Subscriber access provided by TEXAS A&M UNIV COLL STATION
Desiccant efficiency in solvent and reagent drying. 7. Alcohols David R. Burfield, and Roger H. Smithers J. Org. Chem., 1983, 48 (14), 2420-2422• DOI: 10.1021/jo00162a026 • Publication Date (Web): 01 May 2002 Downloaded from http://pubs.acs.org on May 1, 2009
More About This Article The permalink http://dx.doi.org/10.1021/jo00162a026 provides access to: • •
Links to articles and content related to this article Copyright permission to reproduce figures and/or text from this article
The Journal of Organic Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036
J. Org. Chem. 1983,48, 2420-2422
Hz),1.18 (d, 3 H, J = 7 Hz);mass spectrum, m l e 110 (M’).
Acknowledgment. We are grateful to Professor Samuel Danishefsky, Yale University, for encouragement and support through Grant No. AI 16943-03. Registry No. 1, 85736-25-0; 2, 43124-56-7; 3, 15331-05-2; (+)-pulegone,89-82-7.
Desiccant Efficiency in Solvent and Reagent Drying. 7. Alcohols’” David R. Burfield and Roger H. Smithers* Department of Chemistry, University of Malaya, Kuala Lumpur 22-11, West Malaysia Received October 22, 1982
For many of their applications in synthesis, the lower monohydric alcohols are required to be scrupulously dry, and in contrast to alcohols containing six C atoms or more, whose desiccation does not appear difficult,’ alcohols containing one to four C atoms have long been recognized as posing a formidable drying problem. In the case of ethanol for instance, numerous synthetic procedures contain statements such as “traces of water depress the yield considerably”? a practical result of which is that samples dried with CaO, for example (containing up to 5000 ppm of water), are often completely una~ceptable.~There is also no doubt that for this group, as with other cases,l* the problem of dedlccant selection has been compounded by a lack of reliable quantitative data. As an illustration, when discussing the efficiency of magnesium or alkyl phthalateNa as desiccants, many a u t h ~ r i t i e continue s~~ to quote figures for water contents which were obtained in the original investigations1° and which must now be viewed with some scepticism. In the light of this, we now (1) Part 1: Burfield, D. R.; Lee, K. H.; Smithers, R. H. J.Org. Chem. 1977, 42, 3060. (2) Part 2: Burfield, D. R., Gan, G. H.; Smithera, R. H. J.Appl. Chem. Biotechnol. 1978, 28, 23. (3) Part 3: Burfield, D. R.; Smithers, R. H. J. Org. Chem. 1978,43, 3966. (4) Part 4: Burfield, D. R.; Smithers, R. H. J. Chem. Technol. Biotechnol. 1980, 30, 491. (5) Part 5: Burfield, D. R.; Smithers, R. H.; Tan, A. S. C. J. Org. Chem. 1981,46,629. (6)Part 6 Burfield, D. R.; Smithers, R. H.J. Chem. Educ. 1982,59, 703. (7) The use of K2C03as a desiccant is often considered adequate in these cases. See for example: (a) Calzada, J. G.; Hooz, J. Org. Synth. 1974,54,63. (b) Paul, R.; RiobC, 0; Maumy, M. Ibid. 1976,55, 62. (c)
Vogel, A. I. “Vogel’s Textbook of Practical Organic Chemistry”; Longmans: London, 1978; p 270. See also, however: Crandall, J. K.; Rojas, A. C. Org. Synth. 1976,55, 1. (8) See: Weiner, N. “Organic Syntheses”; Wiley: New York, 1943; Collect. Vol. 11, p 279. Also relevant are the following quotations. (a) ‘Yields are poor if the alcohol is not completely dehydrated”; Lycan, W. H.;Puntambeker, S. V.; Marvel, C. S. “Organic Syntheses”; Wiley: New York, 1943; Collect. Vol. 11, p 319. (b) “Moisture in the reagents affects the yield seriously”: Dox, A. W. ’Organic Syntheses”; Wiley: New York, 1941; Collect, Vol. I, p 5. (c) T h e use of 98% alcohol results in a lowering of yield by 1/3”: Kaufmann, W. E.; Dreger, E. E. Ibid., p 258. (d) “The quality of absolute ethanol used has a very marked effect upon the yield”: Adams, R.; Kamm, R. M. Ibid., p 250. (9) Thus, ‘Alcohol dried over lime gives very low yields”: Ford, S. G.; Marvel, C. S. ‘Organic Syntheses”;Wiley: New York, 1943; Collect. Vol. 11, p 373. “Ethanol dried only by a lime process gives a low yield”: Adkins, H.; Gillespie, R. H. Zbid., 1955; Collect. Vol. 111, p 672. (IO) For the use of magnesium see: Lund, H; Bjerrum, J. Ber. Dtsch. Chem. Ges. 1931, 64, 210. For the use of a sodium-alkyl ester combination see: (a) Smith, E. L. J. Chem. SOC.1927, 1288. (b) Manske, R. H. J. Am. Chem. SOC.1931,53, 1106.
report the results of a study of desiccant efficiency for this important group. General Indications.7cJ1 Both chemical and absorptive-type desiccants have been proposed for drying alcohols, and these i n c l ~ d e ~Al, ~ JBaO, ~ * ~CaO, ~ Mg, Na, and K2C03,while molecular sieves of Type 3A and 4A have been recommended for further drying.*lb CaH2 is also widely quoted,lld although some sources11aadvise caution in its use with the lower alcohols. While bearing in mind the well-known hygroscopic and hydrophilic properties of these compounds, the experiments reported below were not carried out by using any special techniques to obviate the entry of the atmosphere into drying systems other than those routinely used by the bench chemist. Methanol. As one authority has observed,’ld water is the dominant impurity in this solvent, and, unless extraordinary care is exercised, this content increases each time MeOH is handled. It is therfore quite remarkable that a survey of the literature on the drying of EtOH and MeOH points up a curious dichotomy: while references to EtOH8vgare characterized by numerous strictures as to the importance of achieving perfectly dry solvent, in contrast, MeOH is treated in a rather offhand manner, even though many of its applications parallel those of ita higher homologue. “All MeOH used must be anhydrous” is a typical comment,12although information on how to realize this is lacking. The cause may lie in the belief that fractionation alone gives a solvent of adequate dryness, although the water content is still an admitted 1000 ppm.7cJ1dWhere drying methods have been indicated, the use of Mg seems most often advised.”J3 The results shown in Table I certainly suggest that, on the whole, MeOH is more difficult to dry than its higher homologue. These figures require little comment, but it is worth highlighting the poor performance of 3A molecular sieve powder for MeOH in comparison with other alcohols. This is almost certainly an effect of molecular size and nonselective adsorption on the large surface area which occurs extremely rapidly with this desiccant.14 On account of ita small size, MeOH is able to compete with water for entry into the sieve pore. Similar reasons also explain the relative ineffectiveness of the 4A and 5A bead forms of molecular sieve. On the other hand, the 3A bead form constitutes a useful desiccant, whose success is presumably due to a much slower rate of absorption which occurs with greater selectivity. I t is also worth noting that a combination of methods is often the most effective strategy: a 1-L sample of MeOH distilled from Mg/12 onto 10% w/v 3A molecular sieve beads and then allowed to stand 48 h had residual water content of only 12 ppm. Finally, the unimpressive results obtained with BaO and CaO tend to support an earlier assessment of these agents as “tedious and wa~teful”.’~ Ethanol. As stated above, references to the use of ethanol as a nucleophile,16 as a solvent for Bouveault-Blanc (11) See: (a) Rickert, H.;Schwartz, H. In “Methoden der Organischen Chemie (Houben-Weyl)”; Mueller, E., Georg Thieme Verlag: Stuttgart, 1968; Band I/2, p 873. (b) Loewenthal, H. J. E. “Guide for the Perplexed Organic Experimentalist”; Heyden: London, 1980; p 50. (c) “Drying in the Laboratory”; E. Merck Co.: Darmstadt. (d) Riddick, J. A.; Bunger, W. B. “Organic Solvents”, 3rd ed.; Wiley-Interscience: New York, 1970 pp 638465. (12) Helferich, B.; Schafer, W. “Organic Syntheses”;Wiley: New York, 1941; Collect. Vol. I, p 365. (13) (a) Murray, J. I. ‘Organic Syntheses”; Wiley: New York, 1963; Collect. Vol. IV.p 744. (b) Baumgarten, H. E.; Petersen, J. M. Ibid., 1973; Collect. Vol. V, p 912. (14) See ref 6 (15) See Ref l l d , p 642.
0 1983 American Chemical Society
J. Org. Chem., Vol. 48, No. 14, 1983
Table 1. Desiccant Efficiency in Dryingaib of Some Common Lower AlcoholsC residual water content, ppm desiccant
3A sieves (bead) 3A sieves (powder)] Mg/I, CdZ Nap Na/dicarboxylic acid esterQ 4A sieves (bead) 5A sieves (bead) CaC, BaO Ca KZCO, CaO KOH powder ion exchange resin
a Static drying modes unless specified otherwise. Water content assayed by the Karl Fischer method. Desiccant loading 5% w/v with a drying period of 24 h unless specified otherwise. Initial water content 1 0 1 0 ppm. e Initial water content 1 5 0 0 ppm. f Initial water content 1000 ppm. g Initial water content 1 0 3 0 ppm. 96-h drying period. 168-h drying period. I Analysis was performed after settling of desiccant, 3-6 h. Weight of magnesium in accord with general See text. Weight of magnesium 2% w/v. " Distilled sample. O Initial water content 1 6 7 0 practice, i.e., 0.5% w/v. ppm, distilled sample. P Weight of sodium 3% w/v. See ref 32. Ratio of sodium to 2-butyl succinate for 2-BuOH and to diethyl phthalate f o r ethanol in accord with general practice (see ref 7c), i.e.; Na, 0.3 mol L-'; dicarboxylic acid ester, N o apparent drying. 0.14 mol L-l. I Stirred sample.
or oxime type reducti0m,1~or as a suitable medium for the preparation of metal ethoxides"' invariably include warnings as to the lowered yields which are obtained through the use of inadequately dried solvent. Desiccants most frequently prescribed16-18 include Ca0,lBMg," Na,2l or Na in conjuction with a high-boiling ethyl ester such as the phthalic acid derivative. The results (Table I) show that the desiccant par excellence both from the viewpoint of residual water levels and convenience is 3A molecular sieve powder.22 Sieves of larger pore size are again seen to be inappropriate. Of the more traditional siccatives, Mg and CaH2 are both efficaceous while the performance of the alkaline earth oxides was once more disappointing. The poor result with sodium is not unexpected and is attributable to the equilibrium shown between alkoxide and NaOH
H 2 0 + RONa
hydroxide. On the other hand, the ploy of refluxing the solution obtained after sodium treatment with a highboiling ethyl ester works well, because saponification of the ester irreversibly removes hydroxide ion. (16) See for example: (a) Croxall, W. J.; Fegley, M. F.; Schneider, H. J. 'Organic Syntheses"; Wiley: New York, 1963; Collect. Vol. IV, p 98. (b) Moffett, R. B. Ibid., p 427. (c) Ford-Moore, A. H.; Perry, B. J. Ibid., p 955. See also ref 8b. (17) See: Maneke, R. H. "Organic Syntheeee";Wiler New York, 1943; Collect. Vol. 11, p 154. See also ref 8a and 9. (18) See for example: (a) Holmes, H. L.; Trevoy, L. W. "Organic Syntheses"; Wiiey: New York, 1955, Collect. Vol. 111, p 300. (b) Meyer, K.; Bloch, H. S. Ibid., p 637. (c) Zuidema, G. D.; Van Tamelen, E.; Van Zyl, G. Ibid., 1963; Collect. Vol. IV,p 10. (c) Daeniker, H. U.; Grob, C. A. Ibid., 1973; Collect. Vol. V, p 989. See also ref 8 and 9. (19) In addition see: (a) Kimball, R. H.; Jeffereson, G. D.; Pike, A. B. "Organic Syntheses"; Wiley New York, 1943; Collect. Vol. 11, p 284. (b) Fuson, R. C.; Wojcik, B. H. Ibid., p 260. (20) In addition see: Lund, H.; Voigt, A. "Organic Syntheses";Wiley New York, 1943; Collect. Vol. 11, p 694. See also ref 8 and 9. (21) In addition see: (a) Marvel, C. S.; Hager, F. D. "Organic Syntheses"; Wiley: New York, 1941; Collect. Vol. I, p 248. (b) Levene, P. A.; Meyer, G. M. Ibid., 1943; Collect. Vol. 11, p 288. See also ref 8. (22) Removal of the laat traces of thie desiccant from a dried solvent probably necesaitates distillationor filtration. However, if left to stand for perioda of several daye,the desiccant usually from a compacted mass at the bottom of the container, leaving a completely clear solvent layer. In many procedures, the traces of powder remaining in suspension are probably of little consequence, and dried material can be dispensed by simple decantation.
Butanols. With some notable exce~tions,2~ perfectly dry 2-butanol is probably less widely required than the other alcohols in this study but is included for purposes of comparison. In contrast, completely anhydrous tertbutyl alcohol is always required for its most common use in the preparation of the potassium derivative. An early source24observes that "it is difficult to prepare t-BuOH free from water". Later authors recommend tackling drying by refluxing over Ca0,26a*b Na, or K25"*d or more recently by distillation from CaHz6 or 3A sieves.27 The results for both butanols on using most of the absorptive type siccatives under static conditions (Table I) indicate a very much slower rate of water absorption than with methanol or ethanol. It seems likely that this is due to the increased viscosity of the C-4 alcohols compared with their lower homologues, and thus the use of 3A sieves in the powder form (which operates as a dispersion) is able to circumvent this difficulty and again appears to be the desiccant of choice.= Interestingly, and in striking contrast to 2-butanol, the use of Na alone does efficiently remove water from tert-butyl alcohol, and this must mean that the equilibrium shown above lies predominantly to the left, presumably as a consequence of the greater disparity between the acidities of the two protonated species in this case.28 I t should be emphasized, however, that sodium (23) The "alcohol method" for the standardization of alkyllithium solutions requires perfectly dry 2-butanol. See: Watson, S. C.; Eastham, J. F. J. Organomet. Chem. 1967, 165. (24) Norrie, J. F.; Olmated, A. "Organic Syntheses";Wiley: New York, 1941; Collect. Vol. I, p 144. (25) (a) Wayne, W.; Adkins, H. "Organic Syntheses"; Wiley: New York. 1955: Collect. Vol. 111. D 48. (b) Hauser. C. R.: Hudson. B. E.: AbrakoviGh, B.; Shivers, J. C.-Ibid.,p 142. (c) Spassow, A. Ibid.; p 145: (d) McElvain, 5. M.; Kundiger, D. Ibid., p 506. (e) Johnson, W. S.; Schneider, W. P. Ibid., 1963; Collect. Vol. IV, p 134. (f) Raha, C. Ibid.,
p 264. (26) (a) Homing, E. C.; Fmelli, A. F. 'Organic Syntheses";Wiley: New York, 1963; Collect. Vol. IV, p 461. (b) Mikol, G. J.; Russell, G. A. Ibid., 1973; Collect. Vol. V, p 937. (c) Ireland, R. E.; Chaykovsky, M. Ibid., p 171. (d) Crowther, G. P.; Kaiser, E. M.; Woodruff, R. A.; Hauser, C. R. Org. Synth. 1971,51, 96. (e) Crockett, G.; Koch, T. H. Ibid., 1979, 59, 132. (27) Itoh, M.; Hagiwara, D.; Kamiya, T. Org. Synth. 1979, 59, 95. (28) pK, values for EtOH and t-BuOH relative to H20 have been reported as -1.94 and -2.6, respectively. See: (a) Lee, D. G.; Cameron, R. J . Am. Chem. SOC.1971,93,4724. (b) Deno, N. C.; Turner, J. 0. J . Org. Chem. 1966,31, 1969.
2422 J. Org. Chem., Vol. 48,No.14,1983
Table 11. Desiccant Efficiency in Dryinga*bof 1,2-Ethanediolc desiccant
residual water content, ppm
3A sieves (bead) 3A sieves (powder) 4A sieves (powder) MgSO, CaC, B203 BaO CaO distillation benzene azeotrope Mg A1
1900 (1200,d 540e) 360f' 1900 (2070)d 3600
residual water content,
residual water content,
Dr ing agent was 3A MS powder at a loading of 5% w/v. Analysis performed by the Karl Fischer method after settling of desiccant, 3-6 h. Initial water content 3600 ppm. Initial water content 4000 ppm. e Initial water content 3700 ppm. f Initial water content 4200 PPm.
a Static drying modes unless specified otherwise. Water content assayed by the Karl Fischer method. Dessicant loading 5% w/v with a drying period of 72 h unless s ecified otherwise; initial water content 2700 168-h drying period. e 600-h drying period. ppm. Analysis was performed after settling of desiccant, - 6 h. Bp 668 Distilled sample, cqntamination by acetylene. 67 "C (0.1 mmHg). ' Sample contaminated by 0.02% v/v benzene. Sample size 1 L; benzene azeotroping Little apparent followed by treatment with Mg/I,. drying.
reflux at an appreciable rate, and, in our opinion, the quite severe solvent losses which can result constitute a drawback serious enough to discourage ita use under these conditions. 1,2-Ethanediol. Although the main thrust of this work was concerned with the lower alcohols, it was t h o u g h t of interest to investigate 1,2-ethanediolas a comparison. This material does in fact possess a high solvation power for particular ions such as fluoride, and the use of dry diol is prescribed for a general synthesis of primary alkyl fluorides which utilizes this solvent property.2e It might be anticipated that the effective drying of ethanediol would prove more problematical than the simple monohydric alcohols, and, indeed, the powerfully hydrophilic nature of this material is indicated by a reference in the patent 1iterature3O where the diol itself is recommended as a dessicant for acetonitrile! There appears to be a paucity of m e t h o d s used for desiccation, due no doubt to the fact that the use of many effective agents such as CaHzis ruled out because of their reaction with this solvent. Most investigators have advocated simple di~tillation,~~ and, in fact, the results show this to be fairly efficaceous (Table 11). With sieves, trends similar to those noted earlier are again observed; thus, due to reasons of slowness of absorption and inappropriate pore size respectively, the 3A bead form and 4A powder form are ineffective, while 3A powder once more demonstrates attributes of powerful desiccant action and simple application. Other noteworthy results were achieved with the use of the benzene azeotrope and with Mg. It would appear that a residual water level of 70 p p m represents the lower limit for t h e desiccation of t h i s material u n d e r practical conditions. Other Diols. Finally, i t was t h o u g h t of interest t o compare the difficulty of drying 1,2-ethanediol with some selected examples of other diols where factors of a changed does in fact react with tert-butyl alcohol under
(29) Vogel, A. I.; Leicester, J.; Macey, W. A. T. 'Organic Syntheses"; Wilev: New York. 1963: Collect Vol. IV. D 525. (36) Derdulla, H. J.; Sommer, M. Ger: (East) 90 136 (Cl. CO~C), May 20,1972; Appl. WP C07c/155952, June 21,1971; Chem. Abstr. 1972, 77,
151525r. (31) See for example: Dickinson, C. L.; Melby, L. R. 'Organic Syntheses";Wiley: New York, 1963; Collect, Vol. IV,p 276. See also ref 28.
Table 111. Relative Difficulty in Dryinga of Some Diolsb
(32) Monson, R. S. "Advanced Organic Synthesis"; Academic Press: New York, 1971; p 141.
positional relationship between the two hydroxyl groups and an increased hydrophobic content might give rise to variations in residual water content. The results (Table 111) suggest that 1,2-diolsare easier to dry than comparable diols where the hydroxyl groups are situated more remotely (cf. 1,2-ethanediol and 1,6butanediol), and this seems reasonable from the viewpoint of preferred intramolecular interactions in 1,2-diols rather than the intermolecular alternatives involving water which are presumably more favorable in other cases. Increasing t h e nonhydrophilic character b y the inclusion of more carbon atoms results in an understandably easier drying process (cf. 1,2ethanediol and 2,3-butanediol; l,4-butanediol and 1,5pentanediol). Extension to higher simple diols was precluded by the limit of their liquid ranges.
Experimental Section Details of the source, activation, and handling of most desiccants have already been described>* Materials such as CiHz and CaCz obtained in an aggregate form were crushed rapidly in a mortar and with a hammer, respectively, immediately prior to use. Where drying methods call for procedures other than simple static interaction, e.g., Mg/Iz, the customary directives were foll~wed.'~ Water contents were determined throughout by the standard Karl Fischer method with a potentiometrically determined end point. Alcohols. Chemically pure or AFt grades of material were used without further purification. In the normal procedure the water content of a freshly opened Winchester bottle was first determined followed by wetting or by drying and wetting to obtain required water concentrations. Drying Experiments. Investigations were carried out on 100-ml samples except where stated otherwise in the tables. Reflux and distillation were carried out under dry nitrogen or argon, and material collected from distillations for analysis was usually a middle cut of 60-75-m1. volume. Where powdered molecular sieves were used as the desiccant, charging of the material into a flask of suitable dimensions was followed by addition of solvent, stoppering of the flask,and vigorous agitation. A clear upper layer suitable for sampling was generally obtained in 3-6 h of standing.
Acknowledgment. W e thank t h e D e p a r t m e n t of Chemistry and University of Malaya for their support of this work under Vote No. F124/80. Registry No. Mg, 7439-95-4;Iz, 7553-56-2; CaHz, 7789-78-8; Na, 7440-23-5; CaC2, 75-20-7; BaO, 1304-28-5; Ca, 7440-70-2; K2C03, 584-08-7; CaO, 1305-78-8; KOH, 1310-58-3; MgS04, 1303-86-2; Al, 7429-90-5; methanol, 67-56-1; 7487-88-9; B203, ethanol, 64-17-5; 2-butanol, 78-92-2; tert-butyl alcohol, 75-65-0; 1,2-ethanediol,107-21-1;2,3-butanediol, 513-85-9; 1,4-butanediol, 110-63-4;1,5-pentanediol,111-29-5;di-2-butyl succinate, 626-31-3; diethyl phthalate, 84-66-2.