Nanoscale Discharge Electrode for Minimizing ... - ACS Publications

Jul 2, 2010 - Shanghai 200092, P. R. China, and Xerox Research Center. Webster, Xerox ... Ground-level ozone emitted from indoor corona devices poses ...
0 downloads 0 Views 3MB Size
Environ. Sci. Technol. 2010, 44, 6337–6342

Nanoscale Discharge Electrode for Minimizing Ozone Emission from Indoor Corona Devices ZHENG BO,† KEHAN YU,† GANHUA LU,† S H U N M A O , † J U N H O N G C H E N , * ,†,‡ A N D FA-GUNG FAN§ Department of Mechanical Engineering, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin 53211, State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, P. R. China, and Xerox Research Center Webster, Xerox Corporation, Webster, New York 14580

Received December 24, 2009. Revised manuscript received May 29, 2010. Accepted June 18, 2010.

Ground-level ozone emitted from indoor corona devices poses serious health risks to the human respiratory system and the lung function. Federal regulations call for effective techniques to minimize the indoor ozone production. In this work, stable atmospheric corona discharges from nanomaterials are demonstrated using horizontally suspended carbon nanotubes (CNTs) as the discharge electrode. Compared with the conventional discharges employing micro- or macroscale electrodes, the corona discharge from CNTs could initiate and operate at a much lower voltage due to the small electrode diameter, and is thus energy-efficient. Most importantly, the reported discharge is environmentally friendly since no ozone (below the detection limit of 0.5 ppb) was detected for area current densities up to 0.744 A/m2 due to the significantly reduced number of electrons and plasma volume generated by CNT discharges. The resulting discharge current density depends on the CNT loading. Contrary to the conventional wisdom, negative CNT discharges should be used to enhance the current density owing to the efficient field emission of electrons from the CNT surface.

Introduction Corona discharge devices such as electrostatic precipitators, photocopiers, and laser printers are known to produce ozone (O3) as a byproduct (1, 2), which is believed to pose serious health hazards to the human respiratory system and to lung function at excessive concentration levels in poorly ventilated areas, causing/aggravating chest pains, asthma, coughing, shortness of breath, and throat irritation (3, 4). Several federal agencies, e.g., Food and Drug Administration (FDA), Occupational Safety and Health Administration (OSHA), and National Institute of Occupational Safety and Health (NIOSH), have proposed regulations (5, 6) or health recommendations (7) to limit ozone emission. The U.S. Environmental Protection Agency (EPA) recently strengthened its National Ambient Air Quality Standards (NAAQS) for ground-level ozone by changing the standard from 80 to 75 ppb (parts per billion) * Corresponding author phone: (414)229-2615; e-mail: jhchen@ uwm.edu. † University of Wisconsin-Milwaukee. ‡ Tongji University. § Xerox Corporation. 10.1021/es903917f

 2010 American Chemical Society

Published on Web 07/02/2010

averaged over eight hours (8), reflecting increasing public concerns about ground-level ozone as a hazardous pollutant and demands for the manufacturers of indoor corona devices to minimize ozone generation following federal regulations (5, 6, 8). Ozone production rate in the atmospheric corona discharge is significantly influenced by the electron number density, the electron kinetic energy, and the corona plasma thickness (9). The corona plasma thickness is referred to herein as the thickness of the plasma region between the discharge electrode surface and the outside plasma boundary, in which the massive volume ionization occurs and the corona plasma-enhanced ozone production is significant. The outside boundary of the corona plasma region is normally defined by the location where the rate of ionization balances the rate of electron attachment. The electrons outside the corona plasma region are not energetic enough to drive the ozone formation. Typically, more ozone is produced in the negative corona than in the positive corona because both the total number of electrons and the plasma thickness of negative corona discharges are larger for the same discharge geometry and supply voltage (9-12). Current industrial practice is to use positive rather than negative corona discharges to minimize ozone production from indoor corona devices such as electronic air cleaners. For both positive and negative discharges, our recent modeling results (9, 10) suggest that the total number of electrons and the plasma thickness, and thus the ozone production, scale with the discharge electrode size, which has been confirmed by many experimental observations (1, 13-15). Therefore, one possible route to further lower ozone production is to use smaller discharge electrodes, such as nanostructures. Carbon nanotubes (CNTs) are graphene sheets rolled-up into cylinders with diameters as small as a few nanometers, while they can be up to several millimeters in length. CNTs have received special research attention in the field of environmental remediation and pollution control (16) because of their unique physical and chemical properties. For instance, CNTs have been shown useful for efficient adsorbents of organic chemicals (17-21), special sorbents in the treatment of recalcitrant-contaminated water (22), rapid and precise sensors for monitoring gaseous pollutants (23), and catalyst support for exhaust emission control (24, 25). Furthermore, CNTs are considered to be a wonderful electrode candidate due to its high aspect ratio (26) and high electrical conductivity (27) that are conducive for electrical discharges, good chemical stability, and superior mechanical strength and durability (28), as evidenced by recent applications of CNTs for field emitters (29), gas sensors (30), gas discharge tube protectors (31), microplasma devices (32), and liquid electrical discharges (33). In this work, we report on the corona discharge from CNTs as a means to minimize ozone production. It is expected that ozone production could be significantly reduced by confining the plasma region in the CNT corona discharge. The novel corona discharge device was fabricated by dispersing horizontally oriented, multiwalled CNTs (MWCNTs) (sitting on their side surfaces) on a holey carbon film to be used as a discharge electrode. The ozone emission from the resulting atmospheric CNT corona discharge was then evaluated and compared with the conventional discharge from microsized discharge electrodes.

Experimental Section Fabrication of CNT Electrodes. Multiwalled CNTs (MWCNTs) purchased from Alfa Aesar (0.5 mg, average diameter: ∼20 VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6337

FIGURE 1. MWCNT corona discharge device together with a scanning electron microscopy (SEM) image of the CNT discharge electrode and a high-resolution transmission electron microscopy (HRTEM) image of an individual MWCNT. nm, length: 5-20 µm) were mixed with ethanol solvent (10 mL, 94-96%, Alfa Aesar) to produce a black suspension (0.05 mg CNTs/mL). The dispersion of CNTs was accomplished by 1-h probe sonication (Virsonic 600 ultrasonic cell disrupter, VirTis Company) at a power of 12 W followed by a 30-min bath sonication treatment (Branson model 1510 ultrasonic cleaner, Branson Ultrasonic Corporation). Electrode samples with various CNT densities (loadings) were prepared by dropcasting different amounts (2-8 drops) of the well-dispersed CNT/ethanol suspension onto the substrate, i.e., a 400 mesh holey carbon film coated copper transmission electron microscopy (TEM) grid (3 mm in diameter, Ted Pella, Inc.). The electrodes were then dried for 1 h with the radiation from an incandescent lamp, followed by a 24-h, roomtemperature drying process through natural convection. With the evaporation of ethanol from the substrate, horizontally oriented CNTs were left behind on the holey carbon film due to the van der Waals interaction between the CNTs and the holey carbon film. No obvious agglomerating CNT was observed after the above fabrication process. The CNT-ongrid was used as the corona discharge electrode. The average density of CNTs on the grid can be controlled by the amount of CNT dispersions used in the drop-casting process. The morphology of the CNTs drop-cast on the carbon film is strongly influenced by the solvent used to disperse CNTs. The ability of various surfactants to promote stable dispersions of CNTs in different aqueous media has been widely reported (34); however, the use of surfactants would lead to the formation of surfactant-functionalized CNTs and possible modification of electrical properties of pristine CNTs. Our experiments showed that ethanol is a good candidate for dispersing CNTs to obtain nonagglomerated CNTs on the TEM grid that can be readily used for corona discharge electrode. CNT Corona Discharge Device. As shown in Figure 1, the large number of holes with varying sizes from ∼1 µm to ∼3 µm in diameter across the entire carbon film makes the holey grid an ideal substrate to support MWCNTs of ∼2.5 µm in length (shortened by the probe-sonication-induced disruption (35)) for discharge applications because the suspended CNTs can lead to a highly enhanced electric field that is critical for the initiation of corona discharges. The prepared MWCNT electrode was placed on the top surface of a brass stage for electrical contact (via the copper grid) with a reversible-polarity, direct-current high voltage supply (DCHV, -10∼+7 kV, EMGO Co.), as shown in Figure 1. To avoid the undesirable discharge from the grid edge, a polytetrafluoroethylene (PTFE) cover was used to wrap the edge. A highly polished brass plate was connected to the ground through a picoammeter and served as the passive electrode. The 6338

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

FIGURE 2. SEM images of (a) the selected meshes along the axes of the grid, (b) an individual selected mesh, and (c) the selected area of the representative mesh for the image analysis. The red boxes in (a) are labeled as X-5∼X-1, XY0, and X1∼X5 from left to right; Y-5∼Y-1, XY0, and Y1∼Y5 from bottom to top. interelectrode gap was maintained as 0.254 mm through a specially designed PTFE spacer to allow the discharge airflow. Calculation of Effective Discharge Length of MWCNTs. The CNT corona discharge occurs between the side surface/ tip of the CNT and the ground electrode. Therefore, for a fixed interelectrode gap and supply voltage, the discharge current is linearly proportional to the effective discharge length of the CNT electrode assuming that the distance between neighboring CNTs is so large that there is no electric field interference between neighboring CNTs. When the CNTs are too close to each other, electrostatic screening effects will occur and suppress the initiation of corona discharges (to be discussed in Figure 3). The effective discharge length of the deposited MWCNTs Lcnt (unit: mm), i.e., the length of MWCNTs above the holes, was statistically determined by analyzing over twenty SEM images for each case. Twentyone meshes along the axis of the grid, labeled as XY0, X-5∼X1, X1∼X5, Y-5∼Y-1, and Y1∼Y5, were selected as representative meshes for statistical analyses, highlighted in red in Figure 2(a). For each selected mesh, the local effective discharge length of CNTs (lcnt(i), mm) in the region of the top

corona discharge in air-cleaning devices is the surface current density at the discharge electrode surface, which affects the charging efficacy of airborne particles (36). For CNT discharges, the surface current density Jcnt (unit: A/m2) was calculated by dividing the discharge current Icnt (unit: nA) by the effective discharge surface area of MWCNTs Jcnt ) 103 ×

FIGURE 3. Surface current density of MWCNT discharges as a function of supply voltage (J-V curve, solid line for positive, and dashed line for negative). left corner (6.325 µm × 4.433 µm or 28.04 µm2 in area; highlighted in green in Figure 2(b)) was measured and used as the average CNT discharge length for that particular mesh, as shown in Figure 2(c). The average value ljcnt (unit: mm) was thus calculated as

lcnt )

∑l

cnt(i)

(1)

n

where n is the sample number, i.e., 21 in the current work. The precision error ∆lcnt was calculated according to the Student t-distribution ∆lcnt ) tR/2

δ √n

(2)

where tR/2 is 2.086 (degree of freedom is set as n - 1 ) 20), assuming a 95% confidence level, i.e., R/2 ) 0.025; the standard deviation δ was calculated as δ)

[ n -1 1 ∑ (l

cnt(i)

1/2

]

- lcnt)2

(3)

As the total area of the carbon film is 0.5776 mm2 (400 meshes, and 38 µm × 38 µm for each mesh), the effective discharge length of CNTs across the entire sample was calculated as Lcnt ) 0.5776 × 106 ×

lcnt ( ∆lcnt 28.04

(4)

The effective length of the CNTs for the discharge was varied by the amount of CNT suspension (CNT loading) used for the casting process. Discharge Procedure. Discharges were conducted in a transparent polymethyl methacrylate (PMMA) chamber with a confined volume of ∼85.8 cm3 at room temperature (298 K) and atmospheric pressure (101,325 Pa). A highly polished brass plate with an area of 78.5 mm2 was used as the passive electrode for all experiments. A high-purity, dry air flow of 1500 sccm was used to produce and sustain the corona discharge. Prior to each CNT discharge experiment, a background current was taken from the holey carbon filmcoated TEM grid discharge. For comparison, conventional tungsten wire discharge electrodes (0.1778 mm, 0.1524 mm, and 0.127 mm in diameter, respectively, and 5 mm in effective length) were also used for controlled experiments. Electrical Characterization. A sensitive picoammeter (range: 20 fA∼21 mA, resolution: 51/2 digit, Model 6485, Keithley Instruments, Inc.) was used to measure the discharge current. A very important parameter for the operation of

Icnt πdcntLcnt

(5)

where the average diameter of MWCNTs dcnt was 20 nm, and the effective discharge length of the MWCNTs Lcnt (unit: mm) was statistically calculated according to eq 4. The corona discharge from the holey carbon film-coated TEM grid was performed as a control, and the discharge current from the grid alone was used as the “background current”. In this work, the discharge current Icnt was obtained by subtracting the background current (without CNTs) from the total discharge current (with CNTs) to eliminate the discharge current contribution from the grid. Ozone Measurement. After-discharge gas flow of 750 sccm was pulled into a vacuum-pump-equipped photometric ozone analyzer (range: 0-10000 ppb, precision: (0.5% of reading, Model 400A, Advanced Pollution Instrumentation, Inc.) for ozone measurements. The ozone generation rate per unit length of the tungsten wire rt and the CNTs rcnt were calculated as rt )

rcnt )

1.96 × 10-6CtQ 60Lt

(6)

1.96 × 10-6CcntQ 60Lcnt

(7)

where Ct and Ccnt are the measured ozone concentration for tungsten wire and MWCNT discharges, respectively, Q is the after-discharge gas flow rate, Lt is the effective length of the tungsten wire, and the units of rt, rcnt, Ct, Ccnt, Q, Lt, and Lcnt are mg/s-m, mg/s-m, ppb, ppb, sccm, mm, and mm, respectively. Material Characterization. SEM analysis of the CNT electrodes was performed with a Hitachi S-4800 SEM; the SEM has a resolution of 1.4 nm at 1 kV acceleration voltage. TEM analysis was performed with a Hitachi H 9000 NAR TEM, which has a point resolution of 0.18 nm at 300 kV in the phase contrast high resolution TEM imaging mode. A Raman spectrometer (Renishaw 1000B) having a spot diameter of 4 µm was used to obtain the Raman spectra of the samples with an excitation wavelength of 633 nm. The Raman measurements were performed at positions where only CNTs were seen in the field of view to eliminate the spectrum contribution from the carbon film on the copper grid.

Results and Discussion Electrical Characterization of CNT Discharges. Figure 3 shows the surface current density from the MWCNT discharge as a function of supply voltage (J-V curve) for both positive and negative discharge polarities. As shown in Figure 3, stable positive and negative corona discharges can be obtained within a certain supply voltage range. Typical corona initiation voltages for the CNTs discharge were on the order of ∼100 V, which is much lower than those of the conventional, microsized, tungsten wire discharge (∼1400 V according to Peek’s law for corona inception field strength (37) and White’s equation for wire-to-plate geometry (38), assuming an ideally smooth wire surface) for the same discharge conditions (298 K, 101,325 Pa, and 0.254 mm of interelectrode gap). Furthermore, CNT discharges can achieve the same (or similar) level of discharge current density at a much lower VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6339

supply voltage, compared with the microsized tungsten wire discharges (see Figure S1 of the Supporting Information). This suggests that corona discharges from nanoscale electrodes could be much more energy-efficient, mainly attributed to the extremely small diameter of the discharge electrode. As shown in Figure 3, for a given effective CNT discharge length Lcnt, the corona discharge surface current density Jcnt increases with increasing supply voltage V. Figure 3 also shows that for both positive and negative discharges there is an optimum CNT length (CNT loading) that maximizes the surface current density. For example, for positive discharges at V ) 0.9 kV, the discharge current density increases from 0 to 0.095 A/m2 with the increase of Lcnt from 0 to 353.13 ( 72.61 mm but then decreases to 0.003 A/m2 with the further increase of Lcnt up to 1959.60 ( 198.04 mm; for negative discharges at V ) 0.9 kV, the discharge current density increases from 0 to 0.744 A/m2 with the increase of Lcnt from 0 to 56.70 ( 8.83 mm but then decreases to 0.001 A/m2 with the further increase of Lcnt up to 1871.47 ( 200.72 mm. For the given discharge geometry, the TEM grid surface area (CNT supporting area) is fixed. With a given supply voltage and a fixed interelectrode gap, the discharge current is not only dependent on the CNT effective discharge length but also strongly influenced by the distance between neighboring CNTs. When there are few CNTs on the grid, the distance between neighboring CNTs is large, and there is no field interference (electrostatic screening) between neighboring CNTs; however, the discharge current is low because of lower effective discharge length. When there are too many CNTs on the grid (higher effective discharge length), the distance between neighboring CNTs is too small and the electrostatic screening effect occurs; and the discharge current is not maximized. This inter-CNT electrostatic screening effect at higher CNT loadings has previously been observed in field-emission devices from densely packed, vertically aligned CNTs (39). For the same supply voltage V (e.g., 1 kV) and the similar effective discharge length Lcnt of the MWCNTs (e.g., 1028.93 ( 198.04 mm for positive discharge and 1053.21 ( 154.20 mm for negative discharge), the surface current density of the negative discharge (0.135 A/m2) was higher than that of the positive discharge (0.026 A/m2). This observation was especially obvious in the relatively large supply voltage conditions. It can be explained by the possible involvement of field emission electrons from the CNT surface in the negative CNT discharge (CNTs as the cathode) (40, 41). The field-emitted electrons serve as additional sources of secondary electrons to enhance the discharge current, in addition to the secondary electrons by photoemission in the conventional negative corona discharges (29, 41). Ozone Emission Evaluation. Figure 4 shows the ozone generation rates per unit length of the tungsten wires and MWCNTs as a function of surface current density for both positive and negative discharges. Obvious ozone generation was observed for tungsten wire discharges. For a given surface current density, the ozone generation rate per unit length of the tungsten wire decreases with the decreasing wire size. However, no ozone (below the detection limit, i.e., 0.5 ppb) was detected in all the MWCNT discharges conducted in the current study for surface current densities up to 0.744 A/m2 (negative discharge). Mechanism of Reduced Ozone Emission from CNT Discharges. The significantly reduced ozone emission from CNT corona discharges should be attributed to the decrease of the discharge electrode diameter. As shown in Scheme 1 for a wire-cylinder DC corona discharge, the interelectrode space of a corona discharge can be divided into two parts by the plasma boundary: the corona plasma region adjacent to the highly curved discharge electrode where energetic 6340

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

FIGURE 4. Comparison of ozone generation rates per unit length of the tungsten wire and the MWCNT as a function of the surface current density for both (a) positive and (b) negative corona discharges. electrons are produced by volume ionization and the unipolar ion region in which the electron-impact reactions are insignificant. For atmospheric air discharges, the ozone production in the corona discharge is mainly (∼80% and ∼90% of the ozone production in positive and negative coronas, respectively) attributed to the reactions involving excited molecules, as shown in Scheme 1, according to our previous modeling results (9, 10). Consequently, the total ozone production rate from the corona discharge is simultaneously determined by the electron number density, electron energy, and the plasma volume. Based on earlier modeling studies (10), the reduced diameter of the discharge electrode leads to a smaller number of electrons and plasma thickness (plasma volume) but higher electron mean kinetic energy, which is primarily a function of the electric field. Ozone formation in the atmospheric corona discharge is initiated by the electron-impact reactions between energetic electrons and background gases (e.g., O2 and N2) in the plasma region. The energy thresholds of the formation of excited molecules (i.e., the initial step shown in Scheme 1) are below 15.6 and 12.06 eV, which are the ionization potential of nitrogen and oxygen molecules, respectively. Although the electron kinetic energy follows a distribution and the specific reaction rate depends on the collision cross-section (which is electron energy dependent), electrons with energy higher than the ionization thresholds will not significantly enhance the massive production of excited molecules and thus ozone. According to the previous modeling result (10), the mean kinetic energy of electrons near the wire surface reaches ∼19.5 eV with a discharge electrode diameter of around 20 µm. Therefore, further increase in the mean electron energy with the reduction of the discharge electrode size to the nanometer regime will not obviously influence the specific ozone production rate (ozone production per electron); however, the significantly

SCHEME 1. Charge Carrier Distribution in a Wire-Cylinder DC Corona Discharge (Not to Scale) and the Reaction Mechanism of Ozone Formation

reduced number of electrons and plasma volume will lead to the severe reduction in ozone production. Physical Stability of CNTs during Corona Discharges. To characterize the physical changes of the CNTs during the corona discharge, Figure 5 shows the Raman spectra of two arbitrary positions (P1 and P2) in the CNTs electrode before and after a 30-min negative discharge. For positions 1 (P1) and 2 (P2), the ratio of the peak intensity of D band (1332 cm-1) and G band (1582 cm-1) before and after the discharge was calculated as 1.052 and 1.101 and 1.049 and 1.052, respectively, which indicates that the physical structure of MWCNT (42) was not obviously damaged by the discharge, and there were no obvious oxidation-induced defects on the sidewalls of MWCNTs (43), suggesting the extremely low ozone concentration. The Raman results are consistent with SEM images of the CNT electrode before and after the corona discharge (see Figure S1 of the Supporting Information), in which no CNT releasing from the substrate was observed during the discharge. The good chemical stability with strong chemical bonds in the graphene sheet (28) ensures the practical use of CNTs for corona discharge applications. Environmental Implication. This study has important environmental implication for manufacturers of indoor corona discharge devices, such as electrostatic precipitators, photocopiers, and laser printers, to follow federal regulations for ozone emission. The energy-efficient (lower corona initiation voltage) and environmentally friendly (near-zero ozone emission) corona discharge through the use of CNTs as discharge electrodes is promising for practical applications in indoor electrostatic devices. The resulting discharge current

density can be varied by varying the CNT loading. Although positive corona discharges are often preferred in conventional corona devices, when CNTs are used as the discharge electrode, negative corona discharges should be used to enhance the current density without the penalty in ozone production. Precautions should be taken in the process of CNT electrode fabrication (e.g., CNT dispersion, coating, and drying) due to the possible negative health impacts (44, 45) of the exposure of CNTs. Although no obvious CNT releasing was observed in the present corona discharge experiments, the long-term stability of the CNT discharge device deserves further investigations. Exploration of vertically aligned CNT arrays and other nanomaterials for the corona discharge electrode will also be attempted in the future.

Acknowledgments We acknowledge the financial support from National Science Foundation (CBET-0741336), Xerox Corporation, University of Wisconsin System, and UWM Research Growth Initiative. The authors thank H. A. Owen for technical support with SEM, A. Andre for technical support with ozone analyzer, and A. V. Skliarov for technical support with Raman. Z. Bo acknowledges the research fellow support from the UWM Research Foundation.

Supporting Information Available Figure S1 shows a comparison of the required supply voltages for corona discharges employing CNTs and microsized tungsten wire electrodes to obtain the same surface current density and the SEM images of a CNT discharge electrode before and after a 30-min negative discharge are shown in Figure S2. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited

FIGURE 5. Raman spectra of position 1 (P1) and position 2 (P2) of the CNT electrode before and after a 30-min negative discharge (inset: SEM image indicating the P1 and P2 position).

(1) Boelter, K. J.; Davidson, J. H. Ozone generation by indoor, electrostatic air cleaners. Aerosol Sci. Technol. 1997, 27 (6), 689– 708. (2) Lee, S. C.; Lam, S.; Fai, H. K. Characterization of VOCs, ozone, and PM10 emissions from office equipment in an environmental chamber. Build. Environ. 2001, 36 (7), 837–842. (3) U. S. Environmental Protection Agency. Air quality criteria for ozone and related photochemical oxidants; EPA: 2006. (4) U. S. Environmental Protection Agency. Health and environmental effects of ground-level ozone; EPA: 1997. (5) Food and Drug Administration. Code of Federal Regulations Title 21 Part 801.415. Available at http://ecfr.gpoaccess.gov/cgi/t/text/ text-idx?c)ecfr&sid)f9bcd1e6e6301af95997562a61d5b8a0&rgn) div8&view)text&node)21:8.0.1.1.2.7.1.3&idno)21 (accessed December 24, 2009). (6) Occupational Safety and Health Administration. Code of Federal Regulations Title 29 Part 1910.137. Available at http://ecfr. gpoaccess.gov/cgi/t/text/text-idx?c)ecfr&sid)f9bcd1e6e6301af95997562a61d5b8a0&rgn)div8&view)text&node)29:5.1.1.1.8.9.34.6& idno)29 (accessed December 24, 2009). VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6341

(7) National Institute for Occupational Safety and Health. NIOSH pocket guide to chemical hazards; NIOSH: 2005. (8) U. S. Environmental Protection Agency. Code of Federal Regulations Title 40 Part 50. Available at http://ecfr.gpoaccess.gov/cgi/t/ text/text-idx?c)ecfr&sid)f9bcd1e6e6301af95997562a61d5b8a0& rgn)div5&view)text&node)40:2.0.1.1.1&idno)40 (accessed December 24, 2009). (9) Chen, J. H.; Davidson, J. H. Ozone production in the positive DC corona discharge: model and comparison to experiments. Plasma Chem. Plasma Process. 2002, 22 (4), 495–522. (10) Chen, J. H.; Davidson, J. H. Ozone production in the negative DC corona: the dependence of discharge polarity. Plasma Chem. Plasma Process. 2003, 23 (3), 501–518. (11) Wang, P. X.; Chen, J. H. Numerical modelling of ozone production in a wire-cylinder corona discharge and comparison with a wire-plate corona discharge. J. Phys. D: Appl. Phys. 2009, 42 (3), 035202. (12) Soria, C.; Pontiga, F.; Castellanos, A. Plasma chemical and electrical modelling of a negative DC corona in pure oxygen. Plasma Sources Sci. Technol. 2004, 13 (1), 95–107. (13) Skalny, J. D.; Orszagh, J.; Matejcik, S.; Mason, N. J. Ozone generation in positive and negative corona discharge fed by humid oxygen and carbon dioxide. Phys. Scr. 2008, 2008 (T131), 014012. (14) Viner, A. S.; Lawless, P. A.; Ensor, D. S.; Sparks, L. E. Ozone Generation in Dc-Energized Electrostatic Precipitators. IEEE Trans. Ind. Appl. 1992, 28 (3), 504–512. (15) Held, B.; Peyrous, R. A systematic parameters study from analytic calculations to optimize ozone concentration in an oxygen-fed wire-to-cylinder ozonizer. Eur. Phys. J. Appl. Phys. 1999, 7 (2), 151–166. (16) Masciangioli, T.; Zhang, W. X. Environmental technologies at the nanoscale. Environ. Sci. Technol. 2003, 37 (5), 102A–108A. (17) Chen, W.; Duan, L.; Zhu, D. Q. Adsorption of polar and nonpolar organic chemicals to carbon nanotubes. Environ. Sci. Technol. 2007, 41 (24), 8295–8300. (18) Yang, K.; Zhu, L. Z.; Xing, B. S. Adsorption of polycyclic aromatic hydrocarbons by carbon nanomaterials. Environ. Sci. Technol. 2006, 40 (6), 1855–1861. (19) Cho, H. H.; Smith, B. A.; Wnuk, J. D.; Fairbrother, D. H.; Ball, W. P. Influence of surface oxides on the adsorption of naphthalene onto multiwalled carbon nanotubes. Environ. Sci. Technol. 2008, 42 (8), 2899–2905. (20) Wang, X. L.; Tao, S.; Xing, B. S. Sorption and Competition of Aromatic Compounds and Humic Acid on Multiwalled Carbon Nanotubes. Environ. Sci. Technol. 2009, 43 (16), 6214–6219. (21) Pan, B.; Xing, B. S. Adsorption Mechanisms of Organic Chemicals on Carbon Nanotubes. Environ. Sci. Technol. 2008, 42 (24), 9005– 9013. (22) Tasis, D.; Tagmatarchis, N.; Bianco, A.; Prato, M. Chemistry of carbon nanotubes. Chem. Rev. 2006, 106 (3), 1105–1136. (23) Kong, J.; Franklin, N. R.; Zhou, C. W.; Chapline, M. G.; Peng, S.; Cho, K. J.; Dai, H. J. Nanotube molecular wires as chemical sensors. Science 2000, 287 (5453), 622–625. (24) Gomes, H. T.; Samant, P. V.; Serp, P.; Kalck, P.; Figueiredo, J. L.; Faria, J. L. Carbon nanotubes and xerogels as supports of welldispersed Pt catalysts for environmental applications. Appl. Catal. B: Environ. 2004, 54 (3), 175–182. (25) Lu, C. Y.; Wey, M. Y.; Chuang, K. H. Catalytic treating of gas pollutants over cobalt catalyst supported on porous carbons derived from rice husk and carbon nanotube. Appl. Catal. B: Environ. 2009, 90 (3-4), 652–661. (26) Zheng, L. X.; O’Connell, M. J.; Doorn, S. K.; Liao, X. Z.; Zhao, Y. H.; Akhadov, E. A.; Hoffbauer, M. A.; Roop, B. J.; Jia, Q. X.; Dye, R. C.; Peterson, D. E.; Huang, S. M.; Liu, J.; Zhu, Y. T.

6342

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

(27) (28) (29)

(30) (31)

(32) (33)

(34)

(35) (36) (37) (38) (39)

(40)

(41) (42) (43)

(44)

(45)

Ultralong single-wall carbon nanotubes. Nat. Mater. 2004, 3 (10), 673–676. Dai, H. J. Carbon nanotubes: opportunities and challenges. Surf. Sci. 2002, 500 (1-3), 218–241. Kolotilo, D. M. Chemical bond and properties of graphite materials. Powder Metall. Met. Ceram. 1968, 7 (12), 966–971. Choi, W. B.; Jin, Y. W.; Kim, H. Y.; Lee, S. J.; Yun, M. J.; Kang, J. H.; Choi, Y. S.; Park, N. S.; Lee, N. S.; Kim, J. M. Electrophoresis deposition of carbon nanotubes for triode-type field emission display. Appl. Phys. Lett. 2001, 78 (11), 1547–1549. Modi, A.; Koratkar, N.; Lass, E.; Wei, B. Q.; Ajayan, P. M. Miniaturized gas ionization sensors using carbon nanotubes. Nature 2003, 424 (6945), 171–174. Rosen, R.; Simendinger, W.; Debbault, C.; Shimoda, H.; Fleming, L.; Stoner, B.; Zhou, O. Application of carbon nanotubes as electrodes in gas discharge tubes. Appl. Phys. Lett. 2000, 76 (13), 1668–1670. Park, S. J.; Eden, J. G.; Park, K. H. Carbon nanotube-enhanced performance of microplasma devices. Appl. Phys. Lett. 2004, 84 (22), 4481–4483. Staack, D.; Fridman, A.; Gutsol, A.; Gogotsi, Y.; Friedman, G. Nanoscale Corona Discharge in Liquids, Enabling Nanosecond Optical Emission Spectroscopy. Angew. Chem., Int. Ed. 2008, 47 (42), 8020–8024. Datsyuk, V.; Landois, P.; Fitremann, J.; Peigney, A.; Galibert, A. M.; Soula, B.; Flahaut, E. Double-walled carbon nanotube dispersion via surfactant substitution. J. Mater. Chem. 2009, 19 (18), 2729–2736. Chen, G.; Shin, D. H.; Kim, S.; Roth, S.; Lee, C. J. Improved field emission stability of thin multiwalled carbon nanotube emitters. Nanotechnology 2010, 21 (1), 015704. Hinds, W. C. Aerosol Technology; Wiley: New York, 1998. Peek, F. W. Dielectric Phenomena in High-Voltage Engineering; McGraw-Hill: New York, 1929. White, H. J. Industrial Electrostatic Precipitation; Addison Wesley: MA, 1963. Nilsson, L.; Groening, O.; Emmenegger, C.; Kuettel, O.; Schaller, E.; Schlapbach, L.; Kind, H.; Bonard, J. M.; Kern, K. Scanning field emission from patterned carbon nanotube films. Appl. Phys. Lett. 2000, 76 (15), 2071–2073. Calderon-Colon, X.; Geng, H. Z.; Gao, B.; An, L.; Cao, G. H.; Zhou, O. A carbon nanotube field emission cathode with high current density and long-term stability. Nanotechnology 2009, 20 (32), 325707. Jung, S. M.; Hahn, J.; Jung, H. Y.; Suh, J. S. Clean carbon nanotube field emitters aligned horizontally. Nano Lett. 2006, 6 (7), 1569– 1573. Ha, B.; Yeom, T. H.; Lee, S. H. Ferromagnetic properties of singlewalled carbon nanotubes synthesized by Fe catalyst arc discharge. Physica B 2009, 404 (8-11), 1617–1620. Simmons, J. M.; Nichols, B. M.; Baker, S. E.; Marcus, M. S.; Castellini, O. M.; Lee, C. S.; Hamers, R. J.; Eriksson, M. A. Effect of ozone oxidation on single-walled carbon nanotubes. J. Phys. Chem. B 2006, 110 (14), 7113–7118. Ryman-Rasmussen, J. P.; Cesta, M. F.; Brody, A. R.; ShipleyPhillips, J. K.; Everitt, J. I.; Tewksbury, E. W.; Moss, O. R.; Wong, B. A.; Dodd, D. E.; Andersen, M. E.; Bonner, J. C. Inhaled carbon nanotubes reach the subpleural tissue in mice. Nat. Nanotechnol. 2009, 4 (11), 747–751. Raja, P. M. V.; Connolley, J.; Ganesan, G. P.; Ci, L. J.; Ajayan, P. M.; Nalamasu, O.; Thompson, D. M. Impact of carbon nanotube exposure, dosage and aggregation on smooth muscle cells. Toxicol. Lett. 2007, 169 (1), 51–63.

ES903917F