4958
-i I
-3
I
-2
-I
I
- 4 - 3 - 2 - 1
I
9
0
I
I
I
I
1
I
,
0
I
2
3
4
5
2
3
4
“/ # I 6
L
T
Figure 5 . “r spectrum at 40“ of compound C (1,3-bisdehydro[16]annulene) (5), measured in carbon tetrachloride at 100 Mcps.
difference in environment of each of these protons in the conformers. The four conformers of 5 are 5a-d,I2 only rotation about the H3,H4 and the H9,H10 trans double bonds being considered. The chemical shifts suggest that H3,H lo occupy an inner position about two-thirds of the
5a
5c
5b
5d
time, whereas H4, H9 occupy an inner position about one-third of the time. Cooling to lower temperatures
than - 80” would presumably give information about the relative contribution of the various conformers. On warming a carbon tetrachloride solution of 5, the chemical shifts of H 3 and HIo become different from each other, as do those of H 4 and Hg. As a result, at 40” (Figure 5 ) the H3, HIo and H4, Hg bands each exhibit eight lines. The substance decomposed relatively quickly at 40°, and it was not possible to investigate the spectrum at higher temperatures. Conclusion
The substances 4 and 5 are the first dehydroannulenes in which interconversion between nonequivalent conformers has been observed. Substance 4 is also the first example of a dehydroannulene in which the protons on a trans double band adjacent to an acetylene are being transferred between internal and external positions. The fact that the inner protons in all three compounds 1, 4, and 5 appear at considerably lower field than the outer protons provides evidence for the existence of a magnetically induced paramagnetic ring current. Each of the three substances has 16 out-of-plane 7~ electrons, and this finding (which parallels that, made with [ 161annulene)’ is in accord with the predictions made for [4n]annulenes and dehydro[4n]annulenes. I6 Acknowledgment. I. C. C. thanks the C.S.I.R.O. (Australia) for an Overseas Postgraduate Studentship. We are also grateful to Dr. C. W. Haigh (Swansea) for valuable discussions. (16) J. A. Pople and K. G. Untch, J . Amer. Chem. Soc., 88, 4811 (1966); F. Baer, H. Kuhn, and W . Regel, Z . Nufurforsch., 22a, 103 (1967); H. C. Longuet-Higgins, Special Publication No. 21, The Chemical Society, London, 1967, p 109.
The Mechanism of Hydrolysis of Methyl Pseudo-2-benzoylbenzoate in Aqueous Sulfuric Acid’ Daniel P. Weeks, Alex Grodski,2 and Roy Fanucci2
Contribution f r o m the Department of Chemistry, Seton Hall University, South Orange, New Jersey 07079. Received February 28, 1968 Abstract: The hydrolysis of methyl pseudo-2-benzoylbenzoate (2) in 1 M aqueous sulfuric acid is characterized by a AS* of - 19.4 eu and a D20solvent isotope effect O f k H z O / k D z O= 0.50. In 5 M sulfuric acid these values become - 19.1 eu and 0.56, respectively. In moderately concentrated sulfuric acid a plot of log k+ us. -Hois linear with a slope of 0.67. A plot of log k+ Hous. Ho log [H+]shows downward curvature. The slopes,4, are 0.66 at 1 M and 0.41 at 5 M . In 90 aqueous acetonitrile containing sulfuric acid (-)-menthyl pseudo-(-)-Zbenzoylbenzoate undergoes racemization much faster than hydrolysis. Thus, the evidence supports a mechanism which is bi-
+
+
molecular.
they could choose to react, as most simple esters
by the AAc2 mechanism shown in Scheme I. On the other hand 1 might prefer to react as most acetals do4 (A1 mechanism, Scheme I). Considering 1 as an ester the A1 mechanism would correspond to the AAl1 mecha-
(1) Presented, in part, at the 154th National Meeting of the American Chemical Society, Chicago, Ill., Sept 1967, Abstract S 175. (2) Taken in part from the M.S. Theses submitted by A. G. (1966) and R. F. (1967) to Seton Hall University.
(3) C. K. Ingold, “Structure and Mechanism in Organic Chemistry,” Cornel1 University Press, Ithaca, N. Y.,1953, p 767; M. L. Bender, Chem. Reu., 60, 53 (1960). (4) F. A. Long and M. A. Paul, ibid., 57, 935 (1957).
everal attractive mechanisms can be written for the S acid-catalyzed hydrolysis of partial acylals, Since these compounds are the esters of hemiacetals
1.
Journal of the American Chemical Society / 90:18 / August 28, 1968
4959
1.1..-
Scheme I
3
0 -2,8: a -3.2
+OH
/I
RC- 0
-
OR'
Hd'
SLOW
H2O
OH I
Q
- Ho OR' + il
OR'
RCOOH
+
CHR"
Figure 1, Hydrolysis of methyl pseudo-2-benzoylbenzoate in sulfuric acid solutions; plot of log k+ against -Ho.The slope is
f0.67.
+AHZ SEVERAL
- H + / STEPS
+
RCOOH
R'OH
+
~'CHO
dOH
+
d'CHO
nism which has been demonstrated for such esters as t-butyl acetate.6 Several other mechanisms can be written and will be considered at the conclusion of this paper. In a study of the mechanisms of hydrolysis of the partial acylals methoxymethyl acetate, ethoxymethyl acetate and methoxymethyl formate, Salomaae concluded that the two alkoxymethyl acetates were undergoing hydrolysis by the unimolecular process. The behavior of methoxymethyl formate indicated that although most of the reaction was unimolecular there was a small bimolecular component.6 More recently, Fife7 has reported that the hydrolysis of y-ethoxy-y-butyrolactone is unimolecular. This paper reports our first efforts to investigate the reliability of the various criteria used to investigate the mechanisms of hydrolysis reactions. We wish to observe these criteria under conditions which are likely to test their credibility. The considerations above led us to believe that partial acylals would provide us with a rare opportunity to observe a change in mechanism over a narrow range of substrate structure. We could then observe the various criteria of mechanism as we swept through the region of changing mechanism. We have undertaken a study of the hydrolysis of methyl pseudo-2-benzoylbenzoate (2) in aqueous sulfuric acid. This compound provides a partial acylal, albeit cyclic, which is reasonably easy to prepare and purify. Further there is ample opportunity for subtle structural variation within the pseudo ester system.
2
3
( 5 ) C. A. Bunton, A. E. Comyns, and J. L. Wood, Research (London),
6, 383 (1951).
(6) P. Salomaa, Acra Chem. Scand., 11, 132, 141, 235 (1957). (7) T. H.Fife, J . Amer. Chem. Soc., 87, 271 (1965).
Results and Discussion Methyl pseudo-2-benzoylbenzoate ( 2 ) reacts in aqueous sulfuric acid to form 2-benzoylbenzoic acid (3) and methanol. The reaction shows straightforward pseudo-first-order behavior. Rate constants under various conditions are shown in Table I. Table I. Methyl Pseudo-2-benzoylbenzoateHydrolysis in Aqueous Sulfuric Acid
No.
lO'[substrate], M
1 2 3 4 5 6 7 8 9 10 11 12 13 14
5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 4.6 8.8 4.6 8.8 4.6
[H2S04],
Temp, "C
104k+,a
M 1.01 2.03 2.59 3.08 3.61 4.12 5.19 1.01 1.01 5.12 5.12 5.12 5.12 5.12
29.8 29.8 29.8 29.8 29.8 29.8 29.8 40.0 50.6 25.8 25.8 34.0 34.0 45.1
1 . 1 1 & 0.03 2.88 f 0.05 4.27 i 0.03 6.64 i 0.13 9.65 i 0.05 13.8 i 0.13 29.6 i 0.05 2.94 f 0.05 7.24 f 0.03 23.7 i 0.50 23.6 i 1.8 1.0 47.8 =I= 48.0 =k 0.90 137 i 13
sec-l
Average of at least two runs. Confidence intervals are based on a 95 confidence level.
Criteria Based on Acidity Functions. When the logarithm of the pseudo-first-order rate constant (log k+)for hydrolysis of 2 is plotted against the Hammett acidity function (- H$ the data show good linearity (Figure 1). The Zucker-Hammett hypothesis4g9 predicts a linear correlation for a reaction in which no water is involved in the transition state of the rate-determining step, Le., A l . The behavior of this reaction is far from ideal, however, because the Zucker-Hammett hypothesis requires a slope of unity. Small deviations from the ideal slope are usually tolerated but the observation of a slope of 0.67 in this work makes a useful application of the Zucker-Hammett hypothesis of doubtful value. Plotting log k+ against log [H+], which should result in linearity if a molecule of water (8) L. P.Hammett and A. J. Deyrup, ibid., 54, 2721 (1932); M. A. Paul and F. A. Long, Chem. Rev., 57,l (1957). (9) L. Zucker and L. P. Hammett, J . Amer. Chem. Soc., 61, 2791 (1939); L. P.Hammett, "Physical Organic Chemistry," McGraw-Hill Book Co., Inc., New York, N. Y., 1940, pp 273-277.
Weeks, Grodski, Fanucci
1 Hydrolysis of Methyl Pseudo-2-benzoylbenzoate
4960
r
c I-4.7
+0-4’q
Y 3.
a -4.5
0
b
b
0
0
0
b
0
0
b
-41
-4.3
I -0.5
00
I -10
LOG -4.1
I
- 0.5
0.0
- 1.5 I
I
- 1.0 H+, L O G [Hi]
I
-2
%o,
Figure 3. Hydrolysis of methyl pseudo-2-benzoylbenzoate in sulfuric acid solutions; plot of (log k+ M H O against ) log U H ~ O ; m = 0.90. Log uH2ovalues taken from ref 10.
+
is involved in the transition state, gives severe curvature. Clearly, the reaction does not fit either of the ZuckerHammett categories. Bunnett lo and Bunnett and Olsen” have suggested alternative methods for treating these data. The more recent of these is the parameter.” Using these cp values Bunnett has assigned a large number of organic reactions occurring in moderately concentrated mineral acid to mechanistic categories with remarkable success, although he cautions against the too literal interpretation of $J values in terms of mechanism. The appropriate treatment of our data is shown in Figure 2. The data show significant curvature. This was disappointtreatment usually results in “arrow ing since the straight” correlations even in cases where the older w p1otsl0 were badly curved. It is interesting to note that r$ for the hydrolysis of 2 lies intermediate between those values resulting from substrates known to hydrolyze cia an A2 mechanism and those from substrates undergoing A1 -type mechanisms. Data collected by Bunnett” allow a wide choice of model compounds. These are presented in Table I1 to show all of the structural features of methyl pseudo-2-benzoylbenzoate. Compound 10 is of special interest. Salomaa6 has postulated that this compound undergoes hydrolysis largely uia an AI mechanism but with an observable A2 component. He supported this with some elegant work in aqueous methanol solution. The value for 10 is noticeably more positive than those for 8 and 9. This indicates that is a sensitive probe for changes in reaction mechanism. We have considered two interpretations of these data. Initially, we believed that the fact that r$ for the hydrolysis of 2 lay intermediate between those values expected for the A1 and A2 mechanisms indicated that the reaction was proceeding cia a mixture of unimolecular and bimolecular pathways. Curvature of the plot indicated that the proportion of molecules undergoing hydrolysis by the two pathways changed as the coricentration of mineral acid changed. That is, at sulfuric acid concentrations of about 1 M a large numa bimolecular pathbcr of molecules hydrolyzed ~ i the way but at about 5 M sulfuric acid the AI pathway had
become more important. An alternative interpretation would be that 2 suffers an hydrolysis reaction cia a bimolecular mechanism which does not obey the acidity function criteria. Table 11. Bunnett Values for Methyl Pseudo-2-benzoyibenzoate, Selected Esters and Selected Acetals
+
+
+
+
+
(10) J. F. Bunnett, J . Amer. Chem. SOC.,83, 4956, 4968, 4973, 4978 (1961). (11) J. F. Bunnett and F. P. 01$en, Can. J . Chem., 44, 1899, 1917 (1966).
Journal of the American Chemical Society
L
I -2.0
+
Figure 2. Hydrolysis of methyl pseudo-2-benzoylbenzoate in sulfuric acid solutions; plot of (log k+ Ha) against (Ho log [H+]). The slopes, 4, are Jr0.66 (1 M ) and $0.41 (5 M ) .
+
I
-I 5
90:18
Slope of log
No. 2
Compound
#Ja
0.66 (1 M HzS04)*
Methyl pseudo-2benzoyl benzoate
k+ us. Mech-H0 anism 0.67b
0.41 (5 M H Q S O ~ ) ~
4 Methyl benzoate 5 a-Glyceryl monobenzoate 6 Ethyl acetate 7 y-Butyrolactone 8 Methoxymethyl acetate 9 Ethoxymethyl acetate 10 Methoxymethyl for mat e 11 Methylal 12 &Butyl acetate 13 Methyl mesitoate a
Reference 11.
0.98 (HC104) 0.99 (HClO4)
A2 A2
0.84 (HzS04)
0.87 (HC1) -0.06 (H2S04)
A2 A2 1.15c A1
-0.44 (HCI)
1.1P A1
0.15 (HCI)
1.OOc A1 (A2)
-0.005 ( H S O J -0.21 (HCl)
A1 A1 A1
-0.25 (HZSO4)
* Present
work.
c
Reference 6.
YatesI2 has suggested a modification of the Bunnett hydration parameter, w.lo The modification involves using an acidity function which is appropriate to the type of substrate being investigated. The Yates treatment eliminates the well-known defi~iencies’~associated with the use of Ho but requires the tedious process of obtaining indicator slopes for each type of substrate. Yates and his coworkers have had striking success in correlating rate with acidity and water activity for amidesI2 and acetate esters.I4 Figure 3 shows our data for the hydrolysis of 2 plotted according to Yates’s method where we have taken n z = 0.90.15 The positive dependence on water activity (12) I
