Annals of Warsaw Agricultural University – SGGW Forestry and Wood Technology No 58, 2006: (Ann. Warsaw Agricult. Univ. – SGGW, For. and Wood Technol. 58, 2006)
Benzene Substitution in Wood Analysis ANDRZEJ ANTCZAK, ANDRZEJ RADOMSKI, JANUSZ ZAWADZKI Department of Wood Science and Wood Protection, Warsaw Agricultural University – SGGW Abstract: Benzene Substitution in Wood Analysis. Various solvent systems are theoretically considered as possible substitutes for benzene – ethanol blend. Hansen solubility parameters are used to evaluate the solvents similarity. Several solvents were experimentally tested. The azeotrope mixture of chloroform and ethanol is the solvent of choice, along with toluene blends with some of alcohols. The blends of toluene with ethanol and isopropanol exhibit unexpected behaviour. Keywords: benzene, wood analysis, extractives, Hansen solubility parameters.
INTRODUCTION Benzene is the solvent used in wood extraction, mixed with ethanol usually in 2:1 (sometimes 1:1) ratio. The Soxhlet apparatus is commonly used for the determination of wood extractives. Unfortunately, benzene in one of the most harmful chemical compound used in laboratory. It is highly inflammable, mutagenic and toxic. It can cause cancer. It also exhibit toxic action when inhaled, swallowed and in contact with skin or eyes. Prolonged exposure creates a serious hazard for health. Many countries imposed restrictions in using benzene. There is no need to use benzene blends in wood analysis. Several other solvents are in use, like ether, ethanol, chloroform, dichloroethylene and others. The only premise to use benzene is (except the habit) assurance of the proper comparison with the results, which had been obtained for many years. For this reason benzene should not be simply substituted by for example toluene. The appropriate solvent or solvent blend needs to exhibit the most similar properties to the original blend of benzene and ethanol. We assume several criteria for the choice of optimum solvent system. First, we looked for the most similar available solvent, taking into consideration Hansen solubility parameters (Radomski et al. 2006). Moreover, we limited our investigations to sole solvents and azeotrope mixtures, for the reasons discussed later. Second, solvent should be inexpensive and easy to recover. Third, the boiling point of solvent has not to exceed about 120°C, to avoid decomposition of extractives. Last, but not least, the safety considerations have to be looked into. THEORETICAL CONSIDERATIONS One of the most significant problems in chemical analysis (not only of wood) is reproducibility of results. If the conditions of analysis are precisely determined and reproduced, the aim mentioned is closer. In the case of extraction the most important parameter is solvent composition. If a blend of solvents differing in volatility is used, the vapour (and of course condensate) composition changes during the distillation process. As the volume of extraction cell of Soxhlet apparatus may vary, it is practically impossible to ensure reproducible composition of extracting solvent, unless an azeotropic mixture or a sole chemical compound is used. That why we do not consider other than azeotropic mixtures of solvents. It should be noticed at this point, that even azeotrope or pure substance do not guarantee the absolute reproducibility. The reason is water present in extracted samples, which is washed out and changes the solvent composition. Moreover, water often forms ternary azeotropes with solvent components. Fortunately, the amount of extractable water generally do not exceed 0.5% of the whole solvent in extraction cell, so this problem may be omitted in our investigations.
The most obvious substitute for benzene is toluene so it is a good start point for research. However, the azeotrope mixture of ethanol and toluene contain 68% of ethanol while the benzene azeotrope contains only 32% of ethanol. Such a significant difference in polarities of these two solvent blends means that they are not interchangeable. The next step is considering other alcohols, forming azeotropes with toluene. Table 1 contains values of Hansen solubility parameters for various toluene-alcohol azeotropes. The solvents are ordered according to decreasing similarity with the benzene-ethanol blend, measured as a distance in Hansen three-dimensional system. The distance between solvent A and B is described by formula below: D = [(d,A – d,B)2 + 4([(p,A – p,B)2 + ([(h,A – h,B)2]½ (1) Where: D - Hansen distance, d,X - dispersion component of total solubility parameter of solvent A and B, respectively, d,X - polar component of solvent A and B, respectively, d,X - hydrogen bonding component of solvent A and B, respectively. Table 1. Solubility parameters for toluene-alcohol azeotropes solvent (composition by weight) 32% ethanol – benzene 45% 2-methyl-1-propanol – toluene 56% 3-methyl-1-butanol – toluene 55% 2-butanol – toluene 27% 1-butanol – toluene 49% 1-propanol – toluene 58% 2-propanol – toluene 68% ethanol – toluene
boiling point, t/°C 67.8 101.2 100.5 95.3 105.6 92.6 80.6 76.7
solubility parameter, /MPa½ hydrogen total, t dispersive polar bonding (Hildebrand d p value) h 19.5 17.5 3.1 8.0 19.0 16.7 3.4 8.5 19.1 16.7 3.6 8.5 19.5 16.8 3.8 9.1 18.6 17.4 2.6 5.9 20.0 17.0 4.1 9.8 20.3 16.7 4.2 10.7 22.7 16.5 6.6 14.2
distance D 0.0 1.1 1.4 1.9 2.3 2.7 3.6 9.4
Higher alcohols are excluded from our considerations, because of too high boiling points. From the data presented a conclusion may be drawn that mixtures containing commonly used alcohols, like ethanol, iso-propanol and n-butanol cannot be expected to give accurate results of the wood extraction. Good conformity is exhibited by toluene blends with alcohols of branched molecules: sec-butyl, iso-butyl and iso-amyl. Unfortunately, these chemicals are relatively expensive. Moreover, their boiling points are about 100°C, i.e. 30° over the boiling point of ethanol-benzene azeotrope. Applying stronger cooling can lower the temperature in extraction cell of Soxhlet apparatus, so the results of determination of extractives content in wood may be sufficiently accurate. More detailed analysis of extractives composition however may be erroneous, as the secondary reactions of extractive compounds may occur. The mixtures of toluene with alcohol are not the only possibilities of solvent type. In Table 2 Hansen data for a few other azeotropes, along with pure substances. In this case more attention should be paid to the similarity of all components of the total solubility parameter. A few examples are given to illustrate this question. On the one hand two exotic pure substances are found of similar solubility behaviour, namely dibenzyl ether and decanol, which of course cannot be used because of extremely high boiling points. On the other hand a few substances or blends can be found of Hildebrand value (total solubility parameter) almost identical to ethanol-benzene azeotrope, but deeper look on them proves substantial incompatibility. Significant differences in component solubility parameters cause distant position of points representing these solvents on Hansen system. The solvent of choice is in this case azeotrope
mixture of chloroform and ethanol, while pure chloroform and azeotrope of cyclohexane and ethanol are the next in turn. Table 2. Solubility parameters for toluene-free solvents boiling point, t/°C
solvent (composition by weight) – benzene
67.8
93% chloroform – ethanol 69% cyclohexane – ethanol 48% heptane – methanol 48% acetone – methyl acetate
59.4 64.9 59.1 55.6
dibenzyl ether 1-decanol chloroform tetrahydrofuran chlorobenzene cyclohexanone
295 231 61.2 65.6 130.6 156
32% ethanol
solubility parameter, /MPa½ hydrogen total, t dispersive polar bonding (Hildebrand d p value) h 19.5 17.5 3.1 8.0 azeotropes 19.5 17.6 3.8 7.4 17.8 16.5 2.7 6.0 19.5 15.2 5.9 10.7 19.3 15.5 8.9 7.3 pure substances 19.3 17.4 3.7 7.4 20.4 17.6 2.7 10.0 18.9 17.8 3.1 5.7 19.5 16.8 5.7 8.0 19.6 19.0 4.3 2.0 19.6 17.8 6.3 5.1
distance D 0.0 1.5 2.4 6.6 11.8 1.3 2.2 2.3 5.2 6.6 7.0
The above considerations show the benefits of applying Hansen solubility parameter theory. In a short time and without an experimental work the set of possible solvent substitutes for benzene-ethanol blend can be confined to narrow limits. A few solvent systems selected are assigned to investigate on the empirical way. We decided to test following solvent mixtures showing Hansen distance closer than 2 units: 3-methyl-1-butanol:toluene and chloroform:ethanol. Additionally, other azeotropes of toluene were included (except npropanol containing) in order to compare the extraction behaviour of more differing solvents. Similarly we included pure chloroform to investigate the influence of small addition of ethanol on chloroform extractive ability. All selected solvents are presented in Hansen threedimensional system on Figure 1. ethanol toluene
6
p
iso-propanol toluene 3 isoamyl alc. toluene
15 chloroform ethanol
chloroform
ethanol benzene
n-butanol toluene
15
d
20
h
5
Figure 1. Positions of possible substitutes of benzene-ethanol in Hansen system.
MATERIALS AND METHODS Samples of wood were collected from about 80 years old Scotch pine stem (Pinus silvestris L.), heartwood zone from the butt end (cross-section at height of 2.0m). Samples were grounded and fractionated. The fraction passing 1-mm sieve and remaining on 0.5-mm sieve were air-dried in slightly elevated temperature (30÷40°C). The experiments were conducted in Soxhlet apparatus for 6 hours, then after determining the extractive content extractions were continued successively for 4 hours. The weight of each sample was about 5g. Three consecutive determinations were made to increase the accuracy of the results. All solvent used were o analytical grade. Moisture content was determined by oven drying to constant weight. RESULTS AND DISCUSSION Figure 2 shows the average extractive contents in Scotch pine heartwood as determined in selected solvents. Toluene – alcohol mixtures are placed first, according to increasing differences of Hansen solubility parameters. Significant differences of extraction behaviour can be observed. In the case of iso-amyl alcohol – toluene blend, the nearest to the original ethanol – benzene mixture, very little decrease of result obtained was noticed. N-butanol – toluene blend exhibits much weaker extractive ability, which is consistent with longer Hansen distance. Unexpectedly, for the next alcohol – toluene mixtures the inversion of this trend can be observed. Iso-propanol blend is more effective than n-butanol one, but still weaker than iso-amyl alcohol, while ethanol blend reaches the accuracy level of original ethanol – benzene mixture. This phenomenon is very interesting, as there is no possibility that distant solvents in Hansen system exhibit more similar solubility behaviour than the near one. The only acceptable reason of this behaviour is varying composition of extractives. The more polar and stronger hydrogen bonding solvents, like ethanol – toluene blend, extract probably different compounds from the wood sample. Maybe the lowest fractions of structural components (hemicelluloses), or inorganic compounds are extracted instead of some waxes, terpenes or other non-polar compounds.
13% 13,0%
13,0%
12,8%
13,0%
12,3%
12% 11,7%
11,4%
11%
3m
be nz en eth e: eth yl -1 an -b ol ut an ol :to 1l.. bu . ta no l:t 2ol pr ue op ne an ol :to lu en et e ha no l:t ol ch ue lo ne ro fo rm :et ha no l ch lo ro fo rm
10%
Figure 2. Extractive contents in Scotch pine heartwood determined by applying various solvents.
In the case of chloroform – ethanol blend, which is the second nearest the original solvent, excellent interchangeability has been found, as the results of both extractions are almost identical. Because pure chloroform is the worst of all solvents tested, another experiment was conducted, using chloroform containing 4% of ethanol (azeotrope contain 7% of ethanol). The average result of three determinations was 12.4%, which is significantly lower value than for azeotrope mixture. CONCLUSIONS The alcohol – toluene blends tested show at first decrease of extractive effectiveness as the Hansen distance lengthen and then inversion of the trend. The nature of this phenomenon requires more detailed analysis, which exceeds the limits of this paper. The blend with isoamyl alcohol can be good substitute for benzene – ethanol, similarly with iso-butanol or secbutanol blends are expected to be. The chloroform – ethanol azeotrope prove to be the excellent substitute. It has low boiling point (even lower than original) and the components are inexpensive. Moreover, chloroform is non-flammable, which is very important advantage as practically eliminates the risk of fire. However, this solvent has to be prepared precisely, as little deviation in ethanol concentration is expected to have a great influence on solubility behaviour of the blend. REFERENCES BARTON, A. F. M., 1983; Handbook of Solubility Parameters and Other Cohesion Parameters Boca Raton, Florida: CRC Press, Inc. HANSEN, C.M., 1967; “The Three Dimensional Solubility Parameter Key to Paint Component Affinities: I. Solvents Plasticizers, Polymers, and Resins” Journal of Paint Technology, Vol. 39, No. 505 HANSEN, C.M., 1967; “The Three Dimensional Solubility Parameter - Key to Paint Component Affinities: II. Dyes, Emulsifiers, Mutual Solubility and Compatibility, and Pigments” Journal of Paint Technology, Vol. 39, No. 511 HANSEN, C.M., 1967; “The Three Dimensional Solubility Parameter - Key to Paint Component Affinities: III. Independent Calculations of the Parameter Components” Journal of Paint Technology, Vol. 39, No. 511 KAČIK F., SOLÁR R., 1999; Analytická chémia dreva. Vyd. TU vo Zvolene. KRUTUL D., 2002; Ćwiczenia z chemii drewna oraz wybranych zagadnień chemii organicznej. Wyd. SGGW, Warszawa. RADOMSKI, A., ZAWADZKI, J., ANTCZAK A., 2006; “Introduction to Solubility Parameters Theory” Ann. Warsaw Agricult. Univ. – SGGW, For. and Wood Technol. 59 (the entire issue) Streszczenie: Zamienniki benzenu w analizie drewna. Przeprowadzono teoretyczną dyskusję możliwości zastąpienia układu benzen – etanol przez inne rozpuszczalniki. Do określenia stopnia podobieństwa rozpuszczalników wykorzystano system parametrów rozpuszczalności Hansena. Kilka rozpuszczalników zostało sprawdzonych doświadczalnie. Najlepszym zamiennikiem jest azeotropowa mieszanina chloroformu z etanolem, dobre wyniki dają też mieszaniny toluenu z niektórymi alkoholami. Mieszaniny toluenu z etanolem i izopropanolem wykazują nieoczekiwane właściwości. Authors’ address: Department of Wood Science and Wood Protection, Warsaw Agricultural University – SGGW ul. Nowoursynowska 159, 02-117 Warsaw, Poland e-mail
[email protected] [email protected] (http://andrzej_radomski.users.sggw.pl)
[email protected] 