New building blocks for fluorinated bioimidazole derivatives II: Preparation of β-fluorourocanic acids

Article abstract of DOI:10.1021/jo0200419

Replacement of vinyl hydrogen with fluorine is based on addition of an FBr equivalent to a double bond followed by HBr elimination. This sequence has been adapted to prepare 3-fluoro-3-imidazolylpropenoic acids (β-fluorourocanic acids), and the related fl

Full text of DOI:10.1021/jo0200419

3468
J . Org. Chem. 2002, 67, 3468-3473
New Bu ild in g Block s for F lu or in a ted Bioim id a zole Der iva tives II:
P r ep a r a tion of â-F lu or ou r oca n ic Acid s
Bohumil Dolensky and Kenneth L. Kirk*
Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases,
National Institutes of Health, Bethesda, Maryland 20892
Kennethk@bdg8.niddk.nih.gov
Received J anuary 18, 2002
Replacement of vinyl hydrogen with fluorine is based on addition of an FBr equivalent to a double
bond followed by HBr elimination. This sequence has been adapted to prepare 3-fluoro-3-imidazolyl-
propenoic acids (â-fluorourocanic acids), and the related fluorinated imidazolyl propenals and prop-
2-en-1-ols, from urocanic acid. Tritylation of the imidazole nitrogen was necessary for successful
addition of “FBr” to the double bond, and prior reduction of the carboxyl group to the alcohol was
required to provide the desired chemoselectivity in the elimination of HX. Reoxidation and
deprotection produced the fluorinated urocanic acids.
In tr od u ction
On the basis of literature data, it seems that the
addition8,9 of FX followed by HX elimination10,11 could
provide a direct route to â-aryl-â-fluoro-acrylic acids.
Furthermore, we have shown previously that this ap-
proach is applicable in imidazole chemistry, as we
demonstrated12 by the preparation of â-fluorohistamines
from 1-trityl-4-vinyl-1H-imidazole. However, applications
of this strategy to compounds with additional functional
groups are not well studied. The limitations, problems,
and resolution of the problems related to the addition of
an FBr equivalent to polyfunctional substrates and a
protocol for preparation of â-aryl-â-fluoro-acrylic acids are
described in this paper. Application of this protocol to
the syntheses of (E)- and (Z)-â-fluorourocanic acid is
detailed.
We have been involved for many years in the synthesis
and biological evaluation of fluorinated analogues of
biologically important imidazoles. As part of this re-
search, we have prepared fluorinated analogues of uro-
canic acid (1), the imidazolyl acrylic acid that is produced
in vivo by histidine catabolism.1 Interest in urocanic acid
and derivatives has increased with the discovery that (Z)-
urocanic acid, formed by photoisomerization of the ini-
tially formed (E)-isomer in the epidermis, functions
systemically as an immunosuppressive agent.2 To study
these and others processes, we have developed methods
or used known ones to prepare (Z)- and (E)-isomers of
R-fluorourocanic,3 2-fluorourocanic, and 4-fluorourocanic
acids.4 To complete the inventory of monofluorinated
urocanic acids, we now have developed synthetic proce-
dures that provide (E)- and (Z)-â-fluorourocanic acids (2a
and 2b, respectively).
Methods to prepare â-fluoro-â-aryl acrylic acids (in
large part â-fluorocinnamate derivatives) can be grouped
into three general procedures. One procedure5 involves
a multistep construction of the carbon skeleton from
CF2dCH-Li, CO2, and aryl-MgBr. A second recently
published6 method is based on catalytic coupling reaction
of aryl-I with CF2BrCH2COOEt. The third procedure is
based on addition of HF to a triple-bond precursor.7 In
general, the applicability of these methods depends on
the availability of starting compounds and the compat-
ibility of functional groups.
Ch em istr y
The facile addition of “FBr” to the double bond of
unsaturated compounds containing a carboxylic acid
remote from the double bond by Et3N‚3HF and NBS is
documented.8,13 Furthermore, the addition of FBr to a
double bond that has the carboxyl function directly
attached (acrylic acids) also has been described.14,15 Issues
to be considered in regard to the presence of the carboxyl
group include the possible intervention of bromodecar-
boxylation (a variant of the Hunsdiecker reaction) as a
competing process.16
(8) Haufe, G.; Alvernhe, G.; Laurent, A.; Ernet, T.; Goj, O.; Kro¨ger,
S.; Sattler, A. Org. Synth. 1999, 76, 159.
(1) Re´tey, J . Arch. Biochem. Biophys. 1994, 314, 1.
(2) (a) DeFabo, E. C.; Noonan, F. P. J . Exp. Med. 1983, 158, 84. (b)
Noonan, F.; DeFabo, E. C. Immunol. Today 1992, 13, 250. (c) Garssen,
J .; Norval, M.; Crosby, J .; Dortant, P.; Van Loveren, H. Immunology
1999, 96, 298; and references therein.
(3) Percy, E.; Singh, M.; Takahashi, T.; Takeuchi, Y.; Kirk, K. L. J .
Fluorine Chem. 1998, 91, 5. In this report, the scheme we constructed
describing the mechanism of action of urocanase was oversimplified
and contained incorrect structures. Please see ref 1 for a thorough
discussion of this mechanism.
(4) Fan, J .; Dolensky, B.; Kim, I. H.; Kirk, K. L. J . Fluorine Chem.,
in press.
(5) Gillet, J . P.; Sauvetre, R.; Normant, J . F. Synthesis 1982, 297.
(6) Peng, S.; Qing, F.-L.; Li, Y.-Q.; Hu, C.-M. J . Org. Chem. 2000,
65, 694.
(9) Boguslavskaya, L. S. Russ. Chem. Rev. 1972, 41, 740. (b)
Boguslavskaya, L. S. Russ. Chem. Rev. 1984, 53, 1178. (c) Yoneda, N.
Tetrahedron 1991, 47, 5329.
(10) Suga, H.; Hamatani, T.; Guggisberg, Y.; Schlosser, M. Tetra-
hedron 1990, 46, 4255.
(11) Eckes, L.; Hanack, M. Synthesis 1978, 217.
(12) Dolensky, B.; Kirk, L. K. J . Org. Chem. 2001, 66, 4687.
Corrections: the 19F NMR shift of compound 8 should be -169.1 ppm;
the page in ref 23 should be 3043.
(13) (a) Michel, D.; Schlosser, M. Synthesis 1996, 1007. (b) Sattler,
A.; Haufe, G. Tetrahedron 1996, 52, 5469.
(14) Gershon, H.; McNeil, M. W.; Bergmann, E. D. J . Med. Chem.
1973, 16, 1407.
(15) Bose, A. K.; Das, K. G.; Funke, P. T. J . Org. Chem. 1964, 29,
1202.
(7) Cousseau, J .; Albert, P. Bull. Soc. Chim. Fr. 1986, 910.
(16) Chowdhury, S.; Roy, S. J . Org. Chem. 1997, 62, 199.
10.1021/jo0200419 This article not subject to U.S. Copyright. Published 2002 by the American Chemical Society
Published on Web 04/24/2002

Preparation of â-Fluorourocanic Acids
J . Org. Chem., Vol. 67, No. 10, 2002 3469
Sch em e 1
Sch em e 2
Ta ble 1
Et3N‚3HF:NBS
time [h]
9a
10
12a
12b
11
Attempts to add FBr to urocanic acid (1) were unsuc-
cessful, and 1 was unchanged (1H NMR of crude product).
The lack of reaction is probably due to insufficient
solubility of 1 in the reaction medium (CH2Cl2). However,
when DMSO was used as a cosolvent to increase solubil-
ity, an intractable tar was obtained. Reaction of methyl
urocanate (3) with FBr in methylene chloride gave a
complex reaction mixture, which had very similar fea-
tures in the 1H NMR to the product mixture obtained
from the reaction of FBr with unprotected 4-vinyl-1H-
imidazole.12 As before, there was no evidence of any
adduct of FBr (on the basis of 19F NMR). Thus, imidazole
ring protection is necessary. On the basis of its effective-
ness in our previous work,12 we used the trityl group for
this purpose.
Treatment of 1-trityl-E-urocanic acid 4 with Et3N‚3HF
and NBS (Scheme 1) gives four products. Under ordinary
conditions of FBr addition (substrate:Et3N‚3HF:NBS
molar ratio ) 1:1.5:1.1), bromoolefin 5 is formed as the
major product and compounds 6-8 as byproducts17 (23%
4; 62% 5; 8% 6; 4% 7; 3% 8). When the ratio of reagents
to substrate was increased (1:3:2), the main product was
7 (0% 4; 0% 5; 18% 6; 57% 7; 25% 8). From these two
experiments, it appears that compound 7 is formed by
addition of FBr to the double bond of initially formed 5
instead of by bromodecarboxylation of 6. This is reason-
able since the carboxyl group of 6 should be relatively
more resistant to bromodecarboxylation initiated by NBS
than 4 (where the carboxyl group is situated on a double
bond).
In light of the problems for FBr addition presented by
the presence of the free carboxyl group, we concluded that
protection of both the imidazole ring and carboxyl group
is required. Accordingly, tritylated methyl urocanate 9a
was next considered as the starting compound. We were
encouraged by the previous demonstration18 of successful
addition of bromine and chlorine to 9a . Furthermore, FBr
addition has been carried out successfully on cinnamate
esters.15,19
Unfortunately, we also encountered serious problems
with the addition of FBr to 9a . Under the ordinary
conditions, four products were formed, including the
desired FBr-adduct 10 and Br2-adduct 11 (Scheme 2). In
a
b
c
d
e
f
1.5:1.1
1.5:1.1
3.0:1.1
3.0:2.2
3.0:2.2
3.0:2.2
12
117
15
16
4
38
49
27
0
7
+
46
33
52
68
72
12
6
18
15
22
5
3
+
6
10
+
14
7
0
0
0
16
0
16
74
addition, R-bromourocanates 12a and 12b, products of
subsequent dehydrofluorination of 10 and/or dehydro-
bromination of 11, were isolated. We found that the
outcome of the reaction is strongly affected the by ratio
of reagents, reaction time, and reaction mixture workup
(Table 1). Under the ordinary conditions, 9a is converted
to FBr-adduct 10 in about 50% yield (row a). When the
reaction time is extended, there is no further conversion
of 9a to the FBr-adduct 10, but there is an increased
formation of bromourocanates 12 (row b). When the
amount of Et3N‚3HF is doubled, the conversion is slightly
better (row c). Finally, doubled amounts of both reagents
(Et3N‚3HF and NBS) produce the best results and the
starting urocanate 9a is converted completely to FBr-
adduct 10 and bromourocanates 12 (row d). In the same
conditions but with a shorter reaction time, the Br2-
adduct 11 could be detected spectroscopically in the crude
product (row e). The product distribution obtained by the
ordinary workup of the reaction is shown in row f.
It is apparent from the above results that the FBr-
adduct 10 is quite sensitive to basic conditions, properties
not reported for FBr- or FCl-adducts of cinnamates15,19
or Br2- of Cl2-adducts of urocanates.18 The FBr-adduct
10 eliminates HF to produce the R-bromourocanates 12
and gives no trace of the desired fluorourocanate 13 (on
the basis of 1H NMR) when treated with base (Et3N,
NaHCO3, or t-BuOK). Elimination to 12 also occurred
under attempts to displace the bromine with potassium
phthalimide or sodium azide. Since displacement of
bromine in FBr-adducts of â-alkylacrylic acids14 by am-
monium can be carried out (10-60%), it is apparent that
the aryl substituent is responsible for this extreme
reactivity. Treatment of FBr- or FCl-adducts of cinnama-
tes with base15,19 or heating neat15 above 50 °C also
results in elimination of â-halogen. Similar behavior is
seen with Br2- or Cl2-adducts of tritylated ethyl urocan-
ate.18 The fact that preferential HBr elimination occurs
with FBr-adducts of styrene or vinyl imidazole reveals
the role of the carboxylic function in the loss of HF in
the present case. Since there is also preferential loss of
HF from FBr-adducts of symmetrically substituted ole-
(17) Compounds 5 and 7 were isolated, while the presence of 6 and
8 was based on MS and NMR spectroscopic evidences.
(18) Cloninger, M.; Frey, P. A. Bioorg. Chem. 1998, 26, 323.
(19) Cabaleiro, M. C.; Garay, R. O.; J . Chem. Soc., Perkin Trans. 2
1987, 1473.

3470 J . Org. Chem., Vol. 67, No. 10, 2002
Dolensky and Kirk
Sch em e 3a
a
(a) (CF3CH2O)2P(O)CH2COOCH3, ref 31, 9b 71-95%; (b) (i) MeOH/H+, ref 32, (ii) TrCl, Et3N, ref 33, 9a 91-95%; (c) DIBAL-H, ref
34, 14a 46-59%, 14b 65-76%; when we used reduction by LAH (ref 35), also reduction of the double bond (18%) was found; (d) NBS,
3HF‚Et3N, 15a 59-76%, 15b 74%; (e) Et3N, 110 °C, 19a 52-62%, 19b 80%; (f) MnO2, 20a 61-80%, 20b 92%; (g) NaClO2, ref 36, 21a
68%, 21b 71%; (h) CH3COOH/HCl/H2O, 2a 100%, overall ca. 10%, 2b 100%, overall ca. 23%.
fins, for example, from R-bromo-R′-fluorosuccinate esters,
special effects of fluorine may also be involved.15,20
The problem of preferential elimination of HF vs HBr
from the product of FBr addition to the furan ring of
khellin was partially solved21 by using KF with DMA as
the elimination agent. Unfortunately, these conditions
again produced no fluorourocanate 13 (1H NMR) when
carried out on 10. In addition, we found that under these
elimination conditions, the initially formed bromouro-
canates 12 are partially converted 4-methoxycarbonyl-
4-ethynyl- and 4-ethynyl-1-trityl-1H-imidazoles22 as re-
sult of loss of HBr or both HBr and COOCH3. A similar
subsequent reaction is not possible for the five-membered
ring of khellin.
It should be noted that, in all cases, the (Z)-isomer 12a
was formed as the major isomer. Since the trans addition
of FBr gives the erythro isomer of FBr-adduct 10, mainly
cis elimination of HF must occur to give 12a , rather than
trans elimination to give 12b. In contrast, we found22 that
treatment of Br2-adduct 11 with base gives mainly 12b,
as a result of expected trans elimination of HBr. The
same selective cis HF elimination was also found with
cinnamates.15,19
From these results, it is clear that the influence of the
carboxyl group precludes formation of the â-fluorouro-
canate 13. To avoid the directive effects of the carboxylic
function in the elimination reaction, compounds 9 were
reduced to the allyl alcohols 14 and the sequence of
reactions to obtain acids 2 was applied (Scheme 3).
At this point, we note that the FBr-adducts 6 and 10
(as well as the Br2-adduct 11) were formed as single
diastereoisomers (1H NMR). Trans addition of FBr to the
(E)-olefin would produce the erythro configuration.9
However, this diastereoselectivity is observed only with
(E)-olefins. The (Z)-olefins produced a mixture of both
diastereoisomers.19,23 Consistent with this previous ob-
servation, the addition of FBr to the (E)-allylalcohol 14a
takes place cleanly and forms the erythro (S*,R*)-isomer
15a in 60-78% yield. The threo isomer 15b was found
only as a minor impurity (<5%) during chromatographic
purification of 15a . On the other hand, the addition of
FBr to the (Z)-allylalcohol 14b proceeded with a good
yield but gave a mixture of diastereoisomers 15a and 15b
in a ratio of 8:2. We note that we did not observe any
isomerization of FBr-adducts 6, 10, or 15, a process that
may occur at higher temperatures.15 We attempted to
alter the ratio of isomer formation by using a larger
excess of Et3N‚3HF or by adding of Et3N in order to form
the more nucleophilic24 Et3N‚2HF. The effect of HF
concentration on the course of FBr addition has been
described.25 As can be seen from Table 2, the ratio of 15a :
15b can be affected slightly by addition of Et3N. However,
when more than 0.5 equiv of Et3N to Et3N‚3HF is used,
the overall conversion of 14b to isomers 15a and 15b is
dramatically lower.
The FBr addition proceeded with Markovnikov selec-
tivity in accordance with all known FBr additions to
â-aryl-acrylates, acrylates, or styrene derivatives wherein
the Markovnikov product is formed as the major or only
one9 (one surprising exception is addition to â-oxiranyl-
styrene26). The total regioselectivity we observe probably
results from reinforcing orientation effects of the aryl and
carboxyl groups. Although not isolated, we found NMR
evidence for the formation 16 (“succinimide-Br” addition)
and 17 (“HO-Br” addition) (see Supporting Information,
Scheme 4), compounds analogous to byproducts formed
from vinylimidazole.12 Compound 18, wherein the OH
and F groups are transposed (see Supporting Informa-
tion, Scheme 4) due to a possible rearrangement involv-
ing the hydroxy group,27 was not observed.
Several conditions were investigated for the dehydro-
bromination of adducts 15. The best conditions proved
to be Et3N in DMSO at 110 °C, which cleanly produced
olefins 19 in yields of 50-80%. We observed no isomer-
ization about the double bond in that dehydrobromina-
tion of mixtures of 15a and 15b gave the same ratio
(24) Giudicelli, M. B.; Picq, D.; Veyron, B. Tetrahedron Lett. 1990,
31, 6527.
(25) Hamman, S.; Beguin, C. G. J . Fluorine Chem. 1983, 23, 515.
Hamman, S.; Beguin, C. G. J . Fluorine Chem. 1979, 13, 163.
(26) Hedhli, A.; Baklouti, A. J . Org. Chem. 1994, 59, 5277. Appar-
ently regioselectivity was established only from 19F NMR. We have
found such data quite sensitive to sample concentration and to
impurities. In contrast, we found analyses of 1H-13C and 19F-13C
coupling constants in 13C NMR to be a reliable basis for structure
assignment.
(20) (a) Hudlicky, M. J . Fluorine Chem. 1972, 2, 1. (b) Hudlicky, M.
J . Fluorine Chem. 1986, 32, 441. (c) Lee, J . G.; Bartsch, R. A. J . Am.
Chem. Soc. 1979, 101, 228.
(21) Nash, S. A.; Gammill R. B. Tetrahedron Lett. 1987, 28, 4003.
(22) Dolensky, B.; Kirk, K. L., manuscript in preparation.
(23) (a) Zupan, M.; Pollak, A. J . Chem. Soc., Perkin Trans. 1 1976,
971. (b) Zupan, M. J . Fluorine Chem. 1977, 9, 177. (c) Gregorcic, A.;
Zupan, M. J . Fluorine Chem. 1984, 24, 291. (d) Gregorcic, A.; Zupan,
M. J . Org. Chem. 1984, 49, 333.
(27) Chehidi, I.; Chaabouni, M. M.; Baklouti, A. Tetrahedron Lett.
1989, 30, 3167.

Preparation of â-Fluorourocanic Acids
J . Org. Chem., Vol. 67, No. 10, 2002 3471
to room temperature, stirred for 2 h, and then evaporated to
dryness at room temperature. H NMR analysis of the crude
of 19a and 19b . Furthermore, we found no NMR evi-
dence for competing HF elimination. The reaction was
slower in DMF or Et3N and did not proceed in toluene
at 80 °C. Other bases examined, including K2CO3, DBU,
or t-BuOK, were less satisfactory, possibly because of
competing oxirane formation.27
1
product indicated that the molar ratio of 4:5:6:7:8 was 23:62:
8:4:3. This mixture was subjected to preparative TLC (9:1
petroleum ether/acetone) to give 115 mg (35%) of compound
5. P r oced u r e B. Using the same conditions as described in
procedure A, 304 mg (0.80 mmol) of 4, 0.40 mL (2.45 mmol) of
Et3N‚3HF, and 281 mg (1.58 mmol) of NBS gave a mixture of
three products. 1H NMR analysis of the crude product showed
the presence of 6, 7, and 8 in a molar ratio of 18:57:25. The
products were partially separated by filtration through 4 g of
silica. NMR and MS spectroscopic analysis of the mixtures
verified the structures of 6 and 8. In another experiment, 151
mg of 7 (37%) was isolated from the crude product by
preparative TLC (CH2Cl2).
The reoxidation of the alcohols to the carboxyl function
was done in two steps. The fluoro alcohols 19 were
conveniently oxidized to aldehydes 20 using activated
MnO2 in yields better than those reported28 for nonflu-
orinated analogues (60-90% vs 30%). However, when the
oxidation was carried out with pyridinium dichromate
in DMF, mild conditions that are reported to give no
isomerization29 of double bonds, we, in fact, observed
substantial isomerization. Thus, the oxidation of (E)-
fluoro alcohol 19a gives (E)-aldehyde 20a and (Z)-
aldehyde 20b in a ratio of 73:27. Under the same
conditions, the oxidation of (Z)-fluoro alcohol 19b gives
aldehydes 20a and 20b in a ratio of 2:98. In view of the
difficulty in preparing 15b caused by the loss of stereo-
selectivity in the FBr addition, this second loss of
stereoselectivity is actually quite useful and provided a
fortuitous alternative route to 2b. The aldehydes 20 are
fairly stable, but we recommend no long-term storage
because of the known7 air oxidation of 3-fluoro-3-phenyl-
propanal to 3-fluoro-3-phenyl-acrylic acid.
1
(E)-4-(2-Br om ovin yl)-1-tr ityl-1H-im id a zole (5). H and
13C NMR characteristics were identical to material we had
characterized previously.12 In addition, the signal for the
1
olefinic protons in H NMR we originally observed as a singlet
is resolved into two doublets at δ 7.01 (1H, d, 13.5) and δ 6.87
(1H, d, 13.5) when measured in a slightly acidic pH. These
interaction constants confirm the earlier assignment12 of the
(E)-configuration.
(2R*,3S*)-2-Br om o-3-flu or o-3-(1-t r it yl-1H -im id a zol-4-
yl)-p r op ion ic Acid (6). 1H NMR characteristic signals: δ 5.92
(1H, dd, 45.4, 6.3), 4.90 (1H, dd, 10.4, 6.3). HRMS FAB-: [M
- H]- calcd for C25H19BrFN2O2, 477.0614, 479.0593; found,
477.0627, 479.0623.
4-(2,2-Dibr om o-1-flu or oeth yl)-1-tr ityl-1H-im id a zole (7).
1H NMR: δ 7.44 (1H, br s), 7.37-7.32 (9H, m), 7.15-7.10 (6H,
m), 7.01 (1H, br s), 6.07 (1H, dd, 13.8, 5.7), 5.65 (1H, dd, 45.6,
5.7). 13C NMR: δ 141.93 (3C, off: m), 139.15 (off: dd, 211.4,
7.2), 135.38 (d, 24.8, off: m), 129.71 (6CH, off: dm), 128.27
(3CH, off: dt), 128.16 (6CH, off: dm), 121.82 (d, 5.2, off: dm,
192), 91.06 (d, 182.8, off: dd, 158.6, 182.8), 75.76 (off: m), 43.91
(d, 30.5, off: dd, 178.5, 30.8). 19F NMR: δ -168.8 (m, ΣJ 150).
HRMS FAB+: M+ calcd for C24H20Br2FN2, 512.9977, 514.9957,
516.9936; found, 512.9962, 514.9951, 516.9921.
4-(1,2,2-Tr ibr om o-et h yl)-1-t r it yl-1H-im id a zole (8). 1H
NMR only characteristic signals: δ 6.18 (1H, d, 6.8), 5.43 (1H,
d, 6.8), predicted30 chemical shifts 6.17 and 5.54. LRMS
FAB+: MH+ calcd for C24H20Br3N2, 573, 575, 577, and 579.
These peaks were observed but too low in intensity to record
the HRMS.
Ad d ition of F Br to Tr ityla ted Ur oca n ic Acid Meth yl
Ester 9a : P r oced u r e A. To 3.00 g (7.61 mmol) of 9a in 45
mL of CH2Cl2, cooled to 0 °C, was added 3.7 mL (22.7 mmol)
of Et3N‚3HF. After 10 min, 2.97 g (16.7 mmol) of NBS was
added in portions. The temperature was allowed to warm to
room temperature (2 h) and stirred for an additional 14 h. The
reaction mixture was poured into a mixture of 250 mL of CH2-
Cl2, 200 mL of water, and 10 mL of concentrated ammonia.
The water layer was extracted with 2 × 100 mL of water, and
the combined organic layers were washed with 3 × 150 mL of
brine and dried over MgSO4. Removal of solvent gave 4.35 g
of solid. This was separated by column chromatography (80
g, 95:5 CH2Cl2/Et2O) to give 0.32 g (9%) of 10, 1.93 g (54%) of
12a , and 0.35 g (9%) of 12b. P r oced u r e B. According to the
same conditions described in procedure A, to 1.00 g (2.54 mmol)
of 9a and 1.2 mL (7.36 mmol) of Et3N‚3HF in 15 mL of CH2-
Cl2 was added 0.91 g (5.11 mmol) of NBS. The reaction mixture
was kept for 10 min at 0 °C and for 4 h at room temperature.
In the second step, the aldehydes 20 were oxidized to
acids 21 with NaClO2. Final removal of trityl groups from
acids 21 produced â-fluorourocanic acids 2. The acid 2a
was isolated as the dihydrochloride monohydrate and
acid 2b as the dihydrochloride dihydrate. These “com-
plexes” are surprisingly stable and, even under extended
drying (100 Pa, 3 days, 40 °C), there was only partial
release of HCl and H2O, as shown by elemental analyses.
Con clu sion
We have completed a synthesis of â-fluorourocanic
acids using, as the key step, addition of FBr (Et3N‚3HF
with NBS) to a vinyl imidazole derivative. On the basis
of our results and literature reports, we make the
following generalizations. Addition of an FBr equivalent
to â-aryl acrylic acids is complicated by decarboxylation/
bromination. The addition of FBr to the N-protectected
imidazoyl acrylate esters is successful. In the next step,
however, the presence of the ester function causes
elimination of HF instead of HBr from the FBr adducts.
Chemoselectivity of the elimination can be changed by
prior reduction of the carboxyl to hydroxymethyl. Fol-
lowing elimination, reoxidation of the alcohol in two steps
gives the â-fluoroacrylates.
Exp er im en ta l Section
The NMR spectra were recorded at frequencies of 300.1 MHz
for 1H, 75.5 MHz for 13C, and 282.2 MHz for 19F spectra. The
solvent is CDCl3 unless otherwise specified. For other details,
see Supporting Information.
Ad d ition of F Br to Tr ityla ted Ur oca n ic Acid 4: P r o-
ced u r e A. To a cooled (0 °C) and stirred solution of 301 mg
(0.79 mmol) of 4 (prepared according to ref 18) dissolved in
CH2Cl2 was added 0.20 mL (1.23 mmol) of Et3N‚3HF. After
15 min, 154 mg (0.87 mmol) of NBS was added in portions.
After an additional 15 min, the mixture was allowed to come
(30) NMR prediction module (Upstream Solutions GmbH, Switzer-
land) of CS ChemDraw Ultra 6.0; CambridgeSoft Corporation: Cam-
bridge, MA.
(31) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(32) Pirrung, M. C.; Pei, T. J . Org. Chem. 2000, 65, 2229.
(33) Kosaka, K.; Maruyama, K.; Nakamura, H.; Ikeda, M. J .
Heterocycl. Chem. 1991, 28, 1941.
(34) Uenishi, J .; Motoyama, M.; Takahashi, K. Tetrahedron: Asym-
metry 1994, 5, 101.
(35) Sellier, C.; Buschauer, A.; Elz, S.; Schunack, W. Liebigs Ann.
Chem. 1992, 317.
(36) Bal, B. S.; Childers, W. E., J r.; Pinnick, H. W. Tetrahedron 1981,
37, 2091.
(28) Lee, K. J .; J oo, K. C.; Kim, E.-J .; Lee, M.; Kim, D. H. Bioorg.
Med. Chem. 1997, 5, 1989.
(29) Corey; E. J ., Schmidt; G. Tetrahedron Lett. 1979, 399.

3472 J . Org. Chem., Vol. 67, No. 10, 2002
Dolensky and Kirk
The mixture was quickly partitioned between 150 mL of CH2-
Cl2 and 150 mL of water containing 2 mL of concentrated
ammonia. The organic layer was washed with 150 mL of brine
and filtered through a pad of silica (5 g), and the solvent was
evaporated to dryness. The residue was separated by prepara-
tive TLC (98:2 CH2Cl2/acetone) to give 0.72 g (58%) of 10 as
an off-white solid. P r oced u r e C. Optimization experiments
were carried out using 129 mg (327 µmol) of 9a in 2 mL of
CH2Cl2 according to procedure A. The reaction mixtures were
evaporated to dryness, and compositions of product mixtures
were estimated by 1H NMR. The equivalents of reagents,
reaction times, and product compositions are given in Table
1.
compositions are given in Table 2. It should be noted that, for
estimation of product composition, the signals at δ 5.72 and
5.76 of the CHF groups cannot be used because their order of
appearance is often switched. This is caused by their high
sensitivity to the conditions under which the NMR measure-
ments are made.
(2S*,3R*)-2-Br om o-3-flu or o-3-(1-t r it yl-1H -im id a zol-4-
yl)-p r op a n -1-ol (15a ). 1H NMR: δ 7.46 (1H, d, 1.5), 7.38-
7.30 (9H, m), 7.16-7.10 (6H, m), 6.96 (1H, dd, 2.4, 1.5), 5.72
(1H, dd, 45.9, 5.4), 4.59 (1H, dddd, 8.1, 6.6, 5.4, 4.8), 4.49 (1H,
br s), 4.11 (1H, dd, 12.5, 6.4), 4.02 (1H, ddd, 12.5, 4.3, 1.5). 13
C
NMR: δ 141.85 (3C, off: m), 138.94 (off: dd, 211.4, 7.1), 135.96
(d, 23.8, off: dddt, 23.8, 10.9, 8.6, 3.1), 129.96 (6CH, off: dm),
128.23 (3CH, off: dt), 128.13 (6CH, off: dm), 121.83 (d, 5.7,
off: ddt, 192.8, 5.9, 3.3), 88.44 (d, 174.3, off: ddm, 174.1, 155.5,
ΣJ 12), 75.61 (off: m), 62.80 (d, 4.9, off: tm, 146.3, ΣJ 13),
54.71 (d, 25.9, off: dd, 152.5, 26.2). 19F NMR: δ -172.3 (dd,
45.6, 12.5). Mp: 142 °C. Anal. Calcd for C25H22BrFN2O: C,
64.52; H, 4.76; N, 6.02. Found: C, 64.61; H, 4.83; N, 6.06.
HRMS FAB+: MH+ calcd for C25H23BrFN2O, 465.0978,
467.0957; found, 465.0956, 467.0966.
(2R*,3S*)-2-Br om o-3-flu or o-3-(1-t r it yl-1H -im id a zol-4-
yl)-p r op a n oic Acid Meth yl Ester (10). 1H NMR: δ 7.48 (1H,
d, 1.2), 7.38-7.30 (9H, m), 7.16-7.09 (6H, m), 6.99 (1H, dd,
3.6, 1.2), 5.73 (1H, dd, 46.5, 9.9), 4.89 (1H, dd, 9.9, 6.0), 3.84
(3H, s). 13C NMR: δ 168.07, 141.76 (3C), 134.54 (d, 21.7),
129.09 (6CH), 128.09 (3CH), 127.98 (6CH), 122.69 (d, 6.3),
87.01 (d, 173.9), 75.94, 53.01, 42.93 (d, 36.5). 19F NMR:
δ
-157.0 (ddd, 46.7, 6.3, 3.8). HRMS FAB+: MH+ calcd for
C26H23BrFN2O2, 493.0927, 495.0906; found, 493.0935, 495.0931.
Mp: 135-145 °C dec.
(2R*,3R*)-2-Br om o-3-flu or o-3-(1-t r it yl-1H -im id a zol-4-
yl)-p r op a n -1-ol (15b). 1H NMR: δ 7.46 (1H, s), 7.36-7.30
(9H, m), 7.16-7.09 (6H, m), 6.98 (1H, s), 5.76 (1H, dd, 45.5,
4.6), 5.35 (1H, bs), 4.45 (1H, ddt, 18.0, 6.9, 4.8), 4.02 (1H, dd,
12.3, 4.8), 3.89 (1H, dd, 12.6, 6.9). 13C NMR: δ 141.79 (3C),
138.89, 136.25 (d, 25.1), 129.61 (6CH), 128.19 (3CH), 128.10
(6CH), 121.28 (d, 5.8), 87.94 (d, 175.0), 75.69, 62.66 (d, 2.8),
55.50 (d, 23.6). 19F NMR: δ -176.9 (dd, 45.8, 17.1).
Elim in a tion of HBr fr om F Br -Ad d u cts 15 to F lu or o-
a llyl Alcoh ols 19. A solution of 357 mg (0.77 mmol) of the
1:4 mixture of isomers 15a and 15b in 30 mL of dry DMSO
was treated with 5 mL of Et3N and stirred for 21 h at 110 °C.
After cooling to room temperature, the mixture was partitioned
between a mixture of 200 mL of brine and 200 mL of water
and extracted with 3 × 20 mL of CH2Cl2. The organic layers
were washed with 100 mL of a 10% aqueous solution of citric
acid and 100 mL of brine, dried over MgSO4, and evaporated
to dryness. Preparative TLC (4:1 CH2Cl2/Et2O) gave 46 mg
(78% from 15a ) of 19a and 190 mg (81% from 15b) of 19b.
The identical ratio of products (1:4) demonstrates that no
isomerization occurs during the elimination.
(E)-3-F lu or o-3-(1-tr ityl-1H-im id a zol-4-yl)-p r op -2-en -1-
ol (19a ). 1H NMR: δ 7.50 (1H, dd, 3.0, 1.5), 7.36-7.30 (9H,
m), 7.17-7.10 (6H, m), 7.06 (1H, t, 1.2), 5.96 (1H, br s), 5.63
(1H, dt, 22.8, 6.9), 4.32 (2H, dd, 6.9, 2.7). 13C NMR: δ 155.46
(d, 235.8, off: ddt, 235.6, 8.2, 5.1), 141.55 (3C), 139.27 (d, 2.0,
off: ddd, 212.1, 7.3, 1.7), 133.48 (d, 41.4, off: m), 129.44 (6CH),
128.13 (3CH), 128.04 (6CH), 119.93 (d, 1.0, off: dd, 194.7, 2.9),
106.09 (d, 18.5, off: ddt, 156.7, 18.4, 4.5), 75.70 (off: m), 55.34
(d, 11.6, tdd, 144.2, 12.5, 2.1). 19F NMR: δ -115.4 (dm, 22.9,
ΣJ 11). Anal. Calcd for C25H21FN2O: C, 78.10; H, 5.51; N, 7.29.
Found: C, 78.04; H, 5.46; N, 7.26. Rf (Et2O) ) 0.77. Mp: 133-
134 °C (from cyclohexane).
(Z)-2-Br om o-3-(1-tr ityl-1H-im id a zol-4-yl)-a cr ylic Acid
1
Meth yl Ester (12a ). H NMR: δ 8.31 (1H, d, 0.6), 7.93 (1H,
dd, 1.3, 0.6), 7.53 (1H, d, 1.3), 7.39-7.31 (9H, m), 7.19-7.11
(6H, m), 3.84 (3H, s). 13C NMR: δ 163.69 (off: dq, 5.0, 3.8),
141.74 (3C), 139.39 (off: dd, 211.5, 7.5), 135.88 (off: d, 157.5),
135.72 (off: ddd, 11.6, 8.7, 2.4), 129.61 (6CH), 128.29 (3CH),
128.17 (6CH), 125.72 (off: ddd, 185.6, 5.5, 3.3), 109.31 (off:
dd, 3.5, 1.0), 76.01, 53.24 (off: q, 147.6). Mp: 200-204.5 °C
(from cyclohexane). Anal. Calcd for C26H21BrN2O2: C, 65.97;
H, 4.47; N, 5.92. Found: C, 65.74; H, 4.61; N, 5.84. The (Z)-
1
configuration was assigned on the basis of predicted H NMR
chemical shifts30 and on analogy with NMR data of the ethyl
ester derivative.18
(E)-2-Br om o-3-(1-tr ityl-1H-im id a zol-4-yl)-a cr ylic Acid
1
Meth yl Ester (12b). H NMR: δ 7.57 (1H, dd, 1.4, 0.6), 7.43
(1H, d, 1.3), 7.36-7.31 (10H, m), 7.17-7.10 (6H, m), 3.70 (3H,
s). 13C NMR: δ 164.37 (off: dq, 10.9, 4.0), 141.87 (3C), 138.96
(off: dd, 211.1, 7.2), 135.53 (off: ddd, 11.6, 8.3, 2.0), 135.13
(off: d, 158.8), 129.61 (6CH), 128.13 (3CH), 128.06 (6CH),
124.79 (off: ddd, 195.7, 5.3, 3.5), 106.81 (off: d, 8.4), 75.68,
52.66 (off: q, 147.8). Mp: 178-185 °C (from chloroform). Anal.
Calcd for C26H21BrN2O2: C, 65.97; H, 4.47; N, 5.92. Found:
C, 65.89; H, 4.47; N, 5.91. The (E)-configuration was assigned
on the basis of predicted 1H NMR chemical shifts30 and on
analogy with NMR data of the ethyl ester derivative.18
(2R*,3S*)-2,3-Dibr om o-3-(1-tr ityl-1H-im id a zol-4-yl)-p r o-
p a n oic Acid Meth yl Ester (11). 1H NMR: δ 7.59 (1H, s),
7.38-7.32 (9H, m), 7.16-7.09 (6H, m), 6.92 (1H, s), 5.37 (1H,
d, 11.4), 5.17 (1H, d, 11.4), 3.86 (3H, s). HRMS FAB+: MH+
calcd for C26H23Br2N2O2, 553.0126, 555.0106, 557.0085; found,
553.0111, 555.0102, 557.0080.
Ad d ition of F Br to Allyla lcoh ols 14a a n d 14b: P r oce-
d u r e A. To 7.88 g (21.5 mmol) of allyl alcohol 14a dissolved
in 360 mL of CH2Cl2 was slowly added via syringe 5.4 mL (33.0
mmol) of Et3N‚3HF. After 10 min, 4.21 g (23.7 mmol) of NBS
was added in two portions and the mixture was stirred for 5
h. The reaction mixture then was washed with brine, dried
over MgSO4, and evaporated to dryness. The solid residue was
separated on column chromatography to give 6.87 g (69%) of
15a . P r oced u r e B. According to the same conditions (proce-
dure A), 1.20 g (3.27 mmol) of allyl alcohol 14b, 0.80 mL (4.91
mmol) of Et3N‚3HF, and 642 mg (3.61 mmol) of NBS in 66
mL of CH2Cl2 gave 1.30 g (74%) of a mixture of isomers 15a
and 15b in a molar ratio of 75:25. Column chromatography
effected only partial separation, and only enriched fractions
were obtained. P r oced u r e C. The optimization reactions were
carried out on 50 mg (0.14 mmol) of 14b in 2 mL of CH2Cl2.
The experiments were done as described for procedure A except
that Et3N was added before the addition of NBS. The reaction
mixtures were evaporated to dryness at room temperature,
and the product compositions were estimated on the basis of
1H and 19F NMR. The equivalents of reagents and product
(Z)-3-F lu or o-3-(1-tr ityl-1H-im id a zol-4-yl)-p r op -2-en -1-
ol (19b). 1H NMR: δ 7.44 (1H, dd, 3.1, 1.5), 7.35-7.29 (9H,
m), 7.16-7.08 (6H, m), 6.93 (1H, m, ΣJ 4), 5.81 (1H, dt, 38.1,
7.3), 4.33 (2H, dd, 7.4, 1.7), 3.53 (1H, bs). 13C NMR: δ 153.35
(d, 242.1), 141.82 (3C), 139.58 (d, 1.6), 133.87 (d, 39.6), 129.58
(6CH), 128.16 (3CH), 128.10 (6CH), 118.96 (d, 1.2), 103.52 (d,
11.3), 75.62, 55.03 (d, 6.9). 19F NMR: δ -122.1 (d, 38.4). Rf
(Et2O) ) 0.36. Mp: 150-152 °C (from petroleum ether).
Th e Oxid a tion of Allyl Alcoh ols 19 to P r op en a ls 20:
P r oced u r e A. To a stirred solution of 422 mg (1.10 mmol) of
allyl alcohol 19a in 30 mL of CH2Cl2 was added 400 mg (4.60
mmol) of activated MnO2. The mixture was stirred at room
temperature and monitored by TLC (9:1 CH2Cl2/Et2O). After
1 h, an additional 400 mg of activated MnO2 was added and
mixture was stirred overnight. The mixture was filtered
through a paper filter to remove most of the MnO2 and then
eluted through 5 g of silica with CH2Cl2. The crude product
was purified by column chromatography (60 g, 95:5 CH2Cl2/
Et2O) to give 336 mg (80%) of propenal 20a . Using an
analogous procedure, propenal 20b was obtained from 19b in
92% yield. P r oced u r e B. To a stirred solution of 26 mg (68

Preparation of â-Fluorourocanic Acids
J . Org. Chem., Vol. 67, No. 10, 2002 3473
µmol) of 19a in 2 mL of dry DMF was added 33 mg (88 µmol)
of pyridinium dichromate (PDC), and the mixture was stirred
for 3 days at room temperature. The mixture then was diluted
with 20 mL of CH2Cl2 and eluted through 4 g of silica with 10
mL of Et2O. After removal of the solvents, the NMR spectrum
of the residue was recorded. The mixture of aldehydes 20a and
20b was estimated to be 73:27. Oxidation of allyl alcohol 19b
using the same procedure gave 20a and 20b in a ratio of 2:98.
(E)-3-Flu or o-3-(1-tr ityl-1H-im idazol-4-yl)-pr open al (20a).
1H NMR: δ 10.85 (1H, ddd, 8.1, 2.4, 0.9), 7.53 (1H, dd, 3.0,
1.5), 7.41-7.35 (9H, m), 7.33 (1H, t, 1.4), 7.16-7.11 (6H, m),
5.82 (1H, dd, 21.0, 8.1). 13C NMR: δ 193.37 (d, 19.3, off: dd,
183.3, 20.2), 168.12 (d, 260.9, off: dd, 261.2, 7.4), 141.38 (3C),
140.87 (d, 2.3, off: ddd, 213.1, 9.1, 2.4), 132.71 (d, 37.4), 129.43
(6CH), 128.36 (3CH, off: dt, 161.6, 7.2), 128.21 (6CH), 124.50
(d, 3.2, off: dm, d 194.7), 109.99 (d, 19.5, off: ddd, 160.4, 26.3,
19.5), 76.03 (off: m). 19F NMR: δ -94.6 (d, 22.0). Rf (9:1 CH2-
Cl2/Et2O) ) 0.69. HRMS FAB+: MH+ calcd for C25H20FN2O,
383.1560; found, 383.1551. Mp: 162.5-164.0 °C. Anal. Calcd
for C25H19FN2O: C, 78.52; H, 5.01; N, 7.32. Found: C, 78.13;
H, 5.10; N, 7.34.
(Z)-3-Flu or o-3-(1-tr ityl-1H-im idazol-4-yl)-pr open al (20b).
1H NMR: δ 10.10 (1H, d, 8.1), 7.52 (1H, dd, 3.2, 1.4), 7.40-
7.34 (9H, m), 7.34 (1H, t, 0.9), 7.17-7.10 (6H, m), 6.21 (1H,
dd, 34.7, 8.1). 13C NMR: δ 188.16 (d, 11.3), 167.51 (d, 266.6),
141.47 (3C), 141.07 (d, 1.9), 132.04 (d, 34.9), 129.55 (6CH),
128.48 (3CH), 128.32 (6CH), 123.49 (d, 2.7), 105.34 (d, 1.3),
76.22. 19F NMR: δ -112.2 (dd, 34.7, 2.8). Rf (9:1 CH2Cl2/Et2O)
) 0.50. HRMS FAB+: MH+ calcd for C25H20FN2O, 383.1560;
found, 383.1568.
7.24 (1H, m, ΣJ 3.2), 7.16-7.09 (6H, m), 6.10 (1H, d, 37.5).
13C NMR: δ 168.56, 162.82 (d, 269.7), 141.53 (3C), 140.55,
132.76 (d, 35.7), 129.59 (6CH), 128.43 (3CH), 128.31 (6CH),
122.84 (d), 94.75, 76.17. The pure compound has very broad
signals in the NMR.
Detr ityla tion of Acid s 21 to Acid s 2. To a solution of 91
mg (228 µmol) of 21b in 5 mL of acetic acid was added 1 mL
of 12 N HCl, and the mixture was stirred for 1 h at room
temperature. The mixture then was evaporated to dryness at
40 °C, and the residual solid was triturated with 3 × 2 mL of
water. The aqueous extracts were combined, filtered, and
evaporated to dryness to give 46 mg (77%) of 2b‚2HCl‚2H2O
as a slightly yellow solid. By the same procedure was prepared
155 mg (94%) of 2a ‚2HCl‚H2O from 234 mg of 21a .
(E)-3-F lu or o-3-(1H-im id a zol-4-yl)-a cr ylic Acid Dih y-
1
d r och lor id e Hyd r a te (2a ). H NMR (CD3OD): δ 9.07 (1H,
s), 8.38 (1H, s), 6.21 (1H, d, 23.1). 1H NMR (CD3SOCD3, rt):
δ 15.0-4.5 (too broad to integrate), 8.97 (1H, s), 8.28 (1H, s),
6.08 (1H, d, 23.7). 1H NMR (CD3SOCD3, 50 °C): δ 11.2-6.6
(6H, br s), 8.78 (1H, s), 8.16 (1H, s), 6.01 (1H, d, 24.0). 1H NMR
(CD3SOCD3, 70 °C): δ 10.6-7.2 (6H, br s), 8.66 (1H, s), 8.09
(1H, s), 5.96 (1H, d, 24.0). 13C NMR: δ 168.55 (d, 21.6), 158.94
(d, 250.2), 137.20, 125.28 (d, 40.0), 124.46 (d, 5.3), 105.30 (d,
29.0). 19F NMR: δ -99.4 (d, 22.9). Mp: decomp. from 160 °C;
melting of residue from 190 to 195 °C. UV: λmax ) 269 nm, ꢀ
) 17,000 L mol-1 cm-1 (log ꢀ ) 4.2) in methanol. HRMS
DCI+: M+ calcd for C6H5FN2O2, 156.0335; found, 156.0334.
Anal. Calcd C6H9Cl2FN2O3: C, 29.17; H, 3.67; N, 11.34; Cl,
28.70. Found: C, 29.22; H, 3.24; N, 11.24; Cl, 22.22.
Oxid a tion of P r op en a ls 20 to Acid s 21. The propenal 20a
(334 mg, 0.87 mmol) was suspended in 21 mL of t-BuOH, and
4.5 mL of isobutylene was added. The stirred mixture was
warmed to 30 °C, and a solution of 740 mg (8.18 mmol) of
NaClO2 and 736 mg (6.13 mmol) of NaH2PO4 in 7.4 mL of
water was slowly added (2 min). The reaction was followed by
TLC (9:1 CH2Cl2/Et2O). After the disappearance of 20a (3 h),
the mixture was poured into a mixture of 10 mL of brine, 10
mL of 10% aqueous citric acid, and 10 mL of CH2Cl2. The
aqueous layer was extracted with 3 × 10 mL of CH2Cl2. The
organic layers were combined, dried over MgSO4, and evapo-
rated to dryness. The residual solid was separated by prepara-
tive TLC (49:1 CH2Cl2/CH3OH) to give 235 mg (68%) of acid
21a . Oxidation of propenal 20b (124 mg, 0.32 mmol) using the
same procedure (1:1 CH2Cl2/Et2O was used for preparative
TLC) gave 21b in yield 91 mg (71%).
(Z)-3-F lu or o-3-(1H-im id a zol-4-yl)-a cr ylic Acid Dih y-
d r och lor id e Dih yd r a te (2b). 1H NMR (CD3OD): δ 9.09 (1H,
t, 1.4), 8.12 (1H, d, 1.3), 6.12 (1H, d, 33.9). 1H NMR (CD3-
SOCD3, rt): δ from 15.5 to -1.1 (too broad to integrate), 8.85
1
(1H, s), 8.08 (1H, s), 6.24 (1H, d, 35.7). H NMR (CD3SOCD3,
50 °C): δ 9.0-3.2 (8H, br s), 8.61 (1H, s), 7.95 (1H, s), 6.17
(1H, d, 35.7). 1H NMR (CD3SOCD3, 70 °C): δ 8.44 (1H, s), 8.0-
4.2 (8H, br s), 7.86 (1H, s), 6.12 (1H, d, 35.7). 13C NMR: δ
165.81 (d, 1.5), 156.42 (d, 268.3), 138.22, 127.29 (d, 39.3),
121.50 (d, 4.0), 101.52 (d, 4.0). 19F NMR: δ -103.3 (dd, 34.0,
1.5). Mp: decomp. from 180 °C; no melting of residue to 250
°C. UV: λmax ) 270 nm, ꢀ ) 29 000 L mol-1 cm-1 (log ꢀ ) 4.5)
in methanol. HRMS DCI+: M+ calcd for C6H5FN2O2, 156.0335;
found, 156.0336. Anal. Calcd C6H11Cl2FN2O4: C, 27.19; H, 4.18;
N, 10.57. Found: C, 26.49; H, 4.01; N, 9.94.
(E)-3-F lu or o-3-(1-tr ityl-1H-im id a zol-4-yl)-a cr ylic Acid
(21a ). 1H NMR: δ 15.96 (1H, br s), 7.66 (1H, dd, 3.0, 1.2),
7.43-7.38 (9H, m), 7.26 (1H, t, 1.2), 7.14-7.08 (6H, m), 5.89
(1H, d, 25.8). 13C NMR: δ 165.55 (d, 22.5), 159.20 (d, 250.7),
140.89 (3C), 138.50 (d, 2.1), 131.17 (d, 41.4), 129.52 (6CH),
128.84 (3CH), 128.55 (6CH), 122.89 (d, 1.9), 103.98 (d, 26.4),
77.00. 19F NMR: δ -104.2 (dd, 25.8, 2.5). Mp: 194-220 °C dec.
(Z)-3-F lu or o-3-(1-tr ityl-1H-im id a zol-4-yl)-a cr ylic Acid
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal conditions, NMR characteristics of certain nonfluorinated
intermediates, Table 2 describing the effects of reaction
conditions on the diastereoselectivity of FBr addition, and
Scheme 4. This material is available free of charge via the
Internet at http://pubs.acs.org.
1
(21b). H NMR: δ 7.54 (1H, dd, 3.0, 1.4), 7.38-7.33 (9H, m),
J O0200419