The effects of ring size and substituents on the rates of acid-catalysed hydrolysis of five- and six-membered ring cyclic ketone acetals

Article abstract of DOI:10.1002/ejoc.200700244

A series of sterically similar five- and six-membered ring cyclic diol-derived ketone acetals have been prepared and their rates of acid-catalysed hydrolysis examined. The rates of hydrolysis are substantially affected by acetal ring conformational stereoelectronic effects and resonance effects depending upon the substituents on the parent ketone; an A1 mechanism of hydrolysis explains the observed effects. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200700244

FULL PAPER
DOI: 10.1002/ejoc.200700244
The Effects of Ring Size and Substituents on the Rates of Acid-Catalysed
Hydrolysis of Five- and Six-Membered Ring Cyclic Ketone Acetals
Jonathan P. Knowles[a] and Andrew Whiting*[a]
Keywords: Acetal / Hydrolysis / Cleavage reactions / Kinetics / Reaction mechanism
A series of sterically similar five- and six-membered ring cy-
clic diol-derived ketone acetals have been prepared and
their rates of acid-catalysed hydrolysis examined. The rates
pending upon the substituents on the parent ketone; an A1
mechanism of hydrolysis explains the observed effects.
of hydrolysis are substantially affected by acetal ring confor- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
mational stereoelectronic effects and resonance effects de-
Germany, 2007)
acetal aldehyde of type 5 was required. During the develop-
ment of a synthesis of systems 5, we discovered surprising
differences in the stability of acetal-functionalised precur-
sors to aldehydes 5, and in this paper, we report these re-
sults and rationalise our findings (Scheme 1).
Introduction
As part of an ongoing project aimed at the construction
of viridenomycin (1),[1,2] we envisaged that a cyclopen-
tenone synthon 2 would be an ideal intermediate from
which we could carry out a stereoselective Michael ad-
ditions for assembly of the northern polyene section. In
turn, the cyclopentenone 2 could be obtained by an intra-
molecular Knoevenagel condensation of the β-keto ester 3a;
the methyl ketone function of which could be derived by an
Results and Discussion
Acetal systems of type 5 have not been reported many
acetal deprotection process. Hence, to access acetal 3b from times in the literature,[3] hence, it was decided that they
an aldol reaction involving a suitable enolate reagent 4, an would be conveniently prepared by ozonolysis of the corre-
Scheme 1. Retrosynthetic analysis of the cyclopentanone core of viridenomycin.
[a] Department of Chemistry, Durham University, Science
sponding vinyl acetals 6, which are acetal derivatives of
methyl vinyl ketone. Attempts to prepare examples of both
5- and six-membered ring acetals was planned using litera-
ture methods[4] from 3-chlorobutanone 7 involving acetal
formation-elimination sequences, as shown in Scheme 2.
Laboratories,
South Road, Durham DH1 3LE, UK
Fax: +44-191-384-4737
E-mail: andy.whiting@durham.ac.uk
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2007, 3365–3368
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3365

J. P. Knowles, A. Whiting
Table 1. Rate constants for the hydrolysis of cyclic acetals.
FULL PAPER
Entry
Acetal
Acid (loading/mol-%)
k1/s–1
1
2
3
4
5
6
7
8
9
8
DCl (50)
DCl (50)
DCl (5)
HCl (50)
DCl (50)
DCl (50)
DCl (5)
0
9
8.8Ϯ0.3ϫ10–4
6.4Ϯ0.1ϫ10–3
4.7Ϯ0.1ϫ10–4
2.25Ϯ0.03ϫ10–6
9.9Ϯ0.6ϫ10–3
6.2Ϯ0.1ϫ10–4
0
9
9
11a
12a
12a
11b
12b
DCl (50)
DCl (50)
7.33Ϯ0.08ϫ10–4
conformation of the propylene glycol ester, such that it re-
sembles the 5-ring acetal system and hence, it is also stable
under these hydrolysis conditions. In addition, clear allylic
effects[5] are observed, i.e. a comparison between acetal
pairs in Entries 1 vs. 2, 5 vs. 6, and 8 vs. 9 highlights this
effect; in the first pair, there is a large difference in the rate
of hydrolysis (stable vs. 10–4), compared with 103 difference
for the second pair, and a repeat stable vs. 10–4 rate for the
last. It is also interesting to note from Entries 2 vs. 3, and
6 vs. 7, that with a ten times reduction in acid concentra-
tion, there is ten-fold decrease in the rate of hydrolysis. The
difference in rate between the five-membered ring vs. six-
membered ring also remains essentially constant even under
these less acidic conditions.
The graphic difference in reactivity moving from alkyl to
allyl substituents supports the most plausible mechanistic
interpretation, which is further supported by the observed
kinetic isotope effect calculated from Entries 2 and 4
(Table 1), i.e. kH/kD = 0.53, which is similar to those pre-
viously reported and is entirely consistent with hydrolysis
occurring by the A1 mechanism as outlined in Scheme 3.[6]
Scheme 2. Formation of 5- and 6-ring acetals 9 and 12.
Initially, vinyl acetal 9 was prepared according to
Scheme 2, however, the corresponding aldehyde resulting
from the ozonolysis proved too volatile to isolate. Attention
was therefore turned to the corresponding six-membered
ring acetals 12 in the hope that the greater ring size might
reduce the resulting aldehyde’s volatility. Vinyl acetals 12a
and 12b were prepared by analogous methods to 9. How-
ever, upon isolation and characterisation of acetal 12a it
was noticed that it was unstable upon standing in CDCl3,
undergoing acid-catalysed acetal hydrolysis to give methyl
vinyl ketone. This is in stark contrast to the five-membered
ring acetal 9, which had shown no such instability. Because
of the marked difference in stability between acetals 9 and
12a, further investigations and kinetic studies of the hydro-
lyses of each of several acetals were undertaken. In this pa-
per, we report the results of these studies.
Acetal pairs 8 and 9, and 11 and 12, were accessed as
described above and outlined in Scheme 2. The rates of the
acid catalysed hydrolysis of each was carried out in a mix-
ture of [D8]THF and D2O using various loadings of DCl
and each of the reactions followed readily by 1H NMR
spectroscopy. In addition, the hydrolysis of vinyl acetal 9
was also carried out using H2O/HCl in order to determine
the kinetic isotope effect. The reactions were performed un-
der pseudo-first order conditions (25 equiv. of D2O) and the
rate constants obtained by plotting ln[product] vs. time.
Scheme 3. General mechanistic scheme for first stages of the acid-
catalysed ring opening of cyclic acetals.
Although it has been previously shown in rigid, bicyclic
Each of the rate constants for all of the acetal hydrolyses systems that the orientation of the oxygen lone pair relative
are summarised in Table 1. to the protonated oxygen leaving group has a substantial
From the results in Table 1, there are several striking ob- effect on the stability of an acetal to hydrolysis,[7] to the
servations. Firstly, there is a considerable ring-size effect best of our knowledge there are no such reports on the
alone on the rate of hydrolysis for unsubstituted acetals, effects of ring-size in such hydrolyses. As discussed above,
i.e. ethylene glycol vs. propylene glycol. Hence, comparing it appears that the conformation of the acetal ring is cer-
Entries 1 and 5, we see a clear transition from a stable ace- tainly an important factor in governing the observed reac-
tal under the 50% acid conditions, to a system which does tion rates and therefore, stereoelectronic effects are the
hydrolyse under these conditions. At first sight, comparing underlying major controlling influences that result from a
Entry 8 with these two Entries (1 and 5) seems to contradict combination of ring size, conformational and acetal substit-
this analysis, however, the 2,2-dimethylpropanediol acetal uents. For unsubstituted five vs. six-ring acetals, the flatter
effectively behaves like ethylene glycol, i.e. the effect of the conformation of a five membered ring reduces the capa-
gem-dimethyl substituents is to flatten the normal chair bility of the pseudo-equatorial oxygen lone pair to interact
3366
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3365–3368

Hydrolysis of 5- and 6-Membered Ring Cyclic Ketone Acetals
FULL PAPER
with the σ* orbital of the protonated oxygen leaving group,
i.e. in structure 14. Hence, ring opening is either slow (vinyl-
substituted) or non-existent (alkyl-substituted).
Addition of a gem-dimethyl substituent on the diol-de-
rived section of the six-ring acetal mimics this effect, by
flattening the chair conformation and causing misalignment
of the equatorial oxygen lone pair to and σ* orbital of the
Experimental Section
General Experimental: All 1H NMR spectra were recorded with
either a 400 or 500 MHz spectrometers. 13C NMR spectra were re-
corded at 100 or 125 MHz. Chemical shifts are expressed as parts
per million downfield from the internal standard TMS. Column
chromatography was performed on silica gel, 60 mesh. TLC was
performed on plastic backed silica gel plates with visualisation
protonated oxygen. Finally, the effect of an additional vinyl achieved using a UV lamp or by using a potassium permanganate
stain. All other materials were purchased directly from standard
chemical suppliers and used without further purification, unless
stated otherwise.
group (i.e. 13 R = CH=CH2, Scheme 3) to the acetal carbon
is to significantly speed up the ring-opening reaction which
can be explained by lowering the energy of the forming ox-
onium ion 15 (R = CH=CH2) due to conjugation with the 2-(1-Chloroethyl)-2-methyl-1,3-dioxolane (8):
A solution of 3-
chloro-2-butenone (23.2 mL, 0.23 mol), ethylene glycol (12.4 mL,
0.22 mol) and pTsOH·H2O (170 mg, 0.89 mmol) in benzene
(330 mL) was heated to reflux in a flask equipped with a Dean–
Stark system. After 21 h the mixture was cooled, the benzene evap-
orated and the crude product purified by kugelrohr distillation to
give the product (27.4 g, 83%) as a clear oil. All spectroscopic and
analytical properties were identical to those reported in the litera-
ture.[4]
alkene and the allyl cation stabilisation effect which occurs
in these systems. The rate increase due to this effect alone
is of the order of 103 or greater. Although such observations
have been previously reported,[8] the combination of ring
size, conformational and substituent effects show how im-
portant it is to ensure the correct choice of acetal for bal-
ancing stability with ease of acetal removal.
2-Methyl-2-vinyl-1,3-dioxolane (9): Compound 8 (10.5 g, 70 mmol)
was added to a solution of potassium hydroxide (24 g, 0.43 mol) in
ethylene glycol (50 mL) at 130 °C, and the mixture stirred at 125 °C
under argon. After 3 h the temperature was increased to 160 °C
and product distilled from the reaction mixture to give the product
(6.62 g, 83%) as a clear oil. All spectroscopic and analytical proper-
ties were identical to those reported in the literature.[4]
Conclusions
This study reinforces previous assertions[9] and unambig-
uous results[6b] that the acid-catalysed hydrolysis of cyclic
acetals involves the formation of a protonated species 14,
followed by discrete bond cleavage to the oxonium ion 15
following an A1 mechanism. Although steric effects relating
to R and RЈ (on 13) have also been studied,[10] these have
little impact upon the series of compounds examined as
part of this study, i.e. 8, 9, 11 and 12, because all have steri-
cally similar functional groups. In the case of these systems,
2-(1-Chloroethyl)-2-methyl-1,3-dioxane (11a):
A solution of 3-
chloro-2-butenone (23.2 mL, 0.23 mol), 1,3-propanediol (10a,
16.5 mL, 0.23 mol) and pTsOH·H2O (1.0 g, 5.2 mmol) in benzene
(330 mL) was heated to reflux in a flask equipped with a Dean–
Stark system. After 20 h the mixture was cooled, the benzene evap-
orated and the crude product purified by kugelrohr distillation to
give the product (33.2 g, 89%) as a clear oil. B.p. 100 °C at 1 Torr.
IR (film): ν
= 2960, 2872, 1480, 1449, 1372, 1247, 1153, 1057
˜
max
acetal ring strain is also not expected to be major,[11] hence, cm–1. 1H NMR (CDCl3, 400 MHz): δ = 1.43 (s, 3 H, Me), 1.45 (d,
3J = 6.8 Hz, 3 H, Me), 1.48–1.57 (m, 1 H, CH2), 1.76–1.86 (m, 1
the observed rates of hydrolysis are substantially affected
3
H, CH2), 3.77–3.97 (m, 4 H, 2ϫCH2O) and 4.12 (q, J = 6.8 Hz,
by acetal ring conformational stereoelectronic effects in all
1 H, CHCl) ppm. 13C NMR (CDCl3, 101 MHz): δ = 15.7, 19.2,
cases, and by resonance effects[11] in the case of conjugated
25.2, 59.9, 60.2 and 99.4 ppm. MS (ES+): m/z = 167 [M + H+,
systems 9 and 12. These conformational stereoelectronic ef-
fects (alignment of the equatorial or pseudo-equatorial lone
37Cl], 165.0677 [M + H+, C7H14O235Cl, requires 165.0677, 100%].
2-Methyl-2-vinyl-1,3-dioxane (12a): To a stirred solution of potas-
pair with the σ* of the breaking C–O bond, see Figure 1)
sium hydroxide (14.4 g, 0.26 mol) in 1,3-propanediol (30 mL) at
explain why a five-membered ring ethanediol-derived acetal
strongly resembles a six-membered ring 2,2-dimethylpro-
pane diol-derived acetal, and why the corresponding pro-
panediol-derived acetals hydrolyse more readily than both
125 °C was added 11a (7.10 g, 43.2 mmol) and the mixture stirred
at 125 °C. After 2 h the temperature was increased to 165 °C and
the product distilled from the reaction mixture and separated from
water to give product (4.73 g, 86%) as a clear oil. B.p. 123 °C at
of these.
760 Torr. IR (film): ν
= 2962, 1645, 1403, 1369, 1249, 1191,
˜
max
1143, 1086 cm–1. 1H NMR (CDCl3, 400 MHz): δ = 1.38 (s, 3 H,
Me), 1.94–2.06 (m, 2 H, CH2), 3.78–3.98 (m, 4 H, 2ϫCH2), 5.36
4
4
(dd, 3J = 10.8, J = 1.2 Hz, 1 H, =CHH), 5.39 (dd, 3J = 17.6, J =
3
3
1.2 Hz, 1 H, =CHH) and 5.80 (dd, J = 17.6, J = 10.8 Hz, 1 H,
CH=CH2) ppm. 13C NMR (CDCl3, 101 MHz): δ = 25.8, 29.0, 61.2,
99.2, 118.1, 138.9 ppm. MS (ES+): m/z 129.0911 [M + H+,
C7H13O2, requires 129.0910], 99, 79 and 60 (100%).
2-(1-Chloroethyl)-2,5,5-trimethyl-1,3-dioxane (11b): A stirred solu-
tion of 3-chlorobutanone (23.2 mL, 0.23 mol), 2,2-dimethyl-1,3-
propanediol (24 g, 0.23 mol) and pTsOH·H2O (1.0 g, 5.2 mmol) in
benzene (330 mL) was heated to reflux in a round bottomed flask
equipped with a Dean–Stark apparatus. After 23 h the mixture was
cooled, the benzene evaporated and the resulting oil dissolved in
Figure 1. Equatorial lone pair to σ* interaction in a protonated
cyclic acetal.
Eur. J. Org. Chem. 2007, 3365–3368
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3367

J. P. Knowles, A. Whiting
FULL PAPER
Et2O (250 mL). The solution was washed with satd. aq. NaHCO3
(150 mL), water (150 mL) and brine (150 mL), dried (MgSO4) and
evaporated to give product (40.3 g, 90%) as a yellow oil. IR (film):
Acknowledgments
We are grateful to the Engineering and Physical Sciences Research
Council (EPSRC) for a DTA award to J. P. K. and to the EPSRC
mass spectrometry service at the University of Wales, Swansea.
ν
= 2980, 1473, 1396, 1329, 1311, 1213, 1080 cm–1. 1H NMR
˜
max
(CDCl3, 500 MHz): δ = 0.83 (s, 3 H, CMeMe), 1.06 (s, 3 H,
3
CMeMe), 1.43 (s, 3 H, MeCO2), 1.51 (d, J = 7 Hz, 3 H, MeCH),
3.44 (d, 3J = 11.5 Hz, 1 H, CHH), 3.46 (d, 3J = 11.5 Hz, 1 H,
CHH), 3.59 (t, 3J = 12 Hz, 2 H, CH2) and 4.11 (q, 3J = 7 Hz, 1 H,
CHCl) ppm. 13C NMR (CDCl3, 126 MHz): δ = 15.2 (CMeMe),
19.3 (CMeMe), 22.4 (MeCO2) 22.9 (MeCHCl), 30.1 (CMe2), 60.5
(CHCl), 70.6 (CH2), 70.8 (CH2) and 99.3 (CO2) ppm. MS (CI+):
m/z (%) = 195 [M+, 37Cl], 193 (100) [M+, 35Cl, ], 159, 129. HRMS
(ES+) 193.0989 [M + H+], C9H18ClO2, requires 193.0990.
[1] a) T. Hasegawa, T. Kamiya, T. Henmi, H. Iwasaki, S. Yamatod-
ani, J. Antibiot. 1975, 28, 167–175; b) M. Nakagawa, K. Furih-
ata, Y. Hayakawa, H. Seto, Tetrahedron Lett. 1991, 32, 659–
662; c) M. Nakagawa, Y. Toda, K. Furihata, Y. Hayakawa, H.
Seto, J. Antibiot. 1992, 45, 1133–1138.
[2] a) J. Ishihara, K. Hagihara, H. Chiba, K. Ito, Y. Yanagisawa,
K. Totani, K. Tadano, Tetrahedron Lett. 2000, 41, 1771–1774;
b) B. M. Trost, C. Jiang, Org. Lett. 2003, 5, 1563–1565; c) N. P.
Mulholland, G. Pattenden, Tetrahedron Lett. 2005, 46, 937–
939; d) G. Pattenden, A. J. Blake, L. Constandinos, Tetrahedron
Lett. 2005, 46, 1913–1915; e) M. P. Arrington, A. I. Meyers,
Chem. Commun. 1999, 1371–1372; f) A. W. Kruger, A. I. Mey-
ers, Tetrahedron Lett. 2001, 42, 4301–4304; g) A. G. Waterson,
A. W. Kruger, A. I. Meyers, Tetrahedron Lett. 2001, 42, 4305–
4308; h) G. N. Maw, C. Thirsk, J.-L. Toujas, M. Vaultier, A.
Whiting, Synlett 2004, 1883–1886; i) A. S. Batsanov, J. P.
Knowles, A. Whiting, J. Org. Chem. 2007, 72, 2525–2532.
[3] a) B. M. Trost, A. Fettes, B. T. Shireman, J. Am. Chem. Soc.
2004, 126, 2660–2661; b) D. Bernard, A. Doutheau, J. Gore,
Tetrahedron 1987, 43, 2721–2732.
[4] a) B. A. Patel, J. I. Kim, D. D. Bender, L.-C. Kao, R. F. Heck,
J. Org. Chem. 1981, 46, 1061–1067; b) P. Page, C. Blonski, J.
Perie, Bioorg. Med. Chem. 1999, 7, 1403–1412.
[5] M. M. Kreevoy, R. W. Taft, J. Am. Chem. Soc. 1955, 77, 5590–
5595.
[6] a) T. H. Fife, Acc. Chem. Res. 1972, 5, 264–272; b) S. R. Wann,
M. M. Kreevoy, J. Org. Chem. 1981, 46, 419–423.
[7] C. M. Evans, R. Glenn, A. J. Kirby, J. Am. Chem. Soc. 1982,
104, 4706–4707.
[8] J. L. Jensen, R. Siegel, J. Org. Chem. 1988, 53, 6105–6106.
[9] D. Drake, R. L. Schowen, H. Jayaraman, J. Am. Chem. Soc.
1973, 95, 454–458.
2,5,5-Trimethyl-2-vinyl-1,3-dioxane (12b): To a stirred solution of
potassium hydroxide (28.8 g, 0.50 mol) in ethylene glycol (60 mL)
at 120 °C was added 11b (8.63 g, 44.4 mmol) and the temperature
increased to 160 °C. After 23 h the mixture was cooled, diluted
with water (300 mL) and extracted with Et2O (3ϫ150 mL). The
combined organic phase was washed with water (100 mL) and
brine (100 mL), dried (MgSO4) and evaporated to give product
(4.79 g, 68%) as a pale yellow oil. IR (film): ν
= 2689, 1645,
˜
max
1472, 1396, 1369, 1239, 1089 cm–1. H NMR (CDCl3, 400 MHz):
1
δ = 0.70 (s, 3 H, CMeMe), 1.16 (s, 3 H, CMeMe), 1.40 (s, 3 H,
3
3
MeCO2), 3.32 (d, J = 11 Hz, 2 H, 2ϫCHH), 3.58 (d, J = 11 Hz,
3
3
2 H 2ϫCHH), 5.32–5.39 (m, 2 H, CH2), 5.75 (dd, J = 17.6, J =
10.8 Hz, 1 H, CH=CH2) ppm. 13C NMR (CDCl3, 101 MHz): δ =
22.1 (CMeMe), 22.8 (CMeMe), 28.7 (MeCO2), 30.2 (CMe2), 71.6
(2ϫCH2), 98.8 (CO2), 118.0 (CH2=CH), 138.6 (CH2=CH) ppm.
MS (EI): m/z = 157, 141, 129, 71, 43. HRMS (ES+) 157.1222
[C9H17O2, M + H+, requires 157.1223].
General Procedure for Acetal Hydrolysis: The appropriate acetal
(0.23 mmol) was dissolved in [D8]THF (0.75 mL) and observed by
1H NMR spectroscopy. D2O (0.100 mL) and the appropriate
amount of DCl in D2O (0.010 mL) were added (giving acid concen-
trations of 0.16  and 0.016  for 50% and 5% loadings respec-
tively), the mixture shaken and the hydrolysis followed over time at
[10] T. H. Fife, L. Hagopian, J. Org. Chem. 1966, 31, 1772–1775.
[11] R. F. Atkinson, T. C. Bruice, J. Am. Chem. Soc. 1974, 96, 819–
825.
1
293 K by H NMR spectroscopy.
Supporting Information (see also the footnote on the first page of
Received: March 19, 2007
this article): All kinetic NMR spectroscopic data is available.
Published Online: May 15, 2007
3368
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3365–3368