BrF3, an underutilized reagent in organic chemistry: A novel C-C-N to C-N-C rearrangement

Article abstract of DOI:10.1021/jo0013094

Little is known about bromine trifluoride in organic chemistry. Under the right conditions, it can be a useful tool and generate a new and unprecedented chemistry. Thus, when reacted with oxime methyl ethers of α-ketoesters, BrF₃ was able to convert the oxime group into a CF₂ group and through a new type of rearrangement cause a shift of the carboxylate group to the nitrogen atom. The novel structure of the α,α-difluorocarbamate was also proven by ¹⁵N NMR as demonstrated for compounds 3, 8, 9, 12, 15, and 18. Another novel "double rearrangement" was observed during the formation of 19. Dynamic ¹⁹F NMR experiments indicate a high nitrogen inversion-rotation (NIR) barrier for these novel carbamates of about 12.5 kcal/mol.

Full text of DOI:10.1021/jo0013094

496
J . Org. Chem. 2001, 66, 496-500
Br F 3, a n Un d er u tilized Rea gen t in Or ga n ic Ch em istr y: A Novel
C-C-N to C-N-C Rea r r a n gem en t
Shlomo Rozen* and Iris Ben-David
School of Chemistry, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel-Aviv University,
Tel-Aviv 69978, Israel
rozens@post.tau.ac.il
Received August 30, 2000
Little is known about bromine trifluoride in organic chemistry. Under the right conditions, it can
be a useful tool and generate a new and unprecedented chemistry. Thus, when reacted with oxime
methyl ethers of R-ketoesters, BrF3 was able to convert the oxime group into a CF2 group and
through a new type of rearrangement cause a shift of the carboxylate group to the nitrogen atom.
The novel structure of the R,R-difluorocarbamate was also proven by 15N NMR as demonstrated
for compounds 3, 8, 9, 12, 15, and 18. Another novel “double rearrangement” was observed during
the formation of 19. Dynamic 19F NMR experiments indicate a high nitrogen inversion-rotation
(NIR) barrier for these novel carbamates of about 12.5 kcal/mol.
Bromine trifluoride has been extensively used for the
there is a chance of developing a new reaction pathway
toward the preparation of R,R-difluorocarboxylic acid
derivatives which are important as enzyme inhibitors.12
Additional support for the notion of using oxime methyl
ethers R to a carboxylic ester was derived from the fact
that BrF3 reacts best when there are at least two
electron-donor groups close to each other in the target
molecule.13 Such donors serve as complexing anchors to
the electron-depleted bromine atom bringing the reagent
close to the reacting site and preventing it from rapid,
undiscriminating radical attacks on most CH bonds,
attacks for which BrF3 is feared and known for.14
There are several literature preparations for pyruvic
acid derivatives starting with acetylene compounds,15
â-keto sulfur derivatives,16 dimethyloxalate,17 and more.
In the present study, we have prepared such substrates
also by direct R-hydroxylation using the HOF‚CH3CN
complex, a novel method we recently published in memory
of the late Prof. Barton,18 followed by oxidation with
CrO3.
To avoid volatility problems, we have prepared the
1,1,1-trichloroethyl pyruvate (1) and reacted it with
methoxylamine forming the corresponding oxime methyl
ether 2 (Scheme 1). This material was subject to a
reaction with 1.5 mol/equiv of BrF3 forming a product in
60% yield. To our surprise, however, this product was
not the anticipated R,R-difluoroester, but proved eventu-
ally to be N-(1,1-difluoroethyl)-N-methoxy-2,2,2-trichlo-
roethyl carbamate (3). The product’s most conspicuous
spectral feature was the 19F NMR, which showed a broad
signal at -77.1 ppm, Wh/2 ) 85 Hz. This chemical shift
is in sharp contrast to the R,R-difluoroester moiety found
synthesis of several modern anesthetics such as desflu-
rane,1 sevoflurane,2 and the promising 1,2-bis(fluo-
romethoxy)-1,1,3,3,3-pentafluoropropane (BFPP).3 Still,
it is seldom found in the organic chemistry literature,
and the existing research concentrates on some oxidative4
and tertiary5 fluorinations as well as on efficient elec-
trophilic bromination of very deactivated aromatic rings.6
It also resembles IF7 in its ability to transform carbonyls,
via the corresponding hydrazones, to the CF2 group, but
it is much more potent than iodine monofluoride, and we
were able for the first time to convert carbonyls to the
CF2 group also through the corresponding azines, DNPs,8
thioesters,9 dithioesters,10 and more. Recently, we have
used bromine trifluoride also for constructing various
aliphatic and aromatic trifluoromethyl ethers from alco-
hols and their respective xanthates.11
Unlike IF, bromine trifluoride can also react with
oxime methyl ethers made with methoxylamine which
in general are more soluble than most other types of
hydrazones.8 Methoxylamine is also somewhat more
selective than hydrazines in differentiating between two
carbonyls in molecules such as R-ketoesters and react
preferentially with the ketonic one. We thought that
* To whom correspondence should be addressed. Fax: +972 3
6409293.
(1) Rozov, L. A.; Huang, C.; Halpern, D. F.; Vernice, G. G. U.S.
Patent 5,283,372, 1994.
(2) Halpern, D. F.; Robin, M. L. U.S. Patent 4,996,371, 1991.
(3) Ruzicka, J . A.; Hrdy, J . B.; Baker, M. T. J . Fluorine Chem. 1995,
75, 191.
(4) Soelch, R. R.; Mauer, G. W.; Lemal, D. M. J . Org. Chem. 1985,
50, 5845.
(5) Boguslavskaya, L. S.; Kartashov, A. V.; Chuvatkin, N. N. Russ.
Chem. Rev. 1990, 59, 501 (translated from Uspekhi Khimii 1990, 59,
865.)
(6) Rozen, S.; Lerman, O. J . Org. Chem. 1993, 58, 239.
(7) Rozen, S.; Zamir, D. J . Org. Chem. 1991, 56, 4695.
(8) Rozen, S.; Mishani, E.; Bar-Haim, A. J . Org. Chem. 1994, 59,
2918.
(12) See, for example: Tsukamoto, T.; Haile, W. H.; McGuire, J . J .;
Coward, J . K. J . Med. Chem. 1996, 39, 2536.
(13) Unpublished results.
(14) Bromine trifluoride can react violently with hexane or other
similar molecules via radical reactions.
(9) Rozen, S.; Mishani, E. J . Chem. Soc., Chem. Commun. 1993,
1761.
(10) Rozen, S.; Mishani, E. J . Chem. Soc., Chem. Commun. 1994,
2081.
(11) Ben-David, I.; Rechavi, D.; Mishani, E.; Rozen, S. J . Fluorine
Chem. 1999, 97, 75.
(15) Muller, P.; Godoy, J . Tetrahedron Lett. 1982, 23, 3661.
(16) Fortes, C. C.; Souto, C. R. O.; Okino, E. A. Synth. Commun.
1991. 21, 2045.
(17) Itoh, K.; Kori, M.; Inada, Y.; Nishikawa, K.; Kawamatsu, Y.;
Sugihara, H. Chem. Pharm. Bull. 1986, 34, 1128.
(18) Dayan, S.; Bareket, Y.; Rozen, S. Tetrahedron 1999, 55, 3657.
10.1021/jo0013094 CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/22/2000

BrF3 as a Reagent in Organic Chemistry
J . Org. Chem., Vol. 66, No. 2, 2001 497
Sch em e 1
Sch em e 2
sible for the additional splitting of the vicinal carbons in
1
1
the 13C NMR 121.9 (CF2, dt, J CF ) 250 Hz, J CN ) 13.5
3
1
Hz) and 152.8 ppm (CO, dt, J CF ) 2.5 Hz, J CN ) 22.5
Hz) in accordance with the proposed structure.
The formation of this novel type of molecule21 can be
explained by the tendency of BrF3 to complex with two
electron rich sites, prior to the nucleophilic attack of the
fluoride on the oxime carbon atom, intermediate A. This
is followed by the formation of intermediate B, which
derives its relative stability from being a fluorocarboca-
tionic one (Scheme 1). It should be noted that adding
radical chain inhibitors such as dinitrobenzene, the
presence of oxygen, or conducting the reaction under
complete darkness does not change the outcome, reducing
the possibility of any radical involving mechanism.
This reaction is not limited to alkyl pyruvates, and
other R-keto esters behave similarly. Methyl 2-ketooc-
tanoate (4) and ethyl 2-ketododecanoate (5) were con-
verted to their methyl oxime ethers 6 and 7, respectively,
and then reacted with BrF3. In both cases, the reactions
proceeded smoothly, and the products were identified as
N-(1,1-difluoroheptyl)-N-methoxy methyl carbamate (8)
and N-(1,1-difluoroundecyl)-N-methoxy ethyl carbamate
(9). Alicyclic compounds containing tertiary hydrogens
also gave the rearranged products, despite the fact that
they are more sensitive to electrophilic attacks.22 Thus,
ethyl 4-cyclohexyl-2-ketobutanoate (10) was transformed
via its methyl oxime ether 11 to N-(3-cyclohexyl-1,1-
difluoropropyl)-N-methoxy ethyl carbamate (12) in 45%
yield.
invariably between -103 and -107 ppm.19 Although it
will be discussed later, we would like to mention now that
at higher temperature (+68 °C) this signal turned into a
quartet at -74.5 ppm (J ) 16 Hz) indicating a presence
of a CH3CF2 moiety. The 1H NMR spectrum clearly shows
a triplet at 2.07 ppm (J ) 16 Hz) for the above group
and two singlets for the methoxy group and the 1,1,1-
An interesting case involving a ring enlargement is the
reaction of dihydro-4,4-dimethyl-2,3-furandione (13)
(Scheme 2). We made the mono-oxime methyl ether (14)
with both 14N and 15N isotopes and reacted it with BrF3.
The product was identified as a six member lactone, 4,4-
difluoro-4,5-dihydro-5,5-dimethyl-N-methoxyoxazine-2-
one (15) whose 15N NMR shows a peak at -203 ppm (t,
2J NF ) 18 Hz).
trichloroethoxy group at 3.90 and 4.85 ppm, respectively.
1
The 13C NMR shows a triplet at 121.9 ppm (CF2, J CF
)
250 Hz), while the carbonyl signal at 152.8 ppm has only
a small coupling constant (t, J CF ) 2.5 Hz) indicating
3
Dimethyl 2-oxoglutarate (16) provided us an interest-
ing insight of the rearrangement process. When its
methyl oxime ether 17 (prepared with both isotopes 14N
and 15N) was reacted with bromine trifluoride, two main
products were isolated and purified. The minor one was
the expected by now rearranged product N-(3-car-
bomethoxy-1,1-difluoropropyl)-N-methoxy methyl car-
bamate (18), an oil at room temperature. In contrast to
all other products described so far, however, the major
product did crystallized at room temperature, and so
apart from the usual battery of analytical tests it was
also subjected to an X-ray analysis.23 All results point
toward yet another rearranged product, N-(4,4-difluoro-
4-methoxybutanoyl)-N-methoxy methyl carbamate (19)
(Scheme 3).
that it is separated by one atom from the CF2 group.
Since the product and the reaction leading to it are not
trivial and involve an unprecedented ester group migra-
tion, we looked for additional proofs beyond the high-
lighted above spectral properties, high-resolution MS,
and microanalysis which are in excellent agreement with
the proposed structure of 3. One such support could have
been provided by 15N NMR, but even with a highly
concentrated sample, we were unable to detect any useful
signal. We therefore repeated the synthesis with 25% 15N-
enriched H2NOMe prepared from Na15NO2, according to
a known procedure.20 The 15N NMR of the enriched
product, CH3CF215N(OMe)COOCH2CCl3 (3), indeed showed
a triplet at -197 ppm (CH3NO2 serving as reference) with
2J NF of 19 Hz, placing the nitrogen atom in the vicinity
of the CF2 group. The 15N atom in its turn was respon-
(21) A somewhat similar family of compounds, CF3N(X)COF (X )
OCF3, OMe and alike), had been synthesized by treating perfluoro
oxaziridines with nucleophiles: Sekiya, S.; DesMarteau, D. D. J .
Fluorine Chem. 1980, 15, 183.
(19) See for example Yang, Z. Y.; Burton, D. J . J . Org. Chem. 1991,
56, 5125.
(20) Hjeds, H. Acta Chem. Scand. 1965, 19, 1764.
(22) Rozen, S.; Gal, C. J . Org. Chem. 1987, 52, 2769 and 4928.

498 J . Org. Chem., Vol. 66, No. 2, 2001
Rozen and Ben-David
Sch em e 3
heating to 154 °C to achieve a complete free inversion-
rotation around the nitrogen atom and the 19F NMR
spectrum shows one signal at -82.6 ppm (t, J ) 14.5 Hz).
These data and further measurements at the tempera-
ture range between -53 and +154 °C enabled us to
calculate25 the ∆Gq of the NIR barrier for 8 to be 12.4 (
0.1 kcal/mol, which indeed is considered to be relatively
high.26 Further variable-temperature 19F NMR studies
for compounds 12 and 18 reveal similar ∆Gq for the NIR
barrier of 12.6 and 12.4 kcal/mol, respectively.
In conclusion, it has been demonstrated that BrF3 can
be used in organic chemistry without destroying sensitive
molecules and perform “counter intuitive” novel reactions
and even “double rearrangements” as in the case of the
formation of 19.
Exp er im en ta l Section
1H, 13C, and 15N NMR spectra were recorded with CDCl3 as
a solvent and Me4Si as an internal standard. 1H and 13C NMR
were measured at 500 and 125.76 MHz, respectively. The 15N
NMR spectra were measured at 50.68 MHz and are reported
in ppm upfield from CH3NO2 which also served as an external
standard. The 19F NMR spectra were measured at 338.8 MHz
and are reported in parts per million upfield from CFCl3
serving as an internal standard. Variable temperature experi-
ments were recorded with acetone as a solvent for low
temperatures and DMSO for high ones. High-resolution mass
spectra were measured with a VG micromass 7070H instru-
ment. IR spectra were recorded in CHCl3 solution or in KBr
pellets on a FTIR spectrophotometer. A Nonius Kappa CCD
diffractometer with Mo KR radiation (λ ) 0.7107 Å) was used
for the crystal structure elucidation.
P r ep a r a tion a n d Ha n d lin g of Br F 3. Although com-
mercially available, we prepare our own BrF3 by simply
passing 0.58 mol of pure fluorine through 0.2 mol of bromine
placed in a copper reactor at 0 to +10 °C. When no excess of
bromine is present, the BrF3 obtained is a pale yellow liquid
that freezes at 7-9 °C and has a density of 2.5.27 At that
reaction temperature, the higher oxidation state derivative -
BrF5 will not form in any appreciable amount,28 although we
always use a small excess of bromine, thereby keeping the
reagent from disproportionation to BrF5. This is also respon-
sible for the reddish coloration of the reagent. We store the
reagent in Teflon containers for long periods, but all reactions
were carried in glassware with no appreciable damage even
after many reactions. BrF3 is a strong oxidizer and tends to
react very exothermically with water and oxygenated organic
solvents. The work with BrF3 should be conducted in a well
ventilated area and caution and common sense be exercised. A
detailed description of our setup for working with fluorine has
been reported.18
As has already been stated, the key step for ionic
reactions with BrF3 is its complexation with electron
donors. There are two favorable ways for such groups in
17 to attract the electrophilic bromine atom through the
participation of the oxime moiety with either the vicinal
or the more remote carbomethoxy group. While the first
pathway gives rise to the expected 18, the very accom-
modating space created by the second possibility encour-
ages the formation of the intermediate A (Scheme 3). The
creation of the five-membered ring (B) and its consecutive
opening are responsible for the overall double rearrange-
ment of both the keto oxygen and the carboxylic group
to give the unexpected 19.24
In most products described so far, the nitrogen atom
is surrounded by three highly electronegative groups
namely CF2, OMe, and COOR. Such a situation is usually
responsible for high nitrogen inversion-rotation (NIR)
barriers. These barriers turn the two geminal fluorine
atoms into diastereotopic ones and indeed, they are not
equivalent at room temperature. The 19F NMR spectrum
of 8 may serve as an illustrative example. At +20 °C we
find two signals, one for each fluorine atom, at -81.5 (bs,
Wh/2 ) 960 Hz) and at -85.9 ppm (bs, Wh/2 ) 950 Hz).
Cooling the sample to -4 °C shifts these signals slightly
to -80.7 and -86.8 ppm, but with narrower Wh/2: 445
and 427 Hz correspondingly. Further cooling to -53 °C
decreases the inversion-rotation around the nitrogen
atom to such an extent that one fluorine atom resonates
P r ep a r a tion of Non com m er cia l r-Ketoester s. These
starting materials have been prepared by several routes.
Although isolated and purified, no attempts have been made
to obtain analytical samples before the reaction with meth-
oxylamine.
at -79.3 ppm (ddd, J FF ) 201 Hz, J H(1)F ) 12 Hz, J H(2)F
)
5 Hz) while the other at -87.7 ppm (dt, J FF ) 201 Hz,
J HF ) 20 Hz). Heating the sample, on the other hand,
reveals a coalescence at +31 °C (-82.8 ppm, bs, Wh/2
)
2000 Hz). Higher temperatures reduces the effect of the
NIR barrier and at +68 °C the two fluorine atoms
resonate at -82.5 ppm (bs, Wh/2 ) 230 Hz). It requires
2,2,2-Tr ich lor oeth yl p yr u va te (1) was prepared by direct
esterification of pyruvic acid and trichloroethanol using DCC.
The compound was mentioned previously in a patent,29 but
was never properly described: bp(2 mm) of 87-92 °C; IR 1737,
1
1758 cm-1; H NMR 4.9 (2 H, s), 2.56 ppm (3 H, s); 13C NMR
(23) Crystal data for C8H13F2NO5 (19): M ) 241.19, triclinic, space
group P1h, a ) 8.9090(4) Å, b ) 11.2080(5) Å, c ) 11.5810(7) Å, R )
103.24(3)°, â ) 103.446(3)°, γ ) 90.875(3)°, U ) 1091.9(1) Å3, z ) 4,
2261 unique reflections were measured (Rint ) 0.0000). The structure
was solved by direct methods and refined by full-matrix least squares
189.9, 158.5, 93.8, 74.8, 26.8 ppm.
(25) Gunther, H. NMR Spectroscopy; J ohn Wiley & Sons: New York,
1995; p 343
(26) Belostotskii, A. M.; Hassner, A. J . Phys. Org. Chem. 1999, 12,
659.
on F2. The final refinement converged at R1 ) 0.0456, and wR2
)
0.1282 [I > 2σ(I)] and at R1 ) 0.0490, and wR2 ) 0.1323 for all data
(see also the Supporting Information section).
(24) A somewhat similar “double” attack of a nucleophilic nitrogen
on a remote carbocationic center resulting on a bicyclic product was
observed in the past with the reaction of 2,5-dibromoadipanilides with
KF. Shahak, I.; Rozen, S.; Bergmann, E. D. J . Org. Chem. 1971, 36,
501.
(27) Simons, J . H. Inorg. Synth.; McGraw-Hill: New York, 1950;
Vol. III, p 184.
(28) Stein, L. J . Am. Chem. Soc. 1959, 81, 1269.
(29) Gleason, J . G.; Holden, K. G.; Huffman, W. F. U.S. Patent 4,-
103,086, 1978; Chem. Abstr. 1979, 90, 54957u.

BrF3 as a Reagent in Organic Chemistry
J . Org. Chem., Vol. 66, No. 2, 2001 499
Eth yl 2-k etod od eca n oa te (5)30 was prepared from decyl-
magnesium bromide and diethyloxalate. A 1 M THF solution
of the Grignard reagent (75 mmol) was added dropwise under
N2 atmosphere to a cold (-75 °C) solution of 10 mL (68 mmol)
of diethyloxalate in 100 mL of dry ether. After the addition
was complete, the reaction mixture was stirred at -75 °C for
an additional 0.5 h. The reaction was quenched with diluted
HCl, the aqueous layer was extracted with ether (3 × 100 mL),
and the combined organic layers were washed with saturated
NaCl, dried over MgSO4, and evaporated: oil; >90% yield; IR
Hz), 4.03 (3 H, s), 2.55 (2 H, dd, J 1 ) 9 Hz, J 2 ) 6.5 Hz), 1.77-
1.66 (5 H, m), 1.4-1.29 (2 H, m), 1.35 (3 H, t, J ) 7 Hz), 1.26
- 1.1 (4 H, m), 1.0-0.85 ppm (2 H, m); 13C NMR 163, 153,
62.8, 61.4, 37.6, 33.2, 32.8, 26.4, 26.1, 23, 14 ppm; HRMS (CI)
calcd for C13H24NO3 242.1756 (M + 1)+, found 242.1754.
3-(O-m eth yl oxim e)d ih yd r o-4,4-d im eth yl-fu r a n -2-on e
(14): 90% yield; mp 110 °C (hexane/CH2Cl2); IR 1767 cm-1
;
1H NMR 4.1 (3 H, s), 4.09 (2 H, s), 1.44 ppm (6 H, s); HRMS
(CI) calcd for C7H12NO3 158.0817 (M + H)+, found 158.0814.
Dim eth yl 2-(O-m eth yl oxim e)glu ta r a te (17):33 oil; 70%
1
1730 cm-1; H NMR 4.32 (2 H, q, J ) 7 Hz), 2.83 (2 H, t, 7
1
yield; IR 1741 cm-1; H NMR 4.07 (3 H, s), 3.87 (3 H, s), 3.68
Hz), 1.63 (2 H, m), 1.37 (3 H, t, J ) 7 Hz), 1.40-1.25 (14 H,
m), 0.88 ppm (3 H, t, J ) 6.5 Hz); 13C NMR 195, 161, 62.2,
39.2, 31.8, 29.4, 29.2, 28.9, 22.9, 22.6, 13.9 ppm; HRMS (CI)
calcd for C14H27O3 243.1960 (M + H)+, found 243.1952.
Eth yl 4-cycloh exyl-2-k etobu ta n oa te (10)17 was prepared
according to a published procedure18 from ethyl 4-cyclohexy-
lbutanoate. This ester was R-hydroxylated with HOF‚CH3CN
and then further oxidized with CrO3 in an overall yield of
(3 H, s), 2.87 (2 H, t, J ) 8 Hz), 2.55 ppm (2 H, t, J ) 8 Hz);
HRMS (CI) calcd for C8H14NO5, 204.0872 (M + H)+, found
204.0874.
Rea ction s w ith Br F 3. The appropriate methyl oxime ether
(5 mmol) was dissolved in 30-40 mL of CFCl3 and cooled to 0
°C. Bromine trifluoride (0.4 mL, 7.5 mmol) dissolved in cold
(0 °C) CFCl3 (20 mL) was added dropwise during 10 min. After
an additional 5 min, the reaction mixture was treated with a
saturated Na2SO3 solution until colorless. The two layers were
separated; the aqueous layer extracted with CH2Cl2, the
combined organic layers were washed with saturated NaCl,
dried, and the solvent was removed. The residues were purified
by flash silica gel chromatography with petroleum ether and
benzene serving as eluent.
40%: oil; IR 1750, 1729 cm-1 1H NMR 4.31 (2 H, q, J ) 7
;
Hz), 2.84 (2 H, t, 7.5 Hz), 1.74-1.67 (5 H, m), 1.52 (2 H, q, J
) 7 Hz), 1.37 (3 H, t, J ) 7 Hz), 1.31-1.15 (4 H, m), 1.0-0.87
ppm (2 H, m); 13C NMR 195, 161, 62, 36.8, 36.6, 32.8, 30, 26.2,
25.9, 13.7 ppm; HRMS calcd for C12H20O3 212.1412 (M)+, found
212.1413.
Meth yl 2-k etoocta n oa te (4) was also prepared by R-hy-
droxylation of methyl octanoate with HOF‚CH3CN. Its physical
properties fully matched those published in the literature.31
Dih ydr o-4,4-dim eth yl-2,3-fu r an dion e (13) and dim eth yl-
2-oxoglu ta r a te (16) were purchased from Aldrich and used
without further purification.
N-(1,1-Diflu or oet h yl)-N-m et h oxy-2,2,2-t r ich lor oet h yl
ca r ba m a te (3) was obtained as an oil in 60% yield: IR 1755
cm-1; HRMS (CI) calcd for C6H9Cl3F2NO3 285.9616 (M + H)+,
found 285.9618. Anal. Calcd for C6H8Cl3F2NO3: C, 25.15; H,
2.81; F, 13.26; N, 4.89. Found: C, 25.46; H, 2.87; F, 13.66; N,
4.78. 1H, 19F, 13C, and 15N NMR were presented in the
Discussion above.
P r ep a r a tion of O-Meth yl Oxim e Eth er s of th e r-Ke-
toester s. The ketoesters (10 mmol) were dissolved in 100 mL
of EtOH. An aqueous solution of methoxylamine hydrochloride
(10 mmol) was added and the reaction mixture cooled to -10
°C. If needed, more ethanol should be added until a clear
solution is obtained. A 3.6 mL portion of 10% aqueous NaOH
was dissolved in EtOH and added dropwise during 30 min,
keeping the temperature below -5 °C. The reaction mixture
was kept overnight at -15 °C, and then all the liquids were
evaporated. Water and ether were added, and the aqueous
layer was extracted with more ether. The combined organic
layers were washed with saturated NaCl, dried, and evapo-
rated. The oxime methyl ethers, mostly unknown, were
purified by chromatography on silica gel. After this procedure
the products were more than 90% pure and were subjected to
the next reaction step without further purification. It should
be noted that if the reaction is carried out at room temperature
or under heating, it is much faster, but the differentiation
between the two present carbonyls is reduced and about 25%
of the attack by the amine occurs on the esteric carbonyl.
2′,2′,2,′-Tr ich lor oeth yl 2-(O-m eth yl oxim e)p yr u va te (2):
oil; 72% yield ; IR 1737 cm -1; 1H NMR 4.91 (2 H, s), 4.11 (3
H, s), 2.11 p p m (3 H, s); 13C NMR 161.8, 147.5, 94.5, 74.3,
63.4, 11.2 p p m .
N-(1,1-Diflu or oh ep tyl)-N-m eth oxy m eth yl ca r ba m a te
(8) was obtained as an oil in 65% yield: IR 1748 cm-1; 1H NMR
3.84 (3 H, s), 3.81 (3 H, s), 2.46-2.23 (2 H, m), 1.55-1.45 (2
H, m), 1.39-1.27 (6 H, m), 0.89 ppm (3 H, t, J ) 7 Hz); 19F
NMR at various temperature was described in the discussion
section above; 13C NMR 155, 123 (t, 1J CF ) 252 Hz), 65.1, 53.6,
2
35.2 (t, J CF ) 26 Hz), 31.4, 28.6, 22.4, 22.3, 13.9 ppm; HRMS
(CI) calcd for C10H20F2NO3 240.1411 (M + H)+, found 240.1399.
Anal. Calcd for C10H19F2NO3: C, 50.20; H, 8.00; N, 5.85.
Found: C, 49.77; H, 7.79; N, 5.47.
N-(1,1-Diflu or ou n d ecyl)-N-m eth oxy eth yl ca r ba m a te
(9) was obtained as an oil in 35% yield: IR 1738 cm-1; 1H NMR
4.28 (2 H, q, J ) 7 Hz), 3.81 (3 H, s), 2.45-2.20 (2 H, m), 1.58-
1.45 (2 H, m), 1.35 (3 H, t, J ) 7 Hz), 1.38-1.25 (14 H, m),
0.88 ppm (3 H, t, J ) 6 Hz); 19F NMR 295 K (22 °C)
(coalescence) -84.5 ppm (Wh/2 ) 1800 Hz); HRMS (CI) calcd
for C15H30F2NO3 310.2194 (M + H)+, found 310.2199. Anal.
Calcd for C15H29F2NO3: C, 58.23; H, 9.45; N, 4.53. Found: C,
58.63; H, 9.52; N, 4.15.
N-(3-Cycloh exyl-1,1-d iflu or op r op yl)-N-m eth oxy eth yl
ca r ba m a te (12): oil; 45% yield; IR 1742 cm-1; 1H NMR 4.28
(2 H, q, J ) 7 Hz), 3.80 (3 H, s), 2.47-2.23 (2 H, m), 1.73-
1.65 (5 H, m), 1.46-1.10 (6 H, m), 1.35 (3 H, t, J ) 7 Hz),
0.98-0.83 ppm (2 H, m),. 19F NMR 250 K (-23 °C) -79.8 (d,
J ) 197 Hz), -87.2 ppm (d, J ) 197 Hz), 306 K (33 °C)
Meth yl 2-(O-m eth yl oxim e)octa n oa te (6): oil; 68% yield;
1
IR 1725 cm-1; H NMR 4.04 (3 H, s), 3.86 (3 H, s), 2.56 (2 H,
(coalescence) -84.5 ppm (Wh/2 ) 1700 Hz), 427 K (154 °C)
t, J ) 8 Hz), 1.48 (2 H, m), 1.38-1.27 (6 H, m), 0.90 ppm (3 H,
t, J ) 6 Hz); 13C NMR 164, 152.5, 63, 52, 31.5, 29, 25.7, 25.2,
22.2, 13.7 ppm; HRMS calcd for C10H19NO3, 201.1365 (M)+,
found 201.1364.
1
-82.4 ppm (bs, Wh/2 ) 38 Hz); 13C NMR 154, 124 (t, J CF
)
2
252 Hz), 65.1, 63.0, 37.1, 33.0 (t, J CF ) 26 Hz), 33.0, 29.6,
26.5, 26.2, 14.2 ppm; HRMS (CI) calcd for C13H24F2NO3
280.1724 (M + H)+, found 280.1727. Anal. Calcd for C13H23F2-
NO3: C, 55.90; H, 8.30; N, 5.01. Found: C, 55.66; H, 8.23; N,
5.04.
Eth yl 2-(O-m eth yl oxim e) d od eca n oa te (7): oil; 87%
1
yield; IR 1720 cm-1; H NMR 4.32 (2 H, q, J ) 7 Hz), 4.04 (3
H, s), 2.54 (2 H, dd, J 1 ) 8 Hz, J 2 ) 7 Hz), 1.47 (2 H, m), 1.35
(3 H, t, J ) 7 Hz), 1.38-1.25 (14 H, m) 0.88 ppm (3 H, t, J )
6 Hz); 13C NMR 164, 153, 63, 61.7, 31.9, 29.6, 29.3, 26, 25.5,
22.7, 14.1 ppm; HRMS (CI) calcd for C15H30NO3 272.2226 (M
+ H)+, found 272.2225.
4,4-Diflu or o-4,5-d ih yd r o-5,5-d im et h yl-N-m et h oxyox-
1
a zin e-2-on e (15): oil; 50% yield; IR 1764 cm-1; H NMR 4.0
(2 H, t, J ) 1 Hz), 3.95 (3 H, s), 1.24 ppm (6 H, s); 19F NMR
299 (26 °C) -98.5 ppm (bs, Wh/2 ) 160), 402K (129 °C) -96.6
3
1
ppm (s, Wh/2 ) 13 Hz); 13C NMR 148.5 (dt, J CF ) 1.5 Hz, J CN
Eth yl 2-(O-m eth yl oxim e)-4-cycloh exylbu ta n oa te (11):
1
1
32
1
oil; 85% yield; IR 1719 cm-1; H NMR 4.31 (2 H, q, J ) 7
) 19.5 Hz), 120.7 (dt, J CF ) 253 Hz, J CN ) 13.5 Hz), 71, 65,
38.0 (t, J CF ) 22 Hz), 18 ppm (t, J CF ) 2.5 Hz); 15N NMR:
2
3
(30) J ahnke, D.; Reinhenkel, H. Ger. Patent DD85071, 1971; Chem.
Abstr. 1973, 78, 110621.
(31) Dubowchik, G. M.; Padilla, L.; Edinger, K.; Firestone, R. A. J .
Org. Chem. 1996, 61, 4676.
(32) Itoh, K.; Kori, M.; Inada, Y.; Nishikawa, K.; Kawamatsu, Y.;
Sugihara, H. Chem. Pharm. Bull. 1986, 34, 1128.
(33) Blomberg, S.; Cronholm, T. Eur. J . Biochem. 1979, 101, 111.

500 J . Org. Chem., Vol. 66, No. 2, 2001
Rozen and Ben-David
2
-203 ppm (t, J NF ) 18 Hz); HRMS (CI) calcd for C7H12F2-
H, s), 3.53 (3 H, s), 3.0 (2 H, t, J ) 8 Hz), 2.36 ppm (2 H, tt, J 1
) 10.5 Hz, J 2 ) 8 Hz); 19F NMR -78.4 ppm (t, J ) 10.5 Hz);
13C NMR 169.4 (d, 1J CN ) 9 Hz), 152.7 (d, 1J CN ) 23 Hz),125.2
NO3, 196.0785 (M + H)+, found 196.0786. Anal. Calcd for
C7H11F2NO3: C, 43.08; H, 5.68; F, 19.47; N, 7.18. Found: C,
43.42; H, 5.65; F, 18.84; N, 6.81.
1
3
(t, J CF ) 261 Hz), 63.8, 54.1, 49.8 (t, J CF ) 4 Hz), 30.5 (t,
2
3J CF ) 3.4 Hz), 30.36 ppm (t, J CF ) 31 Hz); 15N NMR -169
N-(3-Carbomethoxy-1,1-difluoropropyl)-N-methoxy meth-
yl ca r ba m a te (18): oil; 22% yield; IR 1743 cm-1; 1H NMR 3.85
(3 H, s), 3.81 (3 H, s), 3.70 (3 H, s), 2.74 (2 H, tt, J 1 ) 14.5 Hz,
J 2 ) 8 Hz), 2.57 ppm (2 H, t, J ) 8 Hz); 19F NMR 218 K (-55
(s); HRMS (CI) calcd for C8H13FNO5 222.0778 (M - HF + H)+,
found 222.0773. Anal. Calcd for C8H13F2NO5: C, 39.84; H, 5.43;
F, 15.75; N, 5.81. Found: C, 39.54; H, 5.19; F, 15.82; N, 6.10.
°C) -80.1 ppm (dd, J 1 ) 201 Hz, J 2 ) 11 Hz), -88.0 (dt, J 1
)
201 Hz, J 2 ) 18.5 Hz), 301 K (28 °C) (coalescence), 423 K (150
Ack n ow led gm en t. This research was supported by
the USA-Israel Binational Science Foundation (BSF),
J erusalem, Israel.
°C) -84 ppm (bs, Wh/2 ) 48 Hz); 13C NMR 171.8, 154.6 (dt,
1
1
1
3J CF ) 2.5 Hz, J CN ) 20.4 Hz), 122.3 (dt, J CF ) 253 Hz, J CN
2
) 12.7 Hz), 65, 53.5, 51.6, 30.7 (t, J CF ) 27 Hz), 27.3 ppm (t,
3J CF ) 4 Hz); 15N NMR -198 (t, J NF ) 19 Hz); HRMS (CI)
2
calcd for C8H13FNO5 222.0778 (M -HF + H)+, found 222.0779.
Anal. Calcd for C8H13F2NO5: C, 39.84; H, 5.43; N, 5.81.
Found: C, 40.17; H, 5.31; N, 5.53.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystal struc-
ture data and an ORTEP diagram for compound 19. This
material is available free of charge via the Internet at
http://pubs.acs.org.
N-(4,4-Diflu or o-4-m eth oxybu ta n oyl)-N-m eth oxy m eth -
yl ca r ba m a te (19): 35% yield; white solid; mp 39-41 °C
(hexane); IR 1742, 1710 cm-1; H NMR 3.92 (3 H, s), 3.82 (3
J O0013094
1