Products from solvolysis reactions that form (2-phenylcyclopropyl)carbinyl cations

Article abstract of DOI:10.1002/poc.951

Products from solvolytic reactions that form the (2-phenylcyclopropyl) carbinyl cation were determined. The majority of products (> 98%) derived from the 1-phenyl-3-butenyl cation, consistent with reports by Wiberg and co-workers. Small amounts of products derived from the 1-phenyl-1- cyclopropylmethyl cation also were found; these products were previously predicted to be formed from reactions of the title cation. Although the 1-phenyl-1-cyclopropylmethyl cation is considerably more stable than the 1-phenyl-3-butenyl cation, it is not kinetically accessible under a variety of solvolytic conditions. Copyright

Full text of DOI:10.1002/poc.951

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2005; 18: 974–977
Published online 13 June 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.951
Products from solvolysis reactions that form
(2-phenylcyclopropyl)carbinyl cations
R. Esala P. Chandrasena, David Aebisher and Martin Newcomb*
Department of Chemistry, University of Illinois at Chicago, 845 West Taylor Street, Chicago, IL 60607, USA
Received 18 January 2005; revised 30 March 2005; accepted 1 April 2005
ABSTRACT: Products from solvolytic reactions that form the (2-phenylcyclopropyl)carbinyl cation were deter-
mined. The majority of products ( > 98%) derived from the 1-phenyl-3-butenyl cation, consistent with reports by
Wiberg and co-workers. Small amounts of products derived from the 1-phenyl-1-cyclopropylmethyl cation also were
found; these products were previously predicted to be formed from reactions of the title cation. Although the 1-
phenyl-1-cyclopropylmethyl cation is considerably more stable than the 1-phenyl-3-butenyl cation, it is not
kinetically accessible under a variety of solvolytic conditions. Copyright # 2005 John Wiley & Sons, Ltd.
KEYWORDS: solvolysis; 2-(phenylcyclopropyl)carbinyl cations
INTRODUCTION
the experimental and computational work reported by
Wiberg and co-workers,5,6 cation 5a was expected to give
benzylic cation 6 that led to alcohol 4 (Scheme 2). Others
have interpreted evidence for cationic intermediates in
enzyme-catalyzed reactions as implicating the formation
of free carbocations, possibly by H-atom abstraction to
give a radical that is oxidized.7 In the context of probe 1,
such a sequence would give cation 5b, but Shaik and co-
workers recently predicted that cation 5b would rear-
range to cation 7, which would lead ultimately to alcohol
8 (Scheme 2).8,9 If that were the case, then the products
from the enzyme reactions might permit a distinction
between formation of protonated alcohol 5a and carboca-
tion 5b formed by oxidation of a radical. Indeed, Shaik
and co-workers concluded that the formation of 4 and
absence of 8 demonstrated that carbocations were not
formed in P450-catalyzed oxidations of probe 1.8
Cation 7 is well known to be a relatively low-energy
species, with a stability similar to that of the diphenyl-
methyl cation.10 Despite this stability, it is not apparent
that a low-energy pathway to 7 exists such that it should be
formed from cation 5b. In fact, Wiberg and co-workers’
computations did not predict that cation 7 would be
formed,6 and alcohol products such as 8 have not been
reported from previous solvolytic studies that gave 5b.
Nonetheless, early studies of 2-phenylcyclopropylcarbinyl
arenesulfonates and trifluoroacetates focused on rates of
reactions and not products,11 and Wiberg and co-workers’
product studies were conducted before high-performance
gas chromatographic (GC) columns were available.5 To
determine whether alcohol 8 is formed from 5b, we have
conducted a detailed product study of solvolytic reactions
that form the (2-phenylcyclopropyl)carbinyl cation. In
agreement with Wiberg and co-workers’ reports,5,6 the
For more than a decade, trans-2-phenylmethylcyclopro-
pane (1) and related aryl-substituted methylcyclopro-
panes have been employed in mechanistic studies of
hydroxylation reactions catalyzed by cytochrome P450
enzymes (P450s), methane monooxygenase enzymes and
related heme-iron and diiron enzymes.1,2 If radical 2 were
formed in an enzyme-catalyzed hydroxylation reaction,
then rapid ring opening to radical 3 could occur
(k ꢀ 3 ꢁ 1011 sꢂ1 at ambient temperature),3 leading ulti-
mately to rearranged alcohol 4 (Scheme 1). Hence the
detection of homoallyl alcohol 4 in early enzyme me-
chanistic studies was taken as evidence for a radical
intermediate. In quantitative applications of probe 1 and
related cyclopropane substrates in P450 studies, however,
the apparent lifetimes of radical intermediates varied by
several orders of magnitude, leading to the conclusion
that the probe substrates were rearranging at least in part
by a cationic pathway.2 Subsequent studies with probes
that differentiated between radical and cationic inter-
mediates demonstrated that cation-derived products in-
deed were formed in the hydroxylation reactions.2,4
After the discovery that cationic intermediates were
formed in P450 hydroxylations, our group interpreted the
formation of rearranged alcohol 4 in enzyme studies as
likely arising in part from a solvolysis-type reaction of
the protonated alcohol intermediate 5a formed by inser-
tion of the elements of OHþ into a C—H bond in 1. From
*Correspondence to: M. Newcomb, Department of Chemistry, Uni-
versity of Illinois at Chicago, 845 West Taylor Street, Chicago, Illinois
60607, USA. E-mail: men@uic.edu
Contact/grant sponsor: National Institutes of Health; Contact/grant
number: GM-48722.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 974–977

PRODUCTS OF SOLVOLYSIS REACTIONS FORMING (2-PHENYLCYCLOPROPYL)CARBINYL
975
CH3
CH2
A
4
Ph
Ph
Ph
10 + 11
1
2
3
OH
Ph
13
10x
9
4
8
B
12
Scheme 1
8
10
12
14
16
CH3
time (min)
Figure 1. Representative GC trace from a hydrolysis reac-
tion of 11-OTs. The products from the reaction are labeled.
Peak A is the internal standard, 1-phenyl-1-propanol, added
after the reaction, and peak B is an impurity in the sample of
internal standard
Ph
OH+ insertion
OH2
-H, -e-
1
Ph
Ph
Ph
Ph
5a
5b
clopropyl)methanol14 (9), trans-3-phenylcyclobutanol6
(10) and cis-3-phenylcyclobutanol6 (11) were prepared.
Reactions of these alcohols with pyridine–acetic anhy-
dride gave acetates for authentication of the products in
acetolysis reactions. We also prepared authentic samples
of (E)-1-phenyl-1,3-butadiene15 (12), 4-chloro-4-phenyl-
1-butene16,17 (14), trans-2-(phenylcyclopropyl)methyl
chloride16 (15) and (E)-4-chloro-1-phenyl-1-butene18
(16). (Z)-1-Phenyl-1,3-butadiene (13) was formed as a
minor component (ca 5%) in the preparation of 12. GC
analysis of the alcohols and of the acetates showed that
all products were resolved on wide-bore capillary
Carbowax 20M columns, with the exception of alcohols
10 and 11 and their respective acetates, which eluted
together in a broadened peak (Fig. 1). Diene 12 decom-
posed slowly on standing and partially decomposed in the
GC analysis, and we assume that diene 13 also decom-
posed partially in the GC analysis.
7
6
H2O
H2O
- H+
OH
- H+
OH
Ph
Ph
4
8
Scheme 2
reactions gave mainly products derived from benzylic
cation 6. Small amounts of 8 also were produced, but
the low yields indicate that this product cannot be expected
to form at detectable levels in enzyme-catalyzed reactions
that involve cations 5a or 5b. Hence the negative result,
i.e. the absence of 8 in enzyme reactions, does not permit
one to discount the possibility that carbocations were
formed in enzyme-catalyzed reactions as Shaik and co-
workers proposed.8,9
Scheme 3 shows the reactant alcohols and products
from solvolysis reactions, all of which are known com-
pounds. For GC and GC–mass spectral characterization,
authentic samples of 1-phenyl-3-buten-1-ol12 (4), 1-
phenyl-1-cyclopropylmethanol13 (8), (trans-2-phenylcy-
RESULTS AND DISCUSSION
Solvolysis reactions of the tosylates from cyclobutanols
10 and 11, prepared by the method of Wiberg and co-
workers,5,6 were conducted in water and in acetic
acid. The first-formed intermediates of these solvolyses
reactions are 2-phenylcyclobutyl cations that rearrange to
(2-phenylcyclopropyl)methyl cations.6 The acetolysis
reactions essentially reproduced the studies of Wiberg
and co-workers but with an emphasis on products rather
than rates.5,6 Table 1 contains the results. Consistent with
the 1969 report,5 alcohol 4 (in hydrolysis reactions) and
acetate 4-OAc (in acetolysis reactions), from benzylic
cation 6, were by far the major products formed. Alcohol
8 and its acetate also were detected with the aid of high-
performance GC. In the hydrolysis reactions, the ratios of
4 to 8 were 80:1 from 10-OTs and 50:1 from 11-OTs. In
the acetolysis reactions, the ratios of 4-OAc to 8-OAc
were 360:1 and 220:1 from the respective tosylates. The
OH
OH
OH
Ph
Ph
Ph
Ph
4
8
9
Ph
OH Ph
OH
10
11
12 (E-isomer)
13 (Z-isomer)
Cl
Cl
Cl
Ph
Ph
Ph
14
15
16
Scheme 3. Reactant alcohols and products
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 974–977

976
R. E. P. CHANDRASENA, D. AEBISHER AND M. NEWCOMB
Table 1. Relative % yields of products from solvolysis
OMs
reactionsa
Cl
Cl
Reactant Conditions
4
8
9
10 þ 11 12 13
10-OTs Hydrolysisb
11-OTs Hydrolysisb
10-OTs Acetolysisc
11-OTs Acetolysisc
9-OMs Hydrolysisd
96.3 1.2 0.4 1.2
86.5 1.7 0.3 10.8
73.0 0.2 0.5 0.6
67.3 0.3 0.2 7.0
0.8 0.1
0.6 0.1
24.7 1.1
23.9 1.3
26.6 3.3
Scheme 4
and, when the yields of all products were considered, the
average relative yield of alcohol 8 was 0.04%.
66.4 0.14 2.9
0
a Relative percentage yields are averages of two runs.
b Aqueous NaHCO3 solution at reflux for 30 min.
Reactions similar to those described above were con-
ducted with the mesylate 8-OMs to ensure that alcohol 8
was stable. The only products found were alcohol 8
(98.5% relative yield) and chloride 16 (1.5% relative
yield). Importantly, no diene 12 was detected from 8-
OMs; hence diene 12 does not arise from cation 7.
Traces of chloride products 14–16 ( < 1%) were de-
tected in the reactions of 9-OMs and a 1.5% yield of
chloride 16 was found in reactions of 8-OMs. The source
of nucleophilic chloride in these reactions was the
methanesulfonyl chloride used to prepare the mesylates,
and the formation of traces of these chlorides attests to
the mild conditions of the reactions and presumed rela-
tively long lifetimes of cationic transients. In the reaction
of 8-OMs, we speculate that chloride 16 arose in part
from nucleophilic addition to the cyclopropane ring of
the mesylate as shown in Scheme 4.
From the present results, the formation of alcohol 4
from cationic intermediate 5a produced in an enzyme-
catalyzed oxidation of probe 1 clearly is expected. This
fact provides sensibility for the results of mechanistic
probe studies of oxidizing enzymes. If the only route to
homoallylic alcohol products from 2-arylcyclopropyl-
methane probes was a radical pathway, then the quanti-
tative results from studies with several probes would be
confusing. For example, for the P450 2B1 enzyme that
has been studied in detail, the variations in the apparent
lifetimes of the radical transients would be three orders of
magnitude, where some of the radical trapping reactions
had effectively zero activation energy.2
Unfortunately, the low yields of 8 and 8-OAc found in
this work demonstrate that one cannot reach any mean-
ingful conclusions from the absence of 8 in an enzyme-
catalyzed oxidation of probe 1. The total product yields in
the cytochrome P450-catalyzed oxidations of probe 1 are
typically in the range of tens to hundreds of nanomoles,
and yields of rearranged product 4 are < 10 nmol. From
the present results, if all of alcohol 4 arose in a cationic
route, one might expect the formation of alcohol 8 in
yields < 100 pmol and possibly as low as 1 pmol. These
amounts are much smaller than the amounts of numerous
components in the product mixture that derive from the
enzyme mixtures and buffers.
c Acetic acid containing NaOAc at 120 ꢃC for 2–3 h; the acetate derivatives
of the alcohols were the products obtained.
d Solutions of the mesylate were prepared in THF at ꢂ20 ꢃC, and the
mixtures were added to water at ambient temperature.
very low yields of 8-OAc are noteworthy because the
reactions were conducted at 120 ꢃC. Cation 5b is a
requisite intermediate from the cyclobutyl tosylates,6
and the solvolytic conditions were much more severe
than enzyme-catalyzed reactions.
Reactions of the mesylate of alcohol 9 gave similar
results. Mesylate 9-OMs is unstable, but it and related 2-
aryl-substituted-cyclopropylmethanol mesylates can be
produced in situ at ca ꢂ20 ꢃC in THF and reduced with
LiBEt3H to give 2-aryl-1-methylcyclopropanes.19 There-
fore, we prepared THF solutions of 9-OMs at ꢂ20 ꢃC and
added the reaction mixtures to water at ambient tempera-
ture. The results are given in Table 1. Again, a trace of
alcohol 8 could be detected, but the average ratio of
alcohol 4 to alcohol 8 was about 500:1. Alcohol 9 was
found in 2.9% yield, but it is likely that most of this
material resulted from incomplete conversion of 9 to the
mesylate. Chloride 14 (0.5%) and traces of chlorides 15
and 16 (ca 0.05% each) were also detected in these
reactions.
Dienes 12 and 13 were found in all reactions listed in
Table 1, and diene 12 was the second most abundant
product in the acetolysis reactions and in the hydrolysis
reaction of 9-OMs. Dienes 12 and 13 apparently are
primary products formed by deprotonation of cation 6
in competition with nucleophilic capture. In a control
reaction, acetate 4-OAc gave diene 12 in < 1% yield
when subjected to the acetolysis conditions, demonstrat-
ing that 12 and 13 were not secondary products.
A series of reactions with 9-OMs was conducted that
should have been most favorable for formation of cation
7, which is about 7 kcal molꢂ1 (1 kcal ¼ 4.184 kJ) more
stable than cation 6.8 Mesylate 9-OMs was prepared at
ꢂ20 ꢃC as above, and a small amount of water was added
to the reaction mixture, which was then allowed to warm
to room temperature. In principle, cations formed in these
reactions will react mainly with THF to give oxonium
ions that can lose THF to regenerate the cations. The
major product formed under these conditions was diene
12, but small amounts of alcohols 4 and 8 were obtained.
For six reactions, the average ratio of 4 to 8 was 86:1,
Hence a negative result from enzyme studies with 1 or
related aryl-substituted probes, i.e. a failure to detect 8
and analogues in the products, provides no information
about the mechanism of the reactions. Although thermo-
dynamically favored,10 cation 7 simply is not formed
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 974–977

PRODUCTS OF SOLVOLYSIS REACTIONS FORMING (2-PHENYLCYCLOPROPYL)CARBINYL
977
appreciably in typical solvolysis reactions that produce
REFERENCES
the (2-phenylcyclopropyl)carbinyl cation. Moreover, the
formation of alcohol 4 from cation 6 confirmed in the
present work demonstrates that detection of 4 from
enzyme-catalyzed oxidations of 1 does not provide con-
clusive evidence for a radical intermediate. Simple aryl-
substituted methylcyclopropanes might be good radical
clocks when one knows the mechanism, but they are poor
mechanistic probes if one does not. More complex probes
that clearly differentiate between radicals and cations are
better mechanistic tools.4
1. Fu H, Newcomb M, Wong C-H. J. Am. Chem. Soc. 1991; 113:
5878–5880; Liu KE, Johnson CC, Newcomb M, Lippard SJ. J.
Am. Chem. Soc. 1993; 115: 939–947; Atkinson JK, Ingold KU.
Biochemistry 1993; 32: 9209–9214; Atkinson JK, Hollenberg PF,
Ingold KU, Johnson CC, Le Tadic M-H, Newcomb M, Putt DA.
Biochemistry 1994; 33: 10630–10637; Zaks A, Dodds DR. J. Am.
Chem. Soc. 1995; 117: 10419–10424; Toy PH, Newcomb M,
Hager LP. Chem. Res. Toxicol. 1998; 11: 816–823; Austin RN,
Chang HK, Zylstra GJ, Groves JT. J. Am. Chem. Soc. 2000; 122:
11747–11748; Toy PH, Newcomb M, Coon MJ, Vaz ADN. J. Am.
Chem. Soc. 1998; 120: 9718–9719; Newcomb M, Hollenberg PF,
Coon MJ. Arch. Biochem. Biophys. 2003; 409: 72–79; Jin Y,
Lipscomb JD. Biochim. Biophys. Acta Protein Struct. Mol.
Enzym. 2000; 1543: 47–59; Newcomb M, Aebisher D, Shen R,
Chandrasena REP, Hollenberg PF, Coon MJ. J. Am. Chem. Soc.
2003; 125: 6064–6065; Chandrasena REP, Vatsis KP, Coon MJ,
Hollenberg PF, Newcomb M. J. Am. Chem. Soc. 2004; 126: 115–
126.
EXPERIMENT
Solvolysis reactions of 10-OTs and 11-OTs. Solutions of
trans- or cis-3-phenylcyclobutanol tosylates5 (0.01 g,
0.037 mmol) in water (4 ml) with NaHCO3 (0.1 g) were
heated at reflux for 30 min. The cooled solutions were
extracted with diethyl ether (3 ꢁ 25 ml) and the ether
phase was washed with brine and dried over MgSO4.
Products were identified by GC and GC–mass spectral
comparisons with authentic compounds. GC response
factors were obtained with authentic compounds, and
yields in micromoles for each product were determined
against the internal standard 1-phenyl-1-propanol added
after the extractive work-up. Acetolysis reactions were
conducted in a similar manner in solutions of acetic acid
(4 ml) containing 0.1 g of NaOAc, which were heated at
120 ꢃC for 2–3 h. Yields in Table 1 are averages for two
determinations.
2. Newcomb M, Toy PH. Acc. Chem. Res. 2000; 33:
449–455.
3. Newcomb M, Manek MB. J. Am. Chem. Soc. 1990; 112:
9662–9663; Newcomb M, Johnson CC, Manek MB, Varick TR.
J. Am. Chem. Soc. 1992; 114: 10915–10921.
4. Newcomb M, Le Tadic-Biadatti MH, Chestney DL, Roberts ES,
Hollenberg PF. J. Am. Chem. Soc. 1995; 117: 12085–12091; Choi
SY, Eaton PE, Kopp DA, Lippard SJ, Newcomb M, Shen R. J. Am.
Chem. Soc. 1999; 121: 12198–12199; Newcomb M, Shen R, Choi
SY, Toy PH, Hollenberg PF, Vaz ADN, Coon MJ. J. Am. Chem.
Soc. 2000; 122: 2677–2686.
5. Wiberg KB, Nelson GL. Tetrahedron Lett. 1969: 4385–4388.
6. Wiberg KB, Shobe D, Nelson GL. J. Am. Chem. Soc. 1993; 115:
10645–10652.
7. Brazeau BJ, Austin RN, Tarr C, Groves JT, Lipscomb JD. J. Am.
Chem. Soc. 2001; 123: 11831–11837; Auclair K, Hu Z, Little DM,
Ortiz de Montellano PR, Groves JT. J. Am. Chem. Soc. 2002; 124:
6020–6027.
8. Kumar D, de Visser SP, Sharma PK, Cohen S, Shaik S. J. Am.
Chem. Soc. 2004; 126: 1907–1920.
9. Shaik S, Cohen S, de Visser SP, Sharma PK, Kumar D, Kozuch S,
Ogliaro F, Danovich D. Eur. J. Inorg. Chem. 2004; 207–226.
10. Kirmse W, Krzossa B, Steenken S. J. Am. Chem. Soc. 1996; 118:
7473–7477.
11. Sneen RA, Lewandowski KM, Taha IAI, Smith BR. J. Am. Chem.
Soc. 1961; 83: 4843–4848.
Hydrolysis reactions of 9-OMs. Alcohol 9 (0.053 g,
0.36 mmol) was converted to its mesylate by reaction
with MsOCl (35 ml, 0.36 mmol) in THF (5 ml) containing
Et3N (123 ml, 0.88 mmol) at ꢂ20 ꢃC. After 1 h, the reac-
tion mixtures were poured into water (60 ml) at room
temperature. The reaction mixture was worked up and
analyzed as described above.
12. Hart DJ, Kanai K. J. Org. Chem. 1982; 47: 1555–1560.
13. Friedrich EC, Taggart DB, Saleh MA. J. Org. Chem. 1977; 42:
1437–1443.
14. Nozaki H, Takaya H, Moriuti S, Noyori R. Tetrahedron 1968; 24:
3655–3669.
15. Grummitt O, Christoph FJ. J. Am. Chem. Soc. 1951; 73: 3479–
3484.
16. Shono T, Nishiguchi I. Tetrahedron 1974; 30: 2173–2181.
17. Hermans B, Hevesi L. J. Org. Chem. 1995; 60: 6141–6147.
18. Weber WP, Felix RA, Willard AK, Koenig KE. Tetrahedron Lett.
1971; 4710–4704.
19. Newcomb M, Choi SY, Toy PH. Can. J. Chem. 1999; 77: 1123–
1135.
Acknowledgment
This work was supported by a grant from the National
Institutes of Health (GM-48722).
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 974–977