Development and Field-Scale Optimization of a ... - ACS Publications

Nov 17, 2011 - The combined concentrator/oxidizer system has been proposed as an effective physical-chemical option and proven to be a viable solution...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/est

Development and Field-Scale Optimization of a Honeycomb Zeolite Rotor Concentrator/Recuperative Oxidizer for the Abatement of Volatile Organic Carbons from Semiconductor Industry Ji Yang,†,* Yufeng Chen,‡,§ Limei Cao,† Yuling Guo,† and Jinping Jia§ †

State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai 200237, People's Republic of China ‡ Semiconductor Manufacturing International Corporation (SMIC), 18 Zhangjiang Rd, Pudong New Area, Shanghai 201203, People's Republic of China § School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, People's Republic of China ABSTRACT: The combined concentrator/oxidizer system has been proposed as an effective physical-chemical option and proven to be a viable solution that enables Volatile Organic Carbons (VOCs) emitters to comply with the regulations. In this work, a field scale honeycomb zeolite rotor concentrator combined with a recuperative oxidizer was developed and applied for the treatment of the VOC waste gas. The research shows the following: (1) for the adsorption rotor, zeolite is a more appropriate material than Granular Activated Carbon (GAC). The designing and operation parameters of the concentrator were discussed in detail including the size and the optimal rotation speed of rotor. Also the developed rotor performance’s was evaluated in the field; (2) Direct Fired Thermal Oxidizer (DFTO), Recuperative Oxidizer (RO), Regenerative Thermal Oxidizer (RTO) and Regenerative Catalytic oxidizer (RCO) are the available incinerators and the RO was selected as the oxidizer in this work; (3) The overall performance of the developed rotor/oxidizer was explored in a field scale under varying conditions; (4) The energy saving strategy was fulfilled by reducing heat loss from the oxidizer and recovering heat from the exhaust gas. Data shows that the developed rotor/oxidizer could remove over 95% VOCs with reasonable cost and this could be helpful for similar plants when considering VOC abatement.

’ INTRODUCTION In the past decade, the semiconductor manufacturing industry rapidly expanded in China. Consequently, increasing amounts of air pollutants are being generated annually, mostly VOCs which are commonly emitted by semiconductor and optoelectronic manufacturers. The waste gases generally consist of a mixture of chemicals such as acetone, isopropyl alcohol (IPA), propylene glycol monomethyl ether (PGME), and propylene glycol monomethyl ether acetate (PGMEA). Some of the VOCs are environmentally hazardous compounds which are often carcinogens and mutagens and may deplete stratospheric ozone.13 Because of public and corporate pressure, environmental legislation in all major industrialized countries is indeed posing increasingly stringent standards on VOC emissions.4 In addition, the ISO 14000 standard demands the proper treatment of VOC exhaust from these industries. Therefore, the market for the technologies has been increasing to reduce VOCs in the waste gases from the semiconductor industry.5,6 r 2011 American Chemical Society

The air streams containing VOCs from the semiconductor industry differ among various manufacturing processes, and are frequently characterized by large volume and low, varying concentrations, which make VOC recovery not possible. The most pertinent strategy to treat these emissions is to destruct or transfer VOCs physical-chemically or biologically. The physical-chemical technologies include cooling, adsorption, or the combustion process, or by a combination of the processes.79 Biofiltration systems have recently emerged as an efficient and cost-effective technology for the control of VOC emission, which has been used in Europe for many years, while it will not work for the Received: September 10, 2011 Accepted: November 17, 2011 Revised: October 31, 2011 Published: November 17, 2011 441

dx.doi.org/10.1021/es203174y | Environ. Sci. Technol. 2012, 46, 441–446

Environmental Science & Technology

ARTICLE

Table 1. Characteristics of the GAC and Zeolitea

poisonous VOCs and usually requires relatively longer empty bed retention time. Thermal or catalytic oxidation is used as the preferred treatment technique for this purpose. Due to the low concentrated exhaust gas, oxidation requires additional energy which is not to be ignored. Consequently, a combined concentrator/oxidizer system has been proposed as an effective physical-chemical option and proven to be a viable solution that enables VOC emitters to comply with the regulations all over the world.10,11 Combining the concentrator with oxidation processes shows that it is more economical, efficient, and yields fewer secondary emissions as compared with the single oxidation processes such as the regenerative thermal oxidizer, regenerative catalytic oxidizer, or catalytic thermal oxidizer processes.12 For a combined system, the concentrator removes the VOCs from the exhaust air and an oxidizer then mineralizes the concentrated stream of the VOCs emitted from the concentrator. Some research has been performed to explore the removal of VOCs using rotor concentrator alone,10,11,13 however observations and research generally are not available on a sufficient large scale to show the potential for optimization of a combined concentrator/oxidizer system, despite the many studies which have been conducted on removal efficiency and reaction product mineralogies of the concentrator/oxidizer systems. Therefore, the design and control of the major parameters that govern the concentrator/oxidizer frequently raise problems in meeting the required removal of an exhaust gas treatment system, which include the type of absorbent, air flow rate, rotating speed, regeneration temperature, the area ratio of process/cooling/regeneration zone, the flow rate ratio of the process flow to the regeneration flow, and VOC concentrations. Consequently, it is in this context that the present study was undertaken with a view to explore the development and application parameters of the combined concentrator/oxidation system. The major objectives of this study were as follows: (1) screening of commercial adsorbents for the concentrator; (2) selection of proper oxidizer; (3) development of a combined concentrator/oxidation system for a semiconductor plant; and (4) exploring the operating parameters for the combined system. The field scale waste gas treatment facility was developed for SMIC Shanghai, which is one of the leading semiconductor foundries in the world and the largest and most advanced foundry in China, providing integrated circuit (IC) foundry and technology services at 0.35-μm to 45/40-nanometer. The significance of this research is to provide comprehensive design and operating information for the control of waste gas containing VOCs using a combined concentrator/oxidation system. The most relevant application of this information refers to the field treatment of the semiconductor and electro-optical industries to obtain their optimal operation conditions of the concentrator/oxidizer for both efficient VOC controls and energy savings.

SBET (m2/g)

Zeolite

GAC

650.50

1021.23

Vmicro (mL/g)

0.11

0.19

pore diameter (nm)

3.68

2.97

flammability

NA

high

heat resistance performance

high

low

residual quantity after desorption

low

high

low to high

high

suitable VOC concentration

SBET—Specific surface area determined using BET method; Vmicro— Micro-pore volume. a

Figure 1. Schematic diagram of the lab adsorption setup (1. air pump, 2. flowmeter, 3. VOC solution, 4. mixer, 5. adsorption reactor, 6. adsorbent plate, 7. glass beads, 8. gas diffuser, and 9. GC).

A fixed bed adsorption unit given in Figure 1 was employed to study adsorption dynamics of VOCs in a reactor packed with GAC and zeolite. 40 g adsorption material was packed in the reactor, in a shape of a plate, with a radius of 31.5 mm. The bulk density of the adsorption plate was 421 kg/m3, the bed void fraction about 0.129 for GAC, and the bulk density of adsorbent was 456 kg/m3, the bed void fraction about 0.118 for zeolite respectively. Before packing, the adsorbent was washed with DI water and dried in a vacuum oven. The carrier gas was prepurified air. The sorbent used as the stationary phase in the reactor was preconditioned before use for at least 2 h under a stream of the carrier air. In the lab scale adsorption setup, the air was divided into two lines: one was pumped (ACO-5503, Haili, P.R. China) into the container with organic solution to bring VOCs into the adsorption reactor; another line of air was mixed with the carrier gas containing VOCs to adjust the gas concentration by controlling the mixing ratio of the two gas flows. To ensure homogeneous mixing of VOC vapors with carrier gas stream, a static in-line mixer was installed at the inlet section of the adsorption reactor. When the required concentration was reached and kept stable, the gas stream was driven through the reactor to start the adsorption process through a gas diffuser and then the glass beads to diffuse the VOCs evenly. The carrier gas flow rate was measured by a Teflon Flow meter (LZB-4. Yuyao Zhenxing, P.R.C.). During adsorption, the concentration at the exit of the adsorption bed was monitored by a gas chromatograph (GCRAE-1000, U.S.) fitted with a Photoionization Detector. The chromatographic conditions were as follows: 50 cm3/s as the carrier gas (He) flow rate; 40 °C as initial temperature for 5 min; 5 °C/min as

’ EXPERIMENTAL SECTION Materials and Analysis. For lab scale experiments, all chemicals were obtained from Sino-reagent, and are of analytical grade. The detailed characteristics of the GAC and zeolite (SiO2 = 6167%, Al2O3 = 1112.4%, Fe2O3 = 4.306.98%, CaO = 2.523.24%, Na2O = 1.661.79%, K2O = 1.852.07%, MgO = 1.541.64%) are shown in Table 1. 442

dx.doi.org/10.1021/es203174y |Environ. Sci. Technol. 2012, 46, 441–446

Environmental Science & Technology

ARTICLE

Table 2. Characteristics of the Exhaust Gas from Manufacturing Process chemicals

combustion heat (kcal/kg) mole percentage (%)

IPA

7517

30.00%

acetone PGMEA

6965 5931

20.00% 25.00%

PGME

6611

23.00%

monoethyl adipate

4821

1.00%

dimethyl sulfoxide

6823.35

0.40%

N-methyl-2-pyrrolidone

9262

0.50%

hexamethyl disilylamine

8857

0.10%

temperature to 280 °C and keep the temperature for 6 min; 280 °C as injector and transfer-line temperature. For the field experiments, 200 mL gas samples were taken periodically from the exhaust air using a Nutech autosampler system (Nutech 3550, U.S.), including sampling canister, Nafion Dryer, Cryo Trap, Cryo Focuser and the organics were determined using a gas chromatograph (HP6890/5973MSD, U.S.). The total organic concentration in the air steams was also determined online by a hydrocarbon analyzer (51C-HT, Thermo, U.S.) periodically, with the detector temperature at 200 °C, sampling pump pressure at 6.50 psi, air pressure at 15 psi, and hydrogen pressure at 9.5 psi. The removal is defined as the fractional decrease in VOC concentration relative to the feed concentration. The waste gas volume was obtained by measuring the gas flow velocity using a TESTO-400 multifunction measuring instrument (TESTO, Germany).

Figure 2. Adsorption breakthrough curves of GAC and Zeolite (VOC concentration 15 ppmv, gas flow rate 15 L/min).

’ RESULTS AND DISCUSSION Figure 3. Flowchart of the concentrator/oxidizer system (1. inlet VOCs, 2. pump, 3. Zeolite concentrator, 4. adsorption section, 5. regeneration section, 6. cooling section, 7. heat exchanger, 8. exhaust gas, 9. oxidizer, and 10. fuel for the oxidizer).

Data Collection. Waste Gas Data. The production process of semiconductor manufacturing is complicated, in which various different chemical substances are employed for specific purpose in almost every step. Therefore, nearly all manufacturing processes are the likely sources of air pollution. At SMIC Shanghai, VOCs are mainly generated at following processes: (1) etching process, emitting acid gas such as organic acids; and (2) photo resist coating, exposure, and substrate cleaning process, mostly organic solvents. The detailed information of the exhaust gas from SMIC Shanghai is shown in Table 2 as follows: From Table 2, it could be told that the main pollutants in the exhaust gas are IPA, acetone, PGMEA and PGME. It has been determined that the overall concentrations of VOCs are in the range of 5091 ppmv, and the total volume of the exhaust air was around 27751 m3/h. Selection of Adsorbents. The rotor concentrator made of different adsorption materials are often the most cost-effective technology for concentrating high volume, low concentration VOC streams. It has the advantages such as low pressure drop, continuous operation as opposed to a batch type process and low maintenance costs. Crucial to the efficiency of the concentrator is the performance of the selected adsorbent, determined by its capacity, and rate factor which are key indexes in the adsorption. For engineering applications, these two parameters together with the life expectancy of the material constitute the basis for the selection of suitable adsorbents, design, and process optimization of plant adsorption units, and for the analysis of their economical, long-term operation. Consequently, lab scale experiments were performed in order to compare the performance of different

materials. The rotor concentrator generally requires the adsorbent in the form of thin plate and a stable performance is expected under continuous adsorption/regeneration for a long period of time. Therefore, a lab scale reactor was designed as shown in Figure 1 and four typical VOCs in the exhaust gas from semiconductor industry were employed to evaluate the adsorbents, with the contaminant concentration and gas flow rate adjusted to match the field situations. The breakthrough curves of GAC and zeolite are shown in Figure 2, respectively. It is manifested in Figure 2 that for the initial 20 min, satisfying results (over 90% removal) were observed in the VOC adsorption capacities and speed for these two adsorbents, GAC and zeolite. This observation indicates that in the case of physical adsorption both GAC and zeolite could adsorb the chemicals in reasonably short time and large capacity. Furthermore, considering the factors such as flammability of the materials as shown in Table 1, zeolite was chosen to be the adsorbent for the rotor concentrator. Design Development. Rotor Concentrator. Figure 3 shows the flowchart of the VOC abatement system and the schematic setup of a typical rotor concentrator. The high volume, low concentration waste gas passes through a rotor concentrator wheel where the VOCs are stripped from the air and adsorbed onto the wheel. The wheel is generally 400450-mm thick and is 443

dx.doi.org/10.1021/es203174y |Environ. Sci. Technol. 2012, 46, 441–446

Environmental Science & Technology

ARTICLE

Table 3. Performance of the Zeolite Rotor Concentrator under Different Conditions effect of air volume

effect of concentration

air volume (m3/h)

3700

16 000

17 000

air volume (m3/h)

16 000

16 000

16 000

inlet concentration (ppmv)

101.3

101.8

103.2

inlet concentration (ppmv)

101.8

109.3

120.3

outlet concentration (ppmv) removal (%)

4.68 95.4%

3.44 96.6%

4.62 95.5%

outlet concentration (ppmv) removal (%)

3.44 96.6%

3.96 96.4%

4.33 96.4%

Table 4. Comparisons among Different Oxidizers

divided into three sections: the adsorption section, the desorption section, and the cooling section. The majority of this air passes the adsorption section and is then exhausted to atmosphere. A small portion of the air stream passing the cooling section is heated to an elevated temperature (usually higher than the boiling points of the VOCs) to be used as desorption air to generate a low volume, high VOC-laden airstream which will be processed later by an oxidizer. The concentrator wheel rotates continuously, transporting the adsorbed VOCs into the desorption section. The recovery ratio R of enriched VOC leaving the regeneration zone divided by the total feed VOC entering the process and the cooling zones is simply calculated as follows:10 R ¼ ðE  F þ 1Þ=ðF þ 1Þ ¼ E þ ð1  EÞ=ðF þ 1Þ > E

RO waste gas volume (m3/h) operating temperature (°C)

pressure loss (Pa) heat exchange type

9599

9598

8095

4070

300400

600700

>700

recuperative regenerative regenerative average

high

operation cost

average

average

high high

low

average

average

In this development, the speed for the rotor was initially set based on eq 4 and was adjusted slightly later to be within the range of 2.53 round/hour according to field data. The desorption temperature was kept at 180 °C, which is higher than the boiling points of the VOCs. Table 3 shows the removal of VOCs of field tests by the rotor alone at different air volumes and VOC concentrations. It is obvious that the rotor could effectively adsorb most VOCs in the air flow within a wide range of flow rate. The removal is also stable and satisfying when the inlet VOC concentration is as high as 120.3 ppm while the observed VOC concentration of the exhaust air during manufacturing is always lower than 91 ppm. Thermal Incinerator. The concentrated air from the rotor should be disposed by thermal degradation. The available types of oxidizers include Direct Fired Thermal Oxidizer (DFTO), Recuperative Oxidizer (RO), Regenerative Thermal Oxidizer (RTO) and Regenerative Catalytic oxidizer (RCO). Due to the low efficiency and high energy cost of DFTO, only RO, RTO and RCO are considered and their corresponding features are listed in Table 4. It could be seen in Table 4 that all these 3 types of oxidizers could destruct high volume VOC air stream efficiently. A mathematical model was developed to evaluate the oxidizer using two independent parameters, heat recovery factor (HRF) and equipment cost factor (ECF),12 which evaluate separately the merits of energy efficiency and cost effectiveness of VOC oxidation systems. On the basis of the model, generally the RO has a lower ECF compared with RCO and RTO. Therefore, RO was selected as the oxidizer (with maximum flow rate of 5840 m3/h, and operating temperature within 660760 °C) and Figure 3 shows the final flowchart of the combined system for SMIC. No matter what type of oxidizer is chosen, the extra fuel is unavoidable and its consumption could be calculated as follows if natural gas is employed:

ð1Þ

However, considering the potential increasing gas volume (maximum 60 000 m3/h) due to the expansion of the business at SMIC, the diameter of the rotor was designed to be 3.25 m for this case. The removal of VOCs is also closely related to the rotation speed of the rotor.10 The rotor concentrator is a cross-flow moving bed, and the supplying rate of the zeolite is controlled by the rotation speed of the rotor. Higher speed correlates to shorter adsorption/regeneration cycle, which means more adsorbents will be supplied and the pollutants leave the process zone before the breakthrough point of adsorption is reached; However, when the rotation speed is high, the adsorbent temperature will not be different between the adsorption and regeneration zones, and the rotor functions just a heat exchanger but not as a concentrator. In this case, lower rotation speed could ensure that the adsorbent enters the process zone after it is cooled sufficiently in the preceding cooling zone. Therefore, the selection of the rotor speed is critical for the system operation. It has been pointed out that the heat supplied by the regeneration air has to be matched with the one to heat the adsorbent entering the regeneration section. In other words, the optimal rotation speed is determined by a balance of heat capacity supplied to the regeneration zone between flowing regeneration air and rotating zeolite, and following correlation is proposed to for engineering estimations of optimal rotation speed Noptimal:10 Fb cc LN optimal  ¼ 0:55 Fg cp βr Ur

99 6570

capital cost maintenance

In this project, 90% removal is expected and therefore the area ratio among different section is set as 10 (adsorption section):1 (desorption section):1 (cooling section). The diameter of the wheel could be determined as follows: rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Vn 12 ð2Þ  diameter ¼ 3600vπ 11

Δoptimal ¼

RCO

360049800 960043200 960043200 700750 760810 300350

destruction efficiency (%) thermal recovery efficiency (%)

RTO

VG ¼ ðQT  QR  QV Þ=HG

ð4Þ

where QT is expressed as follows:

ð3Þ

QT ¼ ðVf þ VZ Þ  Fg  Cg  ðTf  TV Þ 444

ð5Þ

dx.doi.org/10.1021/es203174y |Environ. Sci. Technol. 2012, 46, 441–446

Environmental Science & Technology

ARTICLE

Table 5. System Performance with Different Oxidizer Temperatures

oxidizer temperature (°C) 3

gas volume (Nm /h) VOCs (ppm)

Table 7. Surface Temperature of the Oxidizer Before and After Re-insulation (the Observation Spots Were Distributed Evenly on the Outside Surface of the Oxidizer)

1

2

3

4

5

760

730

730

730

700

26 000 25 000 28 667 28 465 29 520 78.2 78 86.0 78.2 70.0

VOCs removal (%) natural gas consumption (Nm3/h) 3

fuel/waste gas ratio (10 )

96.0

95.0

95.0

94.5

91.2

54.0

43.5

55.5

45.6

37.2

2.08

1.74

1.93

1.60

1.26

VOCs (ppm) gas volume (Nm3/h) oxidizer temperature (°C) VOCs removal (%) natural gas consumption (Nm3/h)

2

3

200

100

50

46 000 730

46 000 730

46 000 730

95

95

95

51.9

58.3

61.5

ð6Þ

spot 4

average

temperature before (°C)

78.2

77.8

74.5

75.4

76.6

temperature after (°C)

32.8

39.5

34.0

30.8

34.3

ð7Þ

System Performance Evaluation. VOC Removal. The removal of VOCs in the rotor concentrator/oxidizer system relies heavily on factors such as temperature of the oxidizer and the VOC concentration in the inflow, since the working temperature and the VOC concentration in the air stream will directly determine the pollutant removal and the fuel consumption of the system. Table 5 shows the system performance under different oxidizer temperatures. It is obvious that at 760 °C, the VOC removal is 96% which is the highest. However, the fuel/gas ratio is also the highest, indicating that more natural gas is needed to maintain the high removal and temperature. When the temperature is lowered to 730 °C, the removal could still be kept around 95%, while less natural gas is needed. Although a significantly lower amount of fuel is needed when the temperature is decreased to 700 °C, the removal is also the lowest, 91.2%. Therefore, to balance the removal and fuel addition, 730 °C was selected as the operation temperature for the system at SMIC. Another important factor that will impact the system operation is the VOC concentration in the inflow air. As manifested in Table 6, it could be seen that higher VOC concentration in the inflow could decrease the requirement for fuel addition, which makes sense since the combustion heat released by the VOCs could increase the temperature in the oxidizer. Otherwise, more fuel will be injected into the oxidizer to ensure the expected removal at certain working temperature. Heat Recovery. In the above thermal oxidation process, VOCladen air is thermally decomposed at temperatures about 730 °C. This high temperature requirement results in intensive energy consumption, and therefore highest possible heat recovery would

σ  ðt1  t2 Þ  S 1000

ð8Þ

On the basis of the information provided, it could be calculated that before the reinsulation, the heat loss was 72.6 kW, while the heat loss was greatly reduced to 16.6 kW when the oxidizer was insulated again with 50 mm rock-wool. This result is particular meaningful because the consumption of fuel could also be substantially reduced to obtain the same VOC removal. Another phenomenon that was noticed was that although two heat exchangers were employed in the system to avoid energy waste as shown in Figure 3, the temperature of the exhaust gas to the atmosphere was still around 250 °C after the treatment. Consequently, heat recovery from this air stream was developed by adding another heat exchanger in front of the final exhaust chimney to collect the heat in the form of hot water which could be used in other departments of the factor. This heat exchanger lowered the final exhaust air temperature from the oxidizer to about 100 °C, and the heat recovered was 619 709 KJ/h, equal to 241 kg/h steam.

And QV, which is the combustion heat of the VOCs in the air stream could be obtained as follows: QV ¼ CT  HV  Vn

spot 3

Qs ¼

The recovered heat QR could be calculated as follows: QR ¼ QT  heat recovery coefficient

spot 2

be needed in addition to the VOC destruction efficiency.12 For the developed system at SMIC, the fuel costs accounts for over 80% of the operation expenses. Consequently, reducing energy consumption will significantly lower the running cost of the system and make it more attractive as a VOC control technique. It was noticed during the operation that although the oxidizer was covered by a 150 mm thermal insulation layer, the surface temperature of the oxidizer was still over 70 °C, indicating that heat loss was significant. Therefore, another layer of 50 mm 48 K rock-wool (with thermal insulation coefficient, λ = 0.058 J/m.s. °C) was employed to further insulate the oxidizer. The effect is obvious as shown in Table 7. The energy saved could be calculated us following expression,

Table 6. System Performance with Different VOCs Concentrations 1

spot 1

’ AUTHOR INFORMATION Corresponding Author

*Tel: 86-21-64250874; Fax: 86-21-64250874; E-mail: yangji@ ecust.edu.cn.

’ ACKNOWLEDGMENT This research is based upon work supported by the National Natural Science Foundation of China (Project No. 21177037), National 863 program (2009AA062603), Shanghai Leading Academic Discipline Project (Project Number: B506) and SMIC. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the supporting organizations. ’ NOMENCLATURE cc the adsorbent heat capacity (J/kg 3 K) the air heat capacity (J/kg 3 K) cp 445

dx.doi.org/10.1021/es203174y |Environ. Sci. Technol. 2012, 46, 441–446

Environmental Science & Technology t1 t2 v Cg CT E F HG HV L QS QT QV S Tf TV Ur Vf Vn VG VZ βr Δoptimal Fb Fg σ

ARTICLE

the surface temperature of the oxidizer (°C) the room temperature (21.8 °C when tested) gas superficial velocity passing the wheel (m/s) the specific heat of air (0.26 kcal/kg 3 °C) the concentration of VOCs (kg/Nm3) fractional decrease in VOC concentration relative to the feed concentration (dimensionless) the ratio of volumetric flow rate of the process to the regeneration zone (dimensionless) the combustion heat of natural gas, 8500 kcal/Nm3 the combustion heat of VOCs (kcal/kg. °C) the thickness of the rotor (m) energy saved (J/s) the total energy consumed in the oxidizer (kcal/h) the energy released by the combustion of the VOCs in the air (kcal/h) stream (kcal/kg. °C) the cooling area of the oxidizer (53 m2) the temperature in the oxidizer (°C) the exhaust air temperature (°C) the superficial air velocity in the regeneration zones (m/s) the volume of concentrated air entering oxidizer (Nm3/h) the volume of the waste air stream (Nm3/h) the fuel requirement to sustain combustion (Nm3/h) the volume of fresh air introduced into the oxidizer to assist combustion (Nm3/h) the fractional cross-sectional area of the regeneration section (dimensionless) the heat capacity ratio (dimensionless) adsorbent bulk densities (kg/m3) air densities (kg/m3) the heat transfer coefficient on the oxidizer surface (= 25 J/m2 3 s. °C)

(10) Yamauchi, H.; Kodama, A.; Hirose, T.; Okano, H.; Yamada, K. I. Performance of VOC abatement by thermal swing honeycomb rotor adsorbers. Ind. Eng. Chem. Res. 2007, 46 (12), 4316–4322. (11) Yamauchi, H.; Kodama, A.; Hirose, T.; Okano, H.; Yamada, K. Design and operation of adsorptive honeycomb rotor VOC concentrators and improvement of performance by high temperature regeneration. Kagaku Kogaku Ronbun. 2008, 34 (2), 217–223. (12) Warahena, A. S. K.; Chuah, Y. K. Energy recovery efficiency and cost analysis of VOC thermal oxidation pollution control technology. Environ. Sci. Technol. 2009, 43 (15), 6101–6105. (13) Chang, F. T.; Lin, Y. C.; Bai, H.; Pei, B. S. Adsorption and desorption characteristics of semiconductor volatile organic compounds on the thermal swing honeycomb zeolite concentrator. J. Air Waste Manage. Assoc. 2003, 53 (11), 1384–1390.

’ REFERENCES (1) Shen, Y. S.; Ku, Y. Treatment of gas-phase volatile organic compounds (VOCs) by the UV/O(3) process. Chemosphere 1999, 38 (8), 1855–1866. (2) Gonzalez-Velasco, J. R.; Aranzabal, A.; Lopez-Fonseca, R. Enhancement of the catalytic oxidation of hydrogen-lean chlorinated VOCs in the presence of hydrogen-supplying compounds. Appl. Catal., B 2000, 24 (1), 33–43. (3) Neal, A. B.; Loehr, R. C. Use of biofilters and suspended-growth reactors to treat VOCs. Waste Manage. 2000, 20 (1), 59–68. (4) Cai, Z. L.; Sorial, G. A. Treatment of dynamic VOC mixture in a trickling-bed air biofilter integrated with cyclic adsorption/desorption beds. Chem. Eng. J. 2009, 151 (13), 105–112. (5) Kim, B. R.; Adams, J. A.; Klaver, P. R. Biological removal of gaseous VOCs from automotive painting operations. J. Environ. Eng. 2000, 126 (8), 745–753. (6) Chen, M.; Fan, L. P.; Qi, L. Y. The catalytic combustion of VOCs over copper catalysts supported on cerium-modified and zirconiumpillared montmorillonite. Catal. Commun. 2009, 10 (6), 838–841. (7) Schottler, M.; Hottenroth, H.; Schl€uter, B; Schmidt, M. Volatile organic compound abatement in semiconductor and solar cell fabrication with respect to resource depletion. Chem. Eng. Technol. 2010, 33 (4), 638–646. (8) Vartanian, V.; Goolsby, B.; Chatterjee, R.; Kachmarik, R.; Babbitt, D.; Reif, R.; Tonnis, E.; Graves, D. Reduction of semiconductor process emissions by reactive gas optimization. IEEE Trans. Semicond. Manufact. 2004, 17 (4), 483–490. (9) Amari, A.; Chlendi, M.; Gannouni, A.; Bellagi, A. Experimental and theoretical studies of VOC adsorption on acid-activated bentonite in a fixed-bed adsorber. Ind. Eng. Chem. Res. 2010, 49 (22), 11587–11593. 446

dx.doi.org/10.1021/es203174y |Environ. Sci. Technol. 2012, 46, 441–446