Lead-based paint in children's toys sold on China's ...

Environmental Pollution 241 (2018) 311e318

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Lead-based paint in children's toys sold on China's major online shopping platforms* Zhengtao Shen a, b, Deyi Hou a, *, Ping Zhang a, Yinan Wang a, Yunhui Zhang c, Peili Shi a, David O'Connor a a b c

School of Environment, Tsinghua University, Beijing, 100084, China Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, T6G 2E3, Canada Department of Engineering, University of Cambridge, Cambridge, CB2 1PZ, United Kingdom

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 March 2018 Received in revised form 17 May 2018 Accepted 22 May 2018

Lead in paint is a major public health concern and has drawn much attention from international organizations such as the World Health Organization (WHO) and the United Nations Environmental Programme (UNEP). However, there is limited research on lead-based paint, especially its use in toys, which poses an explicit health risk to vulnerable children. The present study sampled 100 different toys purchased from the three largest online shopping platforms in China: JD, Taobao (TB) and Tmall (TM). These selected toys have been sold nearly 3 million times in total, suggesting that they may have reached the dwellings of millions of children. It was found that the toys sold by JD and TM, which are considered organized sellers, had average lead concentrations of 25 mg/kg and 32 mg/kg, respectively, much lower than that of toys sold by unorganized sellers on the TB platform, at 219 mg/kg. Approximately 12% of the toys purchased from TB contained paint with total lead concentrations exceeding China's regulatory standard for paints in toy manufacturing, and nearly 36% of the toys purchased from TB exceeded the equivalent US regulatory standard and EU standard. These results suggest that further action is needed to verify the health and safety standards of toys sold by such unorganized sellers. Moreover, this study found that China's regulatory standard (90 mg/kg) based on soluble lead may underestimate the risk posed by lead in paints, and it is suggested that future regulatory thresholds for lead levels in paints for toy manufacture be based on total rather than soluble lead concentrations. The present study also explored various influencing factors on lead concentration, and found that lead concentrations were related to toy price, age group, color, and sales volume. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Pb Blood lead levels Lead concentration Toy Online shopping

1. Introduction Lead (Pb) is one of the 10 chemicals of highest public health concern. In 2016, lead exposure accounted for nearly half a million deaths and over 9 million disability-adjusted life years (DALYs) (WHO, 2017). The toxic effect of lead exposure on IQ is irreversible and persists for a lifetime (Gilbert and Weiss, 2006). Children are most vulnerable to lead exposure; moreover, children's mouthing behavior has been identified as an important exposure pathway (Lanphear and Roghmann, 1997). Previous studies in the United States (US) show that two major sources of lead exposure for

*

This paper has been recommended for acceptance by B. Nowack. * Corresponding author. E-mail address: [email protected] (D. Hou).

https://doi.org/10.1016/j.envpol.2018.05.078 0269-7491/© 2018 Elsevier Ltd. All rights reserved.

children were airborne lead from leaded gasoline and chips or dust containing deteriorated lead-based paint (Committee on Environmental Health, 2005). After the ban of leaded gasoline, the primary source of children's lead poisoning became deteriorating lead paint (Lofgren et al., 2000). Currently, lead exposure in children incurs an enormous economic cost for developing countries, amounting to approximately $977 billion dollars per year for low-income and middle-income countries, representing 1.9%, 2.0%, and 4.0% of the GDPs of Asia, Latin America, and Africa, respectively (IPEN, 2016). Lead in paint can cause elevated lead concentrations in toys, deteriorated paint chips, dusts and soils, resulting in lead exposure to children at home and in playgrounds. In 2011, the United Nations Environment Programme (UNEP) and World Health Organization (WHO) formed the Global Alliance to Eliminate Lead Paint (GAELP), with a mission to eliminate lead in paint globally by 2020. The

312

Z. Shen et al. / Environmental Pollution 241 (2018) 311e318

GAELP aimed to have a minimum of 70 countries to have adopted legally binding controls on the production, import, sale, and use of lead based paints by 2015, and get all countries to have such legally binding measures by 2020 (UNEP, 2013). Despite regulatory bans of lead-based paint for toy manufacture in many countries, lead exceedance in painted toys has been observed in China (Lin et al., 2009), the US (Greenway and Gerstenberger, 2010), Columbia (Mateus-Garcia and RamosBonilla, 2014), etc. China is the largest exporter of toys in the international market (Anwar, 2014). The US Consumer Product Safety Commission issued recalls of millions of units of lead contaminated toys in 2007, most of which were manufactured in China (Meyer et al., 2008). The use of lead based paint in toys has been the target of criticism, causing damage to the reputation of Chinese manufacturers (Beamish and Bapuji, 2008). On the other hand, the Chinese government has introduced increasingly stringent regulatory thresholds for paint used in toy manufacture (GAQSIQ, 2003, 2009; 2014a; b). The latest regulatory standard became effective on January 1, 2016. However, to our knowledge, although several other countries conducted studies on lead concentrations in toys' paint recently (Table S1), China has not examined this since 2009. Toys sold on China's market generally have higher lead concentrations than those sold in other countries (Table S1). Therefore, it is important to address this problem and find potential solutions. Another aspect of concern is the emergence of online selling of children's toys. In recent years, consumers are rapidly shifting to online shopping. High Street sellers like Toys”R”Us are forced to bankruptcy (Rizzo and Kapner, 2017), with the closure of many retail stores (Thompson, 2018). Online shoppers usually rely upon user-generated ratings to assess product approval (Flanagin et al., 2014); however, user-generated knowledge tends to be related to physical attributes, and it is unlikely to include information regarding risks associated with lead based paint. Product quality in online shopping has become a serious concern, especially with the fast-growing e-commerce market in China (Li et al., 2015). In the present study, we assessed a variety of toy products purchased from different online shopping platforms. The objectives of this study included: 1) to determine the lead levels in lead-based paint in toys assess whether lead-based paint is an issue for toys sold on Chinese online shopping platforms; 2) compare the exceedance of regulatory thresholds, among products purchased from platforms of organized versus unorganized sellers; and 3) explore the various influencing factors which may relate to the lead contents in toys. 2. Methods 2.1. Selection of online shopping platforms The three largest online shopping platforms in China were selected for the present study: JD, Taobao (TB), and Tmall (TM). Among these three platforms, JD (https://www.jd.com/) and TM (https://www.tmall.com/) are considered organized sellers and are analogous of the US counterpart Amazon, while TB (https://www. taobao.com/) hosts third party small businesses and individual entrepreneurs (i.e. unorganized sellers) on their platform, and is often considered China's counterpart of eBay. JD had net revenues of ￥362 billion RMB (US$55.7 billion) in 2017, a 40.3% increase from 2016 (JD, 2018). The revenues of TB and TM (both owned by Alibaba Group Holding Ltd) reached ￥3 trillion RMB (US$462 billion) in 2016 (Lee, 2016). Given the large volumes of consumer products being sold on these platforms, the quality of these products undoubtedly raises a concern. In the context of the present study, we intend to examine and compare the occurrence of leadbased paint on children's toys. We hypothesized that lead based

paint is less often found in products sold by organized sellers JD and TM in comparison with products sold by unorganized sellers on TB. 2.2. Sample selection and processing A total of 100 toys were randomly selected and purchased from JD (n ¼ 24), TM (n ¼ 43) and TB (n ¼ 33). The following criteria were used in selecting toys for purchase: 1) the toys should cover a wide range of colors, to include more than 10 yellow, red, and black toys each (note: previous studies have indicated that these colors tend to have higher lead content than lighter colors); 2) the toys should cover a wide range of costs and sales volumes; and 3) the selected toys should be intended for various age groups. It should be noted that there are myriad of toy products sold on these online platforms. The sample size of the present study (n ¼ 100) is only a tiny fraction of their huge inventory, and by no means would our samples provide a representative picture of the overall quality of products sold on these platforms. Nevertheless, because of the targeted sampling regime used, we believe our study results may provide meaningful insights. Upon delivery, the toys were unwrapped, and the paints were manually scrapped into a receptacle using stainless steel knives. A sufficient quantity of paint (>0.4 g) was collected to meet the analysis need as described in Section 2.3. 2.3. Chemical analyses The paint scrapings were crushed and passed through a 0.5 mm mesh. The soluble lead content was determined by the method specified in GB 6675.4e2014 Safety of toys - Part 4: Migration of certain elements. The procedure involved heating HCl (0.07 mol/L) to 37 ± 2 C, to which the paint sample (0.2 g) was added (50:1 m:m). The mixture was magnetically stirred at 600 rpm for 1 min to reach a pH of 1.0e1.5; if the pH remained >1.5, stirring was continued and HCl (2 mol/L) was added dropwise until pH 1.0e1.5 was reached. The acidified mixture was then stirred in darkness at 37±2 C for 1 h. The mixture was allowed to settle for 1 h, centrifuged at 5000 rpm, and then passed through a 0.22 mm filter. The lead concentration in solution was quantified using Inductively Coupled Plasma - Mass Spectrometry (ICP-MS) (Thermo Fisher XSERIES 2), and the soluble lead concentration was calculated. There is a total lead regulatory standard for paints used in toy manufacturing in China. Therefore, we measured the total lead content by using the method specified in GB 24613-2009 Limit of harmful substances of coatings for toys. The procedure involved: paint scrapings (0.2 g) were crushed and then placed in a polytetrafluoroethylene tube; 5 mL of 65% (m/m) HNO3 and 2 mL of 30% (m/m) H2O2 were added for microwave digestion; the tubes were heated to 180 C and held for 30 min, and then cooled down to room temperature. After digestion, the solution was diluted to 50 mL and passed through a 0.22 mm filter. The lead concentration of the filtrate was measured by ICP-MS, and the total lead concentration in paint was calculated based on mass balance. 2.4. Quality control and statistical analyses The effects of the matrix on the chemical analysis were investigated. Specifically, for the total lead concentrations in the paints, acid digestion without paint was performed as blank, and its influence was calculated as 0.45 mg/kg. Similarly, the influence of the matrix on soluble lead concentrations in the paints was tested as 0.52 mg/kg. All the presented data are subtracted by the concentrations of the matrix (blank). The limit of detection (LOD) of the ICP-MS for Pb is calculated as 0.07 mg/L by testing 10 blank samples. When testing, each

313

0% 0% 0% 0% 0% 0% 0% 0% 7.3 5.0 18.1 18.1 0.4 0.2 0.8 0.4 1.4 0.4 2.8 1.1 2.1 1.0 5.6 2.8 0% 0% 12% 4% 0% 0% 36% 10% 0% 0% 36% 12% 39 44 1677 1677 2 2 69 24 23 29 89 39 24 43 33 100

78 92 44 73 365600 186860 2313277 2982503 JD TM TB Overall

25 32 219 92

Soluble Lead (mg/kg)

% above 90 mg/kg Geometric Standard Max Mean (mg/kg) Error (mg/kg) (mg/ kg) Average (mg/kg)

Total Lead (mg/kg)

Total Sales Average Volume Price (￥) Platform Sample Size

Table 1 shows the total and soluble lead concentrations for paint samples on toys (n ¼ 100) from three online shopping platforms. The average total lead concentration was 92 mg/kg for the 100 paint samples, well below China's regulatory threshold of 600 mg/kg for paints used in toy manufacture (GAQSIQ, 2009) and the EU threshold of 130 mg/kg for migratable lead concentrations (EU Toy Safety Directive (2009/48/EC)), but slightly above the corresponding US regulatory threshold of 90 mg/kg (USCPSC, 2011). The maximum total lead concentration was found to be 1677 mg/kg, nearly three times the regulatory threshold in China. It should be noted that China has regulatory thresholds for both total lead and soluble lead for new paints used for toy manufacture (600 mg/kg and 90 mg/kg, respectively), but for the toy items themselves the regulatory threshold relates only to soluble lead (90 mg/kg). We found that the average soluble lead concentration was only 2.8 mg/kg for the 100 paint samples, well below China's regulatory limit. Overall, no toy paint samples exceeded the directly applicable Chinese lead standard (i.e. 90 mg/kg soluble lead), 4% of paint samples exceeded the indirectly applicable total lead standard (i.e. 600 mg/kg), 8% exceeded the EU standard for migratable lead (i.e. 130 mg/kg total lead), and 12% exceeded the more stringent US regulatory threshold (i.e. 90 mg/kg total lead). These selected toys purchased for this study have been sold nearly 3 million times to date on the three online shopping platforms. The average purchase price of the items was ￥73 RMB (~US$12). There are clear distinctions in the results between the two types of platforms: organized sellers JD (n ¼ 24) and TM (n ¼ 43) versus unorganized sellers on TB (n ¼ 33). The total sales volume of these listed toys was 0.36 million at JD and 0.19 million at TM, while the sales volume at TB was 2.31 million, well above the organized sellers. The average price of the toys purchased from TB was ￥44 RMB (~US$7). In comparison, the average price of the toys purchased from JD and TM was approximately twice as much, at ￥78 RMB (~US$12) and ￥92 RMB (~US$14), respectively. While the toys' price at TB tended to be lower, the lead concentrations of paints in these toys tended to be higher. The average of total lead concentrations of paint samples from toys purchased from TB, JD, and TM was 219 mg/kg, 25 mg/kg, and 32 mg/kg, respectively. As Fig. 1 shows, the majority of samples from TB had paint lead concentrations above most of the samples from JD (P < 0.01) and TM (P < 0.01) (Table S2). Moreover, four samples from TB contained extremely high total lead concentrations (>1000 mg/kg), resulting in a >67% exceedance of the 600 mg/kg threshold. It should be noted that all of these four toys were purchased from one seller on TB. Similarly, the soluble lead concentrations in toys’ paint from TB are significantly higher than those from JD (P < 0.05) and TM (P < 0.05) (Table S3).

Table 1 A comparison of total and soluble lead (mg/kg) in paint on toys from organized (JD and TM) and unorganized (TB) online shopping platforms in China.

3.1. Overall results and comparison between online shopping platforms

% above % above Average 130 mg/kg 600 mg/kg (mg/kg)

3. Results

Geometric Standard Max Mean (mg/kg) Error (mg/kg) (mg/ kg)

% above 90 mg/kg

% above % above 130 mg/kg 600 mg/kg

experimental sample was measured in triplicates by the ICP-MS and the average was reported. The value of the LOD corresponds to 1.75 mg/kg of total lead concentrations and 0.004 mg/kg of soluble lead concentrations in paint, suggesting that the tests were able to examine total and soluble lead concentrations in paint, even at very low values. The fitting of the data was performed using origin 8.5. The significances between groups were examined using LSD Duncan method at a significant level of 0.05 in SPSS20. Other statistical analyses (e.g., geometric mean, standard errors, calculation of the significance of fitting) were also conducted in SPSS20.

0% 0% 0% 0%


314


Fig. 1. Distribution of total lead and soluble lead in paint on toys from three platforms: a) total lead concentration (mg/kg) (vertical axis in logarithmic scale); b) soluble lead concentration (mg/kg).

3.2. Relationship between total lead and soluble lead concentrations In China, soluble lead concentration is used to evaluate the hazard level associated with the bioavailable fraction of lead that can be ingested into the body, and the testing method is different from that for total lead. The relationship between total lead and soluble lead concentration is not well understood. As Fig. 2 shows, we built a quantitative relationship between the determined total lead concentrations and soluble lead concentrations. It appears that the relationship between these two sets of concentrations was best described by a log-linear relationship:

y ¼ 2:8,lnðxÞ 7:5 where y represents the soluble lead concentration in mg/kg and x represents the total lead concentration in mg/kg. This suggests that, for toys' paint with lower total lead concentrations, slight increase

in total lead may result in significant rise in soluble lead concentrations. However, although the correlation is statistically significant (P < 0.01), the correlation coefficient (R2 ¼ 0.61) is relatively low, suggesting the log-linear relationship is relatively weak. Fig. 2 also shows that most of the total lead concentrations were typically an order of magnitude greater than the corresponding soluble lead concentrations. The total and soluble lead in toy's paint from each platform also showed a positive correlation (P < 0.01) (Table S4). Several samples contained total lead concentrations above 1000 mg/kg; however, their corresponding soluble lead concentrations were found to be only around 10 mg/kg, well below the regulatory threshold of 90 mg/kg. This finding differs from the results of Lin et al. (2009), who found that 90% of samples with >600 mg/kg total lead exceeded the 90 mg/kg soluble lead threshold (Lin et al., 2009) using the same testing methods. However, the assessment by Lin was based on a variety of paints (e.g. from furniture, walls, toys, and new paints) with very high lead concentrations, i.e. the maximum total lead concentration in their study was 153,000 mg/kg. Moreover, the study by Lin et al. (2009) was conducted almost a decade ago, and there have been several new regulatory standards established and mandated by the Chinese government since (GAQSIQ, 2003, 2009; 2014a; b). Therefore, the results from the study by Lin et al. (2009) may not be directly comparable to the present study. 3.3. Price

Fig. 2. Relationship between total lead concentration (mg/kg) and soluble lead concentration (mg/kg).

As Fig. 3 shows, a number of products sold at lower prices, i.e. less than ￥50 RMB (~US$8), had total lead concentrations >50 mg/ kg, and a significant number of items had total lead concentrations >90 mg/kg (i.e. above the US regulatory threshold) and 130 mg/kg (i.e. above the EU migratable standard). In comparison, nearly all toys priced above ￥50 RMB (~US$8) had lead concentrations 1000 mg/kg) were excluded as outliers, because those four toys were all from one supplier and their concentrations were similar for four different colors. As Table 3 shows, the average total lead concentration ranged from 28 mg/kg to 75 mg/kg. The highest lead concentrations were found for pink, blue and purple paint samples, at 75 mg/kg, 64 mg/kg and 63 mg/kg, respectively. The lowest average lead concentrations were found for orange, gray, and white paint samples, at 34 mg/kg, 31 mg/kg, and 28 mg/kg, respectively. A number of different color samples exceeded the US threshold of 90 mg/kg and EU migratable threshold of 130 mg/kg. Take the US threshold for example, the highest exceedance found for purple toys was 29%, followed by blue and pink toys (17%), black (9%), and yellow and green color toys (7%). The pattern for soluble lead concentrations was somewhat different, with the highest concentration found for blue toys (3.7 mg/kg), whereas the pink toys average soluble lead concentration was much lower (2.5 mg/kg), despite having highest average total lead concentration. It is of note that the differences between color groups for total or soluble lead concentrations in toys’ paint were not statistically significant generally (P > 0.05), except for soluble lead concentrations between yellow and blue color toys (P < 0.05) (Table S7 and S8).

Total lead concentrations in toys' paint (mg/kg)

1800 1200

China standard

600 330 300 270 240 210 180 150 120 90 60 30 0

EU standard US standard

0

10

20

30

40

315

50

Price (US dollars)

Fig. 3. Relationship between toy price (US dollars) and total lead concentrations (mg/ kg) in toys' paint.

(~US$1.6) increase in price (p ¼ 0.013); for toys purchased from JD, we found that total lead paint concentrations decreased by 1.1 mg/ kg for every ￥10 RMB (~US$1.6) increase in price (p ¼ 0.19); for toys purchased from TM, total lead concentrations decreased by 0.5 mg/kg for every ￥10 RMB (~US$1.6) increase in price (p ¼ 0.047); and, for toys purchased from TB, lead concentrations were found to decrease by 1.9 mg/kg for every ￥10 RMB (~US$1.6) increase in price (p ¼ 0.62). It is of note that all the relationships are generally statistically weak (R2 < 0.07).

3.6. Sales volume Table 4 lists the lead concentrations of toys as grouped by sales volume. Among all samples, 40 toys have been sold more than 10,000 times; 35 toys have been sold 1000 to 10,000 times; and 25 toys have been sold less than 1000 times. As can be seen, the group of least sold toys (i.e.