Chamber tests are usually used to determine the source characteristics of semi-volatile organic compounds (SVOCs) which are critical to quantify indoor exposure to SVOCs. In contrast to volatile organic compounds (VOCs), the sorption effect of SVOCs to chamber surfaces usually needs to be considered due to the much higher surface/air partition coefficients, resulting in a long time to reach steady state, frequently on the order of months, and complicating the mathematical analysis of the resulting data. A chamber test is also complicated if the material-phase concentration is not constant. This study shows how to design a chamber to overcome these limitations. A dimensionless mass transfer analysis is used to specify conditions for (1) neglecting the SVOC sorption effect to chamber surfaces, (2) neglecting the convective mass transfer resistance at sorption surfaces if the sorption effect cannot be neglected, and (3) regarding the material-phase concentration in the source as constant. Several practical and quantifiable ways to improve chamber design are proposed. The approach is illustrated by analyzing available data from three different chambers in terms of the accuracy with which the model parameters can be determined and the time needed to conduct the chamber test. The results should greatly facilitate the design of chambers to characterize SVOC emissions and the resulting exposure.
Citation: Liu C, Liu Z, Little JC, Zhang Y (2013) Convenient, Rapid and Accurate Measurement of SVOC Emission Characteristics in Experimental Chambers. PLoS ONE 8(8): e72445. https://doi.org/10.1371/journal.pone.0072445
Editor: Qinghua Sun, The Ohio State University, United States of America
Received: January 19, 2013; Accepted: July 9, 2013; Published: August 28, 2013
Copyright: © 2013 Liu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was sponsored by the Natural Science Foundation of China (Grant No. 51136002 and 51076079, http://www.nsfc.gov.cn), the National High Technology Research and Development Program 863 (No. 2010AA064903, http://www.most.gov.cn/), the Tsinghua University Initiative Scientific Research Program (Grant No. 20121088010, http://www.tsinghua.edu.cn) and by the National Science Foundation of the United States (Grant No. CBET-1066642, www.epa.gov). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Semi-volatile organic compounds (SVOCs) are ubiquitous in indoor environments, with a significant number present as additives (e.g., plasticizers and flame retardants) in many indoor materials and products –. Due to their extremely low vapor pressure, SVOCs emitted from indoor sources readily partition to indoor media, including interior surfaces –, airborne particles  and settled dust . Human exposure to certain SVOCs is associated with adverse health effects, including asthma, allergies, bronchial obstruction –, reproductive disorders – and endocrine disruption  and partitioning to indoor media plays an important role in determining the dominant route(s) of exposure .
To quantify indoor exposure to SVOCs in specific products, mass balance models are developed that predict indoor SVOC emissions, transport and intake . Small chambers (e.g. Chamber for Laboratory Investigations of Materials, Pollution, and Air Quality (CLIMPAQ), Field and Laboratory Emission Cell (FLEC) and a sandwich-like chamber) are used to determine source parameters – and build confidence in the resulting models . In contrast to VOCs –, the sorption of SVOCs to chamber surfaces needs to be considered due to the much higher surface/air partition coefficients . A strong sink effect results in a long time to reach steady state, frequently on the order of months, and complicates the mathematical analysis of the resulting data.
Depending on the design of the chamber, and the volatility of the SVOCs, there may be conditions for which the sink effect can be neglected. This would simplify the chamber experiments to determine the source characteristics in terms of experiment duration and mathematical analysis of data. If the sink effect must be included, neglecting the convective mass transfer resistance at sorption surfaces under suitable conditions can simplify the mathematical analysis of the experimental data. Finally, chamber studies can be greatly simplified in terms of both experimental duration and mathematical analysis if the material-phase SVOC concentration in the source can be regarded as constant. To improve the efficiency and effectiveness of chamber studies, the conditions under which these three assumptions (neglecting the sink effect, neglecting the convective mass transfer resistance at sorption surfaces, and regarding the material-phase SVOC concentration as constant) are valid need to be identified.
The objectives of this study are therefore to: (1) present a mass transfer analysis to describe the behavior of SVOCs in chambers and identify essential dimensionless parameters; and (2) determine the conditions for which the sink effect and convective mass transfer resistance at sorption surfaces may be neglected, and under which the material-phase SVOC concentration may be considered constant. The analysis is illustrated by analyzing available data from three very different chamber studies in terms of the accuracy with which the model parameters can be determined and the time needed to conduct the chamber test. The results can help quantify improvements in chamber design, facilitating the use of chambers to characterize SVOC emissions and exposure.
Description of the Problem
The concentration of SVOCs used as additives in the source materials can usually be regarded as constant since they constitute 10–40 wt% of the source material –. SVOCs emit slowly from the source materials due to their high molecular weight and low vapor pressure . The emission rate of an SVOC source in a chamber, E (µg/s), is given by:(1)(2)
Similarly, the sorption rate at an interior sink surface, S (µg/s), is:(3)and the gas phase SVOC concentration in a particle-free environment is given by:(4)where hm,so (m/s) is the mass transfer coefficient at the emission surface, Ae (m2) the emission surface area, y0 (µg/m3) the concentration of the SVOC in the air immediately adjacent to the surface, C (µg/m3) the gas-phase SVOC concentration in the chamber, Cm0 (µg/m3) the material-phase SVOC concentration, which is assumed to be constant, K a parameter describing the equilibrium between the material and the air, hm,si (m/s) the mass transfer coefficient at the sink surface, As (m2) the sink surface area, Cs (µg/m2) the SVOC concentration on the sink surface, Ks (m) the SVOC partition coefficient between the sink surface and air, V (m3) the volume of air in the chamber, and Q (m3/s) the ventilation rate.
The analytical solution to equations (1)–(5) is:(6)where X and Y are given by:(7)and α and β are the roots of the following set of equations:(8)with a and b given by the roots of the following equation:
(9)When the sink effect is neglected, equation (3) simplifies to(10)
When convective mass transfer resistance at sorption surfaces is neglected, or hm,si becomes infinite, equation (3) simplifies to(12)
It should be noted that the model excludes airborne particles, a condition which can easily be achieved in laboratory experiments as clean air without particles is usually used. When assessing exposure to SVOCs in a realistic environment, airborne particles can play an important role  and the model needs to be adjusted to include them.
Dimensionless analysis of the source characteristics of VOCs in building materials was first completed by Xu and Zhang . Using the dimensionless formulae, results obtained under one condition can be scaled to another condition, as long as the dimensionless parameters are the same. In addition, after being normalized and made dimensionless, the number of variables can be reduced. The dimensionless parameters for this study are defined as:
The dimensionless model development is presented in Supporting Information S1. Four key dimensionless parameters, instead of the original eight variables, can describe the SVOC concentration and the SVOC source/sink emission/sorption rate. The physical meaning of the four key dimensionless parameters is: T* is the dimensionless time, Ks* is the dimensionless sorption capacity of sink surfaces, while Hm,so* and Hm,si* are the dimensionless mass transfer coefficients for the SVOC source and sink surfaces, respectively.
Results and Discussion
The Conditions for Neglecting the Sink Effect in Chambers
If a chamber is designed to neglect the sink effect, the time to reach steady state can be significantly shortened, and the analysis to determine C* (and consequently y0) can be simplified (equation (S13), Supporting Information S1), compared to when it cannot be neglected (equations (S5)-(S8), Supporting Information S1).
Equation (14) indicates that the influence of neglecting the sink effect on the determination of y0 can be examined by analyzing the influence on C*. As a result, a relative error of C*, εC*, is defined as the difference in gas-phase concentration between that obtained from equations (S5)-(S8) (C1*, with the sink effect) and that obtained from equation (S13) (C2*, without the sink effect), or,(15)
(a) the mass transfer strength; (b) the sorption strength.
Figure 2 reveals that:
- A strong mass transfer resistance at a sorption surface (e.g., Hm,si* = 0.1), or a weak sorption surface (e.g., Ks* = 1), both mean that the sink effect can be neglected. This is either because the strong mass transfer resistance keeps the SVOC from reaching the sorption surface, or because the surface has a weak affinity for the SVOC;
- Sorption to the sink has little influence on C* in the initial stage, but when T*>1, the influence of the sink effect becomes significant. As the gas-phase concentration comes to steady state, the rate of SVOCs sorbing to a sink surface decreases, and the influence of the sink effect becomes less important as the sink approaches equilibrium with the gas-phase concentration;
- For a specific scenario (with given values of Hm,so*, Hm,si* and Ks*), there is a critical time, designated as Tc*, after which the influence of the sink effect on C* can be neglected (εC* is smaller than 10%).
Calculating εC* along with T* identifies the critical time (Tc*) with εC* at that time and beyond being smaller than 10%. A convenient formula (equation (S16), Supporting Information S1) for Tc* is obtained for the range of Hm,so*, Hm,si* and Ks* of 0.1–10, 0.1–100 and 0–106, respectively. A particular case of Ks* = 104 is illustrated in Figure 3.
As Tc* is linearly proportional to Ks* when Ks* is higher than 0.1, it is easy to calculate Tc* for other values of Ks* using this figure.
Equation (S16) indicates that there are three simple and practical ways to design a chamber to neglect the sink effect from the beginning (i.e. Tc* = 0): 1) increase source surface area (i.e. increase Hm,so*); 2) decrease the sink surface area (i.e. decrease Hm,si*); 3) use material with a lower sorptive capacity for chamber surfaces (i.e. decrease Ks*). Other ways are to deliberately design the chamber shape and airflow pattern to decrease the mass transfer coefficient at sink surfaces (i.e. decrease Hm,si*), and/or to increase the mass transfer coefficient at source surfaces (i.e. increase Hm,so*). An additional practical solution is to increase the ventilation rate such that Hm,si* is smaller than 0.11, meaning that Hm,si* is smaller than Hm,so* +0.11 (see equation (S16)).
The Condition for Neglecting the Convective Mass Transfer at Sink Surfaces in Chambers
If a chamber is designed to neglect the mass transfer resistance at sorption surfaces, without neglecting the sink effect, the determination of C* (and consequently y0) can also be greatly simplified (equation (S15)) compared to when the mass transfer resistance cannot be neglected (equations (S5)-(S8)).
Similar to εC*, εC,s* is defined as the difference in gas-phase concentration between that obtained from (S5)–(S8) (C1*, including convection resistance at sorption surfaces) and that obtained from (S15) (C3*, excluding convection resistance at sorption surfaces), or,(16)
Figure 4 shows the influence of Hm,si* on εC,s*.
While T* increases, εC,s* would gradually decrease to zero in a general sense, but with a small increase of less than 10% for the case of Hm,si* = 1.0 and 10 at T* of about 104 (see the insert graph in Figure 4). For a specific scenario (with given values of Hm,so*, Hm,si* and Ks*), there is a critical time, designated as Tc,s*, after which the influence of convective mass transfer resistance at sink surface (Hm,si*) on gas-phase SVOC concentration (C*) can be neglected (εC,s* is smaller than 10%).
Calculating εC,s* along with T* identifies the critical time (Tc,s*) with εC,s* at that time and beyond being smaller than 10%. Results are shown in Figure 5, providing the condition for neglecting convective mass transfer resistance at sink surfaces. To make the determination of Tc,s* convenient, a formula (equation (S17), Supporting Information S1) for Tc,s* (εC,s* = 10%) is obtained for the range of Hm,so*, Hm,si* and Ks* of 0.1–10, 0.1–100 and 10–106, respectively.
As Tc,s* is linearly proportional to Ks*, it is easy to calculate Tc,s* for other values of Ks* using this figure.
Equation (S17) can be used to improve chamber design when the sink effect cannot be neglected. The practical methods presented in Section 4.1 are again applicable, except those for the mass transfer coefficient at sink surfaces and the ventilation rate. Increasing the mass transfer coefficient at sink surfaces is favored while the influence of ventilation rate is complicated and needs to be quantitatively evaluated by equation (S17).
The Condition for Considering Material-phase SVOC Concentration as Constant in Chambers
As mentioned before, if a chamber is designed to be able to regard the material-phase SVOC concentration in the source as constant, the test can be greatly simplified. This has usually been the case in previous studies to investigate the emission mechanism of SVOCs , . The dimensionless model presented here can help determine the general condition under which the assumption holds.
A dimensionless parameter, Vm*, is introduced:(17)where Vm (m3) is the volume of SVOC source material and Vm* is the dimensionless air volume needed to deplete the source material. Based on the previous results ,  Vm* is on the order of 1010–1011. If the emitted mass of SVOC is less than 10% of the initial mass before a chamber test ends, the material-phase SVOC concentration can be considered constant. A critical time, designated as Tc,m*, therefore exists prior to which the material-phase SVOC concentration can be considered constant. For the following range of parameters: Hm,so*: 100–102; Hm,si*: 10−1–102; Ks*: 103–107, Vm*: 109–1012, a formula to determine Tc,m* is obtained by correlating Tc,m* with the other parameters:
(18)Thus, neither Hm,si* nor Ks* has any influence on Tc,m*. By the time of Tc,m*, most of the emitted SVOC is not sorbed to sink surfaces, but has been exhausted from the chamber. Reducing the ventilation rate and increasing the surface area of the source are simple practical ways to lengthen the period during which the material-phase SVOC concentration can be considered constant. Equation (18) can help quantify practical ways to improve the chamber design.
To characterize exposure pathways and assess and control exposure and risk, accurate determination of y0 is necessary. Based on the preceding analysis, an ideal chamber to determine source characteristic parameters is one which has no sink at all. But such an ideal chamber is not easy to achieve in practice. Currently, there are three chambers available in the literature to determine y0: CLIMPAQ , , FLEC  and the specially-designed or “sandwich” chamber , as shown in Figure 6. Detailed description of the experiment work has been provided previously , , and is only briefly reviewed here, with key information listed in Tables 1 and 2. Clean air is passed through the chambers with di-ethylhexyl phthalate (DEHP)-containing source material at room temperature. The air in the outlet is monitored with sorption tubes. The tubes are then thermally desorbed and analyzed with gas chromatography-mass spectroscopy to obtain the evolution of concentration as a function of time. The key parameters (e.g. y0) can be obtained by fitting the models to the evolving gas-phase concentration. In this section, the preceding analysis is illustrated by examining if the sink effect and the convective mass transfer resistance at sorption surfaces can be reasonably neglected for these three chamber studies, and if the material-phase SVOC concentration can be regarded as constant.
The Sink Effect
The test conditions for the three chambers are listed in Table 1. The parameters involved can be obtained by fitting the experimental data with equations (6)-(9). Then the dimensionless parameters can be calculated, as shown in Table 2. Based on equation (S16), the critical dimensional time to neglect the sink effect can be determined, as shown by the red solid line in Figure 7. The experimental data obtained after the critical time is close to the steady-state concentration. Significant uncertainty will be introduced if all the data are used to determine the characteristic parameters by fitting with equation (11). The sink effect for these three chambers cannot be neglected when determining the characteristic parameters. Errors might be smaller if only the data obtained after the critical time are used, but it takes such a long time (see the red solid line in Figure 7) to reach the critical time that the cost of the chamber tests will be significantly increased.
The Convective Mass Transfer Resistance at Sorption Surfaces
Using equation (S17) and values of parameters listed in Table 2, the critical dimensional time to neglect the convective mass transfer resistance at sorption surfaces for CLIMPAQ, FLEC and the sandwich chamber is determined and shown by the red dashed lines in Figure 7. As shown in Figure 7, most experimental data from each chamber was collected after the critical time, indicating that the condition of neglecting convective mass transfer resistance at sorption surfaces is satisfied for most data. Thus neglecting convective mass transfer resistance at sorption surfaces has insignificant influence on determining y0. We calculated y0 by fitting the experimental data with equations (6)–(9) and equation (13), respectively, with results listed in Table 3 showing no significant difference, confirming the preceding analysis.
An effort to accurately determine hm,si is therefore not needed for these three chambers. In addition, equation (13) can be used to fit the experimental data in a much simpler way, compared to using equations (6)-(9). Furthermore, for the sandwich chamber, the condition of neglecting convective mass transfer resistance at sorption surfaces is satisfied much earlier than for the other two. Therefore, this chamber is more time-efficient. The effectiveness of the sandwich chamber can be explained by the fact that it has the smallest Ks* and the highest Hm,so*, which reduces the critical time to neglect the convection resistance, as shown in Figure 5.
The Material-phase SVOC Concentration
The dimensional critical time, tc,m, before which the material-phase SVOC concentration can be treated as a constant, is determined by combining equation (18) and the definition of T*, with results listed in Table 4. The critical time is much longer than the duration of the three chamber runs, as indicated in Figure 7. Therefore, it is reasonable to assume that the material-phase DEHP concentration is constant in all three cases.
This work provided a dimensionless analysis of indoor SVOC source characteristics. Several practical and quantifiable ways to improve chamber design are identified. They can help to shorten the time needed for an experimental test and/or simplify the mathematical analysis of the experimental data. Specifically, it was found that:
- The dimensionless gas-phase SVOC concentration in indoor air, C*, is a function of four dimensionless parameters - T* (the dimensionless time), Ks* (the dimensionless sorption capacity of sink surfaces), Hm,so* and Hm,si* (the dimensionless mass transfer coefficients for the SVOC source and sink surfaces, respectively).
- The applicable conditions of neglecting the sink effect and convective mass transfer resistance at sorption surfaces, and considering material-phase SVOC concentration as constant are: there is a critical time designated as Tc*, such that when T*>Tc*, the sink effect can be neglected; a critical time designated as Tc,s*, such that when T*>Tc,s*, the convective mass transfer resistance at sorption surfaces can be neglected; and a critical time designated as Tc,m*, such that when T*<Tc,m*, the material-phase SVOC concentration can be considered constant. Equations (S16), (S17) and (18) can be used to determine Tc*, Tc,s* and Tc,m*, respectively.
- Several simple practical ways have been proposed to either neglect the sink effect, or the convection at sorption surfaces, or consider the material-phase SVOC concentration as constant. Increasing the surface area of the source is the common solution for all three purposes. One advantage of the present study is that the effect of the practical ways to improve chamber design can be quantified using equations (S16), (S17) and (18).
- Three different chambers used to examine SVOC source characteristics, CLIMPAQ, FLEC and the specially-designed chamber, were evaluated based on the preceding analysis. It was found that for these three methods, the sink effect cannot be neglected, while neglecting convective mass transfer resistance at sorption surfaces and considering the material-phase concentration as constant are reasonable to determine y0. Considering the time to satisfy the condition, the specially-designed chamber method is more time-efficient (2.9 days) than FLEC (22 days) and CLIMPAQ (47 days). This is due to the fact that the method has the smallest Ks* and the highest Hm,so*.
Analyzed the data: CL YPZ JCL ZL. Wrote the paper: CL YPZ JCL. Conceived the idea of the study: ZL.
- 1. Wang L, Zhao B, Liu C, Lin H, Yang X, et al. (2010) Indoor SVOC pollution in China: A review. Chin Sci Bull 55: 1469–1478.
- 2. Wensing M, Uhde E, Salthammer T (2005) Plastics additives in the indoor environment - flame retardants and plasticizers. Sci Total Environ 339: 19–40.
- 3. Weschler CJ, Nazaroff WW (2008) Semivolatile organic compounds in indoor environments. Atmos Environ 42: 9018–9040.
- 4. Weschler CJ (2009) Changes in indoor pollutants since the 1950s. Atmos Environ 43: 153–169.
- 5. Diamond ML, Gingrich SE, Fertuck K, McCarry BE, Stern GA, et al. (2000) Evidence for organic film on an impervious urban surface: Characterization and potential teratogenic effects. Environ Sci Technol 34: 2900–2908.
- 6. Xu Y, Hubal EAC, Clausen PA, Little JC (2009) Predicting Residential Exposure to Phthalate Plasticizer Emitted from Vinyl Flooring: A Mechanistic Analysis. Environ Sci Technol 43: 2374–2380.
- 7. Junge CE (1977) Basic considerations about trace constituents in the atmosphere as related to the fate of global pollutants. Suffet, IH (Ed), Fate of Pollutants in the Air and Water Environments. New York: Wiley Interscience. 7–25.
- 8. Weschler CJ, Nazaroff WW (2010) SVOC partitioning between the gas phase and settled dust indoors. Atmos Environ 44: 3609–3620.
- 9. Bornehag CG, Nanberg E (2010) Phthalate exposure and asthma in children. Int J Androl 33: 333–345.
- 10. Jaakkola JJK, Parise H, Kislitsin V, Lebedeva NI, Spengler JD (2004) Asthma, wheezing, and allergies in Russian schoolchildren in relation to new surface materials in the home. Am J Public Health 94: 560–562.
- 11. Swan SH, Main KM, Liu F, Stewart SL, Kruse RL, et al. (2005) Decrease in anogenital distance among male infants with prenatal phthalate exposure. Environ Health Perspect 113: 1056–1061.
- 12. Matsumoto M, Hirata-Koizumi M, Ema M (2008) Potential adverse effects of phthalic acid esters on human health: A review of recent studies on reproduction. Regul Toxicol Pharm 50: 37–49.
- 13. Rudel RA, Perovich LJ (2009) Endocrine disrupting chemicals in indoor and outdoor air. Atmos Environ 43: 170–181.
- 14. Little JC, Weschler CJ, Nazaroff WW, Liu Z, Cohen Hubal EA (2012) Rapid Methods to Estimate Potential Exposure to Semivolatile Organic Compounds in the Indoor Environment. Environ Sci Technol 46: 11171–11178.
- 15. Xu Y, Hubal EAC, Little JC (2010) Predicting Residential Exposure to Phthalate Plasticizer Emitted from Vinyl Flooring: Sensitivity, Uncertainty, and Implications for Biomonitoring. Environ Health Perspect 118.
- 16. Xu Y, Little JC (2006) Predicting emissions of SVOCs from polymeric materials and their interaction with airborne particles. Environ Sci Technol 40: 456–461.
- 17. Clausen PA, Liu Z, Xu Y, Kofoed-Sorensen V, Little JC (2010) Influence of air flow rate on emission of DEHP from vinyl flooring in the emission cell FLEC: Measurements and CFD simulation. Atmos Environ 44: 2760–2766.
- 18. Liu Z, Xu Y, Little JC (2011) Characterizing emissions of di-2-ethylhexyl phthalate from vinyl flooring in a specially-designed chamber; Austin. pp. Paper No.: 863.
- 19. Xu Y, Liu Z, Park J, Clausen PA, Benning JL, et al. (2012) Measuring and Predicting the Emission Rate of Phthalate Plasticizer from Vinyl Flooring in a Specially-Designed Chamber. Environ Sci Technol 46: 12534–12541.
- 20. Little JC, Hodgson AT, Gadgil AJ (1994) Modeling emissions of volatile organic-compounds from new carpets. Atmos Environ 28: 227–234.
- 21. Xu Y, Zhang YP (2003) An improved mass transfer based model for analyzing VOC emissions from building materials. Atmos Environ 37: 2497–2505.
- 22. Cadogan DF, Howick CJ (1996) Plasticizers. Kirk-Othmer Encycl Chem Technol 19: 258–290.
- 23. Fromme H, Lahrz T, Piloty M, Gebhart H, Oddoy A, et al. (2004) Occurrence of phthalates and musk fragrances in indoor air and dust from apartments and kindergartens in Berlin (Germany). Indoor Air 14: 188–195.
- 24. Hale RC, La Guardia MJ, Harvey E, Mainor TM (2002) Potential role of fire retardant-treated polyurethane foam as a source of brominated diphenyl ethers to the US environment. Chemosphere 46: 729–735.
- 25. Uhde E, Bednarek M, Fuhrmann F, Salthammer T (2001) Phthalic esters in the indoor environment - Test chamber studies on PVC-coated wallcoverings. Indoor Air 11: 150–155.
- 26. Liu C, Morrison GC, Zhang Y (2012) Role of aerosols in enhancing SVOC flux between air and indoor surfaces and its influence on exposure. Atmos Environ 55: 347–356.
- 27. Liu C, Zhao B, Zhang Y (2010) The influence of aerosol dynamics on indoor exposure to airborne DEHP. Atmos Environ 44: 1952–1959.