Determination of the degree of hydrolysis of oil paint samples using a two-step derivatisation method and on-column GC/MS

Citation
Jdj. Van Den Berg et al., Determination of the degree of hydrolysis of oil paint samples using a two-step derivatisation method and on-column GC/MS, PROG ORG C, 41(1-3), 2001, pp. 143-155
Citations number
34
Categorie Soggetti
Material Science & Engineering
Journal title
PROGRESS IN ORGANIC COATINGS
ISSN journal
03009440 → ACNP
Volume
41
Issue
1-3
Year of publication
2001
Pages
143 - 155
Database
ISI
SICI code
0300-9440(200103)41:1-3<143:DOTDOH>2.0.ZU;2-0
Abstract
An elegant method is described for the determination of the degree of hydro lysis of(unsaturated) triacylglycerols, and their oxidation products, prese nt in linseed oil based paints. The analytical strategy, a transethylation of esterified fatty acids followed by a trimethylsilylation of free fatty a cids and their salts, is first tested on reference materials comparable or identical to the compounds present in fresh and aged oil paints. Reproducib ility and repeatability are examined on free- and methylated fatty acids, a sodium- and lead salt and several triacylglycerols. Unwanted trimethylsily lated products could be observed for the triacylglycerols and other esterif ied fatty acids up to a maximum of 6%. Free fatty acids and their salts are shown to be completely trimethylsilylated but incomplete derivatisation wa s observed for azelaic acid. Glycerol, liberated upon transethylation was ( partially) trimethylsilylated. However, the recovery was not uniform and th e results could not be used for quantitative determination of the amount gl ycerol in the paint sample. Studies on the influence of the pigments indigo , lead white, and prussian blue and glycerol showed there is no adverse inf luence on the analytic result, except for lead white and indigo doped refer ence material. The addition of these last compounds led to increased amount s of hydrolysed products up to 37%. The repeatability of the determination of the degree of hydrolysis within oil paint systems was tested on a number of paints. Tt is shown for 5-year-old test paints with a relatively homoge neous composition that the method is reproducible. For less defined paints consisting of multi-layered systems it is shown that the position and way o f sampling can have a significant influence on the spread in the analytical result. This is caused by the large variations that are possible in the sa mpled material in these inhomogeneous systems. The degree of hydrolysis of an oil paint is taken as the average of the values obtained for azelaic-, p almitic- and stearic acid, based on the results presented. Overall, it is o bserved that the relative amount of hydrolysed fatty (di)acids increases in time. Surprisingly, in contrast to the results obtained on reference mater ial, for all lead white pigmented paints lower degrees of hydrolysis were f ound relative to other paints from that particular set. (C) 2001 Elsevier S cience B.V. All rights reserved.