United States Environmental Protection Agency Environmental Sciences Research Laboratory Research Triangle Park NC 27711 Research and Development EPA-600/S7-8M29 Oct. 1981 Project Summary Photoacoustic Detection of Particulate Carbon R. R. Patty and C. A. Bennett, Jr. A photoacoustic technique for mass monitoring of carbonaceous aerosols deposited on filter substrates has been developed. A photoacoustic cell has been designed and calibrated using laboratory generated elemental carbon standards. The nature of the photoacoustic response is examined at several modulation frequencies using these calibration standards, and the physical principles necessary for an adequate interpretation of the experimental results is presented in detail. Practical considerations con- cerning ambient carbon monitoring are outlined; in particular, the pertur- bation due to the presence of scatter- ing particles is examined, and limited experimental quantification of this perturbation is reported. This Project Summary was develop- ed by EPA's Environmental Sciences Research Laboratory. Research Tri- angle Park, NC, to announce key find- ings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Photoacoustic detection has been successfully used in the past as a means of measuring levels of pollutant gases in the atmosphere. The same principle has been used to construct a sensitive real time photoacoustic soot monitor which measures levels of suspended carbon particles within an acoustically reso- nant photoacoustic cell. Other types of chemical species present in the atmos- phere in paniculate form are typically analyzed by the x-ray fluorescence of ambient participate samples which have been deposited on various filter substrates by dichotomous samplers. Unfortunately, the atomic weight of carbon is too low to allow quantitative results from this very sensitive nonde- structive technique. As a consequence, a great deal of data exists, from both past and on-going studies, in the form of ambient particulate samples deposited on filter membranes and for which no information on elemental carbon is available. Soot represents a significant fraction of any ambient particulate sample, either rural or urban, and this fraction is certain to increase as our oil supplies diminish and more and more demand is placed on "less clean" energy sources. Since particulate carbon has been related to adverse health effects, possible climatic pertur- bations and visibility degradation, it is important to establish a procedure by which accurate elemental carbon levels in the atmosphere can be obtained so that appropriate control strategies can be implemented. Perhaps the most common nonde- structive method of elemental carbon detection for samples collected on filter substrates is the integrating plate method (IPM). Although this technique is very simple and economical, experi- ence in our laboratory has indicated that IPM is seriously affected by the pres- ence of nonabsorbing scattering particles such as ammonium sulfate. The photoacoustic technique we have developed is much less effected by scat- ------- tering particles and is also nondestruc- tive, simple, and economical. Even if scattering particles were not present in ambient samples, photoacoustic detec- tion (PAD) would still be much more preferable than IPM. The limit of detec- tion of IPM lies far above that of PAD for the following reason: IPM involves a dif- ference of transmission measurements on filters before and after loading. If the light source - detection system stability is, for example, 1%, then one could not hope to make an IPM measurement on a sample which is 1 % attenuating with an accuracy of any better than + 100%. The photoacoustic signal, on the other hand, is linearly dependent on the light source intensity so that a 1% fluctuation in intensity produces a 1% fluctuation in accuracy of any better than ± 100%. The photoacoustic signal, on the other hand, is linearly dependent on the light source intensity so that a 1% fluctuation in intensity produces a 1% fluctuation in photoacoustic signal. Indeed, examples of photoacoustic measurements on samples with attenuations of 10~4-10~5 abound in the literature. It is clear, then, that the use of PAD will lead to de- creased sample collection times and make possible more accurate measure- ments in clean air environments. The photoacoustic effect in solids is a fascinating subject in its own right. As is evident from our data on pure carbon calibration standards, the very highly absorbing (and hence very thin) carbon' samples prepared for this study illus- trate a structure in the photoacoustic signal due to thermal wave interference. Since ambient samples are typically in the form of very thin deposits, it is essential to have a firm understanding of these phenomena if experimental data obtained by PAD is to be accurately interpreted. For example, thermal wave interference (and the resulting struc- ture in the photoacoustic signal) is highly dependent upon the type of filter substrate used; therefore, it is neces- sary to make separate calibrations for each type of filter material. Results The physical principles underlying the production of the acoustic signal gener- ated within the photoacoustic cell are outlined in detail. The equation fo: the photoacoustic signal (in one dimension) is established via the traditional boundary value approach and by a less vigorous, but more illustrative, ap- proach which better illustrates the physics of the photoacoustic effect. Additional computations are also in- cluded which account for the added photoacoustic signal due to light re- flected from the filter surface (this effect has been ignored in the past ). An analysis of the IPM technique is in- cluded as well as a discussion of the decrease in transmission observed when the particles are sandwiched between the filter and the opal glass. In order to make it possible to obtain measurements on ambient elemental carbon, a set of accurately weighed and evenly deposited elemental carbon cali- bration standards were prepared using Teflon filters. From IPM measurements on these pure soot calibration filters, we determine the absorptance A = (1 - T) where T is the fraction of incident radia- tion transmitted by the carbon particles contained on a filter, and the optical thickness /3I = -ln(T) where ft is the absorption coefficient and I is the thick- ness of the sample. Figure 1 shows both a plot of absorptance vs loading and a plot of ft\ vs loading. Our measured photoacoustic signals for the same calibration filters used for Figure 1 are shown in Figure 2. A comparison of the photoacoustic with the absorptance data suggests (a) that the photoacoustic signal does not saturate with loading as • quickly as does the absorptance, and (b) that there is more scatter in the photo- acoustic data than in the absorptance data. We should mention that if the same filter is repetitively inserted and sampled, the photoacoustical signal for each measurement agrees to within 3%; the error associated with the load- ing measurements is contained within the scatter present in the data shown in Figure 1. It appears that the major source of the scatter in the photoacous- tic data stems from the Teflon filters which were not thermally thick at 100 Hz. Different filters allow variable amounts of thermal energy to be trans-, mitted to the gas from behind the filters. The interesting fact that the photo- acoustic signal does not saturate as the carbon samples become opaque can be explained by considering the interfer- ence of thermal waves generated within the very thin carbon samples. Appar- ently, the thermal waves reflecting from the sample filter backing are shifted in phase by 180° so that the thermal waves at the sample-gas boundary interfere destructively for the thinner samples and interfere more and more 10.0 40 80 120 160 Loading (ng/cmz) Figure 1. Plots of absorptance vs loading and optical thick- ness /3I vs loading. increases, thus causing the photo- acoustic signal to continue to rise instead of reaching saturation as the absorptance approaches unity. Conclusions Photoacoustic detection appears to be a powerful technique for ambient carbon monitoring. The thermal wave mechanism associated with the produc- tion of the acoustic signal introduces additional structure not present when using the optical transmission IPM method. Although, at first glance, this additional structure may appear to be a complication, it is possible that the information provided by this structure will prove useful. It appears that the photoacoustic technique is much less perturbed by the presence of scattering particles than is the IPM method; how- ever, the effect is significant for both methods in the range of loadings ex- pected for ambient samples. The photo- acoustic technique seems to provide a better estimate of the true carbon mass present in ambient samples than does the IPM method. Ultimately, the effect of scattering particles must be theoreti- I I si § 4.0 0.0 *• 120 160 40 80 Loading Figure 2. Plot of photoacoustic re- sponse vs loading for a 1OO Hz chopping frequency. ------- cally and experimentally quantified so that accurate carbon levels may be deduced. Recommendations The photoacoustic technique prom- ises to be an extremely accurate method for determining the absorbing compon- ent of atmospheric aerosols. More work is needed to adequately establish the role of scattering particles in relation to particle size, etc. In the future, more accurate photoacoustic results will be available if suitable attention is given to quality control of the filter substrates. It would be exceedingly attractive to fully automate a photoacoustic system (possibly with a capability for simulta- neous IPM measurements) which could be put into routine operation. R. R. Patty and C. A. Bennett, Jr., are with North Carolina State University, Department of Physics, Raleigh, NC 27650. W. A. McClenny is the EPA Project Officer (see below). The complete report, entitled "Photoacoustic Detection of Paniculate Carbon," (Order No. PB 81-245 425; Cost: $8.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Environmental Sciences Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC 27711 U S GOVERNMENT PRINTING OFFICE, 1981 — 757-012/7363 ------- United .States Center for Environmental Research Environmental Protection Information cnui,-.,m-nt«i A8ency Cincinnati OH 45268 Protection Agency EPA 335 Official Business Penalty for Private Use $300 PS ------- |