1434
Langmuir 1997, 13, 1434-1439
Synthesis, Surface Properties, and Hydrolysis of Chemodegradable Anionic Surfactants: Diastereomerically Pure Sodium cis- and trans-(2-n-Alkyl-1,3-dioxan-5-yl) Sulfates† Andrzej Piasecki,*,‡ Adam Sokołowski,‡ Bogdan Burczyk,*,‡ Roman Gancarz,§ and Urszula Kotlewska‡ Institute of Organic and Polymer Technology and Institute of Organic Chemistry, Biochemistry and Biotechnology, Technical University of Wrocław, 50-370 Wrocław, Poland Received March 7, 1996. In Final Form: December 2, 1996X A systematic study concerning the synthesis, adsorption, micellization, and hydrolytic decomposition of new, chemodegradable and diastereomerically pure sodium cis- and trans-(2-n-alkyl-1,3-dioxan-5-yl) sulfates (alkyl: n-C7H15, n-C9H19, and n-C11H23) has been undertaken. Surface parameters of the compounds under study at the aqueous solution/air interface, i.e., surface tension reduction, surface excess concentration, surface area demand per molecule, and standard free energy of adsorption and micellization, show differences both in the alkyl chain length and in the hydrophilic, i.e., sulfate group configuration at the 1,3-dioxane ring. The cmc values are lower for the trans-isomers than for the cis-isomers, the ∆G°ads and ∆G°cmc values are lower for trans-isomers, and the effectiveness of surface tension reduction is higher for the cis-isomers than for the trans-isomers. The investigated compounds undergo an easy hydrolysis reaction of the acetal function, leading to starting aldehydes and sulfated glycerol. The trans-isomers are hydrolyzed much faster than cis-isomers, and no isomerization reaction of the type cis S trans is observed during the hydrolysis process.
Introduction The largest group of chemodegradable surfactants (for an explanation of the term “chemodegradable surfactants”, see ref 1) are compounds with an acetal grouping in their molecules. Although long chain glycerol acetals have been used for the synthesis of such types of surfactants since 1969,2 they are inconvenient hydrophobic intermediates because they are four-component mixtures of cis- and trans-2-n-alkyl-4-(hydroxymethyl)-1,3-dioxolanes and cisand trans-2-n-alkyl-5-hydroxy-1,3-dioxanes (compounds 3 and 4, respectively, in Scheme 1), which differ considerably in the reactivity of the primary and secondary hydroxyl groups. The various reactivities cause difficulties during their transformation into ionic or nonionic surfactants. Recently, we have developed a very effective method to convert the four-component mixtures of long chain glycerol acetals into two-component mixtures of cis- and trans2-n-alkyl-5-hydroxy-1,3-dioxanes.3,4 The two-component mixtures were then used for the synthesis of surface active anionic sulfate derivatives5-7 with interesting surface * Corresponding authors. † Part 31 in the series Chemical Structure and Surface Activity. Part 30: Piasecki, A.; Sokołowski, A.; Burczyk, B.; Kotlewska, U. J. Am. Oil Chem. Soc., in press. ‡ Institute of Organic and Polymer Technology. § Institute of Organic Chemistry, Biochemistry and Biotechnology. X Abstract published in Advance ACS Abstracts, February 1, 1997. (1) Sokołowski, A.; Piasecki, A.; Burczyk, B. J. Am. Oil Chem. Soc. 1992, 69, 633. (2) Rutzen, H.; Goette, E. German Patent 1542671, 1969. (3) Piasecki, A.; Burczyk, B.; Kotlewska, U. Polish Appl. Patent P-306515, 1994. (4) Piasecki, A.; Sokołowski, A.; Burczyk, B.; Kotlewska, U. Synth. Commun. 1996, 26, 4145. (5) Piasecki, A.; Burczyk, B.; Kotlewska, U. Polish Appl. Patent P-306516, 1994. (6) Piasecki, A.; Burczyk, B.; Kotlewska, U. Polish Appl. Patent P-308929, 1995. (7) Piasecki, A.; Sokołowski, A.; Burczyk, B.; Kotlewska, U. J. Am. Oil Chem. Soc., in press.
S0743-7463(96)00207-7 CCC: $14.00
Scheme 1. Synthesis of Sodium Salts of Sulfated cisand trans-2-n-Alkyl-5-hydroxy-1,3-dioxanes (5a-c)
properties. However, these properties were the properties of the mixture rather than the individual components. Our investigations also proved that nonmicellizing cisand trans-2-n-alkyl-5-hydroxy-1,3-dioxanes clearly differ in surface activity at the aqueous solution/air interface,8,9 and in their susceptibility to hydrolysis in an aqueous solution of hydrochloric acid.10 We have recently extended our investigations by synthesizing diastereomerically pure 2,5-disubstituted (8) Burczyk, B.; Piasecki, A.; We¸ cłas´, L. J. Phys. Chem. 1985, 89, 1032. (9) Lunkenheimer, K.; Burczyk, B.; Piasecki, A.; Hirte, R. Langmuir 1991, 7, 1765. (10) Sokołowski, A.; Burczyk, B. J. Prakt. Chem. 1981, 323, 63.
© 1997 American Chemical Society
Sodium (2-n-Alkyl-1,3-dioxan-5-yl) Sulfates
1,3-dioxane derivatives which are capable of forming micellar solutions and by determining the effect of the surfactant molecules geometric structure on its surface properties and its susceptibility to hydrolytic decomposition in acidic aqueous solution. In this paper, we intend to present the results of our surface activity and hydrolysis investigations of diastereomerically pure sodium cis- and trans-(2-n-alkyl-1,3dioxan-5-yl) sulfates which we obtained according to the synthetic route presented in Scheme 1. The main purpose of the present work was to study the effect of both the alkyl chain length and the polar substituent configuration upon adsorption and micellization processes of the cis- and trans-2,5-disubstituted 1,3-dioxane derivatives under study, as well as their susceptibility to acidic hydrolysis in an aqueous solution.
Langmuir, Vol. 13, No. 6, 1997 1435 Scheme 2. Probable Course of the Hydrolytic Decomposition of 5 in DCl/D2O Solution
Experimental Section General Procedures and Materials. 1H (300 MHz) and 13C (75.5 MHz) NMR spectra were recorded for solutions of the compounds under study in CDCl3 or D2O using a Bruker Avance DRX300 spectrometer with TMS or DSS as standards. Gasliquid chromatography (GLC) was performed using a Giede, GCHF-18.3 (Giede, GDR) apparatus equipped with 10% Silicon XE60 on a GasChrom Q 80/100 packed column (1 m length). Surface tension measurements were performed using a K12e (Kru¨ss, Germany) processor tensiometer for solutions of surfactants in 10-4 M NaHCO3, in triple-distilled water at 31 °C. Equilibrium surface tension data were obtained within 30-60 min. Possible sources of experimental errors were considered11 and eliminated. The accuracy of the measurements was 0.3 mN-1. The aldehydes n-octanal, n-decanal, and n-dodecanal (Merck, Germany) were used after purification by distillation and stabilization with 0.1% hydroquinone. All other reagents were of analytical grade. Synthesis of cis- and trans-2-n-Alkyl-5-hydroxy-1,3dioxanes (cis-4a-c and trans-4a-c). One tenth of a mole of the four-component mixture of glycerol acetals, 3a-c and 4a-c, which was obtained by the known procedure,12 was maintained for 2 days at room temperature in a solution of 50 mL of hexane and then for 5 days at 0-5 °C in the presence of 0.1 g of p-toluenesulfonic acid monohydrate. The precipitated mixture of cis- (cis-4a-c) and trans-2-n-alkyl-5-hydroxy-1,3-dioxanes (trans-4a-c) was filtered off, recrystallized from hexane (7090% yield), and finally subjected to fractional distillation under reduced pressure with separation of pure cis-4a-c as the lower boiling fraction and trans-4a-c as the higher boiling one.3,4 Synthesis of Sodium cis- and trans-(2-n-Alkyl-1,3-dioxan-5-yl) Sulfates (cis-5a-c and trans-5a-c). According to the published procedure,5,7,13 pure geometric isomers of cis(cis-4a-c) and trans-2-n-alkyl-5-hydroxy-1,3-dioxanes (trans4a-c) were sulfated with a sulfur trioxide-pyridine complex in dry carbon tetrachloride solution and neutralized with sodium hydroxide or sodium carbonate to produce sodium salts cis-5a-c and trans-5a-c, respectively, with ca. 90% yields. Hydrolytic Decomposition of Sodium Salts of 1,3-Dioxane Derivatives. A weighed portion of 1,3-dioxane derivative sodium salt cis-5a-b or trans-5a-b was added to 0.5 mL of a 0.1 M DCl/D2O solution in a 5 mm NMR tube. The extent of hydrolysis was determined by 1H NMR analysis of the resultant DCl/D2O solution of surfactant at 27 °C. The calculation was based on electronic integrals for the following signals for cis5a-b: H(a)-2 (t), H(a)-4,6 (d), and H(e)-4,6 (d); for trans-5a-b: H(a)-2 (t), H(a)-4,6 (t), and H(e)-4,6 (dd); and for the hydrolysis reaction product 6: two CH2 quartets and a CH multiplet (see Scheme 2).
Results Syntheses. The hydrophobic intermediates as individual geometric isomers [cis- (cis-4a-c ) and trans-2(11) Lunkenheimer, K.; Wantke, K.-D. Colloid Polym. Sci. 1981, 259, 354. (12) Piasecki, A.; Burczyk, B. Pol. J. Chem. 1980, 54, 367. (13) Piasecki, A. Synth. Commun. 1992, 22, 445.
n-alkyl-5-hydroxy-1,3-dioxanes (trans-4a-c)] were obtained according to the previously published procedure.3,4 Their physicochemical constants are presented in Table 1. The individual geometric isomers cis-4a-c and trans4a-c were converted into sulfate derivatives according to the published procedure,5,7,13 by using a sulfur trioxidepyridine complex as the sulfating reagent. Although the hydrophobic intermediates contain the less reactive secondary hydroxyl group, the sulfate derivatives were obtained in ca. 90 mol % yields. The sulfation reaction products were transformed into sodium salts by neutralization with sodium hydroxide or sodium carbonate (Scheme 1). The physicochemical constants of the resultant sodium cis- (cis-5a-c) and trans-(2-n-alkyl-1,3-dioxan5-yl) sulfates (trans-5a-c) are presented in Table 2. The chemical structure of the sulfate derivatives was confirmed by elemental analysis, while their geometric structure was determined by the analysis of their 1H NMR spectra. The analysis of 1H NMR spectra of the individual geometric isomers of hydrophobic intermediates and their sulfation reaction products, both containing the n-heptyl radical n-C7H15, is presented in Table 3. The chemical shifts and coupling constants determined for the 1,3-dioxane ring protons in the molecules of hydrophobic intermediates and their sulfate derivatives entirely showed the retention of the molecule’s configuration during the sulfation process, i.e., the axial position of the hydroxyl group and the sulfate group at the C-5 carbon atom of the 1,3-dioxane ring in cis-4a-c and cis-5a-c, respectively, and the equatorial position of these functional groups in trans4a-c and trans-5a-c. Additional results of 13C NMR spectra analysis for the selected pair of isomeric sulfate derivatives are presented in Table 4. Surface Activity. The equilibrium surface tension γ of aqueous solutions of the individual surfactants under study was measured at 31 °C because the Krafft point of the undecyl derivative 5c is near 30 °C.7 The obtained values plotted against the logarithm of the molar concentration for cis-5a-c and one pair of cis- and transisomers containing the 2-n-nonyl substituent (cis-5b and trans-5b) are presented in Figures 1 and 2, respectively. Figures 1 and 2 show that aqueous solutions reach the cmc, and that the surface activity of the sodium salts 5a-c is dependent both on the alkyl chain length and on their geometric structures. The surface properties of individual sodium cis- (cis-5a-c) and trans-(2-n-alkyl-1,3-dioxan5-yl) sulfates (trans-5a-c) are presented in Table 5. On the basis of the Gibbs adsorption equation for a 1:1 electrolyte, the surface excess concentration, Γcmc, and then the surface area demand per molecule at the adsorption
1436 Langmuir, Vol. 13, No. 6, 1997
Piasecki et al.
Table 1. Physicochemical Properties of cis- and trans-2-n-Alkyl-5-hydroxy-1,3-dioxanes R ) n-C7H15 bp (°C/mmHg) mp (°C)
R ) n-C9H19
R ) n-C11H23
cis-4a
trans-4a
cis-4b
trans-4b
cis-4c
trans-4c
141.5/10 30-32
161/10 50.5-52
169/10 47-47.5
188/10 63-63.5
170/1.2 54-54.5
188/1.2 72-73
Table 2. Physicochemical Properties of Sodium cis- (cis-5a-c) and trans-(2-n-Alkyl-1,3-dioxan-5-yl) Sulfates (trans-5a-c) elemental analysis [found (calcd) (%)] yielda
entry
alkyl R
isomer
1 2 3 4 5 6
n-C7H15 n-C7H15 n-C9H19 n-C9H19 n-C11H23 n-C11H23
cis-5a trans-5a cis-5b trans-5b cis-5c trans-5c
a
mpb
(mol %)
87.0 89.0 94.5 85.5 88.0 90.5
(°C)
molecular formula
199-203 205-208 195-199 200.5-201 191-195 205-206
C11H21SO6Na C11H21SO6Na C13H25SO6Na C13H25SO6Na C15H29SO6Na C15H29SO6Na
C
H
S
43.2 (43.41) 43.4 (43.41) 46.8 (46.98) 46.7 (46.98) 49.8 (49.98) 49.9 (49.98)
7.1 (6.96) 7.0 (6.96) 7.6 (7.58) 7.5 (7.58) 8.1 (8.11) 8.1 (8.11)
10.5 (10.53) 10.5 (10.53) 9.6 (9.64) 9.5 (9.64) 8.9 (8.89) 8.8 (8.89)
Recrystallized product. b With decomposition. Table 3. Analysis of 1H NMR Spectra of Hydrophobic Intermediates 4a and Their Sulfate Derivatives 5a chemical shiftsa,b,c compound
H(a)-2
H-5
H(e)-4,6
H(a)-4,6
J5.4e
J5.4a
J4gem ) J6gem
cis-4a trans-4a cis-5a trans-5a
4.55 (t) 4.39 (t) 4.63 (t) 4.54 (t)
H(e) ) 3.51 (m) H(a) ) 3.86 (m) H(e) ) 4.33 (m) H(a) ) 4.42 (m)
4.00 (d) 4.16 (t) 4.23 (d) 4.31 (t)
3.88 (d) 3.35 (dd) 4.03 (d) 3.58( dd)
1.5 5.1 1.5 5.2
1.5 10.2 1.5 10.4
11.9 10.2 12.8 11.3
a Downfield from TMS for a solution of 4a in CDCl and from DSS for a solution of 5a in D O. b Chemical shifts for alkyl radical R protons 3 2 were as follows: CH3 ) 0.87 (t); CH3(CH2)5CH2 ) 1.20-1.42 (m); CH3(CH2)5CH2 ) 1.55-1.65. c Data in parentheses denote the following: a, axial; b, broad; d, doublet; e, equatorial; m, multiplet; s, singlet; t, triplet.
Table 4. Analysis of
13C
NMR Spectra of Isomeric Sodium Sulfates 5a carbon positiona
compound
C-2
C-4,6
C-5
C-R
C-β
C-γ
C-δ
C-
C-φ
C-η
cis-5a trans-5a
100.22 100.19
66.96 66.50
69.03 64.43
32.13 31.78
29.46 29.73
26.89 27.28
21.67 22.10
26.89 27.18
20.34 20.51
11.66 11.77
a C-2, C-4,6, and C-5 denote carbon atoms of the 1,3-dioxane ring. C-R to C-η denote the carbon atoms of the alkyl radical R in the R to η positions relative to the 1,3-dioxane ring, respectively.
Figure 1. Equilibrium surface tension-concentration (log c) isotherms of sodium salts of sulfated 1,3-dioxanes: 1, cis-5a; 2, cis-5b; 3, cis-5c.
Figure 2. Equilibrium surface tension-concentration (log c) isotherms of sodium salts of sulfated 1,3-dioxanes: 1, cis-5b; 2, trans-5b.
(1)
Temkin adsorption isotherm.14,15 Some remarks on use of the Temkin-type equation and the Frumkin twodimensional solution approach to find out the assumed standard state and ∆G°ads are given in ref 16. The results, together with some additional data concerning nonionic 2-n-alkyl-5-hydroxy-1,3-dioxanes, are collected in Table 6. The standard free energy of micellization, ∆G°cmc, was
where m ) 1 and 2 for nonionic and ionic surfactants, respectively, and B is an equilibrium constant of the
(14) Parsons, P. Trudy IV Soveshanija po Elektrokhimii (Russ. Ed.); Khimia: Moscov, 1959; p 42. (15) Piasecki, A.; Burczyk, B. Colloid Polym. Sci. 1985, 263, 997. (16) Sokołowski, A. J. Colloid Interface Sci. 1991, 147, 496.
layer, Acmc, were calculated from the γ ) f(log c) plots for points near the cmc. The standard free energy of adsorption, ∆G°ads, was calculated from eq 1:
∆G°ads ) -mRT ln B
Sodium (2-n-Alkyl-1,3-dioxan-5-yl) Sulfates
Langmuir, Vol. 13, No. 6, 1997 1437
Table 5. Surface Properties of Sodium cis- (cis-5a-c) and trans-(2-n-Alkyl-1,3-dioxan-5-yl) Sulfates (trans-5a-c) at 31 °C entry
surfactant
(103)cmc (mol dm-3)
Πcmc (mN m-1)
(106)Γcmc (mol m-2)
(1020)Acmc (m2)
-∆G°ads (kJ mol-1)
-∆G°cmc (kJ mol-1)
1 2 3 4 5 6
cis-5a trans-5a cis-5b trans-5b cis-5c trans-5c
60.2 35.6 14.3 8.43 3.39 2.29
29.3 27.2 30.0 28.5 31.5 29.3
2.91 2.91 2.93 2.65 2.94 2.65
57 ( 2 57 ( 3 57 ( 3 63 ( 3 57 ( 2 63 ( 3
55.7 59.4 63.7 67.9 71.8 74.5
32.7 34.8 38.7 40.9 45.0 46.7
Table 6. Standard Free Energies of Adsorption and Micellization of Nonionic and Anionic 1,3-Dioxane Derivatives entry 1 2 3 4 5 6
alkyl R
polar group
n-C5H11 n-C7H15 n-C5H11 n-C7H15 n-C7H15 n-C7H15
OH OH OH OH OSO3Na OSO3Na
isomer
-∆G°ads (kJ mol-1)
-∆G°ads/CH2 (kJ mol-1)
cis cis trans trans cis trans
29.6a
3.37b
36.2a 31.7a 37.7a 55.7 59.4
3.37b 3.42b 3.42b 4.0c 3.8c
-∆G°cmc/CH2 (kJ mol-1)
3.1c 3.0c
a Data from ref 9, recalculated on the basis of the Temkin adsorption model. The compounds do not form a micellar solution. b Data from ref 9. c Calculated using the data reported in Table 5.
Table 7. Hydrolysis of Sodium cis- (cis-5a-b) and trans-(2-n-Alkyl-1,3-dioxan-5-yl) Sulfates (trans-5a-b) in 0.1 M DCl/D2O Solution at 27 °C surfactant entry
nature
conc (M)
(105) kobs (s-1)
1 2 3 4
cis-5a trans-5a cis-5b trans-5b
0.506 0.506 0.349 0.486
2.15 7.60 0.99 6.0
calculated according to eq 2:17,18
∆G°cmc ) (2 - R)RT ln x(cmc
(2)
where x(cmc ) xcmcy( is the standard mean activity in mole fraction units at the cmc (y( is the mean activity coefficient, log y( ) -0.509J0.5 ) and R is the micellar degree of dissociation (we assume R ) 0.18, as was determined for sodium alkyl sulfates19 ). Surfactants Hydrolytic Decomposition. The kinetics of the hydrolytic decomposition of the individual surfactants containing n-heptyl (5a) and n-nonyl (5b) radicals was investigated in 0.1 M DCl/D2O solution at 27 °C. The results of the investigations are shown in Table 7. All kinetic runs were performed directly in an NMR tube. The extent of the hydrolysis reaction was determined on the basis of integrations for the following signals: the axial methine proton H(a)-2 and axial H(a)-4,6 and equatorial H(e)-4,6 protons of the 1,3-dioxane ring for both geometric isomers of surfactants 5a-b. The linear relationships ln(cA/cA,0) ) f(τ) (where cA, cA,0, and τ denote the running and the initial concentration of surfactant and the reaction time, respectively) were observed in all cases up to at least 80% decomposition. The calculated rate constants of the pseudo-first-order reaction, kobs, are presented in Table 7. Discussion Our synthetic procedure is an effective method for the preparation of diastereomerically pure sulfates of 1,3dioxane derivatives which have the capability to form micellar aggregates in aqueous solution, which makes these compounds easily accessible. The surface activity of the surfactants under study determined from the surface tension measurements at (17) Evans, H. C. J. Chem. Soc. 1956, 7, 579. (18) Sokołowski, A.; Bieniecki, A.; Wilk, K. A.; Burczyk, B. Colloids Surf., A 1995, 98, 73. (19) Philips, J. N. Trans. Faraday Soc. 1955, 51, 561.
the aqueous solution/air interface is dependent not only on the hydrophobic alkyl chain’s length but also on the configuration of the hydrophilic sulfate grouping at the C-5 carbon atom of the 1,3-dioxane ring. All cis diastereomers (cis-5a-c) and trans-5a have identical surface excess concentrations, Γcmc, equal to 2.9 × 10-6 mol m-2, i.e., nearly the same Acmc values. Values of the Γcmc quoted here refer to the values obtained by quadratic fit. However, the Γcmc for cis-5b,c is greater (the surface area Acmc is smaller) than that for trans-5b,c (Table 5), which is opposite to the properties observed for nonmicellizing cis- and trans-2-n-alkyl-5-hydroxy-1,3-dioxanes8,9 (because of the limited solubility in water, the Γ and A values of the latter were determined for points of isotherms close to the solubility limits). Yet, the reasons for this difference are not clear. A limiting area per molecule at the solubility point or at the cmc depends on the method of analysis of the data.20 One explanation of the difference between cis and trans values is that, looking at the γ vs log c curves, we observe a continuous change in the area per molecule right up to the cmc. Moreover, the recent results of Lunkenheimer et al. concerning the cross-sectional areas of sodium alkyl sulfates21 have shown that the areas of the short chain homologues decrease with increasing chain length, reaching minimum values for C10 and C11 alkyl chains, and then increase with alkyl chain lengths Cn with n > 12. The alkyl chains of the sulfates 5a-c under study are C7, C9, and C11; i.e., their Acmc values may represent minimum values. This phenomenon makes the interpretation of our present results difficult. Further studies on this problem will be undertaken in our laboratory. The values of the critical micelle concentration, cmc, and effectiveness of surface tension reduction, Πcmc, are also influenced by the configuration of the polar substituent. trans-Isomers 5a-c are characterized with unequivocally lower cmc values in comparison to cisisomers 5a-c (Table 5), and this difference is greatest for the derivatives with the shortest hydrophobic alkyl chain. Moreover, cis-isomers reduce the surface tension of water more effectively than trans-isomers. Only a few results dealing with the influence of the geometric structure of micellizing surfactants have been presented in the literature written so far. While investigating cis- and trans[(2-n-heptadecyl-2-methyl-1,3-dioxolan-4-yl)methyl]trimethylammonium bromides, Jaeger and co-workers22 (20) Simister, E. A.; Thomas, R. K.; Penfold, J.; Aveyard, R.; Binks, B. P.; Cooper, P.; Fletcher, P. D. I.; Lu, J. R.; Sokołowski, A. J. Phys.Chem. 1992, 96, 1383. (21) Lunkenheimer, K.; Czichocki, G.; Hirte, R.; Barzyk, W. Colloids Surf., A 1995, 101, 187.
1438 Langmuir, Vol. 13, No. 6, 1997
Piasecki et al.
have shown that there is no dependence of cmc values on surfactant stereochemistry within the accuracy of the measurements. Our findings concerning the adsorption parameters of nonmicellizing cis- and trans-2-n-alkyl-4(hydroxymethyl)-1,3-dioxolanes at the aqueous solution/ air interface explain this statement.8 Wang et al.,23 using cis- and trans-[(5-n-dodecyl-1,3-dioxan-2-yl)ethyl]triethylammonium bromides, demonstrated that the cmc of the trans-isomer is only slightly lower than the value for the cis-isomer. Our present results show clearly that the trans-isomers of the homologous series of sodium sulfates bearing a 1,3-dioxane moiety in their molecules exhibit distinctly lower cmc values than the cis-isomers do. The logarithm of the critical micelle concentration (log cmc) is a linear function of the alkyl chain length for both of the series of geometric isomers, i.e., cis-5a-c and trans5a-c (Figure 3), with slopes equal to -0.312 and -0.298, respectively (correlation coefficients 0.999). These values are comparable with data obtained for mixtures of the isomers.7 Moreover, the standard free energies of adsorption, ∆G°ads, and micellization, ∆G°cmc, shown in Table 5, are linear functions of the alkyl chain length. The increments of standard free energies of adsorption and micellization, (∆G°ads/CH2) (kJ mol-1) and (∆G°cmc/CH2) (kJ mol-1), respectively, calculated for one methylene group, CH2, in R are presented in Table 6. We confronted the ∆G°ads values with those of 2-n-alkyl-5-hydroxy-1,3dioxanes (alkyl: n-C5H11 and n-C7H15), taken from the paper by Lunkenheimer et al.,9 which we recalculated on the basis of the Temkin adsorption model. As can be seen, the -∆G°ads values (calculated from eq 1, where m ) 1 and 2 for nonionic and ionic surfactants, respectively) are much higher for sulfates (cis-5a and trans-5a) than for nonionic 1,3-dioxane derivatives. However, the higher surface activity of the trans-isomers than those of the cis ones is evident in both homologous series. The values of (-∆G°ads/CH2) differ for both kinds of surfactants as well; however, these differences are not crucial. On the basis of the literature analogy,24 the hydrolyses of chemodegradable 5a-c are expected to proceed with specific acid catalysis and a rate-determining step involving cleavage of the protonated acetal grouping. Each kobs value for entries 1-4 of Table 7 represents a weighted average of hydrolysis reaction rate constants for unag-
gregated and micellar surfactants, because the kinetic investigations were performed for solutions in DCl/D2O above the cmc (as determined for solutions in H2O). The surfactants under study undergo an easy acidic hydrolysis in aqueous solution. The values of the rate constants of the pseudo-first-order reaction kobs (Table 7) are comparable with those obtained for chemodegradable cationic surfactants bearing the 1,3-dioxolane moiety, i.e., [(2-nalkyl-2-methyl-1,3-dioxolan-4-yl)methyl]trimethylammonium bromides, when hydrolyzed in 0.1 M HBr at 50 °C,22 despite the fact that ketal-type compounds on the one hand and 1,3-dioxolane derivatives on the other are much more susceptible to hydrolytic decomposition than acetaltype 1,3-dioxane derivatives.10,25 The influence of the 1,3dioxacyclane ring structure and the kind of chemodegradable surfactant on the susceptibility to hydrolytic decomposition can be deduced from results published by Wang and co-workers.23,26 [(5-Alkyl-1,3-dioxan-2-yl)ethyl]triethylammonium bromides were completely decomposed in 2 N HCl at 50 °C after 35-90 h depending on the alkyl chain length,23 while dianionic sulfonates bearing the 1,3dioxane moiety were decomposed under the same conditions within 1.5-15 h.26 According to Jaeger’s opinion, the hydrolysis of chemodegradable cationic surfactants of 1,3-dioxolane derivatives above their cmc values in aqueous solutions is much slower than that in the unaggregated form, since the positively charged Stern layer electrostatically repels H3O+.22 The effect of the length of an alkyl substituent in 1,3-dioxacyclane derivatives on the rate of their hydrolysis is not clearly stated. For nonmicellizing glycerol acetals, the susceptibility to hydrolytic decomposition increases with an alkyl chain length increase.10 For micellar solutions of acetal- or ketaltype cationic surfactants, the increase of an alkyl chain length caused the decrease of the rate of hydrolysis22,23 or had a negligible effect.27 The reaction of hydrolysis of isomeric surfactants 5 is so fast that no isomerization reaction cis S trans takes place, and the probable course of decomposition is as presented in Scheme 2. The investigated compounds undergo an hydrolytic cleavage of the acetal function, leading to starting aldehydes 1 and a deuterated derivative of 2-glycerosulfuric acid 6. The chemical structure of 6 was determined by 1H NMR analysis of DCl/D2O solutions after complete decomposition of surfactants. The pseudo-first-order rate constants kobs show that the trans-isomers hydrolyze much faster than the cis-isomers (Table 7). This may be due to the different distances between the acetal oxygen atoms located at positions 1 and 3 of the 1,3-dioxane ring and to the OSO3H group occupying an axial or an equatorial configuration in the cis- and trans-isomers, respectively. If we look at the molecular geometry of the cis-and trans5a-c sulfates (Scheme 2), it is seen that the distance of the oxygen atom of the sulfate group from the acetal oxygen atoms amounts to 0.287 and 0.358 nm in cis- and transisomers, respectively. These dimensions were calculated using the HyperChem computer program, version 4.0 (Hyper Cube, Inc.), on the basis of the force field MM+ approach of molecular mechanics. The acetal oxygen atoms are more shielded by the sulfate group in the cisisomer than in the trans-isomer. In other words, the oxygen atoms of the 1,3-dioxane ring are more accessible to deuterium ion attack in the trans-isomer than in the cis-isomer. It is also probable, on the other hand, that the kinetics of hydrolysis of the sulfates depends on the
(22) Jaeger, D. A.; Mohebalian, J.; Rose, P. L. Langmuir 1990, 6, 547. (23) Wang, G.-W.; Yuan, X.-Y.; Liu, Y.-Ch.; Lei, X.-G.; Guo, Q.-X. J. Am. Oil Chem. Soc. 1995, 72, 83. (24) Cordes, E. H.; Bull, H. G. Chem. Rev. 1974, 74, 581.
(25) Piasecki, A. J. Prakt. Chem. 1985, 327, 731. (26) Wang, G.-W.; Yuan, X.-Y.; Liu, Y.-Ch.; Lei, X.-G. J. Am. Oil Chem. Soc. 1994, 71, 727. (27) Wilk, K. A.; Bieniecki, A.; Burczyk, B.; Sokołowski, A. J. Am. Oil Chem. Soc. 1994, 71, 81.
Figure 3. Logarithm of the critical micelle concentration (log(cmc)) vs the alkyl chain length R: 1, cis-5a-c; 2, trans-5a-c.
Sodium (2-n-Alkyl-1,3-dioxan-5-yl) Sulfates
structures of micellar aggregates formed by cis- and trans5a-c sulfates, which may differ from each other. This assumption requires further investigations which are in progress. What is worth mentioning is that higher ratios of hydrolysis of nonionic, nonmicellizing trans-2-n-alkyl5-hydroxy-1,3-dioxanes in comparison with cis ones were reported earlier,10 yet the differences were not so significant as those for the anionic surfactants 5a-c. Finally, in spite of the decrease in the rate of hydrolysis with an increase of alkyl chain length, the difference between cisand trans-5a,b isomers increases. Conclusions (i) The described synthetic procedure is an effective method for the preparation of diastereomerically pure sodium cis- and trans-(2-n-alkyl-1,3-dioxan-5-yl) sulfates and makes these compounds easily accessible. (ii) The change of polar sulfate grouping configuration at the C-5 carbon atom of the 1,3-dioxane ring from the
Langmuir, Vol. 13, No. 6, 1997 1439
axial position in the cis-isomers to the equatorial position in the trans-isomers of the surfactants under study reveals significant differences in their effectiveness of water surface tension reduction, standard free energies of adsorption, micellization, and hydrolysis parameters. (iii) Besides the investigated susceptibility to hydrolytic cleavage which may solve many problems in, e.g., separation processes using surfactants (foaming, emulsifying), the described surfactants are an interesting object for those studying the polarity and microheterogeneous properties of their micellar solutions. Such studies are in progress in our laboratory. Acknowledgment. Support of this work by the State Committee for Scientific Research, Grant No. 2P30309404, is gratefully acknowledged. LA960207Z
