Annals of Occupational Hygiene Advance Access originally published online on March 30, 2009
Annals of Occupational Hygiene 2009 53(4):391-401; doi:10.1093/annhyg/mep014
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Comparative Evaluation of Overexposure Potential Indices used in Solvent Substitution
Département de santé environnementale et santé au travail, Université de Montréal, PO Box 6128, Main Station, Montréal, Québec H3C 3J7, Canada
* Author to whom correspondence should be addressed. Tel: +(514)343-6134; fax: +(514)343-2200; e-mail: michel.gerin{at}umontreal.ca
Background: Overexposure potential indices (OPIs) can be defined as ratios between different measures of the volatility of a substance and its 8-h occupational exposure limit. Objectives: The overall objective of this study was to compare three OPIs, using a list of commonly used or recommended solvents as both single substances and mixtures. The three indices studied differ in the way they characterize volatility: vapour pressure for the vapour hazard ratio (VHR), calculated emission rate for the ‘SUBstitution FACtor’ (SUBFAC) index obtained from the former Danish SUBTEC software and discrete values based on relative evaporation rate for the Måleteknisk Arbejdshygiejnisk Luftbehov (MAL) index which is part of regulations in Denmark.
Methods: A list of 56 pure solvents and of 50 mixtures of solvents was used for the comparison. For mixtures, VHR was calculated both as VHRmi, where the mixture is considered as ideal, and as VHRmc (corrected) with the introduction of activity coefficients to take into account the nonideal behaviour of components. Activity coefficients were a standard feature of SUBFAC for mixtures (SUBFACm) but are not used when calculating MAL for mixtures (MALm).
Results: Pure solvents rank in overall similar order with the three systems and correlation was excellent between VHR and SUBFAC (R = 0.99) and good between VHR and MAL (R = 0.75). For solvent-based mixtures, correlation is excellent between VHRmi and VHRmc (R = 0.98) and between VHRmc and SUBFACm (R = 0.98) but moderate between VHRmi and MALm (R = 0.52). Ratios between VHRmc and VHRmi varied between 0.57 and 2.7, thus spanning an 
5-fold range, and averaged 1.2. Worse correlations involving the MAL index are attributed to the discrete nature of its numerator.
Conclusions: Overall, these results favour using the simple and more easily available VHR when comparing pure substances. In the case of mixtures, the effect of nonideality may be important in certain cases. As a precaution, since no integrated tool is available presently to easily calculate OPI values corrected for nonideality, substitution should be recommended only as long as the non-corrected VHR value for an alternative solvent is at least 5-fold less than that of the solvent to replace.
Keywords: activity coefficient • MAL • mixture • solvent • SUBFAC • substitution • VHR
Received December 19, 2008; in final form February 23, 2009