ANALYST, JANUARY 1992, VOL. 117 9 Application of Tryptamine as a Derivatizing Agent for the Determination of Airborne lsocyanates Part 5.* Investigation of Tryptamine-coated XAD-2 Personal Sampler for Airborne lsocyanates in Workplacest Weh S. Wu and Virindar S. Gaind Occupational Health Laboratory, Ontario Ministry of Labour, I0 I Resources Road, Weston, Ontario, Canada M9P 3T1 The development of an efficient solid sorbent personal sampler with increased convenience for sample collection in workplaces is described. Several solid sorbents were coated with tryptamine, and sampling tubes were prepared with the coated sorbents. These tubes were evaluated for the collection of phenyl isocyanate vapour generated in a commercial test atmosphere generation system that permits the simultaneous collection of up t o 12 uniformly loaded samples.Tryptamine-coated XAD-2 resin was shown t o be the most efficient solid sorbent for the collection of airborne phenyl isocyanate. The optimum amount of tryptamine needed for coating XAD-2 resin was investigated. Keywords: Personal sampler; airborne isocyanate; tryptamine-coated XA 0-2 For more than a decade, many occupational health labora- tories have attempted to develop a method capable of determining both monomeric and polymeric isocyanates for isocyanate monitoring programmes in workplaces. Despite the slow progress in this area, the method of Bagon et al.1 provided a significant advance. The later method of Wu and co-workers,2-4 based on the concept of the isolation of a selected x-system in a derivative for specific detection,5 lowered the detection limit and also raised the confidence in the analytical results by using two specific detection systems after separation by high-performance liquid chromatography (HPLC).In our earlier work, the airborne isocyanates were collected in impinger solutions containing tryptamine, result- ing in the formation of the corresponding urea derivatives. Determination via the indolyl groups of tryptamine was carried out on a reversed-phase HPLC system equipped with both fluorescence emission and amperometric oxidation detectors. This method is unique because the determination of tryptamine-derivatized isocyanates can be achieved by calib- rating against a single standard such as tryptamine-derivatized toluene diisocyanate.Such manipulation may be necessary for the determination of polymeric isocyanates as information regarding either the type of polymer or the degree of polymerization is often not available. With efforts being concentrated on a search for methods that are capable of determining total isocyanates, researchers have paid less attention to the development of a convenient personal air sampler. Nearly all of the personal samplers for the collection of airborne isocyanates have been limited to measuring the common monomeric isocyanates such as toluene diisocyanate (TDI) ,fF-8 methylene diphenyl diisocy- anate (MDI)9-11 and hexamethylene diisocyanate (HDI).* A diffusive sampler for personal monitoring of TDI has been described and evaluated in the field, but its utilization for polymeric isocyanates was not investigated.12713. There is only one publication14 on the sampling of polymethylene poly- phenylene isocyanate (PMPPI), but its determination was extremely cumbersome. As the personal sampler is attached to a worker, the sampling medium should ideally be a solid sorbent. Unfortunately, this implies that the derivatization of isocyanate during sampling would occur at the solid phase of * For Part 4 of this series, see reference 19. t Presented in part at the 104th AOAC Annual International Meeting and Exposition, New Orleans, LO, USA, 1990. the coated derivatizing reagent in the sampler, for which the efficiency of derivatization is hindered. Tryptamine has previously been demonstrated to be an isocyanate-derivatizing reagent that gives very reliable results for total isocyanate determination.The feasibility of sampling airborne isocyanate using tryptamine-coated solid sorbents for both monomeric and polymeric isocyanates has been investi- gated's.Various types of common solid sorbents, including a glass-fibre filter, were studied for their efficiency in sampling. In order to validate the use of tryptamine-coated solid sorbents in sampling tubes for the collection of airborne isocyanates, a test atmosphere was generated using a commer- cial test atmosphere generation system (TAGS). It was neither practical nor necessary to conduct the simulation of air sampling for all industrial isocyanates owing to their very low vapour pressure and lack of purity. Therefore, phenyl isocyanate was selected as a model for the generation of test atmospheres.Experimental Chemicals and Solid Sorbents Tryptamine was supplied by Sigma (St. Louis, MO, USA) and phenyl isocyanate by Aldrich (Milwaukee, WI, USA). Tryptamine-derivatized phenyl isocyanate used as the calibra- tion standard was synthesized in the laboratory. Amberlite XAD-2, XAD-4 and XAD-7 were purchased from BDH (Toronto, Ontario, Canada). Molecular sieve 13X (45-60 mesh) was obtained from Supelco (Bellefonte, PA, USA) and Glass Beads (80-100 mesh) from Chromatographic Special- ities (Brockville, Ontario, Canada). Glass-fibre filters (Type N E , 37 mm, Cat. No. 225-7) were purchased from SKC (Eighty Four, PA, USA). Silica gel and charcoal were obtained from the corresponding sampling tubes manufac- tured by SKC (Cat.No. 226-22 for the former and 226-16 for the latter). A nylon-66 filter (0.45 pm) was supplied by Rainin Instrument (Woburn, MA, USA). Apparatus The operating conditions of the HPLC system and the associated fluorescence detector have been described in previous parts of this series.2-410 ANALYST, JANUARY 1992, VOL. 117 I solution Fig. 1 Schematic diagram of the TAGS: A, vapour generator; B, cone-shaped chamber; C, multiple sampling ports; D, critical orifices; E, metal bellows pump; F, filter; P, pressure gauge; and R, pressure gauge and regulator. Connections to 1, 2 and 3 of the ration line are identical Solid sorbent sampling tubes Glass tubing (7 cm x 5 mm i d . ) was filled with various solid sorbents to a height of 3 cm with small glass-wool plugs at both ends.Test atmosphere generation system This system for the simulation of air sampling was manufac- tured by SRI International16 (Menlo Park, CA, USA) and is outlined in Fig. 1. It consisted of the following major sections: (a) vapour generator, where the vapour of phenyl isocyanate was generated (at room temperature) by passing compressed nitrogen through liquid phenyl isocyanate; (b) dilution tower, where the incoming phenyl isocyanate vapour carried by the compressed air was further mixed with a large volume of compressed air; (c) cone-shaped chamber, where the mixture of phenyl isocyanate vapour and compressed air was dynamic- ally homogenized; (d) multiple sampling ports, where a maximum of twelve samples can be collected simultaneously; (e) critical orifices, where the flow rate for air sampling was controlled; and (f) vacuum exhaust system, where a vacuum pump was operated for exhaustion.The amount of phenyl isocyanate collected was adjusted by varying the sampling time andor by adjusting the pressure of nitrogen at the vapour generator. It is practical to allow the system to equilibrate for a minimum of 30 min before sampling, which ensures that the whole system has reached dynamic equilibrium. It should be kept in mind that in most instances, the exact time period for sampling on each set of samplers is not crucial, as all recoveries are with reference to the impinger solutions in the set. Preparation of Sampler Containing Tryptamine-coated Solid Sorbent In order to minimize the variation of packed contents between individual samplers, a batch of tryptamine-coated solid sorbent, sufficient for a minimum of ten samplers was prepared.The coating of solid sorbent was carried out by dissolving tryptamine in acetonitrile, mixing with the solid sorbent and removing the solvent on a rotary evaporator under water aspiration. The evaporation was initially conduc- ted at room temperature for 20 min and the temperature was raised to about 40°C until no visible condensation of acetonit- rile was observed. Excessive amounts of unremoved acetonit- rile caused undesirable sputtering during sampling. All precautions were taken to avoid large variations within a batch of tubes being packed. I Time - Fig. 2 HPLC analysis for phenyl isocyanate sampled by a tryptam- ine-coated XAD-2 tube on the TAGS.Column, CSC-Hypersil-ODS (5 km); flow rate, 0.8 ml min-1; eluent, acetonitriIe-0.6% ammonium acetate (55 + 45); retention time, 8.68 min for PI-TP (11.5 ng) The impregnation of tryptamine on glass-fibre filters was performed in a beaker where the filters were soaked with a minimum amount of acetonitrile solution containing tryptam- ine. The filters were air dried at room temperature before use. Investigation of Sampling Efficiency on Solid Sorbent Sampler The investigation was conducted by using the TAGS, where parallel sampling was used on various types of tryptamine- coated solid sorbent samplers. After sampling, the sorbent from each sampler was emptied into a vial and the tryptamine derivative was desorbed with 5 ml of acetonitrile.A portion of the solution was filtered using a 0.45 pm pore size nylon-66 filter. The filtrate was diluted appropriately with acetonitrile before analysis by HPLC. Recovery of Tryptamine-derivatized Phenyl Isocyanate Spiked on XAD-2 and XAD-4 XAD-2 and XAD-4 sampling tubes were individually spiked with 4.44 pg of tryptamine-derivatized phenyl isocyanate and immediately aerated (Bendix Model 44 portable air sampling pump) for 4 h at an air flow of 0.2 dm3 min-1. Recoveries were evaluated at various time intervals up to 12 d. The main purpose was to assess the stability of the isocyanate derivative on XAD resins during and after sampling. Efficiency of Derivatization and Desorption for Sampling Low Levels of Isocyanate on Tryptamine-coated XAD-2 and XAD-4 The tryptamine-derivatized phenyl isocyanate was found to be relatively stable on XAD-2 and XAD-4 sampling tubes before and after aeration.Subsequently, the efficiency of derivatiza- tion and desorption at various time intervals for microgram levels of isocyanates collected on tryptamine-coated XAD-2 and XAD-4 was evaluated. Acetonitrile was used for all the desorptions. Effective Amount of Tryptamine for Coating XAD-2 For the purpose of establishing the relationship between the maximum amount of airborne isocyanate collected for a known amount of coated tryptamine, experiments were conducted on the TAGS by sampling for various time periods using various amounts of tryptamine for coating (Table 4). The findings would serve as a guideline for an individual laboratory to select the amount of tryptamine to be used for coating according to its specific requirements.Before con- ducting this experiment, the TAGS was adjusted mainly by varying the pressure of the purging nitrogen gas and monitor-ANALYST, JANUARY 1992, VOL. 117 11 ing the amount of phenyl isocyanate reaching the sampling port via impingers containing try ptamine solutions. The adjustments were made so that in a reasonable time period, the amount of phenyl isocyanate vapours reaching each of the samplers was slightly less than the equivalent amount of the tryptamine used for sampler coating. Two identical samplers were connected in series during air sampling. Any break- through of phenyl isocyanate would be collected in the back-up sampler. Recovery Study on Simulated Air Sampling for Phenyl Isocyanate As the generation of an exact amount of phenyl isocyanate through the TAGS was not possible, all recoveries were calculated with reference to the amount of phenyl isocyanate collected simultaneously by the impinger solutions containing tryptamine, which was considered to be 100%.The recoveries of phenyl isocyanate from the tryptamine-coated XAD-2 tubes include the evaluation of the effect of the time lapse before desorption on the formation of the isocyanate deriva- tive. Results and Discussion Methods for the determination of isocyanates are based on the determination of the corresponding derivatives resulting from the instability of the isocyanato groups. Collecting samples with solid sorbent samplers for the assessment of exposure to airborne isocyanates in the workplace is a challenging task, because the derivatization at the solid phase of the derivatiza- tion reagent is much slower than the reaction in the impinger solutions.For a worker, however, the solid sorbent sampler is more convenient to carry around during a work-shift. The results in Tables 1-3 show that tryptamine-coated XAD-2 resin was the most efficient sorbent and compared well with the impinger solutions in the recovery study. Table 2 indicates that the desorption of tryptamine-derivatized phenyl isocyanate from XAD-4 was slightly hindered, with con- sistently lower yields. The efficiency of XAD-2 resins over other solid sorbents is probably due to the relatively non-polar chemical structure (polystyrene) and the appropriate pore size Table 1 Sampling efficiency of various tryptamine-coated solid sorbent samplers Amount of Amount of Type of tryptamine for phenyl isocyanate sorbent coating/pg found/yg Impinger Impinger XAD-2 XAD-2 XAD-4 XAD-4 XAD-7 XAD-7 Molecular seive Molecular seive Glass beads* Glass beads* Charcoal Charcoal Silica gel Silica gel Glass-fibre filter? Glass-fibre filter? 100 100 200 200 200 200 200 200 200 200 200 200 400 400 200 200 200$ 200$ 53.3 53.2 59.1 56.6 56.5 56.9 38.2 38.8 1.3 1 .o 20.5 17.3 44.2 41.6 Trace Trace 32.1§ 45.79 * Height of packing in tube 1.5 cm, as the air flow is substantially reduced for 3 cm of packing.T Using two filters in series. $ Amount for each filter. § Combined yield for both filters. (90 A). The extremely low yield with silica gel may reflect the existence of silanol groups, which are likely to be reactive to the isocyanates.The calculated maximum length17 of the MDI and HDI molecules is approximately 15 8, (the length of the tryptamine molecule is about 8 A), whereas for most of the isocyanate pre-polymers the lengths are well under 30 A. For example, Desmodur N, an HDI pre-polymer, is calculated to be slightly longer than 20 A. All isocyanates are expected to penetrate through the porous surface of the XAD-2 resin. With the molecular seive, the average pore size is only 13 A, which would severely hinder the penetration of the isocya- nates. On the other hand, the excessive pore size (lo4 A) of the glass-fibre filters offers very little retention of isocyanate molecules.The threshold limit values’s for the time-weighed average (TLV-TWA) for all isocyanates are at the level of 0.005 ppm. At present, there are no short-term exposure limits (TLV- STEL)18 for isocyanates, except for TDI which has a TLV-STEL of 0.02 ppm. The corresponding TLV-TWA concentration of phenyl isocyanate would be about 0.025 mg m-3 and the TLV-STEL 0.1 mg m-3. It has been common to operate sampling pumps at an air flow rate of 0.2 dm3 min-1 for solid sorbent samplers. For sampling the TLV-TWA concentration of isocyanate in air during one 8 h work shift, the phenyl isocyanate collected will be about 2.4 pg. For the TLV-STEL concentration, the amount of phenyl isocyanate collected during a 15 min sampling will be 0.3 pg. It is apparent that a sampler containing 100 pg of tryptamine is, in general, sufficient for most sampling (Table 5 ) .In order to have a high degree of confidence, practical sampling can be performed by either using a sampler containing 200 pg of tryptamine or two Table 2 Stability of phenyl isocyanate tryptamine derivative on XAD-2 and XAD-4 resins Time lapse Amount of Type of before derivative Recovery sorbent desorption/d found/pg (Yo 1 XAD-2 XAD-2 XAD-4 XAD-4 XAD-2 XAD-2 XAD-4 XAD-4 XAD-2 XAD-2 XAD-4 XAD-2 XAD-2 XAD-4 XAD-4 2 2 2 2 7 7 7 7 9 9 9 12 12 12 12 3.83 4.13 3.45 3.31 3.85 3.92 3.41 3.31 3.92 3.69 3.13 3.62 3.76 3.26 3.41 86.2 93.1 77.8 74.6 86.8 88.4 76.7 74.6 88.4 83.1 70.4 81.5 84.7 73.5 76.7 Table 3 Efficiency of over-all derivatization and desorption for low levels of phenyl isocyanate in XAD resins Amount of Efficiency Type of tryptamine before cyanate recovery Amount of Time lapse phenyl iso- of sorbent coated/yg desorptiodd foundlyg ( Y o ) Impinger XAD-2 XAD-4 XAD-2 XAD-4 XAD-2 XAD-4 XAD-2 XAD-4 XAD-2 XAD-4 100 300 300 0 0 300 300 300 300 0 0 - 0 0 0 0 2 2 7 7 7 7 3.30 3.01 2.32 1.66 0.85 3.05 1.94 2.92 2.30 0.24 Trace 1 00 91.2 70.3 50.3 25.8 92.4 58.8 88.5 69.7 7.3 -12 ANALYST, JANUARY 1992, VOL.117 tryptamine impingers. It should be noted that the efficiency of sampling isocyanate using tryptamine impinger solutions has been verified previously by comparing a well established impinger method involving 1-(2-methoxyphenyI)piperazine .4 Although both solid sorbent and liquid impinger samplers were comparable in over-all efficiency (from sampling isocya- nates in air to analysis by HPLC), the derivatization step was less efficient with the former. This was observed (Table 6) when the desorption process was delayed for 7 d after sampling.The lack of derivatization efficiency at the solid phase of the reagent should not be viewed as a result of the inferiority of the tryptamine in comparison with other commonly used derivatizing reagents. Previous work4.19 has shown that tryptamine is as efficient as 1-(2-methoxyphenyl)- piperazine and 1-(2-pyridyl)piperazine and superior to N-(p- nitrobenzy1)-N-propylamine for derivatizing isocyanates. Table 4 shows that a 100 pg of tryptamine coated XAD-2 sampler can adsorb and derivatize nearly stoichiometric amounts of phenyl isocyanate when the samples are desorbed with acetonitrile immediately after collection.With derivati- zation in the solid state of the reagent, the total surface area of the coated reagent would be drastically diminished compared with that in the solution. Further, the immobile products from the solid-state reaction would shield the surface layer of the reagent, preventing further reaction of newly approaching isocyanate molecules. It should be emphasized that these causes of inefficient derivatization on solid reagents apply not only to tryptamine but all other commonly used derivatizing reagents. In order to ensure the completion of derivatization, it is advisable that after sampling the tryptamine-coated XAD-2 is immediately emptied from each sampler into 10 ml of acetonitrile.The sampling of airborne isocyanates using tryptamine- coated XAD-2 sorbent in conjunction with the already established analytical procedure provides an elegant method for personal monitoring of both monomeric and polymeric isocyanates in the workplace. References Table 4 Effective amount of phenyl isocyanate derivatized by tryptamine-coated XAD-2 resin Amount of phenyl Amount of Phenyl isocyanate found in tryptamine isocyanate sampler*/yg Sampling for coating/ equivalent/ periodh Pg Pg Front Back 1.5 100 100 100 75 75 75 76.1 75.9 79.1 Av. 77.0 99.9 109.1 104.8 Av. 104.6 172.5 148.9 164.6 Av. 162.0 231.4 240.7 232.1 Av. 234.7 0 0 0 200 200 200 148 148 148 0 0 0 300 300 300 222 222 222 0 0 0 4 400 400 400 296 296 296 0 0 0 * Desorption performed immediately after sampling.Table 5 Recoveries of XAD-2 sampled phenyl isocyanate with no time lapse for desorption Amount of phenyl iso- cyanate Recovery foundlpg (Yo) 16.5 15.8 15.0 17.9 12.5 14.5 16.2 17.3 16.2 13.7 Av. 16.2 100 Av. 15.4 k 1.7 95.1 Amount of Type of tryptamine in sampling sampledpg Impinger 100 200 XAD-2 100 100 200 200 300 300 300 300 1 2 Bagon, D. A., Warwick, C. J., and Brown, R. H., Am. Znd. Hyg. Assoc. J., 1984, 45, 39. Wu, W. S . , Nazar. M. A., Gaind, V. S., and Calovini, L., Analyst, 1987, 112, 863. Wu, W. S . , Szklar. R. S . . and Gaind, V. S . , Analyst, 1988,113, 1209. Wu, W. S . , Stoyanoff. R. E., Szklar, R. S . , Gaind, V. S . , and Rakanovic, M . , Analyst, 1990, 115, 801. Wu, W. S., Stoyanoff, R. E., and Gaind, V. S., J. High Resolut.Lake City, 1983. Tucker, S. P., and Arnold, J. E . , Anal. Chem.. 1982,54, 1137. Anderson, K., Gudehn, A., Sevin, J.-O., and Nilsson, C.-A., Am. Ind. Hyg. Assoc. J . , 1983, 44, 802. Lipski, K., Ann. Occup. Hyg., 1982, 25, 1. OSHA Method No. 47 (Revised), OSHA Analytical Labora- tory, Salt Lake City, 1984. Rando, R. J., Hammad, Y. Y . , and Chang, S.-N., Am. Ind. Hyg. Assoc. J.. 1989, 50, 1. Rando, R. J., Hammad, Y. Y., and Chang, S.-N., Am. Znd. Hyg. Assoc. J . , 1989, 50, 8. Beasley, R. K., and Warner, J. M . , Anal. Chem., 1984, 56, 1604. Wu, W. S., Stoyanoff, R. E., and Gaind, V. S . , paper presented at the 104th AOAC Annual International Meeting and Expo- sition, New Orleans, 1990. Gaertner, R. R. W., Appl. Ind. Hyg., 1988, 3, 258. Ketelaar, J. A. A., Chemical Constitution, Elsevier, Amster- dam, 1958, ch. 3, sect. 28, p. 198. Threshold Limit Values and Biological Indices f o r 1990-1991, American Conference of Governmental Industrial Hygienists, Cincinnati, 1990. Wu, W. S., Stoyanoff, R. E., and Gaind, V. S . , Analyst, 1991, 116. 21. Paper 1101324F Received March 19, I991 Accepted August 12, 1991 Table 6 Recoveries of XAD-2 sampled phenyl isocyanate with time lapse for desorption 3 4 5 Amount of Amount of Time lapse phenyl iso- Type of tryptamine in before cyanate sampling samplerlyg desorptiodd foundlyg 8 9 Impinger 100 100 100 100 10.2 9.9 14.9 14.5 Av. 12.4 * 2.7 12.2 11.3 12.3 14.2 Av. 12.5 k 1.5 8.2 6.9 6.4 6.9 Av. 7.1 kO.8 10 11 12 XAD-2 300 300 300 300 13 14 15 300 300 300 300 XAD-2 16 17 18 samplers in series, each containing 100 pg of tryptamine (Tables 4 and 5 ) . The results in Table 3 also indicate that an XAD-2 tube coated with 300 pg of tryptamine can retain about 3 pg of phenyl isocyanate for 7 d without desorption treatment after sampling. Simulated sampling using various amounts of tryptamine again demonstrated (Tables 5 and 6) that the tryptamine-coated XAD-2 sampler is comparable to the 19