Evaluation of thermal extraction GC MS for estimation of emissions from SBR dryers.

The determination of basic air pollutants emitted by a manufacturing facility is extremely important as legislative acts, such as the 1990 amendments to the Clean Air Act, take effect. It will be necessary to reduce emissions from industrial and mobile sources to meet the requirement of the Clean Air Act. The analytical techniques used to ensure that allowed emissions are not exceeded must provide sensitive and definitive measurements of volatile organic compounds (VOCs).

The rubber industry usually measures emissions from dryers by stack gas sampling. Results obtained by this technique are usually very erratic due to difficulties in sampling a gas stream. Parson and Mitzner (ref. 1) have described the use of gas chromatography for analysis of stack emissions. Their technique was used for analysis of relatively simple mixtures of stack emissions, whereas, in many such cases, emissions of several dozens of individual components are involved.

In 1992, the International Institute of Synthetic Rubber Producers, IISRP, source emissions analytical ad-hoc committee developed the static ambient temperature extraction (SATE) test for estimation of styrene emissions from SBR dryers (ref. 2). Crumb samples from the dryer are collected and extracted in isopropyl alcohol (IPA). The EPA extract is analyzed using GC with an FID detector. The test estimates styrene emissions from the difference in residual styrene content of the crumb collected from dryer feed and dryer exit.

In addition to small amounts of parent monomers, synthetic polymers contain residues of auxiliary materials such as antioxidants, organic acids and oils. It is essential to extend the SATE test to include the VOCs originating from these raw materials.

A few reports have described methods for evaluation of organic vapors released from rubber (refs. 3-5) and plastic (ref. 6) under mild heating. In addition, the application of on-column cryogenic focusing to the analysis of volatiles released from plastic has also been reported (ref 7).

This present work shows that thermal extraction coupled with GC/MS and on-column cryogenic focusing can be used for identification of the major VOCs produced from styrene/butadiene rubber. Coupled with the SATE test, it can also be used to estimate VOCs emitted from dryer stacks in an SBR plants.

Experimental

Instrumentation

The instrumentation used for this study consisted of a Ruska Laboratories ThermeEx inlet system interfaced to a Hewlett-Packard 5890 Series 11 gas chromatograph fitted with a 30 meter, 0.32 mm id DB-5 column and Hewlett-Packard 5971 Mass Selective Detector. The ThermEx unit heats the samples in fused quartz crucibles in a stream of helium. The evolved compounds then pass through a transfer line into the inlet of the gas chromatograph.

The ThermEx unit was programmed with an initial temperature of 55[degrees]C for two minutes to sweep air and water from the system. The sample crucible was then heated to 100[degrees]C and held for ten minutes, then allowed to cool back to 55[degrees]C. The transfer line to the GC was maintained at 325[degrees]C. The effluent from the Thermex was split 1:30 at the GC injection port. The GC was programmed to hold -10[degrees]C for ten minutes, then ramped to 150[degrees]C at 10[degrees]C/minute. The mass spectrometer was operated in scan mode, scanning the mass range from 39 to 550 amu.

Procedure

Samples were collected and stored using the same protocol described in the SATE test (ref. 2). Crumb samples were collected when the dryer was operating at equilibrium conditions taking into account dryer holdup time. Enough crumb was collected to completely fill an 8 oz. jar. The jar was immediately sealed to minimize loss to the head space. The crumb samples were brought to the lab and allowed to cool to room temperature. Samples not analyzed within three hours were refrigerated. Stack gas sampling was done by Southwestern Laboratories, Houston. TX, on two consecutive days. During the same period, duplicate samples of crumb were collected from the dryer feed and dryer exit for analysis by thermal and solvent extraction.

To determine the recovery efficiency of styrene, 100 mg of crumb were placed in a quartz crucible and manually inserted in the sample loader assembly of the thermal analyzer. During the sample loading period, the sample holder was cooled to room temperature to avoid loss of volatile materials on contact with the sample holder and carrier gas flow was cutoff to prevent air from entering the mass spectrometer. Once sample and crucible were inserted in place, the system was sealed and flow of helium to the pyrocell was resumed. The sample was then heated to 100[degrees]C at a rate of 30[degrees]C/ minute. The volatile organic compounds emitted were swept out of the pyrocell through the transfer line into the injection port. The temperature of the transfer line was kept at 325[degrees]C.

To determine the total residual styrene content of crumb, 0.5 g of crumb was cut into small pieces and accurately weighed. 50 mL of tetrahydrofuran (THF) was pipetted into an 8 oz. jar and tightly capped. Alpha-methyl styrene was used as an internal standard. The 8 oz. jar with solvent and sample was placed on a Eberbach shaker until the rubber was completely dissolved. The sample was then removed from the shaker and one micro liter of the solution was manually injected into the GC-MS. The concentration of the residual styrene was calculated using the following equation:

Residual styrene, ppm = (X1*X2*X3* 100/%X4)/(X5*X6) Where:

X1 = micrograms of alpha-methyl styrene in internal standard;

X2 = styrene peak area;

X3 = response factor for styrene;

X4 = extraction efficiency;

X5 = alpha-methyl styrene peak area;

X6 = crumb rubber sample weight in grams.

Results and discussion

The thermal analyzer used in this work can be compared to a combination of headspace, purge trap and thermal extraction system. The approach taken in this development was driven by three major factors. First, to develop a relatively easy test for characterization of the VOCs emitted from crumb, so that the components identified might be quantified using the SATE test. Secondly, to explore the possibility of developing a solvent free test for the extraction of crumb samples eliminating the time and expense incurred in the sample preparation and the expense of solvent disposal. Finally, to see if thermal extraction coupled with the on-column cryofocusing could be used to estimate styrene emissions from SBR dryers. To this end, effects of experimental parameters such as size of crumb, extraction time, rubber additives and recovery efficiency were examined to establish an optimum set of conditions.

Sample size

In order to minimize errors due to the selection of unrepresentative samples size, it was essential to find the optimum size for thermal extraction. The upper limit on the sample size was governed by the capacity of the furnace and rate of heat transfer. The lower limit was dependent on sample homogeneity and weighing accuracy. Sample sizes of 10 mg to 100 mg were found to be most suitable. A plot of detector response vs. weight of crumb rubber used yielded a straight line with a correlation coefficient r = 0.99 (figure 1).

Extraction temperature

Extractions were conducted at 40, 80, 100 and 135[degrees]C using a 100 mg crumb sample. The peak areas increase with temperature up to 100[degrees]C. There was no significant change when the temperature was further increased (figure 2). It was concluded that 100[degrees]C is the optimum temperature under the experimental conditions employed in this work.

Extraction time

The detector response increased with the time allowed for extraction for up to five minutes. No significant increase in peak area was observed when the time was increased further.

Recovery efficiency

In order to determine the recovery efficiency of thermal extraction (defined in this study as the PPM of styrene obtained by thermal extraction of crumb divided by the PPM of styrene obtained in THF dissolution), approximately 100 mg of rubber were cut into small pieces, placed in a crucible and accurately weighed. The sample was then thermally extracted using the procedure outlined above.

This study shows that, approximately the same amount of styrene is recovered by the thermal extraction procedure and THF-dissolution, giving a recovery efficiency of over 95%.

Effects of rubber additives

Effects of percent carbon black, percent oil and percent bound styrene on recovery efficiency of styrene were also studied. Varying the oil and carbon black content of the crumb from 0 to 40% had no effects on styrene recovery efficiency. Percent bound styrene in polymer also has no significant effect on the recovery efficiency of styrene in crumb (table 1).

Table 1 - effects of carbon black,oil and bound
 styrene on the recovery efficiency of crumb
% Bound   % Oil   % Carbon   % Recovery
styrene             black     efficiency
  23.5     0.0       0.0          96.8
  23.5    27.2       0.0          97.3
  23.5    22.0      33.0          97.4
  40.0    27.6       0.0         101.0

Comparison of results obtained by thermal extraction with those obtained by SATE and stack gas sampling The correlation of the result of thermal extraction with the SATE test and the dryer stack gas sampling test was verified by analyzing crumb samples collected from the dryer feed and dryer exit. The limited results obtained in this work show that, if crumb samples from dryer feed and exit are collected when the dryer is operating under equilibrium conditions and loss of styrene to headspace is kept to a minimum, a good correlation can be obtained between results obtained by these test methods (table 2 and figure 3).

Table 2 - styrene emissions (lb/hr)(*)
SATE     Stack gas       Thermal
          sampling     extraction(**)
3.02       3.531           3.33
2.86       3.110           3.07
3.09       2.921           3.30
3.10                       3.31
(*) Styrene-butadiene rubber with 32% carbon black
and 6% oil.
(**) Standards for the thermal extraction were prepared
using the SATE test

The VOCs emitted from SBR can easily be identified using the thermal extraction-GC/MS system used in this work. Table 3 lists some of the common VOCs identified. This technique can also be used to identify the source of a given VOC in a finished product by direct comparison of the total ion chromatogram (TIC) of the VOCs produced from the product with those produced from raw materials added during manufacturing. Figure 4 illustrates this point, by comparing the TIC of the VOCs emitted from a carpet underlay and the SBR sample used in its formulation.

Table 3 - common VOCs emitted from SBR
Compound                                  Comments
4-vinyl cyclohexene                     Butadiene dimer
1,5 cyclooctadiene                      Butadiene dimer
1,2-divinyl cyclobutadiene              Butadiene dimer
Methyl benzene
Styrene
Xylene (o, p & m)
Propyl benzene
Benzaldehyde
Cyclohexane, 1-methyl-4-(1-methylethyl)
Napthalein
Benzothiazole
4-phenyl cyclohexene               Styrene-butadiene dimer
Indenes
Tetramethylthiourea

Conclusion

Results of this work show that thermal extraction-GC/MS coupled with on-column cryogenic focusing has the necessary precision to be used for estimation of styrene emission from SBR dryers. The technique allows the determination of emission level without solvent extraction, eliminating the time and expense incurred in sample preparation and the expense incurred in solvent disposal.

The ThermEx can serve as a combination of headspace, purge and trap and thermal extraction. It is essential, however, to remember that extremely volatile compounds can be lost during sample introduction. Analysis of crumb with high amounts of water may require a pre-drying stage which may result in lost volatiles. Finally, for samples with very low styrene content it may be necessary to find ways of extracting larger samples to obtain a large enough signal.

References

[1.] Parsons, J.S. and Mitzner S., Environ, Sci. Technol. 9, 1053 (1975). [2.] IISRP Source Emission Analytical Ad-hoc Committee: "Determination of residual styrene in crumb rubber" by L. Levi, R. Wu, G. Ghebremeskel, C Leonard, G. Frisone, G. Collins, D. Daugherty and T. Gurley, Houston, TX (1992). [3.] Willoughby, B.G., Rubber and plastics processing, Q.M.C Symposium on "Airborne hazards at work, " London, (1976). [4.] Willoughby, B.G., and Lawson, G., Laboratory,, vulcanization as an aid to factory air analysis," ACS Rubber Division, 116th meeting, Cleveland Ohio, (1979). [5.] Hu, J. CA, "Chromatography analysis of rubber and other high polymers, " J. of Chrom. Sci. 19: 634 (1981). [6.] Rigby, C.J., "The collection and identification of toxic volatiles from plastic under thermal stress," Am. of Occup. Hyg. 3:331 (1981). [7.] Moncun, J.G., TE. Sharp and E.R. Byrd, "Focused cryogenic trapping for dynamic headspace and pyrolytic analysis of polymers on fused silica capillary column," J. of HRC and CC. 4: 603 (1981).