Ann. occup. Hyg., Vol. 46, No. 7, pp. 637-641, 2002
© 2002 British Occupational Hygiene Society
Published by Oxford University Press
Determination of Dicumyl Peroxide in Workplace Air
Department of Chemical Environmental Science, Chalmers University of Technology, SE-412 96 Göteborg, Sweden
Received 22 February 2002; in final form 31 May 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
A method for the determination of dicumyl peroxide in workplace air was developed and applied. Micro-impinger bottles and personal air sampling pumps were used for the sampling. Gas chromatography/mass spectrometry was used for the analytical separation and quantitative determination. The technique makes it possible to monitor peroxide concentrations down to 5 µg/m3 in air.
Keywords: dicumyl peroxide; gas chromatography; GC/MS; workplace air
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Dicumyl peroxide is an organic peroxide. The peroxide bond with its two oxygen atoms is easily broken up. For this reason, organic peroxides are often used as free radical sources in polymerizations and similar reactions. Dicumyl peroxide is among the less reactive of the organic peroxides. Large amounts of dicumyl peroxide are used as cross-linking agents in the polymer industry. Exposure of the workers to this compound may occasionally be high, for example when sacks of dicumyl peroxide are cut open and the peroxide is poured into hot melting tanks. This operation is sometimes performed without any protective equipment.
Because of their reactive behaviour, organic peroxides affect living organisms in a negative way. Workers at a polymer producing plant experienced problems with nose bleeds when handling dicumyl peroxide and an investigation showed the formation of new blood vessels in the nose after exposure to the peroxide (Petruson and Järvholm, 1983). Subsequent acute and long-term exposure studies on animals confirmed that structural and functional changes in the nasal mucosa appear even after a short time of exposure to dicumyl peroxide (Hansson and Petruson, 1986). Dicumyl peroxide is also suspected of causing occupational asthma (Stenton et al., 1989). Efforts to investigate health effects of dicumyl peroxide exposure have suffered from a lack of methods for measurement of peroxide concentrations in air (Petruson and Järvholm, 1983; Hong et al., 1998).
Surveys of different methods for the determination of organic peroxides have been reported (see for example Horner and Jürgens, 1963; Mair and Graupner, 1964; Johnson and Siddiqi, 1970; Swern, 1970–72; Cornish et al., 1981; Funk and Baker, 1985; Baj, 1994; Hong et al., 1998; Wang and Glaze, 1998). For dicumyl peroxide the number of methods described in the literature is much smaller. Iodimetric titration (Mair and Graupner, 1964; Swern, 1970–72; Thomson and Bell, 1974; Hudec et al., 1976; Silbert, 1992; Zawadiak et al., 1993), NMR (Swern et al., 1969; Swern, 1970–72), polarographic determination (Swern, 1970–72; Gregorowicz et al., 1977) and thin layer chromatography (Johnson and Siddiqi, 1970; Swern, 1970–72) have been suggested. For the identification and quantitative determination of low concentrations of dicumyl peroxide in a solvent, chromatographic methods have been described (Hudec et al., 1976; Baj, 1994). However, none of the methods mentioned have been used for measurements in workplace air.
In this study, the objective was to measure dicumyl peroxide in workplace air. A method for sampling of dicumyl peroxide from air and subsequent quantitative determination was developed. Gas chromatography is often the preferred method for the determination of organic compounds with sufficient volatility and thermal stability, like dicumyl peroxide (Swern, 1970–72), and has been shown to be successful for the quantitative determination of dicumyl peroxide in clay after extraction with toluene (Hudec et al., 1976). In the present study, the combination of gas chromatography/mass spectrometry (GC/MS) was chosen since it allows positive identification and quantification of low concentrations of the analysed substance.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Sampling
For the sampling, a micro-impinger bottle (Micro-impinger 2.5 ml; SKC) was used. The bottle was filled with toluene (Merck). A personal air sampling pump (Gilian Low Flow Sampler LFS113DC) was used to draw air through the solvent during sampling. The air flow was normally
30 ml/min for 3–5 h. Toluene vapour is detrimental to the pump used and therefore a protective sorbent trap (Micro-impinger Trap; SKC) was placed between the bottle and the pump. The capacity of the trap was sufficient to absorb the total amount of solvent in the micro-impinger bottle. Prior to use, each micro-impinger bottle was carefully cleaned and weighed. Approximately 2.5 ml of toluene was pipetted into the bottle. The pump flow was measured and adjusted before sampling. The sampling equipment was either carried by a person or used for stationary sampling. All parts of the system (pump, filter and bottle) were equipped with clips so that a person, without changing normal working actions, could easily carry them around. After sampling, the bottle was weighed again to determine the mass of the remaining solvent and the pump flow was measured again to ensure a constant flow. The sample was pipetted into a vial that was sealed and put into a refrigerator pending analysis within a few hours.
Analysis
The analytical set-up for industrial hygiene application made use of GC/MS and single ion monitoring (SIM) applying a HP 5890 gas chromatograph. The analytical conditions are given in Table 1.
|
External standard calibration solutions were prepared by dissolving dicumyl peroxide (Nippon Oil and Fat) in toluene at room temperature. The calibration solutions were stored in a refrigerator and no degradation of the dicumyl peroxide was observed over a week.
An autosampler (Hewlett Packard) was used for injection of all solutions.
The relative intensity of the molecular ion (mol. wt 270) formed in the mass spectrometer is low. However, the relative intensity of the dominating fragment ion (m/z 119) is high. This makes it favourable to use SIM to obtain high sensitivity and specificity. SIM gives a significantly higher response than total ion current (TIC) monitoring. Figure 1 shows a typical chromatogram from the analysis of dicumyl peroxide in workplace air.
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Measurements of the concentration of dicumyl peroxide in air were performed in different parts of a polymer producing plant where dicumyl peroxide was used. The data and comments given in Table 2 summarize the results. Unless indicated in the comments, the data refer to stationary sampling and normal process conditions.
|
Since dicumyl peroxide is comparatively stable, the concentration in air can be high in a closed area. Table 2 shows that measurable concentrations of the peroxide occur along the polymer production line but that concentrations are negligible outdoors and in some rooms of the building, e.g. the product testing room. When there was visible peroxide spillage on the floor, the concentration of peroxide in the air of the peroxide blender room was high. The peroxide concentration in air rises when the ambient temperature rises and vice versa. Therefore the concentration can differ significantly from one day to another. Concentrations are high in the peroxide melting room where the peroxide itself is subjected to high temperatures. Other factors that influence the amount of peroxide in air are changes in ventilation and service work on the peroxide system. The highest concentration of peroxide was measured when work was done on the melting tank.
Reproducibility and sensitivity
In one location six parallel samples were taken to evaluate the reproducibility of the method at high concentration (Table 2). In this case, the standard deviation was 6% of the mean value. In the present study no internal standard was used. The autosampler minimizes the error in injected volume and increases the reproducibility, but the use of an internal standard is recommended.
The sensitivity of the method is high. It is possible to measure concentrations of dicumyl peroxide in air down to 5 µg/m3 if volumes of 10 l are sampled. Splitless injection is used to obtain high sensitivity. A test of the linearity of the GC/MS method was made by analysing reference solutions (1, 10 and 100 ng/µl). Each solution was analysed three times. The results showed that the response is linear at least up to 100 ng/µl (correlation coefficient to a straight line >0.99).
Sampling
The concentrations of dicumyl peroxide in workplace air are normally too low to be measured by gas chromatography by direct injection of an air sample. Dicumyl peroxide consequently has to be extracted from the air. Experiments with adsorbent sampling and thermal desorption were previously performed in the polyethylene producing plant where this study was carried out. However, at the high temperature necessary for desorption, the peroxide decomposed and it was not possible to obtain reliable results from the analysis of the complex mixture of breakdown products and interfering background compounds. Furthermore, during adsorbent sampling there is a risk of reactions with other compounds. In the present study, absorption in a solvent in a micro-impinger bottle was chosen. Even in this case the risk of reactions with other compounds exists, but to a lesser extent according to our general experience of air sampling.
In selecting the solvent for sample collection, the prerequisites for the solvent were as follows.
· High solubility of dicumyl peroxide.
· Low volatility to prevent evaporation during sampling. A few hours of sampling are often required for determination of occupational exposure.
· No impurities that could interfere with the analysis.
· A suitable boiling point in order to make it possible to use the ‘solvent effect’ in the gas chromatographic analysis. Then the column temperature is set below the boiling point of the solvent. The solvent thus becomes partly condensed in the beginning of the column and the sample is refocused giving narrow chromatographic peaks.
Toluene proved to be acceptable in all respects. About 50% of the initial volume was lost when air was pumped through the solvent for 5 h (flow rate 40 ml/min). Depending on external and analytical parameters (ambient temperature, total sampling volume, injector temperature, etc.), other solvents may be preferred if the conditions differ from those of this work. It is vital to use fresh absorbent for solvent trapping, to avoid the person carrying the sampling equipment being exposed to toluene vapours.
In this study, sampling time ranged between 4 and 5 h. If the sample collection is to be extended to cover a full working shift of 8 h, the pump flow should be lowered to give the same total volume of air as for the shorter sampling time. If the temperature of the air exceeds 20°C, the sampling volume should also be lowered.
The sample collection system was tested for breakthrough using air with dicumyl peroxide at a concentration of 100 µg/m3. The tests were performed with two bottles in series. At moderate flows (30–50 ml/min) the concentration in the second bottle was less than 5% (typically 1–2%) of the concentration in the first bottle after 4 h sampling.
The micro-impinger bottle contains 2.5 ml of solvent, which is about one-tenth of the volume of a conventional impinger bottle. Since the bottle is very small, a person can easily carry it around during their daily work. It is designed to tolerate excessive tilting during operation without spillage and loss of sample. The micro-impinger bottle had an additional holster to protect the glass as well as to protect the solution from light.
Stability
The relatively high thermal stability of dicumyl peroxide makes it possible to use temperatures up to 120°C in the gas chromatographic separation. The half-life for dicumyl peroxide is 5 h at 120°C (Swern, 1970–72). No decomposition (<2%) was observed when a solution of dicumyl peroxide in toluene was stored in a refrigerator in the dark for 1 week. The presence of other compounds from the air sampling may influence the stability. Thus, analyses should always be performed as soon as possible after sampling.
GC analysis
Because of the thermal instability of the peroxide, the temperatures in the gas chromatographic analysis had to be moderate. Extensive testing was required to evaluate the optimal injector and oven temperatures. The injector temperature had to be sufficiently low to avoid unacceptable breakdown of the peroxide, yet high enough to avoid carryover in the injector. An injector temperature of 100°C proved to be optimal. The oven temperature was not as critical, probably because of a more inert surface. At 120°C the loss by thermal decomposition was kept below 2%.
Initially, a flame ionization detector was used. This set-up also proved efficient. Low concentrations of dicumyl peroxide could be detected (down to 5 µg/m3). A drawback was the limited specificity since the compounds have to be identified from their retention times. The GC/MS method was developed to enable positive identification of the analysed compound. The advantages of mass spectrometric detection by single ion monitoring are high specificity and high sensitivity due to the fact that the dicumyl peroxide gives rise to one dominant fragment ion (m/z 119). Monitoring of this ion gave a clean and easy to interpret chromatogram (see Fig. 1).
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
The concentration of dicumyl peroxide in workplace air was determined using standard equipment. The method is suitable for the determination of occupational exposure. A person can easily carry the sampling equipment during a full shift under normal working conditions. Dicumyl peroxide concentrations in air down to 5 µg/m3 can be determined.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: +46-31-772-3005; fax: +46-31-772-2999
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Baj S. (1994) Quantitative determination of organic peroxides. Fresenius J Anal Chem; 350: 159–61.
Cornish LA, Ferrie R, Patersson JE. (1981) High pressure liquid chromatography of some organic peroxides. J Chromatogr Sci; 19: 85–7.
Funk MO Jr, Baker WJ. (1985) Determination of organic peroxides by high performance liquid chromatography with electrochemical detection. J Liquid Chromatogr; 8: 663–75.
Gregorowicz Z, Cebula J, Górka P. (1977) Polarografische Bestimmung von Cumolhydroperoxid und Dicumylperoxid nebeneinder in technologischen Mischungen. Z Anal Chem; 284: 287–8.
Hansson HA, Petruson B. (1986) Nasal mucosa changes after acute and long-term exposure to dicumylperoxide. Acta Otolaryngol; 101: 102–13.[Medline]
Hong J, Maguhn J, Freitag D, Kettrup A. (1998) Determination of H2O2 and organic peroxides by high-performance liquid chromatography with post-column UV irradiation, derivatization and fluorescence detection. Fresenius J Anal Chem; 360: 124–8.
Horner L, Jürgens E. (1963) Quantitative Bestimmung organischer Perverbindungen in Gemischen. Angew Chem; 70: 266–8.
Hudec P, Novotna B, Petruj J. (1976) Determination of dicumenyl peroxide by gas chromatography. Analyst; 101: 379–80.
Johnson RM, Siddiqi IW. (1970) The determination of organic peroxides. Oxford: Pergamon Press.
Mair RD, Graupner AJ. (1964) Determination of organic peroxides by iodine liberation procedures. Anal Chem; 36: 194–204.
Petruson B, Järvholm B. (1983) Formation of new blood vessels in the nose after exposure to dicumylperoxide at a chemical plant. Acta Otolaryngol; 95: 333–9.[Medline]
Silbert LS. (1992) Peroxides. Part 11. Iodimetric analysis of dialkyl and dicumenyl peroxides. Analyst; 117: 745–9.
Stenton SC, Kelly CA, Walters EH, Hendrick DJ. (1989) Occupational asthma due to a repair process for polyethylene-coated electrical cables. J Soc Occup Med; 39: 33–4.[Medline]
Swern D. (ed.) (1970–72) Organic peroxides, Vols I–III. New York: Wiley-Interscience.
Swern D, Clements AH, Luong TM. (1969) Nuclear magnetic resonance spectra of organic peroxides. Anal Chem; 41: 412–6.
Thomson IJ, Bell GC. (1974) Determination of traces of peroxides in hydrocarbon process streams. Anal Chem; 46: 1505–8.
Wang K, Glaze WH. (1998) High-performance liquid chromatography with postcolumn derivatization for simultaneous determination of organic peroxides and hydrogen peroxide. J Chromatogr A; 822: 207–13.
Zawadiak J, Gilner D, Kulicki Z, Baj S. (1993) Concurrent iodimetric determination of cumene hydroperoxide and dicumenyl peroxide used for reaction control in dicumenyl peroxide synthesis. Analyst; 118: 1081–3.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||