New approaches to side-chain fluorinated bioimidazoles: 4-alkynylimidazoles as substrates for fluorination

Article abstract of DOI:10.1016/S0022-1139(03)00197-0

4-Alkynylimidazoles have been prepared and their behavior as substrates for "FBr" addition (NBS+Et₃N·3HF) have been studied. Facile Markownikov addition to ethylnylimidazole 2 gave fluorobromoolefin 9a, a potential synthon for the 2-imidazolyl-

Full text of DOI:10.1016/S0022-1139(03)00197-0

Journal of Fluorine Chemistry 124 (2003) 105–110
New approaches to side-chain fluorinated bioimidazoles:
4-alkynylimidazoles as substrates for fluorination
Bohumil Dolensky, Kenneth L. Kirk*
Laboratory of Bioorganic Chemistry, National Institute of Diabetes, and Digestive and Kidney Diseases,
National Institutes of Health, DHHS, Bethesda, MD 20892, USA
Received 24 June 2003; received in revised form 16 July 2003; accepted 17 July 2003
Abstract
4-Alkynylimidazoles have been prepared and their behavior as substrates for ‘‘FBr’’ addition (NBS þ Et3NÁ3HF) have been studied. Facile
Markownikov addition to ethylnylimidazole 2 gave fluorobromoolefin 9a, a potential synthon for the 2-imidazolyl-2-fluoro-1-ethenyl moiety.
Reaction of the lithium salt of 2 with diethyl oxalate produced imidazolylpropynoic ester 3b. However, because of the deactivating effect of
the ester functionality, all attempts to carry out addition of ‘‘FBr’’ to 3b met with failure. Reduction of the ester gave the hydroxymethyl-
substituted acetylene 4. Addition of ‘‘FBr’’ to this substrate and reductive removal of bromine from the product produced fluoroolefins 12,
precursors to E- and Z-b-fluorourocanic acids. The same fluoroolefins can be used as intermediates in the synthesis of b-fluorohistidinols.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Alkynes; ‘‘FBr’’ addition; Urocanic acid
1. Introduction
E-fluoroolefin. This complicated the synthesis of Z-b-fluoro-
urocanic acid (1a). During the development of this synthesis,
we had considered different approaches that would be more
convenient for the synthesis of 1a and had made significant
progress to that end. This work was based on preparation and
functional manipulation of alkynylimidazoles and involved
the preparation of several imidazole intermediates that have
potential for serving as synthons for additional targets. We
herein report the details of the chemistry of alkynylimida-
zoles that was developed and the successful completion of
this alternative approach to acids 1a and 1b.
Interest in urocanic acid and derivatives has increased
with the discovery that Z-urocanic, formed by photoisome-
rization of the initially formed E-isomer in the epidermis,
functions systemically as an immunosupressive agent. We
have had a long-standing interest in the chemistry and
biology of fluorinated imidazole derivatives, including
fluorinated analogues of urocanic acid. Thus, to study the
possible consequences of side-chain fluorination on such
properties as photoisomerization, we prepared Z- and E-a-
fluorourocanic acids and we more recently have described
preparation of Z- and E-b-fluorourocanic acid (1a and 1b)
[1]. The key to this latter synthesis was the addition of
an FBr equivalent to the double bond of Z- and E-isomers
of 1-trityl-4-(3-hydroxy-1-propenyl)imidazoles. Subsequent
elimination of HBr gave fluoroolefins that were oxidized
to the urocanic acid derivatives. A complicating factor in
this approach was undesired isomerization during ‘‘FBr’’
addition to the Z-isomer of the olefin. For this reason,
dehydrobromination of this adduct gave predominantly the
E-fluoroolefin rather than the expected Z-fluoroolefin. Since
no similar loss of stereochemistry occurred in the sequence
starting with the E-starting olefin, this also produced the
2. Chemistry
2.1. Preparation of alkynylimidazoles
There are only limited reports of preparation of 4-alky-
nylimidazoles in the literature [2]. For the purposes of these
studies, we prepared the four alkynylimidazoles 2, 3a, 3b
and 4. Ethynylimidazole 2 was prepared by addition of
* Corresponding author. Tel.: þ1-301-496-2610; fax: þ1-301-402-4182.
E-mail address: kennethk@bdg8.niddk.nih.gov (K.L. Kirk).
0022-1139/$ – see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-1139(03)00197-0

106
B. Dolensky, K.L. Kirk / Journal of Fluorine Chemistry 124 (2003) 105–110
Br
Br
Br
t-BuOK
Br2
t-BuOK
84%
TrN
N
TrN
TrN
N
TrN
TrN
N
N
6
5
CHO
F
1. BuLi
2. DMF
"FBr"
63%
N
Br
+
2
TrN
N
9a
2
10
Scheme 1. Preparation of alkynylimidazoles from vinylimidazole.
COOMe
COOMe
COOMe
Br
1. Br2
2. Et3N
a) LDA
N
TrN
TrN
a) 10%
b) 14%
N
or b) KF
TrN
N
64% 8a (E)
9% 8b (Z)
3a
Scheme 2. Preparation of alkynyl ester 3a from methyl urocanate.
bromine to the double bond of 1-trityl-4-vinylimidazole [3]
followed by elimination of two molecules of HBr from
bromine addition product 5 using t-BuOK. This approach
worked very well and ethynylimidazole 2 was obtained in an
overall yield of 84%. In addition, when insufficient excess of
t-BuOK was used, the intermediate a-bromovinylimidazole
6 could be isolated (Scheme 1).
We tried a similar approach for the preparation of methyl
ester of the 1-trityl-4-imidazoyl propynoic acid (3a)
(Scheme 2). The addition of bromine to tritylated urocanic
acid methyl ester takes place with no problem and the adduct
7 is formed in almost quantitative yield as a single diaster-
eoisomer. The first molecule of HBr is eliminated readily by
treatment with Et3N to produce isomers 8a and 8b in a ratio
of 88:12. The major isomer 8a has the expected E config-
uration that would be predicted from anti addition of Br2 and
anti elimination of HBr. (This contrasts to the observed syn
elimination of HF from the analogous FBr-adduct [1].)
However, we met serious problems with elimination of
the second HBr molecule. Use of bases such as t-BuOK
or KOH/MeOH was unsuccessful, possibly because of
nucleophilic attack on the ester carbonyl and/or triple bond,
as suggested by crude NMR data. Warming in Et3N causes
only isomerization of 8a to 8b. Use of DBU also was
unsuccessful. About 10% of alkynylimidazole 3a is formed
from 8a or 8b by treatment with LDA. A longer reaction
time or bigger excess of LDA resulted in degradation of both
starting bromoolefins 8 and alkynylimidazole 3a.
It is also known that KF treatment can lead to HBr
elimination from bromoethylenes [4]. However, in the lit-
erature we found no examples of this transformation wherein
the substrate also contained an ester function. In our case, we
found that treatment of bromoolefin 8a with KF in DMF
gives three products: isomer 8b, the desired alkynylimidazol
3a and, surprisingly, also ethynylimidazole 2. The product
ratios depend on reaction time and temperature. In general, a
longer reaction time or higher temperature leads to a higher
yield of ethynylimidazole 2 (Table 1, see Section 4) that
suggests that formation of 2 is a result of decomposition of
alkynylimidazole 3a. Unfortunately, it is impossible to force
Table 1
Conditions vs. results for dehydrobromination of 8
Starting
compound
Reaction conditions
Product ratios
KF (eq.)
Temperature (8C)
Time (h)
8a
8b
3a
2
8a
8b
8a
8a
8b
8b
2
2
125
125
80
15
15
40
45
14
64
47
Traces
26
44
84
73
56
63
9
9
16
1
Traces
Traces
0
10
11
9
130
140
140
32
Traces
Traces
11
16
1
Traces
21
10
90
The preparative yield of 2 is about 25%.

B. Dolensky, K.L. Kirk / Journal of Fluorine Chemistry 124 (2003) 105–110
CH2OH
107
COOEt
1. BuLi
2. ClCO2Et
DIBAL
61%
2
TrN
N
71%
TrN
N
4
3b
F
CH2OH
CH2OH
F
ref. 1
Bu3SnH
54%
"FBr"
50%
1a
+
1b
Br
N
TrN
TrN
12a + 12b 2:3
Scheme 3. Preparation of fluoroolefins 12a and 12b from ethynylimidazole 2.
N
11a
the reaction to full conversion to 2, and the separation of 3a
from 2 is difficult. Thus, this approach is not suitable for the
preparation of either 2 or 3a. This process nonetheless is
intriguing because reported formation of propynes by dec-
arbomethoxylation of methyl propynoates occurs only under
condition of vacuum flash pyrolysis at 750 8C [5]. The fact
that we observed no loss of the carbomethoxy group when
alkynylimidazole 3b was treated with KF suggests that HBr
produced in an initial step may play an important role in loss
of the carbomethoxy group.
With these largely negative results, we abandoned
attempts to carry out direct conversion of N-tritylurocanate
esters to the corresponding propynoates. Having a conve-
nient method of preparation of ethynylimidazole 2 in hand,
we used this as our source of imidazoylpropynoates. Treat-
ment of 2 with butyllithium followed by reaction of the
lithium salt with ethyl chloroformate produced 3b in a yield
of 71%. Reduction of ethyl ester 3b with DIBAL-H pro-
duced the corresponding alcohol 4 (Scheme 3).
combines to block addition. This low reactivity can be
overcome via using the authentic mixed halogen, FBr [8].
In contrast to the deactivated acetylenic ester 3b, ‘‘FBr’’
addition to ethynylimidazole 2 occurs readily to produce the
E adduct 9a and Z adduct 9b in a ratio of 96:4 (Scheme 1).
We had hoped that lithiation of 9a with BuLi followed by
treatment with electrophiles (CO2 or DMF) would produce
the a-bromo-b-fluoropropenoic acid or aldehyde [9]. Sub-
sequent reductive elimination of bromine would be provide a
route to our target compounds. However, such reactions of
bromofluoroalkene 9a were unsuccessful. For example, the
reaction of 9a with t-BuLi followed by addition of DMF to
the reaction mixture resulted in loss of both fluorine and
bromine, and formation of aldehyde 10 and ethynylimida-
zole 2 (Scheme 1). Even under very short reaction times
there were no traces of fluorinated compounds (19F NMR),
suggesting that no fluorinated intermediates capable of
being intercepted are being formed. We are currently explor-
ing substitution of bromine via transition metal catalyzed
reactions of 9a as an approach to the synthesis of imidazole-
containing structures linked by the ethylene unit.
2.2. Introduction of fluorine
Turning next to a substrate that contains the necessary
carbon skeleton, but is not deactivated by the ester function,
we examined ‘‘FBr’’ addition to acetylenic alcohol 4. We
were pleased to find that addition occurs readily to give 11a,
the expected product of trans ‘‘FBr’’ addition, in 50% yield
(Scheme 3). The bromine in 11a was removed by Bu3SnH
reduction [10] to give a 2:3 mixture of the fluoropropenols
12a [E-isomer] and 12b [Z-isomer]. These are the same
fluoropropenols we had prepared in our previous synthesis
of b-fluorourocanic acids 1 [1]. Thus, the fluoropropenols
12a and 12b can be oxidized and deprotected to give the
corresponding isomers of 1, as we demonstrated in our
previous work. The lack of stereochemical control in the
reductive removal of bromine is actually very welcome
because the isomers of 12 can be easily separated by
chromatography and both isomers of b-fluorourocanic acids
1 thus can be prepared from a single precursor. In addition,
the fluoropropenols 12 were key intermediates in the synth-
esis of b-fluoro- and b,b-difluoro-histidinols [11].
The most direct route to the target acids 1 would consist
of HF addition to the triple bond of esters 3. Indeed, facile
HF addition to the triple bond of phenyl analogues by
À
Bu4NþH2F3 gives the corresponding b-fluorocinnamates
in high yields [6]. However, in the case of ester 3b we
observed destruction of starting material using published
conditions [6] (120 8C, 4 h), and no reaction occurred at a
lower reaction temperature (80 8C, 3 h).
We also explored the addition of an FBr equivalent to
the triple bond of ester 3b as a second approach to fluorine
introduction. This addition to acetylenic compounds is
known and typically gives products resulting from anti
addition [7]. Reductive removal of bromine would provide
a facile route to the target compounds. However, attempted
‘‘FBr’’ addition to ester 3b gives no fluorinated products
(19F NMR). The lower reactivity of triple bonds in electro-
philic reactions compared to olefins together with the elec-
tron-withdrawing effects of the carbonyl group obviously

108
B. Dolensky, K.L. Kirk / Journal of Fluorine Chemistry 124 (2003) 105–110
3. Summary
4.2. 4-(Ethynyl)-1-trityl-1H-imidazole (2)
1
We have developed an alternative approach to the pre-
paration of E- and Z-isomers of b-fluorourocanic acid. One
advantage over the previous method [1] is that this approach
overcomes the problem we experienced related to selectivity
of ‘‘FBr’’ addition to the cis isomers of alkenes. Although
this synthetic pathway has more steps than the previous one
[1], the effort extended to make both isomers actually is less.
This is because the same readily prepared starting compound
can be used to make both isomers. From a practical point of
view, we feel this is the superior approach. In addition, the
work represents an alternative approach to other biologically
important imidazoles, including b-fluorohistidinols. Finally,
intermediates prepared in this work, such as the bromo-
fluoroolefin 9a, hold promise for additional synthetic appli-
cations. We are exploring these possibilities.
Melting point 186.5–187.0 8C. H NMR: 7.40 (1H, d,
1.5), 7.36–7.30 (9H, m), 7.16–7.09 (6H, m), 7.08 (1H, d,
P
1.5), 3.03 (1H, s). 13C NMR: 141.76 (3C, JCH 18), 138.85
(JCH 211.8 d, 7.1 d, C2imi), 129.58 (6CH), 128.16 (3CH),
128.08 (6CH), 126.26 (JCH 195.0 d, 8 m, C5imi), 122.27
(C4imi), 77.45 (JCH 252.1 d, BCH), 77.44 (JCH 50.5 d, –CB),
75.66 (Tr). Anal. Calcd. for C24H18N2: C, 86.20; H, 5.42; N,
8.38. Found: C, 85.79; H 5.60; N, 8.34.
4.3. 4-(1,2-Dibromoethyl)-1-trityl-1H-imidazole (5)
1H NMR: 7.48 (1H, d, 1.5), 7.36–7.31 (9H, m), 7.16–7.10
(6H, m), 6.89 (1H, d, 1.2), 5.17 (1H, dd, 10.8, 4.8), 4.24 (1H,
dd, 10.8, 9.9), 3.99 (1H, dd, 9.9, 4.8). 13C NMR: 141.82
(3C), 139.34 (C2imi), 137.96 (C4imi), 129.65 (6CH), 128.19
(3CH), 128.08 (6CH), 120.61 (C5imi), 75.66 (Tr), 44.54 (JCH
160.7 d, CHBr), 34.40 (JCH 158.4 t, CH2Br). HRMS FABþ
Calcd. for MHþ e.i. C24H21N2Br2: 495.0071, 497.0053,
499.0037. Found: 495.0073, 497.0049, 499.0053.
4. Experimental
The NMR spectra were recorded at frequencies of
1
300.1 MHz for H, 75.5 MHz for 13C and 282.2 MHz for
4.4. 4-(1-Bromovinyl)-1-trityl-1H-imidazole (6)
19F spectra. The solvent is CDCl3 unless otherwise specified.
On selected carbons the interaction constants with hydrogen
(JCH), important for assignment, are given.
1H NMR: 7.43 (1H, d, 1.5), 7.37–7.31 (9H, m), 7.18–7.11
(6H, m), 7.04 (1H, d, 1.5), 6.50 (1H, d, 1.5), 5.45 (1H, d, 1.8).
13C NMR: 141.98 (3C), 139.59 (JCH 210.6 d, 7.5 d, C2imi),
139.57 (C4imi), 129.67 (6CH), 128.19 (3CH), 128.13 (6CH),
121.87 (JCH 193.5 d, 3.3 d, C5imi), 121.52 (JCH 5.5 t, CBr),
114.15 (JCH 166.8 d, 160.5 d, CH2), 75.60 (Tr).
4.1. Preparation 2, 5 and 6
4.1.1. Procedure A
To a stirred solution of 250 mg (0.74 mmol) of vinylimi-
dazole [3] in 8 ml of CH2Cl2 was added drop wise 38 ml
(0.74 mmol) of bromine at 0 8C. After 15 min the light
yellow solution was evaporated to dryness and crude dibro-
moimidazole 5 was obtained in quantitative yield. Dibromide
5 was suspended in a mixture of 20 ml of petroleum ether and
10 ml of 2-propanol and 502 mg (4.47 mol, 6 eq.) of t-BuOK
was added. After 20 h of stirring at room temperature the
mixture was evaporated to dryness and the resulting solid was
partitioned between 20 ml of CH2Cl2 and a 1N aqueous
solution of citric acid and brine. The organic portion was
dried over MgSO4 and evaporated to dryness. The solid that
was obtained (287 mg) was separated by preparative TLC to
give 48 mg (16%) of bromovinylimidazole 6 and 96 mg
(39%) of ethynylimidazole 2.
4.5. Preparation of 7, 8a and 8b
To a solution of 5.67 g (14.4 mmol) of methyl urocanate
[1] in 60 ml of CH2Cl2 was adding solution of bromine in
CH2Cl2 (3 g in 10 ml) until an orange color persisted for
5 min. A small sample of the reaction mixture was evapo-
1
rated to dryness to give a solid. The H NMR of the solid
showed that it was essentially pure dibromo adduct 7. The
sample was returned to the reaction mixture and 6 ml
(43.0 mmol) of Et3N was added. The mixture was stirred
at room temperature for 3 days and then evaporated to
dryness. Separation by column chromatography (100 g,
CH2Cl2:Et2O 1:0 ! 8:2) gave 4.37 g (64%) of pure 8a
(E) and 0.58 g (9%) of pure 8b (Z). All characteristics of
compounds 7, 8a and 8b were identical with values published
previously [1].
4.1.2. Procedure B
To a stirred solution of 250 mg (0.74 mmol) vinylimida-
zole [3] in 8 ml of CH2Cl2 was added by drop wise 38 ml
(0.74 mmol) of bromine at 0 8C. After 15 min the light
yellow solution was evaporated to dryness to give crude
dibromoimidazole 5 in quantitative yield. Crude dibromide
5 suspended in a mixture of petroleum ether and isopropyl
alcohol was treated with 6 eq. of t-BuOK at 40 8C for 3 h.
The ethynylimidazole 2 was obtained in 84% yield as the
only product.
4.6. (1-Trityl-1H-imidazol-4-yl)-propynoic acid
methyl ester (3a)
4.6.1. Procedure A
To a solution of 100 mg (211 mmol) of 8a in 5 ml of THF
was added 105 ml (210 mmol) of LDA solution (2 M in
heptane:THF:EtPh) at À68 8C. After 1 h, the mixture was
allowed to warm to room temperature and after an additional

B. Dolensky, K.L. Kirk / Journal of Fluorine Chemistry 124 (2003) 105–110
109
1 h evaporated to dryness. The resulting solid was parti-
tioned between CH2Cl2 and brine. The organic layer was
dried over MgSO4 and evaporated to dryness. The resulting
solid was subjected to repeat preparative TLC to give a low
yield (>10%) of 3a.
(1H, d, 1.4), 4.45 (2H, br d, 4.9, CH2), 2.00 (1H, br t, 4.9,
OH).
4.9. (Z)-4-(2-Bromo-1-fluoroethenyl)-1-trityl-1H-
imidazole (9a)
4.6.2. Procedure B
To a stirred solution of 999 mg (2.99 mmol) of ethyny-
limidazole 2 and 750 ml (4.60 mmol) of Et3NÁ3HF in 20 ml
of CH2Cl2 was added 585 mg (3.28 mmol) of NBS at 0 8C.
After 30 min, the mixture was allowed to come to room
temperature and stirred an additional 2 h. The mixture was
then partitioned between CH2Cl2 and water/brine. The
organic fraction was dried over MgSO4 and evaporated to
dryness. The solid that was obtained (1.233 g) was separated
by column chromatography (100 g silica gel, CH2Cl2 !
CH2Cl2:Et2O 100:1) and to give 817 mg (63%) of Z-fluoro-
bromoimidazole 9a, 99 mg (10%) of starting ethynyli-
midazole 2, and 37 mg (3%) of E-fluorobromoimidazole
9b3. Attempts to recrystalization of 9a from methanol led to
loss of trityl group to give unprotected derivative 13 together
with tritylmethylether. The unprotected derivative 13 was
re-tritylated to 9a under the usual conditions.
To 854 mg (1.80 mmol) of 8b dissolved in 25 ml of dry
DMF was added 960 mg (16.5 mmol) of KF. The mixture
was stirred at 140 8C for 15 h. Solvent was removed by
evaporation and the residue was extracted with CH2Cl2. 1H
NMR indicated three compounds, 8b, 3a and 2, in a ratio of
63:16:21. By repeated column chromatography (100 g,
silica gel, petroleum ether:EtOAc 6:4) there was obtained
3a in low yield (<15%). (For effects on variation of the
reaction conditions on the dehydrobromination of 8a and 8b
1
see Table 1.) H NMR: 7.45 (1H, d, 1.2), 7.39–7.27 (10H,
m), 7.16–7.10 (6H, m), 3.78 (3H, s). 13C NMR: 154.42 (CO),
141.51 (3C), 139.87 (C2imi), 129.97 (C5imi), 129.60 (6CH),
128.42 (3CH), 128.27 (6CH), 120.32 (C4imi) 81.51 (CC),
81.36 (CC), 76.15 (Tr), 52.60 (CH3). HRMS DCIþ Calcd.
for Mþ e.i. C26H20N2O2: 392.1525. Found: 392.1518.
4.7. (1-Trityl-1H-imidazol-4-yl)-propynoic
acid ethyl ester (3b)
4.10. (Z)-4-(2-Bromo-1-fluoroethenyl)-1-trityl-1H-
imidazole (9a)
1
To a stirred solution of ethynylimidazole 2 (0.75 g,
2.24 mmol) in 60 ml of THF at À65 8C was added a solution
of n-BuLi (1.6 M in hexane, 2.1 ml, 3.36 mmol). After
30 min ethyl chloroformate (0.33 ml, 3.45 mmol) was added
and the mixture was allowed to warm to room temperature
and stirred overnight. The reaction mixture was evaporated
to dryness and the resulting solid was separated by column
chromatography (10 g silica gel, CH2Cl2:Et2O 9:1) to give
0.65 g (71%) of 3b. Melting point 212–215 8C (from petrol
Melting point 139 8C (decomp.). H NMR: 7.53 (1H, t,
1.4), 7.45 (1H, d, 1.5), 7.36–7.32 (9H, m), 7.17–7.10 (6H,
m), 6.06 (1H, d, 15.4). 13C NMR: 153.78 (d, 243.0, CF),
141.78 (3C), 139.13 (d, 0.9, C2imi), 131.66 (d, 33.1, C4imi),
129.60 (6CH), 128.23 (3CH), 128.13 (6CH), 123.34 (d, 5.9,
C5imi), 84.73 (d, 51.7, CHBr), 75.82 (Tr). 19F NMR: À107.5
(d, 15.4). Anal. Calcd. for C24H18BrFN2: C, 66.52; H, 4.19;
N, 6.46. Found: C, 67.30; H, 4.18; N, 6.49.
1
ether). H NMR: 7.45 (1H, d, 1.4), 7.38–7.28 (10H, m),
4.11. (Z)-4-(2-Bromo-1-fluoroethenyl)-1H-imidazole (13)
7.16–7.09 (6H, m), 4.24 (2H, q, 7.2), 1.30 (3H, t, 7.2). 13C
NMR: 153.93 (CO), 141.50 (3C, Tr), 139.78 (C2imi), 129.81
(C5imi), 129.55 (6CH), 128.37 (3CH), 128.22 (6CH), 120.37
(C4imi), 81.65 (C¼), 80.95 (C¼), 76.08 (Tr), 61.77 (CH2),
13.99 (CH3). HRMS DCIþ Calcd. for Mþ e.i. C27H22N2O2:
406.1681. Found: 406.1669.
1H NMR (CD3OD): 7.82 (1H, s), 7.79 (1H, s), 6.30
(1H, d, 15.0).
4.12. 3-(1-Trityl-1H-imidazol-4-yl)propynal (10)
Bromofluoroolefin 9a (179 mg, 413 mmol) was dissolved
in 16 ml of dry THF and chilled to À65 8C and 1.7 M t-BuLi
in pentane (0.41 ml, 697 mmol) was added dropwise. After
5 min 0.6 ml of DMF (7.75 mmol) in 2 ml of dry THF was
added. After 10 min was mixture allowed to warm to room
temperature and then kept at this temperature for 1 h. The
reaction mixture was partitioned between water and CH2Cl2,
the organic layer part was separated and evaporated to
dryness. The residue was subjected to preparative TLC
(CH2Cl2:Et2O 95:5) to give 60 mg (34%) of starting olefin
9a and 79 mg of a mixture of ethynylimidazole 2 and
aldehyde 10 in ratio of about 1:1. This mixture was separated
on by a second preparative TLC (CH2Cl2:Et2O 95:5) to give
4.8. 3-(1-Trityl-1H-imidazol-4-yl)-prop-2-yn-1-ol (4)
To solution of 390 mg of 3b (0.96 mmol) in 60 ml of
CH2Cl2 was added 5 ml of DIBAL-H (1.0 M in THF,
5 mmol) at À65 8C. The reaction was stirred for 4 h at the
same temperature. Then 20 ml of aqueous solution NH4Cl
was added, the mixture was allowed to warm to room
temperature and was partitioned between CH2Cl2 and water
with brine. The organic fraction was dried over MgSO4
and evaporated to dryness. The resulting solid was separated
by column chromatography (20 g silica gel, CH2Cl2:Et2O
99:1 ! 0:1) to give 215 mg (61%) of propynol 4. 1H NMR:
7.40 (1H, d, 1.4), 7.37–7.31 (9H, m), 7.15–7.09 (6H, m), 7.02
1
a low yield (<20% based on H NMR) of aldehyde 10.

110
B. Dolensky, K.L. Kirk / Journal of Fluorine Chemistry 124 (2003) 105–110
1H NMR: 9.36 (1H, s), 7.50 (1H, d, 1.5), 7.40–7.32 (10H,
m), 7.14–7.07 (6H, m). 13C NMR: 176.27 (JCH 192.7 d,
CHO), 141.40 (3C, JCH 17 m), 140.40 (JCH 213.4 d, 7.3 d,
C2Imi), 130.86 (JCH 190.8 d, 3.7 d, C5Imi), 129.55 (6CH),
128.40 (3CH), 128.30 (6CH), 120.16 (JCH 13.0 d, 8.5 d),
90.48 (JCH 5.7 d), 90.33 (JCH 32.3 d), 76.36. HRMS DCIþ
Calcd. for Mþ e.i. C25H18N2O: 362.1419. Found: 362.1425.
110 8C, after which time the NMR spectrum indicated full
conversion of 11a (and formed 11b). After evaporation to
dryness there was obtained 11 mg (21%) of 12a and 17 mg
(33%) of 12b, isolated by preparative TLC (CH2Cl2:Et2O
7:3). Physical and spectral data of 12a and 12b were
identical to the values for 19a and 19b, respectively, pre-
pared and reported in our previously published work [1].
Evidence for the formation of the Z-isomer 11b is based only
on a signal at À108.0 ppm (t, 4.5) in 19F NMR spectrum.
4.13. E-2-Bromo-3-fluoro-3-(1-trityl-1H-imidazol-4-yl)-
prop-2-en-1-ol (11a)
References
The same procedure described for the preparation of 9
was used, employing 100 mg (274 mmol) of alcohol 4, 70 ml
(429 mmol) of Et3NÁ3HF, 54 mg (303 mmol) of NBS and
4 ml of CH2Cl2. The crude product was separated on pre-
parative TLC (CH2Cl2:Et2O 3:2) to give the FBr-adduct 11a
(63 mg, 50%) as a waxy white solid. No formation of Z-
[1] B. Dolensky, K.L. Kirk, J. Org. Chem. 67 (2002) 3468–3473.
[2] (a) T. Sakamoto, H. Nagata, Y. Kondo, M. Shiraiwa, M.H.
Yamanaka, Chem. Pharm. Bull. 35 (1987) 823–828;
(b) I.H.P. De Esch, R.C. Vollinga, K. Goubitz, H. Schenk, U.
Appelberg, U. Hacksell, S. Lemstra, O.P. Zuiderveld, M. Hoffmann,
R. Leurs, W.M.P.B. Menge, H. Timmerman, J. Med. Chem. 42 (1999)
1115–1122.
1
isomer 11b was observed. H NMR: 9.58 (1H, br s), 7.52
(1H, dd, 1.4, 1.0), 7.47 (1H, d, 1.4), 7.37–7.31 (9H, m), 7.17–
7.10 (6H, m), 4.56 (2H, d, 4.1). 13C NMR: 149.99 (d, 246.4,
JCH 3.1 t, CF), 141.71 (3C), 138.91 (JCH 212.8 d, 7.4 d,
C2imi), 131.56 (d, 32.8, C4imi), 129.61 (6CH), 128.28 (3CH),
128.18 (6CH), 124.3 (d, 6.6, JCH 196.0 d, 12.9 m, C5imi),
106.46 (d, 38.3, JCH 2.8 t, ¼CBr–), 75.93 (Tr), 61.24 (d, 6.9,
JCH 147.2 t, CH2OH). 19F NMR: À106.7 (t, 4.1). HRMS
FABþ Calcd. for MHþ e.i. C26H23BrFN2O2: 493.0927,
495.0906. Found: 493.0935, 495.0931.
[3] B. Dolensky, K.L. Kirk, J. Org. Chem. 66 (2001) 4687–4691.
[4] J. Yamawaki, T. Kawate, T. Ando, T. Hanafusa, Bull. Chem. Soc.
Jpn. 56 (1983) 1885–1886.
[5] (a) R.A. Aitken, C.E.R. Horsburgh, J.G. McCreadie, S. Seth, J.
Chem. Soc. Perkin Trans. I (1994) 1727–1732;
(b) R.A. Aitken, S. Seth, J. Chem. Soc. Perkin Trans. I (1994)
2461–2466.
[6] J. Cousseau, P. Albert, Bull. Soc. Chim. Fr. (1986) 910–915.
[7] R.E.A. Dear, J. Org. Chem. 35 (1970) 1703–1705.
[8] S. Rozen, M. Brand, J. Org. Chem. 51 (1986) 222–225.
[9] (a) J.P. Gillet, R. Sauvetre, J.F. Normant, Synthesis (1986) 355–
360;
4.14. E- and Z-3-Fluoro-3-(1-trityl-1H-imidazol-4-yl)-
prop-2-en-1-ol (12a and 12b)
(b) T. Dubuffet, C. Bidon, P. Martinet, R. Sauvetre, J.F. Normant,
Organomet. Chem. 393 (1990) 161–172;
(c) L.M. Yagupol’skii, P.G. Cherednichenko, M.M. Kremlev, J. Org.
Chem. USSR 23 (1987) 246–248;
A pressure tube containing 63 mg (0.14 mmol) of FBr-
adduct 11a, 44 ml (0.16 mmol) of Bu3SnH, 4 ml of dry
benzene, and 1 mg (7 mmol) of 1,10-azobis(cyclohexanecar-
bonitrile) were placed under N2 and warmed in an oil bath
at 90 8C for 2.5 h. A small sample of the mixture (0.1 ml)
(d) J.P. Gillet, R. Sauvetre, J.F. Normant, Synthesis (1986) 538–543;
(e) F. Tellier, R. Sauvetre, J.F. Normant, J. Organomet. Chem. 292
(1985) 19–28;
(f) S. Martin, R. Sauvetre, J.F. Normant, J. Organomet. Chem. 264
(1984) 155–161.
1
was evaporated to dryness and H and 19F NMR indicated
[10] C. Chen, D. Crich, Tetrahedron 49 (1993) 7943–7954.
[11] B. Dolensky, J. Narayanan, K.L. Kirk, J. Fluorine Chem., 2003, in
press.
four fluorinated compounds 11a:11b:12a:12b in a ratio
83:10:1:6. The mixture was stirred additional 9 h at