Potential mercury emissions from fluorescent lamps production and obsolescence in mainland China

Abstract

The use of fluorescent lamps has expanded rapidly all over the world in recent years, because of their energy-saving capability. Consequently, however, mercury emissions from production, breakage, and discard of the lamps are drawing increasing concern from the public. This article focuses on evaluating the amount of mercury used for fluorescent lamp production, as well as the potential mercury emissions during production and breakage, in mainland China. It is expected to provide a comprehensive understanding about the risks present in the mercury from fluorescent lamps, and to know about the impacts of the policies on fluorescent lamps after their implementation. It is estimated that, in 2020, mercury consumption will be about 11.30–15.69 tonnes, a significant reduction of 34.9%–37.4% from that used in 2013, owing to improvement in mercury dosing dosage technology and tighter limitations on mercury content in fluorescent lamps. With these improvements, the amount of mercury remaining in fluorescent lamps and released during production is estimated to be 10.71–14.86 and 0.59–0.83 tonnes, respectively; the mercury released from waste fluorescent lamps is estimated to be about 5.37–7.59 tonnes. Also, a significant reduction to the mercury emission can be expected when a collection and treatment system is well established and conducted in the future.

Keywords

Fluorescent lamps mercury consumption emission distribution

Introduction

The demand for lighting is an important part of electricity consumption; it accounts for about 12% of the total electricity consumption in China, 10% in Germany, and 19% in the United States (Frondel and Lohmann, 2011; National Development and Reform Commission et al., 2011; US Department of Energy, 2012). The advantages of the fluorescent lamp (FL) include energy-savings of 65% and a lifespan over ten times as long as that of the traditional incandescent lamp (IL) (US Department of Energy, 2013). Many countries, including Canada and the US, as well as the European Union countries (Waide, 2010), have been working on eliminating ILs by forbidding their import and sale, ever since Australia began phasing them out in 2007 (United Nations Environment Programme, 2014).

China also announced a national plan on 1 November 2011, and began the process of prohibiting ILs on 1 October 2012 (National Development and Reform Commission et al., 2011). China is now the largest FL producer in the world. In 2011, China produced 7.024 billion FL units, more than 28 times the number produced in 1994 (National Development and Reform Commission, 2002; Ye et al., 2012). An accelerated flow of obsolescent FLs into China’s solid waste stream can be expected during the coming years, owing to the gradual elimination of ILs and their replacement with FLs. Based on the authors’ previous study, it is estimated that the FL production in mainland China would reach up to 9.40 billion units in 2015, and the amount of waste FLs in the same year would be about 4.13 billion units; the production and waste FL generation in 2020 would be 11.90 and 5.84 billion units, respectively (Tan and Li, 2014) (Figure 1).

However, mercury, an indispensable component of FLs, is also a heavy metal with high toxicity. It converts the electron flow into ultraviolet radiation, after which the fluorescent phosphors coating inside the glass envelope of the lamp is excited to generate visible light via ultraviolet radiation (European Commission, 2013). Several milligrams of mercury are usually contained in a FL to make it work properly; the emission of mercury occurs not only in the production process, but also at the end-of-life stage of the FL.

According to a study conducted by Shao et al. (2012) in Zhejiang, China’s largest FL production province, the soils and vegetables from sites near compact fluorescent lamp (CFL) manufacturing plants showed higher mercury concentrations than samples from control sites, and the surrounding environment was significantly affected by the production processes. In addition, when FLs are broken, mercury is released into the atmosphere and biosphere, posing a risk to human health. It was demonstrated in a study by Nance et al. (2012) that in certain scenarios, the risk of mercury emission from CFL breakage in a closed room could exceed the relevant risk targets. Sarigiannis et al. (2012) also pointed out that the indoor air concentration of mercury vapour may exceed the reference exposure limit for mercury vapour set by the Environmental Protection Agency of California (2008).

Environmental benefits can be obtained from the recycling of metals and cullet (Tan et al., 2015b). At present, the technologies for waste FLs treatment can be divided into two approaches, the end-cutting/air-pushing and crushing/sieving (European Commission, 2006; Tan et al., 2015a); the latter approach can be categorised into dry and wet route. Glass and phosphors obtained in waste FLs treatment are heated under high temperature (more than 400 °C), to remove and recover the mercury (Chang et al., 2009; Jang et al., 2005; Raposo et al., 2003). The glass without mercury can be reused for FLs production. In the wet route of the crushing/sieving approach, FLs are immersed in an acid solution during the crushing and washing processes to avoid mercury releasing into atmosphere. The glass can also be reused for FL production after washing, while the mercury in the solution can be precipitated and then solidified and stabilised for landfilling. Researches also have been carried out to improve the performance of heating, washing, and recycling of mercury processes (Bussi et al., 2010; Durão Jr et al., 2008; Rey-Raap and Gallardo, 2013).

The FL industry has been one of the major centres of mercury consumption in China (China Council for International Cooperation on Environment and Development, 2011), and in recent years, increasing public concern has been devoted to the potential risk of mercury exposure caused by FL production, breakage, and disposal. In response to this concern, and as preparation for the upcoming Minamata Convention on mercury, the Chinese government issued a national plan to reduce the mercury content in FLs, on 18 February 2013 (Ministry of Industry and Information Technology et al., 2013), aiming to eliminate the obsolescent liquid mercury dosing technology by replacing it with amalgam technology.

Despite the actions noted above, the need to manage the FL production and FL waste along an environmentally sound path is becoming increasingly urgent in China, especially as there is no collection or disposal system established for the waste FL stream, most of which is entering the municipal solid waste treatment facilities, making it difficult to recover rare earth elements from their phosphors (Tan et al., 2015a). Based on the authors’ previous study of projected FL production, as well as waste FLs generation and distribution in mainland China, this article takes the further step of evaluating the potential amount of mercury used for FL production and the potential emissions from production, breakage, and disposal in mainland China. It is expected to provide a comprehensive understanding of the mercury risk from FLs in mainland China by quantified estimation, as well as what proportion of mercury consumption and emission can be reduced with the implementation of national plans.

Materials and methods

The time-varying extended logistic model proposed by Trappey and Wu (2007, 2008) was applied for predicting the production of FL from 2012 to 2020 on the basis of historical data of FL production from 1995 to 2011; while the market supply method was selected and modified for the estimation of waste FL generation during 2011–2020. An extended logistic model has the advantage of improving the accuracy of the market share forecast when compared with other growth curve methods, such as simple logistic curve and Gompertz curve (Bengisu and Nekhili, 2006; Meyer and Ausubel, 1999). The time-varying extended logistic model adopted in this study is adapted based on the extended logistic model, and has been successfully applied to the production prediction of 22 electronic products (Trappey and Wu, 2007, 2008). The market supply method is chosen between other e-waste generation prediction methods owing to the data availability of FL and waste flow in its end-of-life stage (Kim et al., 2013; Liu et al., 2006; Zhang et al., 2012). The formulas of these two methods can be found in the Supplementary Material equations (S1) and (S3).

As for the potential mercury consumption for FL production ( $Q_{c o n}$ ), the calculation method was presented in equation (1). The limits on mercury content in various types of FLs in each year were summarised ( $C_{i}$ ). The proportion of FL applying the solid amalgam and liquid mercury technology for mercury dosing during the production was investigated ( $P_{H g, f, i}$ ). Furthermore, the ratio of mercury that would be released during the dosing process ( $P_{r e l, f}$ ), which was related to the dosing technology, was investigated and confirmed by interviewing the FL manufacturers. Mercury content in FL varies obviously with the brand; there is also differences among FLs produced in different years. The threshold values in standards on mercury content in FLs are used for the estimation:

\begin{array}{l} Q_{c o n} = \sum_{i = 1, 2, 3} A_{r e m, i} + \sum_{i = 1, 2, 3} A_{r e l, i} \\ = \sum_{i = 1, 2, 3} N_{T, i} \times C_{i} + \sum_{i = 1, 2, 3} A_{r e m, i} \times \sum_{f = 1, 2} P_{H g, f, i} \times \frac{P_{r e l, f}}{1 - P_{r e l, f}} \\ = \sum_{i = 1, 2, 3} N_{T, i} \times C_{i} \times (1 + \sum_{f = 1, 2} P_{H g, f, i} \times \frac{P_{r e l, f}}{1 - P_{r e l, f}}) \end{array}

(1)

Where, i represents the type of FL, 1-compact, 2-linear, and 3-circular; f represents the form of mercury used for FL production, 1-solid and 2-liquid. $A_{r e m, i}$ and $A_{r e l, i}$ means the amount of mercury remaining and released out during FL (i) production. $N_{T, i}$ is the production of a specific type of FL (the same as it in Appendix equation (S5)).

As for the estimation of mercury emission from waste FLs ( $Q_{e m i}$ ), the studies focusing on the mercury emissions from waste FLs were reviewed to determine the percentage ( $P_{e m i}$ ) of mercury that could potentially be released after waste FLs breakage. $C_{i}$ represents the content of mercury in FL, and $Q_{i}$ is the quantity of waste FL generated in each year:

Q_{e m i} = P_{e m i} \sum_{i = 1, 2, 3} C_{i} \times Q_{i}

(2)

The values for different specifications of FL (a specific type) are used as the upper and lower limit of the estimations on mercury consumption and emission. The specification of each type of FL is classified according to the characteristic used by state standards (Figure 2): for a compact FL, the characteristic for classification is rated power − 30 W; for a linear and circular FL, it is the diameter of tube − 17 mm.

Results and discussion

Trends of mercury content in FLs

According to a report for the Natural Resources Defense Council, a significant amount of mercury content reduction has been achieved since the 1990s. Before the early 1990s, the mercury content in a T12 FL was slightly over 40 mg. It was reduced to about 30 mg, and then to slightly over 20 mg in the 1990s, and went a step further to about 10 mg in the 2000s. For the T8 FL, the mercury content was about 30 mg in the early 1990s, was reduced by about half in the late 1990s, and then to 8 mg in the early 2000s; for the CFLs, mercury content was less than in the linear FLs – under 8 mg in the early 2000s (Dunmire et al., 2003). In 2010, it was reported that the mercury content in linear FLs in the Chinese market was usually 5–10 mg, and a few enterprises were able to reach 3–4 mg or even better; it was 3–5 mg for compact FLs, with some containing less than 3 mg, or even 1 mg (Li et al., 2013).

Mercury content limits in FLs were set to 5 mg and 10 mg for compact and linear products, respectively, by the Chinese government in 2006; stricter requirements of 3 mg and 5 mg, respectively, were announced in July 2012 and enacted in October of the same year. A detailed roadmap for gradually reducing the mercury content in FLs even further was announced in February 2013, rising to the challenge of the increasingly stringent restriction of mercury usage in FLs from the recast European Directive on the restriction of the use of certain hazardous substances in electrical and electronic equipment and the approaching Minamata Convention on Mercury. The progress of requirements for content limits of mercury in FLs is illustrated in Figure 2.

A total of 26 pieces of FLs were bought from the market. They belonged to nine brands, with different specifications and produced in different years. The results of the tests conducted on them are given here. There were two FLs produced in 2011, their mercury content was especially higher, 8.8 and 9.5 mg, but still they met the requirements in 2011. The mercury content in the other 24 tested FLs also met the requirements in the standards of their production year. Their average mercury content was 2.38 mg per piece of FL, and their minimum content was 0.6 mg. The mercury content in FLs varied among the different brands, but all of them met the standards, and it indicated that using the threshold values in standards for the estimation could represent the reality.

FL production technology progress and mercury liberation potential

At present, solid amalgam and liquid mercury are the two forms being used for FL production in China. The solid amalgam mercury dosing technology causes less mercury to be released than the liquid mercury technology during production (Megaman, 2010), because of the highly volatile character of mercury. It was found that about 5.26% of the mercury would be lost during the solid amalgam mercury dosing process, while the average mercury loss percentage was 9.09% for the liquid mercury technology (Li et al., 2013). An interview with the Beijing manufacturer of a famous lighting products brand confirmed that the level of mercury lost during its production process ranged from 2.5% to 5% when the solid amalgam mercury was used, but there were no data available about the liquid mercury dosing technology. Thus, the values of average mercury loss percentage presented in the study by Li et al. (2013), 5.26% and 9.09%, were adopted for this evaluation. The proportion of FL products using the solid amalgam mercury has been increasing continuously in recent years, to about 77%, 83%, and 95% for compact, linear, and circular FLs, respectively, in 2010 (Li et al., 2013).

Studies have been carried out by several researchers evaluating the potential exposure risk from FL breakage. It was demonstrated that mercury in FLs would diffuse through the fluorescent phosphors during their lifetime. Although some differences exist between different studies of mercury distribution, more than 80% of the mercury in FLs is contained in the fluorescent phosphors, while over 10% of the mercury remains in the glass of FLs (Jang et al., 2005; Rey-Raap and Gallardo, 2012; Rhee et al., 2014), indicating that the mercury emissions from broken FLs would last for weeks or even months (Li and Jin, 2011). The largest proportion of mercury liberated to the atmosphere during experiments, under the ambient temperature and air circulation, was more than 75% in weeks. It was estimated that more than 95% of the mercury in broken FLs could be released as soon as in within 53 days, if no measures were taken to control its emissions (Aucott et al., 2003; Li and Jin, 2011).

Potential mercury consumption and emission evaluation

An evaluation of mercury consumption during FL production, and potential emission from waste FL breakages, was made based on the FL production, and waste FL generation and regional distribution obtained in the authors’ previous study (Tan and Li, 2014). Considering the pressure from stricter limits on mercury content, the authors made assumptions for the proportion of FLs using solid amalgam mercury technology, of 80%, 85%, and 96% for compact, linear, and circular FLs in 2012, respectively; moreover, it was assumed that the annual growth of these proportions would be 5%, 5%, and 2%, respectively, with a maximum value of 100%. The worst-case scenario for mercury emissions from waste FLs after breakage was set for the evaluation by assuming that all the mercury in waste FLs would eventually be released into the environment. Estimated mercury consumption of the FL industry and potential mercury emission from domestic waste FLs after breakage are illustrated in Figure 3.

It may be concluded that the mercury consumption by the FL industry and potential emissions from waste FLs could obviously be reduced by good implementation of policies and/or regulations on FL mercury content reduction. It is estimated that in 2015, the last year of the reduction roadmap for mercury content in FLs, the amount of mercury consumption by the FL industry could be reduced to 8.72–11.87 tonnes, of which 8.25–11.24 tonnes would remain in the FLs while about 0.47–0.64 tonnes would be released into the environment. The estimated amount of mercury consumption for FL production in 2015 would be only about 11.15%–15.18% of its 2007 level (78.2 tonnes) (China Council for International Cooperation on Environment and Development, 2011). The potential mercury emissions from waste FLs in 2015 is about 12.90–25.25 tonnes, with the average of 19.07 tonnes. This is approximately the same amount of atmospheric mercury emission from solid waste in 2007 (China Council for International Cooperation on Environment and Development, 2011). It would drop sharply to about 8.58–11.00 tonnes in 2017, about 43.56%–66.51% of its 2015 level, probably because of the limits on FL mercury content enacted in 2012.

Subsequently, mercury consumption would gradually increase with the growth of domestic FL production and use. In 2020, the estimated amount of mercury consumption for FL production could reach 11.30–15.69 tonnes, of which the amount of mercury remaining in FLs and released during production would be about 10.71–14.86 and 0.59–0.83 tonnes, respectively. The amount of mercury consumed would be about 14.45%–20.06%, of that in 2007. The emission could potentially reach 5.37–7.59 tonnes and would be just about 27.84%–39.35% of the atmospheric mercury emissions from solid waste in 2007 (China Council for International Cooperation on Environment and Development, 2011).

It was found that the environment surrounding FL factories was significantly affected by the mercury released during the FL production processes. Higher mercury concentrations were found in the soil and food from FL production sites than from control sites (Shao et al., 2012). Decreasing the mercury emissions during FL production, by promoting the technology substitution and mercury content requirements, promises to show appreciable environmental benefits. The amount of mercury lost during FL mercury production in 2010 (2.18 tonnes) could be decreased to an average of about 0.55 tonnes in 2015, a reduction of approximately 75%; in 2020, it would be about one-third of what it was in 2010. Figure 4 shows the distribution and trend of mercury loss during FL production in China. These eight provinces accounted for about 96% of the total amount of mercury loss. Zhejiang province, where the largest Chinese FL production base is located (Shao et al., 2012), released the most mercury during FL production, followed by Guangdong, Jiangsu, and Fujian provinces. The mercury released in Zhejiang province was about 0.81 tonnes in 2010, and it could be reduced to about 0.21 tonnes in 2015, and 0.27 tonnes in 2020.

The regional distribution of potential mercury emissions from waste FLs was calculated based on the regional distribution of waste FLs (Tan and Li, 2014). Guangdong was found to be the province with the highest potential mercury emission from waste FLs. The emission amount was estimated to be about 1,710 kg in 2015, and would drop to about 560 kg in 2020. Jiangsu, Shandong, Zhejiang, and Sichuan are the other four of the top five mercury emission provinces in mainland China. The specific amounts of potential mercury emissions from waste FLs in these five provinces are presented in Figure 5. The overall distribution of the potential mercury emissions in mainland China is exhibited in Figure 6. Tibet (abbreviated as ‘XZ’) has the least amount of mercury emissions; the amount in 2015 is expected to be only about 11.84 kg, or just about 0.7% of that in Guangdong in the same year.

Currently, the thermal desorption process for the decontamination and recovery of mercury in FLs is well and widely used (Chang et al., 2009; Durão Jr et al., 2008; Jang et al., 2005). Mercury-containing residues generated in the waste FL disposal process were distilled at 500 °C for 5 h, and the mercury content was reduced to below 1 mg kg⁻¹ (Jang et al., 2005). In an aggressive scenario, if a take-back system is well established and operated, under the most promising scenario that all the waste FLs are collected, the amount of mercury released from waste FLs could be reduced to less than 0.94 tonnes. Therefore, establishing a collection system and taking back waste FLs, as well as recovering the mercury, should improve the prevention and control of mercury pollution from waste FLs significantly, and promote the recycling of waste glass, rare earth elements, and other metallic components, which would also contribute to environmentally sound management of waste FLs.

However, as of August 2015, there are only five licensed hazardous waste operation facilities covering the category of mercury-containing lamps in China. The facility in Jiangsu province, using a wet treatment approach, has the greatest capacity at 5000 tonnes per year, while the other four facilities are located in Beijing, Fujian Guangdong, and Zhejiang, respectively, which take a thermal approach. The capacity of the five facilities is 13,500 tonnes, which can process merely 2.2% of the estimated mass of waste FLs that will be generated in mainland China in 2015. Even including the facility in the phase of trial production in Shanghai (with a capacity of about 1700 tonnes), the ratio of capacity/demand is only 2.5% for the same year.

In Jiangsu and Beijing, for example, the facility in each region can only cover about 10.1% and 6.2%, respectively, of the regional waste FL generation in 2012. However, the actual amount of waste FLs treated in these two facilities are only about 1200 and 350 tonnes, about 24% and 23% of their capacities, respectively. Therefore, establishing a collection system for waste FLs is the key to environmentally sound management. The prevention and control of mercury pollution, as well as the relevant risk from waste FLs, should be significantly improved with good implementation of waste FL take-back and mercury recovery.

Conclusion

The potential health risk caused by mercury exposure from FL production and the management of waste FLs is attracting increasing attention and concern from the public. In order to gain a realistic understanding of the issue of FL mercury emissions, and to contribute to the establishment of a management system for waste FLs, we have evaluated the amount of future mercury consumption and release during FL production, as well as emissions from waste FLs.

It is estimated that about 11.30–15.69 tonnes of mercury would be consumed in FL production in 2020, of which 10.71–14.86 tonnes would remain in the FLs, while about 0.59–0.83 tonnes would be released into the environment. The amount consumed drops sharply with the implementation of stricter limitations on FL mercury content. There is a significant estimated reduction in 2020 of 34.9%–37.4% when compared with 2013 (17.35–25.05 tonnes). This is equal to only about 14.45%–20.06% of the consumption amount for FL production in 2007.

Meanwhile, the implementation of reduced FL mercury limits for production can also reduce the corresponding mercury emissions from waste FLs. For the end-of-life stage of FLs, it is estimated that about 19.07 tonnes of mercury could be released from waste FLs discarded in 2015, which will be reduced to about 6.48 tonnes in 2020, and about 33.59% of the atmospheric mercury emissions from solid waste in 2007.

As for the regional mercury emissions and releases from FL production and waste FLs, Zhejiang and Guangdong merit the most attention, as Zhejiang is the province with the greatest release of mercury during FL production, and Guangdong province has the highest amount of potential mercury emissions from waste FLs.

Establishing collection and take-back waste FL systems, and recovering the contained mercury, can contribute significantly to the prevention and control of relevant mercury pollution. The amount of mercury emissions is estimated to decrease to below 0.94 tonnes with well-conducted collection and recovery systems for most of the waste FLs.

Footnotes

Acknowledgements

The authors thank Brenda Lopez for reviewing the grammar of the manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by the National Key Technologies R&D Program [NO. 2014BAC03B04] and Environmental Protection Public Welfare Project [2011467035].

References

Aucott

McLinden

Winka

(2003) Release of mercury from broken fluorescent bulbs. Journal of the Air & Waste Management Association 53: 143–151.

Bengisu

Nekhili

(2006) Forecasting emerging technologies with the aid of science and technology databases. Technological Forecasting and Social Change 73: 835–844.

Bussi

Cabrera

Chiazzaro

. (2010) The recovery and recycling of mercury from fluorescent lamps using photocatalytic techniques. Journal of Chemical Technology and Biotechnology 85: 478–484.

California Environmental Protection Agency (2008) Mercury reference exposure levels, pp.476–501. Available at: http://oehha.ca.gov/air/hot_spots/2008/AppendixD1_final.pdf#page=476 (accessed 24 October 2013).

Chang

You

. (2009) Treating high-mercury-containing lamps using full-scale thermal desorption technology. Journal of Hazardous Materials 162: 967–972.

China Council for International Cooperation on Environment and Development (2011) Final report: Special policy study on mercury management in China. Beijing, China: China Council for International Cooperation on Environment and Development.

China National Institute of Standardization (2002) Technical supporting report for China energy efficiency standard of double-capped fluorescent lamps. Beijing, China: China National Institute of Standardization.

Dunmire

Calwell

Jacob

. (2003). Mercury in fluorescent lamps: Environmental consequences and policy implications for NRDC. New York, NY: Natural Resources Defense Council.

Durão

Jr de Castro

Windmöller

(2008) Mercury reduction studies to facilitate the thermal decontamination of phosphor powder residues from spent fluorescent lamps. Waste Management 28: 2311–2319.

10.

European Commission (2006) Integrated pollution prevention and control – reference document on best available techniques for the waste treatments industries. Brussels, Belgium: European Commission.

11.

European Commission (2013) Energy-saving lamps & health. Available at: http://ec.europa.eu/health/opinions/en/energy-saving-lamps/l-3/3-cfl-characteristics.htm (accessed 17 October 2013).

12.

Frondel

Lohmann

(2011) The European Commission’s light bulb decree: Another costly regulation? Energy Policy 39: 3177–3181.

13.

Jang

Hong

Park

(2005) Characterization and recovery of mercury from spent fluorescent lamps. Waste Management 25: 5–14.

14.

Kim

Oguchi

Yoshida

. (2013) Estimating the amount of WEEE generated in South Korea by using the population balance model. Waste Management 33: 474–483.

15.

Wang

Zhang

. (2013) Analysis on reduction trend of mercury releases from Chinese light source production industry for execution the global legally binding instrument on mercury. China Illuminating Engineering Journal 24: 23–26, 30.

16.

Jin

(2011) Environmental release of mercury from broken compact fluorescent lamps. Environmental Engineering Science 28: 687–691.

17.

Liu

Tanaka

Matsui

(2006) Generation amount prediction and material flow analysis of electronic waste: A case study in Beijing, China. Waste Management & Research 24: 434–445.

18.

Megaman (2010) Compact fluorescent lamps and the dangers of liquid mercury: A Megaman White Paper.

19.

Meyer

Ausubel

(1999) Carrying capacity: A model with logistically varying limits. Technological Forecasting and Social Change 61: 209–214.

20.

Ministry of Industry and Information Technology, Ministry of Science and Technology and Ministry of Environmental Protection (2013) Roadmap on reducing mercury content in fluorescent lamps in China. Beijing, China.

21.

Nance

Patterson

Willis

. (2012) Human health risks from mercury exposure from broken compact fluorescent lamps (CFLS). Regulatory Toxicology and Pharmacology 62: 542–552.

22.

National Development and Reform Commission, Ministry of Commerce, General Administration of Customs, State Administration for Industry & Commerce and General Administration of Quality Supervision, Inspection and Quarantine (2011) Roadmap on phasing-out of incandescent lamps in China. Beijing, China.

23.

Raposo

Windmoller

Junior

WAD

(2003) Mercury speciation in fluorescent lamps by thermal release analysis. Waste Management 23: 879–886.

24.

Rey-Raap

Gallardo

(2012) Determination of mercury distribution inside spent compact fluorescent lamps by atomic absorption spectrometry. Waste Management 32: 944–948.

25.

Rey-Raap

Gallardo

(2013) Removal of mercury bonded in residual glass from spent fluorescent lamps. Journal of Environmental Management 115: 175–178.

26.

Rhee

S-W

Choi

H-H

Park

H-S

(2014) Characteristics of mercury emission from linear type of spent fluorescent lamp. Waste Management 34: 1066–1071. DOI: 10.1016/j.wasman.2013.07.029.

27.

Sarigiannis

Karakitsios

Antonakopoulou

. (2012) Exposure analysis of accidental release of mercury from compact fluorescent lamps (CFLS). Science of the Total Environment 435–436: 306–315. DOI: 10.1016/j.scitotenv.2012.07.026.

28.

Shao

Liang

. (2012) A human health risk assessment of mercury species in soil and food around compact fluorescent lamp factories in Zhejiang Province, PR China. Journal of Hazardous Materials 221–222: 28–34. DOI: 10.1016/j.jhazmat.2012.03.061.

29.

Tan

(2014) A study of waste fluorescent lamp generation in mainland China. Journal of Cleaner Production 81: 227–233.

30.

Tan

Zeng

(2015a) Rare earth elements recovery from waste fluorescent lamps: A review. Critical Reviews in Environmental Science and Technology 45: 749–776.

31.

Tan

Song

(2015b) The environmental performance of fluorescent lamps in China, assessed with the LCA method. The International Journal of Life Cycle Assessment 20: 807–818.

32.

Trappey

H-Y

(2007) An evaluation of the extended logistic, simple logistic, and gompertz models for forecasting short lifecycle products and services. In: Loureiro

Curran

(eds) Complex Systems Concurrent Engineering. London: Springer, pp.793–800.

33.

Trappey

H-Y

(2008) An evaluation of the time-varying extended logistic, simple logistic, and Gompertz models for forecasting short product lifecycles. Advanced Engineering Informatics 22: 421–430.

34.

US Department of Energy (DOE) (2012) 2010 US Lighting market characterization.

35.

US Department of Energy (DOE) (2013) Fluorescent lighting. Available at: http://energy.gov/energysaver/articles/fluorescent-lighting (accessed 19 October 2013).

36.

United Nations Environment Programme (UNEP) (2014) Global efficient lighting forum convened to address accelerated transition to advanced lighting technologies. Available at: http://www.unep.org/newscentre/Default.aspx?DocumentID=2812&ArticleID=11067&l=en (accessed 3 January 2015).

37.

Waide

(2010) Phase out of incandescent lamps: Implications for international supply and demand for regulatory compliant lamps. IEA Energy Papers, No. 2010/05, OECD Publishing.

38.

Wang

Chen

(2012) Situation of discharge lamp production and market in 2011. China Light & Lighting 18: 1–5.

39.

Zhang

Yuan

. (2012) Estimating future generation of obsolete household appliances in China. Waste Management & Research 30: 1160–1168.