household weee generation estimate in sao jose dos
HOUSEHOLD WEEE GENERATION ESTIMATE IN SAO JOSE DOS CAMPOS, BRAZIL R.G. SOUZA*, M.N.M. ABBONDANZA* AND C.F. PALANCA* * São Paulo State University (Unesp), Institute of Science and Technology, Department of Environmental Engineering - Rodovia Presidente Dutra, Km 137,8, 12247-004, São José dos Campos, Brazil
SUMMARY: WEEE generation estimates in cities are relevant for planning adequate take-back and treatment schemes, but face the challenge of lack of trustable local data. Population surveys are a recommended technique to support data collection for WEEE estimating models. The objective of this study is to determine the EEE acquisition, WEEE lifespan and disposal profiles in the city of Sao Jose dos Campos, Brazil, in order to allow applying the Market Supply WEEE estimating model. A stratified sample with error margin of 5% was calculated and interviewed in the city. The lifespan probabilistic profiles for different WEEE appliances indicate variations that represent the social and economic aspects of each city Zone. The amounts of WEEE per year from current products in use were estimated, but a quantification of WEEE generation from further products to be acquired requires marketing projections. The disposal behaviour of population in the city shows that most common practices are the destination to selective and special collection schemes, but also the disposal on walkways. Regarding WEEE (second-hand) donation, the main receivers are acquitances and waste pickers.
1. INTRODUCTION Waste Electrical and Electronic Equipment (WEEE) generation in Brazil and worldwide is a critical issue in environmental, social and economic aspects. It is the fastest growing waste stream (Lundgren, 2012). It is estimated that Brazil produced 7 kg WEEE/capita in 2014, a higher rate than in countries like China or India (StEP, 2016). Such estimate can be explained by factors like the Purchasing Power of Population, what can also provoke variations in WEEE generation within different regions of the country. The computers market in Brazil had a large growth in the last 10 years; in 2011, the percentage of households with computers reached 43%, double the proportion in 2006 (IBGE, 2011). Electrical and Electronic Equipment (EEE) enter the Brazilian market not just through legal production and retailing chains, but also with significant contribution of the informal market. In 2014, unnofficial markets accounted for 1.5 million computers sold, 15% of the total in that year, and 35% considering only desktops (ABINEE, 2013). After the lifespan of these appliances, the large amount of WEEE produced, when not adequately treated, has high toxicity and contamination risks. There are more than 1,000 hazardous substances associated with WEEE, whilst the most commonly reported are dozens of toxical metals (eg. Ba, Be, Cd, Co, Cr, Cu, Fe, Pb, Li, La, Hg, Mn, Mo, Ni, Ag) and Persitant Organic Pollutants (POPs) like Brominated Flame Retardants (BFRs), Policyclic Aromatic Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Hidrocarbons (PAHs) and Polyvinyl Chloride (PVC) (Kidee et al., 2013). In Brazil, critical factors for the exposure to such WEEE agents are: • Low coverage of selective collection: 15% of Brazilian population attended in 2016 (CEMPRE, 2016); • Inadequate waste disposal: 60% cities or 42% MSW in Brazil (ABRELPE, 2014); • Large-scale activity of informal waste pickers: estimated 70.5k waste pickers in urban areas in 2008, of which 18% in the State of Sao Paulo, and 5.5 children under 14 y.o.; 27% cities acknowledged waste pickers at the final waste disposal areas (IPEA, 2012). On the other hand, WEEE has a high economic potential by recovery of valuable materials such as Gold, Paladium, Silver and Berilium. Because of this, in developing countries like Brazil, informal sector has widely applied rough recycling techniques such as acid leaching of PCBs and open burning of cables, what poses significant risks to the health of these informal workers and surrounding communities (Robinson, 2009; Lundgren, 2012; Kidee et al., 2013). Nevertheless, adequate WEEE recycling in Brazil is estimated in only 2% (Bandini, 2009 apud Araujo, 2012). WEEE accounts for less than 1% of waste collected by selective collection schemes (CEMPRE, 2016). Another common route for Brazilian WEEE is to be landfilled mixed with MSW. In the city of Rio de Janeiro in 2012, 3.7k tonnes WEEE were destined to the urban landfill (COMLURB, 2013). Considering both the hazardousness and economic value potentials for WEEE, the Brazilian National Solid Waste Policy (Brasil, 2010) established as mandatory the implementation of WEEE take-back schemes under the shared responsibility of EEE producers, importers, distributors and retailers, as well as the Ministry of Environment (MMA). In practice this is not implemented due to lack of agreement among these parties on some critical points. A very important information for the planning of efficient WEEE take-back schemes is the amount and types of WEEE generated within a particular area, especially the municipalities. Most actual studies focused WEEE estimates in the country level, some concerned with specific types of appliances (ex.: LCD – Lee & Cooper, 2008; mobile phones – Polak & Drapalova, 2012), whilst others only estimate a total WEEE amount in the countries (e.g. StEP, 2015). Estimates in the country level do not provide resourceful information for the planning of takeback schemes in municipalities and regions, because they do not detail the different social and economic contexts. Estimating total amounts of WEEE also do not provide useful information for planning because the variety of appliances require different collection and treatment strategies. In Brazil, few studies have engaged in estimating WEEE generations. Araujo et al. (2012) estimated the generation of some kinds of WEEE in the country level, and a more general estimation was carried out by ABDI (2013). Rodrigues et al. (2015) estimated the WEEE generation in the city of São Paulo, by applying a survey to a sample population. All these studies used estimation methods based on the volume of “Products-on-Market” (POM), i.e. the number of each type of appliance purchased in the country every year; and an (discrete) average lifespan of such appliances, in all cases, obtained from international literature (Table 1), therefore not reflecting the actual lifespan profiles in each region of the country. Table 1. Some WEEE Lifespan averages (in years) adopted in Brazilian studies WEEE Araujo et al. (2012) ABDI (2013) Rodrigues et al. (2015) Refrigerator 12 15 11 Mobile phone 4 3 4 Washing machine 10 11 11
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
The objective of this study is to determine the EEE consumption, WEEE lifespan and disposal profiles in the city of Sao Jose dos Campos, Brazil, ir order to support planning of an adequate take-back scheme. Specific objectives are: • To produce regional data on WEEE lifespan and disposal profiles; • To determine lifespan and disposal profiles reflecting the different social and economic contexts in each Zone of the city; and • To obtain lifespan probabilistic distributions for the object of study. • 2. REVIEW ON WEEE ESTIMATES METHODS AND STUDIES 2.1 WEEE estimation methods Some WEEE estimation studies adopted correlation methods that associate the amount generated with the GDP per capita or the Purchasing Power Parity (PPP) of population (e.g. Huisman et al.., 2008; StEP, 2015). On one hand, these studies require the previous availability of data regarding WEEE generation in order to find a regression equation that allows for further projections, and on the other hand, the correlation coefficient obtained may not be significative enough to allow trustable estimates (e.g. in Huisman, 2008 the coefficient R for WEEE/head x GDP/head was around 0.55 with data from some EU Member States). Most WEEE estimating studies apply mass balance methods, which rely on data such as POM, current stocks (equipment in use) in year t, and lifespan profiles (Figure 1). WEEE estimates using mass balance methods are recommended by UNEP (2007). There are several available mass balance methods for WEEE estimates (Table 2), each adopting some of the data indicated in Figure 1. The choice of the most adequate method to be used depends at large extent on the available data. Some of them vary slightly depending on the adopted variable; for example, the Carnegie Mellon method adopts a discrete average lifespan value, rather than a lifespan distribution as in the Market Supply method.
Figure 1. Multiple variables and data points for enhanced e-waste estimation methods (Source: Wang et al., 2013). Mass balance models require trustable data. Data quality is a critical issue in WEEE estimates using and of these models. Particularly regarding the Lifespan variable, data are
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
usually obtained from previous literature, in many cases reflecting lifespan profiles from different countries and in previous decades (e.g. Araujo et al., 2012; ADBI, 2013; Rodrigues et al., 2015). In this sense, surveys with population sampes can be useful to determine not only local lifespan profiles, but stocks composition and average weights of appliances (Wang et al., 2013). The use of surveys has been a methodological resource in recent studies. Kim et al. (2013) applied a survey to 1,000 persons to determine the lifespan of EEE, supported by the Weibull distribution. The same study has also collected data regarding the number of existing appliances at households (stocks), by age of the appliances. Other studies using surveys combined with the Weibull distribution for this purpose are those of Kunacheva et al. (2009), Steubing et al. (2010) and Chirapat et al. (2012).
Figure 2. Procedural guideline for estimating e-waste generation (Source: Wang et al., 2013). Table 2. Mass balance models for WEEE estimates Model Equation Time Step Market Supply
Variables W(n): WEEE generation in year n S(n); S(n-1): No. appliances in stock in years n and n-1
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Stock and Lifespan
Combines Time Step and Market Supply with the condition:
“Leaching” Sales-StockLifespan (multivariate analysis)
Composition at stock age: Composition at disposal age: Lifespan distribution (Weibull)
Stock composition in different years:
POM(t): Product sales in year t; Lp(t,n): Probabilistic obsolescence rate of a class of products (discarded in n/ acquired in t); t0: Initial year of analysis; L(av): Average lifespan (obsolescence distribution average) S(t,n), S(m,n): Stock in year n of products sold in year t or m L(c)(t’, t): Cumulative distribution of lifespan from year t to n α,β: Weibull distribution parameters
Source: Wang et al. (2013) 2.2 WEEE estimates in Brazil Few studies have focused WEEE generation estimates in Brazil. Araujo et al. (2012) estimated nationwide generation of some WEEE categories: TVs, refrigerators, freezers, washing machines, audio systems (mature markets); and computers and mobile phones (nonmature markets). For mature markets they adopted the “leaching” model, whilst for non-mature markets the Time-Step model was used. Lifespan data adopted were obtained from diverse literature, whilst stock data was gathered from the national census. The study from ABDI (2013) is a technical reference for the feasibility analysis of WEEE take-back in Brazil, hired by the Federal Goverment. They adopted the Carnegie Mellon method (Market Supply with discrete average lifespan). POM data was acquired from industry and commerce representatives, whilst the lifespan averages do not correspond to actual data from Brazil. Rodrigues et al. (2015) applied a survey in the São Paulo city in order to obtain data such as stock and age of appliances, but they still adopted lifespan averages from the international literature in past decades.
3. MATERIALS AND METHODS 3.1 Scope of the study This study aims at developing WEEE lifespan and disposal profiles of the city of Sao Jose dos Campos, in the State of Sao Paulo, Brazil, with estimated population of 695,992 inhabitants in 2016 (IBGE, 2017). Such primary data must reflect the contexts of each of the seven great Zones of the city (Table 3).
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 3. Population density in the Zones of Sao Jose dos Campos, Brazil, year 2015 Zone No. Households % Households No. Population % Population Center 27855 13.3% 80779 11.7% North 19702 9.5% 66984 9.7% East 53039 25.3% 180331 26.2% Southeast 14251 6.9% 51303 7.5% South 76939 37.4% 261592 38.0% West 14874 7.3% 46109 6.7% S.F.Xavier District 486 0.2% 1504 0.2% TOTAL 207146 100% 688602 100% Source: S.J.CAMPOS (2016) The types of EEE/WEEE appliances in the scope of the study were selected aiming at covering a wide variety of those considered the most common equipments at households: (a) Refrigerators/Freezers; (b) Air conditioners; (c) CRT TV; (d) LCD TV; (e) landline phones; (f) smartphones; (g) non-smartphone mobiles; (h) notebooks; (i) desktops; (j) LCD monitors; (k) CRT monitors; (l) computer keyboards; (m) washing machines; (n) basic printers; (o) multiuse printers; (p) radios. 3.2 WEEE estimate method According to the guidelines in Fig. 1 and Table 2, Market Supply is the first recommended method that adopts a variable for lifespan profiles. One advantage of this method is that the lifespan duration values are represented according to their probabilistic distributions, rather than adopting a discrete average value (as adopted in many of the studies cited in Sections 1 and 2). This permits a better representation of the variations of WEEE disposal ages according to social, technological and economic aspects. The other variable in this Model is POM(n), what can be more easily quantified at the national level or in the level of organizations, where there is a high control on EEE purchases (or sales) and imports. 3.3 Survey and analysis methods The questions on the survey form adopted in the research are listed below. It is worth noticing that Question 5 is to obtain values for POM(n) in the Market Supply Method, whilst Question 10 is to determine the lifespan distributions for each appliance. •
•
•
General questions: o 1) What municipal Zone? o 2) How many people living at household? Current stock (for each type of appliance mentioned in Section 3.1): o 3) How many do you actually have? o 4) How many are working? o 5) In which year did you acquire the appliances? o 6) How many were new when acquired? o 7) Do you intend to change, fix or donate any of them? o 8) When to change/fix/donate? Previous stock (for each type of appliance mentioned in Section 3.1): o 9) How many did you change or disposed of from year 2010? o 10) How many years did it last?
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
•
o 11) How many were working? o 12) How many did you change or dispose of from year 2010? Disposal: o 13) Where do you usually dispose of WEEE? (walkway; mixed MSW; selective collection; cooperative; special collection; others) o 14) To whom do you usually donate second-hand EEE? (Acquitances; NGOs; waste pickers; cooperatives; others).
In order to represent the social and economic profiles of the different municipal Zones (Table 3), the selected population sampling method were: stratified sampling to estimate a proportion of a finite population (Eq. 1), and stratified sampling to estimate an average of a finite population. Both methods are selected because the survey form contains both quantitative and qualitative variables. The highest sample size between both Equations was adopted for each Zone.
(Eq. 1)
(Eq. 2) Where: n = Sample size of the stratified zone Ni = Population size of stratum i σi2 = Population variance of stratum i wi = Ni/n D = d2/Z2, where d = sample error (adopted: 5%), and Z = abscissa of the standard normal distribution = Estimate of the true proportion of stratum (when there is no previous estimate, adopt pi= 0,50, this way obtaining the higher possible sample size for i. Equation 1 was first applied because it does not depend on variance, with a sample error of 5%, what resulted in a sample size of 399 households. The survey started in August 2016, being applied alternatively in each city zone, seeking to adjust the sample proportions to the true Zone households proportions (Table 2). After the first 100 responses, the variance values were adjusted to Eq. 2, estimating a sample size of 401 households. After 8 months with the work of six students, the realised sample of the survey was 459 interviews, with maximum variation of 3% among the true household proportions in Table 2. The S.F.Xavier District was not considered in this sample because its calculated sample size equals to 1 would not reflect a minimum acceptable distribution for the values.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4. RESULTS AND DISCUSSION 4.1 Lifespan profiles From the total sample of 459 interviewed households, the obtained sample sizes of discarded equipment, for each type of appliance, are presented in Table 4, together with their average lifespans and respective standard deviations. From this Table, it is possible to notice that the most discarded equipment by the population of the city from year 2010 are smartphones, refrigerators or freezers, and washing machines. For this reason, deeper analyses of the results focus on these appliances. In comparison to Table 1, it is possible to notice that currently adopted lifespan averages in Brazilian studies are much higher than those obtained in Table 4, indicating that these studies are either adopting data that may not reflect Brazilian reality, of that these may be obsolete. This highlights the importance of studies like the present one to produce local data. Such higher lifespan averages, if adopted for example in planning WEEE take-back systems in Sao Jose dos Campos, would lead to underestimated WEEE generation estimates in the short term. Another interesting observation from Table 4 is that smartphones have an average lifespan of less than two years, reflecting their fast obsolescence, and have been discarded at a rate of 2.8 appliances per household, reflecting high usage and disposal rates by all family members. On the other hand, the highest lifespans are those for CRT monitors (10 years, but for a very small sample), refrigerators and freezers (7.8 years), and CRT TVs (6.9 years). For the selected WEEE estimation method (Market Supply), the lifespan profiles are represented by a probabilistic distribution, i.e. the % of households that discards a specific appliance after 1, 2, ... , n years. In the present study, the obtained lifespan profile for smartphones is presented in Fig. 3. In order to estimate the WEEE generation from the current EEE in stock (acquired EEE as answered by population sample and projected to total households), the Market Supply method was applied, and the result for smartphones is presented in Figure 4. The lifespan profiles and disposal estimates for smartphones per city Zone are illustrated in Fig. 5. In can be seen in Fig. 5 that in some Zones the most likely lifespan for smartphones is 1 year, whilst in most Zones this is 2 years. Interestingly, the Zones with a 1year lifespan of smartphones are one high-income (North) and one low-income (East) area, what may indicate possibilities like: both low- and high-income households have the same lifespan profile; North has high obsolescence rate and East is a receiver of second-hand mobiles; high obsolescence at North and low-quality products acquired by clients from East; or others. Lifespan variations among Zones also occurred to other types of appliances. Fig. 6 presents lifespan distributions of other types of appliances (for the total sample).
Figure 3. Smartphones lifespan (years) distribution for the total sample of the study
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 4. Obtained sample sizes, average lifespans and standard deviations for each type of discarded appliance (from year 2010) WEEE
Discarded
Discarded 2010+: %
No.
No. households
2010+:
households
appliances
discarded
Avg.
household
discarded
/
Average (years)
lifespan
Lifespan deviation
Refrigerator/Freezer
130
28,3%
176
0.39
7.78
4.43
Air Conditioner
0
0%
0
0
-
-
CRT TV
25
5.4%
36
0.08
6.94
2.51
LCD TV
63
13.7%
93
0.21
4.34
2.38
Landline phone
21
4,6%
53
0.12
3.30
2.49
Smartphone
329
71.7%
1243
2.77
1.78
0.91
Non-Smart. mobile
24
5.2%
48
0.11
2.29
1.24
Notebook
63
13.7%
89
0.20
4.02
2.25
Desktop
9
2.0%
16
0.04
5.94
3.07
LCD monitor
8
1.7%
14
0.03
5.11
2.94
CRT monitor
1
0.2%
2
0.01
10
0
Keyboard
9
2.0%
16
0.04
5.93
3.07
Washing machine
115
25.1%
119
0.27
6.76
2.92
Basic printer
6
1.3%
18
0.04
2.94
1.39
Multiuse printer
18
3.9%
30
0.07
4.31
2.56
Radio
5
1.1%
7
0.02
4.14
0.69
standard
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 4. WEEE smartphones generation estimate from current stock (per year of acquisition)
Central
West
East
Southeast
North
South
Figure 5. Smartphones lifespan profiles according to different city Zones
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 6. Lifespan distributions (total sample) for some WEEE types
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4.2 WEEE generation projection for further years (future Products-on-Market): The survey is a powerful tool to collect primary data for WEEE estimate models, like lifespan profiles and current stocks (what are the sum POM(n) for n from previous years to now). For the variable POM(n) in the Market Supply model, the current survey could collect data for products acquired from past years to now, as declared by the citizens. However, for the medium- and long-term planning of WEEE take-back schemes, it is necessary to apply marketing techniques in order to estimate the further acquisition of EEE products, and thus estimate when these further acquired products will probably become WEEE to be collected and treated. This escapes the scope of this study, but as an exercise, Fig. 7 shows a regression line obtained for the POM(n) behaviour presented by the study sample for current smartphone stocks and their respective acquisition ages. The curve with highest correlation index was exponential (R2 = 0.93), but this projection needs to be taken with extreme caution, because the increase rate in smartphone acquisitions in the latest years were a particular phenomenon that is biasing the curve, leading to unreal values for the near future – for example, around 5,800,000 WEEE smartphones in 2021, whilst in the same year the estimated population is 752,370, with a 1.57% population growth rate per year (IBGE, 2017), leading to a WEEE smartphone/capita rate of 7.7, compared to 0.3 in 2016. Because of this, Fig. 7 shows a linear regression line with a reasonable correlation index of 0.75, that seems more plausible, indicating a WEEE smartphones generation per capita shifting from 0.28 in 2016 to 0.37 in 2021. Other feasible POM(n) estimates using regression curves are obtained for refrigerators and washing machines (Fig. 8), although in these cases the correlation index is around 0.7. For these appliances, respectively, from 2016 to 2021 the estimated WEEE per capita is 0.02 to 0.05 (refrigerators), and steady at 0.03 (washing machines).
Figure 7. Smartphones POM(n) projection by linear regression based on survey data
Washing machines
Refrigerators / Freezers
Figure 8. POM regression curves for washing machines and refrigerators/freezers
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4.3 WEEE disposal behaviour The survey also searched to identify what is the behaviour of the population regarding WEEE disposal. Table 5 presents the results (total and per zone). It is relevant to highlight in Table 5 that many people may have mistaken the existing special WEEE collection system with the selective collection for recyclables, as these are carried out by the same MSWM company, what may have biased the results. It is also worth noticing that S.J. dos Campos is considered a benchmark for selective recyclable collection schemes in Brazil, as in the last 10 years this service has a coverage of 90+% in the city (CEMPRE, 2016), what is by far higher than the Brazilian average coverage of 15% (see Section 1), indicating that citizens in this city have a historical tradition in separating waste. It can also be seen that East Zone, a low-income region, has the highest disposal rate on the walkways, and the lowest to the selective collection system. The highest rate to selective and special collection was at the North Zone, a high-income region in the city. Table 6 presents results for de usual destination of WEEE donations prior to disposal, the most common being donation to acquitances or to waste pickers. Table 5. WEEE disposal behaviour in S.J. dos Campos, Brazil (total and per Zones) Disposal Total Central S E N W Walkway 8.0% 5.9% 8.4% 12.7% 9.8% 3.4% Special collection 6.75% 8.1% 5.9% 7.5% 4.5% 8.9% Selective collection 75.5% 73.9% 77.5% 70.5% 79.3% 78.6% Mixed with MSW 5.8% 7.25% 5.6% 5.8% 4.2% 3.3% Others 3.9% 4.9% 2.6% 3.5% 2.2% 5.8%
SE 7.8% 5.5% 73.4% 8.6% 4.5%
Table 6. WEEE donation receivers in S.J. dos Campos, Brazil (total and per Zones) Disposal Total Central S E N W Waste pickers 11.0% 6.4% 9.3% 14.7% 9.2% 17.5% Acquitances 71.3% 68.0% 74.8% 68.9% 71.3% 71.3% Cooperatives 4.2% 5.9% 3.9% 5.4% 3.2% 2.2% NGOs 9.5% 14.6% 9.1% 7.7% 9.9% 6.1% Others 4.0% 5.1% 2.9% 3.5% 6.4% 2.8%
SE 8.9% 73.5% 4.3% 9.9% 3.5%
5. CONCLUSIONS In summary, we can conclude that: • Considering the Brazilian actual studies on WEEE generation estimates, adopting lifespan profiles from diverse literature may represent significantly different disposal ages than the currently taking place at the regions of study, and thus mislead the accurate WEEE generation estimate for the next few years; • Surveys are a powerful tool to produce local primary data on EEE consumption, WEEE lifespan and other variables of the WEEE estimate models; • Market projections for EEE sales are needed to produce data for POM(n) considering the future products acquisitions by citizens (the current study could only estimate WEEE for current products in stock, i.e. already purchased by citizens); • The city of Sao Jose dos Campos presented significant variation in WEEE lifespan profiles and disposal behaviour among its different Zones, reflecting the influences of diverse social and economic aspects.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Further steps are: • To apply the Market Supply model using market estimations of POM(n) in the next years in the city; • To dimension WEEE take-back schemes based on WEEE generation estimates in the city; and • To Estimate the WEEE recycling and hazardousness potentials - amounts of valuable and hazardous components in the WEEE composition - in the city.
AKNOWLEDGEMENTS The authors thank the support by UNESP via its call 09/2016-PROPe, “Primeiros Projetos”. The authors also acknowledge the volunteer students who helped in the survey: Kauê Moura, Maitê Packer, and Nathalia Rodrigues.
REFERENCES ABDI (2013). Logística Reversa de Equipamentos Eletroeletrônicos: Análise de Viabilidade Técnica e Econômica. Available at: http://www.abdi.com.br/Estudo/Logistica%20reversa%20de%20residuos_.pdf ABINEE (2013). Panorama econômico e Desempenho Setorial – 2013. São Paulo: Associação Brasileira da Indústria Elétrica e Eletrônica – ABINEE. Available at: http://www.abinee.org.br/programas/50anos/public/panorama/#20/z ABRELPE (2014). Panorama dos Resíduos Sólidos no Brasil 2014. São Paulo: ABRELPE. Araújo, M. G., Magrini, A., Mahler, C. F., & Bilitewski, B (2012). A model for estimation of potential generation of waste electrical and electronic equipment in Brazil. Waste Management, 32(2), 335-342. doi:10.1016/j.wasman.2011.09.020 BRASIL (2010). Law. No. 12.305/2010 – Establishes the National Solid Waste Policy. CEMPRE (2016). Pesquisa Ciclosoft 2016. Available at: http://cempre.org.br/ciclosoft/id/8. CEMPRE. Chirapat, P., Kittinan, A., Kiattiporn, W., Nawanuch, T., & Surus, T. (2012). Development of forecasting model for strategically planning on E-waste management in Thailand. In Electronics Goes Green 2012+, ECG 2012 – Joint International Conference and Exhibition, Proceedings. Huisman, J.; Magalini, F.; Kuehr, R.; Maurer, C.; Ogilvie, S.; Poll, J.; Delgado, C.; Artim, E.; Szlezak, J.; Stevels, A (2008). 2008 Review of Directive 2002/96 on Waste Electrical and Electronic Equipment (WEEE). Bonn: United Nations University. IBGE (2011). Pesquisa Nacional por Amostra de Domicílios – PNAD 2011. Rio de Janeiro: IBGE, 2011. IBGE (2017). http://www.cidades.ibge.gov.br IPEA (2012). Diagnóstico sobre Catadores de Resíduos Sólidos: Relatório de Pesquisa. Brasília: IPEA. Kiddee, P., Naidu, R., Wong, M. H (2013). Electronic waste management approaches: an overview. Waste Management, 33(5), 1237–50. doi:10.1016/j.wasman.2013.01.006 Kim, S., Oguchi, M., Yoshida, A., Terazono, A. Estimating the amount of WEEE generated in South Korea by using the population balance model. Waste Management 33 (2013) 474-483. Doi: 10.1016/j.wasman.2012.07.011
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Kunacheva, C., Juanga, J. P., & Visvanathan, C. (2009). Electrical and electronic waste inventory and management strategies in Bangkok, Thailand. International Journal of Environment and Waste Management, 3(1/2), 107. doi:10.1504/IJEWM.2009.024703 Lee, S.J., & Cooper, J. (2008). Estimating regional material flows for LCDs. In 2008 IEEE International Symposium on Electronics and the Environment-ISEE: Proceedings (pp. 320– 325). Lundgren, K (2012). The global impact of e-waste: Addressing the challenge. Geneva: International Labour Office, Programme on Safety and Health at Work and the Environment (SafeWork), Sectoral Activities Department (SECTOR). Polak, M., & Drapalova, L (2012). Estimation of end of life mobile phones generation: The case study of the Czech Republic. Waste Management, 32(8), 1583–1591. doi:10.1016/j.wasman.2012.03.028 Robinson, B. H. (2009). E-waste: An assessment of global production and environmental impacts. Science of the Total Environment, 408(2), 183–191. doi:10.1016/j.scitotenv.2009.09.044 Rodrigues, A.C.; Gunther, W.M.R.; Boscov, M.E.G. (2015). Estimativa da geração de resíduos de equipamentos elétricos e eletrônicos de origem domiciliar: proposição de método e aplicação ao município de São Paulo, São Paulo, Brasil. Eng. Sanit. Ambient., v.20 n.3, jul/set 2015, 437-447. doi: 10.1590/S1413-41522015020000133701 S.J.CAMPOS (2016). http://www.sjc.sp.gov.br/. Accessed in 01/04/2016 StEP (2015). E-Waste World Map. Solving the E-Waste Problem (StEP). Available at: www.step-initiative.org/step-e-waste-world-map.html Steubing, B., Böni, H., Schluep, M., Silva, U., & Ludwig, C. (2010). Assessing computer waste generation in Chile using material flow analysis. Waste Management, 30(3), 473–82. doi:10.1016/j.wasman.2009.09.007 Wang, F., Huisman, J., Stevels, A., & Baldé, C. P. (2013). Enhancing e-waste estimates: improving data quality by multivariate Input-Output Analysis. Waste Management, 33(11), 2397–407. doi:10.1016/j.wasman.2013.07.005
SUMMARY: WEEE generation estimates in cities are relevant for planning adequate take-back and treatment schemes, but face the challenge of lack of trustable local data. Population surveys are a recommended technique to support data collection for WEEE estimating models. The objective of this study is to determine the EEE acquisition, WEEE lifespan and disposal profiles in the city of Sao Jose dos Campos, Brazil, in order to allow applying the Market Supply WEEE estimating model. A stratified sample with error margin of 5% was calculated and interviewed in the city. The lifespan probabilistic profiles for different WEEE appliances indicate variations that represent the social and economic aspects of each city Zone. The amounts of WEEE per year from current products in use were estimated, but a quantification of WEEE generation from further products to be acquired requires marketing projections. The disposal behaviour of population in the city shows that most common practices are the destination to selective and special collection schemes, but also the disposal on walkways. Regarding WEEE (second-hand) donation, the main receivers are acquitances and waste pickers.
1. INTRODUCTION Waste Electrical and Electronic Equipment (WEEE) generation in Brazil and worldwide is a critical issue in environmental, social and economic aspects. It is the fastest growing waste stream (Lundgren, 2012). It is estimated that Brazil produced 7 kg WEEE/capita in 2014, a higher rate than in countries like China or India (StEP, 2016). Such estimate can be explained by factors like the Purchasing Power of Population, what can also provoke variations in WEEE generation within different regions of the country. The computers market in Brazil had a large growth in the last 10 years; in 2011, the percentage of households with computers reached 43%, double the proportion in 2006 (IBGE, 2011). Electrical and Electronic Equipment (EEE) enter the Brazilian market not just through legal production and retailing chains, but also with significant contribution of the informal market. In 2014, unnofficial markets accounted for 1.5 million computers sold, 15% of the total in that year, and 35% considering only desktops (ABINEE, 2013). After the lifespan of these appliances, the large amount of WEEE produced, when not adequately treated, has high toxicity and contamination risks. There are more than 1,000 hazardous substances associated with WEEE, whilst the most commonly reported are dozens of toxical metals (eg. Ba, Be, Cd, Co, Cr, Cu, Fe, Pb, Li, La, Hg, Mn, Mo, Ni, Ag) and Persitant Organic Pollutants (POPs) like Brominated Flame Retardants (BFRs), Policyclic Aromatic Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Hidrocarbons (PAHs) and Polyvinyl Chloride (PVC) (Kidee et al., 2013). In Brazil, critical factors for the exposure to such WEEE agents are: • Low coverage of selective collection: 15% of Brazilian population attended in 2016 (CEMPRE, 2016); • Inadequate waste disposal: 60% cities or 42% MSW in Brazil (ABRELPE, 2014); • Large-scale activity of informal waste pickers: estimated 70.5k waste pickers in urban areas in 2008, of which 18% in the State of Sao Paulo, and 5.5 children under 14 y.o.; 27% cities acknowledged waste pickers at the final waste disposal areas (IPEA, 2012). On the other hand, WEEE has a high economic potential by recovery of valuable materials such as Gold, Paladium, Silver and Berilium. Because of this, in developing countries like Brazil, informal sector has widely applied rough recycling techniques such as acid leaching of PCBs and open burning of cables, what poses significant risks to the health of these informal workers and surrounding communities (Robinson, 2009; Lundgren, 2012; Kidee et al., 2013). Nevertheless, adequate WEEE recycling in Brazil is estimated in only 2% (Bandini, 2009 apud Araujo, 2012). WEEE accounts for less than 1% of waste collected by selective collection schemes (CEMPRE, 2016). Another common route for Brazilian WEEE is to be landfilled mixed with MSW. In the city of Rio de Janeiro in 2012, 3.7k tonnes WEEE were destined to the urban landfill (COMLURB, 2013). Considering both the hazardousness and economic value potentials for WEEE, the Brazilian National Solid Waste Policy (Brasil, 2010) established as mandatory the implementation of WEEE take-back schemes under the shared responsibility of EEE producers, importers, distributors and retailers, as well as the Ministry of Environment (MMA). In practice this is not implemented due to lack of agreement among these parties on some critical points. A very important information for the planning of efficient WEEE take-back schemes is the amount and types of WEEE generated within a particular area, especially the municipalities. Most actual studies focused WEEE estimates in the country level, some concerned with specific types of appliances (ex.: LCD – Lee & Cooper, 2008; mobile phones – Polak & Drapalova, 2012), whilst others only estimate a total WEEE amount in the countries (e.g. StEP, 2015). Estimates in the country level do not provide resourceful information for the planning of takeback schemes in municipalities and regions, because they do not detail the different social and economic contexts. Estimating total amounts of WEEE also do not provide useful information for planning because the variety of appliances require different collection and treatment strategies. In Brazil, few studies have engaged in estimating WEEE generations. Araujo et al. (2012) estimated the generation of some kinds of WEEE in the country level, and a more general estimation was carried out by ABDI (2013). Rodrigues et al. (2015) estimated the WEEE generation in the city of São Paulo, by applying a survey to a sample population. All these studies used estimation methods based on the volume of “Products-on-Market” (POM), i.e. the number of each type of appliance purchased in the country every year; and an (discrete) average lifespan of such appliances, in all cases, obtained from international literature (Table 1), therefore not reflecting the actual lifespan profiles in each region of the country. Table 1. Some WEEE Lifespan averages (in years) adopted in Brazilian studies WEEE Araujo et al. (2012) ABDI (2013) Rodrigues et al. (2015) Refrigerator 12 15 11 Mobile phone 4 3 4 Washing machine 10 11 11
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
The objective of this study is to determine the EEE consumption, WEEE lifespan and disposal profiles in the city of Sao Jose dos Campos, Brazil, ir order to support planning of an adequate take-back scheme. Specific objectives are: • To produce regional data on WEEE lifespan and disposal profiles; • To determine lifespan and disposal profiles reflecting the different social and economic contexts in each Zone of the city; and • To obtain lifespan probabilistic distributions for the object of study. • 2. REVIEW ON WEEE ESTIMATES METHODS AND STUDIES 2.1 WEEE estimation methods Some WEEE estimation studies adopted correlation methods that associate the amount generated with the GDP per capita or the Purchasing Power Parity (PPP) of population (e.g. Huisman et al.., 2008; StEP, 2015). On one hand, these studies require the previous availability of data regarding WEEE generation in order to find a regression equation that allows for further projections, and on the other hand, the correlation coefficient obtained may not be significative enough to allow trustable estimates (e.g. in Huisman, 2008 the coefficient R for WEEE/head x GDP/head was around 0.55 with data from some EU Member States). Most WEEE estimating studies apply mass balance methods, which rely on data such as POM, current stocks (equipment in use) in year t, and lifespan profiles (Figure 1). WEEE estimates using mass balance methods are recommended by UNEP (2007). There are several available mass balance methods for WEEE estimates (Table 2), each adopting some of the data indicated in Figure 1. The choice of the most adequate method to be used depends at large extent on the available data. Some of them vary slightly depending on the adopted variable; for example, the Carnegie Mellon method adopts a discrete average lifespan value, rather than a lifespan distribution as in the Market Supply method.
Figure 1. Multiple variables and data points for enhanced e-waste estimation methods (Source: Wang et al., 2013). Mass balance models require trustable data. Data quality is a critical issue in WEEE estimates using and of these models. Particularly regarding the Lifespan variable, data are
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
usually obtained from previous literature, in many cases reflecting lifespan profiles from different countries and in previous decades (e.g. Araujo et al., 2012; ADBI, 2013; Rodrigues et al., 2015). In this sense, surveys with population sampes can be useful to determine not only local lifespan profiles, but stocks composition and average weights of appliances (Wang et al., 2013). The use of surveys has been a methodological resource in recent studies. Kim et al. (2013) applied a survey to 1,000 persons to determine the lifespan of EEE, supported by the Weibull distribution. The same study has also collected data regarding the number of existing appliances at households (stocks), by age of the appliances. Other studies using surveys combined with the Weibull distribution for this purpose are those of Kunacheva et al. (2009), Steubing et al. (2010) and Chirapat et al. (2012).
Figure 2. Procedural guideline for estimating e-waste generation (Source: Wang et al., 2013). Table 2. Mass balance models for WEEE estimates Model Equation Time Step Market Supply
Variables W(n): WEEE generation in year n S(n); S(n-1): No. appliances in stock in years n and n-1
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Stock and Lifespan
Combines Time Step and Market Supply with the condition:
“Leaching” Sales-StockLifespan (multivariate analysis)
Composition at stock age: Composition at disposal age: Lifespan distribution (Weibull)
Stock composition in different years:
POM(t): Product sales in year t; Lp(t,n): Probabilistic obsolescence rate of a class of products (discarded in n/ acquired in t); t0: Initial year of analysis; L(av): Average lifespan (obsolescence distribution average) S(t,n), S(m,n): Stock in year n of products sold in year t or m L(c)(t’, t): Cumulative distribution of lifespan from year t to n α,β: Weibull distribution parameters
Source: Wang et al. (2013) 2.2 WEEE estimates in Brazil Few studies have focused WEEE generation estimates in Brazil. Araujo et al. (2012) estimated nationwide generation of some WEEE categories: TVs, refrigerators, freezers, washing machines, audio systems (mature markets); and computers and mobile phones (nonmature markets). For mature markets they adopted the “leaching” model, whilst for non-mature markets the Time-Step model was used. Lifespan data adopted were obtained from diverse literature, whilst stock data was gathered from the national census. The study from ABDI (2013) is a technical reference for the feasibility analysis of WEEE take-back in Brazil, hired by the Federal Goverment. They adopted the Carnegie Mellon method (Market Supply with discrete average lifespan). POM data was acquired from industry and commerce representatives, whilst the lifespan averages do not correspond to actual data from Brazil. Rodrigues et al. (2015) applied a survey in the São Paulo city in order to obtain data such as stock and age of appliances, but they still adopted lifespan averages from the international literature in past decades.
3. MATERIALS AND METHODS 3.1 Scope of the study This study aims at developing WEEE lifespan and disposal profiles of the city of Sao Jose dos Campos, in the State of Sao Paulo, Brazil, with estimated population of 695,992 inhabitants in 2016 (IBGE, 2017). Such primary data must reflect the contexts of each of the seven great Zones of the city (Table 3).
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 3. Population density in the Zones of Sao Jose dos Campos, Brazil, year 2015 Zone No. Households % Households No. Population % Population Center 27855 13.3% 80779 11.7% North 19702 9.5% 66984 9.7% East 53039 25.3% 180331 26.2% Southeast 14251 6.9% 51303 7.5% South 76939 37.4% 261592 38.0% West 14874 7.3% 46109 6.7% S.F.Xavier District 486 0.2% 1504 0.2% TOTAL 207146 100% 688602 100% Source: S.J.CAMPOS (2016) The types of EEE/WEEE appliances in the scope of the study were selected aiming at covering a wide variety of those considered the most common equipments at households: (a) Refrigerators/Freezers; (b) Air conditioners; (c) CRT TV; (d) LCD TV; (e) landline phones; (f) smartphones; (g) non-smartphone mobiles; (h) notebooks; (i) desktops; (j) LCD monitors; (k) CRT monitors; (l) computer keyboards; (m) washing machines; (n) basic printers; (o) multiuse printers; (p) radios. 3.2 WEEE estimate method According to the guidelines in Fig. 1 and Table 2, Market Supply is the first recommended method that adopts a variable for lifespan profiles. One advantage of this method is that the lifespan duration values are represented according to their probabilistic distributions, rather than adopting a discrete average value (as adopted in many of the studies cited in Sections 1 and 2). This permits a better representation of the variations of WEEE disposal ages according to social, technological and economic aspects. The other variable in this Model is POM(n), what can be more easily quantified at the national level or in the level of organizations, where there is a high control on EEE purchases (or sales) and imports. 3.3 Survey and analysis methods The questions on the survey form adopted in the research are listed below. It is worth noticing that Question 5 is to obtain values for POM(n) in the Market Supply Method, whilst Question 10 is to determine the lifespan distributions for each appliance. •
•
•
General questions: o 1) What municipal Zone? o 2) How many people living at household? Current stock (for each type of appliance mentioned in Section 3.1): o 3) How many do you actually have? o 4) How many are working? o 5) In which year did you acquire the appliances? o 6) How many were new when acquired? o 7) Do you intend to change, fix or donate any of them? o 8) When to change/fix/donate? Previous stock (for each type of appliance mentioned in Section 3.1): o 9) How many did you change or disposed of from year 2010? o 10) How many years did it last?
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
•
o 11) How many were working? o 12) How many did you change or dispose of from year 2010? Disposal: o 13) Where do you usually dispose of WEEE? (walkway; mixed MSW; selective collection; cooperative; special collection; others) o 14) To whom do you usually donate second-hand EEE? (Acquitances; NGOs; waste pickers; cooperatives; others).
In order to represent the social and economic profiles of the different municipal Zones (Table 3), the selected population sampling method were: stratified sampling to estimate a proportion of a finite population (Eq. 1), and stratified sampling to estimate an average of a finite population. Both methods are selected because the survey form contains both quantitative and qualitative variables. The highest sample size between both Equations was adopted for each Zone.
(Eq. 1)
(Eq. 2) Where: n = Sample size of the stratified zone Ni = Population size of stratum i σi2 = Population variance of stratum i wi = Ni/n D = d2/Z2, where d = sample error (adopted: 5%), and Z = abscissa of the standard normal distribution = Estimate of the true proportion of stratum (when there is no previous estimate, adopt pi= 0,50, this way obtaining the higher possible sample size for i. Equation 1 was first applied because it does not depend on variance, with a sample error of 5%, what resulted in a sample size of 399 households. The survey started in August 2016, being applied alternatively in each city zone, seeking to adjust the sample proportions to the true Zone households proportions (Table 2). After the first 100 responses, the variance values were adjusted to Eq. 2, estimating a sample size of 401 households. After 8 months with the work of six students, the realised sample of the survey was 459 interviews, with maximum variation of 3% among the true household proportions in Table 2. The S.F.Xavier District was not considered in this sample because its calculated sample size equals to 1 would not reflect a minimum acceptable distribution for the values.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4. RESULTS AND DISCUSSION 4.1 Lifespan profiles From the total sample of 459 interviewed households, the obtained sample sizes of discarded equipment, for each type of appliance, are presented in Table 4, together with their average lifespans and respective standard deviations. From this Table, it is possible to notice that the most discarded equipment by the population of the city from year 2010 are smartphones, refrigerators or freezers, and washing machines. For this reason, deeper analyses of the results focus on these appliances. In comparison to Table 1, it is possible to notice that currently adopted lifespan averages in Brazilian studies are much higher than those obtained in Table 4, indicating that these studies are either adopting data that may not reflect Brazilian reality, of that these may be obsolete. This highlights the importance of studies like the present one to produce local data. Such higher lifespan averages, if adopted for example in planning WEEE take-back systems in Sao Jose dos Campos, would lead to underestimated WEEE generation estimates in the short term. Another interesting observation from Table 4 is that smartphones have an average lifespan of less than two years, reflecting their fast obsolescence, and have been discarded at a rate of 2.8 appliances per household, reflecting high usage and disposal rates by all family members. On the other hand, the highest lifespans are those for CRT monitors (10 years, but for a very small sample), refrigerators and freezers (7.8 years), and CRT TVs (6.9 years). For the selected WEEE estimation method (Market Supply), the lifespan profiles are represented by a probabilistic distribution, i.e. the % of households that discards a specific appliance after 1, 2, ... , n years. In the present study, the obtained lifespan profile for smartphones is presented in Fig. 3. In order to estimate the WEEE generation from the current EEE in stock (acquired EEE as answered by population sample and projected to total households), the Market Supply method was applied, and the result for smartphones is presented in Figure 4. The lifespan profiles and disposal estimates for smartphones per city Zone are illustrated in Fig. 5. In can be seen in Fig. 5 that in some Zones the most likely lifespan for smartphones is 1 year, whilst in most Zones this is 2 years. Interestingly, the Zones with a 1year lifespan of smartphones are one high-income (North) and one low-income (East) area, what may indicate possibilities like: both low- and high-income households have the same lifespan profile; North has high obsolescence rate and East is a receiver of second-hand mobiles; high obsolescence at North and low-quality products acquired by clients from East; or others. Lifespan variations among Zones also occurred to other types of appliances. Fig. 6 presents lifespan distributions of other types of appliances (for the total sample).
Figure 3. Smartphones lifespan (years) distribution for the total sample of the study
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Table 4. Obtained sample sizes, average lifespans and standard deviations for each type of discarded appliance (from year 2010) WEEE
Discarded
Discarded 2010+: %
No.
No. households
2010+:
households
appliances
discarded
Avg.
household
discarded
/
Average (years)
lifespan
Lifespan deviation
Refrigerator/Freezer
130
28,3%
176
0.39
7.78
4.43
Air Conditioner
0
0%
0
0
-
-
CRT TV
25
5.4%
36
0.08
6.94
2.51
LCD TV
63
13.7%
93
0.21
4.34
2.38
Landline phone
21
4,6%
53
0.12
3.30
2.49
Smartphone
329
71.7%
1243
2.77
1.78
0.91
Non-Smart. mobile
24
5.2%
48
0.11
2.29
1.24
Notebook
63
13.7%
89
0.20
4.02
2.25
Desktop
9
2.0%
16
0.04
5.94
3.07
LCD monitor
8
1.7%
14
0.03
5.11
2.94
CRT monitor
1
0.2%
2
0.01
10
0
Keyboard
9
2.0%
16
0.04
5.93
3.07
Washing machine
115
25.1%
119
0.27
6.76
2.92
Basic printer
6
1.3%
18
0.04
2.94
1.39
Multiuse printer
18
3.9%
30
0.07
4.31
2.56
Radio
5
1.1%
7
0.02
4.14
0.69
standard
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 4. WEEE smartphones generation estimate from current stock (per year of acquisition)
Central
West
East
Southeast
North
South
Figure 5. Smartphones lifespan profiles according to different city Zones
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Figure 6. Lifespan distributions (total sample) for some WEEE types
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4.2 WEEE generation projection for further years (future Products-on-Market): The survey is a powerful tool to collect primary data for WEEE estimate models, like lifespan profiles and current stocks (what are the sum POM(n) for n from previous years to now). For the variable POM(n) in the Market Supply model, the current survey could collect data for products acquired from past years to now, as declared by the citizens. However, for the medium- and long-term planning of WEEE take-back schemes, it is necessary to apply marketing techniques in order to estimate the further acquisition of EEE products, and thus estimate when these further acquired products will probably become WEEE to be collected and treated. This escapes the scope of this study, but as an exercise, Fig. 7 shows a regression line obtained for the POM(n) behaviour presented by the study sample for current smartphone stocks and their respective acquisition ages. The curve with highest correlation index was exponential (R2 = 0.93), but this projection needs to be taken with extreme caution, because the increase rate in smartphone acquisitions in the latest years were a particular phenomenon that is biasing the curve, leading to unreal values for the near future – for example, around 5,800,000 WEEE smartphones in 2021, whilst in the same year the estimated population is 752,370, with a 1.57% population growth rate per year (IBGE, 2017), leading to a WEEE smartphone/capita rate of 7.7, compared to 0.3 in 2016. Because of this, Fig. 7 shows a linear regression line with a reasonable correlation index of 0.75, that seems more plausible, indicating a WEEE smartphones generation per capita shifting from 0.28 in 2016 to 0.37 in 2021. Other feasible POM(n) estimates using regression curves are obtained for refrigerators and washing machines (Fig. 8), although in these cases the correlation index is around 0.7. For these appliances, respectively, from 2016 to 2021 the estimated WEEE per capita is 0.02 to 0.05 (refrigerators), and steady at 0.03 (washing machines).
Figure 7. Smartphones POM(n) projection by linear regression based on survey data
Washing machines
Refrigerators / Freezers
Figure 8. POM regression curves for washing machines and refrigerators/freezers
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4.3 WEEE disposal behaviour The survey also searched to identify what is the behaviour of the population regarding WEEE disposal. Table 5 presents the results (total and per zone). It is relevant to highlight in Table 5 that many people may have mistaken the existing special WEEE collection system with the selective collection for recyclables, as these are carried out by the same MSWM company, what may have biased the results. It is also worth noticing that S.J. dos Campos is considered a benchmark for selective recyclable collection schemes in Brazil, as in the last 10 years this service has a coverage of 90+% in the city (CEMPRE, 2016), what is by far higher than the Brazilian average coverage of 15% (see Section 1), indicating that citizens in this city have a historical tradition in separating waste. It can also be seen that East Zone, a low-income region, has the highest disposal rate on the walkways, and the lowest to the selective collection system. The highest rate to selective and special collection was at the North Zone, a high-income region in the city. Table 6 presents results for de usual destination of WEEE donations prior to disposal, the most common being donation to acquitances or to waste pickers. Table 5. WEEE disposal behaviour in S.J. dos Campos, Brazil (total and per Zones) Disposal Total Central S E N W Walkway 8.0% 5.9% 8.4% 12.7% 9.8% 3.4% Special collection 6.75% 8.1% 5.9% 7.5% 4.5% 8.9% Selective collection 75.5% 73.9% 77.5% 70.5% 79.3% 78.6% Mixed with MSW 5.8% 7.25% 5.6% 5.8% 4.2% 3.3% Others 3.9% 4.9% 2.6% 3.5% 2.2% 5.8%
SE 7.8% 5.5% 73.4% 8.6% 4.5%
Table 6. WEEE donation receivers in S.J. dos Campos, Brazil (total and per Zones) Disposal Total Central S E N W Waste pickers 11.0% 6.4% 9.3% 14.7% 9.2% 17.5% Acquitances 71.3% 68.0% 74.8% 68.9% 71.3% 71.3% Cooperatives 4.2% 5.9% 3.9% 5.4% 3.2% 2.2% NGOs 9.5% 14.6% 9.1% 7.7% 9.9% 6.1% Others 4.0% 5.1% 2.9% 3.5% 6.4% 2.8%
SE 8.9% 73.5% 4.3% 9.9% 3.5%
5. CONCLUSIONS In summary, we can conclude that: • Considering the Brazilian actual studies on WEEE generation estimates, adopting lifespan profiles from diverse literature may represent significantly different disposal ages than the currently taking place at the regions of study, and thus mislead the accurate WEEE generation estimate for the next few years; • Surveys are a powerful tool to produce local primary data on EEE consumption, WEEE lifespan and other variables of the WEEE estimate models; • Market projections for EEE sales are needed to produce data for POM(n) considering the future products acquisitions by citizens (the current study could only estimate WEEE for current products in stock, i.e. already purchased by citizens); • The city of Sao Jose dos Campos presented significant variation in WEEE lifespan profiles and disposal behaviour among its different Zones, reflecting the influences of diverse social and economic aspects.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Further steps are: • To apply the Market Supply model using market estimations of POM(n) in the next years in the city; • To dimension WEEE take-back schemes based on WEEE generation estimates in the city; and • To Estimate the WEEE recycling and hazardousness potentials - amounts of valuable and hazardous components in the WEEE composition - in the city.
AKNOWLEDGEMENTS The authors thank the support by UNESP via its call 09/2016-PROPe, “Primeiros Projetos”. The authors also acknowledge the volunteer students who helped in the survey: Kauê Moura, Maitê Packer, and Nathalia Rodrigues.
REFERENCES ABDI (2013). Logística Reversa de Equipamentos Eletroeletrônicos: Análise de Viabilidade Técnica e Econômica. Available at: http://www.abdi.com.br/Estudo/Logistica%20reversa%20de%20residuos_.pdf ABINEE (2013). Panorama econômico e Desempenho Setorial – 2013. São Paulo: Associação Brasileira da Indústria Elétrica e Eletrônica – ABINEE. Available at: http://www.abinee.org.br/programas/50anos/public/panorama/#20/z ABRELPE (2014). Panorama dos Resíduos Sólidos no Brasil 2014. São Paulo: ABRELPE. Araújo, M. G., Magrini, A., Mahler, C. F., & Bilitewski, B (2012). A model for estimation of potential generation of waste electrical and electronic equipment in Brazil. Waste Management, 32(2), 335-342. doi:10.1016/j.wasman.2011.09.020 BRASIL (2010). Law. No. 12.305/2010 – Establishes the National Solid Waste Policy. CEMPRE (2016). Pesquisa Ciclosoft 2016. Available at: http://cempre.org.br/ciclosoft/id/8. CEMPRE. Chirapat, P., Kittinan, A., Kiattiporn, W., Nawanuch, T., & Surus, T. (2012). Development of forecasting model for strategically planning on E-waste management in Thailand. In Electronics Goes Green 2012+, ECG 2012 – Joint International Conference and Exhibition, Proceedings. Huisman, J.; Magalini, F.; Kuehr, R.; Maurer, C.; Ogilvie, S.; Poll, J.; Delgado, C.; Artim, E.; Szlezak, J.; Stevels, A (2008). 2008 Review of Directive 2002/96 on Waste Electrical and Electronic Equipment (WEEE). Bonn: United Nations University. IBGE (2011). Pesquisa Nacional por Amostra de Domicílios – PNAD 2011. Rio de Janeiro: IBGE, 2011. IBGE (2017). http://www.cidades.ibge.gov.br IPEA (2012). Diagnóstico sobre Catadores de Resíduos Sólidos: Relatório de Pesquisa. Brasília: IPEA. Kiddee, P., Naidu, R., Wong, M. H (2013). Electronic waste management approaches: an overview. Waste Management, 33(5), 1237–50. doi:10.1016/j.wasman.2013.01.006 Kim, S., Oguchi, M., Yoshida, A., Terazono, A. Estimating the amount of WEEE generated in South Korea by using the population balance model. Waste Management 33 (2013) 474-483. Doi: 10.1016/j.wasman.2012.07.011
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Kunacheva, C., Juanga, J. P., & Visvanathan, C. (2009). Electrical and electronic waste inventory and management strategies in Bangkok, Thailand. International Journal of Environment and Waste Management, 3(1/2), 107. doi:10.1504/IJEWM.2009.024703 Lee, S.J., & Cooper, J. (2008). Estimating regional material flows for LCDs. In 2008 IEEE International Symposium on Electronics and the Environment-ISEE: Proceedings (pp. 320– 325). Lundgren, K (2012). The global impact of e-waste: Addressing the challenge. Geneva: International Labour Office, Programme on Safety and Health at Work and the Environment (SafeWork), Sectoral Activities Department (SECTOR). Polak, M., & Drapalova, L (2012). Estimation of end of life mobile phones generation: The case study of the Czech Republic. Waste Management, 32(8), 1583–1591. doi:10.1016/j.wasman.2012.03.028 Robinson, B. H. (2009). E-waste: An assessment of global production and environmental impacts. Science of the Total Environment, 408(2), 183–191. doi:10.1016/j.scitotenv.2009.09.044 Rodrigues, A.C.; Gunther, W.M.R.; Boscov, M.E.G. (2015). Estimativa da geração de resíduos de equipamentos elétricos e eletrônicos de origem domiciliar: proposição de método e aplicação ao município de São Paulo, São Paulo, Brasil. Eng. Sanit. Ambient., v.20 n.3, jul/set 2015, 437-447. doi: 10.1590/S1413-41522015020000133701 S.J.CAMPOS (2016). http://www.sjc.sp.gov.br/. Accessed in 01/04/2016 StEP (2015). E-Waste World Map. Solving the E-Waste Problem (StEP). Available at: www.step-initiative.org/step-e-waste-world-map.html Steubing, B., Böni, H., Schluep, M., Silva, U., & Ludwig, C. (2010). Assessing computer waste generation in Chile using material flow analysis. Waste Management, 30(3), 473–82. doi:10.1016/j.wasman.2009.09.007 Wang, F., Huisman, J., Stevels, A., & Baldé, C. P. (2013). Enhancing e-waste estimates: improving data quality by multivariate Input-Output Analysis. Waste Management, 33(11), 2397–407. doi:10.1016/j.wasman.2013.07.005
