跳到主要內容

臺灣博碩士論文加值系統

(44.192.20.240) 您好!臺灣時間:2024/02/26 03:51
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:吳振雄
研究生(外文):Chen-Hsiung Wu
論文名稱:矽膠採樣管過濾特性研究
論文名稱(外文):Aerosol Penetration through Silica Gel Tubes
指導教授:陳志傑陳志傑引用關係
指導教授(外文):Chih-Chieh Chen
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:職業醫學與工業衛生研究所
學門:醫藥衛生學門
學類:公共衛生學類
論文種類:學術論文
論文出版年:2001
畢業學年度:89
語文別:英文
中文關鍵詞:矽膠硫酸
外文關鍵詞:silica gelsulfuric acid
相關次數:
  • 被引用被引用:0
  • 點閱點閱:370
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
無中文摘要

Silica gel is commonly used by industrial hygienists to collect gases and vapors in the workplace, in particular air contaminants with high polarity. The collected air pollutants are then treated and analyzed to identify their type and to determine the concentration using various methods and instrumentations. In addition to collection of gaseous pollutants, the silica gel tubes are also used for acid mist collection according to the listed official analytical methods (e.g., NIOSH method 7903 and OSHA method ID-165SG). However, the filtration characteristics of silica gel tubes have not been thoroughly investigated.
A constant output aerosol generator and an ultrasonic atomizing nozzle were used to generate submicrometer-sized and micrometer-sized aerosol particles, respectively. A scanning mobility particle sizer and an aerodynamic particle sizer were used to measure particles smaller and larger than 0.6 μm, respectively. Potassium sodium tartrate and dioctylphthalate were used as the solid and liquid test agents, respectively. Two types of SKC silica gel tubes (Cat No. 226-10 and 226-10-03) were examined for aerosol penetration, air resistance and loading characteristics. The results show that the aerosol penetration through the silica gel tubes could be as high as 80% at the penetration maximum (or collection minimum) under the normal sampling flow of 0.5 L/min, well within the inertial impaction dominated region. Two glass wool plugs and one urethane plug between sorbent sections and at the back end of the SKC 226-10 contributed about 22% of the total air resistance, and the remaining 78% of the air resistance was caused by the silica gel. When the filtration efficiency by these separators was deduced, the aerosol penetration at the most penetrating size was as high as 90%. The aerosol penetration increased and the penetration curve shifted to smaller particle size as the sampling flow increased. However, this increase in aerosol penetration of particles smaller than the penetration maximum reached a maximum and thereafter decreased as the sampling flow was increased beyond 1.5 L/min (equivalent filtration velocity of 93 cm/sec), a clear evidence of inertial impaction surpassing the diffusion deposition. As a result, the use of silica gel tubes for acid mist collection may not be appropriate if the behavior of the complete aerosol size distribution is not considered as part of the assessment of these devices.

Introduction
Silica gel and activated charcoal are the two commonly used solid adsorbents for continuous or integrated sampling of gaseous hazards in the workplace. Silica gel has an advantage over charcoal due to its higher collection efficiency and recovery rate of polar contaminants by a variety of common solvents (Brown and Monteith, 1995). In addition, silica gel also generally has several advantages over activated charcoal for sampling gases and vapors. Firstly, polar contaminants collected by the silica gel can be more easily removed from the adsorbent by a variety of common solvents. Secondly, amines and some inorganic substances for which the use of charcoal may not be suitable can be collected quite efficiently. Thirdly, the use of highly toxic carbon disulfide is avoided during desorption procedures.
Silica gel is a granular form of air sampler and is used mainly in sampling gases and vapors. In addition to collection of gaseous pollutants, silica gel tubes are occasionally used for acid mist collection (e.g., sulfuric acid), according to the analytical manuals of the National Institute for Occupational Safety and Health (NIOSH, 1994) and the Occupational Safety and Health Administration (OSHA, 1993). However, at room temperature, sulfuric acid and phosphoric acid are very likely to be in the aerosol phase, not in gas or vapor phase, because sulfuric acid has a vapor pressure of less than 0.001mmHg and phosphoric acid has a vapor pressure of 0.03mmHg. Moreover, very little experimental data are available on aerosol collection efficiency for the commercially available silica gel tubes
The preparation of tubes containing silica gel is documented in the appendix of NIOSH method 7903. The silica gel tube preparation procedures can be divided into two main steps, silica gel cleaning and silica gel tube packing. After finished silica gel cleaning procedure, the silica gel is packed in a glass tube. The silica gel tube length is 11 cm with an outer diameter of 7 mm and inner diameter of 4.8 mm. After sieving with 20/40 mesh, the silica gel is then packed into a glass tube. The front section is packed with 400 mg silica gel, and the back section with 200 mg silica gel. The front section is hold in place with a 6-mm diameter, 1-mm thick glass fiber filter plug, and urethane foam plugs are used between sorbent sections and the back end. More recent use has concentrated on smaller silica gel tubes (shown in Figure 1) used at room temperature. NIOSH recommends such tubes for a variety of more polar chemicals such as amines, phenols, amides and inorganic acids.
Classical filtration theory begins with an isolated fiber, and the collection efficiency of this fiber is defined by the ratio of the inlet height of the limiting particle trajectory to the fiber diameter. The theoretical aerosol penetration of a particle with n elementary charges, Pn, through a filter is normally expressed in terms of total single fiber efficiency, ES,n (Hinds, 1999)
(1)
where ES,n represents the combined effect of several individual single fiber filtration mechanisms, a is the packing density, x is the filter thickness, df is the fiber diameter, and ES,n is given by (Lathrache and Fissan, 1987; Tennal et al., 1991)
(2)
where Ed is due to diffusion (Lee and Liu, 1982); Er is due to interception (Lee and Liu, 1982); Ei is due to impaction (Hinds, 1982); Eg is due to gravitational settling (Hinds, 1982); Ep is due to dielectrophoretic force (Lathrache and Fissan, 1987; Tennal et al., 1991); Ec,n is due to Coulombic force (Lathrache and Fissan, 1987; Tennal et al., 1991); and Em,n is due to image force (Pich, 1966). The above Equation is an approximation based on the assumptions that the electret filters have a uniform charge on their fibers and all mechanisms are independent.
Similar to the single fiber theory used to predict the filtration efficiency of fibrous filters, the single grain theory is used to calculate the collection efficiency of granular bed filters (Schmidt, et al., 1978; Kogan, et al., 1993; Hinds, 1999). Aerosol particles may be collected in a granular bed filter by a number of individual filtration mechanisms, such as diffusion (Gutfinger and Tardos, 1979), interception (Gutfinger and Tardos, 1979), gravitational settling (Tardos, et al., 1979) and inertial impaction (D’Ottavio and Goren, 1983). Aerosol filtration may be supplemented by electrostatic attraction if the filter media are electrically charged. Most of the interactions among these individual filtration mechanisms have not yet been quantified and, therefore, direct summation must assume there is no interaction (Schmidt, et al., 1978; Lathrache and Fissan, 1987). D’Ottavio and Goren (1983) investigated the aerosol capture in granular beds in the impaction-dominated regime, pointing out that single grain capture efficiency depends on three dimensionless factors: the Stokes number Stk (= rp CpU/18mDG), the grain Reynolds number Re (= rDGU/m), and the packing density of the granular bed, a; here rp is the particle density, Dp is the particle diameter, Cp is the Cunningham slip correction for the particle, m is the gas viscosity, U is the superficial velocity, DG is the diameter of spherical grains, and r is density of the fluid. They also observed that aerosols exhibit bouncing and breakage at sufficient filtration velocities.
The sorption efficiency of gaseous pollutants by granular bed filters has been studied intensively. However, the aerosol filtration characteristics of the silica gel tube has had few reports and needs further study. Therefore, the main objectives of the present study are (1) to investigate the aerosol collection efficiency of each component in the silica gel tube by using polydisperse particles, (2) to determine the overall aerosol collection efficiency of the silica gel tube, and (3) to identify the major filtration mechanisms pertinent to the silica gel sorbent beds.
Experimental Materials and Methods
Two types of silica gel tubes manufactured by SKC Inc. were tested in the present study, as shown in Figure 1. The smaller one on the right was SKC 226-10 and larger one was SKC 226-10-03. Both tubes were chosen for their commercial availability and, according to the manufacturer, they were constructed in conformance to both NIOSH and OSHA. The larger one (SKC 226-10-03) was 11 cm long with an outer diameter of 7 mm and inner diameter of 4.8 mm. The packing materials, in order of flow direction, were glass wool, a filter plug, 400 mg of silica gel, glass wool, 200 mg of silica gel and glass wool. The smaller tube (SKC 226-100) was 7 cm long, with outer diameter of 6 mm and inner diameter of 4 mm. The front section had 150 mg silica gel, and there was 75 mg in the back section. Notice that the SKC 226-10-03 tubes are fabricated based on NIOSH 7903 and OSHA ID-165SG for collecting the volatile inorganic acid (HF, H2SO4…) as well as their particulate salts, whereas the SKC 226-10 tubes are used for sampling organic vapors such as amines, acetamide, tetrabormoethane, etc.. The small SKC 226-10 tubes were selected for demonstrating the shift of aerosol filtration mechanisms from diffusion to inertial impaction under very high filtration velocity, as will be discussed below.
To conduct the aerosol penetration test, aerosol particles with different size distributions were generated by two aerosol generation systems, as shown in Figure 2. A constant output atomizer (model 3076, TSI Inc., St. Paul, Minn.) and an ultrasonic atomizing nozzle (model 8700-120, Sonotek Inc., Highland, N.Y.) were used to generate polydisperse submicrometer-sized and micrometer-sized aerosol particles, respectively. For these two systems, in order to avoid the particle charge effect, the particles were then passed through a 22.5-mCi Po-210 radioactive source (model P-2001, NRD Inc.) to neutralize the aerosol particles to the Boltzmann charge equilibrium. An aerosol electrometer (model 3068, TSI, Inc.) was used downstream of the test chamber to monitor the aerosol charge neutralization. Two different spectrometers were used to measure the aerosol concentration and size distribution: a scanning mobility particle sizer (SMPS, model 3934U, TSI Inc.) for particles smaller than 0.7 mm, and an aerodynamic particle sizer (APS, model 3320, TSI Inc.) for particles larger than 0.7 mm. Potassium sodium tartrate tetrahydrate (PST) and dioctylphthalate (DOP) were chosen as the solid and liquid test agents, respectively. Five small and five large silica gel tubes were tested for both aerosol penetration and air resistance. For aerosol penetration test, five replicates were conducted for each specimen.
Testing flows ranging from 0.1 to 2.0 L/min were applied to study flow dependency. All air flows were controlled and monitored by mass flow controllers (Hastings Instruments, Hampton, VA), and calibrated using an electronic bubble meter (Gilibrator, Gilian Instrument Corp., Wayne, NJ). The pressure drop across the silica gel tube was recorded by using a pressure transducer, which was calibrated against an inclined manometer. The challenge aerosol concentration and the sampling time were not considered as important parameters affecting the aerosol penetration, because the silica gel tubes were designed for collecting liquid aerosols or organic vapor, so after loading, both the increase in packing density of the granular bed and the shift in aerosol penetration were expected to be insignificant (Chen, et al., 1998). Therefore, only the flow dependency was investigated in the present study.
The aerosol penetration through the sorbent bed could not be conducted without the supporting separator. Therefore, the aerosol penetration through separators was measured first, and then the aerosol penetration through silica gel could be obtained by back-calculation, i.e., dividing the aerosol penetration through whole tube by the aerosol penetration of the separators, as follows,
(3)
The silica gel was removed in such a way that the configuration of the separators was not changed.
The filter plug in the front part of the SKC 226-10-03 was tested for air resistance and aerosol penetration because its appearance indicated that it might be important for both aspects. The filter disc in the silica gel tube was not normally placed in a well-sealed manner, but it was difficult to characterize the arbitrary orientation of the filter disc. Therefore, we tested the filter plug only when it was snugly fitted in the tube.
Results and Discussion
The recommended sampling flow for the silica gel tube (SKC 226-10) is less than 300 cm3/min, which is equivalent to the filtration velocity of about 35 cm/sec, well within the filtration regime dominated by impaction for micrometer-sized particles penetrating through a characteristic dimension of 2 mm (Hinds, 1999). The pressure drop across the SKC 226-10 is shown in Figure 3 as a function of sampling flow and corresponding filtration velocity. The upper line indicates the pressure drop for the whole tube. The pressure drop across the whole tube and each element increased in a linear relationship with increasing filtration velocity, indicating that the flow inside the tube was laminar. The silica gel contained in both front and rear (back-up) sections contributed about 78% ((117+38) / 200 = 78%) of the air resistance. The remaining 22% of the air resistance was caused by the foam plug and two pieces of glass wool.
Under the same sampling flow, the pressure drop across the larger SKC 226-10-03 was about 3 times higher than SKC 226-10, primarily due to the larger amount of packing material (600 mg vs. 225 mg). Figure 4 shows the pressure drop across the SKC 226-10-03 as a function of sampling flow rate. The pressure drop again increased in a linear relationship with increasing sampling flow, indicating that flow was laminar. The largest share of air resistance was contributed by the 400 mg silica gel (363/569 = 64%). The front sorbent bed of 400 mg silica gel appeared to be more compact (higher packing density) than the back-up sorbent bed of 200 mg silica gel because the air resistance caused by unit mass was higher. All separators, including three pieces of glass wool and a filter plug contributed about 19% (106/569 = 19%) of the pressure drop.
Aerosol penetration through SKC 226-10 was tested using DOP particles as the challenge agent, under sampling flows ranging from 0.1 to 1 L/min. Each curve was an average of 25 replicates. As shown in the upper plot of Figure 5, the aerosol penetration (through the whole silica gel tube) around the most penetrating size (MPS, or the collection minimum) was about 80%. Generally speaking, the penetration curve shifted to the upper left as the flow rate increased. This is because, as the flow rate increased, the inertial impaction became more significant and resulted in lower aerosol penetration for particles larger than the most penetrating size. For particles smaller than the most penetrating size, the increasing flow rate shorten the retention time, which is critical for aerosol deposition by diffusion, and therefore resulted in progressively higher aerosol penetration.
The aerosol penetration through silica gel was obtained by using Equation 3 and shown in the bottom plot of Figure 5. The aerosol penetration patterns of separators and silica gel were similar to that of the whole tube. Yet, the aerosol penetration of separators was slightly lower than that of silica gel, indicating that the separator(s) might collect more aerosol particles than the silica gel. Also, the most penetrating size (MPS, also referred to as collection minimum) was about 0.2 to 0.6 mm, and decreased with increasing sampling flow. The maximum of the aerosol penetration of the MPS was around 90%.
The aerosol penetration through the whole tube of SKC 226-10-03 was much lower than that through SKC 226-10 apparently due to the larger amount of silica gel materials, as shown in the upper plot of Figure 6. The aerosol penetration curves moved toward the upper left as the sampling flow increased, similar to that of SKC 226-10. The aerosol penetration rates of the collection minimum were only about 4-5%. The aerosol penetration patterns of separators and silica gel were quite similar, with the penetration maximum around 20-25%.
Figure 7 shows the pressure drop across the silica gel tube as a function of loading time. The target silica gel tube was SKC 226-10 and the sampling flow was 0.1 L/min. The increasing trend in pressure drop was much more significant for silica gel tube challenged with solid PST particles. Whereas, the increase in pressure drop was less noticeable when challenged with liquid DOP particles. Note that, the size distribution and the mass concentration were not exactly the same between solid and liquid challenge particles (1.88 vs. 0.85 mg/m3). Yet the difference in the rate of increase in pressure drop was too significant to neglect. This is mainly because the solid particles are likely to form dendrites, which tend to increase the air resistance, and in theory collect more incoming aerosols at the same time (Bhutra and Payatakes, 1979; Okuyama and Payatakes, 1981). In contrast, the liquid aerosols may form as a layer of coating film, which can cause only slight increase in air resistance due to the slight increase in packing density (Kirsch, 1978). For the large SKC 226-10-03 tubes, they are designed for sampling liquid acid mists and, therefore, the loading curve is expected to be similar with the DOP curve.
The pressure drop across the filter plug linearly increased with increasing flow rate, as shown in Figure 8. Under the snugly fit condition, the pressure drop across the filter plug (352 mmH2O, Figure 8) was equivalent to silica gel of 400 mg (363 mmH2O, Figure 4). The air resistance induced by the filter plug was about 9 times higher than the rest of separators made of glass wool. The aerosol penetration of submicrometer-sized particles through this filter plug was in the range from 5% to 8% when tested at 0.3 to 0.5 L/min, as shown in Figure 9. This aerosol penetration was only slightly higher than the 5% of the penetration maximum (or collection minimum) shown in the upper plot of Figure 6.
In testing the SKC 226-10, we found that the entire penetration curve shifted to the upper left with increasing flow rate (shown in Figure 5), but the rate of increase in aerosol penetration decreased if the flow rate continued to increase. We then investigated whether there would be particle bounce-off for large particles and a shift in the collection mechanism from diffusion to inertial impaction for small particles, in particular particles smaller than the penetration maximum.
It should be noted that the pressure drop across the whole SKC 226-10 under 1 L/min was almost the upper limit for the Scanning Mobility Particle Sizer. In order to investigate the filtration characteristics of the granular sorbent bed of silica gel under high filtration velocity, only the backup section of SKC 226-10 was tested for aerosol penetration under sampling flow ranging from 0.3 to 2 L/min. The corresponding filtration velocity of the highest sampling flow of 2 L/min was 116 cm/sec, high enough to enhance the effect of inertial impaction. Figure 10 indeed shows the expected decrease in aerosol penetration of particles smaller than the penetration maximum when the sampling flow increased from 1.5 to 2.0 L/min, an evidence of inertial impaction exceeding diffusion deposition. However, the bounce-off of the large particles was not evident, probably due to the fact that the silica gel (75 mg) bed was too thick, and therefore the downstream portion of the sorbent bed and the following foam piece would probably have collected the particles that bounced off in the front portion of the silica gel.
Aerosol penetration efficiency is strongly dependent on the size distribution of the challenge aerosols. For micrometer-sized particles, inertial impaction and gravitational settling tend to be the dominant filtration mechanisms, while for submicrometer-sized particles, diffusion is the principal collection mechanism. The size distributions of aerosol particles in the workplace can be very diverse, ranging from monodisperse (GSD < 1.2) to very polydisperse (GSD>>3). To cover both ends, GSD of 2 was used in the following simulations. Based on the data of aerosol penetration through separators and silica gel layers, two simulations of the aerosol measurements were performed for SKC 226-10, as shown in Figure 11. The first simulation had a count median diameter (CMD) of 0.3 mm and the second one has a larger CMD of 1.0 mm, both with the same geometric standard deviation of 2.0. In order to demonstrate the size dependency, both challenge aerosols were assumed to have the same weight of 100 g. For the first case of CMD of 0.3 mm, the glass wool collected about 50% of the particles in terms of weight; the following sorbent layer held only 11%, the second glass wool retained another 9%, with the backup sorbent layer intercepting 2.66%, and the last foam with trivial 0.38%. Overall 25% of the test agents penetrated. Whereas for the case of CMD of 1.0 mm, a surprisingly high (92%) of the test agents were collected by the first piece of glass wool, and only 2% penetrated the back-up foam. For both cases, the highest collection efficiency was found in the first piece of glass wool. Silica gel layers collected less than 15% of the challenge agents.
Conclusions and Recommendations
With respect to overall aerosol collection efficiency, the SKC 226-10 apparently is not suitable for collecting aerosol particles because of poor aerosol collection efficiency. The SKC 226-10-03 might be acceptable for collecting aerosol particles, because the collection efficiency was up to 95% for the most penetrating size.
Aerosol penetration through silica gel tubes is strongly dependent on aerosol size. For particles larger than 1 mm, the aerosol penetration decreased as the filtration velocity increased, apparently due to stronger inertial impaction. For particles smaller than 1 mm, the aerosol penetration increased with increasing filtration velocity due to shorter retention time for aerosol deposition by diffusion. For filtration velocity over 116 cm/sec, the aerosol penetration decreased with increasing filtration velocity, apparently due to the shifting of principal filtration mechanism from diffusion to inertial impaction.
In terms of aerosol collection efficiency and air resistance, the filter disc in the front part of the SKC 226-10-03 might play a significant role if snugly fitted in the silica gel tube. The filter disc was found to collect over 90% of aerosol particles of all sizes. That means only less than 10% of the challenge particles had a chance to penetrate to the downstream silica gel section. Therefore, it is important to include this filter plug in the further analytical procedures, although the filter plug was seldom in the perfectly sealed position. This filter plug itself also caused a significant air resistance if snugly fitted in the tube, which exerted unnecessary loading on the personal sampling pump. We therefore recommend not using this filter plug as the separator.

Bhutra, S. and Payatakes, A.C. (1979). Experimental Investigation of Dentritic Deposition of Aerosol Particles. J. Aerosol Sci. 10:445-464.
Brown, R.H. and Monteith, L.E. (1995). Gas and Vapor Sample Collectors” in Air Sampling Instruments — for evaluation of atmospheric contaminants, American Conference of Governmental Industrial Hygienists. p369-400.
Chen, C.C., Lai, C.Y., Shih, T.S. and Yeh, W.Y. (1998). Development of respirable aerosol samplers using porous foams, American Industrial Hygiene Association Journal, 59(11): 766-773.
D’Ottavio, T., and Goren, S.L. (1983). Aerosol Capture in Granular Beds in the Impaction Dominated Regime, Aerosol Sci. Technol. 2: 91.
Gutfinger, C. and Tardos, G.L. (1979). Theoretical and Experimental Investigation on Granular Bed Dust filters, Atmos. Environ. 13: 853.
Hinds, W.C. (1999). Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles, 2nd Edition, John Wiley and Sons, Inc., New York, p 182-205.
Kirsch, A.A. (1978). Increase in Pressure Drop in a Model Filter during Mist Filtration. J. Coll. Int. Sci. 64:120-125.
Kogan, V., Kuhlman, M.R., Coutant, R.W., and Lewis, R.G. (1993). Aerosol Filtration by Sorbent Beds, J. Air Waste Manage. Assoc. 43:1367-1373.
Lathrache, R. and Fissan, H.J. (1987). Enhancement of Particle Deposition in Filters due to Electrostatic Effects. Proc. of Filtration Society. 418-422.
Lee, K.W. and Liu, B.Y.H. (1982). Theoretical Study of Aerosol Filtration by Fibrous Filter. Aerosol Sci. Technol. 1: 147-161.
National Institute for Occupational Safety and Health (1994). NIOSH Manual of Analytical Methods, 4th ed. DHEW (NIOSH) Pub. No. 94-113.
Occupational Safety and Health Administration (1993). OSHA Analytical Methods Manual. OSHA Analytical Laboratories, Salt Lake City, UT. Available from ACGIH, Cincinnati, OH (1985); (Supplements 1991, 1993).
Okuyama, K. and Payatakes, A.C. (1981). Comparison between Theory and Experiment in Dendritic Deposition: a Revision. J. Aerosol Sci. 12:269-274.
Pich, J. (1966). Theory of Air Filtration by Fibrous and Membrane Filters, in Aerosol Science, C.N. Davies, ed., Academic Press, London and New York.
Schmidt, E.W., Gieseke, J.A., Gelfand, P., Lugar, T.W., and Frulong, D.A. (1978). Filtration Theory for Granular Beds. J. of the Air Pollution Control Association. 28(2):143-146.
Tardos, G.I., Yu, E., Pfeffer, R., and Squires, A.M. (1979). Experiments on Aerosol Filtration in Granular Sand Beds, J. Colloid and Interface Sci. 71: 616.
Tennal, K.B., Mazumder, M.K. Siag, A. and Reddy, R.N. (1991). Effect of Loading with an Oil Aerosol on the Collection Efficiency of an Electret Filter. Particulate Science and Technology 9: 19-29.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top