more procedures
designed to restore its functions and to remove any connection with its previous user (deletion of data)9.
5
The rare metals war – Guillaume Pitron
6
Digital Technologies in Europe: an environmental life cycle approach – 7 December 2021 – Study commissioned by
the French Green Party/EFA – sponsored by GreenIT.fr together with the NegaOctet consortium
7 Renouvellement des terminaux mobiles et pratiques commerciales de DISTRIBUTION (Renewing mobile terminals
and sales practices for DISTRIBUTION) – ARCEP – June 2021
8
ADEME. J. Lhotellier, E. Less, E. Bossanne, S. Pesnel. 2018. LCA modelling and assessment for consumer products and
capital goods – Report. 186 pages
9
Decree 2022-190 of 17 February 2022 on the conditions for using the terms ‘refurbished’ and ‘refurbished product’

Refurbishment therefore allows
consumers access to products that
have been tested under warranty and
offer the same services as a new,
recent product, for a lower price. Even
though consumers’ purchasing
decisions are guided primarily by
financial gain, they are also guided by
a feeling of having made a gain in
terms of impact on the environment.
This study provides data on

environmental impacts that illustrates
that these products are preferable to
their new equivalents, and proposes
avenues for improvement in order to
optimise performance in the
refurbishment of digital devices.
Figure 1 – Composition of smartphones
and computers - The hidden side of the
digital sector - ADEME 2021
1.2. Aims of the study

Although refurbishment is no longer
limited to telecommunications
devices, the practice is still the most
widespread in this sector. Thus, this
study will involve assessing the impact
of refurbishing desktop and laptop
computers, smartphones and
tablets10. There are several types of
refurbishment for each of these product families: cleaning only, or cleaning with some components being
replaced. Regardless, refurbishment includes mandatory procedures such as deleting data, testing for
conformity, functional testing, warranty, customer service and after-sales service, and supplier and
subcontractor traceability. It can be carried out in one or several locations – France, Europe, Asia, USA,
etc. – which will affect logistics, particularly with regard to transporting the parts, performing the
refurbishment procedures and shipping from the terminal.
All of the products and their refurbished versions will now be examined.
10
An additional assessment was carried out for video game consoles and servers, and is included here as an appendix.

Figure 2 – Product families and versions studied
This study will use the life cycle analysis method to:
 Assess the difference between using refurbished devices and purchasing new devices;
 Assess the multi-criteria environmental impact of the refurbishment of digital devices destined
for the general public (smartphones, tablets, desktop computers and laptop computers);
 Identify the main contributing factors to the environmental impact of refurbishing and
distinguish the best practices that allow the environmental performance of refurbishment
procedures to be improved;
 Compare the impact of refurbishment depending on the origin of the products and where the
refurbishment is carried out (France, Europe, Asia, USA, etc.).
1.3. Methodology
1.3.1. Overview
Life cycle analysis is a method used to evaluate the environmental impact of products, services or
organisations. There are other methods for evaluating environmental impact, such as the carbon
footprint or impact studies. However, some of LCA’s characteristics make its holistic approach unique.
This method, which has been in use since the late 1990s and is standardised in the ISO 14040:200611 and
ISO 14044:200612 series, uses several key concepts to determine the ecological burden imposed by a
product or service:
Multi-criteria: multiple environmental indicators should be considered systematically, including global

warming potential; the depletion of abiotic resources; the creation of photochemical ozone; water, air
and soil pollution; human ecotoxicity; and biodiversity. The list of these indicators is not definitive and
depends on the business sector in question.
11
ISO 14040:2006 - Environmental management — Life cycle assessment — Principles and framework
12
ISO 14044:2006 - Environmental management — Life cycle assessment — Requirements and guidelines

48
Figure 3 - Basics of life cycle analysis
Life cycle: in order to incorporate the impacts caused at every stage in the life cycle of a device, from
the extraction of raw materials (which are often difficult to access) to the production of waste, and even
the consumption of energy in the use phase, etc., we try to ensure that we consider the device’s entire
service life as a whole.
Quantitative: each indicator is quantified numerically, so that all externalities for a given product or
service can be placed on the same scale, and decisions can be taken objectively.
Functional: the aim of the study is defined by the function it performs, so as to make it possible to
compare different technical solutions.
Attributional or consequential:
 Attributional: the study describes the potential environmental impacts that can be attributed
to a system (such as a product) over the course of its life cycle, i.e. upstream from the supply
chain and downstream from the use of the system, as well as at the end of its lifespan. It aims
to quantify the direct effects connected with a system.
 Consequential: the study seeks to identify the consequences of a decision on a broad

ecosystem – economic consequences, energy consequences, social consequences, etc. For
example, it might investigate the consequences of the development of refurbished products
on the WEEE (Waste from Electrical and Electronic Equipment) industry, the rare earths
industry and the relocation of employment. A consequential analysis can be used to model
the target system in a manner that will fully incorporate these bundles of consequences. It
takes account of the indirect effects connected with a system.
Thus, an environmental audit can identify and prevent pollution being transferred from one phase to
another as well as from one indicator to another.
1.3.2. The different stages in an LCA

As set out in the ISO 14040:200613 standard, an LCA study is composed of 4 interdependent stages:
1. Defining the objective and the scope
2. Life cycle inventory analysis (LCI)
3. Life cycle impact assessment (LCIA)
4. Interpreting the results of the life cycle
LCA is an iterative technique whereby each stage uses the results from the other stages, thereby
contributing to the integrity and the consistency of the study and its results. It is a holistic approach and,
13
ISO 14040:2006 - Environmental management — Life cycle assessment — Principles and framework

therefore, must be used transparently in order to ensure that the results obtained are adequately
interpreted.
N.B.: LCA addresses potential environmental aspects, and thus does not forecast the real or absolute
impact on the environment.
Defining the objective and the scope

The definition of the study’s objective must describe the aim of the study and the decision-making
process for which it will provide support with taking decisions concerning the environment. The objective
of an LCA must determine the planned application, the reasons for which the study is being carried out,
the target audience (i.e., the people to whom the results of the study are intended to be communicated)
and whether the results are to be used for comparative purposes that are to be disclosed to the public.
The scope of an LCA – including the limits of the system, the level of detail, the quality of the data, the
hypotheses formed, the limits of the study, etc. – will depend on the subject and the intended use of the
study. The depth and scope of a field of application can vary considerably depending on the specific
objective being pursued.
An LCA has a structured approach, pertaining to a functional unit and/or a declared unit. All subsequent
analyses are therefore connected to this unit. Where a comparison is required – solely between products
or services performing the same function – a functional unit must be chosen that refers to the function
performed by the products or services in question.
Life cycle inventory analysis

Data collection
This stage involves collecting data and the calculation procedures for quantifying the relevant entry and
exit data for the system being studied. The data to be included in the inventory must be collected for
each unit process under consideration, within the limits of the system being studied.
Inventory of elementary flows
In an LCI, the elementary flows must be recorded within the limits of the system, i.e., account must be
taken of the flows of materials and energy that come from the environment without prior transformation
by people (e.g. the consumption of oil, coal, etc.) and the flows targeting the environment without any
other transformation taking place (e.g. atmospheric emissions of CO2, SO214, etc.). Elementary flows
include the use of resources, atmospheric emissions and waste discharged into the water and soil
associated with the system.
The data collected, whether it is measured, calculated or estimated, is used to quantify all incoming and
outgoing materials and energies from the various processes involved.
Rules for attribution and allocation
Reality has shown that very few industrial processes have one single output. In fact, industrial processes
typically produce more than one product, and/or intermediary products for which the waste is recycled.
Where several outgoing flows are produced, criteria for attributing the environmental burden to each
individual product must be applied, as in the present study.
Assessing data quality
The LCA and LCI data associated with digital services and devices continues to be present challenges.
Most studies inspired by LCA use single-criterion data (such as energy or global warming) or sets of
heterogeneous data. This project uses the NegaOctet database. This database was still under
development and undergoing external testing during the study (early 2022), and an assessment of the
data quality was proposed for the specific data used in the study.
Nonetheless, we chose this database for several reasons:
 It allows new and refurbished subassemblies to be modelled homogeneously;
14
Sulphur dioxide

 The NegaOctet database can be used to calculate all the indicators for the E.F. method15
(Environmental Footprint), and it is in line with E.F. recommendation 3.0. This database was
developed by a consortium composed of experts in the digital sector and in life cycle analysis:
APL Datacentre, Ddemain, GreenIT.fr and LCIE. This database was provided to these clients in
March 2022 and was reviewed internally and externally by the scientific community. The
method used for compiling the database is available on request;
 The NegaOctet database will populate Ademe’s Impact database by March 2022.
In this study, this database will be used alongside the ESR, Ecoinvent 3.6, EIME and ELCD databases, all
four of which are mutually compatible with regard to their method and the scope of their data.
Life cycle impact assessment
Selection, classification and characterisation of impacts

This stage aims to assess the significance of potential environmental impact based on the results from
the inventory. This process involves selecting impact categories and associating inventory data with those
impact categories (e.g. climate change) and with impact category indicators (such as climate change over
100 years according to the CML impact model) via the characterisation factor. This stage provides
information for the interpretation stage.
Standardisation & weighting

The numerical results from the indicators can potentially be ordered, standardised, consolidated and
weighted. This approach makes interpretation easier, but there is no scientific consensus on a robust
method for carrying out this assessment.
Interpreting the results of the life cycle

Interpretation is the final stage of the LCA.
During this stage, the results from the inventory and/or the assessment are summarised and discussed in
an understandable manner. This part is used by the recipients of the study as a basis for making
conclusions, recommendations and decisions, in accordance with the established objective and scope.
Sensitivity & uncertainty analysis

Given that part of data collection is based on a literature review, the model relies on secondary data,
which may be uncertain. In order to determine the order of magnitude of the variation in the results, a
sensitivity and uncertainty analysis must be carried out.
1.4. Study procedure

The study was organised in the following stages:
 An outline stage common to all the product families to be studied, to define the scope of the
study and gain an understanding of the refurbishment process;
 A pilot stage for the method used for smartphones:

‐ Identifying the reference products;
‐ Collecting specific data from multiple refurbishers;
‐ Developing a tool for configuring refurbishment scenarios;
‐ Creating a reference model for refurbishment;
‐ Evaluating the results and analysing variations;
15
The E.F. method is defined by the European Commission based on the Commission Recommendation of 9 April
2013 on the use of common methods to measure and communicate the life cycle environmental performance of
products and organisations throughout the life cycle (2013/179/EU).

 Critical review of the first model;
 Adapting the methodology to the other product families under consideration.
A presentation was given to the steering committee at the end of each stage.
A review was organised by the critical review committee to present the method used and the results for
each product family.
1.5. Critical review

Critical review is a procedure used to certify that the Life Cycle Analysis (LCA) complies with international
standards and complementary national regulations to meet the objectives of the study. In this case, ISO
14040 and ISO 14004 are the standards that must be observed. Where possible and appropriate here, the
methodological framework will also refer to complementary standards such as:
 ITU L140 – Methodology for environmental life cycle assessments of information and
communication technology goods, networks and services16;
 PEF17 and PEFCR guidance documents on IT equipment18.
A critical review is mainly carried out where the results are to be released to the general public or where
there are comparative claims. Its aim is to verify the:
 Consistency between the objective, the data collection and the results of the study;
 Validity of the hypotheses considered;
 Validity of the results of the study;
 Validity of the communication carried out.
In the present context, critical review also strives to:
 Lend a third-party perspective on the models obtained;
 Identify the key elements and limits of the study;
 Ensure that the information considered is suitable and reliable;
 Assure ADEME that the elements on which they will rely, taken from this study, are robust,
within the limits of the available data.
The critical review of the study performed for ADEME was realised by:
 Stéphane Le Pochat, EVEA – 1st iteration
 Roland Hischier, EMPA – full review
The critical review was carried out progressively in the following stages:
 Framing;
 Discussion and validation of the general principles applicable to smartphones;
 Validation of the results for smartphones;
 Validation of the results for the other product families.
The report from the critical review is appended to this report.
16
https://www.itu.int/rec/T-REC-L.1410/en
17
https://ec.europa.eu/environment/eussd/smgp/pdf/PEFCR_guidance_v6.3.pdf
18
https://ec.europa.eu/environment/eussd/smgp/pdf/PEFCR_ITequipment_Feb2020_2.pdf

1.6. Project stakeholders
1.6.1. Target audience

The target audience is mainly:
 ADEME;
 French and international refurbishers;
 Market platforms;
 The general public.
The final study and the data generated are to be protected by Creative Commons licence (CC BY-SA) to
ensure they are widely accessible. The comparative statements will be valid solely in the context set out
in this study.
1.6.2. Study backers and organisation

This study was carried out on behalf of ADEME.
ADEME is thus the backer of the study and the party steering the project. To ensure that the work is valid,
a Steering Committee, composed of professional organisations, businesses and associations from the
refurbishment sector, was established.
For the technical aspects of the LCA, the project was steered by DDemain and carried out by DDemain,
Hubblo and LCIE.
The critical review was conducted by Evea and EMPA.
Figure 4 – Roles and responsibilities of the study stakeholders
1.7. Validity of the results

The results are valid solely for the situation described by the hypotheses set out in this report. The
conclusions are subject to change under different conditions. The suitability and reliability of a usage by
third parties or for other purposes than those mentioned in this report therefore cannot be guaranteed
by the parties performing the LCA.
This falls solely under the responsibility of the study backer.

2. Scope of the study
2.1. Scope of the study

The aim of our study is to propose an assessment of the environmental impact of refurbished electronic
devices, and to enable that impact to be compared with the impact of new devices.
The study focuses mainly on the direct impacts of refurbishment. Some consideration was given to
certain indirect impacts (particularly the trend towards overperformance) but there is room for further
consideration (on the impact of the development of the refurbishment sector on the end-of-life sector,
the labour market in France, the relocation of certain impacts, etc.). This is therefore an attributional LCA.
The following paragraphs give details of the scope of the study, namely:
 Functional unit;
 System limits: inclusion, exclusion, cut-off criteria;
 Geographic, temporal and technological representativeness;
 Life cycle phase considered;
 Quantified environmental impact and characterisation methods;
 Types and sources of data;
 Data quality requirements.
2.1.1. Function and functional unit

The functional unit is the reference unit used to connect inputs and outputs, as well as the environmental
performance ratings for one or more product systems. We considered the following functional unit:
‘Owning and using a device for one year’
All of the impacts considered will be scaled to one year’s usage. We will therefore divide the
environmental impacts by the number years associated with the life cycle:
𝑨𝒏𝒏𝒖𝒂𝒍 𝒆𝒏𝒗𝒊𝒓𝒐𝒏𝒎𝒆𝒏𝒕𝒂𝒍 𝒊𝒎𝒑𝒂𝒄𝒕𝒔 𝒅𝒆𝒄𝒍𝒂𝒓𝒆𝒅

𝑬𝒏𝒗𝒊𝒓𝒐𝒏𝒎𝒆𝒏𝒕𝒂𝒍 𝒊𝒎𝒑𝒂𝒄𝒕𝒔 𝒄𝒂𝒍𝒄𝒖𝒍𝒂𝒕𝒆𝒅 𝒇𝒐𝒓 𝒕𝒉𝒆 𝒅𝒖𝒓𝒂𝒕𝒊𝒐𝒏 𝒐𝒇 𝒕𝒉𝒆 𝒍𝒊𝒇𝒆 𝒄𝒚𝒄𝒍𝒆 𝒄𝒐𝒏𝒔𝒊𝒅𝒆𝒓𝒆𝒅
𝑨𝒔𝒔𝒐𝒄𝒊𝒂𝒕𝒆𝒅 𝒖𝒔𝒂𝒈𝒆 𝒑𝒆𝒓𝒊𝒐𝒅 𝒇𝒐𝒓 𝒕𝒉𝒆 𝒄𝒚𝒄𝒍𝒆 𝒄𝒐𝒏𝒔𝒊𝒅𝒆𝒓𝒆𝒅
Usage periods may be different for new and refurbished devices.
2.1.2. System limits
System limits and life cycle phases considered

During this study, we will consider the following elements:
 The production stage includes:
‐ For new devices: extracting the raw materials, upstream transport, and manufacturing
processes;
‐ For refurbished equipment: collecting devices that are out of service, sourcing of
equipment to be refurbished, the refurbishment process, producing spare parts, and
disposing of worn parts.
 The distribution stage includes:
‐ Manufacturing packaging and accessories;

‐ Distribution between the producer (manufacturer or refurbisher) and the distributer (first
logistics platform);
‐ Travel by the end user, or the distribution scenario to the place of use.
 The use phase includes:
‐ The electricity consumed when used in France.
 The end-of-life stage (modelled solely for new products) includes:
‐ Collecting devices that are out of service;

‐ Pre-processing devices;
‐ Eliminating the elements that are not reused: incineration or landfill.
In our study, the following stages have been excluded for comparison or suitability reasons:
 After-sales service and Customer service;
 In-store presentation;
 End-of-life for devices that were not refurbished: to prevent the same devices being counted
twice, the end-of-life stage for devices supplied by refurbishers but not processed was
excluded from the system, since it is associated with marketing new devices.
NOTE 1 – We are currently unable to measure consumption reliably for the various types of terminal, since
the data used to characterise the use phase is averaged out and intended to give an order of magnitude
of the relative impact of the use phase as compared with the production stage.
NOTE – As part of a study on indirect effects, it might be interesting to consider a usage scenario that
differentiates between basic (single-function) telephones, smartphones (calls, SMS, social media, etc.) and
high-end smartphones (video calls, SMS19, voice messages, HD films, etc.).
NOTE 3 – In the substitution approach, the impacts of end of life are attributed to the life cycle of new
devices. We consider that refurbishment shifts this end of life to another point in time, and it is still
attributed to the device’s first life cycle.
With regard to refurbishment, an initial data set was collected for all refurbishment processes done by all
refurbishers.
19
Short Message Service

Figure 5 – Scope of the study for refurbished products
For new products, generic configurable data taken from the NegaOctet project were taken into
consideration. The reference models were selected in collaboration with the refurbishers, based on their
sales statistics for 2020.

Figure 6 – Scope of the study for new products
Exclusion
The following flows were excluded from the study during data collection:
 Lighting, heating, cleaning and tidying the plants producing the devices (in some cases, this
data could not be differentiated and was therefore incorporated);
 Transport for employees deemed to fall outside the scope of the study;
 The manufacture and maintenance of production tools;
 The construction and maintenance of infrastructures;
 Flows in administrative, management and R&D departments;
 The sale of the products;
 Meals for the staff.
In a manner more specific to this study, we excluded:
 Impacts associated with sourcing control;
 Impacts associated with points of sale and storage locations;
 The consumption of data and resources associated with the use phase;
 The manufacture of protective equipment for the devices: cases, screen protectors, bags, etc.;
 The manufacture of associated accessories such as controllers for video games, etc.;
Cut-off criteria
Generally, environmental modelling must cover a specific percentage (greater than or equal to 95%) of
the devices or systems;
 The mass of the intermediary flows not taken into account must be less than or equal to 5% of
the mass of the elements for the reference product corresponding to the functional unit;

 The energy flows not taken into account must be less than or equal to 5% of the total primary
energy used during the life cycle of the reference product corresponding to the functional
unit.
2.1.3. Temporal representativeness

The data collected from refurbishers was collected in 2021 for the whole year of 2020. It should be noted
that the pandemic had a significant impact on business for some of them. Where this was the case, data
was collected for 2019.
To ensure that the data collected was suitably representative, it was collected for a one-year period.
2.1.4. Geographical limit

This study covers the refurbishment of digital devices aimed at the general public in France or abroad,
and sold in France.
2.2. Managing the comparison between new and refurbished

Two approaches were considered when performing the comparison:
 A substitution approach (reference approach) – No weight is attributed to a refurbished

product with regard to its first life, and it is deemed that purchasing that product completely
replaces the manufacture of a new device;
 A depreciation approach (alternative approach) – If the refurbishment occurs before the

theoretical end of the device’s first life, part of the impact of manufacturing the new device is
reattributed to the refurbished device.
 Further, several consumption profiles were conceptualised in order to guide consumption

choices over the long term.
2.2.1. Reference approach - Substitution approach

Reference approach – An approach indicating the substitution of a new product with a refurbished
product. We adopted the starting assumption that selling a refurbished product prevented a new
product having to be manufactured. We also deemed that the impacts associated with the end-of-life
processing of the devices should be associated with the manufacture of the new devices, since
refurbishment merely delayed this end-of-life process instead of preventing it.
𝐸𝑛𝑣𝑖𝑟𝑜𝑛𝑚𝑒𝑛𝑡𝑎𝑙 𝑖𝑚𝑝𝑎𝑐𝑡 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑓𝑢𝑟𝑏𝑖𝑠ℎ𝑒𝑑 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑝𝑒𝑟 𝑦𝑒𝑎𝑟

𝐼𝑚𝑝 𝐼𝑚𝑝 𝐼𝑚𝑝
𝐷

Where
𝐼𝑚𝑝 – Manufacturing impact of the refurbished product
𝐼𝑚𝑝 – Distribution impact of the refurbished product
𝐼𝑚𝑝 – Usage impact of the refurbished product
𝐷 – Usage period of the refurbished device
Figure 7 – Comparative scope between new and refurbished products under a substitution approach (example used:
smartphones)
2.2.2. Alternative approach - Depreciation approach

Since the first model was liable to encourage consumers to renew their devices more regularly and did
not necessarily promote an increase to the overall lifespan of a device, a second approach was put
forward.
This approach involves transferring and depreciating some of the impacts of manufacturing and end of
life of the new device to the refurbished product, if refurbishment takes place before the theoretical end
of the usage period in the first cycle. This approach makes it possible to:

 Distinguish the stakeholders who refurbish older products with a lower economic value from
those who prefer products with a higher monetary value but with a shorter initial lifespan;
 Direct consumers towards the products and practices with the lowest possible impacts,
thereby avoiding a rebound effect and overconsumption.
Using this approach makes it possible to address an issue raised by ARCEP in its report on ‘Renewing
mobile terminals and commercial distribution practices’ of June 2021 and shown by the data collected as
part of this study, which is that:
 Currently, a considerable percentage of devices are collected outside of France. The supply of
devices for refurbishment comes mainly from abroad, which restricts any local circularity;
 The markets that supply the refurbishment market are the US and Asian markets, which are
characterised by an overconsumption of terminals combined with an aggressive annual
renewal policy (particularly for smartphones). At this stage, refurbishment is therefore based
on an assumption of overproduction.
The impact of refurbishment in terms of a logic of depreciating the first lifetime is calculated as follows:
𝐼𝑚𝑝𝑎𝑐𝑡 𝐼𝑚𝑝𝑎𝑐𝑡
𝐼𝑚𝑝𝑎𝑐𝑡 𝐼𝑚𝑝𝑎𝑐𝑡
𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑
where:
𝐼𝑚𝑝𝑎𝑐𝑡 .
𝐼𝑚𝑝𝑎𝑐𝑡 𝐷 𝐷
𝐷
𝐷 = Usage period in the theoretical first lifetime observed on the market
= Usage period beyond which it can be considered that refurbishment automatically extends the lifespan
𝐷
= Usage period in the actual first lifespan
= Usage period prior to collection and change of owner
Figure 8 – Taking account of impacts as part of a depreciation approach
2.2.3. Calculation of impacts avoided

 A calculation of the impacts avoided was also carried out. This calculation presumes that
purchasing a refurbished device prevents a new device having to be manufactured, or defers

this manufacture to a later date (if the new and refurbished usage periods are not entirely
equivalent).
 This calculation considers that purchasing a refurbished device prevents or delays a new
device having to be manufactured. This total or partial non-manufacture of the new device is
counted so as to highlight the benefits associated with avoiding or delaying manufacture.
𝐼𝑚𝑝𝑎𝑐𝑡 𝑜𝑓 𝑡ℎ𝑒 𝑟𝑒𝑓𝑢𝑟𝑏𝑖𝑠ℎ𝑒𝑑 𝑑𝑒𝑣𝑖𝑐𝑒 𝐼𝑚𝑝𝑎𝑐𝑡 𝑜𝑓 𝑡ℎ𝑒 𝑛𝑒𝑤 𝑑𝑒𝑣𝑖𝑐𝑒
𝐴𝑛𝑛𝑢𝑎𝑙𝑖𝑠𝑒𝑑 𝑖𝑚𝑝𝑎𝑐𝑡𝑠 𝑎𝑣𝑜𝑖𝑑𝑒𝑑
𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑠𝑒𝑐𝑜𝑛𝑑 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒 𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑓𝑖𝑟𝑠𝑡 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒
2.2.4. Conceptualising uses

To illustrate the impact of these approaches and uses, various renewal behaviours were simulated in order
to identify which behaviours had the most positive impact. These simulations were adapted to each
family of devices.
For smartphones and tablets, we identified 6 major types of consumer and conceptualised their impacts
over a 6-year period:
 Compulsive buyers of refurbished but recent smartphones/tablets: purchase a 1-year-old

smartphone/tablet once every 2 years;
 Measured buyers of refurbished smartphones/tablets: purchase a 2-year-old

smartphone/tablet once every 2 years;
 Environmentally conscious buyers of refurbished smartphones/tablets: purchase a

smartphone/tablet over 3 years old once every 3 years;
 Compulsive buyers of new smartphones/tablets: purchase every year;
 Measured buyers of new smartphones/tablets: purchase every 3 years;
 Environmentally conscious buyers of new smartphones/tablets: purchase every 6 years.
For laptop and desktop computers, we modelled 6 purchasing behaviours over a 10-year period:
 Regular buyers of refurbished but recent computers: purchase a 2-year-old computer once
every 3 years;
 Measured buyers of refurbished computers: purchase a 3-year-old computer once every 4

years;
 Environmentally conscious buyers of refurbished: purchase a computer over 5 years old once
every 5 years;
 Measured buyers of new computers: purchase every 5 years;
 Environmentally conscious buyers of new computers: purchase every 8 years;
 Compulsive buyers of new computers: purchase every 3 years.

3. Environmental impact indicators and data
3.1. Types and source of data

Three different types of information are required when calculating an LCA:
 Data on the physical and usage characteristics of the systems under consideration (such as
the technical parameters of the reference devices: screen size, RAM, etc., and the lifespans to
be taken into account). For this project, this data was taken from the refurbishers’ own sales
statistics.
 Data on incoming and outgoing flows involved in the refurbishment process. This data is
collected directly in the field from several refurbishers;
 Data on the impacts of the life cycle of computer equipment, or of energy flows entering the
system under consideration. This data is taken from the databases available in the EIME
software package:
‐ In its generic database;
‐ In the database from the NegaOctet20 project, where this data has been critically reviewed;
‐ In the Ecoinvent database, version 3.6, where data was missing or obsolete.
‐ In the ELCD database.
3.2. Data on physical properties
3.2.1. Reference models and technical characteristics

Data on the devices’ technical characteristics is taken from:
 For the refurbished reference models selected: refurbishers’ sales statistics for 2020;
 For the new reference models selected: the market data used in the ADEME-ARCEP study ‘The
environmental impact of the digital sector in France’, published in December 2021.
3.2.2. Lifespans considered

Defining the lifespan of electrical and electronic devices is no easy matter, yet it is a crucial aspect of
assessing the environmental impacts of the devices. In this study, the key parameter considered is the
duration of use. The duration of use, as defined in the study on the lifespan of electrical and electronic
devices21, corresponds to the period of time for which the product is used, i.e. is functional and ready for
use, by a given user. Precise durations of use are not known. The table below sets out the identified data
taken from literature. This data does not indicate any changes in ownership or any refurbishment.
Smartphones Tablets Desktop Laptop

COMPUTERS COMPUTERS
Total duration of use taken 20 months to 6 5 years19 3–12 years19 5–8 years19
from literature years19
Table 1 – Lifespans identified in literature
20
Negaoctet.org
21
2012; ÉTUDE SUR LA DURÉE DE VIE DES ÉQUIPEMENTS ÉLECTRIQUES ET ÉLECTRONIQUES (STUDY ON THE
LIFESPAN OF ELECTRICAL AND ELECTRONIC DEVICES), Study carried out on behalf of ADEME by BIO Intelligence
Service S.A.S. (Shailendra Mudgal – Benoît Tinetti – Thibault Faninger – Sarah Lockwood – Gina Anderson) Technical
coordination: Erwann Fangeat - EPR and Recycling Department - Sustainable Consumption and Waste Division -
ADEME Angers

Thus, the durations considered in this study are taken from professional opinions, market analyses and
analyses of literature so as to achieve a conceptualisation that is realistic. A sensitivity analysis was
performed on this data, as was a behavioural renewal analysis.
Smartphones Tablets Desktop Laptop

COMPUTERS COMPUTERS
Duration of use 1 3 years 3 years 5 years 5 years

–new product
Duration of use 2 2 years 2 years 3 years 3 years

–refurbished product
Table 2 – Lifespans considered within the study context
3.3. Data on incoming and outgoing flows involved in the refurbishment

process
3.3.1. Overview
The data to be processed for each stage in the synopsis is the following:
Incoming flows: Outgoing flows:
 Supply scenario: masses of devices, masses  Processing packaging waste;

and materials of the shipping packaging,
 Processing cleaning waste;
distances covered, supply models, load
rates;  End-of-life processing of worn parts;
 Consumption data per stage of the  Emissions into the air.

process:
‐ Energy: Electricity, Gas, etc.;
‐ Water;
 Consumption of auxiliary materials;
‐ Consumption of detergent;
‐ Consumption of cloth;
 Supply and manufacture of spare parts;
 Supply and manufacture of accessories;
 Supply and manufacture of packaging

materials.

Figure 9 – Data collected during the modelling of the refurbishment process

3.3.2. Refurbishment process – Stakeholders and data collected
The refurbishment synopsis is set out below. It applies to each device family:
Step Description Type of data
MANUFACTURE
Step 0 – Upstream Logistics – This step represents the process of Secondary data – Default
Collection from users collecting the devices from the private scenario
from the previous life and transport to individuals/professionals at the first
the centralisation centre.
centralisation centres It includes individual or batch transport
from the home/workplace of the user
and centralised transport between the
collection point and the centralisation
centre.
Step 1 – Supply – This step corresponds to transport from Primary data completed and
From the centralisation centre the centralising party to the refurbisher averaged out for all
to the refurbisher and the end-of-life processing of the refurbishers
packaging in which the devices are
supplied.
Step 2 – Reception/Testing/Sorting This step corresponds to receiving and Primary data completed and
testing the products on unloading by the averaged out for all
haulier, deleting the data and checking refurbishers
the products’ identity. The impacts
associated with this step are the impacts
from the refurbishment site.
Step 3 – Diagnostics/Expert Assessment This step corresponds to the detailed Primary data completed and
analysis of faults on the devices. The averaged out for all
impacts associated with this step are the refurbishers
impacts from the refurbishment site.
Step 4A – Refurbishing This step represents the repair and Primary data completed and
the devices restoration of the devices. averaged out for all
refurbishers
Step 4.1 – Testing/Reset Primary data completed and

Step 4.2 – Cleaning averaged out for all
Step 4.3 – Replacing the Screen refurbishers. Extrapolation of
Step 4.3 – Replacing the Battery the type of part depending on
Step 4.3 – Other the reference models.
Step 4B – Refurbishing This stage corresponds to the impacts of This process is deemed
spare parts refurbishing spare parts (which may be identical to the process of
done by an external service provider). refurbishing a complete
device.
Step 5a – Placing in stock for sale – This step represents the outgoing checks Primary data completed and
Outgoing checks before the device is placed on the averaged out for all
market. These impacts are included in refurbishers
the site’s consumption.
Step 5b – Placing in stock for sale This step covers the incorporation of Primary data averaged out for
and order preparation accessories and packaging as well as all refurbishers
transport from the refurbishment site to
the refurbisher, if the latter is a separate
party.
DISTRIBUTION
Step 6 – Distribution This step includes: Secondary data – Default

‐ Transport to the distribution scenario
centre where necessary (where
subcontractors from abroad
are involved, for example);
‐ Transport from the distribution
platform to the end user.
USE
Step 7 – Use of the smartphone This step corresponds to the usage of the Secondary data – Default
device. We considered several scenario
hypotheses. The aim of incorporating this
stage is to put the results into
perspective in view of the use phase
impacts.

Table 3 – Description of the refurbishment synopsis considered for each product family
This collection synopsis enabled us to model the impacts of refurbishment thanks to the contribution
from national and international refurbishers:
Smartphones Tablets Laptop Desktop computers

computers
Primary or secondary Primary Extrapolation Primary Primary

data based on
smartphones
Number of 11 9 7
refurbishers who
provided
data
Parties involved in Agora Place, Ateliers du Ateliers du bocage; ATF bis repetita,
collection Backmarket, Bak2, Itancia, Largo, Le bocage; ATF bis ATF GAIA, Atoutek, Ingram, LM
GSM, repetita, ATF Ecoproduction, Recyclea
Recommerce, GAIA, CRS,
Sofigroupe, Riitekpro + 2 Emmaus
refurbishers (CONFIDENTIAL) Connect,
Ingram, LM
Ecoproduction,
Okamac,
Recyclea
Location of France – Germany – Romania – France – United France

refurbishers China – Hong Kong Arab Emirates
questioned
Year in which data 2021 2021 2021

was collected
Collection period 2020 2020 2020
Table 4 – Collection process per device family
As part of this study, we carried out a complementary study on servers and consoles. These product
categories were extrapolated. They are set out in an appendix.
3.3.3. Transport data

Data pertaining to upstream and downstream transport can be of 2 types:
 Data released by manufacturers relating to means of transport, locations and distances;
 Data estimated by the LCA analyst based on the refurbishment location and the supply
location.
Freight Transport
For all freight transport, we considered the hypotheses set out in the table below. Generally speaking,
intercontinental transport takes place by plane. A sensitivity analysis for supply by ship was carried out.

Destination France Germany Europe - Romania – Bucharest Asia - Shenzhen United Arab Emirates - Dubai
ORIGIN
PLANE (BERLIN)
(CLASSIC.SEAROUTES.COM)
PLANE (PARIS)
(HAMBURG)
(LE HAVRE)
LORRY
LORRY
LORRY
LORRY
LORRY
PLANE
PLANE
PLANE
SHIP
SHIP
SHIP
SHIP
SHIP
Asia-Plane (Shenzhen base) 10000 1000 10000 1000 10000 1000 5000 1000 6000 1000
Asia-Ship (Shenzhen) 1000 22000 1000 18000 1000 15000 1000 2500 1000 9000
US-Plane (Denver) 8000 1000 8000 1000 10000 1000 12000 1000 12000 1000
US-Ship (average between 1000 10000 1000 10000 1000 11000 1000 15000 1000 23000
NY
and San Francisco)
Australia-Plane (Adelaide) 19000 1000 15000 1000 15000 1000 5000 1000 11000 1000
Australia-Ship (Adelaide) 1000 20000 1000 22000 1000 22000 1000 10000 1000 11500
France-Plane (Paris) - - - 1000 1000 2000 1000 10000 1000 5200 1000
France-Ship (Le Havre) - - - - - - - - - 1000 22000 1000 11000
France-Lorry (average) 1000 2000 3500 - - - 6500
Europe-Plane (Frankfurt) 2000 1000 2000 1000 2000 1000 10000 1000 4800 1000
Europe-Lorry 3500 35000 3500 - - - 6500
Europe-Ship (Rotterdam) - - - - - - - - - 1000 22000 1000 11000
Table 5 – Transport distance matrix
A NOTE TO THE READER: the table is read as follows, for the data circled in red. For transport from Asia to France by ship, we reckoned on 1000 km by lorry and
22,000 km by ship.

Transport scenario – Purchase and drop-off by a private individual
– B2C
We additionally elected to model the transport taken by private individuals in the following reference
scenario.
%clients Purchase Scenario Means of Distance Unit

concerned - attribution transport
Reference
scenario22*
Shop/ 54% 50% Urban Public 2 km Passenger.km
Collection point transport return trip
Shop/ 24% 50% Peri-urban Car 10 km Passenger.km
Collection point return trip
Shop/ 11% 50% Rural Car 20 km Passenger.km

Collection point return trip
Home 11% pooling Rural Van 20 km kg.km

delivery return trip
Table 6 - Reference scenario - means of transport for acquiring the equipment.
This scenario is created by combining:
 the selected transport hypotheses according to estimations based on prior work carried out
by the service providers;
 a client distribution equal to the population distribution.
Data on modelling for new devices
New devices were not modelled separately for this study; we adopted the hypotheses and models used
as part of the 2021 ADEME-ARCEP study23. This data has been critically reviewed and will be provided in
the Impact® database (in progress).
Details of how the data was broken down and analysed are appended to this report.
The input parameters are:
 Size and type of screen;
 CPU capacity, RAM, storage;
 Device mass;
 Associated accessories.
3.3.4. Data on modelling energy impacts during the use phase

Electrical data is assessed as a French electricity mix. In order to be consistent with the data collection
period (2019), we selected the most recent reliable electricity data available, taken from RTE for 201824.
22
Source for the figures on distribution throughout the population of France:
http://www.observationsociete.fr/territoires/lieu-de-vie_terri/ville-campagne-periurbain-qui-vit-ou.html
23
‘The environmental impact of the digital sector in France’, published in December 2021
24
https://www.iea.org/data-and-statistics/

France Europe China Romania Germany United Arab
Emirates
Inventory Electricity Electricity Electricity Electricity Electricity Production

name in EIME Mix; Mix; Mix; Mix; Mix; mix
Production Production Production Production Production reconstructe
mix; Low mix; Low mix; Low mix; Low mix; Low d based on
voltage; FR voltage; EU- voltage; CN voltage; RO voltage; GE IAE data
27
Inventory CODDE-2548 CODDE-2537 CODDE-2572 CODDE-2562 CODDE-2538
reference in
EIME
Type of Manufacture Manufacture Manufacture Manufacture Manufacture Manufacture

electricity
mix
Reference 2018 2018 2018 2018 2018 2018
year
Coal 1.8% 20.2% 66.5% 20.9% 37.3% 0%
Fuel oil 1.0% 1.7% 0.2% 0.8% 0.8% 1%
Gas 5.3% 19.0% 3.1% 14.0% 13.0% 98%
Bioenergy - 0.0% 0.0% 0.0% 0.0% 0.0% 0%

Biomass
Bioenergy - 1.1% 5.1% 1.3% 14.0% 7.0% 0%
Biogas
Bioenergy - 0.8% 1.5% 0.2% 0.6% 2.1% 0%

Waste
Nuclear 71.0% 25.3% 4.1% 15.1% 11.9% 0%
Pumping 0.0% 0.0% 0.0% 0.0% 0.0% 0%
Hydraulic 12.2% 11.6% 17.2% 24.0% 3.7% 0%
Solar 1.9% 3.9% 2.5% 2.3% 7.1% 1%
Wind 4.9% 11.5% 5.1% 8.4% 17.1% 0%

turbines
Other 0.1% 0.2% 0.0% 0.0% 0.0% 0%
sources
Table 7 - Information on mixes
The CODDE-2022-01 databases were used to assess the impact of electricity consumption during
refurbishment and during the use phase.
3.3.5. Data on the impacts of the life cycle

Data on the impacts of the life cycle of computer equipment or of energy flows is broken down into the
following categories:
 Primary data (also known as ‘site-specific data’) – data collected from a product
manufacturing/processing plant, and data taken from an analysis of the stages involved in the
life cycle of the product and of the system in which it is incorporated.
‐ E.g.:
 materials or electricity as measured by the service providers supplying them;
 transport, assessed based on fuel consumption and the associated actual emissions.
 Secondary data, divided into:

‐ selected secondary data – data taken from commonly available data sources that meet
the data quality levels required for precision, exhaustiveness and representativeness;
‐ proxy data – data taken from commonly available data sources (that do not meet all the
data quality levels required for ‘selected secondary data’).
In accordance with the objectives and limits of the system, no specific data was given priority. Most of
the data is taken from the databases available in the EIME software package (selected secondary data):
the NégaOctet25, CODDE26 and Ecoinvent 3.627 databases.
The DQR method was used to characterise the quality of all the data used.
TiR- Temporal representativeness TeR- Technological GR- Geographical

representativeness representativeness
1 The data’s publication date was The technology covered by the data The process covered by the data
during the validity period for the is exactly the same as the takes place in the country
study. technology necessary for the study. corresponding to the geographical
scope of the study.
2 The data’s publication date was a The technology necessary for the The process covered by the data
maximum of 2 years after the study is included in the data takes place in the geographical
validity period for the study. covered by the life cycle inventory. region corresponding to the
geographical scope of the study (e.g.
Europe).
maximum of 4 years after the study is partially covered by the takes place in one of the valid
validity period for the study. data included in the life cycle geographical regions for the
inventory. purposes of the geographical scope
of the study.
maximum of 6 years after the study is similar to that covered by does not take place in one of the
validity period for the study. the data included in the life cycle valid geographical regions for the
of the study but sufficient
similarities have been estimated
based on an expert opinion.
5 The data’s publication date wa The technology necessary for the The process covered by the data
smore than 6 years after the validity study is different from that covered does not take place in one of the
period for the study. by the data included in the life cycle valid geographical regions for the
of the study.
Table 8 - Description of the DQR method applied during the study
The data used is appended to this report.
3.3.6. Attribution procedures
Managing missing data

The collected data received is of heterogeneous quality. Some refurbishers have complete, documented
information, whereas others track certain headings only. Thus, where data was missing, the values
adopted correspond to the maximum value of the unit data for all refurbishers who did declare their
values.
General allocations – Data aggregation

With regard to data collection relating to refurbished devices, data was collected from several
refurbishers. Once the data had been collected, it was:
25
https://negaoctet.org/
26
https://codde.fr/en/our-services/software-tools
27
https://ecoinvent.org/

1. scaled down to one product unit based on the number of units processed and refurbished by
each refurbisher over the year. The allocation factor is the number of units refurbished arriving
at the end of the process. The data on the supply of defective and non-’refurbishable’ devices
was thus incorporated into the process:
2. supplemented by maximising data where no data was available following a check with the
refurbisher;
3. averaged to an arithmetical average for all refurbishers.
Thus, a fictional reference refurbishment model was created. This model will be used as a basis for
performing all extrapolations and sensitivity analyses.
The formula below summarises the method:
𝑉
𝐼𝑚𝑝𝑎𝑐𝑡 𝐼𝑚𝑝𝑎𝑐𝑡 𝐶 𝐼𝑚𝑝𝑎𝑐𝑡
𝑁
𝐼𝑚𝑝𝑎𝑐𝑡
𝐼𝑚𝑝𝑎𝑐𝑡
𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑
Where
𝑉 = Average values collected from refurbishers and supplemented where there was missing data
= Annual consumption for the process, transport distances, mass of consumables, etc.
𝐼𝑚𝑝𝑎𝑐𝑡 = Unit impacts taken from life cycle inventory databases for the manufacturing, distribution and
end of life stages
𝑁= Number of devices refurbished
= Number of devices supplied (Number of devices for parts + Number of devices out of service - Number
of devices for direct resale (Brokers)
𝐶= Electricity consumption over the lifespan in kWh
𝐼𝑚𝑝𝑎𝑐𝑡 = Unit impacts of the electricity mix
The figure below clarifies the concepts and how the different fractions were processed, as well as whether
or not they were incorporated within the limits of the system.
Figure 10 – Choice of the allocation factor
Method for creating refurbishment models

The method for creating refurbishment models is as follows:

 Creation of a reference model:
‐ Data collected by refurbisher;

‐ Each incoming and outgoing flow is allocated to the unit impact28 per manufacturer;
‐ Supplement for missing data with the maximum value collected;
‐ Arithmetical averages created per unit flow;
‐ Reference model created.
 Adaptation of the reference model to each refurbishment scenario:
‐ The ‘change parts’ section is extracted and a basic cleaning model created;
‐ Variants added:
Figure 11 – Creating refurbishment variants
 Adaptation of the model to each location, in terms of both the supply and the location of the
refurbisher:
‐ Supply scenario adapted;
‐ Energy mixes adapted.
Taking account of second-hand parts

In some cases, refurbishers incorporate second-hand parts. To model the impacts of these parts, we
adopted the following hypothesis:
 Second-hand parts follow the same logistical pattern as devices for refurbishment;
 They are disassembled from the devices arriving at the refurbisher’s site;
 Their impact is equivalent to a reference basic cleaning without packaging or accessories.
28
Impact scaled to the number of units refurbished at the end of the process

Allocation for end of life
Electrical and electronic equipment and subassembly

Recycling and reclaiming materials taken from electrical and electronic equipment at the end of its life
cycle are considered, using the circular footprint formula (CFF), as applied by the Ecosystem end-of-life
database29. There are two approaches in this method: the ‘without benefits’ approach, and the ‘with
benefits’ approach. The approach adopted is the ‘with benefits’ approach. This method assumes that
recycling or recovering the energy from materials at the end of their lifespan offers a benefit associated
with substituting out virgin materials or primary energy sources.
The CFF applied for the ‘with benefits’ approach is as follows:
 Impacts of materials recycling;
𝑄
1 𝐴 𝑅2 ∗ 𝐸𝑟𝑒𝑐𝑦𝑐𝑙𝑖𝑛𝑔𝐸𝑜𝐿 𝐸∗ ∗
𝑄
Where:
‐ A: distribution factor for the charges and credits between the supplier and the user of the
recycled materials. Under the Ecosystem approach, A=0.
‐ R2: the proportion of materials in the product that will be recycled (or reused) in a
subsequent system.
‐ ErecyclingEoL: specific emissions and resources consumed (per functional unit) and
resulting from the process of end-of-life recycling, including the collection, sorting and
transport processes.
‐ E*v: specific emissions and resources consumed (per functional unit) and generated by the
acquisition and production of the virgin materials replaced by the recycled materials.
‐ QSub: quality of the recycled material at the substitution point.
‐ Qp: quality of the raw material.
 Impact of the energy recovery:
1 𝐵 𝑅3 ∗ 𝐸𝐸𝑅 𝐿𝐻𝑉 ∗ 𝑋 , ∗𝐸 , 𝐿𝐻𝑉 ∗ 𝑋 , ∗𝐸 ,

Where:
‐ B: allocation factor for the energy recovery processes. Applies both to charges and to
credits. In the Ecosystem approach, B=0.
‐ R3: the proportion of materials in the product that is used as part of end-of-life energy
recovery.
‐ EER: specific emissions and resources consumed (per functional unit) associated with the
energy recovery process (e.g., incineration with energy recovery, landfill dumping with
energy recovery, etc.).
‐ LHV: lower heating value of the material from which energy is being recovered.
‐ XER,heat: efficiency of energy recovery for heat.
‐ ESE,heat: specific emissions and resources consumed (per functional unit) that would have
been generated by the heat replaced by the energy recovered.
‐ XER,elec: efficiency of energy recovery for electricity.
‐ ESE,elec: specific emissions and resources consumed (per functional unit) that would have
been generated by the electricity replaced by the energy recovered.
29
http://weee-lci.ecosystem.eco/Node/

 Impacts of waste disposal:
1 𝑅2 𝑅 3 ∗ 𝐸𝐷
Where:
subsequent system.
‐ R3: the proportion of materials in the product used as part of end-of-life energy recovery.
‐ ED: specific emissions and resources consumed (per functional unit) and resulting from
waste disposal at end of life for the product analysed, without energy recovery.
Industrial and domestic waste

For present purposes, industrial and domestic waste is defined as packaging waste (paper, cardboard,
pallets, plastic) and cleaning waste, particularly cloths.
The approach considered employs the stocks/without benefits method. This method assumes that
recycling or recovering the energy from materials at the end of their lifespan does not offer any benefit
associated with substituting out virgin materials or primary energy sources, and that all the impacts from
recycling operations are attributed to the new material.
Impacts of materials recycling;
𝑅2 ∗ 𝐸𝑟𝑒𝑐𝑦𝑐𝑙𝑖𝑛𝑔𝐸𝑜𝐿
Where:
 R2: the proportion of materials in the product that will be recycled (or reused) in a subsequent
system.
 ErecyclingEoL: specific emissions and resources consumed (per functional unit) and resulting from
the process of end-of-life recycling, including the collection, sorting and transport processes.
In the present case, the impacts of recycling are attributed to the production of the recycled
material. The impact considered is from collecting the waste (100 km by lorry) and from pre-
processing.
 Impact of the energy recovery:
𝑅3 ∗ 𝐸𝐸𝑅
Where:
‐ R3: the proportion of materials in the product that is incinerated at end of life.
‐ EER: specific emissions and resources consumed (per functional unit) associated with the
incineration process. Since the yields from domestic waste incineration centres were often
<65%, we considered that incineration took place without energy recovery as a matter of
course.
‐ Impacts of dumping waste in landfill:
1 𝑅2 𝑅 3 ∗ 𝐸𝐷
Where:
subsequent system.
‐ R3: the proportion of materials in the product that is incinerated.
‐ ED: specific emissions and resources consumed (per functional unit) and resulting from
waste disposal at end of life for the product analysed, without energy recovery.

Values of R2 & R3 factors
Wood Paper/Cardboard Plastic Soiled fabric
R2 31%+5% 8% 52%
R3 21% 8% 42% 100%
1-R2-R3 43% 85% 6%
Source Reuse of packaging in France - 2019 data – ADEME Hypothesis
Table 9 - Rates of recycling and incineration for packaging waste
3.4. AILC methodology and types of impact
3.4.1. Selection, classification and characterisation of impacts

The indicators adopted as part of this study are the mandatory indicators from the most recent reference
guide to environmental labelling published by ADEME for digital services30. This list of indicators was
drawn up using the EF 3.0 characterisation/weighting method.
This list was supplemented by the material footprint (MIPS), water requirements (WU) and production of
WEEE, to enable a better understanding of the environmental impacts.
The following indicators were thus selected:
Priority of Category of impact Model Unit LCIA method

interpretation recommendation
level
IMPACT INDICATORS
I Climate change IPCC 2013, GWP 100 kg CO2 eq I
I Depletion of abiotic, mineral and ADP for mineral and metal kg Sb eq III
metal resources resources, based on van Oers et al.
2002 as implemented in CML, v.
4.8 (2016)
I Fine particle emissions Fantke et al., 2016 disease I

incidence
I Ionising Radiation Frischknecht et al., 2000 kBq U235 eq II
II Depletion of abiotic resources ADP for energy carriers, based on MJ III

and fossil fuels Oers et al. 2002, as applied in CML,
v. 4.8 (2016)
II Acidification Posch et al., 2008 mol H+ eq II
SUPPLEMENT
I Raw materials MIPS – Material Input per Service Kg N/A

Unit, Schmidt-Bleek, 1994 and
Ritthoff et al., 2002
I Waste production from electric Kg N/A

and electronic devices
II Water requirements AWARE 100 (based on Boulay et al., m3 world eq III

2018)
Table 10 - Selection of appropriate indicators based on standardisation and weighting
To give a better understanding of the indicators selected, the following table sets out each indicator
together with an explanation of the environmental aspects associated with it:
Depletion of natural resources Climate change

(minerals and metals)  Type of indicator: Problem-orientated impact
INDICATORS
 Type of indicator: Problem-orientated indicator (mid-point)

impact indicator (mid-point)  Unit: kg CO2 equivalent (kg CO2 eq.)
IMPACT
 Unit: kg Sb equivalent (kg Sb eq.)  Assessment method: IPCC 2013 method

 Assessment method: CML 2002  Definition: Greenhouse gases (GHGs) are gaseous
compounds that absorb the infrared radiation
30
General principles for the environmental labelling of consumer products - Methodological reference guide for
assessing digital services, ADEME, July 2021

 Definition: Industrial exploitation causes emitted by the Earth’s surface. An increase in
a reduction in the finite amount of their concentration in the Earth’s atmosphere
resources available. This indicator contributes to global warming.
assesses the quantity of mineral and
metal resources extracted from the
ground as antimony-equivalent.
Acidification Fine particle emissions

 Type of indicator: Problem-orientated  Type of indicator: Problem-orientated impact
impact indicator (mid-point) indicator (mid-point)
 Unit: mol H+ eq.  Unit: Disease incidence
 Assessment method: Accumulated  Assessment method: PM method
Exceedance (Seppälä et al. 2006, Posch recommended by UNEP (UNEP 2016)
et al, 2008)  Definition: The presence of fine particles of low
 Definition: The acidification of the air has diameter in the air – particularly particles with
been linked to emissions of nitrogen a diameter of under 10 microns – constitutes a
oxides, sulphur oxides, ammonia and problem for human health, as breathing in
hydrochloric acid. These pollutants are these particles can cause respiratory and
converted to acids when exposed to cardiovascular issues.
humidity, and their precipitation can
harm ecosystems as well as buildings.
Ionising Radiation Depletion of (fossil) abiotic resources

 Type of indicator: Problem-orientated  Type of indicator: Problem-orientated impact
impact indicator (mid-point) indicator (mid-point)
 Unit: kBq U235 eq.  Unit: MJ
 Assessment method: Human health  Assessment method: CML 2002
effect model as developed by Dreicer et  Definition: This indication represents the
al. 1995 (Frischknecht et al, 2000) primary energy consumption from different
 Definition: Radionuclides can be released non-renewable sources (oil, natural gas, etc.).
during several human activities. When The calculations are based on the Lower
radionuclides decay, they release ionising Heating Value (LHV) for the types of energy
radiation. When human beings are considered, expressed in MJ/kg. For example, 1
exposed to ionising radiation, this can kg of oil will contribute 41.87 MJ to the
result in damage to their DNA, which in indicator considered.
turn can lead to various types of cancer
and birth defects.
Water usage
 Type of indicator: Problem-orientated
impact indicator (mid-point)
 Unit: m3 eq.
 Assessment method: AWARE 100 (based
on Boulay et al., 2018)
 Definition: The water used in an open
loop is taken into consideration.
Mass of electrical and electronic waste generated (EW)

 Type of indicator: Flow indicator
 Unit: kg
 Definition: Quantity of computer equipment mobilised per functional unit and that will virtually have
to be reprocessed.
Raw materials - MIPS

 Type of indicator: Resource consumption indicator
FLOW INDICATORS
 Unit: kg
 Assessment method: MIPS - Material Input per Service Unit
 Definition: The MIPS indicator is used to calculate the resources used to produce one product or
service unit, taking a life cycle analysis approach (Schmidt-Bleek, 1994). Five types of resources
are considered: abiotic resources (materials, fossil energy, etc.), biomass, earth moved
mechanically or by erosion, water, and air (Rithoff et coll., 2002). These consumption figures are
then simply added together, to give a resource consumption indicator (extracted raw materials
and energy raw materials).
Table 11 – Description of indicators
3.4.2. Standardisation
The numerical results from the indicators can be ordered, standardised, consolidated and weighted. This
approach makes interpretation easier, but there is no scientific consensus on a robust method for
carrying out this assessment.
In our study, we used the standardisation factors provided by JCR in the PEF/OEF method (EF 3.0)
published on 20 November 2019, as set out in the table below. No weighting is envisaged.

Category of impact Standardisation factor Unit
Climate Change 8.10E+03 kg CO2 eq. per person
Depletion of abiotic, mineral and 6.36E-02 kg Sb eq. per person

metal resources
Fine particle emissions 5.95E-04 disease incidences per person
Depletion of abiotic resources and 6.50E+04 MJ per person

fossil fuels
Ionising radiation and human health 4.22E+03 kBq U-235 eq. per person
Acidification 5.56E+01 mol H+ eq. per person
Water usage 1.15E+04 m3 water eq of deprived water per person
Table12 – Standardisation factors proposed by JRC
3.4.3. Planetary boundaries

An appraisal was carried out using the planetary boundaries method as part of this study. The planetary
boundaries approach is a concept that makes it possible to compare environmental impacts with
planetary boundaries, which is a framework used to estimate the extent to which human activities respect
or exceed an operating space sustainable for humanity.
The planetary boundaries approach has been fully incorporated into LCAs. Nonetheless, JRC provided
factors that enabled us to connect the LCA results and the planetary boundaries31.
The factors used are set out below:
Category of impact Factor - Planetary boundaries Unit
Climate Change 985 kg CO2 eq. per person
Depletion of abiotic resources, 3.18E-2 kg Sb eq. per person

minerals and metals
Fine particle emissions 7.47E-5 disease incidences per

person
Depletion of abiotic resources, 32,400 MJ per person

fossil fuels
Ionising radiation and human health 7,620 kBq U-235 eq. per person
Acidification 145 mol H+ eq. per person
Water usage 26,300 m3 water eq of deprived

water per person
Table 13 – Factors for planetary boundaries
3.5. LCA modelling tool

For each refurbishment scenario, the material and energy resource flows from nature to the technical
system, and the emissions from the technical system into the air, the soil and the water were taken into
account using a special Microsoft Excel calculator. All of the emission factors used were taken from
version 5.9.1. of the EIME software and the associated databases: CODDE-2020-12, ELCD and
NEAGAOCTET. EIME conforms to the ILCD manual (level I entry).
With regard to missing or obsolete data, we used version 3.6 of the Ecoinvent database.
31
2019. Consumption and consumer Footprint: methodology and results – Indicators and assessment of European
consumption, JRC

4. Smartphones and tablets family
4.1. Overview
Definition of a smartphone: a smartphone is a mobile telephone that performs multiple computing
functions and typically has a touchscreen, WiFi and mobile internet access, a GPS connection and an
operating system (OS) capable of running downloaded applications.
Definition of a tablet: A tablet is a type of portable computer with a touchscreen and potentially a
physical keyboard connected to it.”32
4.2. Defining reference models

The reference models were designed so as to:
 Represent the reality of the markets of the refurbishment stakeholders that participated in the
study;
 Be compatible with all the work coordinated by ADEME in early 2022 so as to provide a
harmonised view of responsible practices in the digital sector.
Refurbished smartphone New smartphone
Sources of data Refurbishers (collected from 11) ADEME/ARCEP study - NegaOctet

database
Refurbishers Agora Place, Backmarket, Bak2,

Itancia, Largo, Le GSM, Recommerce,
Sofigroupe, Riitekpro + 2 other
refurbishers (confidential)
Types of data Primary Secondary
Reference model Average product based on refurbisher Average product based on the French
data collected market
Characteristics of the reference Refurbishment location: Manufacturing location:

model France (48%) Asia (100%)
Europe (26%)
Asia (26%)
Supply location: Supply location:

France (18%) Europe (44%) Asia (11%) Asia (100%)
US (25%)
Average characteristics: Average characteristics:

Mass: 180g Mass: 200g
Screen size: 5.6 inches Screen size: 6.61 inches
Type of screen: 60%LCD, 40%OLED Type of screen: 50%LCD,
Battery weight: 35g 50%AMOLED/OLED
RAM: 7.3GB
Storage: 180 GB
Circuit board area: 180cm²
Battery weight: 81g
Average part replacement:
12.28 cm2 of LCD screen
8.19 cm2 of OLED screen
9g of battery
0.5 non-electronic units
0.21 electronic units
0.03 printed circuit boards
Consumption during use phase 3.9 kWh per year 3.9 kWh per year
Location of use France France
Lifespan 2 years 3 years
Table 14 - Presentation of the characteristics of reference smartphones
Definition from the ICT report: European Commission, ICT Impact study, Final report, prepared by VHK and
32
Viegand Maagøe for the European Commission, July 2020, p.132

A discrepancy can be observed in this table between the technical characteristics of refurbished and new
reference models, which corresponds to the actual state of the market, as new devices are renewed based
on the latest innovations in the sector.
Electricity consumption: Electricity consumption is based on the average value for kWh per year as
determined in the ICT 2020 report, based on the endurance (tested by GSMArena33) of the 8 best-selling
smartphones in Europe in 2019, divided by the number of hours per year. The ICT report explains that
“The theoretical number of charges has then been multiplied by two to provide data for a more realistic
life scenario. The charges per year is multiplied by the battery capacity in Wh to give energy consumption
per year. The energy consumption is then divided by an efficiency of 75 % to estimate the losses in the
phone charger”.’ This gives 3.9 kWh per year.
N.B.: We cannot differentiate between consumption for new and refurbished devices. We thus opted for
an average consumption. It will nonetheless be noted that, in spite of an improvement in energy
consumption by the devices, the devices distributed are becoming ever-increasingly powerful, which
reduces the gap in energy consumption between new and refurbished devices.
Typical lifespan:
 For a new telephone: 3 years34:
 For a refurbished telephone: 2 years.
4.3. Hypotheses and entry data
4.3.1. Refurbishment
Profile of refurbishers
Data was collected from 11 national and international refurbishers. This work highlighted the fact that
profiles varied widely between refurbishers. The variant parameters are:
 Locations of refurbisher;
 Locations of refurbisher’s suppliers;
 Refurbishment practices:
‐ Number of parts changed;
‐ Types of parts changed;

‐ Use of refurbished parts.
The table below summarises the various profiles.
French refurbishers Asian European

refurbishers refurbishers
Company H
Company G
Company D
Company C
Company A
Company K
Company B
Company E
Company F
Company J
Company I
Locations France France France France France France France China Hong Germa Romani
Kong ny a
Number of units refurbished in 13 40 65 126 53 44 136 35 150 37 293

2020
- in thousand units
33
https://www.gsmarena.com/battery-test.php3
34
Renouvellement des terminaux mobiles et pratiques commerciales de distribution (Renewing mobile terminals and
sales practices for distribution) – ARCEP – June 2021

Type of refurbishment 10% 11% 17% 42% 50% 90% 126%* 95%** 170%* 99%** 42%
- Proportion of parts
changed
Type of parts - 3% 0% 50%*** 37% 4% 28% 33% 10% 0% 36% 37%

% of parts taken from
refurbishment
Main source of France 99% 85%

products supplied
Europe 100% 9% 1% 75% 8% 75% 24% 10% 95% 75%
Asia 44% 15% 15% 10% 30% 5% 15%
US 47% 10% 8% 10% 66% 60% 80% 10%
Australia 20%
*more than one part changed per mobile

**parts nearly always changed
***half of the parts replaced came from refurbishment
Table 15 – Reference refurbished smartphone - Presentation of refurbisher profiles - Summary
It should be noted that the refurbishment location/supply location combination has a direct influence on
the distance covered by the devices and the spare parts, as well on the distribution of the impacts
between upstream and downstream transport. The distances set out in Table 2 were used to characterise
the upstream and downstream transport for the various fractions.
Default characteristics
In order to define the impacts associated with each piece of data collected, hypotheses were drawn up
for the constituent elements of refurbished smartphones, spare parts and accessories.
Parts Characteristics Parts Characteristics
Mass of telephone (g) 180 Mass of cord (g) 13
Screen
Screen size (inches): 5.6 Screen size (cm2) 83.4
screen/% LCD 60% screen/% OLED 40%
Mass of screen (g) 45
Other characteristics of spare parts;
Mass of battery (g) 35 Mass of SIM card 3

extractor (g)
Mass of camera (g) 10 Mass of tempered 17

glass
Mass of miscellaneous part 10 Mass of 15

(g) miscellaneous
accessory (g)
Mass of charger (g) 35 Mass of earbuds (g) 10
Table 16 – Refurbished smartphone - Physical characteristics of the various elements considered for the reference
model
Data collected per FU

The table below sets out the entry data used in the model following processing for one refurbished
product unit.

Sub-step Ref. Min Max Unit
INVARIABLE PART
Supply – previous user to centralisation
Transport from the user to the collection point – estimated
Public transport 0.68 0.54 1.08 Passenger.km
Car 2.88 2.3 4.6 Passenger.km
Transport from the collection point to centralisation
Lorry 225.74 179.74 360 kg.km
Supply – centralisation to refurbishers
Transport from the centralisation centre to the refurbisher
Lorry 415.45 179.29 900 kg.km
Plane 1,185 97.91 3,888 kg.km
Packaging
Cardboard 0.06061 0.00358 0.12243 Kg
Plastic 0.01279 0.00423 0.02131 Kg
Paper 0.0031 0.003 0.00317 Kg
Pallet 0.03465 0.00057 0.05286 Kg
Miscellaneous Packaging (1/3 cardboard, 0.0263 0.0263 0.0263 Kg

1/3 paper, 1/3 plastic)
Consumption for the site
Electricity 1.06 0.28 1.24 kWh
Water 0.16 0 0.3 m3
Gas 1.93 0.01 3.86 MJ
Control
Paper (kg) 1.55E-03 7.73E-04 1.63E-03 Kg
Cleaning
Rubbing alcohol - Detergent - 0.21648 0.00096 3.54E-01 L

Disinfectant, etc.
Cloth 0.0097 0.00003 2.11E-02 Kg
Accessories
Charger
Manufacture 0.73 0.28 1 Unit
Supply
Lorry 30.65 9.75 87.5 kg.km
Ship 409.29 236.84 630 kg.km
Plane 240.66 97.48 380.75 kg.km
USB cable
Supply
Lorry 13.89 3.54 35 kg.km
Ship 248.86 85.94 630 kg.km
Plane 100.92 35.37 138.16 kg.km
Earbuds
Supply
Lorry 8.03 1.5 12.81 kg.km

Ship 106.79 27.07 180 kg.km
Plane 105.8 92.32 128.07 kg.km
Upstream packaging
Manufacture Cardboard 8.92E-02 1.09E-02 3.94E-01 Kg
Plastic 2.53E-02 2.11E-03 4.88E-02 Kg
Tempered glass 1.54E-04 5.68E-05 2.50E-04 Kg
Paper 5.99E-03 5.99E-03 5.99E-03 Kg
Supply
Lorry 1.21E+02 1.91E+01 4.49E+02 kg.km
Ship 1.82E+03 2.39E+02 5.27E+03 kg.km
Plane 6.33E+02 6.33E+02 6.33E+02 kg.km
Refurbishment to sale stock
Lorry 497.17 0 811.62 kg.km
Plane 2,118.00 0 2,277.00 kg.km
Distribution to points of sale
Lorry 215.47 180 237.7 kg.km
Points of sale to users
Public transport 0.54 0.54 0.54 Passenger.km
Car 2.3 2.3 2.3 Passenger.km
Truck 1.05 0.4 2.2 kg.km
CHANGING REFERENCE PARTS
Replacing screen
Manufacture
LCD 7.484 0.53 35.03 cm2
OLED 4.989 0.35 23.35 cm2
2nd-hand 7.998 0.11 44 cm2
Supply
Lorry 14.992 0.51 80.94 kg.km
Ship 20.446 20.45 20.45 kg.km
Plane 66.898 5.1 244.76 kg.km
Changing battery
Manufacture
Lithium 8.712 1.29 28 G
2nd-hand 0.305 0.07 0.7 G
Estimated transport
Lorry 15.886 1.84 82.38 kg.km
Ship 83.26 83.26 83.26 kg.km
Plane 65.243 18.36 188.99 kg.km
Changing printed circuit boards
Manufacture
Board 0.03 0.03 0.03 Unit
Estimated transport
Lorry 0.26 0.26 0.26 kg.km

Plane 2.58 2.58 2.58 kg.km
Changing miscellaneous electronics
Manufacture
Component 0.15 0.01 0.5 Unit
2nd-hand 0.06 0.01 0.11 Unit
Estimated transport
Lorry 3.51 0.11 13.75 kg.km
Plane 15.9 1.11 46.33 kg.km
Changing miscellaneous non-electronic components
Manufacture
miscellaneous (case, etc.) 0.5 0.41 0.59 Unit
Estimated transport
Lorry 5 4.08 5.91 kg.km
Plane 49.97 40.79 59.15 kg.km
Table 17 – Refurbished smartphone - Entry data for modelling the reference model scenario
4.3.2. New smartphones
To ensure that the results from this study are homogeneous with the studies already carried out on behalf
of ADEME as well as with Base Impacts® data, the reference smartphone model corresponds to a
weighted mix of three configurations (average model take from Base Impacts®), with the following
characteristics:
Configuration 1 2 3
Distribution (hypothesis) 50% 30% 20%
Device weight (kg) 0.204 0.189 0.202
Screen size (inches): 6.59 6.57 6.72
Screen technology LCD AMOLED OLED
Processor MediaTek Helio G85 SnapdragonTM 730G Exynos 990
RAM (GB) 6 7 11
SSD capacity (GB) 128 160 341
PWB surface area (cm²) 125.16 117.94 122.36
Battery weight (g) 84 79 77
Table 18 - Characteristics of the new smartphone considered
The data associated with the modelling of this device is set out in an appendix.
4.4. Results
4.4.1. Foreword
As set out above, a reference model was created. Based on this model, several variation parameters were
identified and will be examined in this report. These parameters correspond to:

 The location of the refurbishers;
 The consumption involved in the refurbishment process;
 The origin of the products to be refurbished;
 The type of refurbishment, from simple cleaning to the replacement of all used parts.
Figure 12 – Presentation of the variable elements in the refurbishment process
Subsequently, an analysis will be carried out on two levels:
 The full life cycle as broken down in the graph below by sectors and by sources of impact;
 The level of refurbishment only (scope delineated in red in the graph below):

Figure 13 – Reminder of the parameters for analysis and correspondence with the data collected.
The results will themselves be presented using the lifespan of the device or functional unit as a reference
scale.

Reminder - Abbreviations for indicators
ADP - Elements = Depletion of natural abiotic GWP = Climate change

resources - metals and minerals
IR = Ionising Radiation
ADP - fossil = Depletion of natural abiotic
PM = Fine particles
resources - fossil energy
WU = Water usage
AP = Acidification
MIPS = Ecological load
4.4.2. Assessment of the impact of a refurbished reference smartphone

used for 1 year
All impacts will be presented below, relative to the following functional unit:
‘Owning and using a smartphone for one year’
To scale the environmental impacts to the usage period, the impacts must be multiplied by the reference
usage period, i.e. 2 years.
Total results at a functional unit level (1 year of use)

For each year of use, the life cycle impacts of a reference refurbished mobile phone are as follows:
Impact Value Unit
REFERENCE PERIOD 2 Years
Climate change (GWP) 3.80 kgeqCO2 per FU
Ecological load (MIPS) 13 Kg per FU
Production of WEEE 11 g per FU
Depletion of natural abiotic resources kgeqSb per FU

- metals and semi-metals (ADPe) 2.98E-04
Depletion of natural fossil resources (ADPf) 102.64 MJ per FU
Water usage (WU) 6.79 m3eq per FU
Acidification (AP) 2.11E-02 kgeqH+ per FU
Fine particles (PM) disease occurrence

2.01E-07 per FU
Ionising radiation (IR) 8.95 kgU235eq per FU
Table 19 – Reference smartphone - Environmental impacts per functional unit for 1 year

Results broken down by stage of the life cycle
The environmental impacts are broken down as shown in the following figure:
Figure 14 - Reference refurbished smartphone - Results broken down by life cycle impact
Findings:
The refurbishment of devices represents the majority of the impacts (23%–94%). As for new devices, and
in spite of a lower impact during the manufacturing stage, the use phase shows lower impacts (excluding
ionising radiation and the depletion of fossil resources). The upstream and downstream share of logistics
represent 8–16% (for the indicators for which it has a significant impact).
Analysis:
Concentrating on the part incumbent on refurbishers and distributors is appropriate because the impact
of this stage is determined by the refurbishment location, the supply location and the type of
refurbishment.
The upstream, downstream and usage impacts are independent from the impacts of refurbishment, but
do show a greater impact relative to the entire life cycle, compared with a new scenario.

Results broken down by stage of refurbishment
To gain a better understanding of the environmental impacts over the refurbishment stage, a detailed
analysis was performed.
Figure 15 - Reference refurbished smartphone - Impacts broken down by stage - Focus on the refurbishment process
Findings:
Accessories and screen replacements are the main contributors to the impact on the depletion of natural
resources and on MIPS indicators.
Batteries and consumption on the site itself are the main contributors to water usage.
As for the other impacts, the contributions are spread evenly between the different headings.
Analysis:
Caution should be taken as to whether accessories need to be added and the battery needs to be
changed for every single device.
The approach adopted does allow a consistent reference model to be created, but the variations from
one company and from one scenario to another must be observed in order to identify the
environmentally optimum outcome.

Range of variation by type of refurbishment
Once the reference model was established, the average profile for part replacement was deleted in favour
of profiles corresponding to the replacement of a full screen, a battery or other parts. This analysis sets
out the variations based on the type of replacement made.
Figure 16 – Refurbished smartphone - Impact of replacing parts - Life cycle vision
Findings:
Significant variations in impact of between -28% and +79% were observed.
It should be noted that the reference screen corresponds to a mix composed of 60% LCD screens and
40% OLED screens. We observed that changing the screen had a very major impact, in particular LCD
screens on the depletion of natural resources, and OLED screens on climate change.
Changing the battery had a significant impact, particularly on water resources.
Analysis:
The type of parts replaced has a significant impact on all indicators, especially for screens.
With the incorporation of the use phase taken into account, the difference between the reference
scenario and the other scenarios is less significant than originally imagined, owing to the presence of new
accessories and to consumption during the use phase.

Once the reference model had been established, the supply profile was modified to create a variation for
the various impacts depending on the supply and refurbishment locations.
Figure 17 – Refurbished smartphone - Impact of the supply and refurbishment location - Life cycle vision (100% =
reference scenario)
Findings:
An increase in impacts on the depletion of fossil resources, climate change, acidification and fine particles
can be observed of up to +24% at the life cycle level, and +34% where the supply and refurbishment
locations are far away.
An increase in the impact of ionising radiation can also be observed.
Analysis:
The first increase mentioned above is associated with the contribution made by transport to the overall
impact of refurbishment, and with the combustion of fossil fuels (since the preferred means of transport
was express delivery by plane). A sensitivity analysis for means of transport must be carried out in order
to consolidate these discrepancies.
The increase in ionising radiation is associated with the French electricity mix.
Range of variation of impact depending on the refurbishers

The reference model is fictional. This model represents a weighted average of the market data
transmitted. Within the sample used to generate the reference model, the profiles for refurbishment
companies/methods are very different. Namely:
 The replacement rates for parts vary from 10% (0–1 part replaced, infrequently) to 170%
(multiple parts replaced, more frequently) from one refurbisher to another.
 The usage rate for second-hand parts is variable: from 0 to 50% of parts.
 The estimated average distances for the total route travelled by a smartphone vary by a factor
of up to 24 from one refurbisher to another, depending on the location of the refurbisher and
its suppliers.

Thus, the range of variation between unit data per company and per reference model was evaluated:
Impact Reference Value Value Unit

value (min) (max)
Reference period 2 2 2 years
Climate change (GWP) 3.80 2.43 4.94 kgeqCO2 per FU
Ecological load (MIPS) 13 6 17 Kg per FU
Depletion of natural abiotic resources – metals and 2.98E-04 1.45E-04 3.69E-04 kgeqSb per FU
semi-metals (ADPe)
Depletion of natural fossil resources (ADPf) 102.6 84.6 116.2 MJ per FU
Water usage (WU) 6.79 1.61 13.63 m3eq per FU
Acidification (AP) 2.11E-02 1.37E-02 2.85E-02 kgeqH+ per FU
Fine particles (PM) 2.01E-07 1.56E-07 2.29E-07 disease

occurrence per
FU
Ionising radiation (IR) 8.96 7.82 16.60 kgU235eq per FU
Table 20 – Smartphone - Environmental impacts per functional unit for 1 year - Variations between companies
Figure 18 - Refurbished smartphone - Variation of impacts depending on the companies - Impacts on climate change

Figure 19 - Refurbished smartphone - Variation of impacts depending on the companies - Impacts on the depletion
of abiotic mineral resources
Figure 20 - Refurbished smartphone - Variation of impacts depending on the companies - Impacts on the depletion
of abiotic fossil resources

Figure 21 - Refurbished smartphone - Variation of impacts depending on the companies - Impacts on ionising
radiation
Very significant variations in impacts can be observed from one company profile to another (from a factor
of 4 to 10 between the minimum and the maximum, depending on the indicators).
For climate change:
 company E has the lowest impact – this is a French refurbisher with a supply coming almost
exclusively from France and a low rate of part replacement, with 50% of replaced parts coming
from refurbishment;
 company G has the highest impact. It is located in Hong Kong, gets its supplies from the US
and Australia, and changes 1.7 parts per device using new spare parts.
For the depletion of natural resources:
 Company D has the lowest impact: it adds the lowest amount of new cables and chargers.
 Company B has the highest impact, replacing numerous screens and routinely adding chargers,
earbuds and cables.
For ionising radiation:
 Company D has the lowest impact. Company K has the highest impact. Impact is in direct
correlation with site’s consumption and the use of the French energy mix. However, with
regard to this parameter, the question arises of whether the variation is associated with actual
differences in consumption or with a difference in the scope of the data (in spite of
considerable consolidation work).

Findings:
A very significant range of variation of impact can be observed from one refurbisher to another. This can
be explained by different locations, logistical scenarios and methods of refurbishment.
The range over the four reference indicators is relatively similar. However, the extremes do not
correspond to the same company profiles.
Analysis:
Refurbishers were selected based on their location so as to represent the French market: marketplace
suppliers and refurbishers retailing under their own name.
Over and above their geographic location, refurbisher profiles vary based on their supply models (more
or less selective) and their varied methods of refurbishment (from a simple cleaning model with no repairs
to a very high-security model where parts are systematically replaced), and this is reflected directly in the
analysis results . Using a reference model is appropriate, but the variation in impacts from one refurbisher
to another needs to be kept in mind. This variation is shown below:
Range of variation based on the lifespan

The second lifespan is a key parameter for the environmental impact of a refurbished mobile phone. In
the reference scenario, we chose a lifespan lower than the first lifespan so as to be in a realistic situation
with regard to both the behaviour and the total lifespan of the product.
Figure 22 - Refurbished smartphone - Variation of impacts depending on the lifespan of the refurbished smartphone
Findings:
Given the relatively low impact of a smartphone’s use phase, the environmental impact of a refurbished
smartphone varies extensively based on the second lifespan (the rule of proportionality on the
manufacture, distribution and end-of-life phases).

Analysis:
The second lifespan is a key parameter in terms of dictating the size of the environmental impact of
refurbishment. As for new smartphones, the longer the lifespan, the lower the impact.
Additional sensitivity analyses
Focus on screen type

ADP - ADP - AP (mol GWP IR (kg PM (Disease WU MIPS
elements fossil H+ eq) (kg CO2 U235 occurrence) (m3) (kg)
(kg SB eq) (MJ) eq) eq)
LCD – Screen 105% 100% 99% 97% 101% 99% 101% 102%
replacement in
the reference model
Reference smartphone 2.98E-04 102.64 2.11E-02 3.80 8.96 2.01E-07 6.79 12.64
(60% LCD 40% OLED) in
the scenario
model
OLED – Screen 93% 101% 102% 104% 98% 101% 99% 97%
replacement in
the reference model
Table 21 - Refurbished smartphone - Sensitivity analysis for screen type - For replacement in the reference model
ADP - ADP - AP (mol GWP (kg IR (kg PM WU MIPS

element fossil H+ eq) CO2 eq) U235 (Disease (m3) (kg)
s (kg SB (MJ) eq) occurre
eq) nce)
LCD – Screen replaced 123% 97% 93% 86% 108% 97% 107% 108%
with an LCD screen
Reference smartphone 4.23E-04 122.49 2.88E-02 5.26 10.14 2.47E-07 5.43 18.56
(60% LCD 40% OLED)
in the screen replacement
scenario
OLED – Screen replaced 65% 104% 110% 121% 88% 105% 90% 88%
by an OLED screen
Table 22 - Refurbished smartphone - Sensitivity analysis for screen type - For full screen replacement
Findings:
A limited variation was observed in the reference scenario, since the contribution represented by the
replaced screen is small. On the other hand, more significant variations in impact were observed for
screen replacements, including a strong variation in the ADP-elements indicator.
Analysis:
The screen type has a significant impact on all indicators when the screen is replaced. An OLED screen
will have a higher impact on the GWP indicator (associated with its complex manufacture and how recent
the technology is), but a lower impact on the ADP-elements indicator. Conversely, an LCD screen will have
a higher impact on the ADP-elements indicator (associated with the iridium contained in the display) but
a more limited impact on the GWP indicator.

Focus on batteries
ADP - ADP - AP (mol GWP (kg IR (kg PM (Disease WU MIPS
elements fossil (MJ) H+ eq) CO2 eq) U235 eq) occurrence) (m3) (kg)
(kg SB
eq)
Reference 2.98E-04 102.64 2.11E-02 3.80 8.96 2.01E-07 6.79 12.64

scenario
(35g)
75g 101% 101% 103% 102% 101% 101% 113% 103%
95g 101% 102% 106% 104% 102% 103% 127% 105%
Table 23 - Refurbished smartphone - Sensitivity analysis for battery mass
The average battery mass declared by refurbishers was 35g. This mass has been increasing over
generations, and this is generating an additional impact on products (particularly for the water criterion).
However, since changing the battery typically allows the second lifespan to be extended, its impact
should be qualified.
Focus on means of upstream transport

ADP - ADP - AP (mol GWP (kg IR (kg PM WU (m3) MIPS (kg)
element fossil H+ eq) CO2 eq) U235 (Disease
eq) nce)
Transcontinental 100% 116% 123% 133% 101% 105% 100% 121%

transport 100% Plane
Reference 2.98E-04 102.64 2.11E-02 3.80 8.96 2.01E-07 6.79 12.64

transport 100% Ship
Table 24 - Refurbished smartphone - Sensitivity analysis - means of upstream transport
Findings:
The choice of means of transport has a significant impact (-14% to +15%) on the environmental impacts
of a refurbished smartphone over 1 year.
Analysis:
Refurbishment with a short supply chain has a lower environmental impact. The choice of the means of
supply is key in terms of reducing the environmental impact of refurbishment.

Focus on means of distribution by the private individual
ADP - ADP - AP (mol GWP (kg IR (kg PM (Disease WU (m3) MIPS
elements fossil H+ eq) CO2 eq) U235 eq) occurrence) (kg)
(kg SB eq) (MJ)
Reference 2.98E-04 1.03E+02 2.11E-02 3.80E+00 8.96E+00 2.01E-07 6.79E+00 1.26E+01
Urban Distribution 90.62% 95.77% 95.65% 91.96% 99.77% 94.03% 99.42% 100.00%
> Shop/Collection
Point
Periurban 132.05% 117.55% 124.40% 133.05% 100.95% 132.41% 102.11% 100.00%

distribution >
Shop/Collection
Point
Rural Distribution 173.79% 140.91% 157.74% 177.01% 102.21% 176.61% 104.88% 100.01%
> Shop/Collection
Point
> Home delivery
Table 25 - Refurbished smartphone - Sensitivity analysis - Means of distribution
Findings:
The choice of means of transport to private individuals has a very significant impact (-10% to +70%) on
the impacts of a refurbished smartphone over 1 year.
Analysis:
The choice of the means of distribution is key in terms of limiting the environmental impact of
refurbishment. In particular, trips by private car to pick up the parcel should be avoided as much as
possible (as in the Peri-urban>Shop and Rural>Shop scenarios).
4.4.3. Standardisation of the results
Standardised results – world inhabitant approach

Standardisation involves bringing each indicator to a shared reference scale. In the present case, this scale
is the consumption of an average world inhabitant. All indicators are thus shown on the same scale.
Standardisation shows contrasting results depending on the indicators considered. The priority indicators
to be taken into account are the following:
 For the refurbishment phase (the priority phase for the purposes of this study):
‐ Depletion of natural mineral and fossil resources;

‐ Water usage;
‐ Climate change;
 For the use phase:
‐ Depletion of fossil resources;

‐ Ionising radiation (chiefly linked to the use phase and to use of the French energy mix).
Thus, some comparative analyses will be carried out on the ADPe, ADPf, IR and GWP indicators only.

Figure 23 – Reference refurbished smartphone model - Results for impacts after standardisation (world inhabitant)
- For the entire life cycle
Results scaled to planetary boundaries

Standardisation to planetary boundaries makes it possible to incorporate the concept of a sustainable
budget into standardisation, which is why we decided to incorporate it into the results presented. Thanks
to this standardisation, we can see a variation in priorities compared with the previous standardisation,
with a focus on:
 Depletion of natural mineral and fossil resources;
 Climate change;
 Fine particle emissions.
Figure 24 – Reference refurbished smartphone - Results for impacts after standardisation to planetary boundaries for
the entire life cycle

A note to the reader - the value shown in the graph is the percentage of the budget available for each
world inhabitant, respecting the planetary boundaries. Thus, we can say that using a refurbished
smartphone for 1 year consumes 0.94% of one world inhabitant’s budget in abiotic mineral resources.
ADP - ADP - AP GWP IR PM W

elements fossil
Standardisation factor 3E-02 3E+04 1E+02 1E+03 8E+04 7E-05 3E+04
Reference refurbished 0.94% 0.32% 0.01% 0.39% 0.01% 0.27% 0.03%

smartphone - Standardised
to planetary boundaries
Table 26 – Reference refurbished smartphone - Results for impacts after standardisation to planetary boundaries for
Thus, in order to present the results in a simpler manner, the choice was made to focus on:
 Climate change.
4.4.4. Comparison between new and refurbished smartphones
Comparison of the impacts of new and refurbished smartphones –

Substitution approach
Presentation of results for the reference scenarios

The table below acts as a reminder of the data for the 2 scenarios that will be used as references for the
remainder of the comparison.
Impact over the entire life cycle New Refurbished Unit

smartphone smartphone
Climate change (GWP) 85.2 7.61 kgeqCO2
Ecological load (MIPS) 268.6 25.3 Kg
Production of WEEE 200.20 22 G
Depletion of natural abiotic resources - metals and 2.50E-03 5.97E-04 kgeqSb

semi-metals (ADPe)
Depletion of natural fossil resources (ADPf) 1,269.1 205 MJ
Water usage (WU) 89.2 13.6 m3eq
Acidification (AP) 4.89E-01 4.23E-02 kgeqH+
Fine particles (PM) 2.91E-06 4.02E-07 disease

occurrence
Ionising radiation (IR) 36.3 17.92 kgU235eq
Table 27- Reference smartphone - Environmental impact for the entire life cycle for the reference lifespans (over the
entire life cycle)

Presentation of results – Impacts avoided
For the reference scenario, purchasing a refurbished smartphone and using it for 2 years instead of
purchasing a new smartphone and using it for 3 years allowed the following impacts to be avoided
annually:
Impact Value (per FU) Value (per market Percentage

in 2021)
ABSOLUTE VALUES ABSOLUTE VALUES
Reference period for new devices 3 years
Reference period for refurbished devices 2 years
Number of units sold 2.8 million units
Climate change (GWP) -24.6 kgeqCO2 per year -68,878 TeqCO2 per -87%
year
Ecological load (MIPS) -76.9 Kg per year -215,289 T per year -86%
Production of WEEE -55.75 g per year -156 T per year -84%
Depletion of natural abiotic resources - metals -5.36E-04 kgeqSb per year -2 TeqSb per year -64%
and semi-metals (ADPe)
Depletion of natural fossil resources (ADPf) -321 MJ per year -897,517,153 MJ per -76%
year
Water usage (WU) -22.9 m3eq per year -64,242,212 m3eq per -77%
year
Acidification (AP) -1.42E-01 kgeqH+ per year -397 TeqH+ per year -87%
Fine particles (PM) -7.70E-07 disease -2 disease occurrence -79%

occurrence per year per year
Ionising radiation (IR) -3.13 -8,768 TU235eq per -26%

year
Table 28 - Smartphone comparison - Substitution approach - Impacts avoided for the reference scenarios - 100%
corresponds to the impact of the reference device
The impact avoided varies depending on the scenario chosen, and is largely dependent on the lifespan
of the new product and the refurbished product. Avoided impacts are calculated as the impacts per FU
for refurbished smartphones minus the impacts per FU for new smartphones:

𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑠𝑒𝑐𝑜𝑛𝑑 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒 𝑈𝑠𝑎𝑔𝑒 𝑝𝑒𝑟𝑖𝑜𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑓𝑖𝑟𝑠𝑡 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒
At the scale of a private individual, substituting a new smartphone with a refurbished smartphone can
prevent:
 the emission of 24.6 kgeqCO2 per year, or 2.50% of the sustainable annual carbon footprint
budget, or 73.65 km by car;
 the extraction of 76.9 kg of materials;
 the production of 55.75 g of electronic waste.
At country level, the French market can prevent:
 the emission of 68,000 TeqCO2 per year, or 0.1% of the sustainable annual carbon footprint
budget, or 206 million km by car;
 the extraction of 215,290 tonnes of raw materials;
 the production of 156 T of electronic waste.
Variants and ranges of variation

Given the variability of the results from one refurbishment scenario to another, the comparison was
carried out using a substitution approach between:
 An average new product with a lifespan of 3 years;
 A fictitious reference refurbished product with a usage period of 2 years;
 A product refurbished with a short supply chain*, without accessories and with basic cleaning;
 A product refurbished with a short supply chain*, with new replacement parts, and
accessories;
 A product refurbished with a short supply chain*, with second-hand replacement parts, and
accessories;
 A product refurbished on a globalised market **, without accessories and with basic cleaning;
 A product refurbished on a globalised market **, with new replacement parts, and accessories;
 A product refurbished on a globalised market **, with second-hand replacement parts, and
accessories.
* Product collected and refurbished in France ** Product collected in the USA and refurbished in China
Figure 25 - Smartphone comparison - Substitution approach - Results’ - (100% = impact of new reference device)
Despite the conservative hypotheses for new devices, we observed that refurbished smartphones have a
significantly lower impact: -95 to -48% annual impact, excluding ionising radiation. The impact on ionising
radiation is explained by the part of the industrial process carried out in France for refurbishment.
Regardless of the configuration for refurbishment, the variation in impacts between the refurbishment
scenarios does not cancel out the beneficial effect of refurbishment practices.
Comparison of the impacts of new and refurbished smartphones –

Depreciation approach
Under the depreciation approach, we consider that if refurbishment takes place before the theoretical
end of the first lifespan, a residual environmental impact from that first life should be imputed to the
second life.
The following results thus illustrate the outcome of this approach. In the interests of greater readability,
the analysis was carried out on the four priority indicators, i.e. climate change (GWP), depletion of natural
mineral and metal resources (ADPe), depletion of natural fossil resources (ADPf) and ionising radiation (IR).

Figure 26 - Smartphone comparison - Depreciation approach - Variation in environmental impact by duration of first
life.
The first figures show the variation in impact based on the duration of the first life. It can be observed
that the later the refurbishment takes place, the more favourable the environmental impact of the
refurbished mobile phone. If refurbishment takes place before the first year has elapsed, refurbishment
will not be favourable. This reinforces the general conclusion that making your smartphone last is better
than replacing it. In addition, in order to have a lower impact, the market for refurbished devices must
not be based on a market of overconsumption (the US market in particular).
A second analysis was carried out. This analysis combines the concepts of the duration of the first life and
the type of refurbishment. The minimum refurbishment scenario involves cleaning only within the
national territory, and the maximum refurbishment scenario involves replacement of all parts. It can be
seen even more clearly that if refurbishment involves replacing a large number of parts, it can be less
favourable prior to the end of the second year of the lifespan.

Figure 27 - Smartphone comparison - Depreciation approach - Variation by duration of first life and by type of
refurbishment
The earlier the refurbishment takes place, the more its impact needs to be compensated for by a long
second life. There is therefore no excuse for keeping your refurbished smartphone for a shorter time just
because it is second-hand.

Comparison – Analysis of behavioural scenarios
The results set out below use a behavioural approach to illustrate the benefit of acquiring a refurbished
device based on the behaviour of its initial and subsequent users.
In this conceptualisation, the impact on climate change depends on the type of purchase (new or
refurbished), the frequency of purchase, the length of the 1st life of the refurbished device and the period
for which the device was kept/used.
 regular buyers of refurbished but recent smartphones: purchase a 1-year-old smartphone

once every 2 years.
 reasonable buyers of refurbished smartphones: purchase a 2-year-old smartphone once

every 2 years
 environmentally conscious buyers of refurbished smartphones: purchase a smartphone

over 3 years old once every 3 years.

Figure 28 - Comparison with reference smartphone - Depreciation approach - Market results over 6 years for climate
change
New and refurbished devices are alternatively lower-impact and higher-impact, depending on the
scenario.
The scenario that systematically minimises impacts on Global Warming Potential (GWP) is measured
purchasing: purchasing a refurbished device that had a first life of 3 years (or more), every 3 years (or
more).
N.B.: The quantified results are set out in an appendix (9.5.1).
4.5. Extrapolation to tablets

The situation for tablets was extrapolated from smartphone data collected from refurbishers, since tablet
refurbishment volumes are low, and these devices follow the same procedure as smartphones.
4.5.1. Reference data

Tablets are collected and processed in the same circuits and processes as smartphones. Thus, the
smartphone reference model was extrapolated to obtain a reference model for tablet-style devices. To
do this, the mass, size and characteristics of the parts were modified to correspond to the characteristics
of tablets.
Refurbished tablet New tablet
Type of modelling Extrapolation
Sources of data Refurbishers (collected from 11) ADEME/ARCEP study -

NegaOctet database

Refurbished tablet New tablet
Refurbishers Agora Place, Backmarket, Bak2,

Itancia, Largo, Le GSM, Recommerce,
Sofigroupe, Riitekpro + 2 refurbishers
(CONFIDENTIAL)
Reference model Average product based on Average product based on

products collected by refurbishers the French market
Characteristics of the reference Refurbishment location: Manufacturing location:

model France (48%) Asia (100%)
Europe (26%)
Asia (26%)

France (18%) Europe (44%) Asia (11%) Asia (100%)
US (25%)

Mass: 478g Mass: 478g
Screen size: 10.5 inches Screen size: 10.5 inches
Type of screen: 100% LCD Type of screen: 100% LCD
Battery weight: 193g Battery weight: 193g
Storage: 256GB Storage: 266GB

Screen: 71.52 cm²
Battery: 41g
0.5 non-electronic units
0.21 electronic units
0.03 printed circuit boards
Consumption during use phase 18.6 kWh per year 18.6 kWh per year
Table 29 – Reference characteristics for tablets
In addition, the new reference model was adapted for tablets
TABLET
PRIVATE USE CAT. 1 CAT. 2 CAT. 3 AVERAGE
Distribution 33% 33% 33% 100%
Weight of device (kg) 0.465 0.498 0.471 0.478
Weight of packaging (kg) 0.093 0.0996 0.0942 0.0956
Screen size (inches): 10.2 10.3 11.1 10.53333333
Screen technology LCD touchscreen LCD touchscreen LCD

touchscreen
Type of processor Huawai Kirin 659 A10 Fusion Apple A12

Bionic
RAM (GB) 4 4 6 4.666666667
SSD storage (GB) 32 256 512 266.6666667
Motherboard surface area (cm²) 400.84 421.35 461.14 427.7766667
Battery mass (g) 156.2 205.8 216.4 192.8
Consumption considered annually 0.5 hrs per day (active mode), 23.5 hrs per day (stand-by) 18.6
(Wh per year)
Reference model example Huawei Mediapad M3 Samsung Galaxy Apple iPad Air
Lite 10.1 Tab S5e 4

Table 30 – Characteristics for new tablets considered
4.5.2. Results
Impacts of a refurbished reference tablet at a functional unit level (1

year of use)
For each year of use, the life cycle impacts of a reference refurbished tablet are as follows:
Impact Value Unit
Depletion of natural abiotic resources 4.19E-04 kgeqSb per FU

- metals and semi-metals (ADPe)
Fine particles (PM) 4.59E-07 disease

occurrence per
FU
Table 31 – Reference tablet - Environmental impacts per functional unit for 1 year
Comparison of the impacts of new and refurbished tablets

Since the calculation performed for tablets was an extrapolation of the calculation for smartphones, the
general analysis of the results is very similar to the one obtained for smartphones. A comparative summary
was therefore produced for this product family, considering:
 A product refurbished with a short supply chain*, without accessories and with basic cleaning;
 A product refurbished with a short supply chain*, with new replacement parts, and
accessories;
 A product refurbished with a short supply chain*, with second-hand replacement parts, and
accessories;
 A product refurbished on a globalised market **, without accessories and with basic cleaning;
 A product refurbished on a globalised market **, with new replacement parts, and accessories;
 A product refurbished on a globalised market **, with second-hand replacement parts, and
accessories.
* Product collected and refurbished in France
** Product collected in the USA and refurbished in China

Figure 29 - Tablet comparison - Substitution approach - Results for one year - (100% = impact of new reference
device)
Despite the conservative hypotheses for new devices, we observed that refurbished tablets have a
significantly lower impact: -98 to -26% annual impact, excluding ionising radiation. The impact on ionising
radiation is explained by the part of the industrial process carried out in France for refurbishment.
In most cases, the variation in impacts between the refurbishment scenarios does not cancel out the
beneficial effect of refurbishment practices.
The excess impact on IR and WU is associated with the replacement of the battery and the use of the
French energy mix for full refurbishment.
Impacts avoided by purchasing a refurbished tablet – Substitution

approach
For the reference scenario, purchasing a refurbished tablet and using it for 2 years instead of purchasing
a new tablet and using it for 3 years allowed the following impacts to be avoided annually:
Impact Value Unit %
REFERENCE PERIOD FOR NEW DEVICES 3 Years
REFERENCE PERIOD FOR REFURBISHED DEVICES 2 Years
Climate change (GWP) -20.3 kgeqCO2 per -78%

year
Ecological load (MIPS) -80.1 Kg per year -80%
Production of WEEE -120 g per year -75%
Depletion of natural abiotic resources - metals and semi- -4.80E-04 kgeqSb per year -70%
metals (ADPe)
Depletion of natural fossil resources (ADPf) -260 MJ per year -46%
Water usage (WU) -22.9 m3eq per year -62%
Acidification (AP) -1.26E-01 kgeqH+ per year -79%
Fine particles (PM) -6.79E-07 disease -60%

occurrence per
year
Ionising radiation (IR) -7.98 kgU235eq per -18%

year

Table 32 - Tablet comparison - Substitution approach - Impacts avoided for the reference scenarios - 100% being the
impact of the reference device.
of the new product and the refurbished product. Avoided impacts are calculated as the impacts per FU
for refurbished smartphones minus the impacts per FU for new smartphones:
𝐼𝑚𝑝𝑎𝑐𝑡 𝑟𝑒𝑓𝑢𝑟𝑏𝑖𝑠ℎ𝑒𝑑 𝑑𝑒𝑣𝑖𝑐𝑒 𝐼𝑚𝑝𝑎𝑐𝑡 𝑛𝑒𝑤 𝑑𝑒𝑣𝑖𝑐𝑒

𝐴𝑛𝑛𝑢𝑎𝑙 𝑎𝑣𝑜𝑖𝑑𝑒𝑑 𝑖𝑚𝑝𝑎𝑐𝑡
𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑢𝑠𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑠𝑒𝑐𝑜𝑛𝑑 𝑙𝑖𝑓𝑒 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑢𝑠𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑓𝑖𝑟𝑠𝑡 𝑙𝑖𝑓𝑒

5. Computer family
5.1. Laptop computers
5.1.1. Overview
Definition: ‘A computer designed specifically for portability and to be operated for extended periods of
time either with or without a direct connection to an AC power source. It has an integrated display.’35
5.1.2. Defining reference models

‐ Represent the reality of the markets of the refurbishment stakeholders that participated in
the study;
‐ Be compatible with all the work coordinated by ADEME in early 2022 so as to provide a
Refurbished laptop computers New laptop computers
Sources of data Refurbishers (collected from 8) ADEME/ARCEP study - NegaOctet database
Refurbishers Ateliers du bocage; ATF bis repetita, ATF

GAIA, CRS, Emmaus Connect, Ingram, LM
Ecoproduction, Okamac, Recyclea
Reference model Average product based on refurbisher data Average product based on the French
collected market
Characteristics of the Refurbishment location: Manufacturing location:

reference model France (84.35%), United Arab Emirates Asia (100%)
(15.65%)

France (66.67%), Europe (18.67%), US Asia (100%)
(10.56%), Australia (3.00%), United Arab
Emirates (1.11%)

Mass: 2 kg Mass: 1.6 kg
Screen size: 13 inches Screen size: 14.5 inches
Type of screen: 88% LCD, 12% OLED Type of screen: 88% LCD, 12% OLED
Storage: 500GB (SSD: 34% HDD: 66% Storage: 500GB (100% SSD)
RAM: 16GB (20g) RAM: 13GB (20g)
Battery weight: 300g Battery weight: 244g

8.97 cm2 LCD screen / 5.98 cm2 OLED
screen
0.05 unit: RAM / 0.08 unit: hard disk
7.58 g battery / 0.17 electronic units
Consumption during 29.1 kWh per year 29.1 kWh per year
use phase (personal)
Table 33 - Presentation of the characteristics of reference laptops
reference models. This corresponds to the actual state of the market, as new devices are renewed based
35

Electricity consumption: 29.1 kWh per year in 2020. The electricity consumption of laptop computers is
taken from the ICT 2020 report.
N.B.: We cannot differentiate between consumption for new and refurbished devices. We thus opted for
an average consumption. It will nonetheless be noted that, in spite of an improvement in the energy
efficiency of devices, those distributed are becoming ever-increasingly powerful, which reduces the gap
in energy consumption between new and refurbished devices.
Typical lifespan:
 For a new laptop computer: 5 years36;
 For a refurbished laptop: 3 years.
5.1.3. Hypotheses and entry data
Refurbishment
Data was collected from 8 international refurbishers. This work highlighted the fact that profiles varied
considerably between refurbishers. The variant parameters are:
 Location of refurbisher;
 Location of refurbisher’s suppliers;

Compa Compa Compa Compa Compa Compa Compa Compa Compa

ny A ny B ny C ny D ny E ny G ny H ny I ny J
Location France France France France France France UAE France France
Number of units 44.9 23.2 5 0.5 7 31.2 40 46.8 7.5

refurbished in 2020
(in thousand units)
Type of 52.03 0% 140% 2% 0% 0% 229% 50% 0% 46%

refurbishment - %
Proportion of
parts
replaced
Type of parts - % 40.18 0% 58% 100% 0% 0% 23% 39% 67% 75%

of parts taken %
from
refurbishment
Main France 66.67 100% 0% 100% 100% 100% 0% 0% 100% 100%

source %
of
produc Europe 18.67 0% 60% 0% 0% 0% 88% 20% 0% 0%
ts %
Asia 0.00% 0% 0% 0% 0% 0% 0% 0% 0% 0%
US 10.56 0% 40% 0% 0% 0% 5% 50% 0% 0%

%
Australi 3.00% 0% 0% 0% 0% 0% 7% 20% 0% 0%

a
36
Renouvellement des terminaux mobiles et pratiques commerciales de distribution (Renewing mobile terminals and
sales practices for distribution) – ARCEP – June 2021

Compa Compa Compa Compa Compa Compa Compa Compa Compa
ny A ny B ny C ny D ny E ny G ny H ny I ny J
United 1.11% 0% 0% 0% 0% 0% 0% 10% 0% 0%

Arab
Emirat
es
*more than one part replaced per COMPUTER

**parts nearly always replaced
***half of the parts replaced came from refurbishment
Table 34 – Reference refurbished laptops - Presentation of refurbisher profiles - Summary
It should be noted that the refurbishment location/supply location combination has a direct influence on
the distance covered by the devices and the spare parts, as well on the distribution of the impacts
between upstream and downstream transport. The distances set out in Table 4 were used to characterise
the upstream and downstream transport for the various fractions.
In order to define the impacts associated with each piece of data collected, hypotheses were drawn up
for the constituent elements of refurbished laptops, spare parts and accessories.
Screen
Screen 800 G Screen size 465.9 Cm²
Screen / % LCD 88% Screen / % OLED 12%
Storage
SSD 60 G SSD drive volume 512 GB
HDD 115 G
RAM:
RAM 20 G RAM volume 16 GB
Other
Keyboard 200 G
Battery 300 G Charger 300 g
Table 35 – Refurbished laptop - Physical characteristics of the various elements considered for the reference model

Collection data per FU
Ref. Min Max Unit
INVARIABLE PART
Supply - Previous user to centralisation
Transport from the user to the collection point
Public transport 0.85 0.54 1.73 passenger.km
Car 3.62 2.29 7.36 passenger.km
Van 6.94 4.36 14.09 kg.km
Lorry 3,150.60 1,987.18 6,404.79 kg.km
Supply - Centralisation to refurbishers
Transport from the centralisation centre to the refurbishers
Lorry 3,624.96 1,987.18 6,570.51 kg.km
Ship 3,833.33 0 34,500.00 kg.km
Plane 1,707.56 3,557.69 11,810.34 kg.km
Packaging
Cardboard 0.3 0 0.68 kg
Plastic 0.02 0 0.05 kg
Pallet 0.19 0.01 0.46 kg
On-site consumption
Electricity 6.12 0.23 13.11 kWh
Water 0.010 0.000 0.020 m3
Gas 4.740 0.090 33.270 MJ
Control
Paper 0.010 0.000 0.010 kg
Cleaning (if no info given, max values for alcohol and cotton)
Rubbing alcohol - Detergent - Disinfectant, etc. (l) 0.010 0.000 0.010 kg
Accessories
Charger (Power supply)
Manufacture 0.35 0 1 unit
Supply
Lorry 48.58 0.34 270 kg.km
Ship 58.43 0 525.86 kg.km
Plane 269.94 3.40 1620 kg.km
Mouse
Manufacture 0.11 0 1 unit
Packaging
Cardboard
Manufacture 0.24 0.08 0.27 unit
Supply
Lorry 414.61 78.28 1,462.50 kg.km
Plane 2,399.66 782.78 2,723.04 kg.km
Plastic
Manufacture 0.15 0 0.22 unit
Supply

Ref. Min Max Unit
Lorry 177.93 1.21 284 kg.km
Ship 162.5 162.5 162.5 kg.km
Plane 2,150.47 2,150.47 2,150.47 kg.km
Tempered glass
Manufacture 2.04E-01 0 1.84 kg
Supply
Lorry 2.04E-01 0 1.84 kg.km
Paper
Manufacture 1.79E-04 1.5E-05 1.61E-03 kg
Supply
Lorry 1.79E-01 0.1 1.61 kg.km
Refurbishment to sale stock
Lorry 271.67 0 2,445.00 kg.km
Plane 1,411.11 0 12,700.0 kg.km
Lorry 2,248.75 1,329.06 4,447.40 kg.km
Van 4.4 2.35 9.67 kg.km
VARIABLE PART
Screen replacement
Manufacture
LCD 1.54 0.17 10.64 cm²
OLED 0.21 0.02 1.45 cm²
second-hand 13.2 0.12 108.8 cm²
Estimated transport
Lorry 3.09 1.12 20.73 kg.km
Plane 27.04 36 207.33 kg.km
Battery replacement
Manufacture
Lithium Battery 5.58 0.07 31.69 g
2nd-hand 2 0.97 17.06 g
Supply
Lorry 18.86 0.07 102.56 kg.km
Plane 155.83 0.71 1,025.64 kg.km
RAM replacement
Manufacture
RAM 0.03 0.01 0.2 unit
2nd-hand 0.02 0 0.1 unit
Estimated transport
Lorry 2.18 0.05 16.55 kg.km
Plane 19.85 0.45 165.54 kg.km
Hard disk replacement
Manufacture

Ref. Min Max Unit
SSD 0.04 0.01 0.17 unit
HDD 0.01 0 0.03 unit
2nd-hand 0.04 0.07 0.22 unit
Estimated transport
Lorry 6.11 0.53 22.77 kg.km
Plane 23.68 5.26 133.65 kg.km
Replacement of miscellaneous electronics
Manufacture
Component 0.15 0 1.32 unit
2nd-hand 0.03 0 0.23 unit
Estimated transport
Lorry 1.47 0 13.21 kg.km
Plane 14.67 0 132.05 kg.km
Table 36 - Collection data for refurbished laptop computers
New laptop computers
The model for laptop computers is a weighted mix of three configurations, with the following
characteristics:
Cat. 1 Cat. 2 Cat. 3 Average
Profile Chromebook Office Gaming
Distribution 40% 49% 12%
Weight of device (kg) 1.5 1.54 2.3 1.61
Weight of packaging (kg) 0.3 0.3 0.46 0.323
Screen size (inches): 14.5 14.5 15.6 14.6287
Screen technology LCD LCD LED
Type of processor AMD Ryzen 5 Intel® Core™ i7, 8th Intel Core I5
generation
RAM (GB) 16 8 16 13.33333333
SSD storage (GB: 427 564 512 503.801
Motherboard surface area 52.93 62.03 88.22 61

(cm²)
Type of graphics card Built-in Built-in NVIDIA Ampere

GeForce RTX 2060
6Go
Battery mass (g) 330 165 287 244
External power supply (g) 245 211 208 224
Table 37 – Reference characteristics for new computers

5.1.4. Results
Foreword
 The location of the refurbishers;
Subsequent analysis will be carried out on two levels:
 The full life cycle as broken down in the graph below by sector and by sources of impact;
 Refurbishment only (scope delineated in red in the graph below).

Figure 31 – Reminder of the analysis parameters and correspondence with the data collected
The results will themselves be presented using the lifespan of the device or functional unit as a reference
scale.

Reminder – Abbreviations for indicators
ADP - Elements = Depletion of natural abiotic GWP = Climate change

resources - metals and minerals
IR = Ionising Radiation
ADP - Fossil = Depletion of natural abiotic
PM = Fine particles
WU = Water usage
AP = Acidification
Assessment of the impact of a refurbished reference laptop used for 1

year
‘Owning and using a laptop computer for personal use for one year’
usage period, i.e. 3 years for refurbished devices and 5 years for new devices.

For each year of use, the life cycle impacts of a reference refurbished laptop computer are as follows:
Impact Value (personal Unit

use)
REFERENCE PERIOD 3 YEARS
Depletion of natural abiotic resources - metals and semi-metals (ADPe) 9.57E-05 kgeqSb per FU
Fine particles (PM) 6.92E-07 disease

occurrence per
FU
Table 38 – Reference laptop - Environmental impacts per functional unit for 1 year

Results broken down by life cycle impact
Figure 32 - Reference refurbished laptop - Results broken down by life cycle impact
Findings:
Despite a larger share for the use phase, refurbishing devices represents the majority of impacts for 5
indicators and a significant impact for 3 others.
Analysis:
of this stage is determined by the refurbishment location, the supply location and the type of the
refurbishment.

To gain a better understanding of the environmental impacts across the refurbishment process, a detailed
Figure 33 - Reference refurbished laptop computer - Impacts broken down by stage - Focus on the refurbishment
process
Findings:
Impacts are divided between supply, on-site consumption, packaging and part replacements (RAM and
disks).
Consumption on the refurbishment site is the main contributor for the IR indicator.
Disk replacements are the main contributors for ADP.
Using second-hand parts can reduce the impact generated from replacing parts.
Analysis:
one company and from one scenario to another must be observed in order to identify an environmentally
optimum outcome.
Analysis:
Caution should be taken when it comes to systematically adding accessories and replacing parts,
particularly with new parts.

Once the reference model was established, the average profile for part replacement was deleted in
favour of profiles corresponding to the replacement of a full screen, a battery or other parts. This analysis
sets out the variations based on the type of replacement made.
Figure 34 – Refurbished laptop computer - Impact of replacing parts - Life cycle vision -(100% = impact of the
reference refurbished device)
Findings:
Significant variations in impact were observed between the different scenarios (from a factor of 10 to 17).
The largest variations are associated with:
 Replacing the hard disk (the disk considered is an SSD, which is made of semiconductors, the
manufacture of which requires considerable energy consumption);
 Replacing the screen, specifically LCD screens for the ADPe indicator and OLED screens for the
GWP indicator;
 A significant impact from replacing the battery, specifically on water (associated with the
mining of lithium).
Analysis:
The type of parts replaced has a significant impact on all indicators, especially for screens. If parts were
replaced frequently, this would have a considerable impact on the reference model. If second-hand parts
were used, the impact of parts replacement would be drastically reduced.
Taking account of the incorporation of the use phase, the difference between the reference scenario and
the other scenarios is less significant than could be imagined, owing to the presence of new accessories
and to consumption during the use phase.

The reference model is fictitious. This model represents a weighted average of the market data
 The replacement rates for parts vary from 0% (0–1 part replaced, infrequently) to 229%
(multiple parts replaced, more frequently) from one refurbisher to another.
 The usage rate for second-hand parts is variable: from 0 to 100% second-hand parts.
 The estimated average distances for the total route travelled by a desktop computer vary by a
factor of up to 16 from one refurbisher to another. This depends on the location of the
refurbisher and its suppliers.
Figure 35 - Refurbished laptop computer - Variation of impacts depending on the companies - Impacts on climate
change

Figure 36 - Refurbished laptop computer - Variation of impacts depending on the companies - Impacts on the
depletion of abiotic mineral resources
Figure 37 - Refurbished laptop computer - Variation of impacts depending on the companies - Impacts on the
depletion of abiotic fossil resources

Figure 38 - Refurbished laptop computer - Variation of impacts depending on the companies - Impacts on ionising
radiation
of 6 to 14 between the minimum and the maximum, depending on the indicators):
 The refurbishments performed by Company A had the least impact on average: A is a French
company, supplied exclusively in France, that does not replace parts.
N.B.: - A part replacement may be deemed low impact if it extends the second life of the device but this
cannot be seen from these graphs.
 The refurbishments performed by Company G had the highest impact on average. G carries
out its refurbishments in France, but the products are sourced from abroad, the part
replacement rate is high (more than 2 parts per device) and the parts are only rarely second-
hand (23% of parts).
With regard to the depletion of natural resources, the refurbishments performed by companies that
replace a lot of parts have a significantly larger impact than the others. However, this impact is limited
where the percentage of second-hand parts is high.
 Company A is the lowest impact. Company B has the highest impact. Impact is in direct
correlation with the on-site consumption and the use of the French energy mix. However, with
Findings:
The sample of refurbishers who participated in this study is less varied than for smartphones. Most of
them are French refurbishers who source their devices from neighbouring regions.
A very considerable range of impact variation can however be observed from one refurbisher to another.
This can be explained by the use of very different refurbishment models, from cleaning only to parts
replacement, whether or not extensive use is made of second-hand parts, and by different logistics
scenarios.
Analysis:

Despite a good level of participation in the study, the degree to which the refurbishers represent the
French market is low. The scenario of devices refurbished abroad for marketplaces is only partially
covered. Nonetheless, strong variations can already be seen. Performing an analysis based on a fictitious
scenario would be appropriate and will round off this initial sample analysis.
However, refurbisher profiles vary based on their supply models (more or less selective) and their varied
methods of refurbishment (from a simple cleaning model with no repairs to a very high-security model
where parts are systematically replaced), which is reflected directly in the results of the analyses.
Using a reference model is appropriate, but the variation in impacts from one extreme scenario to
another needs to be kept in mind, and these impacts need to be supplemented by fictitious scenarios.
The impacts of the reference model and extreme scenarios are set out below:
Impact Reference Value Value Unit

value (min) (max)
Reference period 3 years

3 3

4.62 13.60
6 31
Depletion of natural abiotic resources - metals and 9.57E-05 kgeqSb per FU

semi-metals (ADPe) 7.07E-05 1.51E-04
412.0 528.8
0.73 6.28
2.22E-02 7.14E-02
Fine particles (PM) 6.92E-07 disease occurrence
5.62E-07 8.30E-07 per FU

50.91 57.81
Table 39 – Laptop computer - Environmental impacts per functional unit for 1 year - Variations between companies

Figure 39 – Refurbished laptop computer - Impact of refurbishment and supply location - Life cycle vision - (100% =
impact of the reference refurbished device)
Findings:
The reference scenario includes few replacement parts. The supply portion is therefore considerable. The
variation in these 4 indicators is strongly associated with the combustion of fossil fuels (it should be noted
that the preferred means of transport was express delivery by plane).
Analysis:
The increases are linked to the impact of transport as part of the overall impact of refurbishment. A
sensitivity analysis for means of transport must be carried out in order to consolidate these discrepancies.

The second lifespan is a key parameter for the environmental impact of a refurbished computer. In the
reference scenario, we chose a lifespan lower than the first lifespan so as to be in a realistic situation with
regard to both the behaviour and the total lifespan of the product.

Figure 40 - Refurbished laptop computer - Variation of impacts depending on the lifespan of the refurbished laptop
Findings:
For the indicators that change considerably based on energy consumption (ADPf, IR, PM, etc.), the
variation in impact is slight. For ADPe-type indicators, the variation in impact is very significant.
Analysis:

Focus on screens - Type
elements fossil H+ eq) CO2 eq) U235 occurrence) (m3) (kg)
(kg SB eq) (MJ) eq)
LCD – substitution of 100% 100% 100% 100% 100% 100% 100% 100%
the screen portion in
the reference model
Reference laptop 9.57E-05 472.00 4.06E-02 8.09 53.96 6.92E-07 2.16 14.68
computer
(15cm² replaced –
88% LCD
12% OLED)
OLED – substitution 97% 100% 100% 100% 100% 100% 99% 100%
of the screen portion
in the reference
model
Table 40 - Refurbished laptop computer - Sensitivity analysis per type of screen - For replacement in the reference
model

elements fossil (MJ) H+ eq) CO2 eq) U235 eq) (Disease
(kg SB eq) occurren
ce)
LCD – screen 112% 99% 97% 94% 101% 99% 110% 104%
replaced by
an LCD
Full screen 9.05E-04 547.88 6.97E-02 12.43 61.77 8.72E-07 3.78 43.04
replacement
(100% replaced –
88% LCD
12% OLED)
OLED – screen 11% 105% 123% 147% 89% 108% 23% 71%
replaced by an
OLED
Table 41 - Refurbished laptop computer - Sensitivity analysis per type of screen - For full screen replacement
Findings:
A limited variation was observed in the reference scenario, since the contribution represented by the
replaced screen is small. On the other hand, very significant variations in impact were observed for screen
replacements, including a strong variation in the ADP-elements indicator and WU.
Analysis:
The type of screen has a significant impact on all indicators when the screen is replaced. An OLED screen
will have a higher impact on the GWP indicator (but a lower impact on the ADP-elements indicator owing
to its complex manufacture and how recent the technology is). Conversely, an LCD screen will have a
higher impact on the ADP-elements indicator (associated with the iridium contained in the display) but a
more limited impact on the GWP indicator.
Focus on RAM
element fossil H+ eq) CO2 eq) U235 eq) (Disease
s (kg SB (MJ) occurren
eq) ce)
Reference scenario 9.57E-05 469.43 3.96E-02 7.90 53.96 6.86E-07 2.07 14.25
(RAM replaced
in 5% of cases – 16GB
capacity)
8GB 101% 99% 98% 98% 100% 99% 109% 98%
32GB 101% 101% 105% 105% 100% 102% 122% 106%
64GB 101% 103% 115% 114% 100% 105% 140% 117%
Table 42 - Refurbished laptop computer - Sensitivity analysis for RAM capacity

(kg SB eq) occurrenc
e)
Full RAM
replacemen
t scenario 9.57E-05 472.00 4.06E-02 8.09 53.96 6.92E-07 2.16 14.68
(16GB)
8GB
101% 99% 98% 98% 100% 99% 109% 98%
32GB
101% 101% 105% 105% 100% 102% 122% 106%
64GB
101% 103% 115% 114% 100% 105% 140% 117%
Table 43 - Refurbished laptop computer - Sensitivity analysis for RAM capacity (full RAM replacement)
Findings:
When we consider a modification in the reference scenario (RAM replaced in 5% of cases), we observe
that the choice of RAM capacity has a significant, visible impact (+1% to +22% for indicators showing a
variation). No variation is observed for the ADPe indicator.
This variation is heightened in the scenario that includes a full RAM replacement. Variations in impact of
between -44% and +245% are observed.
Analysis:
An increase in RAM capacity correlates to an increase in the quantity of silicon wafer and the number of
masks, and thus to a significant increase in the impacts associated with the manufacture of
semiconductors. It should be noted that there is no change to the type of encapsulation or the number
of sticks of RAM, which explains a stagnation in the impact on the depletion of natural resources. The
RAM capacity should therefore be adjusted to meet the requirements of the end user, thereby ensuring
good performance for the product and a genuine extension to its lifespan.
Focus on hard disks

eq) ce)
Reference scenario
(disk replaced in 8%
of cases –512GB 9.57E-05 472.0 4.06E-02 8.09 54.0 6.92E-07 2.16 14.68
capacity)
256GB
100.5% 99% 94% 95% 100% 98% 104% 91%
1024GB
100.6% 102% 110% 109% 100% 103% 131% 108%
2048GB
100.6% 106% 131% 128% 100% 110% 167% 131%
Table 44 - Refurbished laptop computer - Sensitivity analysis for hard disk capacity

e)
Full disk 3.70E-04 614.4 1.03E-01 19.0 55.8 1.03E-06 6.1 48.9
replacemen
t scenario
(512GB)
256GB 100.0% 100% 100% 100% 100% 100% 100% 100%
1024GB 100.2% 169% 275% 272% 100% 196% 341% 244%
2048GB 100.5% 262% 509% 502% 101% 324% 661% 437%
Table 45 - Refurbished laptop computer - Sensitivity analysis for hard disk capacity (full hard disk replacement)
The manufacturing techniques for SSD hard disks are very similar to those used for manufacturing RAM.
Considerable variations can thus be seen here, as in the previous sensitivity analysis.
As with RAM, the storage capacity of the computer should be adjusted according to the end user’s needs.
Focus on the means of upstream transport

elements (kg fossil H+ eq) (kg U235 occurrence) (m3) (kg)
SB eq) (MJ) CO2 eq)
eq)

transport 100% Plane
Reference 9.57E-05 472.00 4.06E- 8.09 53.96 6.92E-07 2.16 14.68

02

transport 100% Ship
Table 46 - Refurbished laptop computer - Sensitivity analysis - Means of transport
The choice of means of transport has a fairly insignificant impact (-14% to +4%) on the environmental
impacts of a refurbished laptop over 1 year. However, it should be noted that, in the reference model,
few devices originated from abroad. Thus, the impact of changing the means of supply could be
heightened if that supply comes from abroad.

Focus on the means of distribution by the private individual

elements fossil H+ eq) CO2 eq) U235 eq) occurrence) (kg)
(kg SB eq) (MJ)
Reference 9.56E-05 4.71E+02 4.06E-02 8.09E+00 5.40E+01 6.67E-07 2.29E+00 1.47E+01
Urban distribution 80.64% 99.39% 98.49% 97.50% 99.97% 98.83% 98.85% 100.00%
> Shop/Collection
Point
Peri-urban 166.10% 102.51% 108.33% 110.16% 100.10% 106.20% 104.18% 100.00%

Distribution >
Shop/Collection
Point
> Shop/Collection
Point
> Home delivery
Table 47 - Refurbished laptop computer - Sensitivity analysis - Means of distribution
Findings:
The choice of means of transport to private individuals has a very significant impact (-3% to +23% for the
GWP indicator and -20% to +152% for the ADPe indicator) on the impacts of a refurbished laptop
computer over 1 year.
Analysis:
The choice of the means of distribution is key to limiting the environmental impact of refurbishment. In
particular, trips by private car to pick up the parcel should be avoided as much as possible (the
Periurban>Shop and Rural>Shop scenarios).
It should be noted that the variation in the ADPe indicator is greater than for smartphones, mainly
because the reference model for laptops contains proportionally fewer new electronic components.
Standardisation of the results
Standardised results – world inhabitant approach

to be considered are the following:

‐ Climate change;
‐ Ionising radiation (chiefly linked to the use phase and to use of the French energy mix);

‐ Ionising radiation (chiefly linked to the use phase and to use of the French energy mix);
Thus, some comparative analyses will be carried out on the ADPe, ADPf, IR and GWP indicators only.

Figure 41 – Reference refurbished laptop - Results for impacts after standardisation for the entire life cycle

to this standardisation, we can see a reverse in the trend for priority indicators, and focus should be
placed on:
 Depletion of natural fossil resources;
 Climate change;
Given the low rate of part replacement, the impact on depletion of mineral resources is low. However,
we chose to reinstate it, as whenever a screen or disk is replaced, this criterion becomes significant.
 Climate change.

Figure 42 – Reference refurbished laptop - Results for impacts after standardisation to planetary boundaries over the
entire life cycle
Comparison of the impacts of new and refurbished laptops
Comparison of the impacts of new and refurbished laptops -

Presentation of results for the reference models

New laptop Refurbished laptop Unit
computers computers
usage included usage included
LIFESPAN
Climate change (GWP) 177.6 24.28 kgeqCO2
Ecological load (MIPS) 706.8 44.0 kg
Production of WEEE 1,610 25 G
Depletion of natural abiotic resources - metals and semi- 8.09E-03 2.87E-04 kgeqSb
metals (ADPe)
Depletion of natural fossil resources (ADPf) 4,110.6 1416 MJ
Water usage (WU) 192.4 6.5 m3eq

E
Acidification (AP) 1.01 +00 1.22E-01 kgeqH+
E
Fine particles (PM) 7.76 -06 2.07E-06 disease
occurrence
Ionising radiation (IR) 327.9 161.88 kgU235eq
Table 48- Reference laptop - Environmental impacts for the entire life cycle for the reference lifespans (over the
entire life cycle) - Personal use
For the reference scenario, purchasing a refurbished laptop (used for 3 years) instead of purchasing a new
laptop allowed the following impacts to be avoided annually:
Value (personal Unit %

use)
Reference period for new devices 5 Years
Reference period for refurbished devices 3 Years

year
Depletion of natural abiotic resources - metals and -1.52E-03 kgeqSb per year -94%
semi-metals (ADPe)
Acidification (AP) -1.61E-01 kgeqH+ per year -80%

occurrence per
year

year

Table 49 - Laptop comparison - Substitution approach - Impacts avoided for the reference scenarios - 100% being the
impact of the reference desktop computer.
of the new product and the refurbished product as well as on the refurbishment scenario. This is therefore
a reference evaluation. Avoided impacts are calculated as the impacts per FU for refurbished laptops
minus the impacts per FU for new laptops:

𝑈𝑠𝑎𝑔𝑒 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑖𝑛 𝑠𝑒𝑐𝑜𝑛𝑑 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑢𝑠𝑒 𝑖𝑛 𝑓𝑖𝑟𝑠𝑡 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒
At the scale of a private individual, substituting a new laptop with a refurbished laptop can prevent:
budget;
 82 km travelled by car;
 the extraction of 126 kg of materials;
 the production of 314 g of electronic waste.

carried out using the substitution approach between:
 A product refurbished with a short supply chain*, with basic cleaning;
 A product refurbished with a short supply chain*, with new replacement parts;
 A product refurbished with a short supply chain*, with second-hand replacement parts;
 A product refurbished on a globalised market**, with basic cleaning;
 A product refurbished on a globalised market**, with new replacement parts;
 A product refurbished on a globalised market**, with second-hand replacement parts;
* Product collected and refurbished in France ** Product collected in the USA and refurbished in China

Figure 43 - Laptop comparison - Substitution approach - Results for 1 year (100% = impact of reference new device)
At the functional unit scale (considering a first lifespan of 5 years and a second lifespan of 3 years), we
can observe that:
 Refurbishment is lower impact if:

‐ No replacements are made, or only minor replacements.
‐ Major parts are replaced using second-hand parts.
 Refurbishment may be disadvantageous if:
‐ Multiple parts are replaced (screen, RAM, etc.) using new parts. Since the screen, the RAM
and the disk represent the majority of the impacts from the device (20% to 57% of the
impact of a new device), if the increase in lifespan is shorter than 5 years (the length of the
first lifespan), refurbishment has a greater impact than the manufacture of a new device
used for 5 years. In addition, replacing parts is often used as a way of increasing the capacity
of the components.
Moreover, a change in the location and logistics scenario has a relatively low impact at the level of the
device’s entire life cycle, as the use phase and the manufacture of the parts have a preponderant impact.
Comparison of the impacts of new and refurbished laptops -

Depreciation approach
second life.
The following results thus illustrate the outcome of this approach. In the interests of greater readability,
the analysis was carried out on the 3 priority indicators, i.e. climate change (GWP), depletion of natural
mineral and metal resources (ADPe) and depletion of natural fossil resources (ADPf).

Figure 44 - Laptop comparison - Depreciation approach - Variation in environmental impact based on duration of
first life.
The first figures show the variation in impact based on the duration of the first life. It can be observed
refurbished mobile phone. If refurbishment takes place before the third year has elapsed, refurbishment
will not be favourable. This reinforces the general conclusion that making your device last is better than
replacing it.
seen even more clearly that if refurbishment involves replacing a large number of parts, it is less favourable
unless the second lifespan is at least equal to the first lifespan.

Figure 45 - Laptop comparison - Depreciation approach - Variation in environmental impact based on duration of
first life and type of refurbishment.
Analysis:
If we assume a first lifespan of 3 years and if refurbishment takes place before those 3 years are up, it
generates an additional impact. In order to be beneficial, refurbishment must allow the lifespan after
refurbishment to be extended by more than 4 years.
N.B.: - The reference scenario includes few replacement parts and/or second-hand replacement parts.
The reference scenario is thus close to the minimum scenario.
If all parts are replaced, refurbishment must occur after 4 years and enable the lifespan to be extended
by over 5 years. Over 60% of the impact from a computer comes from its screen, battery, RAM and disk.
If all parts are replaced, this produces an impact similar to that of new devices, and this replacement
should be depreciated over a long lifespan.

The results set out below use a market approach to illustrate the benefit of purchasing a refurbished
device based on the behaviour of its initial and subsequent users.
READING THE GRAPHS:
This graph shows a simple illustration of the purchasing scenarios for refurbished laptops that have
more or less impact than buying a new phone (from a climate change perspective), based on a market
approach. This involves transferring some of the impacts of manufacturing of the new device to the
refurbished product. This market approach is connected to the theoretical usage period: 5 years for a
new laptop and 3 years for a refurbished one. We modelled 6 purchasing behaviours over a period of
10 years:
 regular buyers of refurbished but recent laptop computers: purchase a 2-year-old computer
once every 3 years;
 reasonable buyers of refurbished laptop computers: purchase a 3-year-old computer once

every 4 years;
 environmentally conscious buyers of refurbished laptop computers: purchase a computer

over 3 years old once every 5 years;
 reasonable buyers of new laptop computers: purchase every 5 years;
 environmentally conscious buyers of new laptop computers: purchase every 8 years;
 compulsive buyers of new laptop computers: purchase every 3 years.
In this paragraph, the impact on climate change depends on the type of purchase (new or refurbished),
the frequency of purchase, the 1st lifespan of the refurbished device and the period for which the device
was kept/used.
Figure 46 - Comparison with reference laptop - Depreciation approach - Market results over 10 years for climate
change
New and refurbished devices are alternatively lower-/higher-impact depending on the scenarios.
The scenario that systematically minimises impact on global warming is measured purchasing: purchasing
a refurbished device that had a first life of 5 years (or more), every 5 years (or more).

N.B.: - all quantified results for laptop computers are set out in Appendix 9.
5.2. Desktop computers
5.2.1. Overview
Definition: A computer where the main unit is intended to be in a permanent location and is not
designed for portability. It is only operational with external equipment such as display, keyboard and
mouse.37 The model considered does not include a screen, keyboard or mouse.
5.2.2. Defining reference models

 Represent the reality of the markets of the refurbishment stakeholders that participated in the
study;
 Be compatible with all the work coordinated by ADEME in early 2022 so as to provide a
Refurbished desktop computer New desktop computer
Sources of data Refurbishers (collected from 7) ADEME/ARCEP study - NegaOctet database
Refurbishers Ateliers du bocage; ATF bis repetita, ATF

GAIA, Atoutek, Ingram, LM Ecoproduction,
Recyclea
Reference model Average product based on refurbisher data Average product based on the French
collected market
Characteristics of the Refurbishment location: Manufacturing location:

reference model France (100%) Asia (100%)
France (81.43%), Europe (18.57%) Asia (100%)
Mass: 6 kg Mass: 5.43 kg
Storage: 500GB (SSD: 34% HDD: 66%) Storage: 1,500 GB (27% SSD - 73% HDD)
RAM: 16GB (20g) RAM: 10 GB
Weight of packaging (kg): 1.81 kg

0.18 units RAM
0.2 units hard disk
Consumption during use 100 kWh per year 100 kWh per year
phase (personal)
Table 50 - Presentation of the characteristics of reference desktop computers
reference models, which corresponds to the actual state of the market, as new devices are renewed based
Electricity consumption: 100 kWh per device per year The energy consumption of desktop computers is
taken from the ICT 2020 report.
N.B.: We cannot differentiate between consumption for new and refurbished devices. We therefore
opted for an average consumption. It will nonetheless be noted that, in spite of an improvement in the
37

energy efficiency of devices, those distributed are becoming ever-increasingly powerful, which reduces
the gap in energy consumption between new and refurbished devices.
Typical lifespan:
 For a new desktop computer: 5 years;
 For a refurbished desktop computer: 3 years.
5.2.3. Hypotheses and entry data
Refurbishment
Data was gathered from 7 national refurbishers. The variant parameters are:
 Location of refurbisher’s suppliers;
N.B.: - We did not obtain any data collected from international refurbishers. Fictitious profiles will
therefore be created for them.
Company Company Company Company Company Company Company

A B C E F I J
LOCATION FRANCE FRANCE FRANCE FRANCE FRANCE FRANCE FRANCE
Number of units refurbished in 50.2 6.9 1.7 2 27.5 66.2 4.9

2020 (in thousand units)
Type of refurbishment 45.16% 0% 10% 2% 0% 240% 0% 64%

- Proportion of parts
replaced
Type of parts - 58.86% 0% 95% 100% 0% 60% 100% 57%

% of parts taken from
refurbishment
Main source France 81.43% 100% 20% 100% 100% 50% 100% 100%
of products
Europe 18.57% 0% 80% 0% 0% 50% 0% 0%
Table 51 – Reference refurbished desktop computers - Presentation of refurbisher profiles - Summary
Collection data per FU

Ref. Min Max Unit
INVARIABLE PART
Supply - Previous user towards centralisation
Transport from the user to the collection point - pro
Van 1,589.64 600.91 4,019.14 kg.km
Transport from the user to the collection point – personal
Van 38.14 13.23 88.71 kg.km

Ref. Min Max Unit
Lorry 5,278.88 2,009.06 13,444.98 kg.km
Supply - Centralisation to refurbishers
Estimated transport
Lorry 15,856.57 6,012.08 40,287.08 kg.km
Packaging
Cardboard 0.50 0 0.96 Kg
Plastic 0.03 0.02 0.09 Kg
Pallet 0.23 0.03 0.3 Kg
Energy consumption
Electricity 0.88 0 13.11 kWh
Water 0 0 0.02 m3
Gas 3.57 0 31.7 MJ
Cleaning (if no info given, max values for alcohol and cotton)
Rubbing alcohol - Detergent - 0.01 0 0.03 Kg

Disinfectant, etc.
Accessories
Power supply
Supply
Lorry 89.83 7.23 600 kg.km
Ship 924.68 0 6,472.73 kg.km
Plane 10.33 0 72.29 kg.km
Packaging
Cardboard
Manufacture 0.2 0.2 0.2 Kg
Supply
Lorry 203 203 203 kg.km
Plane 2,036 2,036 2,036 kg.km
Plastic
Manufacture 0.27 0 0.43 Kg
Supply
Lorry 272 1.21 430 kg.km
Plane 3762 1,605 4,300 kg.km
Paper
Manufacture 2.2E-04 1.54E-03 0 Kg
Supply
Lorry 0.22 1.54 1.54 kg.km
Lorry 9973.92 5,996.02 31,124.50 kg.km
Van 21.43 13.15 67.59 kg.km
VARIABLE PART
RAM replacement

Ref. Min Max Unit
Manufacture
RAM 0.05 0.03 0.16 Unit
2nd-hand 0.15 0 0.84 Unit
Supply
Lorry 97.75 0.07 684.21 kg.km
Plane 60.26 0.74 421.05 kg.km
Hard disk replacement
Manufacture
SSD 0.05 0.01 0.19 Unit
HDD 0.02 0 0.08 Unit
2nd-hand 0.15 0 0.84 Unit
Supply
Lorry 27.42 0.61 172.73 kg.km
Plane 43.67 6.12 185.74 kg.km
Replacement of miscellaneous electronics
Manufacture
Component 0.0128 0 0.09 Unit
2nd-hand 0.0101 0 0.07 Unit
Supply
Lorry 0.4343 0 3.04 kg.km
Ship 0.6597 0 4.62 kg.km
Table 52 - Desktop computer - Collection data associated with refurbishment
New desktop computers
The model for desktop computers is a weighted mix of five configurations, with the following
characteristics:
Cat. 1 Cat. 2 Cat. 3 Cat. 4 Cat. 5 Average
Profile Basic PC Family PC Gaming PC Power Gaming Power User PC

PC
Distribution 18% 16% 31% 14% 21% 100%
Weight of 2.2 2.4 4.8 6.8 10.5 5.43

device (kg)
Weight of 0.44 0.48 0.96 1.36 2.1 1.09

packaging (kg)
Type of Intel Celeron Intel AMD Ryzen AMD Ryzen 5 AMD Ryzen 7
processor G3930 Pentium 5 1500X 1600 1700X
G4560
RAM (GB) 4 8 8 16 16 10
SSD storage 250 250 250 500 1000 443

(GB)
HDD storage 1000 1000 2000 2000 1172

(GB)
Motherboard 289 359.1 590.49 686.25 686.25 533.24

surface area
(cm²)

Type of Built-in MSI GeForce Sapphire GeForceRTX2080 GeForceRTX3080
graphics card GTX 1050 Radeon RX
2GT LP 570 Nitro+
4Go
External power 0.34 1.2 3.27 1.7 1.66 1.85

supply (kg)
Table 53 - Desktop computer - Description of the reference model
5.2.4. Results
Foreword
Subsequently, an analysis will be carried out on two levels:
 The full life cycle as broken down in the graph below by sectors and by sources of impact;
 The level of refurbishment only (scope delineated in red in the graph below).

Figure 48 – Reminder of the parameters for analysis and correspondence with the data collected
The results themselves will be presented using the lifespan of the device or functional unit as a reference
scale.

Reminder - Abbreviations for indicators
GWP = Climate change

ADP - Elements = Depletion of natural abiotic
resources - metals and minerals IR = Ionising Radiation
ADP - fossil = Depletion of natural abiotic PM = Fine particles

WU = Water usage
AP = Acidification
Assessment of the impact of a refurbished reference desktop

computer used for 1 year
‘Owning and using a desktop computer for personal use for one year’
usage period, i.e. 3 years for refurbished devices and 5 years for new devices.
Impacts of a refurbished reference desktop computer used for 1 year

For each year of use, the life cycle impacts of a reference refurbished desktop computer are as follows:
Value unit
Reference period 3 years
Depletion of natural abiotic resources 1.88E-04 kgeqSb per FU

Depletion of natural fossil resources (ADPf) 1,426.9 MJ per FU
Fine particles (PM) 1.95E-06 disease occurrence per FU
Table 54 – Reference desktop computer - Environmental impacts per functional unit for 1 year

Results broken down by life cycle impact
Figure 49 - Reference refurbished desktop computer - Results broken down by life cycle impact
Findings:
Refurbishment of devices represents a preponderant share of the impacts for 4 indicators (38–71%)
outside the use phase (significant for 6 indicators, at 20–90% of impacts).
Upstream and downstream logistics represent 6–43% (for the indicators for which they have a significant
impact). The impact of upstream logistics on the ADPe indicator is linked to the use of planes as well as a
low rate of replacement of new parts in the reference model.
Analysis:
of this stage is determined by the number of parts replaced, the logistics scenario and the use of second-
hand parts.

To gain a better understanding of the environmental impacts on the refurbishment stage, a detailed

Figure 50 - Reference refurbished desktop computer - Impacts broken down by stage - Focus on the refurbishment
process
Findings:
The environmental impact is divided between four major contributors:
 Supply;
 On-site consumption;
 Disk replacement;
 Packaging.
Analysis:
one company and from one scenario to another must be observed in order to consolidate the analysis
and extrapolate it to a broader reality in the field.

Once the reference model was established, the average profile for part replacement was deleted in favour
of profiles corresponding to full part replacements. This analysis sets out the variations based on the type
of replacement made.

Figure 51 – Refurbished desktop computer - Impact of replacing parts - Life cycle vision -(100% = impact of the
reference refurbished device)
Findings:
Significant variations in impact of between -13% and +422% can be observed, with disk and RAM
replacements having the highest impacts.
Analysis:
The type of parts replaced has a significant impact on all indicators, especially for disks.
Taking account of the incorporation of the use phase, the difference between the reference scenario and
the other scenarios is less significant than could be imagined, owing to consumption during the use phase.
Once the use phase has been discarded, the discrepancy between scenarios widens. The scenarios with
part replacements show higher impacts.

The reference model is fictitious. This model represents a weighted average of the market data
 Parts replacement rates vary from 0% (0–1 parts replaced, infrequently) to 240% (multiple parts
replaced, more frequently) from one refurbisher to another.
 The usage rate for second-hand parts is variable: from 0 to 100% second-hand parts.
 The estimated average distances for the total route travelled by a telephone [sic] vary by a
factor of up to 24 from one refurbisher to another. This depends on the location of the
refurbisher and its suppliers.

Figure 52 - Refurbished desktop computer - Variation of impacts depending on the companies - Impacts on climate
change
Figure 53 - Refurbished desktop computer - Variation of impacts depending on the companies - Impacts on the
depletion of abiotic mineral resources

Figure 54 - Refurbished desktop computer - Variation of impacts depending on the companies - Impacts on the
depletion of abiotic fossil resources
Figure 55 - Refurbished desktop computer - Variation of impacts depending on the companies - Impacts on ionising
radiation
of 5 to 9 between the minimum and the maximum, depending on the indicators):
For climate change:

 Company C has the lowest impact – this is a French refurbisher with a low rate of part
replacement, with 100% of replaced parts being second-hand and a supply sourced exclusively
from France;
 Company F has the highest impact. It has a part replacement rate higher than 2 parts per
device, uses only 60% second-hand parts and gets 50% of its supply from Europe.

For the depletion of natural resources:
 Company I has the lowest impact, as it does not replace any parts and gets its supply
exclusively in France;
 Company J has the highest impact, replacing numerous parts and using a lower percentage of
second-hand parts (57%).

 Company A has the lowest impact. Company F has the highest impact. Impact is in direct
correlation with the site’s consumption and the use of the French energy mix. However, with
Findings:
A very significant range of variation of impact can be observed from one refurbisher to another. This can
be explained by different methods of refurbishment and a variable percentage of second-hand parts
used.
Analysis:
In spite of a smaller sample of refurbishers and similar locations (France), the diversity of practices
involved causes marked variations in the impact results. Taking a reference model into account is
appropriate but must be done in a controlled manner. Fictitious refurbishment scenarios were therefore
established: from simple cleaning with a short supply chain to full parts replacement with a global supply
chain.
Impact Reference Minimum Maximum Unit

value value value
Reference period 3 3 3 years
Climate change (GWP) 15.42 11.45 21.46 kgeqCO2 per FU
Ecological load (MIPS) 27 19 40 Kg per FU
Depletion of natural abiotic resources - metals 1.88E-04 1.37E-04 3.12E-04 kgeqSb per FU
Depletion of natural 1,426.9 1,359.6 1,523.4 MJ per FU

fossil resources (ADPf)
Water usage (WU) 2.11 0.90 3.98 m3eq per FU
Acidification (AP) 7.82E-02 5.86E-02 1.06E-01 kgeqH+ per FU
Fine particles (PM) 1.95E-06 1.76E-06 2.21E-06 disease occurrence

per FU
Ionising radiation (IR) 178.12 173.46 182.18 kgU235eq per FU
Table 58 – Desktop computer - Environmental impacts per functional unit for 1 year - Variations between companies

Figure 56 – Refurbished desktop computer - Impact of the supply and refurbishment location - Life cycle vision
Findings:
A significant variation can be observed in impacts in the international scenario and a lesser variation in
the local scenario.
Analysis:
Given the low rate of part replacement and the local ‘supply chain’ in the reference model, the local
scenario does show a lower impact but there is less contrast in the variation than for the reference
scenario.
The international scenario, on the other hand, displays a far greater variation, owing to:
 The energy mix considered for the refurbisher’s site;
 Increased transport distances;
 The use of planes for international transport.

The second lifespan is a key parameter for the environmental impact of a refurbished computer. In the
reference scenario, we chose a lifespan shorter than the first lifespan so as to be in a realistic situation
with regard to both the behaviour and the total lifespan of the product.

Figure 57 - Desktop computer - Variation of impacts depending on the lifespan of the refurbished desktop computer
Findings:
For the indicators that change considerably based on energy consumption (ADPf, IR, PM, etc.), the
variation in impact is slight. For ADPe-type indicators, the variation in impact is very extensive.
Analysis:
Focus on RAM
eq) ce)
Reference scenario
(RAM replaced
in 5% of cases –16GB 1.88E-04 1,426.92 7.82E-02 15.42 178.12 1.95E-06 2.11 26.52
capacity)
8GB
100% 100% 99% 99% 100% 100% 96% 98%
32GB
100% 100% 102% 102% 100% 101% 107% 103%
64GB
100% 101% 107% 107% 100% 102% 122% 108%
Table 55 - Refurbished desktop computer - Sensitivity analysis for RAM capacity

e)
Full RAM 2.00E-04 1,475.97 9.92E-02 19.03 177.88 2.05E-06 4.03 38.80
replacemen
t scenario
(16GB)
8GB 100.19% 94.83% 74.62% 75.74% 99.97% 91.55% 60.64% 74.43%
32GB 99.78% 109.78% 148.41% 146.02% 100.05% 115.97% 174.74% 148.53%
64GB 99.28% 128.78% 241.71% 235.15% 100.16% 147.01% 319.96% 241.87%
Table 56 - Refurbished desktop computer - Sensitivity analysis for RAM capacity (full RAM replacement)
Findings:
When we consider a modification in the reference scenario (RAM replaced in 5% of cases), we observe
that the choice of RAM capacity has a visible but fairly insignificant impact (-+1% to +8% for indicators
showing a variation). No variation is observed for the ADPe indicator.
This variation is heightened in the scenario that includes a full RAM replacement. Variations in impact of
between -25% and +219% are observed.
Analysis:
An increase in RAM capacity correlates to an increase in the quantity of silicon wafer and the number of
masks, and thus to a significant increase in the impacts associated with the manufacture of
semiconductors. It should be noted that there is no change to the type of encapsulation or the number
of RAM sticks, which explains a stagnation in the impact on the depletion of natural resources. The RAM
capacity should therefore be adjusted to meet the requirements of the end user, thereby ensuring good
performance for the product and a genuine extension to its lifespan.
Focus on hard disks

eq) ce)
Reference scenario
(disk replaced in 8%
of cases –512GB
capacity) 1.88E-04 1,426.92 7.82E-02 15.42 178.12 1.95E-06 2.11 26.52
256GB
100% 99% 96% 96% 100% 99% 88% 93%
1024GB
100% 101% 106% 106% 100% 101% 121% 105%
2048GB
100% 103% 121% 119% 100% 105% 164% 122%
Table 57 - Refurbished desktop computer - Sensitivity analysis for hard disk capacity

ADP - ADP - AP (mol GWP (kg IR (kg PM WU (m3) MIPS
elements fossil H+ eq) CO2 eq) U235 (Disease (kg)
(kg SB eq) (MJ) eq) occurre
nce)
Full disk replacement 4.58E-04 1,562.37 1.38E-01 25.74 179.89 2.26E-06 6.73 60.09
scenario (512GB)
256GB 99.93% 90.60% 66.23% 67.59% 98.84% 85.16% 56.84% 60.85%
1024GB 100.07% 115.25% 152.49% 152.80% 98.99% 122.83% 177.09% 132.07%
2048GB 100.36% 148.34% 267.71% 266.95% 99.19% 173.17% 336.70% 227.59%
Table 58 - Refurbished desktop computer - Sensitivity analysis for hard disk capacity (full hard disk replacement)
The manufacturing techniques for SSD hard disks are very similar to those used for manufacturing RAM.
Considerable variations can thus be seen here, as in the previous sensitivity analysis.
As with RAM, the computer’s storage capacity should be adjusted to meet the end user’s needs.
Focus on the means of distribution by the private

individual
(kg SB (MJ) eq)
eq)
Reference 1.88E-04 1 426.92 7.82E-02 15.42 178.12 1.95E-06 2.11 26.52
Urban Distribution 83.68% 99.61% 98.15% 97.50% 99.99% 99.05% 98.03% 100.00%
> Shop/Collection
Point
Peri-urban 127.11% 100.65% 103.26% 104.20% 100.02% 101.60% 103.24% 100.00%

Distribution >
Shop/Collection
Point
Rural Distribution > 170.87% 101.75% 109.19% 111.37% 100.07% 104.55% 108.61% 100.00%
Shop/Collection
Point
Rural Distribution > 84.25% 99.62% 97.87% 97.51% 99.99% 99.05% 98.18% 100.01%
Home delivery
Table 59 - Refurbished desktop computer - Sensitivity analysis - Means of distribution to/by the private individual
Findings:
The choice of means of transport to private individuals has a very significant impact (-3% to +11% for the
GWP indicator and -16% to +70% for the ADPe indicator) on the impacts of a refurbished desktop
computer over 1 year.
Analysis:
The choice of the means of distribution is key in terms of limiting the environmental impact of
refurbishment. In particular, trips by private car to pick up the parcel should be avoided as much as
possible (the Periurban>Shop and Rural>Shop scenarios).
Standardisation of the results
Standardised results per world inhabitant

to be taken into account are the following:


‐ Ionising radiation (chiefly linked to use of the French energy mix);
‐ Climate change;

‐ Ionising radiation (chiefly linked to the use phase and to use of the French energy mix).
Thus, some comparative analyses will be carried out for the ADPe, ADPf and GWP indicators only.
Figure 58 – Reference refurbished desktop computer - Results for impacts after standardisation for the entire life
cycle

to this standardisation, we can see a reverse in the trend for priority indicators, and focus should be
placed on:
 Climate change;
 Climate change.

Figure 59– Reference refurbished desktop computer - Results for impacts after standardisation to planetary
boundaries for the entire life cycle
Comparison of the impacts of new and refurbished desktop computers
Comparison of the impacts of new and refurbished desktop -

5.2.4.5.1.1. Assessment results for the reference models at a lifespan
scale
Impact over the entire life cycle New desktop computer Refurbished desktop Unit
computer
Usage Excluding Usage Excluding

included usage included usage
Climate change (GWP) 289.8 256.4 46.25 26.21 kgeqCO2
Ecological load (MIPS) 1,429.0 1,348.0 79.6 29.1 Kg
Production of WEEE 5,430.21 5,430.2 17 17 G
Depletion of natural abiotic resources - 7.97E-03 7.95E-03 5.65E-04 5.55E-04 kgeqSb

metals and semi-metals (ADPe)
Depletion of natural 11 130.8 4 730.8 4281 440.8 MJ

fossil resources (ADPf)
Water usage (WU) 162.9 160.6 6.3 3.1 m3eq
Acidification (AP) 1.53 1.34 2.34E-01 1.18E-01 kgeqH+
Fine particles (PM) 1.59E-05 8.40E-06 5.84E-06 1.34E-06 disease

occurrence
Ionising radiation (IR) 1 661.5 796.5 534.35 15.35 kgU235eq
Table 60- Reference desktop computer - Environmental impacts for the entire life cycle for the reference lifespans
(over the entire life cycle)

For the reference scenario, purchasing a refurbished desktop computer (used for 3 years) instead of
purchasing a new desktop computer allowed the following impacts to be avoided annually:
Value Unit
(personal use)
Reference period for new devices 5 Years
Reference period for refurbished devices 3 Years

year
Depletion of natural abiotic resources - metals and semi-metals -1.41E-03 kgeqSb per -88%
(ADPe) year
Acidification (AP) -2.28E-01 kgeqH+ per -74%

year

occurrence per
year

year
Table 61 - Desktop computer comparison - Substitution approach - Impacts avoided for the reference scenarios
of the new product and the refurbished product, as well as on the refurbishment scenario. This is
therefore a reference evaluation. Avoided impacts are calculated as the impacts per FU for refurbished
laptops minus the impacts per FU for new laptops:
𝐼𝑚𝑝𝑎𝑐𝑡 𝑟𝑒𝑓𝑢𝑟𝑏 𝑑𝑒𝑣𝑖𝑐𝑒𝑠 𝐼𝑚𝑝𝑎𝑐𝑡 𝑛𝑒𝑤 𝑑𝑒𝑣𝑖𝑐𝑒𝑠
𝐼𝑚𝑝𝑎𝑐𝑡𝑠 a𝑣𝑜𝑖𝑑𝑒𝑑 𝑎𝑛𝑛𝑢𝑎𝑙𝑙𝑦
𝐷𝑢𝑟at𝑖𝑜𝑛 𝑜𝑓 𝑢𝑠𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑠𝑒𝑐𝑜𝑛𝑑 𝑙𝑖𝑓𝑒 𝐷𝑢𝑟at𝑖𝑜𝑛 𝑜𝑓 𝑢𝑠𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑓𝑖𝑟s𝑡 𝑙𝑖𝑓𝑒
At the scale of a private individual, substituting a new desktop computer with a refurbished computer
can prevent:
budget;
 127.19 km travelled by car;
 the extraction of 245 kg of materials;
 the production of 1.07 g of electronic waste.

carried out using the substitution approach between:
An average new product with a lifespan of 5 years;
A fictitious reference refurbished product with a usage period of 3 years;
A product refurbished with a short supply chain*, with basic cleaning;
A product refurbished with a short supply chain*, with new replacement parts;
A product refurbished with a short supply chain*, with second-hand replacement parts (taken from
recycling);
A product refurbished on a globalised market**, with basic cleaning;

A product refurbished on a globalised market**, with new replacement parts;
A product refurbished on a globalised market**, with second-hand replacement parts;
*Product collected and refurbished in France **Product collected in the USA and refurbished in China
Figure 60 - Desktop computer comparison - Substitution approach - Results for 1 year
At the functional unit scale (considering a first lifespan of 5 years and a second lifespan of 3 years), we can
observe that:
Refurbishment is low impact if:
 No replacements are made, or only minor replacements;
 Major parts are replaced using second-hand or locally sourced parts;
Refurbishment may be disadvantageous if:
 Multiple parts are replaced; RAM and disk are replaced using new parts and involving international
logistics.
Comparison of the impacts of new and refurbished desktop

computers - Depreciation approach
second life.
The table below illustrates the results of this approach. In the interests of greater readability, the analysis
was carried out on the 3 priority indicators, i.e. climate change (GWP), depletion of natural mineral and
metal resources (ADPe) and depletion of natural fossil resources (ADPf).

Figure 61 - Desktop computer comparison - Depreciation approach - Variation in environmental impact by duration
of first life.
The first figures show the variation in impact based on the duration of the first life. It will be observed
refurbished desktop computer. If refurbishment takes place before the third year has elapsed,
refurbishment will not be favourable. This reinforces the general conclusion that making your device last
is better than replacing it.
seen even more clearly that if refurbishment involves replacing a large number of parts, it is less favourable
except where the second lifespan is at least equal to the first lifespan.

Figure 62 - Desktop computer comparison - Depreciation approach - Variation in environmental impact by duration
of first life and type of refurbishment.
Analysis:
If we assume a second lifespan of 3 years and if refurbishment takes place before those 3 years have
passed, it generates an additional impact. In order to be beneficial, refurbishment must allow the lifespan
after refurbishment to be extended by over 4 years.
N.B.: - in the current market for refurbished goods, operations involve few part replacements or
substitutions for second-hand equipment, which indicates that the reference model is close to the
minimum model.
If all parts are replaced, refurbishment must occur after 4 years and enable the lifespan to be extended
by over 5 years.

The results set out below use a market approach to illustrate the benefit of acquiring a refurbished device
based on the behaviour of its initial and subsequent users.
READING THE GRAPHS:
This graph shows a simple illustration of the purchasing scenarios for refurbished desktop computers
that are lower or higher impact than buying a new desktop computer (from a climate change
perspective), based on a market approach. This involves transferring some of the impacts of
manufacturing the new device to the refurbished product. This market approach is connected to the
theoretical usage period: 5 years for a new desktop computer and 3 years for a refurbished one. We
modelled 6 purchasing behaviours over a period of 10 years:
regular buyers of refurbished but recent desktop computers: purchase a 2-year-old computer once
every 3 years;
reasonable buyers of refurbished desktop computers: purchase a 3-year-old computer once every 4
years;
environmentally conscious buyers of refurbished desktop computers: purchase a computer over 5

years old once every 5 years;
reasonable buyers of new desktop computers: purchase every 5 years;
environmentally conscious buyers of new desktop computers: purchase every 8 years;
compulsive buyers of new desktop computers: purchase every 3 years.
In this paragraph, the impact on climate change depends on the type of purchase (new or refurbished),
the frequency of purchase, the length of the 1st life of the refurbished device and the period for which
the device was kept/used.
Figure 63 - Comparison with reference desktop computer - Depreciation approach - Market results over 10 years for
climate change
New and refurbished devices are alternatively lower-impact and higher-impact depending on the
scenarios.

The scenario that systematically minimises impact on global warming potential (GWP) is measured
purchasing: purchasing every 5 years (or more) a refurbished device that had a first life of 5 years (or
more).
N.B.: - all quantified results for desktop computers are set out in Appendix 9.
6. Limits
Determining the environmental impacts of refurbished products in a precise and exhaustive manner is a
complex task confronted with numerous limits, owing to the variability of the refurbishment process, the
types of stakeholder (from businesses in the social and solidarity-based economy to international
businesses) and the fact that the business sector is so new. The following limits in particular have been
identified:
Limits associated with knowledge of user behaviour in terms of duration of use
Little information is available on the duration of use of devices. The data chosen was taken from literature
or from the experience of professionals. However, the differences in behaviour are very significant and
lead to extremely wide variations in the results. Changes in lifespans may be the origin of a radical
modification to conclusions.
To take account of this point, we:
 Incorporated a sensitivity analysis on the subject;
 Incorporated the depreciation approach with different first lifespans;
 Chose to conceptualise behaviours taking account of different first and second lifespans, in
order to issue recommendations.
Generally speaking, the results need to be communicated accompanied by the first and second lifespans
considered.
Limits associated with the degree to which the stakeholders represent the market
Smartphones - The stakeholders who participated in this study sold almost 990,000 units in 2020, 35% of
the French market. The results of the study, and thus the reference model, could be improved by
increasing this coverage rate.
This proportion was not evaluated for the other product categories.
So as to be able to extend the study to other situations, we conceptualised specific situations based on:
 The type of refurbishment;
 The origin of the products and the location of the suppliers.
This enabled us to establish ranges of variation in the results. In addition, we incorporated both the
reference scenario and its variants in all the comparisons.
Limits associated with data quality
In spite of the involvement of the participants, the data collected is of variable quality levels, owing to
disparities in the way in which each participant keeps track of processes.
We thus harmonised the data and employed a homogeneous (penalising) strategy for processing missing
data and for processing distances.

Uncertainties as to user behaviours and lifespans
Given the number of devices in users’ possession but not in use, the frequencies at which devices are
renewed is very difficult to assess. We therefore had to consider realistic but fictitious lifespans.
Parts replacements for new devices
To extend the lifespan of new devices, parts replacements (particularly new batteries) may be useful.
These replacements were not taken into consideration, owing to a lack of data.
Evaluating the period by which the second life is extended
One might imagine that renewal including parts replacement would allow for a greater extension of a
device’s lifespan. However, this aspect was not taken into consideration, owing to a lack of data. We did
add a sensitivity analysis on the issue.
Evaluating transport scenarios
Logistics scenarios were evaluated using maximising generic data so as to treat all stakeholders in the same
manner. Sensitivity analyses showed that this approach was acceptable.
Uncertainties linked to the maturity of indicators
The indicators for water usage and ionising radiation sometimes show extreme results that can raise
questions. These indicators, although relevant to the digital sector, will need to be improved in the future.
In addition, the indicator for water usage was removed with regard to the end of life of electrical and
electronic devices, as it showed an anomaly that will need to be corrected.
All of these limits were taken into account via sensitivity analyses and the conceptualisation of behaviours,
particularly for comparisons with new devices. We are therefore in a position to present
recommendations that are pertinent and are not confined to the reference scenario. A variability can be
observed in the conclusions and results but this does not undermine the environmental benefits of
refurbishment, as long as the latter is practised in a non-premature manner and replaces parts only where
strictly necessary.

7. Conclusions
The purpose of this study is to assess the multi-criteria environmental impact of the refurbishment of
digital devices destined for the general public (smartphones, tablets, desktop computers and laptop
computers) and to identify the difference in impact between using refurbished devices and purchasing
new devices.
To do this, we implemented a 2-stage approach:

 Creation of a reference model based on data collected from refurbishers
 Conceptualisation of these models to:
Assess the impact of refurbishing practices
Assess the impact of user behaviours
We were thus able to determine the impacts of a reference refurbished device, compare them to the
impacts of a new device in the same category and identify the main contributors. This gives us a multi-
criteria and comparative view of the impact of this practice.
Impact Refurbished Refurbished tablet Refurbished laptop Refurbished desktop

smartphone computers computer
DIRECT IMPACTS AVOIDED IMPACTS AVOIDED IMPACTS AVOIDED IMPACTS AVOIDED
Reference 2 3 2 5 3 5 3 5
lifespan
Climate change 3.80 -87% 5.63 -78% 9.52 -73% 23.65 -60%
(GWP)
Ecological load 12.64 -86% 20 -80% 15 -90% 40 -86%

(MIPS)
Production of 10.99 -84% 40 -75% 8 -97% 16 -99%

WEEE
Depletion of 2.98E-04 -64% 4.19E-04 -70% 1.18E-04 -93% 4.72E-04 -71%

natural abiotic
resources -
metals and
semi-metals
(ADPe)
Depletion of 102.64 -76% 303.3 -46% 491.3 -40% 1557.5 -31%

natural fossil
resources
(ADPf)
Water usage 6.79 -77% 13.97 -62% 2.41 -94% 4.06 -88%
(WU)
Acidification 2.11E-02 -87% 3.38E-02 -79% 4.88E-02 -76% 1.20E-01 -61%

(AP)
Fine particles 2.01E-07 -79% 4.59E-07 -60% 8.16E-07 -48% 2.40E-06 -26%
(PM)
Ionising 8.96 -26% 35.57 -18% 54.06 -18% 181.01 -46%

radiation (IR)
Table 62 - Summary of the results for the various product families.
N.B.: - The percentages of impacts prevented are calculated based on the impact value (100%) of the
new device for the reference lifespan, scaled to 1 year of use.

For the reference cases chosen, and considering that refurbishment can extend the lifespan of the device
over and above the reference first lifespan, refurbishment allows for significantly lower resources and
emissions than the manufacture of a new device, regardless of the product family, of between -18 and -
87% annually.
However, this conclusion is liable to vary, based on 5 major factors:
 The device’s first and second lifespan: Generally speaking, the lifespan can diminish the
impacts of manufacture, distribution and end of life. The longer the lifespan, the lower the
impact. Where refurbishment takes place before the end of the theoretical first lifespan
(according to the depreciation approach) and/or involves the replacement of several parts, the
impact of a refurbished product can be equivalent to or even greater than that of a new device
that has been kept for several years. It is therefore essential to ensure that the market for
refurbished goods is based on a principle of extending the lifespan and reducing the
consumption of devices.
 Rate and type of parts replacement (screens, batteries, RAM and disk): When new, spare parts
and accessories form the main contributor to the impact of refurbishment. The manufacture of
screens, batteries, sticks of RAM and hard disks requires a large quantity of energy and water
resources, and these have strong impacts. For laptop computers, these 4 components represent
over 50% of the impact of a new device. Parts should therefore be replaced only where
necessary. Using second-hand parts (where possible) is a promising avenue in terms of
guaranteeing good-quality service.
 The supply country: The country supplying the devices for refurbishment has a significant
impact. Indeed, distances and means of transport have a significant impact. This is especially
true as products as often sent by plane. In addition, some pools are favoured by some very
aggressive renewal offers (every year in the US) and the industry then relies on overconsumption
in certain markets in order to function.
 Adding new accessories: In some cases, the addition of accessories is compulsory and/or
systematic. Manufacturing these accessories represents a significant impact and could be
avoided by sending accessories only on request.
 The location of the refurbisher: The closer a refurbisher is to their market, the lower the impact
of final distribution.
These five topics lead to variations of up to a factor of 11 between the environmental impacts of the
lowest-impact refurbishers and those with the highest impact. So as to be able to identify beneficial and
harmful scenarios more easily, we chose to conceptualise refurbishment methods based on different
profiles, from simple cleaning with a short supply chain to the replacement of all parts carried out
internationally. For smartphones, the benefits of refurbishment are relatively obvious. However, for the
refurbishment of laptops involving parts replacement, care should be taken to ensure that the
refurbishment allows the lifespan to be extended considerably.
In addition, comparisons with new devices were made using 2 approaches:
 Substitution approach
 Depreciation approach
If we focus on the substitution approach, refurbishment proves to be positive in all cases for some families
of products, such as smartphones, tablets, servers and consoles, in spite of a shorter second lifespan. For
other product families, this conclusion is not as clear-cut.
The depreciation approach can be used to complement and qualify this principle. Since the
environmental impact of mobile phones and tablets is chiefly associated with the manufacture of these
devices, the end user’s renewal behaviour has a heavy influence on the environmental impact of their
device. Ideally, refurbishment should take place at the end of the first lifespan, and the second user
should keep the device for as long as possible. It is therefore recommended that the products sourced
should be those that have had a real second life, instead of making the market for refurbished goods an
endorsement of the overconsumption associated with certain economic models (US premium offer with
a new mobile phone being provided every 6 months to 1 year).

Thus, where the first lifespan has been shortened, the second lifespan should be extended in order to
absorb the additional impacts associated with refurbishment. The impact on climate change thus
depends on the type of purchase (new or refurbished), the frequency of purchase, the length of the 1st
life of the refurbished device and the period for which the device was kept/used.

8. General recommendations
In order to maximise the reductions in environmental impacts generated by the industry, we recommend
that:
 Users:
‐ Protect and repair their devices and make them last

‐ Buy accessories only when necessary
‐ Opt for local refurbishment with a short supply chain
‐ Choose older devices to ensure they are in a genuine dynamic of circular economy and
second life, so as not to encourage a premature end to the devices’ first life, and so as not
to promote a market for refurbished goods based on overconsumption
‐ Keep their refurbished devices for as long as possible
‐ Choose refurbishers who are genuinely committed to evaluating the impact of their
products and who use the best practices (see above)
‐ Choose the right level of device for their needs
 Refurbishers:
‐ Do not replace parts systematically

‐ Opt to use second-hand parts for replacements
‐ Optimise the volume, mass and materials of packaging
‐ Refurbish close to their market, using products taken from that same market
‐ Set up an after-sales or product-service system to prevent parts being replaced as a matter
of course
‐ Offer customers the right level of devices for their needs
‐ Identify the thresholds beyond which repairs no longer have any environmental benefit
(particularly for PCs and derivatives)
‐ Investigate opportunities for determining the age of devices for refurbishment and
highlight this information to customers
 Distribution platforms:
‐ Organise the trade-in of replaced devices so as to increase the size of the pool
‐ Promote products with a local supply chain
‐ Work with refurbishers to set up an after-sales or product-service system to prevent parts
being replaced as a matter of course upstream, but instead to offer a service for repairing
or renewing the battery so as to make the device last as long as possible
 Legislators/public authorities:
‐ Make adding a charger optional (on request, for example)

‐ Implement measures to improve the level of devices collected (make sure there is a deposit;
give a baseline repurchase value; subsidise the refurbishment of equipment that is older
than the depreciation period for its first life; etc.)
‐ Organise the traceability of devices and record the date on which they are released onto
the market
‐ Implement measures to extend lifespans.
 Today, the limiting factor for the second life is the ability to prolong usage beyond 6
years and the loss of equipment value after 3–4 years. Legislating on maintaining
operating systems for longer than 5 years and on the ecological design of mobile

applications (for public services as a minimum) so as to make them lighter and more
backwards-compatible would limit the degree of material obsolescence combined with
software obsolescence.
 Prohibit manufacturers from engaging in practices to limit the use of second-hand spare
parts – such as pairing spare parts – and ensure that refurbishment or repair are available
at reasonable rates outside recognised circuits.
 Tighten the requirements of the repairability index to facilitate parts replacement by
users or refurbishers.
 Help the industry structure itself so as to be in a genuine position to control its impact.

9. Appendices
9.1. Critical review report
Empa
Technology & Society Lab (TSL) Empa
Lerchenfeldstrasse 5 Materials Science and Technology
CH‐9014 St. Gallen
T +41 58 765 76 72
www.empa.ch/tsl
Report
Critical Review by the firms of DDemain, EFL

Consulting and LCIE Bureau Veritas
“Assessment of the environmental impact of a
set of refurbished products”
Client: ADEME
20, avenue du Grésillé, 49004 Angers (France)
Number of pages: 6
Version: V1.0

Final Critical Review Report Page 5/6
“Assessment of the environmental impact of a set of refurbished products”
1 INFORMATION ON THE PROCESS
The critical review process in accordance with ISO 14’040ff standards, as described in this document,
took place between June 2021 and June 2022, and was commissioned by ADEME, the ordering party for
the “Assessment of the environmental impact of a set of refurbished products”, drawn up by the staff of
DDemain, EFL Consulting and LCIE Bureau Veritas.
This critical review report is based on the study’s final report, dated June 2022. The final version of this
final critical review report will be incorporated into the final version of the study.
1.1 Critical review committee

The members of the critical review committee are:
Roland Hischier (June 2021 -> Empa

June 2022) Technology & Society Lab (TSL)
Lerchenfeldstrasse 5
9014 ST-GALLEN
(Switzerland)
Stéphane Le Pochât EVEA Evaluation & Accompagnement

(June 2021 -> February 2022) 11 rue Voltaire
Supported during the second 44000 NANTES
phase by: Nicolas Béalu (France)
1.2 Process for the critical review

The critical review took place as a review supporting the entire study, i.e. the authors from DDemain, EFL
Consulting and LCIE Bureau Veritas had already incorporated the members of the critical review shortly
after the study was launched. A total of 5 (telephone) meetings were held between the members of the
critical review committee, the authors of the study and the ordering party.
During the first phase, over the course of a series of telephone meetings (mainly concerning presentations
on the context and methodology of the study), the authors of the study agreed with the members of the
critical review on the main methodological choices to be adopted for this study.
During the second phase (accompanied by an intermediate report on smartphones with its Excel
calculation sheet – sent to the critical review committee following the third and fourth meetings in
October 2021), an individual critical review was carried out by each member of the critical review
committee and their comments on this work were sent to the authors of the study.
During the third phase (accompanied by an initial version of the full report, again with Excel calculation
sheets – sent to the critical review committee following the fifth meeting in March 2022), an individual
critical review

was carried out by Roland Hischier, acting as the sole member of the critical review. He sent his comments
to the authors of the study.
The same procedure was applied during the fourth phase, with a provisional final report on the study
released in June 2022. A fresh individual critical review was carried out by Roland Hischier, who sent his
comments to the authors of the study. Based on an amended version of the report, Roland Hischier then
drafted this critical review report.
The entire critical review procedure – in spite of its length – took place in an open and constructive
atmosphere. The ordering party for the study (ADEME) was involved in all the discussions mentioned
above and showed particular interest in ensuring that the entire study was performed professionally and
impeccably.
2 GENERAL REMARKS
The entire critical review procedure consists of complying with the critical review procedure set out in
the ISO 14044:2006 standard. Based on the questions set out in that standard, the critical review for this
study examined the following aspects:
Question Verdict Focal point for the reader

Are the methods used to carry out Yes
this study consistent with the
According to the report, the purpose of this study is ‘to assess
objectives and scope of the study
the environmental impact of the refurbishment of digital
and in compliance with the ISO
devices destined for the general public (smartphones, tablets,
14040:2006 and ISO 14044:2006
desktop computers and laptop computers) and to identify the
standards?
difference in impact between using refurbished devices and
purchasing new devices’. To achieve this goal, LCA – life cycle
analysis – is the methodology of choice.
Are the methods used to carry out Yes
this study valid from a scientific The aim of the study is to provide an assessment of the
and technical perspective? environmental impacts of various refurbished electronic
products, focusing primarily on the direct impacts. The
authors chose to use ‘Owning and using a device for one year’
as a functional unit – which is consistent with the aim of the
study.
For the boundaries of the system, the authors took account of
the whole life cycle of a new device, and took account of the
fact that there was no end‐of‐life stage for refurbished
devices. The latter represent a sort of extension of the use of
devices – and the end of life is thus connected to the device.
The comparison between refurbished and new was drawn
using two different approaches – a ‘traditional’ substitution
approach, where the refurbished device simply

replaces the new one, and, as a second approach, a

depreciation approach, where part of the impact of the new
device is carried over and depreciated by the refurbished
device, if the new device has not yet reached its end of life.
This second approach allows the study to include the issue of
the rate of exchange within the sector and to take account of
the topic of sustainable consumption.
For the assessment, the study not only focuses on greenhouse
gases (as do a lot of studies currently) but also presents an
approach – relying on a report from ADEME – featuring a total
of 6 impact indicators and 3 additional indicators representing
different aspects of resource consumption and waste
production. All this is aligned with the aim of the study – to
assess the environmental impacts of various refurbished
products.
Are the data and hypotheses used Yes
in this study appropriate and
reasonable for the objectives of
The data was collected from stakeholders in the
the study?
refurbishment sector, meaning that it is primary data.
In light of the extensive variations between stakeholders, the
authors drew up different scenarios for refurbishment as well
as for use.
The authors present the data and the hypotheses used in the
study in a very transparent manner. As a backdrop, they used
a range of databases so as to ensure that an adequate data
set was available for each aspect.
All this gives the readers of this study a very comprehensive
view of the subject – in perfect harmony with the purpose of
the study.
Do the interpretations reflect the Yes
objectives and also any potential As mentioned above, the aim of the study is to provide an
limits identified for the study? assessment of the environmental impacts of various
refurbished electronic products. The results for the 4 devices
in question (smartphones, tablets, laptop computers and
desktop computers) are shown logically and completely
transparently. This allows readers to form their own
interpretations and comparisons.
Basically, the documentation in this report gives a transparent
overall view of the data used and the limits of that data. The
limits are described in particular detail in Chapter 6 of the
report, in a clear and well‐structured manner.

Final Critical Review Report Page 5/6

Is the study report transparent Yes
and consistent with the objectives
The report’s clear and complete structure gives readers the
of the study?
necessary framework and information to understand the
entire report on refurbishment easily.
There are a few imprecise aspects of the report that we feel
could be corrected (see the table with the remaining
comments at the end of the report).
Are the conclusions of the report Yes
compatible with the objectives of In their conclusions, the authors illustrate in a very clear and
the study? structured manner, in only a few pages, the ecological
benefits of refurbishment and the key elements to be taken
into account in order to achieve these benefits.
3 Summary and Conclusion
The study complies with the international standards for life cycle analyses (LCA) (ISO 14040 and 14044)
and, in my capacity as a critical reviewer, I can endorse the publication of the results of this study.
St Gallen (Switzerland), 30 June 2022
[signature]
R. Hischier
Head of the ‘Advancing Life Cycle Assessment’ Group

9.2. Collection file example – synopsis

9.3. Data used for modelling
Source EIME/Ecoinvent Name Date Time Location TiR GR TeR Quality
validity
Transport Car - Passenger ecoinvent 3.6 market for transport, passenger car; RER // 2015 2022 Europe 2 1 1 1.3
transport transport, passenger car
Transport Transport bus ecoinvent 3.6 market for transport, regular bus; GLO // 2015 2022 World 2 1 3 2
transport, regular bus
Transport Lorry 17T ecoinvent 3.6 market for transport, freight, lorry 16-32 metric ton, EURO1; ZA // 2015 2022 Europe 2 1 1 1.3
transport, freight, lorry 16-32 metric ton, EURO1
market for transport, freight, lorry 16-32 metric ton, EURO2; ZA //
market for transport, freight, lorry 16-32 metric ton, EURO3; RER //
market for transport, freight, lorry 16-32 metric ton, EURO2; ZA //
Transport Lorry 3.3T ecoinvent 3.6 market for transport, freight, light commercial vehicle; Europe 2015 2022 Europe 2 1 3 2
without Switzerland // transport, freight, light commercial vehicle
Transport Container ship ecoinvent 3.6 market for transport, freight, sea, container ship; GLO // transport, 2015 2022 World 2 1 1 1.3
freight, sea, container ship
Transport Plane ecoinvent 3.6 market for transport, freight, aircraft, unspecified; GLO // transport, 2015 2022 World 2 1 1 1.3
freight, aircraft, unspecified

validity
Transport Train ecoinvent 3.6 market for transport, freight train; Europe without Switzerland // 2015 2022 Europe 2 1 1 1.3
transport, freight train
Equipment Smartphone NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
Base Impact in France’, 2021
Equipment Tablets NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
Equipment Laptop computer NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
Equipment Desktop NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
computer Base Impact in France’, 2021
Equipment Server NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
Equipment Home games NEGAOCTET - ADEME-ARCEP study - ‘The environmental impact of the digital sector 2021 2028 China 1 2 2 1.7
console Base Impact in France’, 2021
Packaging Cardboard (50% EIME-2020-12 & ADEME-ARCEP study - ‘The environmental impact of the digital sector 2012 2019 Europe 5 2 2 3
recycled) ELCD in France’, 2021
Packaging Cardboard- EIME-2020-12 & Corrugated cardboard; 5 layers; production mix, at plant; primary 2012 2019 Europe 2 2 2 2
PrimaryProd ELCD production; RER
Packaging Cardboard- EIME-2020-12 & Corrugated cardboard; 5 layers; production mix, at plant; secondary 2012 2019 Europe 2 2 2 2
SecondaryProd ELCD production; RER
Packaging Shaped plastic EIME-2020-12 & 5 2 2 3

ELCD
Packaging Polyethylene Film ecoinvent 3.6 polyethylene production, low density, granulate; RER // polyethylene, 2017 2024 Europe 2 2 2 2
low density, granulate
Packaging Polystyrene EIME-2020-12 & Polystyrene expandable granulate (EPS); production mix, at plant; RER 2011 2018 Europe 2 2 2 2
ELCD
Packaging Paper (50% EIME-2020-12 &

recycled) ELCD
Packaging Paper (from virgin EIME-2020-12 & Paper; production mix, at plant; from virgin fibre; RER 2000 2007 Europe 5 2 2 3
fibre) ELCD

validity
Packaging Paper (recycled) EIME-2020-12 & Paper; production mix, at plant; with deinking, 100% recycled, from 2014 2021 Europe 1 2 2 1.7
ELCD wastepaper; RER
Packaging Pallet ecoinvent 3.6 Pine wood; timber; production mix, at saw mill; 40% water content; 2011 2018 Germany 5 3 3 3.7
DE
Cleaning Cloth EIME-2020-12 & Cotton fabric, from intensive production; production mix; RER 2008 2015 Europe 4 2 3 3
ELCD
Cleaning Detergent EIME-2020-12 & Unspecified organic chemicals; average production; production mix, 2000 2007 Europe 5 3 5 4.3
ELCD at plant; RER
Spare Tempered glass EIME-2020-12 & Glass; for photovoltaic application; production mix, at plant; RER 2000 2007 Europe 5 3 4 4
parts; ELCD
Spare Smartphone LCD EIME-2020-12 & Capacitive touchscreen LCD display panel, colour, high-tech 2020 2027 China 1 1 1 1
parts; screen ELCD application
Spare Smartphone OLED NegaOctet Capacitive touchscreen OLED display panel, colour 2020 2027 China 1 1 3 1.7
parts; screen Projet
Spare Spare parts - Environmental impact of refurbishment (data collected as part of the 2021 2028 France 1 1 1 1
parts refurbished - study)
screen
Spare Spare parts - Environmental impact of refurbishment (data collected as part of the 2021 2028 France 1 1 1 1
parts refurbished study)
Spare Battery EIME-2020-12 & Nickel cobalt manganese li-ion (NCM) battery, prismatic; production 2013 2020 GLO 2 1 2 1.7
parts ELCD mix, at plant; GLO
Spare Processor NegaOctet + 2021 2028 China 1 1 3 1.7

parts French PCR
Spare RAM; DDR5, 2GB NegaOctet + Random-access memory, RAM; DDR5, 2GB; production mix, at plant; 2021 2028 China 1 1 3 1.7
parts French PCR CN

validity
parts French PCR CN
parts French PCR CN
parts French PCR CN
parts French PCR CN
parts French PCR CN
Spare RAM; DDR5, NegaOctet + Random-access memory, RAM; DDR5, 128GB; production mix, at 2021 2028 China 1 1 3 1.7
parts 128GB French PCR plant; CN
Spare Solid State Drive NegaOctet + Solid State Drive (SSD); 2.5", SLC, 256GB; production mix, at plant; CN 2021 2028 China 1 1 3 1.7
parts (SSD); 2.5", SLC, French PCR
256GB
Spare Solid State Drive NegaOctet + Solid State Drive (SSD); 2.5", SLC, 512GB; production mix, at plant; CN 2021 2028 China 1 1 3 1.7
parts (SSD); 2.5", SLC, French PCR
512GB
Spare Solid State Drive NegaOctet + Solid State Drive (SSD); 2.5", SLC, 1024GB; production mix, at plant; 2021 2028 China 1 1 3 1.7
parts (SSD); 2.5", SLC, French PCR CN
1024GB

validity
Spare Solid State Drive NegaOctet + Solid State Drive (SSD); 2.5", SLC, 2048GB; production mix, at plant; 2021 2028 China 1 1 3 1.7
parts (SSD); 2.5", SLC, French PCR CN
2048GB
Spare Solid State Drive NegaOctet + Solid State Drive (SSD); 2.5", MLC, 256GB; production mix, at plant; 2021 2028 China 1 1 3 1.7
parts (SSD); 2.5", MLC, French PCR CN
256GB
Spare Solid State Drive NegaOctet + 2021 2028 China 1 1 4 2

parts (SSD); 2.5", MLC, French PCR
512GB
1024GB
2048GB
Spare HDD NegaOctet + Hard disk drive; mix of 2.5" and 3.5", mix of aluminium and glass disks 2021 2028 China 1 1 3 1.7
parts French PCR
Spare Server power NegaOctet + Power supply unit (PSU), computers, rack and mainframe servers; 2021 2028 China 1 1 3 1.7
parts supply French PCR input AC 100-240V, output DC 12V
Spare DVD player NegaOctet + DVD-ROM Drive 2021 2028 China 1 1 3 1.7
parts French PCR
Accessorie Power Supply NegaOctet + External power supply (EPS); production mix, at plant; for mobile 2017 2024 China 1 1 2 1.3
s French PCR device, input AC 100-240V, output DC 5V
Accessorie USB cable EIME-2022-01 EIME modelling 1 1 2 1.3

s

validity
Accessorie Earbuds EIME-2022-01 2021 2021 Europe/C 1 3 3 2.3

s hina
Accessorie SIM card extractor EIME-2022-01 Steel electrogalvanised; 35% recycled; production mix, at plant; 2006 2013 Europe 5 3 3 3.7
s thickness 0,3 to 3 mm, width 600 to 2100 mm; GLO
Consumpti Water ecoinvent 3.6 Drinking water; water purification treatment; production mix, at 2012 2019 Europe 5 2 1 2.7
on plant; from surface water; RER
Consumpti Gas - France EIME-2022-01 Process steam; consumption mix, at consumer; MJ; FR 2009 2016 France 4 1 3 2.7
on
Consumpti Gas - China EIME-2022-01 Process steam; consumption mix, at consumer; MJ; CN 2009 2016 China 4 1 3 2.7
on
Consumpti Gas - US EIME-2022-01 Process steam; consumption mix, at consumer; MJ; US 2009 2016 US 4 1 3 2.7
on
Consumpti Gas - Europe ecoinvent 3.6 Process steam; consumption mix, at consumer; MJ; RER 2008 2015 Europe 4 1 3 2.7
on
Consumpti Electricity mix - EIME-2022-01 2018 2025 France 2 1 1 1.3

on France
Consumpti Electricity mix - EIME-2022-01 2018 2025 Europe 2 1 1 1.3

on Europe
Consumpti Electricity Mix - EIME-2022-01 2018 2025 China 2 1 1 1.3

on China
Consumpti Electricity Mix - EIME-2022-01 2018 2025 Romania 2 1 1 1.3

on Romania
Consumpti Electricity Mix - EIME-2022-01 2018 2025 Germany 2 1 1 1.3

on Germany
Consumpti Electricity mix - EIME + IAE 2018 2025 Asia- 2 1 1 1.3

on Asia-Pacific Pacific

validity
Spare Controller EIME-2022-01 2021 2028 China 1 1 3 1.7

parts
End of life Paper/cardboard- EIME-2020-12 & Europe 2 3 2.5

EoL-Tot ELCD
End of life Paper/cardboard- ecoinvent 3.6 Waste incineration of paper fraction in municipal solid waste (MSW); 2006 2013 Europe 5 2 3 3.3
EoL-incineration average European waste-to-energy plant, without collection, transport
and pre-treatment; at plant; EU-27
End of life Paper/cardboard- EIME-2020-12 & Landfill of paper waste; landfill including landfill gas utilisation and 2005 2012 Europe 5 2 3 3.3
EoL-landfill ELCD leachate treatment and without collection, transport and pre-
dumping treatment; at landfill site; EU-27
End of life Paper/cardboard- EIME-2020-12 & Europe 2 3 2.5

EoL-recycling ELCD
End of life Plastic-EoL-Tot EIME-2020-12 & Europe 2 3 2.5

ELCD
End of life Plastic-EoL- ecoinvent 3.6 Waste incineration of plastics (PE, PP, PS, PB); average European waste- 2011 2012 Europe 5 2 3 3.3
incineration to-energy plant, without collection, transport and pre-treatment; at
plant; EU-27
End of life Plastic-EoL-landfill EIME-2020-12 & Landfill of plastic waste; landfill including landfill gas utilisation and 2011 2018 Europe 5 2 3 3.3
dumping ELCD leachate treatment and without collection, transport and pre-
treatment; at landfill site; EU-27
End of life Plastic-EoL- EIME-2020-12 & Europe 2 3 2.5

recycling ELCD
End of life woodEoL EIME-2020-12 & Europe 2 3 2.5

ELCD
End of life wood-EoL-reuse EIME-2020-12 & Europe 2 3 2.5

ELCD
End of life wood-EoL- EIME-2020-12 & Europe 2 3 2.5

recycling ELCD

validity
End of life wood-EoL- EIME-2020-12 & Waste incineration of untreated wood (10.7% water content); average 2005 2012 Europe 5 2 3 3.3
incineration ELCD European waste-to-energy plant, without collection, transport and
pre-treatment; at plant; EU-27
End of life wood-EoL-landfill EIME-2020-12 & Landfill of untreated wood; landfill including landfill gas utilisation and 2005 2012 Europe 5 2 3 3.3
dumping ELCD leachate treatment and without collection, transport and pre-
treatment; at landfill site; EU-27
End of life Soiled fabric-EoL
End of life fabric-EoL- EIME-2020-12 & Europe 2 3 2.5

recycling ELCD
End of life fabric-EoL- EIME-2020-12 & Waste incineration of textile fraction in municipal solid waste (MSW); 2005 2010 Europe 2 3 2.5
incineration ELCD average European waste-to-energy plant, without collection, transport
and pre-treatment; at plant; EU-27
End of life textile-EoL-landfill EIME-2020-12 & Landfill of textiles; landfill including landfill gas utilisation and leachate 2005 2010 Europe 5 2 3 3.3
dumping ELCD treatment and without collection, transport and pre-treatment; at
landfill site; EU-27
End of life Waste collection EIME-2020-12 & Lorry transport; Euro 0, 1, 2, 3, 4 mix; 22 t total weight, 17.3t max 2005 2010 Europe 5 2 3 3.3
ELCD payload; RER
End of life End of life - WEEE - EIME-2020-12 & EoL, Flat Screens | LCD panel, Substitution benefits included; From 2015 2022 France 2 2 2 2
Screen - cm² ELCD collection to final destinations; In the context of French WEEE take-
back schemes; FR
End of life End of life - WEEE - EIME-2020-12 & EoL, Flat Screens | LCD panel, Substitution benefits included; From 2015 2022 France 2 2 2 2
Screen – g ELCD collection to final destinations; In the context of French WEEE take-
back schemes; FR
End of life End of life - WEEE EIME-2020-12 & Waste treatment of Lithium-ion battery; by hydrometallurgical 2010 2017 Europe 2 2 2 2
– Battery ELCD process; technology mix, at waste treatment plant; RER

validity
End of life End of life - WEEE EIME-2020-12 & CODDE model - based on ESR data 2015 2022 France 2 2 2 2
– Other ELCD
End of life End of life - WEEE - EIME-2020-12 & CODDE model - based on ESR data 2015 2022 France 2 2 2 2
Other non- ELCD
electronic
Screen - ELCD
Smartphone
Screen - Tablet ELCD
Screen - Laptop ELCD
computer
Battery - ELCD
Smartphone
Battery - Tablet ELCD
Battery – Laptop ELCD
Miscellaneous ELCD
Parts -
Smartphone
Miscellaneous ELCD
Parts – Tablet
Miscellaneous ELCD
Parts - Laptop
computer
Miscellaneous ELCD
Parts - Desktop
computer

validity
2nd-hand Parts - ELCD
Smartphone
Tablet
Laptop computer
Desktop
computer
End of life Motherboard WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation
End of life Processor WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation
End of life RAM; DDR5, 2GB WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation
calculation
calculation
calculation
calculation
calculation
End of life RAM; DDR5, WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
128GB calculation
End of life Solid State Drive WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
(SSD); 2.5", SLC, calculation
256GB

validity
512GB
1024GB
2048GB
(SSD); 2.5", MLC, calculation
256GB
1024GB
512GB
2048GB
End of life HDD WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation
End of life Server power WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
supply calculation
End of life DVD player WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation
End of life Controller WEEE mass CODDE model - based on ESR data 2015 2022 France 2 2 2 2
calculation

9.4. Impact data – new equipment
9.4.1. Summary of entry data

Source: ADEME –ARCEP study – Impact of the digital sector in France – November 2021
SMARTPHONE TABLET Laptop computer Desktop computer
PRIVATE USE Cat. 1 Cat. 2 Cat. 3 Average Cat. 1 Cat. 2 Cat. 3 Average Cat. 1 Cat. 2 Cat. 3 Average Cat. 1 Cat. 2 Cat. 3 Cat. 4 Cat. 5 Average
Profile Chromebook Office Gaming Basic PC Family Gaming Power Power
PC PC Gaming User PC
PC
Distribution 33% 33% 33% 100% 33% 33% 33% 100% 40% 49% 12% 18% 16% 31% 14% 21% 100%
Weight of 0.204 0.189 0.202 0.198 0.465 0.498 0.471 0.478 1.5 1.54 2.3 1.61 2.2 2.4 4.8 6.8 10.5 5.43
device (kg)
Weight of 0.0408 0.0378 0.0404 0.039 0.093 0.0996 0.0942 0.0956 0.3 0.308 0.46 0.323 0.44 0.48 0.96 1.36 2.1 1.09
packaging (kg)
Screen size 6.59 6.57 6.72 6.626 10.2 10.3 11.1 10.5 14.5 14.5 15.6 14.6287
(inches)
Screen LCD AMOLED OLED LCD LCD LCD LCD LCD LED
technology touchscreen touchscreen touchscreen touchscreen
Type of MediaTek SnapdragonTM Exynos Huawai A10 Fusion Apple A12 AMD Ryzen 5 Intel® Intel Intel Intel AMD AMD AMD
processor Helio G85 730G 990 Kirin 659 Bionic Core™ i7, 8th Core I5 Celeron Pentium Ryzen 5 Ryzen 5 Ryzen 7
generation G3930 G4560 1500X 1600 1700X
RAM (GB) 6 7 11 8 4 4 6 4.6 16 8 16 13.3 4 8 8 16 16 10
SSD storage 128 160 341 209.6 32 256 512 266.6 427 564 512 503.801 250 250 250 500 1000 443
(GB)
HDD storage 0 0 0 0 1000 1000 2000 2000 1172
(GB)
Motherboard 125.16 117.94 122.36 121.8 400.84 421.35 461.14 427.7 52.93 62.03 88.22 61 289 359.1 590.49 686.25 686.25 533.24
surface area
(cm²)
Mass of 84 79 77 80 156.2 205.8 216.4 192.8 330 165 287 244 0
battery (g)
External 245 211 208 224 0.34 1.2 3.27 1.7 1.66 1.85
power supply
(g)
Reference Xiaomi Xiaomi Mi Note Apple Huawei Samsung Apple iPad Honor Lenovo Acer
product Redmi 10 Lite iPhone Mediapad Galaxy Tab Air 4 MagicBook ThinkPad Nitro 5
example Note9 12 Pro M3 Lite 10.1 S5e Pro (2020) X390 AN515-
Max AMD 55-51QY

9.5. Additional quantified results for the main product families
9.5.1. Smartphone family

Impact over the entire life cycle New smartphone Refurbished smartphone unit
USAGE OUT OF USAGE OUT OF

INCLUDED USE INCLUDED USE
Ecological load (MIPS) 268.6 266.7 25.3 23.1 Kg
Production of WEEE 200.20 200.2 22 22 g
Depletion of natural abiotic 2.50E-03 2.50E-03 5.97E-04 5.96E-04 kgeqSb

resources
Depletion of natural fossil 1,269.5 1,119.8 205 105.4 MJ

resources (ADPf)
Acidification (AP) 4.89E-01 4.84E-01 4.23E-02 3.92E-02 kgeqH+

occurrence
Ionising radiation (IR) 36.3 16.0 17.92 4.43 kgU235eq
Table 62–Impact over the entire lifespan of a reference refurbished and new smartphone
ADP - ADP - AP (mol GWP IR (g PM WU MIP EW (kg)

elements fossil H+ eq) (kg U23 (Disease (m3) S
(kg SB eq) (MJ) CO2 5 occurren (kg)
eq) eq) ce)
Upstream logistics: 12% 8% 12% 15% 0% 15% 1% 0% 0%

from the previous
user to centralisation
Refurbishment 78% 38% 72% 68% 24% 44% 98% 95% 100%
Distribution from the 10% 6% 9% 11% 0% 11% 1% 0% 0%

distributor to the
user
Use by the user 0% 49% 7% 7% 75% 29% 0% 5% 0%
TOTAL 2.98E-04 102.64 2.11E-02 3.80 8.96 2.01E-07 6.8 12.6 1.10E-02
Table 63 - Reference refurbished smartphone - Results broken down by life cycle impact

ADP - ADP - AP (mol GWP IR (kg PM WU MIPS EW (g)
elements fossil H+ eq) (kg U235 (Disease (m3) (kg)
(kg SB (MJ) CO2 eq) occurre
eq) eq) nce)
Supply: Supply -
Centralisation to
refurbishers 3.70E-05 7.95 2.48E-03 0.56 0.04 3.11E-08 5.92E-02 0.01 0.00E+00
Consumption of the
refurbishment site 1.12E-06 5.34 2.02E-03 0.49 0.02 8.35E-09 9.81E-02 0.79 0.00E+00
Standard
refurbishment:
Checking - Reset – 3.55E+0
Cleaning 7.06E-08 7.29 1.85E-03 0.34 0.49 1.39E-08 0 0.72 0.00E+00
Screen replacement
1.16E-08 5.07 1.91E-03 0.30 0.01 1.01E-08 9.93E-01 0.37 0.00E+00
Battery replacement
2.42E-05 5.65 2.17E-03 0.41 0.27 1.33E-08 8.20E-02 1.19 4.50E-03
Replacement of
other parts 3.83E-06 1.71 1.32E-03 0.16 0.15 5.86E-09 1.81E+00 0.64 4.48E-03
Accessories
8.40E-07 1.02 3.31E-04 0.07 0.02 2.17E-09 8.29E-04 0.03 8.98E-04
Packaging
1.98E-04 6.81 3.17E-03 0.37 1.16 1.56E-08 6.99E-02 7.69 1.10E-03
Placing in stock for

sale 4.40E-07 3.57 1.66E-03 0.36 0.02 6.32E-09 1.05E-01 0.55 0.00E+00
Distribution from the

refurbisher to the
distributor 1.19E-06 0.75 4.90E-04 0.05 0.00 3.19E-09 2.95E-03 0.01 0.00E+00
Figure 64 - Reference refurbished smartphone - Impacts broken down by stage - Focus on the refurbishment process
- Numbers

elements fossil H+ eq) CO2 eq) U235 (Disease (m3) (kg)
(kg SB (MJ) eq) occurre
eq) nce)
Reference replacement 2.98E-04 102.64 2.11E-02 3.80 8.96 2.01E-07 6.79 12.64
Reference Screen 141% 120% 137% 138% 113% 124% 80% 147%
replacement
LED Screen replacement 174% 116% 127% 119% 122% 120% 85% 159%
OLED Screen replacement 92% 125% 150% 167% 100% 130% 72% 129%
Battery replacement 95% 97% 105% 98% 102% 100% 179% 105%
Miscellaneous 90% 93% 83% 85% 95% 90% 72% 86%

replacement
Cleaning Only 90% 92% 82% 83% 95% 89% 72% 85%
Figure 65 – Refurbished smartphone - Impact of replacing parts - Life cycle vision

elements fossil H+ eq) CO2 eq) U235 (Disease (m3) (kg)
(kg SB (MJ) eq) occurre
eq) nce)
International Scenario 100.05% 112.58% 124.38% 121.29% 100.89% 111.25% 48.66% 116.30%
(US-CN-FR) - Reference
replacement
Local Scenario (FR-FR-FR) - 99.35% 101.87% 91.09% 90.91% 110.00% 100.03% 48.17% 95.22%
Reference replacement
Table 66 – Refurbished smartphone - Impact of the supply and refurbishment location - Life cycle vision
ADP - ADP - AP (mol GWP (kg IR (kg PM WU

elements fossil H+ CO2 U235 (Disease (m3/Euro
(kg SB (MJ/Europ eq/Europ eq/Europ eq/Europ occurrenc pean
eq/Europ ean ean ean ean e/Europea average)
ean average) average) average) average) n average)
average)
Upstream logistics: 5.82E-04 1.22E-04 4.47E-05 6.88E-05 9.06E-06 5.23E-05 5.15E-06

from the previous user
to centralisation
Refurbishment 3.66E-03 5.98E-04 2.75E-04 3.18E-04 5.09E-04 1.49E-04 5.79E-04
Distribution from 4.51E-04 9.06E-05 3.35E-05 5.12E-05 6.67E-06 3.82E-05 3.94E-06

the distributor
to the user
Use by the user 1.94E-06 7.68E-04 2.71E-05 3.22E-05 1.60E-03 9.83E-05 1.60E-06
TOTAL 4.69E-03 1.58E-03 3.80E-04 4.70E-04 2.12E-03 3.38E-04 5.90E-04
Table 67 – Reference refurbished smartphone - Results for impacts after standardisation (eq. world inhabitant) over
ADP - ADP - AP (mol GWP (kg IR (kg PM (Disease WU (m3) MIPS (kg)
elements fossil (MJ) H+ eq) CO2 eq) U235 eq) occurrence)
(kg SB eq)
Lifespan 1 5.97E-04 1.55E+02 4.08E-02 7.35E+00 1.12E+01 3.43E-07 1.36E+01 2.46E+01

(years)
2 2.98E-04 1.03E+02 2.11E-02 3.80E+00 8.96E+00 2.01E-07 6.79E+00 1.26E+01
3 1.99E-04 8.51E+01 1.46E-02 2.62E+00 8.22E+00 1.53E-07 4.53E+00 8.64E+00
4 1.49E-04 7.63E+01 1.13E-02 2.03E+00 7.85E+00 1.30E-07 3.40E+00 6.64E+00
Table 68 - Smartphone - Comparison of impacts depending on the lifespan of the smartphone - substitution
approach, all stages

element fossil (MJ) H+ eq) CO2 eq) U235 eq) (Disease
s (kg SB occurre
eq) nce)
Refurbished smartphone 2.98E-04 1.03E+02 2.11E-02 3.80E+00 8.96E+00 2.01E-07 6.79E+00 1.26E+01
model
Lifespan: 2 years
Cleaning only, without 2.65E-04 9.51E+01 1.52E-02 2.79E+00 9.41E+00 1.76E-07 1.34E+00 1.02E+01
accessories, supply FR,
refurbishment FR
Lifespan: 2 years
accessories, supply US,
refurbishment Asia
Lifespan: 2 years
Replacement of all parts, 4.34E-04 1.30E+02 3.19E-02 5.53E+00 1.16E+01 2.68E-07 9.18E+00 2.05E+01
with accessories, supply
FR, refurbishment FR
Lifespan: 2 years
US, refurbishment Asia
Lifespan: 2 years
Replacement of all parts 3.03E-04 1.34E+02 2.72E-02 5.30E+00 1.03E+01 2.71E-07 1.34E+00 1.02E+01
with 2nd-hand parts, with
refurbishment FR
Lifespan: 2 years
Replacement of all parts 3.04E-04 1.44E+02 3.39E-02 6.41E+00 9.45E+00 2.89E-07 1.38E+00 1.29E+01
refurbishment Asia
Lifespan: 2 years
Reference new 8.34E-04 4.23E+02 1.63E-01 2.84E+01 1.21E+01 9.71E-07 2.97E+01 8.95E+01
smartphone
Lifespan: 3 years
Table 69 - Smartphone comparison - Substitution approach - Results

ADP - ADP AP (mol GWP IR (kg PM (Disease WU MIPS
elements - H+ eq) (kg CO2 U235 occurrence) (m3) (kg)
(kg SB fossil eq) eq)
eq) (MJ)
6 Residual 1E-03 467 2E-01 35.18 6.68 1E-06 37.14 111.12

months impact
from the 1st
life
Refurbishment 2.98E-04 103 2.11E-02 3.80 8.96 2.01E-07 6.79 12.64

+ use
1 year Residual 8E-04 373 2E-01 28.14 5.35 9E-07 29.71 88.90
impact
from the 1st
life

+ use
2 years Residual 4E-04 187 8E-02 14.07 2.67 5E-07 14.86 44.45
impact
from the 1st
life

+ use
3 years Residual 0 0 0 0.00 0.00 0 0.00 0.00

impact
from the 1st
life

+ use
New FU 8.34E-04 423 1.63E-01 28.40 12.09 9.71E-07 29.73 89.53
Table 70 - Smartphone comparison - Substitution approach - Quantified results
Regular Regular Reasonable Reasonable Environmentally Systematic

purchase purchase purchase use (3 years) conscious use (6 purchase
(every 2 years) (every 2 years) (every 3 years) of a new years) of a new (every year)
of a recent of a of an older smartphone smartphone of new items
refurbished refurbished refurbished
model (1 year) model (2 model (3 years)
years)
Year 0 0 0 0 0 0 0
Year 1 32 18 4 28 28 85
Year 2 64 22 8 57 57 169
Year 3 96 40 8 85 85 254
Year 4 128 43 12 114 86 339
Year 5 160 61 15 142 86 424
Year 6 192 65 16 170 86 508
Table 71 - Comparison with reference smartphone - Depreciation approach - Cumulative results over 6 years for
climate change

9.5.2. Tablet family
New tablet Refurbished tablet Unit

Ecological load (MIPS) 301 292 40.5 2.2 Kg
Production of WEEE 479.46 479.46 80 80 g
Depletion of natural abiotic resources - metals 4.20E-03 4.19E-03 8.38E-04 8.37E-04 kgeqSb
Depletion of natural fossil resources (ADPf) 1 690 976 607 130.4 MJ
Acidification (AP) 0.48 0.46 6.77E-02 5.33E-02 kgeqH+

occurrence
Table 72- Reference tablet - Environmental impact for the entire life cycle for the reference lifespans (over the entire
life cycle)
ADP - ADP - AP (mol GWP IR (kg PM WU MIPS

element fossil H+ eq (kg CO2 U235 (Disease (m3 (kg per
s (kg SB (MJ per per eq per eq per occurrenc per year)
eq per year) year) year) year) e per year)
year) year)
Reference refurbished tablet 4.19E-04 303.27 0.03 5.63 35.57 4.59E-07 13.97 20.25
0.21 LCD screen
0.4 OLED screen
0.26 Battery
Lifespan: 2 years
Cleaning only, without 2.65E-04 282.45 0.02 3.73 34.48 3.95E-07 1.40 11.83
refurbishment FR
Lifespan: 2 years
Cleaning only, without 2.67E-04 308.31 0.03 5.97 33.74 4.26E-07 1.45 16.87
refurbishment Asia
Lifespan: 2 years
Replacement of all parts, with 1.21E-03 394.99 0.08 11.15 45.59 7.04E-07 37.77 55.37
refurbishment FR
Lifespan: 2 years
Replacement of all parts, with 1.22E-03 420.86 0.09 13.39 44.85 7.34E-07 37.82 60.40
refurbishment Asia
Lifespan: 2 years
Replacement of all parts with 3.05E-04 347.88 0.05 8.53 34.61 5.22E-07 7.15 19.17
2nd-hand parts, with
refurbishment FR
Lifespan: 2 years
Replacement of all parts with 3.05E-04 347.88 0.05 8.53 34.61 5.22E-07 7.15 19.17
refurbishment Asia
Lifespan: 2 years
Reference new tablet 1.40E-03 563 0.16 25.93 43.55 1.14E-06 36.93 100.32
Lifespan: 3 years
Table 73 - Tablet comparison - Substitution approach - Results for one year

9.5.3. Laptop family
Impact on the entire life cycle New laptop computers Refurbished laptop Unit
computers

Ecological load (MIPS) 706.8 683.2 44.0 28.1 kg
Production of WEEE 1 610.35 1 610.4 25 25 G
Depletion of natural abiotic resources - metals 8.09E-03 8.08E-03 2.87E-04 2.84E-04 kgeqSb
Depletion of natural fossil resources (ADPf) 4 113.1 1,119.8 1416 298.6 MJ
Acidification (AP) 1-01E+00 9.54E-01 1.22E-01 8.81E-02 kgeqH+

occurrence
Table 74- Reference laptop - Environmental impact for the entire life cycle for the reference lifespans (over the
entire life cycle) - Personal use
ADP - ADP AP (mol GWP IR (g PM (Disease WU MIPS EW (kg)

elements - H+ eq) (kg U235 occurrence) (m3) (kg)
(kg SB fossil CO2 eq)
eq) (MJ) eq)
Upstream logistics: 46% 3% 10% 12% 0% 9% 4% 1% 0%

from the previous
user
to centralisation
Refurbishment 33% 17% 59% 61% 7% 23% 89% 67% 100%

distributor to the user
Use by the user 1% 79% 28% 24% 93% 65% 6% 32% 0%
TOTAL 9.57E-05 472 4.06E-02 8.09 54 6.92E-07 2.16 14.7 8.18E-03
Table 75 - Reference refurbished laptop - Results broken down by life cyce impact
ADP - ADP AP (mol GWP IR (kg PM WU MIPS EW (kg)

elements - H+ eq) (kg U235 (Disease (m3) (kg)
(kg SB eq) fossil CO2 eq) occurrenc
(MJ) eq) e)
Supply: 18% 10% 15% 17% 1% 15% 8% 11% 0%

centralisation to
refurbishers
Consumption of the 0% 40% 17% 18% 85% 29% 2% 15% 0%

refurbishment site
Standard 0% 0% 1% 0% 0% 1% 9% 0% 0%
refurbishment:
France - Reset -
Cleaning
Screen replacement 10% 1% 1% 1% 1% 1% 1% 1% 18%
Battery replacement 5% 1% 3% 2% 2% 2% 38% 3% 23%
RAM replacement 4% 7% 9% 8% 0% 9% 8% 10% 5%
Disk replacement 43% 15% 22% 19% 6% 20% 19% 30% 39%

Replacement of 0% 0% 0% 0% 0% 1% 0% 0% 13%
other parts
Accessories 2% 1% 1% 1% 3% 1% 0% 3% 1%
Packaging 4% 18% 24% 27% 2% 14% 16% 21% 0%
Placing in stock for 2% 4% 5% 4% 0% 2% 0% 4% 0%

sale

refurbisher to the
distributor
Total 3.13E-05 81.5 2.41E-02 4.9 3.5 1.79E-07 1.9 9.9 8.18E-03
Table 76 - Reference refurbished laptop computer - Impacts broken down by stage - Focus on the refurbishment
process - Numbers

elements fossil H+ eq) CO2 eq) U235 eq) (Disease
(kg SB eq) (MJ) occurrenc
e)
Reference 9.57E-05 472.00 4.06E-02 8.09 53.96 6.92E-07 2.16 14.68

replacement
Reference 947% 116% 172% 153% 114% 127% 171% 293%

Screen
replacement
LED Screen 1062% 115% 166% 143% 116% 126% 188% 305%
replacement
OLED Screen 102% 122% 211% 225% 102% 137% 44% 209%
replacement
Battery 168% 103% 147% 121% 106% 111% 1857% 165%

replacement
RAM 116% 129% 240% 225% 100% 147% 276% 279%

replacement
Disk 387% 165% 426% 383% 107% 209% 504% 597%

replacement
Miscellaneous 80% 96% 79% 82% 99% 93% 42% 70%

replacement
Cleaning Only 80% 96% 79% 82% 99% 93% 42% 70%
Table 77 – Refurbished laptop - Impact of replacing parts - Life cycle vision

(kg SB eq) (MJ) eq)
Reference 9.57E-05 472.00 4.06E-02 8.09 53.96 6.92E-07 2.16 14.68

scenario
Cleaning only, 35% 75% 63% 69% 88% 67% 34% 52%
without
accessories
Replacing all parts 4019% 830% 1193% 977% 538% 1036% 2996% 1624%
Table 78 – Refurbished laptop computer - Impact of the supply and refurbishment location - Refurbisher vision

SB eq) (MJ) CO2 eq)
eq)
Supply 112% 115% 188% 174% 95% 116% 107% 196%

USA/refurbishment
China

SB eq) (MJ) CO2 eq)
eq)
Supply 97% 98% 90% 89% 101% 98% 99% 91%

France/refurbishment
France
Table 79 – Refurbished laptop computer - Impact of the supply and refurbishment location - Life cycle vision
ADP - ADP - AP (eq GWP (eq IR PM (eq WU (eq

elements fossil (eq Inhab) Inhab) (eq Inhab) Inhab)
(eq Inhab) Inhab)
Inhab)

to centralisation
Distribution from the 3.01E-04 6.08E-05 2.25E-05 3.43E-05 4.48E-06 2.57E-05 2.64E-06
TOTAL 1.50E-03 7.25E-03 7.30E-04 1.02E-03 1.28E-02 1.12E-03 2.00E-04
Table 80 – Reference refurbished laptop - Results for impacts after standardisation over the entire life cycle
(kg SB eq)

(years)
2 9.57E-05 4.72E+02 4.06E-02 8.09E+00 5.40E+01 6.92E-07 2.16E+00 1.47E+01
3 6.41E-05 4.39E+02 3.08E-02 6.04E+00 5.28E+01 6.07E-07 1.49E+00 1.14E+01
4 4.83E-05 4.22E+02 2.59E-02 5.02E+00 5.22E+01 5.64E-07 1.15E+00 9.70E+00
5 3.88E-05 4.12E+02 2.30E-02 4.40E+00 5.18E+01 5.39E-07 9.47E-01 8.70E+00
Table 81 - Laptop - Comparison of impacts depending on the lifespan - substitution approach, all stages
ADP – ADP – AP (mol GWP (kg IR (kg PM (Disease WU
elements (kg fossil H+ eq) CO2 eq) U235 occurrence) (m3)
SB eq) (MJ) eq)

laptop computer

laptop computer -
Standardised
Table 82 – Reference refurbished laptop - Results for impacts after standardisation to planetary boundaries over the
entire life cycle

element fossil H+ eq) CO2 eq) U235 (Disease (kg)
eq) nce)
Reference refurbished 5.91% 57.30% 20.09% 23.16% 82.27% 42.98% 5.96% 10.39%
laptop computer

Cleaning only, without 4.50% 53.76% 13.95% 16.25% 82.59% 38.94% 2.42% 6.30%
refurbishment FR
Cleaning only, without 5.45% 63.24% 33.65% 36.02% 78.06% 46.71% 2.89% 17.19%
refurbishment Asia
Replacement of all parts, 81.36% 127.66% 148.08% 145.15% 106.63% 135.38% 159.79% 115.83%
Replacement of all parts, 82.31% 137.14% 167.78% 164.93% 102.10% 143.15% 160.27% 126.72%
Replacement of all parts 3.42% 59.70% 21.70% 28.38% 82.94% 49.74% 2.68% 6.41%
with 2nd-hand parts,
Replacement of all parts 4.37% 69.18% 41.40% 48.16% 78.42% 57.52% 3.15% 17.30%
Reference new laptop 1.62E-03 822 0.20 35.52 66 1.55E-06 38.49 141.35
computer
Table 83 - Laptop computer comparison - Substitution approach - Results for 1 year

(kg SB (MJ) eq)
eq)
1 Residual impact 2.16E-03 6.00E+02 2.54E-01 4.47E+01 2.03E+01 1.51E-06 5.10E+01 1.82E+02
year from the 1st life
Refurbishment 9.57E-05 4.72E+02 4.06E- 8.09E+00 5.40E+01 6.92E-07 2.16E+00 1.47E+01

02
2 Residual impact 1.62E-03 4.50E+02 1.91E-01 3.35E+01 1.52E+01 1.13E-06 3.83E+01 1.37E+02
years from the 1st life

02
4 Residual impact 5.39E-04 1.50E+02 6.36E- 1.12E+01 5.08E+00 3.77E-07 1.28E+01 4.55E+01
years from the 1st life 02

02
5 Residual impact 0 0 0 0 0 0 0 0
years from the 1st life

02
New FU 1.62E-03 4.50E+02 1.91E-01 3.35E+01 1.52E+01 1.13E-06 3.83E+01 1.37E+02
Table 84 - Laptop comparison - Depreciation approach - Quantified results
Regular Reasonable Environmentally Reasonable Environmentally Systematic

purchase purchase conscious use (5 years) conscious use purchase
(every 3 years) (every 4 years) purchase (every of a new (8 years) of a (every 3 years)
of a recent of a 5 years) of an laptop new laptop of new items
refurbished refurbished older
model (2 model (3 refurbished
years) years) model (over 5
years old)
Year 0 0 0 0 0 0 0
Year 1 40 28 6 35 35 58
Year 2 79 57 12 71 71 116
Year 3 119 63 18 106 106 173

Year 4 159 65 20 142 142 231
Year 5 198 94 22 177 177 289
Year 6 238 122 28 213 179 347
Year 7 278 128 35 248 181 404
Year 8 317 130 41 284 183 462
Year 9 357 159 43 319 218 520
Year 10 396 187 45 354 254 578
Figure 85 - Comparison with reference laptop - Depreciation approach - Cumulative impacts over 10 years for
climate change
9.5.4. Desktop computer family

ADP - ADP - AP (mol H+ GWP IR (kg PM (Disease WU (m3)
elements fossil (MJ) eq) (kg U235 eq) occurrence)
(kg SB eq) CO2
eq)
Upstream logistics: 43% 1.8% 10% 11% 0.1% 6% 6%

to centralisation
Refurbishment 38% 7.9% 38% 43% 2.8% 15% 71%
Distribution from the 17% 0.5% 3% 3% 0.0% 1% 2%

Use by the user 2% 90% 50% 43% 97% 78% 20%
TOTAL 1.88E-04 1 426.92 7.82E-02 15.42 178.12 1.95E-06 2.11
Table 86 - Reference refurbished desktop computer - Results broken down by impact over the life cycle
ADP - ADP - AP (mol GWP IR (kg PM (Disease WU MIPS EW (kg)

elements fossil H+ eq) (kg U235 occurrence) (m3) (kg)
(kg SB (MJ) CO2 eq)
eq) eq)
Supply: 43.5% 18.5% 21.7% 27.5% 2.2% 32.5% 20.0% 10.2% 0.0%
centralisation to
refurbishers
Consumption of the 0.1% 33.3% 13.0% 9.6% 87.5% 18.8% 1.6% 13.5% 0.0%
refurbishment site
Checking - Reset - 0.0% 0.4% 0.7% 0.4% 0.0% 0.4% 10.4% 0.5% 0.0%
Cleaning
RAM replacement 2.9% 6.9% 9.5% 9.2% 0.4% 8.3% 9.8% 9.3% 26.9%
Disk replacement 30.1% 16.7% 25.6% 22.0% 6.4% 18.1% 30.2% 41.1% 70.3%
Replacement of 0.1% 0.2% 0.2% 0.2% 0.0% 0.3% 0.0% 0.0% 1.6%
other parts
Accessories 0.6% 0.3% 0.6% 0.3% 1.4% 0.3% 0.2% 1.3% 1.2%
Packaging 1.9% 16.5% 21.3% 22.5% 1.3% 7.9% 25.7% 23.1% 0.0%
Distribution from 20.7% 7.4% 7.4% 8.2% 0.9% 13.5% 2.1% 1.0% 0.0%
the refurbisher to
the distributor
Total 7.22E-05 114.26 2.98E-02 6.51 4.96 3.08E-07 1.43 10.14 5.82E-03

Table 87 - Reference refurbished desktop computer - Impacts broken down by stage - Focus on the refurbishment
process - Numbers

elements fossil H+ eq) CO2 eq) U235 eq) occurrence)
(kg SB eq) (MJ)
Reference 1.88E-04 1 425.13 7.81E-02 15.53 178.12 1.93E-06 2.35 1.88E-04

replacement
RAM 106% 109% 170% 163% 100% 115% 300% 197%

replacement
Disk 244% 121% 267% 246% 102% 137% 522% 372%

replacement
Miscellaneous 87% 98% 87% 87% 100% 96% 71% 81%

replacement
Cleaning Only 87% 98% 87% 87% 100% 96% 71% 81%
Table 88 – Refurbished desktop computer - Impact of replacing parts - Life cycle vision

eq) nce)
Reference scenario 1.88E-04 1,426.92 7.82E-02 15.42 178.12 1.95E-06 2.11 1.88E-04
Supply 113% 115% 220% 212% 98% 112% 116% 242%

USA/refurbishment
China
Supply 83% 99% 94% 93% 100% 96% 96% 99%

France/refurbishment
France
Table 89 – Refurbished desktop computer - Impact of the supply and refurbishment location - Life cycle vision
ADP - ADP - AP (mol GWP (kg IR (kg PM WU

elements fossil H+ CO2 U235 (Disease (m3/Europ
(kg SB (MJ/Europe eq/Europe eq/Europe eq/Europe occurrenc ean
eq/Europ an an an an e/Europea average)
ean average) average) average) average) n average)
average)

from the previous
user to centralisation
Distribution from the 4.51E-04 9.06E-05 3.35E-05 5.12E-05 6.67E-06 3.81E-05 3.94E-06
distributor to the
user
TOTAL 4.74E-03 1.56E-03 3.77E-04 4.69E-04 2.12E-03 3.23E-04 5.95E-04
Table 90 – Reference refurbished desktop computer - Results for impacts after standardisation over the entire life
cycle

elements fossil (MJ) H+ eq) CO2 eq) U235 eq) occurrence) (kg)
(kg SB eq)

(years)
2 1.88E-04 1.43E+03 7.82E-02 1.54E+01 1.78E+02 1.95E-06 2.11E+00 2.65E+01

3 1.27E-04 1.38E+03 6.50E-02 1.25E+01 1.76E+02 1.80E-06 1.56E+00 2.31E+01
4 9.57E-05 1.35E+03 5.84E-02 1.10E+01 1.76E+02 1.72E-06 1.29E+00 2.14E+01
5 7.72E-05 1.34E+03 5.45E-02 1.02E+01 1.75E+02 1.68E-06 1.13E+00 2.03E+01
Table 91 - Desktop computer - Comparison of impacts depending on the lifespan - substitution approach, all stages
ADP - ADP - AP (mol GWP (kg IR (kg PM (Disease WU (m3)
(kg SB eq)
Reference refurbished 1.88E-04 1.43E+03 7.81E-02 1.55E+01 1.78E+02 1.93E-06 2.35E+00

desktop computer
Reference refurbished 5.92E-03 4.40E-02 5.39E-04 1.58E-02 2.34E-03 2.58E-02 8.94E-05

desktop computer -
Standardised
Table 92– Reference refurbished desktop computer - Results for impacts after standardisation to planetary
boundaries for the entire life cycle

eq) nce)
Reference refurbished 1.88E-04 1 426.92 7.82E-02 15.42 178.12 1.95E-06 2.11 26.52
desktop computer
0.18 RAM
0.2 Disk
Lifespan: 3 years
Cleaning only, without 1.33E-04 1 382.79 6.30E-02 12.19 177.69 1.79E-06 1.36 21.20
refurbishment FR
Lifespan: 3 years
Cleaning only, without 1.61E-04 1 601.75 1.58E-01 29.69 174.61 2.03E-06 1.74 58.93
refurbishment Asia
Lifespan: 3 years
Replacement of all parts, 4.62E-04 1 863.13 2.69E-01 48.84 182.05 2.93E-06 17.33 129.28
Lifespan: 3 years
Replacement of all parts, 4.91E-04 2 082.09 3.64E-01 66.34 178.97 3.17E-06 17.71 167.01
Lifespan: 3 years
Replacement of all parts 1.51E-04 1 439.90 7.95E-02 17.06 177.96 2.04E-06 1.42 21.25
Lifespan: 3 years
Replacement of all parts 1.80E-04 1 658.86 1.74E-01 34.57 174.88 2.28E-06 1.80 58.99
Lifespan: 3 years
Reference new desktop 1.59E-03 2 227.06 3.06E-01 57.90 332.30 3.19E-06 32.46 285.80
computer
10GB RAM
443GB SSD
1272GB HDD
Lifespan: 5 years
Table 93 - Desktop computer comparison - Substitution approach - Results for 1 year

(kg SB eq) occurren
ce)
1 Residual 2.12E-03 1.26E+03 3.57E-01 6.83E+01 2.12E+02 2.25E-06 4.27E+01 3.59E+02

year impact from
the 1st life
Refurbishme 1.88E-04 1.43E+03 7.82E-02 1.54E+01 1.78E+02 1.95E-06 2.11E+00 2.65E+01

nt

year impact from
s the 1st life

nt

year impact from
s the 1st life

nt
5 Residual 0 0 0 0 0 0 0 0
year impact from
s the 1st life

nt
New FU 1.59E-03 9.47E+02 2.67E-01 5.12E+01 1.59E+02 1.69E-06 3.20E+01 2.70E+02
Table 94 - Desktop computer comparison - Substitution approach - Quantified results

Regular Reasonable Environmentally Reasonable Environmentally Systematic
purchase purchase aware purchase use (5 years) aware use (8 purchase
(every 3 years) (every 4 years) (every 5 years) of a new years) of a new (every 3 years)
of a recent of a of an older laptop laptop of new items
refurbished refurbished refurbished
model (2 model (3 model (over 5
years) years) years old)
Year 0 0 0 0 0 0 0
Year 1 60 43 9 58 58 92
Year 2 120 86 17 116 116 184
Year 3 180 94 26 174 174 276
Year 4 240 101 33 232 232 368
Year 5 300 144 40 289 289 460
Year 6 360 187 48 347 296 552
Year 7 420 196 57 405 303 644
Year 8 480 202 66 463 310 736
Year 9 540 245 72 521 367 828
Year 10 600 288 79 579 425 920
Table 95 - Comparison with reference desktop computer - Depreciation approach - Cumulative impacts over 10
years for climate change
9.6. Results extrapolated to other product families
N.B.: - The initial plan for the ADEME study included conducting LCAs for the box, server and console
product families. However, during the study, we were unable to gather enough primary information to
carry out a study of the same quality for the other, more mature product families. We have thus set out
below extrapolations adapted to these product families but based on the primary data collected for
laptop computers.
This extrapolation gives an idea of the orders of magnitude for these two product families.
9.6.1. Server family
Overview
Definition: a server is a piece of computer hardware or software (computer program) that provides
functionality for other programs or devices, called ‘clients’. This architecture is called the client–server
model. Servers can provide various functionalities, often called "services", such as sharing data or
resources among multiple clients, or performing computation for a client. A single server can serve
multiple clients, and a single client can use multiple servers. Client–server systems are most frequently
implemented by (and often identified with) the request–response model: a client sends a request to the
server, which performs some action and sends a response back to the client, typically with a result or
acknowledgment. Typical servers are database servers, file servers, mail servers, print servers, web servers,
game servers, and application servers.38
Reference data
In the absence of any field data, we considered that a server was refurbished in the same manner as a
desktop computer. We therefore extrapolated the desktop computer model for the server model.
However, no reference models were modelled here.
38
https://en.wikipedia.org/wiki/Server_(computing)

Refurbished server New server
Type of modelling
Sources of data Extrapolation from the desktop ADEME/ARCEP study - NegaOctet

computer model database
Refurbishers Refurbishers (gathered from 7)
Types of data Ateliers du bocage; ATF bis repetita, Secondary

ATF GAIA, Atoutek, Ingram, LM
Ecoproduction, Recyclea
Reference model Secondary Reference product representative of

business usage
Characteristics of the reference Average product based on refurbisher Serveur Rack with 2 processors, 4 SSD
model data collected of 2048 GB each, 4x16 GB RAM; 1 GPU
Parts replaced RAM, hard disks
Consumption during use phase Refurbishment location: Manufacturing location:

France (100%) Asia (100%)
Mass 18 kg 18 kg
Annual consumption (kWh) 10512 kWh 10512 kWh
Table 96 – Reference characteristics for servers
Results - Comparison of the impacts of new and refurbished servers -

Since the calculation performed for servers was an extrapolation of the calculation for desktop
computers, the general analysis of the results is very similar to the one obtained for desktop computers.
A comparative summary was therefore produced for this product family, considering:
 An average new product with a lifespan of 5 years;
 A fictitious reference refurbished product with a usage period of 3 years;
 A product refurbished with a short supply chain*, with basic cleaning;
 A product refurbished with a short supply chain*, with new replacement parts;
 A product refurbished with a short supply chain*, with second-hand replacement parts;
 A product refurbished on a globalised market**, with basic cleaning;
 A product refurbished on a globalised market**, with new replacement parts;
 A product refurbished on a globalised market**, with second-hand replacement parts;

N.B.: - replacing all parts involves replacing the hard disk and the RAM
ADP - ADP - AP (mol GWP (kg IR (kg PM WU (m3 MIPS (kg
elements fossil (MJ H+ eq CO2 eq U235 eq (Disease per year) per year)
(kg SB eq per year) per year) per year) per year) occurrence
per year) per year)
Cleaning only, without 5.73E-04 1.35E+05 4.12E+00 7.14E+02 1.82E+04 1.58E-04 5.11E+01 1.71E+03
refurbishment FR
Lifespan: 3 years
Cleaning only, without 6.89E-04 1.35E+05 4.34E+00 7.66E+02 1.82E+04 1.59E-04 5.21E+01 1.79E+03
refurbishment Asia
Lifespan: 3 years
Replacement of all parts, 3.17E-03 1.47E+05 9.21E+00 1.64E+03 1.82E+04 1.87E-04 4.67E+02 4.00E+03
with accessories, supply FR,
refurbishment FR
Lifespan: 3 years
Replacement of all parts, 3.29E-03 1.47E+05 9.42E+00 1.69E+03 1.82E+04 1.87E-04 4.68E+02 4.08E+03
with accessories, supply US,
refurbishment Asia
Lifespan: 3 years
Replacement of all parts 5.91E-04 1.35E+05 4.14E+00 7.19E+02 1.82E+04 1.59E-04 5.12E+01 1.71E+03
refurbishment FR
Lifespan: 3 years
Replacement of all parts 7.07E-04 1.35E+05 4.35E+00 7.71E+02 1.82E+04 1.59E-04 5.21E+01 1.79E+03
refurbishment Asia
Lifespan: 3 years
Reference new server 1.83E-02 1.56E+05 1.32E+01 2.34E+03 1.88E+04 2.08E-04 7.64E+02 6.15E+03
Lifespan: 5 years
Table 97 - Server comparison - Substitution approach - Results for one year
Figure 64 - Server comparison - Substitution approach - Results for one year
Despite the conservative hypotheses for new devices, we observed that refurbished servers have a lower
impact than the purchase of new devices, regardless of the refurbishment scenario. The variation in
impacts between the refurbishment scenarios will never cancel out the beneficial effect of refurbishment
practices.

9.6.2. Game console families
Overview
Definition: according to the SRI39, cited in the report on ICT, “A games consoles is a computing device
whose main function is to play video games. Games consoles share numerous functions and components
of material architecture with general personal computers (such as central processing units, system
memory, video architecture, optical readers and/or hard disks or other forms of internal memory).’
Reference data
In the absence of any field data, we considered that a home games console was refurbished in the same
manner as a desktop computer. We therefore extrapolated the desktop computer model for the console
model. However, no reference models were modelled here.
Refurbished console New console
Type of modelling
Sources of data Extrapolation from the desktop ADEME/ARCEP study - NegaOctet

computer model database
Refurbishers Refurbishers (collected from 7)
Types of data Ateliers du bocage; ATF bis repetita, Secondary

ATF GAIA, Atoutek, Ingram, LM
Ecoproduction, Recyclea
Reference model Secondary Average product based on the French

market
Characteristics of the reference Average product based on refurbisher See table below
model data collected
Consumption during use phase Refurbishment location: Manufacturing location:

France (100%) Asia (100%)
Parts changed Controllers + Hard disks
Annual consumption (kWh) 55.88 kWh 55.88 kWh
Table 98 – Reference characteristics for consoles
39
In the European Union, video games consoles are governed by a Self-Regulatory Initiative (SRI) under the eco-design
directive (ENTR lot 3). The signatories are the three main manufacturers: Microsoft (Xbox), Sony (PlayStation) and
Nintendo. The most recent version is SRI 2.6.3 (2018) and the latest conformity report from an independent
inspector (Intertek) was published in October 2019 (Intertek, Independent Inspector Annual Compliance Report –
Games Consoles Self-Regulatory Initiative, Reporting Period 2018, Oct. 2019). All information on the SRI is available
from the dedicated website: www.efficientgaming.eu.

PRIVATE USE Cat. 1 Cat. 2 Cat. 3 Cat. 4 Average
Distribution 79.50% 3.30% 2.10% 15.10% 100%
Weight of device (kg) 2.8 4.78 2.9 3.5 2.97
Weight of packaging (kg) 0.56 0.956 0.58 0.7 0.59
Type of processor AMD Jaguar AMD Zen 2 AMD Jaguar AMD Jaguar
RAM (GB) 8 16 8 8 8.3
SSD storage (GB) 0 825 0 0 27
HDD storage (GB) 500 0 500 500 483.5
Motherboard surface area (cm²) 838.75 904.72 686
Type of graphics card AMD Radeon AMD RDNA 2 AMD Radeon AMD Radeon
Reference product example Playstation 4 Playstation 5 Xbox One S Xbox One
Table 99 – Reference characteristics for new consoles
Results - Comparison of the impacts of new and refurbished consoles -

Since the calculation performed for consoles was an extrapolation of the calculation for desktop
computers, the general analysis of the results is very similar to the one obtained for desktop computers.
A comparative summary was therefore produced for this product family, considering:
 An average new product with a lifespan of 5 years;
 A fictitious reference refurbished product with a usage period of 3 years;
 A product refurbished with a short supply chain*, with basic cleaning;
 A product refurbished with a short supply chain*, with new replacement parts;
 A product refurbished with a short supply chain*, with second-hand replacement parts;
 A product refurbished on a globalised market**, with basic cleaning;
 A product refurbished on a globalised market**, with new replacement parts;
 A product refurbished on a globalised market**, with second-hand replacement parts;
N.B.: - replacing all parts involves replacing the hard disk and the controllers

ADP - ADP - AP (mol GWP (kg IR (kg PM (Disease WU (m3 MIPS (kg
elements fossil (MJ H+ eq CO2 eq U235 eq occurrence per year) per year)
(kg SB eq per year) per year) per year) per year) per year)
per year)
refurbishment FR
Lifespan: 3 years
refurbishment Asia
Lifespan: 3 years
with accessories, supply FR,
refurbishment FR
Lifespan: 3 years
with accessories, supply US,
refurbishment Asia
Lifespan: 3 years
Replacement of all parts with 1.29E-04 8.62E+02 5.93E-02 1.35E+01 1.02E+02 1.30E-06 1.40E+00 1.40E+01
refurbishment FR
Lifespan: 3 years
Replacement of all parts with 1.57E-04 1.02E+03 1.29E-01 2.63E+01 9.82E+01 1.53E-06 1.74E+00 4.06E+01
refurbishment Asia
Lifespan: 3 years
Reference new console 4.86E-03 1.47E+03 3.19E-01 5.85E+01 1.93E+02 2.68E-06 2.08E+01 3.12E+02
Lifespan: 5 years
Table 100 - Console comparison - Substitution approach - Results for one year
Figure 65 - Console comparison - Substitution approach - Results for one year
Despite the conservative hypotheses for new devices, we observed that refurbished consoles have a lower
impact than the purchase of new devices, regardless of the refurbishment scenario. The variation in
impacts between the refurbishment scenarios will never cancel out the beneficial effect of refurbishment
practices.

10. Index of tables and figures
TABLES
Table 1 – Lifespans identified in literature ................................................................................................ 34
Table 2 – Lifespans considered as part of the studies ............................................................................. 34
Table 3 – Description of the refurbishment synopsis considered for each product family ................ 36
Table 4 – Collection process per device family ........................................................................................ 37
Table 5 – Transport distance matrix .......................................................................................................... 38
Table 6 - Reference scenario - means of transport for acquiring the equipment. ................................ 38
Table 7 - Information on mixes .................................................................................................................. 40
Table 8 - Description of the DQR method applied during the study ..................................................... 41
Table 9 - Rates of recycling and incineration for packaging waste ......................................................... 46
Table 10 - Selection of appropriate indicators based on standardisation and weighting .................... 47
Table 11 – Description of indicators .......................................................................................................... 48
Table 12 – Standardisation factors proposed by JRC ............................................................................... 49
Table 13 – Factors for planetary boundaries ............................................................................................. 50
Table 14 - Presentation of the characteristics of reference smartphones ............................................. 51
Table 15 – Reference refurbished smartphone - Presentation of refurbisher profiles - Summary ...... 53
Table 16 – Refurbished smartphone - Physical characteristics of the various elements considered for
the reference model .......................................................................................................................... 53
Table 17 – Refurbished smartphone - Entry data for modelling the reference model scenario .......... 55
Table 18 - Characteristics of the new smartphone considered .............................................................. 56
Table 19 – Reference smartphone - Environmental impacts per functional unit for 1 year ................. 58
Table 20 – Smartphone - Environmental impacts per functional unit for 1 year - Variations between
companies ........................................................................................................................................... 63
Table 21 - Refurbished smartphone - Sensitivity analysis per screen type - For replacement in the
reference model ................................................................................................................................. 67
Table 22 - Refurbished smartphone - Sensitivity analysis for screen type - For full screen
replacement ........................................................................................................................................ 67
Table 23 - Refurbished smartphone - Sensitivity analysis for battery mass ........................................... 68
Table 24 - Refurbished smartphone - Sensitivity analysis - Upstream means of transport .................. 68
Table 25 - Refurbished smartphone - Sensitivity analysis - Means of distribution ................................ 69
Table 26 – Reference refurbished smartphone - Results for impacts after standardisation to
planetary boundaries for the entire life cycle ................................................................................ 71
Table 27- Reference smartphone - Environmental impact for the entire life cycle for the reference
lifespans (for the entire life cycle) .................................................................................................... 72
Table 28 - Desktop computer comparison - Substitution approach - Impacts avoided for the
reference scenarios ............................................................................................................................ 72
Table 29 – Reference characteristics for tablets ...................................................................................... 78
Table 30 – Characteristics for new tablets considered ........................................................................... 79
Table 31 – Reference tablet - Environmental impacts per functional unit for 1 year ........................... 79
reference scenarios ............................................................................................................................ 80
Table 33 - Presentation of the characteristics of reference laptops ...................................................... 83
Table 34 – Reference refurbished laptops - Presentation of refurbisher profiles - Summary .............. 84
Table 35 – Refurbished laptop - Physical characteristics of the various elements considered for the
reference model ................................................................................................................................. 84
Table 36 - Data collected for refurbished laptop computers ................................................................. 86
Table 37 – Reference characteristics for new computers ....................................................................... 87
Table 38 – Reference laptop - Environmental impacts per functional unit for 1 year .......................... 89
Table 39 – Laptop computer - Environmental impacts per functional unit for 1 year - Variations
between companies ........................................................................................................................... 96

Table 40 - Refurbished laptop computer - Sensitivity analysis per type of screen - For replacement in
the reference model .......................................................................................................................... 98
Table 41 - Refurbished laptop computer - Sensitivity analysis per type of screen - For full screen
replacement ........................................................................................................................................ 99
Table 42 - Refurbished laptop computer - Sensitivity analysis per RAM capacity ................................ 99
Table 43 - Refurbished laptop computer - Sensitivity analysis per RAM capacity (full RAM
replacement) ....................................................................................................................................... 99
Table 44 - Refurbished laptop computer - Sensitivity analysis per hard disk capacity ...................... 100
Table 45 - Refurbished laptop computer - Sensitivity analysis per hard disk capacity (full hard disk
replacement) ..................................................................................................................................... 100
Table 46 - Refurbished laptop computer - Sensitivity analysis - Means of transport ........................... 101
Table 47 - Refurbished laptop computer - Sensitivity analysis - Means of distribution ....................... 101
Table 48- Reference laptop - Environmental impact for the entire life cycle for the reference
lifespans (for the entire life cycle) - Personal use .......................................................................... 103
reference scenarios .......................................................................................................................... 104
Table 50 - Presentation of the characteristics of reference desktop computers ............................... 109
Table 51 – Reference refurbished desktop computers - Presentation of refurbisher profiles -
Summary ............................................................................................................................................ 111
Table 52 - Desktop computer - Data collected associated with refurbishment .................................. 112
Table 53 - Desktop computer - Description of the reference model.................................................... 112
Table 54 – Reference desktop computer - Environmental impacts per functional unit for 1 year .... 115
Table 55 - Refurbished desktop computer - Sensitivity analysis per RAM capacity............................ 123
Table 56 - Refurbished desktop computer - Sensitivity analysis per RAM capacity (full RAM
replacement) ..................................................................................................................................... 124
Table 57 - Refurbished desktop computer - Sensitivity analysis per hard disk capacity .................... 124
Table 58 - Refurbished desktop computer - Sensitivity analysis per hard disk capacity (full hard disk
replacement) ..................................................................................................................................... 125
Table 59 - Refurbished desktop computer - Sensitivity analysis - Means of distribution towards/by
the private individual ....................................................................................................................... 125
Table 60- Reference desktop computer - Environmental impact for the entire life cycle for the
reference lifespans (for the entire life cycle) ................................................................................. 127
reference scenarios .......................................................................................................................... 128
Table 62–Impact over the entire lifespan of a reference refurbished and new smartphone ............ 155
Table 63 - Reference refurbished smartphone - Results broken down by life cycle impact .............. 156
Figure 64 - Reference refurbished smartphone - Impacts broken down by stage - Focus on the
refurbishment process - Numbers .................................................................................................. 156
Figure 65 – Refurbished smartphone - Impact of replacing parts - Life cycle vision ........................... 156
Table 66 – Refurbished smartphone - Impact of the supply and refurbishment location - Life cycle
vision .................................................................................................................................................. 156
Table 67 – Reference refurbished smartphone - Results for impacts after standardisation (eq. world
inhabitant) for the entire life cycle ................................................................................................. 157
Table 68 - Smartphone - Comparison of impacts depending on the lifespan of the smartphone -
substitution approach, all steps ...................................................................................................... 157
Table 69 - Smartphone comparison - Substitution approach - Results ................................................ 157
Table 70 - Smartphone comparison - Substitution approach - Quantified results ............................. 158
Table 71 - Comparison with reference smartphone - Depreciation approach - Cumulative results
over 6 years for climate change ...................................................................................................... 158
Table 72- Reference tablet - Environmental impact for the entire life cycle for the reference
lifespans (for the entire life cycle) .................................................................................................. 158
Table 73 - Tablet comparison - Substitution approach - Results for one year .................................... 159
Table 74- Reference laptop - Environmental impact for the entire life cycle for the reference
lifespans (for the entire life cycle) - Personal use .......................................................................... 159
Table 75 - Reference refurbished laptop - Results broken down by life cycle impact ....................... 160

Table 76 - Reference refurbished laptop computer - Impacts broken down by stage - Focus on the
Table 77 – Refurbished laptop - Impact of replacing parts - Life cycle vision ...................................... 161
Table 78 – Refurbished laptop computer - Impact of the supply and refurbishment location -
Refurbisher vision .............................................................................................................................. 161
Table 79 – Refurbished laptop computer - Impact of the supply and refurbishment location - Life
cycle vision ......................................................................................................................................... 161
Table 80 – Reference refurbished laptop - Results for impacts after standardisation for the entire life
cycle ................................................................................................................................................... 161
Table 81 - Laptop - Comparison of impacts depending on the lifespan - substitution approach, all
stages .................................................................................................................................................. 161
Table 82 – Reference refurbished laptop - Results for impacts after standardisation at planetary
boundaries for the entire life cycle ............................................................................................... 162
Table 83 - Laptop computer comparison - Substitution approach - Results for 1 year ...................... 162
Table 84 - Laptop comparison - Approach by substitution - Quantified results ................................. 162
Figure 85 - Comparison with reference laptop - Depreciation approach - Cumulative impacts over 10
years for climate change.................................................................................................................. 163
Table 86 - Reference refurbished desktop computer - Results broken down by life cycle impact ... 163
Table 87 - Reference refurbished desktop computer - Impacts broken down by stage - Focus on the
Table 88 – Refurbished desktop computer - Impact of replacing parts - Life cycle vision ................ 164
Table 89 – Refurbished desktop computer - Impact of the supply and refurbishment location - Life
cycle vision ........................................................................................................................................ 164
Table 90 – Reference refurbished desktop computer - Results for impacts after standardisation for
the entire life cycle .......................................................................................................................... 165
Table 91 - Desktop computer - Comparison of impacts depending on the lifespan - substitution
approach, all stages .......................................................................................................................... 165
Table 92– Reference refurbished desktop computer - Results for impacts after standardisation to
planetary boundaries for the entire life cycle ............................................................................... 165
Table 93 - Desktop computer comparison - Substitution approach - Results for 1 year .................... 166
Table 94 - Desktop computer comparison - Substitution approach - Quantified results .................. 166
Table 95 - Comparison with reference desktop computer - Depreciation approach - Cumulative
impacts over 10 years for climate change ...................................................................................... 167
Table 96 – Reference characteristics for servers .................................................................................... 168
Table 97 - Server comparison - Substitution approach - Results for one year .................................... 169
Table 98 – Reference characteristics for consoles ................................................................................. 170
Table 99 – Reference characteristics for new consoles .......................................................................... 171
Table 100 - Console comparison - Substitution approach - Results for one year................................ 172
FIGURES
Figure 1 – Composition of smartphones and computers - The hidden side of the digital sector -
ADEME 2021 ........................................................................................................................................ 14
Figure 2 – Product families and versions studied ...................................................................................... 15
Figure 3 - Basics of life cycle analysis ......................................................................................................... 16
Figure 4 – Roles and responsibilities of the stakeholders in the study ................................................... 22
Figure 5 – Scope of the study for refurbished products .......................................................................... 25
Figure 6 – Scope of the study for new products ...................................................................................... 26
Figure 7 – Comparative scope between new and refurbished products under a substitution
approach (example used: smartphones) .......................................................................................... 29
Figure 8 – Taking account of impacts as part of a depreciation approach ........................................... 31
Figure 9 – Data collected during the modelling of the refurbishment process ..................................... 35
Figure 10 – Choice of the allocation factor ............................................................................................... 42
Figure 11 – Creating refurbishment variants .............................................................................................. 43
Figure 12 – Presentation of the variable elements in the refurbishment process ................................. 56

Figure 13 – Reminder of the parameters for analysis and correspondence with the data collected .. 57
Figure 14 - Reference refurbished smartphone - Results broken down by life cycle impact ............... 59
Figure 15 - Reference refurbished smartphone - Impacts broken down by stage - Focus on the
refurbishment process ....................................................................................................................... 60
Figure 16 – Refurbished smartphone - Impact of replacing parts - Life cycle vision ............................. 61
Figure 17 – Refurbished smartphone - Impact of the supply and refurbishment location - Life cycle
vision .................................................................................................................................................... 62
Figure 18 - Refurbished smartphone - Variation of impacts depending on the companies - Impacts
on climate change .............................................................................................................................. 63
on the depletion of abiotic mineral resources ................................................................................ 64
on the depletion of abiotic fossil resources .................................................................................... 64
Figure 21 - Refurbished smartphone - Variation of impacts depending on the lifespan of the
refurbished smartphone .................................................................................................................... 67
Figure 22 – Reference refurbished smartphone - Results for impacts after standardisation (eq. world
inhabitant) for the entire life cycle ................................................................................................... 70
Figure 23 – Reference refurbished smartphone - Results for impacts after standardisation to
planetary boundaries for the entire life cycle ................................................................................ 71
Figure 24 - Smartphone comparison - Substitution approach - Results ................................................. 74
Figure 25 - Smartphone comparison - Depreciation approach - Variation in environmental impact
by duration of first life. ...................................................................................................................... 75
Figure 26 - Smartphone comparison - Depreciation approach - Variation by duration of first life and
by type of refurbishment................................................................................................................... 76
Figure 27 - Comparison with reference smartphone - Depreciation approach - Market results over 6
years for climate change.................................................................................................................... 77
Figure 28 - Tablet comparison - Substitution approach - Results for one year ..................................... 80
Figure 29 – Presentation of the variable elements in the refurbishment process ................................. 87
Figure 30 – Reminder of the parameters for analysis and correspondence with the data collected . 88
Figure 31 - Reference refurbished laptop - Results broken down by life cycle impact ......................... 90
Figure 32 - Reference refurbished laptop computer - Impacts broken down by stage - Focus on the
refurbishment process ....................................................................................................................... 91
Figure 33 – Refurbished laptop - Impact of replacing parts - Life cycle vision ...................................... 92
Figure 34 - Refurbished laptop computer - Variation of impacts depending on the companies -
Impacts on climate change ............................................................................................................... 93
Impacts on the depletion of abiotic mineral resources ................................................................. 94
Impacts on the depletion of abiotic fossil resources...................................................................... 94
Impacts on ionising radiation ............................................................................................................ 95
Figure 38 – Refurbished laptop computer - Impact of the supply and refurbishment location - Life
cycle vision .......................................................................................................................................... 97
Figure 39 - Refurbished laptop computer - Variation of impacts depending on the lifespan of the
refurbished laptop ............................................................................................................................. 98
Figure 40 – Reference refurbished laptop - Results for impacts after standardisation for the entire
life cycle ........................................................................................................................................... 102
Figure 41 – Reference refurbished laptop - Results for impacts after standardisation to planetary
boundaries for the entire life cycle ............................................................................................... 103
Figure 42 - Laptop comparison - Substitution approach - Results for 1 year (100% = impact of new
reference device) .............................................................................................................................. 105
Figure 43 - Laptop comparison - Depreciation approach - Variation in environmental impact by
duration of first life. ......................................................................................................................... 106
Figure 44 - Laptop comparison - Depreciation approach - Variation in environmental impact by
duration of first life and type of refurbishment. ........................................................................... 107
Figure 45 - Comparison with reference laptop - Depreciation approach - Market results over 10 years
for climate change ........................................................................................................................... 108

Figure 46 – Presentation of the variable elements in the refurbishment process ................................ 113
Figure 47 – Reminder of the parameters for analysis and correspondence with the data collected 114
Figure 48 - Reference refurbished desktop computer - Results broken down by impact on the life
cycle .................................................................................................................................................... 116
Figure 49 - Reference refurbished desktop computer - Impacts broken down by stage - Focus on the
refurbishment process ...................................................................................................................... 117
Figure 50 – Refurbished desktop computer - Impact of replacing parts - Life cycle vision ................ 118
Figure 51 - Refurbished desktop computer - Variation of impacts depending on the companies -
Impacts on climate change .............................................................................................................. 119
Impacts on the depletion of abiotic mineral resources ................................................................ 119
Impacts on the depletion of abiotic fossil resources.................................................................... 120
Impacts on ionising radiation .......................................................................................................... 120
Figure 55 – Refurbished desktop computer - Impact of the supply and refurbishment location - Life
cycle vision ........................................................................................................................................ 122
Figure 56 - Desktop computer - Variation of impacts depending on the lifespan of the refurbished
desktop computer ............................................................................................................................ 123
Figure 57 – Reference refurbished desktop computer - Results for impacts after standardisation for
the entire life cycle .......................................................................................................................... 126
Figure 58– Reference refurbished desktop computer - Results for impacts after standardisation to
planetary boundaries for the entire life cycle ............................................................................... 127
Table 59 - Desktop computer comparison - Substitution approach - Results for 1 year .................... 129
Figure 60 - Desktop computer comparison - Depreciation approach - Variation in environmental
impact by duration of first life. ....................................................................................................... 130
Figure 61 - Desktop computer comparison - Depreciation approach - Variation in environmental
impact by duration of first life and type of refurbishment. .......................................................... 131
Figure 62 - Comparison with reference desktop computer - Depreciation approach - Market results
over 10 years for climate change .................................................................................................... 132
Figure 63 - Server comparison - Substitution approach - Results for one year ................................... 169
Figure 64 - Console comparison - Substitution approach - Results for one year ................................ 172

12. Abbreviations and acronyms
LCA Life Cycle Analysis

Agence de l’Environnement et de la Maîtrise de l’Énergie (French Agency for Ecological
ADEME
Transition)
WEEE Waste from Electrical and Electronic Equipment
DMS Desktop Managed Services
EN Standards from the European Committee for Standardization
EPEAT Electronic Product Environmental Assessment Tool
ErP Energy-related Products
ETSI European Telecommunications Standards Institute
EuP Energy-using Product
EPIC Public Industrial and Commercial Establishment
ES ETSI Standard
MSB Mid-Sized Business
ISP Internet Service Provider
GG Greenhouse Gas
GeSI Global-enabling Sustainability Initiative
GHG GreenHouse Gas
HDD Hard Disk Drive
HPC High-Performance Computing
HW Hardware
ICT Information and Communication Technology
EE Environmental Engineering
IEC International Electrotechnical Commission
ILCD International Reference Life Cycle Data system
iNEMI International Electronics Manufacturing Initiative
IoT Internet of Things
ISO International Organization for Standardization
IT Information Technology
ITU International Telecommunication Union
KPI Key Performance Indicator
LCA Life Cycle Analysis
LCIA Life Cycle Impact Assessment
MFA Material Flow Analysis
MIPS Material Input per Service Unit
PCR Product Category Rule
PEF Product Environmental Footprint
PEFCR Product Environmental Footprint Category Rule
SME Small and Medium-sized Enterprise
GWP Global Warming Potential
PUE Power Usage Effectiveness
REACH Registration, Evaluation, Authorisation and Restriction of Chemicals
REF Renewable Energy Factor
RoHs Restriction of Hazardous substances in electrical and electronic equipment
CSR Corporate Social Responsibility
SDIS Departmental Fire and Rescue Service
SSD Solid State Drive
SW Software

TIA Telecommunications Industry Association
ICT Information and Communication Technologies
TNS Telecommunications Network Services
TS Technical Specification
FU Functional unit

THE COLLECTIONS OF
ADEME
FACTS AND FIGURES

ADEME as a specialist: Provides
objective analyses based on
ADEME IN BRIEF quantified indicators that are
At ADEME - the French Agency for Ecological updated regularly.
Transition - we are firmly committed to fighting global
warming and the depletion of resources. KEYS TO ACTION
ADEME as a facilitator: Compiles
We mobilise citizens, economic stakeholders and practical guides to help
regions on all fronts, and give them the tools to move stakeholders complete their
forward and become a society that is resource- projects in a methodical manner
efficient, more reserved with carbon, fairer and more and/or in compliance with
harmonious. regulations.
In every field – energy, air, circular economy, food, SUCCESS STORIES

waste, soil, etc. – we advise on, facilitate and help ADEME as a catalyst: Stakeholders
finance numerous projects from research to solution- speak of their experiences and
share their know-how.
sharing.
At every level, we devote our abilities, expertise and EXPERTISE

ADEME as an expert: Reports on
forward planning to serving public policy.
the results of research, studies and
collective achievements carried out
ADEME is a public establishment under the authority under its supervision.
of the French Ministry for Ecological Transition and
Solidarity and the Ministry of Higher Education,
HORIZONS
Research and Innovation. ADEME facing the future: Proposes
a forward-facing and realistic vision
of the challenges posed by the
energy and ecological transition,
for a desirable future that we can
build together.

ASSESSMENT OF THE
ENVIRONMENTAL IMPACT OF A
SET OF REFURBISHED PRODUCTS
ADEME commissioned a study to establish the environmental
reports of refurbished digital devices compared with their new
equivalents, over their entire life cycles.
This shows that purchasing a refurbished device instead of a

new one can prevent the following (excluding ionising
radiation):
- Smartphones: 87-64% of the annual impact

- Tablets: 46-80% of the annual impact
- Laptop computers: 43-97% of the annual impact Refurbishment can prolong the
- Desktop computers: 43-97% of the annual impact duration of use for products and
thus reduce their environmental
Within the industry, significant variations in impact can be
observed, from a factor of 2 up to 11, owing to differences in impacts.
practices concerning: This study assesses and analyses
the environmental impacts caused
- extending the lifespan by the refurbishment of products
- adding new accessories and assesses the environmental
- whether or not parts are systematically replaced benefits associated with using
- the use of second-hand parts refurbished products as against
- the volume of the packaging and the materials constituting it new products.
- the supply market (France, Europe, Asia, United Arab
Emirates, US)
- the refurbishment location
The lifespan of these devices is a major contributor to their

impacts. Ideally, refurbishment should take place at the end
of the device’s first lifespan (smartphones/tablets: 3 years. PC:
4/5 years), and the second user should keep the device for as
long as possible.
In order to maximise the extent to which refurbished devices

prevent impacts, several best practices for users, refurbishers,
platforms and legislators are proposed in this study.
www.ademe.fr

Environmental Impact of Refurbished Devices

Uploaded by

Copyright:

Available Formats

Environmental Impact of Refurbished Devices

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Environmental Impact of Refurbished Devices

Uploaded by

Copyright:

Available Formats

ASSESSMENT OF THE

Assessment of the environmental impact of a set of refurbished products | 1 |

Emma Carre (Recommerce)

Marie Castelli (Back Market)

Sophie Chavignon (Itancia)

Lucylle Dennequin (Dupli)

Rachel Dethier (Sirrmiet)

Jean Christophe Estoudre (Sofi Groupe)

Erwann Fangeat (ADEME)

Frédéric Gandon (Largo)

Jean-Lionel Laccoureye (Bak2)

Etienne Lees Perasso (Tide Environnement)

Michel Maggi (Sofigroupe)

Camille Richard (Back Market)

Regis Robin (Itancia)

François Perabo (Sofi Groupe)

Veronique Vallejo (Le GSM)

Benoit Varin (Recommerce/Rcube)

Hugo Verger (Dupli)

Nicolas Flipo (Recyclea)

Nathalie Isach (Lm Eco Production)

Didier Moiroud (Ingram Micro Services)

Garik Maureemootoo (Emmaus Connect)

Camille Bardou (Emmaus Connect)

Sarah Maisonneuve (Ateliers Du Bocage)

Miryam Nachat (Atoutek)

Alban Regnier (Okamac)

Thomas Bardou (ATF)

Valérie Brouard (ATF)

Assessment of the environmental impact of a set of refurbished products | 2|

This report is available online at https://librairie.ademe.fr/

This document is distributed by ADEME

Contract number: 2021MA000057

Department: DECD / SER

Assessment of the environmental impact of a set of refurbished products | 3|

Selection, classification and characterisation of impacts ...............................................18

1.4. Study procedure ....................................................................................................................................... 19

2.2. Managing the comparison between new and refurbished ................................................................ 26

Electrical and electronic equipment and subassembly .................................................. 43

4.4. Results ....................................................................................................................................................... 55

Assessment of the environmental impact of a set of refurbished products | 4|

Presentation of results for the reference scenarios..........................................................71

4.5. Extrapolation to tablets.......................................................................................................................... 76

Profile of refurbishers ........................................................................................................... 82

Default characteristics ......................................................................................................... 85

Total results at a functional unit level (1 year of use) ...................................................... 88

Standardised results – world inhabitant approach ....................................................... 100

Comparison of the impacts of new and refurbished laptops - Substitution approach

5.2.1. ............................................................................ Overview

Assessment of the environmental impact of a set of refurbished products | 5|

Default characteristics ........................................................................................................ 111

Total results at a functional unit level (1 year of use) ..................................................... 114

Standardised results per world inhabitant...................................................................... 124

Comparison of the impacts of new and refurbished laptops - Substitution approach

6. LIMITS ................................................................................................................................................ 131

BIBLIOGRAPHICAL REFERENCES .............................................................................................................. 171

Assessment of the environmental impact of a set of refurbished products | 6|

Assessment of the environmental impact of a set of refurbished products | 7|

To do this, we implemented a 2-stage approach:

 Creating reference models based on data collected from refurbishers