Oxidation of Isopropylcyclopropane by Chromyl Chloride: Ring-Opened Products Support a Hydrogen Atom Abstraction Mechanism

Article abstract of DOI:10.1021/jo962348b

Chromyl chloride (CrO₂Cl₂) reacts with neat isopropylcyclopropane at 65 °C to give a dark precipitate along with at least 20 organic products. Both cyclopropyl products and ring-opened products are observed: 2-cyclopropyl-2-chloropropane (1,0.4% yield based on chromium), 2-cyclopropyl-2-propanol (2, 0.2%), 5-chloro-2-methyl-2-pentene (3, 0.3%), and 4-methyl-3-penten-1-ol (4, 0.5%) as well as other ring-opened products. Authentic samples of 1-4 were synthesized, and their GC and GC/ MS data were compared with the reaction mixture. Other organic products (5-10) were tentatively assigned by GC/MS on the basis of their m / z and fragmentation patterns. The ratio of (1 + 2) vs (3 + 4) increases by a factor of 2 when the initial concentration of CrO₂Cl₂ increases from 0.3 to 1.12 M. The reaction was also carried out in the gas phase, and essentially all the products from the liquid phase reaction were observed. The products are explained by a mechanism involving initial hydrogen atom abstraction from the substrate. The resulting dimethylcyclopropylcarbinyl radical can either be trapped by CrO₂Cl₂ (to form 1 and 2) or ring-open to give 4-methyl-3-pentenyl radical, which reacts with CrO₂Cl₂ to form 3 and 4 as well as further oxidized products. The oxidation of isopropylcyclopropane by MnO₄^- in pyridine was also examined. Acetone, an expected ring-opened product, was the only product observed by our analytical techniques. Me₂C¹⁸O is produced from ¹⁸O-labeled MnO₄^-. These results suggest that the reactions of CrO₂Cl₂ and MnO₄^- with isopropylcyclopropane proceed by hydrogen atom transfer to form organic radical intermediates.

Full text of DOI:10.1021/jo962348b

4248
J . Org. Chem. 1997, 62, 4248-4252
Oxid a tion of Isop r op ylcyclop r op a n e by Ch r om yl Ch lor id e:
Rin g-Op en ed P r od u cts Su p p or t a Hyd r ogen Atom Abstr a ction
Mech a n ism
Kun Wang and J ames M. Mayer*
Department of Chemistry, Box 351700, University of Washington, Seattle, Washington 98195-1700
Received December 16, 1996X
Chromyl chloride (CrO2Cl2) reacts with neat isopropylcyclopropane at 65 °C to give a dark precipitate
along with at least 20 organic products. Both cyclopropyl products and ring-opened products are
observed: 2-cyclopropyl-2-chloropropane (1, 0.4% yield based on chromium), 2-cyclopropyl-2-propanol
(2, 0.2%), 5-chloro-2-methyl-2-pentene (3, 0.3%), and 4-methyl-3-penten-1-ol (4, 0.5%) as well as
other ring-opened products. Authentic samples of 1-4 were synthesized, and their GC and GC/
MS data were compared with the reaction mixture. Other organic products (5-10) were tentatively
assigned by GC/MS on the basis of their m/ z and fragmentation patterns. The ratio of (1 + 2) vs
(3 + 4) increases by a factor of 2 when the initial concentration of CrO2Cl2 increases from 0.3 to
1.12 M. The reaction was also carried out in the gas phase, and essentially all the products from
the liquid phase reaction were observed. The products are explained by a mechanism involving
initial hydrogen atom abstraction from the substrate. The resulting dimethylcyclopropylcarbinyl
radical can either be trapped by CrO2Cl2 (to form 1 and 2) or ring-open to give 4-methyl-3-pentenyl
radical, which reacts with CrO2Cl2 to form 3 and 4 as well as further oxidized products. The
oxidation of isopropylcyclopropane by MnO4- in pyridine was also examined. Acetone, an expected
ring-opened product, was the only product observed by our analytical techniques. Me2C18O is
produced from 18O-labeled MnO4-. These results suggest that the reactions of CrO2Cl2 and MnO4
-
with isopropylcyclopropane proceed by hydrogen atom transfer to form organic radical intermediates.
In tr od u ction
MnO4- is similarly known to trap radicals at close to the
diffusion limit.5 In the oxidation of toluene by perman-
The development of selective hydrocarbon oxidation
reactions will be aided by mechanistic understanding.
Chromyl chloride (CrO2Cl2), permanganate (MnO4-), and
other metal-oxo species are well known to oxidize C-H
bonds.1 Our recent studies of alkane and arylalkane
ganate in organic solvents, the presence of O2 was found
-
to significantly affect the rate of disappearance of MnO4
suggesting that O2 is competing with MnO4 in the
trapping of benzyl radical.3
,
-
Radical probes have been widely used in elucidating
reaction mechanisms.6 The probe is a precursor to a
reactive intermediate that undergoes a characteristic
reaction, usually a structural rearrangement or isomer-
ization. Detection of rearranged products from the probe
provides evidence that the intermediate is formed. We
2
- 3
oxidation by CrO2Cl2 and MnO4 in nonpolar solvents
indicate that the rate-limiting step in all cases is transfer
of a hydrogen atom from the hydrocarbon to a metal oxo
-
group. In other words, CrO2Cl2 and MnO4 behave like
organic free radicals, abstracting H• from the substrate
RH, although they are both closed-shell species with no
unpaired electrons.
report here the reactions of isopropylcyclopropane with
-
CrO2Cl2 and MnO4
. Abstraction of the most weakly
The organic radical formed by hydrogen atom abstrac-
bound hydrogen atom7 would generate the dimethyl
cyclopropylcarbinyl radical,8 a “radical-clock” that un-
dergoes ring-opening with a rate constant of 2.3 × 108
s-1 at 65 °C.9 The observation of products derived from
the ring-opened species provides further evidence for
radical intermediates in these reactions.
tion, R•, rapidly undergoes further oxidation by CrO2Cl2
- 2-5
or MnO4
.
In the oxidation of cyclohexane by CrO2-
Cl2, the intermediate cyclohexyl radicals are trapped by
CrO2Cl2 (k = 3 × 109 M-1 s-1 at 80 °C) in competition
with trapping by CBrCl3, which gives bromocyclohexane.2b
X Abstract published in Advance ACS Abstracts, May 15, 1997.
(1) (a) Comprehensive Organic Synthesis (Oxidation); Trost, B. M.,
Ed.; Pergamon: New York, 1991; Vol. 7. (b) Oxidation in Organic
Chemistry; Wiberg, K. B., Ed.; Academic Press: New York, 1965; Part
A. (c) Oxidation in Organic Chemistry; Trahanovsky, W. S., Ed.;
Academic Press: New York, 1973; Part B. (d) Fatiadi, A. J . Synthesis
(Stuttgart) 1987, 85-127. (e) Arndt, D. Manganese Compounds as
Oxidizing Agents in Organic Chemistry; Open Court Publishing: La
Salle, IL, 1981. (f) Organic Syntheses by Oxidation with Metal
Compounds; Mijs, W. J ., de J onge, C. R. H. I., Eds.; Plenum: New
York, 1986.
(6) (a) Newcomb, M. Tetrahedron 1993, 46, 1151-1176. (b) Griller,
D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323. (c) Bowry, V. W.;
Lusztyk, J .; Ingold, K. U. J . Am. Chem. Soc. 1991, 113, 5687-5698.
(d) Horner, J . H.; Martinez, F. N.; Musa, O. M.; Newcomb, M.; Shahim,
H. J . Am. Chem. Soc. 1995, 117, 11124-11133. (e) Newcomb, M.; Le
Tadic-Boadatti, M.-H.; Chestney, D. L.; Roberts, E. S.; Hollenberg, P.
F. J . Am. Chem. Soc. 1995, 117, 12085-12091. (f) Newcomb, M.;
Chestney, D. L. J . Am. Chem. Soc. 1994, 116, 9753-9754. (g) Liu, K.
E.; J ohnson, C. C.; Newcomb, M.; Lippard, S. J . J . Am. Chem. Soc.
1993, 115, 939-947. (h) Bowry, V. W.; Ingold, K. U. J . Am. Chem.
Soc. 1991, 113, 5699-5707. (i) Bullock, R. M.; Samsel, E. G. J . Am.
Chem. Soc. 1990, 112, 6886-6898.
(7) Although the C-H bond strengths in isopropylcyclopropane are
not known, presumably the weakest bond is the tertiary one in the
isopropyl group. Cf. H-cyclopropyl, 106 kcal/mol; H-isopropyl, 98 kcal/
mol. Colussi, A. J . In Chemical Kinetics of Small Organic Radicals;
Alfassi, Z. B., Ed.; CRC Press: Boca Raton, FL, 1988; pp 25-43.
(8) Alternative names: dimethylcyclopropylmethyl radical, 2-cyclo-
propyl-2-propyl radical.
(2) (a) Cook, G. K.; Mayer, J . M. J . Am. Chem. Soc. 1995, 117, 7139-
7156. (b) Cook, G. K.; Mayer, J . M. J . Am. Chem. Soc. 1994, 116, 1855-
1868. (c) Correction, ibid. 1994, 116, 8859.
(3) (a) Gardner, K. A.; Mayer, J . M. Science 1995, 269, 1849-1851.
(b) Gardner, K. A.; Kuehnert, L. L.; Mayer, J . M. Inorg. Chem., in press.
(c) Gardner, K. A. Ph.D. Thesis, University of Washington, 1996.
(4) Aqueous chromium(VI) is also known to trap alkyl radicals at
close to diffusion-limited rates: Al-Sheikhly, M.; McLaughlin, W. L.
Radiat. Phys. Chem. 1991, 38, 203-211.
(5) Steenken, S.; Neta, P. J . Am. Chem. Soc. 1982, 104, 1244-1248.
(9) Calculated from the Arrhenius parameters in ref 6c.
S0022-3263(96)02348-1 CCC: $14.00 © 1997 American Chemical Society

Oxidation of Isopropylcyclopropane by CrO2Cl2
J . Org. Chem., Vol. 62, No. 13, 1997 4249
Rea ction of Cr O2Cl2 w ith Isop r op ylcyclop r op a n e. In
the glovebox under a red “safe” light, 10 µL of CrO2Cl2 (0.12
mmol) was syringed into a 70 mL Pyrex bomb that was then
sealed with a Teflon stopcock and wrapped in foil. Isopropyl-
cyclopropane (0.4 mL; ca. 7.5 M) was later vacuum transferred
into this bomb at -78 °C. (Caution: Direct addition of
isopropylcyclopropane to CrO2Cl2 at ambient temperatures
results in violent reaction and should be avoided.) The bomb
was placed in a 65 °C water bath for ca. 30 min to ensure
complete reaction of CrO2Cl2. The initial orange-red color of
CrO2Cl2 turned brown, and a precipitate formed (the EÄ tard
complex2). The organic volatiles were then vacuum transferred
out and analyzed by GC and low-temperature GC/MS. The
EÄ tard complex was hydrolyzed by treatment with 1.0 mL of 1
M aqueous solution of Na2S2O3 and passage through a column
of alumina (to remove chromium), and the aqueous solution
was also analyzed by GC and GC/MS.
Oth er Rea ction s of Cr O2Cl2. CrO2Cl2 (10 µL, 0.13 mmol)
was vacuum transferred onto 50 µL of 2-cyclop r op yl-2-
p r op a n ol (2, ca. 0.43 mmol) at -78 °C. The mixture was
allowed to warm to room temperature and react overnight.
Organic volatiles were vacuum transferred out, and the EÄ tard
complex was destroyed by treatment with an aqueous solution
of Na2S2O3. The aqueous solution was then extracted with
Et2O. Both the organic volatiles and the ether layer were
analyzed by GC. Similar reaction and workup conditions were
used for the reactions of CrO2Cl2 with m eth yl cyclop r op yl
k eton e [2 M CrO2Cl2 in neat substrate (ca. 8.4 M)] and
2-cyclop r op ylp r op en e [0.9 M CrO2Cl2 in neat substrate (ca.
8.4 M)].
Ga s-P h a se Rea ction s of Cr O2Cl2. CrO2Cl2 (5.0 µL, 0.065
mmol) was placed in a 40-mL Pyrex bomb in the glovebox, and
isopropylcyclopropane (20.0 µL, ca. 0.18 mmol) was placed in
a 100 mL bomb. Both reactants were completely in the gas
phase at room temperature. The bombs were then attached
to the vacuum line through a short-path T-joint and freeze-
pump-thawed. The contents in the bombs were allowed to
diffuse to each other at room temperature. Instant reaction
occurred with consumption of the orange CrO2Cl2 vapor and
formation of a brown precipitate. After about 1 h, the bombs
were detached from the vacuum line and the EÄ tard complex
destroyed by treatment with an aqueous solution of Na2S2O3.
The aqueous solution was then extracted with Et2O and the
ether layer collected and analyzed by GC. Reactions with
2-cyclopropylpropene and 2-cyclopropyl-2-propanol (2) were
carried out similarly.
Rea ction of [P P N]Mn O4 w ith Isop r op ylcyclop r op a n e.
In a glovebox, [PPN]MnO4 (26 mg, 0.04 mmol) was dissolved
in 0.6 mL of a 3.1 M solution of isopropylcyclopropane in
pyridine in a Pyrex bomb wrapped in foil. A control reaction
of [PPN]MnO4 in pyridine without substrate was also pre-
pared. The bombs were placed in a 65 °C water bath. After
ca. 4 h, the solution turned from dark purple to rusty brown,
while there was no observable color change in the control. The
bombs were then taken out of the water bath, and the volatiles
were vacuum transferred out. The residue was dissolved in
an aqueous Na2S2O3/HCl solution, with reduction of the solid
MnO2, and extracted with Et2O. The volatiles, the ether layer,
and the aqueous layer were all analyzed by GC and GC/MS.
For the 18O-labeling experiment, a solution of KMn18O4, 16 mM
in 4:1 (v/v) pyridine/isopropylcyclopropane, was heated at 65
°C for 2.5 h, and product analysis was carried out as above.
Exp er im en ta l Section
All reaction mixtures were prepared under an N2 atmo-
sphere either in a glovebox or by using vacuum-line techniques
with Teflon-sealed Pyrex reaction vessels, and all were shielded
from light. CrO2Cl2 (99.99%, Aldrich) was sealed in a grease-
less, light-free glass vessel and was vacuum transferred prior
to use in a vacuum-line greased with KRYTOX fluorinated
grease (DuPont). Caution: CrO2Cl2 is a corrosive and carci-
nogenic volatile liquid that should be handled with extreme
caution. Isopropylcyclopropane (99%, Wiley Organics) was
dried over Na, vacuum transferred into a sealed vessel, and
stored in a freezer in the glovebox. GC/MS analysis of the
isopropylcyclopropane showed no unsaturated olefinic species.
[PPN]MnO4 (PPN ) (Ph3PNPPh3)+) was prepared according
to a literature method10 and recrystallized from CH2Cl2/Et2O.
KMn18O4 (ca 10% 18O enriched) was prepared by exchange of
KMnO4 with H218O.3c Pyridine was purified by standard
procedures,11 stored over CaH2, and vacuum transferred prior
to use. 4-Methyl-3-penten-1-ol (4, 97%, Aldrich) was used as
received. Methyl cyclopropyl ketone (99%; Aldrich) was dried
over Na2SO4 for use in the reaction with CrO2Cl2. Product
analyses were performed on an HP5890 series GC equipped
with a 30-m cross-linked PH ME silicone column and an FID
detector; GC/MS were obtained on a Kratos mass spectrometer
equipped with an HP5890 GC and a cooling unit for low-
temperature analysis.
2-Cyclopropylpropene was prepared following a reported
procedure12 and was dried over sodium. NMR and IR data
are consistent with literature values.12 MS (70 eV) m/ z: 83
(4), 82 (58), 67 (100), 41 (90). 2-Cyclopropyl-2-chloropropane
(1) was prepared, following ref 13, by reacting 2-cyclopropyl-
propene with HCl in CH2Cl2. MS (70 eV) m/ z: 121 (18), 119
(65), 103 (12), 83 (39), 79 (28), 78 (12), 77 (100), 76 (32), 69
(15), 67 (20), 56 (34), 43 (36), 41 (64). 2-Cyclopropyl-2-propanol
(2) was synthesized from methylcyclopropyl ketone and MeMg-
Br by a literature method14 and dried over 4 Å sieves. MS
(70 eV) m/ z: 85 (67), 83 (14), 72 (61), 67 (13), 59 (49), 57 (45),
55 (18), 43 (100), 41 (60); the fragmentation pattern matches
that reported.15
The reported procedure16 for 5-chloro-2-methyl-2-pentene
(3), from 4-methyl-3-penten-ol and SOCl2 in pyridine, gives a
very low yield. Therefore, 3 was prepared from 2 plus HCl by
a modification of the reported procedure.17 Compound 2 (3.0
mL, 27 mmol assuming 100% purity) was dissolved in 20 mL
of dry CH2Cl2 under N2 in a 50 mL round-bottom flask and
cooled to -10 °C. HCl gas (99%, Aldrich) was passed through
the solution for 15 s, and the solution was allowed to slowly
warm to room temperature while stirring was maintained for
ca. 30 min. CH2Cl2 was removed in vacuo, and the residue
was dried over K2CO3. A slightly yellow liquid (∼2.5 mL) was
obtained after vacuum transfer. In contrast to the prior
report,17 GC and GC/MS indicated the liquid was a mixture
of 3 and 1 (ca. 3:1 by GC), suggesting a yield of 3 of 60%.
Separation of the two by distillation was not successful
(presumably due to their similar boiling points: 3, 128 °C;16
1, 105 °C13), so the mixture was used without purification. MS
(70 eV) m/ z: 120 (72), 118 (86), 69 (100), 103 (11), 83 (67), 79
(47), 77 (66), 75 (41), 69 (100), 65 (44), 56 (64), 55 (72), 53 (64),
51 (49), 43 (45), 42 (47).
(10) Martinsen, A.; Songstad, J . Acta Chem. Scand. A 1977, 31, 645-
650.
(11) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon Press: New York, 1992; pp 267-268.
(12) Farneth, W. E.; Thomsen, M. W. J . Am. Chem. Soc. 1983, 105,
1843-1848.
Resu lts a n d Discu ssion
Rea ction of Cr O2Cl2 w ith Isop r op ylcyclop r op a n e.
The reaction of CrO2Cl2 with isopropylcyclopropane was
carried out in the neat substrate to avoid involvement of
solvent in the reactions. Direct addition of neat substrate
to CrO2Cl2 at ambient temperatures results in a violent
reaction accompanied by a large release of heat. There-
fore, reaction solutions were prepared by vacuum transfer
of the hydrocarbon to a -78 °C sample of CrO2Cl2 in a
Pyrex vessel with a Teflon stopcock. Reactions were
(13) Shellhamer, D. F.; McKee, D. B.; Leach, C. T. J . Org. Chem.
1976, 41, 1972-1976.
(14) J ulia, M.; J ulia, S.; Gue´gan, R. Bull. Soc. Chim. Fr. 1960, 1072-
1079.
(15) Hamuise, J .; Puttemans, J . P.; Smolders, R. R. Tetrahedron
1969, 25, 1757-1769.
(16) Grob, C. A.; Waldner, A. Helv. Chim. Acta 1979, 62, 1854-
1865.
(17) (a) J ulia, S.; J ulia, M.; Brasseur, L. Bull. Soc. Chim. Fr. 1962,
1634-1638. (b) Corey, E. J .; Hartmann, R.; Vatakencherry, P. A. J .
Am. Chem. Soc. 1962, 84, 2611-2614.

4250 J . Org. Chem., Vol. 62, No. 13, 1997
Wang and Mayer
The observed products are readily explained by a
mechanism involving initial hydrogen atom abstraction
from the tertiary position (Scheme 1). The resulting
c-C3H5C˙ Me2 radical either is trapped by CrO2Cl2 or ring-
opens to form the 4-methyl-3-pentenyl radical, and
products are observed from both pathways. Reaction of
c-C3H5C˙ Me2 with CrO2Cl2 should be similar to the
reaction of the tert-butyl radical, which occurs by (i)
chlorine atom abstraction to give t-butyl chloride, (ii)
C-O bond formation yielding tert-butoxide ligands that
are observed as the alcohol on hydrolysis, and (iii) a
second hydrogen atom transfer to give isobutylene.2a In
the isopropylcyclopropane reaction, 2-cyclopropyl-2-chlo-
ropropane (1) is the product of Cl abstraction and
2-cyclopropyl-2-propanol (2) is the result of C-O bond
formation.19 Similar trapping of the 4-methyl-3-pentenyl
radical gives the observed 5-chloro-2-methyl-2-pentene
(3) and 4-methyl-3-penten-1-ol (4).
F igu r e 1. Typical chromatograph of the reaction mixture from
isopropylcyclopropane and CrO2Cl2. GC oven temperatures: 35
°C (12 min) f 10 °C/min f 200 °C (20 min).
The reaction is much more complex than these four
products because of the rapid and nonselective oxidation
of alkenes by CrO2Cl2.2,20,21 Isobutylene, for instance,
reacts with CrO2Cl2 to give a variety of products including
isobutyraldehyde, 2-chloroisobutyraldehyde, chlorohy-
drins, and acetone (by CdC bond cleavage).2a So 3 and
4 are predominantly oxidized further, as are the products
resulting from hydrogen atom abstraction from the
intermediate radicals 2-cyclopropylpropene (11) and
4-methyl-1,3-pentadiene. Reasonable pathways to the
observed products are given in Scheme 2, but this is by
no means an exhaustive description of products or
pathways. Oxidation of the double bond in 3 gives
chlorohydrins, chloro ketone, acetone, aldehydes, and
likely carboxylic acids and CO2. Independent oxidation
of a mixture of 3 and 1 (0.3 mL; ∼3:1 by GC) with limited
CrO2Cl2 (40 µL, roughly 0.2 equiv) resulted in the
formation of dichloro ketone 5 (0.4%), acetone, chloro-
aldehyde 6 (0.1%), and chlorohydrin 7 (4%). [These
estimated yields, and those given below, are per mole of
CrO2Cl2 and assume that all species have the same
response factor in the GC. The other products, as noted
above, were not observed.] CrO2Cl2 reacts preferentially
with 3 over 1, as expected, because the ratio of 3 to 1
decreased to 3:2 after reaction as revealed by GC. These
data support 3 being on the route to other ring-opened
products.
protected from light and were prepared using a red “safe”
light. Reaction mixtures were heated to 65 °C to ensure
complete consumption of CrO2Cl2. Reactions run in the
gas phase indicate that a similar mix of products is
formed at ambient temperatures (see below). The orange-
red color of CrO2Cl2 is replaced by a colorless solution
and a brown precipitate (the EÄ tard complex). The organic
products are present in solution and bound to chromium
in the precipitate, as is characteristic of these reactions.2
GC and GC/MS analysis of the solution and an aqueous
workup of the precipitate showed a complex spectrum of
at least 20 products (Figure 1). It should be emphasized
that some of the reaction products, such as carboxylic
acids, are not observed by this analytical procedure.2
A subset of the observed products have been identified
by GC/MS on the basis of their m/ z and fragmentation
patterns (Table 1). Some of the assignments are tenta-
tive; for instance, stereoisomers cannot typically be
distinguished on the basis of mass spectral data. Four
of the compounds have been identified by comparison of
their GC and GC/MS data with authentic samples:
2-cyclopropyl-2-chloropropane (1), 2-cyclopropyl-2-pro-
panol (2), 5-chloro-2-methyl-2-pentene (3), and 4-methyl-
3-penten-1-ol (4). The yields of these four products, per
mole of CrO2Cl2, are given in Table 1. For the other
products, approximate yields are given based on the
yields for 1-4 and assuming that all of the compounds
have roughly the same response factor in the GC (FID
detection). While this assumption is not very accurate,
the calculation provides a rough measure of the product
distribution. The sum of these yields is not a measure
of the reaction mass balance as multiple CrO2Cl2 mol-
ecules are consumed for many of the products. With
further assumptions about the number of CrO2Cl2 mol-
ecules consumed per oxidized product, the compounds
listed in Table 1 represent roughly a quarter of the CrO2-
Cl2 oxidizing equivalents expended in the reaction.18 The
remaining oxidative equivalents are likely expended in
the formation of carboxylate products, which have proven
very difficult to liberate from the chromium(III) products
in this system.2
The reaction of CrO2Cl2 and 2-cyclopropyl-2-propanol
(2) has been examined both in the neat liquid substrate
and in the gas phase. GC analysis indicated quite similar
product distributions from the two reaction conditions:
3 (77%), 6 (10%), 7 (1%), 5 (0.3%), 1 (0.2%), and 4 (trace;
yields are from the solution reaction). The formation of
these products likely occurs by initial dehydration of the
alcohol, presumably via chromium-alkoxide formation,2a
and subsequent reactions of 2-cyclopropylpropene (11).
It should be noted that alcohol products such as 2 are
present only as alkoxide ligands to reduced chromium
during the alkane oxidation and are liberated as the
alcohol only on protic workup. The reaction of 2 and
(19) The alcohol is not formed by autoxidation, as O2 is rigorously
excluded from these reactions and there is no formation of methylcy-
clopropyl ketone, the facile fragmentation product from the tertiary
alkoxy radical.
(18) A final chromium oxidation state of 3.5 is assumed, by analogy
with the reaction of Me3CH.2a This is likely a lower estimate because
FID response factors are smaller for lower carbon-number products.
The issue of mass balances in CrO2Cl2 reactions is discussed in more
detail in ref 2a,b.
(20) (a) Sharpless, K. B.; Teranishi, A. Y.; Ba¨ckvall, J .-E. J . Am.
Chem. Soc. 1977, 99, 3120-3128. (b) Miyaura, N.; Kochi, J . K. Ibid.
1983, 105, 2368-2378.
(21) (a) Wiberg, K. B. In ref 1b, pp 69-184. (b) Freeman, F. In ref
1f, pp 41-118.

Oxidation of Isopropylcyclopropane by CrO2Cl2
J . Org. Chem., Vol. 62, No. 13, 1997 4251
Ta ble 1. P r od u cts fr om th e Rea ction of Cr O2Cl2 w ith Isop r op ylcyclop r op a n ea
a
b
Reaction conditions: 3.1 M of CrO2Cl2 in neat isopropylcyclopropane at 65 °C for 30 min. From GC/MS. c Fragments most useful in
d
structure elucidation, not the strongest MS peaks. Absolute yields, from GC data, per mole of CrO2Cl2 using authentic samples as
calibration. e Approximate relative yields, from GC data, assuming equal response factors for all products and that all products were
observed (see text). f Identified by spiking the reaction mixture with an authentic sample; MS data is for the authentic sample.
Sch em e 1
Sch em e 2
CrO2Cl2 is an approximation to the behavior of these
alkoxides during alkane oxidation.
2-Cyclopropylpropene (11) is not observed in the oxida-
tion of isopropylcyclopropane, but as noted above, it is a
likely intermediate, both from dehydration of the alcohol
2 and by hydrogen atom abstraction from c-C3H5C˙ Me2.
Reaction of CrO2Cl2 (0.9 M) in neat 11 yields 3 (6%), 6
(2%), methylcyclopropyl ketone (0.3%), and trace amounts
of 7 (0.05%), 1 (0.07%), and 5 (0.03%). Methylcyclopropyl
ketone, the product derived from CdC bond cleavage in
11, is formed in low yield apparently because it reacts

4252 J . Org. Chem., Vol. 62, No. 13, 1997
Wang and Mayer
further with CrO2Cl2, as confirmed by an independent
reaction. In sum, 2 and 11 are precursors to some of the
ring-opened products observed on oxidation of isopropyl-
cyclopropane. But the product distributionssfor in-
stance, the lack of formation of 8 or 9sindicate that these
are not the major pathway to ring-opened products. The
data are most consistent with formation of ring-opened
products via the dimethylcyclopropylcarbinyl radical
“clock” (Scheme 1).
Rea ction of Mn O4- w ith Isop r op ylcyclop r op a n e.
Reactions of permanganate with isopropylcyclopropane
cannot be done in the neat substrate as the CrO2Cl2
oxidations were, due to the insolubility of the ionic
reagent in the hydrocarbon. [PPN]MnO410 has been used
as the permanganate salt because of its relative stability
and solubility. According to control experiments, pyri-
dine is a good solvent for [PPN]MnO4 since there is
limited decay of permanganate: about 10% decomposi-
tion was observed by UV/vis spectroscopy after 5 h at 65
°C. Reactions were performed in mixed pyridine/isopro-
pylcyclopropane solvent (2/1 by volume, 3.1 M in the
substrate). Reactions were run at 67 mM [PPN]MnO4,
close to the saturation point. Other solvents such as
acetonitrile and o-dichlorobenzene also provide solubility
but are too reactive with permanganate. The reaction
A prediction of Scheme 1 is that the partitioning
between cyclopropyl products and ring-opened products
depends on the relative rates of trapping vs ring-opening
of c-C3H5C˙ Me2. Ring-opening occurs with a rate constant
of 2.3 × 108 s-1 at 65 °C.9 Reaction of CrO2Cl2 with the
tertiary radical likely occurs at ca. 109 M-1 s-1, assumed
to be a little slower than the reaction with cyclohexyl
radical (3 × 109 M-1 s-1 2b) for steric reasons. With an
initial CrO2Cl2 concentration of 0.15 M, it is predicted
that the trapping of the radical by CrO2Cl2 should be
competitive with ring opening. This is consistent with
the observation of significant products from both path-
ways. The prediction cannot be quantitatively tested,
however, because not all of the products were detected
or identified and because of the complexity of the kinetic
scheme, with further oxidation of many of the products.
This analysis implies that increasing the initial con-
centration of CrO2Cl2 should give a corresponding in-
crease in the trapped, cyclopropyl-containing products.
Using the four products for which standards are avail-
able, it is found that the ratio of (1 + 2) vs (3 + 4)
increases from 0.6 to 1.0 when [CrO2Cl2]initial increases
from 0.3 to 1.12 M in neat isopropylcyclopropane. This
is qualitatively consistent with the proposed mechanism
(Scheme 1) that a second-order trapping reaction is
competing with first-order ring-opening. As above, quan-
titative agreement should not be expected.
The reaction of CrO2Cl2 and isopropylcyclopropane was
also run in the gas phase by allowing the vapor of CrO2-
Cl2 and isopropylcyclopropane to diffuse to each other in
a vacuum manifold. Although the relative yields vary,
the same products were observed in the gas phase as
found in solution. This strongly argues against the
involvement of charged species in the reaction, as the
formation of ions from neutral species in the absence of
a dielectric medium is a high energy process. Therefore,
the ring-opened products are not formed via a carbocation
rearrangement, for instance in c-C3H5C+Me2. Similarly,
charged species are not likely to be involved in the
oxidation of 2 by CrO2Cl2 on the basis of the similarity
of the gas- and solution-phase reactions.
-
of MnO4 with isopropylcyclopropane does not occur as
readily as that of CrO2Cl2, requiring heating at 65 °C for
4 h.
Product analyses are problematic for this reaction
because of the small concentrations involved and because
the ring-opened products, if any, are likely to be pre-
dominantly carboxylic acids that have been difficult to
detect.22 Quenching a reaction mixture of 67 mM MnO4
-
and 3.1 M isopropylcyclopropane with aqueous NaHSO3/
HCl and analysis by GC/MS revealed only the formation
of acetone. The possibility that acetone was introduced
inadvertently was ruled out after carefully scrutinizing
the control sample. In addition, reaction of partially
isotopically enriched Mn18O4 (ca. 10% 18O enriched)3c
-
gave Me2C18O (ca. 10%). In a control experiment, no
exchange was observed between Me2CO and Mn18O4
-
under these conditions. Neither [18O]methyl cyclopropyl
ketone nor [18O]-2-cyclopropyl-2-propanol were detected.23
Acetone is one of the expected products if ring-opening
occurs, as permanganate is known to cleave CdC double
bonds to the corresponding aldehydes and ketones.1
Aldehydes and primary or secondary alcohols would most
likely be oxidized further by permanganate under these
conditions.
Con clu sion
The reaction of chromyl chloride with isopropylcyclo-
propane gives a large number of products, including both
compounds that retain the cyclopropane ring and ring-
opened compounds. The nature and yields of the prod-
ucts are consistent with formation of the dimethylcyclo-
propylcarbinyl radical clock, which is trapped by CrO2Cl2
in competition with ring opening. The only material
observed from permanganate oxidation of this substrate
is acetone, which is a logical product of vigorous oxidation
of the ring-opened radical. These observations support
the contention that alkane oxidation by chromyl chloride
and permanganate proceeds by initial hydrogen atom
abstraction to give an organic radical intermediate.
(22) The carboxylate products are bound to the MnO2 colloid. They
are liberated on acidic reduction. Direct detection by GC has been
unsuccessful, as were attempts to esterify carboxylic acid products by
diazomethane treatment,3c possibly because of manganese ion interfer-
ence of CH2N2 reactions.
Ack n ow led gm en t. We are grateful to the National
Institutes of Health for financial support of this work.
We thank Bob Morley and J im Roe for invaluable
technical support, Andrea Collier, Teri Blubaugh, Darin
DuMez, and Tom Crevier for experimental assistance,
and Dr. K. A. Gardner for helpful discussions.
(23) Addition of 2-cyclopropyl-2-propanol (2) to the reaction mixture
showed that this compound is stable to the reaction conditions and
would have been detected if formed. (Tertiary alcohols are known to
be stable to high valent manganese complexes: Mu¨ller, P. In The
Chemistry of the Functional Groups: Ethers, Crown Ethers and
Hydroxyl Groups; Patai, S., Ed.: Wiley: New York, 1980; Vol. 1, p
-
469). However, trapping of the 2-cyclopropyl-2-propyl radical by MnO4
would give the manganese(VI) alkoxide, [O3Mn{OCMe2(C3H5)}-],
which is less likely to be stable.
J O962348B