The mechanism of base-promoted HF elimination from 4-fluoro-4-(4′-nitrophenyl)butan-2-one: A multiple isotope effect study including the leaving group 18F/19F KIE

Article abstract of DOI:10.1021/ja002513b

Leaving-group fluorine as well as the primary and secondary deuterium kinetic isotope effects (KIEs) have been determined for the base-promoted elimination of hydrogen fluoride from 4-fluoro-4-(4′-nitrophenyl)butan-2-one in aqueous solution. The elimination was studied for formate, acetate, and imidazole as the catalyzing base. The fluorine KIEs were determined using the accelerator-produced short-lived radionuclide ¹⁸F in combination with natural ¹⁹F. The ¹⁹F substrate was labeled with ¹⁴C in a remote position to enable radioactivity measurement of both isotopic substrates. The elimination reaction exhibits large primary deuterium KIEs: 3.2, 3.7, and 7.5 for formate, acetate, and imidazole, respectively, thus excluding the E1 mechanism. The corresponding C₄-secondary deuterium KIEs are 1.038, 1.050 and 1.014 and the leaving group fluorine KIEs are 1.0037, 1.0047 and 1.0013, respectively. The size of the fluorine KIEs corresponds to 5-15% of the estimated maximum of 1.03 for complete C-F bond breakage. No H/D exchange is observed during the reaction. The size and trends of the KIEs for the different bases are consistent with an E1cB-like E2 or an E1cB_ip mechanism.

Full text of DOI:10.1021/ja002513b

2712
J. Am. Chem. Soc. 2001, 123, 2712-2718
The Mechanism of Base-Promoted HF Elimination from
4-Fluoro-4-(4′-nitrophenyl)butan-2-one: A Multiple Isotope Effect
18
Study Including the Leaving Group F/19F KIE
Per Ryberg and Olle Matsson*
Contribution from the Institute of Chemistry, Uppsala UniVersity, PO Box 531, SE-751 21 Uppsala
ReceiVed July 10, 2000
Abstract: Leaving-group fluorine as well as the primary and secondary deuterium kinetic isotope effects (KIEs)
have been determined for the base-promoted elimination of hydrogen fluoride from 4-fluoro-4-(4′-nitrophenyl)-
butan-2-one in aqueous solution. The elimination was studied for formate, acetate, and imidazole as the catalyzing
base. The fluorine KIEs were determined using the accelerator-produced short-lived radionuclide 18F in
combination with natural 19F. The 19F substrate was labeled with 14C in a remote position to enable radioactivity
measurement of both isotopic substrates. The elimination reaction exhibits large primary deuterium KIEs:
3.2, 3.7, and 7.5 for formate, acetate, and imidazole, respectively, thus excluding the E1 mechanism. The
corresponding C4-secondary deuterium KIEs are 1.038, 1.050 and 1.014 and the leaving group fluorine KIEs
are 1.0037, 1.0047 and 1.0013, respectively. The size of the fluorine KIEs corresponds to 5-15% of the
estimated maximum of 1.03 for complete C-F bond breakage. No H/D exchange is observed during the
reaction. The size and trends of the KIEs for the different bases are consistent with an E1cB-like E2 or an
E1cBip mechanism.
Introduction
Scheme 1
Previously Schultz et al. have shown that the catalytic
antibody 43D4-3D12 promotes elimination from 4-fluoro-4-(4′-
nitrophenyl)butan-2-one1 and have reported significant primary
deuterium KIEs (PD KIEs) for both the antibody- and acetate-
promoted reactions.1b These results rule out an E1 mechanism
in which the rate-limiting detachment of the leaving group pre-
cedes a fast proton-transfer step, but are consistent with a rate-
limiting bond breakage of the proton being transferred in either
a concerted E2 or a stepwise E1cB mechanism (Scheme 1).
Leaving-group kinetic isotope effects (LG KIEs) can be used
to determine whether the bond breakage of the leaving group
is rate-limiting and may also be used to distinguish between
stepwise and concerted elimination mechanisms.2 Fluorine has
commonly been employed as a leaving group in mechanistic
studies of elimination and substitution reactions.3 However, since
nature provides 100% of 19F, experimental F KIEs had not been
available until a few years ago when Matsson et al. reported
the first fluorine KIEs4 which were determined for studies of
nucleophilic aromatic substitution reactions, where steric effects5
or change in solvent6 were demonstrated to cause a switch in
the rate-limiting step. The F KIEs were determined by using a
combination of the accelerator-produced short-lived tracer
nuclide 18F with the naturally abundant 19F. Theoretical F KIEs
on model E2 and E1cB elimination reactions have recently been
reported by Saunders, Jr.7 Therefore, it was of interest to use
our recently developed method for the determination of leaving-
group 18F/19F KIEs for an elimination reaction. The base-
promoted HF elimination from 4-fluoro-4-(4′-nitrophenyl)butan-
2-one appeared to be a suitable candidate for such a study. When
combined with the PD KIEs, the LG KIEs for a series of bases
with different basicity could provide enough information to
distinguish between the mechanistic alternatives. In addition,
the secondary R-deuterium KIEs (C4-SKIEs) for the same
reactions have also been determined, since they are expected
to exhibit a similar trend as the LG F KIEs.
Results
Synthesis. The 18F-labeled 4-fluoro-4-(4′-nitrophenyl)butan-
2-one was synthesized according to Scheme 2.
Prior to each kinetic experiment compound 3 was radioflu-
orinated via nucleophilic substitution to give compound 4 which
was then deprotected to give the 18F substrate 5.
(1) (a) Uno, T.; Schultz, P. G. J. Am. Chem. Soc. 1992, 114, 6573-
6574. (b) Shokat, K. M.; Uno, T.; Schultz, P. G. J. Am. Chem. Soc. 1994,
116, 2261-2270. (c) Romesberg, F. E.; Flanagan, M. E.; Uno, T.; Schultz,
P. G. J. Am. Chem. Soc. 1998, 120, 5160-5167.
(2) Saunders, W. H., Jr. Acc. Chem. Res. 1976, 9, 19-25.
(3) See e.g.: (a) de la Mare, P. B. D.; Swedlund, B. E. In The Chemistry
of Functional Groups. The Chemistry of the Carbon-Halogen Bond; Patai,
S., Ed.; John Wiley and Sons: London, 1973; Part 1, Chapter 7. (b) Parker,
R. E. In AdVances in Fluorine Chemistry; Stacey, M., Tatlow, J. C., Sharpe,
A. G., Eds.; Butterworth: London, 1963; Vol. 3. (c) Koch, H. F.; Koch, J.
G. In Fluorine Containing Molecules; Liebman, J. F., Greenberg, A.,
Dolbier, W. R., Jr., Eds.; VCH: New York, 1988; Chapter 6.
(4) Matsson, O.; Persson, J.; Axelsson, S.; Långstro¨m, B. J. Am. Chem.
Soc. 1993, 115, 5288-5289.
The 19F substrate with a remote 14C label was synthesized
from [14C]methyliodide according to Scheme 3. The [14C]-
methyliodide was converted to [14C]methyllithium by treatment
with n-butyllithium in hexane.8 The [14C]methyllithium was
treated with lithium(2-thienyl)cyanocuprate to give [14C]meth-
(7) (a) Saunders, W. H., Jr. J. Org. Chem. 1997, 62, 244-245. (b)
Saunders, W. H., Jr. J. Org. Chem. 1999, 64, 861-865.
(8) Aberhart, D. J.; Lawrence, J. L. J. Chem. Soc., Perkin Trans. 1 1974,
2320.
(5) Persson, J.; Matsson, O. J. Org. Chem. 1998, 63, 9348-9350.
(6) Persson, J.; Axelsson, S.; Matsson, O. J. Am. Chem. Soc. 1996, 118,
20-23.
10.1021/ja002513b CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/28/2001

Mechanism of Base-Promoted HF Elimination
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2713
Scheme 2a
a a) acetone, NaOH(aq); b) ethyleneglycol, PPTs, benzene (Dean-Stark); c) p-TsCl, pyridine; d) Kryptofix(2.2.2)K+18F-, MeCN; e) p-TsOH,
acetone/H2O.
Scheme 3a
a a) n-BuLi(hexane); b) (2-Th)CuCNLi(Et2O); c) 4-nitrocinnamoyl chloride (THF); d) H2O2, NaOH, MeOH; e) (PhSe)2, NAC, NaOH, MeOH;
f) DAST, CH2Cl2.
Scheme 4
Scheme 5
Table 1. Rate Constants, Primary Deuterium KIEs, C4-secondary
yldilithio(2-thienyl)cyanocuprate9 which upon reaction with
Deuterium KIEs, and Leaving-Group Fluorine KIEs for Formate-,
4-nitrocinnamoyl chloride gave the R, â- unsaturated ketone 6.
Acetate-, and Imidazole-Promoted Dehydrofluorination of the
Fluorobutanone in 75% Aqueous Methanola
The deuterated substrate 12 was synthesized from 4-nitro-
(R-2H)benzaldehyde 10 via the same route as that for compounds
11 and 14, described below (see Schemes 4 and 5). The 4-nitro-
(R-2H)benzaldehyde was obtained by reduction of 4-nitrobenzoyl
chloride with lithium tri-tert-butoxyaluminodeuteride.10
k
primary
secondary
leaving group
pKa base min-1 M-1 kH/kD
kH/kD
k18/k19
3.7 formate
4.7 acetate
6.95 imidazol 0.55
0.0073 3.2 ( 0.1 1.038 ( 0.013 1.0037 ( 0.0020
0.028
3.7 ( 0.1 1.050 ( 0.014 1.0047 ( 0.0012
7.45 ( 0.1 1.014 ( 0.017 1.0013 ( 0.0012
The 4-fluoro-4-(4′-nitrophenyl)-butan-2-one 11 and 4-fluoro-
4-(4′-nitrophenyl)-(1,1,1,3,3-2H5)butan-2-one 14 were synthe-
sized as described by Schultz et al.1b Compound 13 which was
needed as a reference for HPLC was obtained by treatment of
the hydroxyacetal 2 with DAST, according to Scheme 4. An
interesting observation is that 4-nitrostyrene was formed as the
major product.
a The KIES are average values from several separate kinetic
experiments as described under Experimental Section. The experimental
uncertainty was calculated as the standard deviation of the mean.
of confidence. The values for acetate and formate are, however,
not significantly different according to the t-test.
Kinetic Isotope Effects. The kinetic isotope effects observed
in the base-promoted HF elimination experiments are shown
in Table 1. A t-test on the F KIEs and S D KIEs shows that
both acetate and formate differ from imidazole at the 95% level
All experiments were run at 38 °C under pseudo-first-order
conditions using a 75% aqueous methanol kinetic buffer. The
sodium formate and sodium acetate concentrations were 1 M.
The pH was adjusted with HCl to 5.2 and 6.0, respectively.
The imidazole solution was 120 mM in imidazole and 0.94 M
in NaCl; the pH was adjusted to 7.0 by addition of HCl. In all
cases the substrate concentration was 5 mM or less.
(9) Kihlberg, T.; Neu, H.; Långstro¨m, B. Acta Chem. Scand. 1997, 51,
791-796.
(10) Brown, H. C.; Subba Rao, B. C. J. Am. Chem. Soc. 1958, 80, 5377.

2714 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Ryberg and Matsson
The possibility of H/D-exchange was investigated for the
acetate-promoted elimination reaction. No isotopic exchange was
observed. In all cases the (E)-alkene was the only product
formed in the elimination reaction.
KIEs with high precision. The most frequently utilized elements
have been carbon, nitrogen, oxygen, sulfur, and chlorine.
Recently Matsson et al. added fluorine to the list of elements
for which KIEs can be determined.4 The fluoride ion is a
commonly employed leaving group in mechanistic investigations
of elimination and substitution reactions.3 Leaving-group KIEs
are fairly easy to interpret, particularly for one-atom groups,
and have been utilized for a long time in mechanistic investiga-
tions.
Kinetic Methods. The kinetic method for F KIE determi-
nation is a one-pot technique based on HPLC separation of
reactants and products in a series of samples representing
different extents of reaction followed by liquid scintillation
counting of the collected radioactive fractions.
A leaving-group KIE is expected to monotonically increase
with the increasing degree of bond breakage between the
isotopic leaving-group atom and the R-carbon in the transition
state of the rate-limiting step.17,18 Several investigators have
utilized leaving-group KIEs in mechanistic studies of elimination
reactions, including KIEs of nitrogen for the bimolecular
elimination of arylethylammonium ions,19 of sulfur for elimina-
tion of 2-phenylethyldimethylsulfonium ion,20 of chlorine for
dehydrochlorination reactions,21 and that of oxygen for the
enzymatic fumarase and crotonase reactions.22
As suggested by Thibblin and Ahlberg23 and demonstrated
by Saunders in computational studies,7 a secondary leaving-
group fluorine KIE may appear on the deprotonation step in an
E1cB elimination reaction due to negative ion hyperconjugation.
Calculations by Saunders suggest that the TS of the E2 reaction
of hydroxide ion with ethyl fluoride was very similar to that of
the proton-transfer step of the E1cB reaction of 1,1,1-trifluo-
roethane.7 Fluorine isotope effects (18F/19F) as large as 1.0058
(corresponding to 18% of the estimated maximal F KIE for
complete breakage of a C-F bond4) were calculated for the
E1cB proton-transfer TS.
The method used for F KIE determination was similar to the
methods for determination of KIEs using short-lived 11C in
combination with 14C and short-lived 18F in combination with
natural 19F which were developed earlier in our laboratory.4,6,11
In this study the method has been modified by introducing a
remote 14C label in the naturally abundant 19F substrate. This
modification was done because of difficulties in obtaining high
enough precision in preliminary experiments using the previous
method with integration of the UV-detector signals. We also
wanted to evaluate the potential of the remote labeling tech-
nique12 since this might be expected to possess several
advantages, that is (i) both substrates can be measured quanti-
tatively using the same instrument and (ii) in the HPLC
fractionation, it is of little importance that the reactant and
product fractions are free from impurities, as long as they are
radiochemically pure. Measurement of the 18F and 14C radio-
activity were performed by liquid scintillation counting. The F
KIEs were calculated from the 18F/19F(14C) isotopic ratios at
0% and 80-95% reaction. For each base studied at least three
separate kinetic experiments were performed and 2-4 point
KIEs were determined for each experiment. Then the average
value was calculated, and the uncertainty was calculated as the
standard deviation of the mean.
The size of primary deuterium KIEs for proton-transfer
processes, such as elimination reactions, are generally considered
as a measure of the symmetry of the transition state for the
transfer. On the basis of the transition state theory, the
Melander-Westheimer postulate predicts that the maximal
primary KIE occurs for the most symmetrical activated complex,
that is where the proton is bound with equal strength to both
the donor and acceptor.24 This usually occurs when the donor
and acceptor are of equal pKa strength.25 Several examples of
Values for the primary and secondary deuterium KIEs were
calculated from the ratio of the rate constants for the light and
heavy reactants. The data used for constructing the rate curves
were obtained by integration of the UV peak at 270 nm for the
reactant in samples taken at different extents of reaction. The
radiochromatograms showed only the reactant peak (retention
time tR ) 8.1 min). In the UV chromatogram only the internal
standard (phenol, tR ) 3.6 min) and the product (tR )10.5 min)
were observed in addition to the reactant peak.
(14) (a) Weiss, P. M. In Enzyme Mechanism from Isotope Effects; Cook,
P. F., Ed.; CRC Press: Boca Raton, 1991; Chapter 11. (b) Paneth, P. In
Isotopes in Organic Chemistry; Buncel, E., Saunders, W. H., Jr., Eds.;
Elsevier: Amsterdam, 1992; Vol. 8, Chapter 2.
(15) Singleton, D. A.; Thomas, A. A. J. Am. Chem. Soc. 1995, 117,
9357-9358.
(16) (a) Parkin, D. W. In Enzyme Mechanism from Isotope Effects; Cook,
P. F., Ed.; CRC-Press: Boca Raton, 1991; Chapter 10. (b) Bergson, G.;
Matsson, O.; Sjo¨berg, S. Chem. Scr. 1977, 11, 25-31.
(17) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; Chapter 9.2.
(18) (a) Saunders, W. H. Chem. Scr. 1975, 8, 27-36. (b) Glad, S. S.;
Jensen, F. J. Org. Chem. 1997, 62, 253-260. (c) Glad, S. S.; Jensen, F. J.
Am. Chem. Soc. 1994, 116, 9302-9310.
(19) (a) Smith, P. J.; Bourns, A. N. Can. J. Chem. 1970, 48, 125-132.
(b) Smith, P. J.; Bourns, A. N. Can. J. Chem. 1974, 52, 749-760.
(20) (a) Saunders, W. H., Jr.; Cockerill, A. F.; Asˇperger, S.; Klasinc,
L.; Stefanovic′, D. J. Am. Chem. Soc. 1966, 88, 848. (b) Cockerill, A. F.;
Saunders, W. H., Jr. J. Am. Chem. Soc. 1967, 89, 4985.
Discussion
The determination and interpretation of kinetic isotope effects
provides one of the most powerful tools in physical organic
and bioorganic chemistry.13 The observation of a KIE demon-
strates the rate-limiting change of bonding to the isotopic atom.
Traditionally deuterium kinetic isotope effects have been the
most frequently studied due to their importance, large magnitude
and relative ease of determination. However, the development
of very accurate methods based on mass spectrometry,14 NMR,15
polarimetry,16 or radioactivity measurements14b,16a has enabled
chemists to determine mechanistically significant heavy-atom
(21) (a) Grout, A.; McLennan, D. J.; Spackman, I. H. J. Chem. Soc.,
Perkin Trans. 2. 1977, 1758. (b) Koch, H. F.; Koch, J. G.; Koch, N. H.;
Koch, A. S. J. Am. Chem. Soc. 1983, 105, 2388. (c) Koch, H. F.; Koch, J.
G.; Tumas, W.; McLennan, D. J.; Dobson, B.; Lodder, G. J. Am. Chem.
Soc. 1980, 102, 7955-7956. (d) Koch, H. F.; McLennan; Koch, J. G.;
Tumas, W.; Dobson, B.; Koch, N. H.. J. Am. Chem. Soc. 1983, 105, 1930-
1937.
(22) (a) Blanchard, J. S.; Cleland, W. W. Biochemistry 1980, 19, 4506-
4513. (b) Bahnson, B. J.; Anderson, V. E. Biochemistry 1989, 28, 4173-
4181.
(11) (a) Axelsson, B. S.; Långstro¨m, B.; Matsson, O. J. Am. Chem. Soc.
1987, 109, 7233-7235. (b) Matsson, O.; Axelsson, S.; Husse´nius, A.;
Ryberg, P. Acta Chem. Scand. 1999, 53, 670-79.
(12) (a) Stark, G. R. J. Biol. Chem. 1971, 246, 3064. (b) O’Leary, M.
H.; Marlier, J. F. J. Am. Chem. Soc. 1978, 100, 2582. (c) O’Leary, M. H.;
Marlier, J. F. J. Am. Chem. Soc. 1979, 101, 3300.
(13) See e. g. (a) Melander, L.; Saunders, W. H., Jr. Reaction Rates of
Isotopic Molecules; Wiley-Interscience: New York, 1980. (b) Enzyme
Mechanism from Isotope Effects; Cook, P. F., Ed.; CRC Press: Boca Raton,
1991.
(23) Thibblin, A.; Ahlberg, P. J. Am. Chem. Soc. 1977, 99, 7926-7930.

Mechanism of Base-Promoted HF Elimination
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2715
Scheme 6
The C4-SKIEs and leaving-group F KIEs observed are very
small and decrease with increasing pKa-value when going from
acetate or formate to imidazole. However, the differences in
these KIEs between the acetate- and formate-promoted reactions
are within experimental error. The combination of large PD
KIEs and small LG F KIEs and SD KIEs together with the
absence of any detectable H/D-exchange fits with either the
E1cB-like E2 or the E1cBip.
the use of primary deuterium KIEs in the investigation of the
mechanisms of elimination reactions have been reported.26
Secondary deuterium KIEs are usually used to monitor
rehybridization in the TS.27 On the basis of theoretical calcula-
tions it has been suggested that the secondary KIEs for the
R-position are more reliable than that for the â-position as a
measure of rehybridization in elimination reactions.18b
Base-promoted â-eliminations may follow either the E2
(Scheme 6) or the E1cB (Scheme 7) pathway. The actual
pathway chosen will be dependent on the stability of the
carbanion intermediate. It has been proposed by Dewar,28 that
since generally two-bond activations cost more energy than one-
bond activation, an E1cB mechanism would be energetically
more favorable than an E2. If, however, the putative carbanion
is too unstable to have any significant lifetime, the reaction
would then be forced to go via the E2 mechanism. Since both
mechanisms may have similar kinetic characteristics, it can
prove to be difficult to distinguish between either of them.
Solvent H/D exchange is indicative for an E1cB mechanism
where the carbanion is reprotonated by the solvent.
For the present reaction system no H/D exchange was
observed when acetate was used as the base. The elimination
reaction exhibits deuterium KIEs that are large enough to be
safely classified as primary (3.2-7.45) and increases with base
strength. Since the substrate is doubly deuterated in the
3-position, the deuterium not being transferred may introduce
a secondary KIE affecting the size of the observed primary KIE.
Experimentally determined secondary KIEs for the correspond-
ing position have been reported for the dehydrobromination and
dehydrotosylation of cyclohexyl derivatives.29 These studies
suggest that the size of such a KIE may be in the range of 1.15-
1.25. Somewhat larger values have been obtained in theoretical
calculations.18b Due to the relatively small size of the C3-SKIE
as compared to that of the primary KIE, the trend in the observed
values (Table 1) may be assumed to reflect the variation of the
primary KIE. The absolute magnitude is, however, slightly too
large. The three deuteriums on the terminal methyl group are
remote from the reaction center and are not expected to introduce
any significant KIE.
The E2 Mechanistic Alternative. According to the variable
TS structure model the effect of a stronger base on a symmetric
E2 TS would be an unchanged degree of C-H bond breakage
and a decrease in C-F bond breaking. This would be observed
as an unchanged primary deuterium KIE. The effect of less bond
breakage to the leaving group would result in a smaller LG F
KIE. For the E1cB-like E2 mechanism the change to a stronger
base would be accompanied by a decrease in C-H bond
breakage and also a decrease in C-F bond breakage. This
should be observed as an increase in PD KIE, provided that
the conjugate acid of the catalyzing base is stronger than the
donor. The size of the LG KIE would decrease just as for the
case of a symmetric E2 TS. The degree of rehybridization at
the R-carbon may be considered to correlate with the degree of
bond breaking to the leaving group resulting in a corresponding
decrease of the secondary deuterium KIE. However, on the basis
of experimental as well as computational studies Bernasconi
and other workers have demonstrated cases of nonperfect
synchronization between proton-transfer and rehybridization.30
If such a time-lag is allowed in the present reaction system,
then a smaller than expected secondary deuterium KIE would
be observed.
The observed primary deuterium, secondary deuterium ,and
leaving group F KIEs conform with the trends predicted by the
variable TS structure model for the E1cB-like E2-mechanism.
Both the C4-SKIEs and the leaving-group KIEs decrease
significantly, that is from 1.050 to 1.014 and from 1.0047 to
1.0013, respectively, when changing from acetate to imidazole
as the catalyzing base. The primary deuterium KIE at the same
time changes in the direction expected for the E1cB-like TS
from 3.7 to 7.45. Thus, these observations are consistent with
the predictions based on the variable TS structure model for an
E1cB-like E2 mechanism.
The E1cB Mechanistic Alternative. For the E1cB case the
interpretation of the KIEs is more complicated due to the kinetic
complexity. The relationship between the mechanistic KIEs and
the size of the observed leaving-group KIEs for an E1cB reaction
are given by eq 1. The same holds for the C4-SKIEs of the k2
step. For the observed primary deuterium KIE the situation is
more complex and will be discussed separately in detail below.
(24) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; pp 29 ff.
(25) (a) More O’Ferrall, R. In Proton-Transfer Reactions; Caldin, E. F.,
Gold, V., Eds.; Chapman and Hall: London, 1975; Chapter 8. (b) Melander,
L.; Saunders, W. H., Jr. Reaction Rates of Isotopic Molecules, Wiley-
Interscience: New York, 1980; pp 131 ff.
(26) (a) Smith, P. J. In Isotopes in Organic Chemistry; Buncel, E., Lee,
C. C., Eds.; Elsevier: Amsterdam, 1976; Vol. 2, Chapter 6. (b) Melander,
L.; Saunders, W. H., Jr. Isotope Effects on Reaction Rates; Wiley-
Interscience: New York, 1980; p 136. (c) McLennan, D. J. In Isotopes in
Organic Chemistry; Buncel, E., Lee, C. C., Eds.; Elsevier: Amsterdam,
1987; Vol. 7, Chapter 6. (d) Fry, A. Chem. Soc. ReV. 1972, 1, 163-210. (e)
Saunders, W. H., Jr.; Cockerill, A. F. Mechanisms of Elimination Reactions;
Wiley: New York, 1973. (f) Cook, D.; Hutchinson, R. E. J.; MacLeod, J.
K.; Parker, A. J. J. Org. Chem. 1974, 39, 534-539. (g) More O’Ferrall, R.
A.; Warren, P. J. J. Chem. Soc., Chem. Commun. 1975, 483-484. (h)
Baciocchi, E. Acc. Chem. Res. 1979, 12, 430-436. (i) Thibblin, A. J. Am.
Chem. Soc. 1988, 110, 4582-4586. (j) Thibblin, A. J. Am. Chem. Soc.
1989, 111, 5412-5416.
kL2 kH2 + k-1[BH]
kH2 kL2 + k-1[BH]
KIEOBS
)
×
(1)
(L ) isotopically lighter species, H ) isotopically heavier
species).
From eq 1 it is clear that the size of the observed LG KIEs
are dependent on the ratio between the rate of reprotonation of
(30) (a) Bernasconi, C. F. Acc. Chem. Res. 1987, 20, 301-308. (b)
Bernasconi, C. F. Acc. Chem. Res. 1992, 25, 9-16. (c) Bernasconi, C. F.
AdV. Phys. Org. Chem. 1992, 27, 119-238. (d) Bernasconi, C. F.; Wenzel,
P. J. J. Am. Chem. Soc. 1994, 116, 5405-5413. (e) Bernasconi, C. F.;
Wenzel, P. J. J. Am. Chem. Soc. 1996, 118, 10494-10504. (f) Bernasconi,
C. F.; Wenzel, P. J.; Keeffe, J. R.; Gronert, S. J. Am. Chem. Soc. 1997,
119, 4008-4020. (g) Harris, N.; Wei, W.; Saunders, W. H., Jr.; Shaik, S.
J. Phys. Org. Chem. 1999, 12, 259-262.
(27) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; pp 172 ff.
(28) Dewar, M. J. S. J. Am. Chem. Soc. 1984, 106, 209-219.
(29) Cook, D.; Hutchinson, R. E. J.; Parker, A. J. J. Org. Chem. 1974,
39, 3029-3038.

2716 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Ryberg and Matsson
Scheme 7
the carbanion and the rate of elimination of the leaving group.
Two extreme situations may be identified.
(a) When reprotonation of the carbanion is fast compared to
elimination of the LG.
(b) When reprotonation of the carbanion is slow compared
to elimination.
would be independent of the base, and consequently the leaving-
group and secondary deuterium KIEs should not vary with base
strength.
Assuming an E1cBip mechanism, it is not surprising that the
smallest leaving-group and secondary deuterium KIEs are
observed for the strongest base imidazole. One may guess that
the stability of the hydrogen-bonded carbanion-protonated base
complex increases with base strength. It would also be reason-
able to assume that the rate of reprotonation is dependent on
the stability of this complex, that is the more stable complex
gives the slower reprotonation. Slower reprotonation would be
observed as smaller leaving-group and secondary deuterium
KIEs (eq 1). When changing from acetate (pKa ) 4.7) to
imidazole (pKa ) 6.95) as catalyzing base, the leaving,group
and secondary deuterium KIEs decrease from 1.0047 to 1.0013
and from 1.050 to 1.014, respectively. When changing from
formate to acetate, the trend is reversed but is much less
pronounced.
The variation in size of the primary D KIEs is more likely a
result of changes both in the intrinsic KIEs and the rates of
reprotonation. The effect of a stronger base on the deprotonation
step should result in a more symmetrical TS. This would be
observed as a larger KIE for the proton transfer, provided that
the conjugate acid of the catalyzing base is stronger than that
of the proton donor. The large PD KIE of 7.45 for imidazole is
probably close to the intrinsic value, whereas for the acetate,
which is a weaker base, a smaller intrinsic value is assumed.
The observed PD KIE (kH/kD ) 3.7) would then be smaller than
the intrinsic KIE due to internal return. The same arguments
should also account for formate.
In case (a) where reprotonation is fast compared to elimina-
tion, and reprotonation occurs from the solvent, the size of the
observed primary deuterium KIE will be affected by H/D-
exchange, that is the deuterated reactant will rapidly be
converted to the 1H analogue. Consequently, a very small
primary deuterium KIE which varies with the extent of reaction
will be observed. The magnitude of the observed leaving group
and secondary deuterium KIEs would, on the other hand, be
close to the intrinsic KIEs on the elimination step (k2) as
predicted from eq 1. This type of mechanism is usually classified
as a reversible E1cB (E1cBrev).
In case (b) exchange of 2H for 1H with solvent is slow
compared to elimination. Thus, the observed primary deuterium
KIE will be close to the intrinsic value. According to eq 1 the
leaving-group KIE and C4-SKIEs are predicted to be negligible.
This mechanism is denoted irreversible E1cB (E1cBirr).
A third situation occurs when the carbanion remains inti-
mately associated with the newly protonated base, either as an
ion pair if the protonated base is positively charged or as a
hydrogen-bonded pair if the protonated base is neutral. This
mechanism is denoted as the ion pair E1cB (E1cBip). If internal
return is allowed (rehydronation occurs with the previously
removed Hydron), then eq 2, where [IP - H] and [IP - D]
denotes the steady-state concentrations of the protic and
deuterated hydrogen-bonded complexes, is valid. Assuming that
there is a KIE on the reprotonation step, this would lead to a
situation where the size of the primary deuterium KIE is
dependent on the ratio between k-1[IP - H] and k2.
On the basis of these arguments one may conclude that an
E1cB-like E2 or an E1cBip mechanism are the only alternatives
which are fully consistent with the observations reported herein.
Experimental Section
Our observation of the large primary deuterium KIEs together
with the absence of H/D-exchange is consistent with either an
E1cBirr or an E1cBip mechanism but not with an E1cBrev
mechanism.
General. The 18F-fluoride was prepared by use of the Scanditronix
MC-17 cyclotron at the Uppsala University PET Centre. The 18O(p,n)18F
reaction was performed in a high-pressure silver target containing 18O-
enriched water bombarded with 17 MeV protons.
HPLC was performed with a Beckman 126 gradient pump and a
Beckman 166 variable wavelength or a Beckman 168 diode array UV-
detector in series with a â+ flow detector.
kH1 k2 + kD-1[IP - D]
kD1 k2 + kH-1[IP - H]
KIEOBS
)
×
(2)
For analytical HPLC a Phenomenex spherisorb 5 ODS(2) 250 ×
4.6 mm ID C18 column was used. For semipreparative HPLC a
Phenomenex spherisorb 10 ODS(2) 250 × 10 mm ID C18 column was
used. Datacollection and HPLC control were performed with the use
of a Beckman System Gold chromatography software package.
Sample injection and fraction collection were performed by a Gilson
231 XL sampling injector in combination with a Gilson 401C dilutor.
Liquid scintillation counting was performed with a Beckman LS 6000
LL liquid scintillation counter in wide open mode, using Zinsser
Quicksafe A as scintillation cocktail in Zinsser 20 mL poly vials.
1H- and 13C NMR spectra were recorded on a Varian Unity 400
MHz spectrometer at 400 and 100.6 MHz, respectively, with (2H)-
chloroform as internal standard.
The small leaving-group and secondary deuterium KIEs may
be explained by an E1cBirr or an E1cBip mechanism. Since the
intermediate from which the elimination occurs is identical for
the different bases, it is reasonable to assume that the intrinsic
leaving- group and secondary deuterium KIEs on the k2 step
are independent of base strength. Then, the variation in leaving-
group and secondary deuterium KIEs when the catalyzing base
is changed would be a consequence of different rates of
reprotonation. This observation is only consistent with the
E1cBip mechanism, primarily because the very observation of
a primary leaving-group KIE for any E1cB mechanism except
for the E1cBip requires reprotonation of the carbanion by solvent.
This would be accompanied by H/D-exchange. Moreover, the
rate of reprotonation of the carbanion in an E1cBirr mechanism
Kinetic Procedure. The kinetic experiments were run under pseudo-
first-order conditions. Analytical HPLC system was as descibed under
General. Solvent system was methanol/50 mM ammonium formate pH
3.5, 0-9 min 55% methanol, 9-10 min 55-95% methanol, 14-15
min 95-55% methanol 1 mL/min. In this system the reactant eluted at

Mechanism of Base-Promoted HF Elimination
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2717
tR ) 8.1 min, and the product eluted at tR ) 10.5 min, and they were
baseline separated for 1.4 min.
ratio between the 18F and 19F(14C) reactants and R is the corresponding
ratio at (1 - f).
The 18F substrate in 0.5 mL of 50% aqueous methanol and the 14C-
labeled 19F substrate 35 KBq in 50 µL of acetonitrile were mixed in a
2 mL septum-covered HPLC vial, and four 10 µL aliquots were injected
to the HPLC. For each injection three fractions were collected into 20
mL liquid scintillation vials containing 14 mL of liquid scintillation
cocktail. Fraction 1 (tR ) 6-7 min) was collected just before the
reactant fraction, fraction 2 (tR ) 7.7-8.7 min) was the reactant fraction,
and fraction 3 (tR ) 9.5-11.2 min) was the product fraction. Then the
base was added to the substrate mixture, and the reaction was allowed
to proceed to 80-95% conversion at 38 °C. Then 700 µL out of the
total 1000 µL of kinetic solution was withdrawn and passed through a
100 mg C-18 SPE column. The column was washed with 1 mL of
H2O and then eluted with 1 mL of acetonitrile.The eluate was
concentrated in vacuo and redissolved in 0.4 mL 50% aqueous
methanol. From this eluate three 100 µL aliquots were injected to the
HPLC, and fractions were collected as above. The remaining 300 µL
of the kinetic solution was allowed reach complete reaction, and then
methanol 100 µL was added, and three 100 µL aliquots were injected
on the HPLC. The reactant and product fractions were collected as
above.
ln(l - f)
ln((l - f)R/R0)
KIE )
(7)
(8)
14
C
react
l - f ) 14
14
Creact
+ C
prod
H/D Exchange Experiment. In the H/D exchange experiment
4-fluoro-4-(4′-nitrophenyl)-butan-2-one (300 mg) was dissolved in 0.5
M sodium acetate in D2O/MeOD/THF 1:1:2 (40 mL). The mixture was
stirred at rt for 110 min (corresponding to 60% reaction), then saturated
aqueous NaCl (50 mL) was added, and the mixture was extracted with
Et2O. The Et2O phase was dried over MgSO4 and concentrated in vacuo.
The residue was dissolved in chloroform, and unreacted 4-fluoro-4-
2
(4′-nitrophenyl)-butan-2-one was analyzed by H NMR spectroscopy.
No H incorporation could be detected.
2
Synthesis. 4-(4′-Nitrophenyl)-[1-14C]but-3-ene-2-one (6). The 14C-
labeled methyl iodide 37 MBq (2.04 Gbq/mmol) was vacuum trans-
ferred to dry hexane 1 mL at -100 °C. To 0.5 mL of this solution
(18.5 MBq) was added normal methyl iodide (20 µL, 0.4 mmol). n-BuLi
1.6 M in hexane (0.44 mL, 0.7 mmol) was added to a 3 mL oven dried
tube covered with a rubber septum and flushed with N2. The methyl
iodide solution was added to the BuLi at rt, and the tube was agitated
for 10 min while a white precipitate of MeLi(s) was formed. The tube
was centrifuged for 5 min at 3000 rpm, giving a pellet of MeLi at the
bottom of the tube. The supernatant containing excess BuLi and butyl
iodide was removed via a syringe, and n-BuLi (0.16 M, 1.3 mL in
diethyl ether) was added, giving a clear solution. This solution was
added to (2-Th)CuCNLi (0.25 M, 2.5 mL, 0.65 mmol)in dry THF (2.5
mL) at -78 °C. The mixture was stirred for 10 min and then added to
p-nitrocinnamoyl chloride (158 mg, 0.75 mmol) in THF (3 mL) at -78
°C. After 20 min the reaction was quenched with methanol 1 mL and
poured into saturated NaCl(aq) (20 mL). The mixture was extracted
with diethyl ether (3 × 20 mL). The combined organic extracts were
concentrated in vacuo, and the product was purified by prep. TLC using
diethyl ether/hexane 1:1 as eluent. Rf ) 0.26. 6 (15 MBq, 81%) was
obtained with >99% radiochemical purity according to radio TLC. The
product was identified by HPLC.
As a consequence of the 18F-fluoride temporarily binding to the
HPLC column a 18F background radioactivity appeared, which had to
be corrected for. Thus, samples from fraction 1 were collected. It was
also observed that the reactant fraction contained a small, although often
negligible, 14C background which also had to be corrected for. These
corrections were performed according to the procedure described below.
When all of the fractions had been collected, the 18F background
radioactivity, 18Fbg, and the total reactant radioactivity, (18F + 14C)react
,
were determined by measuring the radioactivity in all fraction 1 samples
and fraction 2 samples simultaneously. Thus, by measuring the samples
from each injection pairwise with a counting time of 2 min/vial, no
decay correction was needed for further calculations.
To obtain a high precision value for the total reactant radioactivity,
(18F + 14C)*react, needed for KIE calculations, the fraction 2 samples
were remeasured with a counting time of 20-40 min/vial.
When all 18F had decayed (48 h), the 14C radioactivity in the fraction
2 and fraction 3 samples were measured again to give the reactant 14
C
14
14
radioactivity,
C
, and the product 14C radioactivity,
C
. The
react
prod
counting was continued until a precision of 2σ ) 0.1-0.2% was
obtained.
The 14C radioactivity in the reactant and product fractions for the
3,4-Epoxy-4-(4′-nitrophenyl)-[1-14C]butan-2-one (7). 6 (15 Mbq)
was dissolved in methanol (6 mL). Aqueous 27% H2O2 (0.1 mL) and
NaOH(aq) (6 M) 4 drops were added, and the mixture was stirred at rt
for 10 min and then poured into saturated NaCl(aq) (15 mL). Extraction
with CH2Cl2 (3 × 10 mL) and concentration in vacuo gave 7 11 MBq
(73%) as a yellow oil. The product was analyzed by radio TLC and
HPLC, showing both high radiochemical and UV purity.
4-Hydroxy-4-(4′-nitrophenyl)-[1-14C]butan-2-one (8). 7 (11 Mbq)
dissolved in methanol 3 mL was added to a solution of N-acetylcysteine
(100 mg, 0.61 mmol), (PhSe)2 (3 mg, 0.0096 mmol, and NaOH(aq)
(0.1 mL 6 M, 0.6 mmol) in methanol 4 mL.32 The mixture was stirred
under N2 at rt for 13 h and then poured into saturated NaCl(aq) (20
mL); extraction with diethyl ether (3 × 10 mL) and concentration in
vacuo gave 8 as yellow crystals. The product was purified by preparative
TLC using diethyl ether/hexane 3:1 as eluent; Rf ) 0.57. Only the
middle part of the fraction containing 8 was recovered, yielding 9 MBq
(82%) of pure product according to HPLC.
4-Fluoro-4-(4′-nitrophenyl)-[1-14C]butan-2-one (9). 8 (2.1MBq)
was dissolved in dry CH2Cl2 (2 mL) and cooled to -78 °C under N2.
Diethylaminosulfur trifluoride (DAST) (10 µL) was added, and the
mixture was stirred at -78 °C for 20 min and then heated to rt. Water
(1 mL) was added, and the mixture was extracted with CH2Cl2 (3 × 2
mL). The product was purified by semipreparative HPLC on a
Phenomenex C-18 column using 50 mM ammonium formate pH 3.5/
acetonitrile as eluent system (4 mL/min.). The gradient was 0-2 min
10% acetonitrile, 2-21 min 10-65% acetonitrile. A fraction between
tR 17.3-19.5 was collected. The collected fraction was poured into
saturated aqueous NaCl (20 mL) and extracted with (3 × 5 mL) CH2-
Cl2. Drying over MgSO4 and removal of solvent in vacuo yielded 1.53
samples taken at complete reaction were also measured to give the 14
C
background radioactivity in the reactant fraction at complete reaction,
14
C
C
∞, and the product 14C radioactivity at complete reaction,
.
react,
14
prod,
∞
The data were also corrected for instrumental background before
any further calculations were made.
18
The relative 18F background,18Frel, bg, was obtained from the
F
bg
and the (18F + 14C)react according to eq 3. The reactant 18F radioactivity,
18
F
react
, was obtained according to eq 4. The 18Freact was then corrected
for decay according to eq 5, where t is the elapsed time from the start
of the radioactivity measurements in minutes and k ) ln 2/t1/2. This
18
gave the true reactant 18F radioactivity,
F
react,
true, used for KIE
14
calculation. The
C
was corrected for 14C background and instru-
react
mental background, eq 6, to give the true reactant 14C radioactivity,
14
Creact, true, used for the KIE calculations.
18
18
14
F
)
F
/[(18F + 14C)react
-
C ]
react
(3)
(4)
(5)
rel,bg
bg
18
14
F
) [(18F + 14C)* react
-
C
] × (1 - 18Frel,bg
)
react
react
18
18
F
)
F
× ekt
react,true
react
14
14
14
C
)
C
× [1 - 14Creact,∞/(14Creact,∞
+
C
)] (6)
prod,∞
react,true
react
The point F KIE for each individual sample was calculated according
to eq 731 where 1 - f was calculated according to eq 8, R0 is the initial
(31) Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic
Molecules; Wiley-Interscience: New York, 1980; p 95.
(32) Engman, L.; Stern, D. J. Org. Chem. 1994, 59, 5179-5183.

2718 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Ryberg and Matsson
MBq (73%) of radiochemically pure 9. The product was dissolved in
acetonitrile (3 mL) containing 5% of methanol and stored at -15 °C.
4-Hydroxy-4-(4′-nitrophenyl)-butan-2-one ethylene acetal (2).
4-Hydroxy-4-(4′-nitrophenyl)-butan-2-one (2.15 g, 10 mmol), 1,2-
ethanediol (3 g, 48 mmol), and pyridinium p-toluenesulfonate (0,5 g,
2 mmol) were weighed into a round-bottomed flask equipped with a
Dean-Stark water separator. Benzene (75 mL) was added, and the
mixture was refluxed until no more water separated (2 h). The benzene
was removed in vacuo, and diethyl ether (100 mL) was added to the
brownish residue. This ether solution was washed with saturated
aqueous NaHCO3 (50 mL) and H2O (50 mL) and then dried over
MgSO4. Removal of the ether in vacuo yielded a brown oil that
crystallizes to yellow needles. The product was purified by chroma-
tography on silica using diethyl ether/pentane/triethylamine 100:50:1
as eluent. 2 (1.53 g, 60%) was obtained. 1H NMR δ 1.41 (s, 3H), 2.03
(m, 2H), 4.06 (m, 4H), 4.28 (s, 1H), 5.11 (m, 1H), 7.55-8.19 (m, 4H).
13C NMR δ 151.7, 147.1, 126.5, 123.6, 109.9, 64.9, 64.4, 47.2, 24.2.
Anal. Calcd for C12H15NO5: C, 56.91%; H, 5.97%; N, 5.53%. Found:
C, 56.84%; H, 5.94%; N, 5.58%.
4-(4′-Nitrophenyl)-4-p-toluenesulfoxybutan-2-one ethylene acetal
(3). 2 (800 mg, 3.2 mmol), p-toluenesulfonyl chloride (900 mg, 4.7
mmol), N,N-(dimethylamino)pyridine (390 mg, 3.2 mmol), and tri-
ethylamine (0.45 g, 4.4 mmol) were stirred in dichloromethane (10
mL) for 30 h. The mixture was diluted with diethyl ether (60 mL) and
filtered. The etheral extract was washed with saturated aqueous NaHCO3
and saturated aqueous NaCl and then dried over MgSO4. The product
was purified by chromatography on silica using diethyl ether/pentane/
triethylamine 80:20:2 as eluent. Yield: 400 mg (28%).
mixture was stirred for 30 min and then slowly cannulated (20 min) to
4-nitrobenzoyl chloride (2 g) in diglyme (10 mL) at -78 °C. The
mixture was stirred for an additional 30 min, and then H2O (2 mL)
was added. The mixture was filtered through a plug of silica using
Et2O/pentane 1:1 as eluent. The eluate was cooled to 0 °C, and water
was added. The product was filtered off and dried in a vacuum
desiccator and used without further purification.
4-Hydroxy-4-(4′-nitrophenyl)-(4-2H)butan-2-one. 4-Nitro-(R-2H)-
benzaldehyde (1 g, 6.5 mmol) was dissolved in acetone 20 mL at 0 °C
and 1% aqueous NaOH (0.5 mL) was added. After 10 min the mixture
was neutralized with aqueous HCl and concentrated in vacuo. The
brown residue was dissolved in ether (30 mL) and washed with H2O
(30 mL). Drying over MgSO4, concentration in vacuo and flash
1
chromatography on silica gave 0.8 g (57%) of product. H NMR δ
2.11 (s, 3H), 2.76 (m, 2H,), 4.02 (s, 1H), 7.45-8.05 (m, 4H). 13C NMR
δ 208.3, 150.4, 146.5, 126.2, 123.3, 68.0 (t, J(C2H) ) 30.7), 65.5, 51.3.
Isotopic purity > 98%.
4-Fluoro-4-(4′-nitrophenyl)-(4-2H)butan-2-one (12). 4-(4′-Nitro-
phenyl)-4-hydroxy-(4-2H)butan-2-one (0.2 g, 0.93 mmol) was dissolved
in CH2Cl2 (10 mL) and cooled to -78 °C. Then DAST (0.15 mL, 1
mmol) was added, and the mixture was stirred for 20 min, poured into
H2O (30 mL), and extracted with CH2Cl2 (20 mL). After this solution
was dried over MgSO4, a small amount of silica was added to the
solution. The CH2Cl2 was evaporated off, and the silica was applied to
a column prepacked with silica. The product was eluted with Et2O/
1
pentane 1:1. Yield: 0.18 g (90%). H NMR δ 2.21 (s, 3H), 2.87 (dd,
1H, J ) 16.8, 29.2), 3.22 (dd, 1H, J ) 16.8, 16.8), 7.5-8.25 (m, 4H)
13C NMR δ 203.7 (s), 147.8 (s), 146.2 (d, J(CF) ) 20), 126.3 (d, J(CF)
) 6.9), 124.1 (d, J(CF) ) 30.4), 88.7 (dt, J(CF) ) 170, J(C2H) )
30.9), 50.3 (d, J(CF) ) 25), 30.9(s)
1H NMR δ 1.25 (s, 3H), 2.12 (dd, 1H, J ) 5.58, 15.0), 2.34 (s, 3H),
2.45 (dd, 1H, J ) 7.02, 15.0), 3.85 (m, 4H), 5.78 (dd, 1H, J ) 5.53,
7.01), 7.11-7.6 (m, 4H), 7.32-8.06 (m, 4H). 13C NMR δ 148.0, 146.4,
145.0, 134.4, 129.7, 128.0, 127.9, 123.7, 107.9, 79.4, 64.7, 45.8, 24.7,
21.7. Anal. Calcd for C19H21NO7S: C, 56.01%; H, 5.19%; N, 3.44%.
Found: C, 55.88%; H, 5.20%; N, 3.42%.
4-Fluoro-4-(4′-nitrophenyl)-butan-2-one ethylene Acetal (13). 2
(202 mg, 0.8 mmol) was dissolved in CH2Cl2 (10 mL) and cooled to
-78 °C. DAST (154 µL, 0.8 mmol) was added; the mixture was stirred
for 15 min and then poured into H2O (30 mL) and extracted with CH2-
Cl2 (30 mL). The CH2Cl2 phase was dried over MgSO4 and then
concentrated in vacuo. Purification by chromatography on silica using
Et2O/pentane/Et3N 50:50:1 gave 86 mg (43%) of pure product. 1H NMR
δ 1.43 (s, 3H), 2.08 (ddd, 1H, J ) 2.84, 15.22, 34.73), 2.38 (ddd, 1H,
J ) 8.77, 15.14, 18.75), 3.89-4.03 (m, 4H), 5.78 (ddd, 1H, J ) 2.90,
8.71, 48.12), 7.49-8.25 (m, 4H). 13C NMR δ 148.0 (d, J(CF) ) 20),
126.3 (d, J(CF) ) 7.6), 124.0 (s), 108.4 (s), 90.2 (d J(CF) ) 172.2),
64.9 (d, J(CF) ) 4.6), 46.7 (d, J(CF) ) 22.8), 22.9 (d, J(CF) ) 3.1).
Anal. Calcd for C12H14NO4F: C, 56.47%; H, 5.53%; N, 5.49%.
Found: C, 56.64%; H, 5.53%; N, 5.55%.
[4-18F]Fluoro-4-(4′-Nitrophenyl)-butan-2-one (5). The aqueous
[18F]F- solution 1-2 GBq (0.5-1 mL) was added to K2CO3 (1.5-2
mg) and Kryptofix(2.2.2) 3-4 mg in a 3 mL septum covered vial. The
water was removed by repeated azeotropic evaporation with acetonitrile
under a flow of nitrogen at 100 °C. When the vial was completely dry,
the nitrogen flow was stopped, and 3 (3 mg) dissolved in acetonitrile
(0.2 mL) was added. The resulting violet solution was kept at 80-90
°C for 30 min and then cooled to 70 °C. Then p-TsOH in H2O/acetone
1:5 (20 mg/ml, 0.5 mL) was added, and the mixture was kept at 70 °C
for 10 min. The mixture was cooled to rt, H2O (0.5 mL) was added,
and the solution was injected to a semipreparative Phenomenex C-18
HPLC column. Eluent system: acetonitrile, 50 Mm ammonium formate
pH 3.5, 4 mL/min. Gradient: 0-2 min 10% acetonitrile, 2-21 min
10-65% acetonitrile. The fraction between tR 17.6-19.2 was collected.
The collected fraction was diluted with H2O (10 mL) and passed through
a Supelco Supelclean ENVI-18 SPE column. The column was washed
with H2O (3 mL) and then eluted with acetonitrile (2 mL). The
acetonitrile was removed in vacuo, and the residue was dissolved in
H2O (0.5 mL). Typically 150-250 MBq of radiochemically pure 5
was obtained, corresponding to 30-50% decay corrected radiochemical
yield.
Acknowledgment. Professor Bengt Långstro¨m, Uppsala
University PET-center, is gratefully acknowledged for general
radiochemical support and for placing the excellent facilities
of the PET-center at our disposal. Mr. Markus Johnsson is
acknowledged for his contribution to the initial phase of this
investigation. We thank Dr. Nicholas Power for comments on
the manuscript. This project is supported by the Swedish Natural
Science Research Council (NFR).
4-Nitro-(r-2H)benzaldehyde (10). LiAlD4 (0.6 g) was suspended
in diglyme (7 mL), and tert-butyl alcohol (5 mL) was added. The
JA002513B