Photoinitiated carbonylation with [11C]carbon monoxide using amines and alkyl iodides

Article abstract of DOI:10.1021/jo049934m

Photoinitiated radical carbonylation with [¹¹C]carbon monoxide at low concentration was employed in syntheses of carbonyl-¹¹C- labeled amides using alkyl iodides and amines as precursors. Eleven ¹¹C-amides were synthesized

Full text of DOI:10.1021/jo049934m

P h otoin itia ted Ca r bon yla tion w ith [11C]Ca r bon Mon oxid e Usin g
Am in es a n d Alk yl Iod id es
Oleksiy Itsenko,† Tor Kihlberg,‡ and Bengt Långstro¨m*,†,‡
Department of Organic Chemistry, Institute of Chemistry, Uppsala University, Box 531,
S-751 21 Uppsala, Sweden, and Uppsala Imanet AB, Box 967, S-751 09 Uppsala, Sweden
bengt.langstrom@uppsala.imanet.se
Received J anuary 9, 2004
Photoinitiated radical carbonylation with [11C]carbon monoxide at low concentration was employed
in syntheses of carbonyl-11C-labeled amides using alkyl iodides and amines as precursors. Eleven
11C-amides were synthesized in up to 74% decay-corrected radiochemical yields with reaction times
of 400 s and with up to 95% conversion of carbon monoxide. Starting with 26.3 GBq of [11C]carbon
monoxide, 10.6 GBq of 1-cyclohexane [11C]carbonyl-4-phenyl-piperazine (15) was obtained within
35 min from the end of bombardment (33 µA) and with a specific radioactivity of 192 GBq/µmol at
the same time point. The influence of solvents was investigated. The described procedure extends
the range of accessible labeling methods. The method may also be useful for preparation of 13C-
and 14C-substituted compounds.
In tr od u ction
The extensive and successful studies by Sonoda and
Ryu3 on radical carbonylation of alkyl iodides using
carbon monoxide inspired us to adapt this approach to
11C-labeling. To run reactions on the microscale utilizing
submicromolar amounts of [11C]carbon monoxide, a spe-
cial small-scale apparatus was developed. The apparatus
consists of a small autoclave equipped with a sapphire
window and a mercury lamp with a focusing mirror. This
device was used in the present work to establish the
principles for photoinduced labeling chemistry employing
[11C]carbon monoxide. This project is still in progress both
with regard to chemistry research and technical develop-
ment. The synthetic methods and results using sub-
micromolar quantities of [11C]carbon monoxide differ from
those previously described with carbon monoxide at 20-
25 atm and will be explained in this paper dealing with
11C-labeled amides.
This paper describes utilization of a fixed reaction time
in the labeling synthesis. To achieve the best possible
radiochemical yield in a specific labeling synthesis, an
important factor is maximal conversion of carbon mon-
oxide, which increases with reaction time. On the other
hand, the short lifetime of [11C]carbon should also be
taken into account.4 Considering this and other practical
matters related to our experimental setup, a reaction
time of 400 s was selected in order to simplify the
comparison of experimental results. Generally, the op-
timal reaction time for each synthesis is determined by
taking into account the reaction kinetics, the character-
istics of the light source, scale of synthesis, etc.
With increasing utilization of positron emission tomog-
raphy (PET) as a noninvasive molecular imaging tech-
nique in medical and biological research and especially
as a tool in drug development,1 there is a growing
demand for new tracers and consequently new methods
for preparation of radioactive tracers. The labeling of
biologically active compounds with 11C imposes the
requirement of fast reactions, preferably by one-pot
procedures, as dictated by the short half-life of 11C (t1/2
) 20.3 min). The submicromolar amounts of labeled
substance often put restraints on the reaction conditions,
but this may sometimes be turned into an advantage.
Recent technical development for the handling and use
of [11C]carbon monoxide has made this compound useful
in labeling syntheses. [11C]Carbon monoxide may be
obtained in high radiochemical yields from cyclotron-
produced [11C]carbon dioxide and used to yield target
compounds with high specific radioactivities. A broad
range of carbonyl compounds has been labeled using
either palladium- or selenium-mediated reactions.2 The
use of transition-metal-mediated reactions is restricted
by problems related to the competing â-hydride elimina-
tion reaction excluding or at least severely limiting
utilization of organo electrophiles having a hydrogen
atom in the â-position. We have therefore explored other
methods in order to circumvent the problem with â-
hydride elimination and to widen the range of structures
accessible for labeling.
A stimulus for the work was the objective to synthesize
* Corresponding author.
a range of 11C-carbonyl labeled WAY-1003655 analogues
† Uppsala University.
‡ Uppsala Imanet AB.
(1) Bergstro¨m, M.; Grahne´n, A.; and Långstro¨m, B. Eur. J . Phar-
macol. 2003, 357-366.
(2) (a) Kihlberg, T.; Långstro¨m, B. J . Org. Chem. 1999, 64, 9201-
9205. (b) Kihlberg, T.; Karimi, F.; Långstro¨m, B. J . Org. Chem. 2002,
67, 3687-3692.
(3) Ryu, I.; Nagahara, K.; Kambe, N.; Sonoda, N.; Kreimerman, S.;
Komatsu, M. Chem. Commun. 1998, 1953-1954.
(4) Långstro¨m, B.; Kihlberg, T.; Bergstro¨m, M.; Antoni, G.; Bjo¨rk-
man, M.; Forngren, B. H.; Forngren, T.; Hartvig, P.; Markides, K.;
Yngve, U.; O¨ gren, M. Acta Chem. Scand. 1999, 651-669.
10.1021/jo049934m CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/02/2004
4356
J . Org. Chem. 2004, 69, 4356-4360

Photoinitiated 11C-Carbonylation
The conversion of carbon monoxide to products was
indeed favored when the syntheses were carried out in
more polar solvents. Thus, the conversion was about 30%
with acetone and acetonitrile and about 85% with DMF
and DMSO (Table 1, entry 2). However, in the case of
DMF and DMSO the amounts of labeled side-products
also increased (25-40% estimated by LC) which may be
the result of the ability of these solvents to serve as
hydrogen donors. When 1-methyl-2-pyrrolidinone (NMP)
and 1,3-dimethyl-2-imidazolidinone (DMI) were used as
solvents the conversion of carbon monoxide was still high
and the purity of the crude product was increased (Table
1, entries 1 and 3).
F IGURE 1. Halides used in the study.
Attempts to label compounds 17 and 18 using DMSO
as solvent resulted in radiochemical yields below 4%. The
reason is probably that the nucleophilicity of aniline (13)
and 2-aminopyridine (14) is not sufficient to trap the acyl
iodide efficiently. Changing the solvent to NMP resulted
in 44% radiochemical yield of amide 17 (Table 1, entry
5) and 2% of 18 (Table 1, entry 6) correspondingly. It is
worth mentioning that the amine WAY-100634 showed
a reactivity similar to that for 14.
The reaction of iodo alcohol 7 with [11C]carbon mon-
oxide in the presence of amine 10 led to the prevailent
formation of amide 24 over butyrolactone via intra-
molecular acylation (Table 1, entry 13). This selectivity
may serve as evidence for the importance of nucleo-
philicity in determining the synthesis outcome.
Carboxamidation of iodomethane in NMP and aceto-
nitrile gave only 1% isolated yield of 20 (Table 1, entry
8). The conversion of [11C]carbon monoxide in n-hexane
increased to 27% (yield 17%), and in THF the conversion
and the yield reached 11% and 8%, respectively. In THF,
the purity of the crude product determined by HPLC was
the best (91%). The highest decay-corrected radiochemical
yield was obtained in ethanol (Table 1, entry 9).
When amino alcohol 12 was used, the acylation pro-
ceeded almost exclusively on the nitrogen atom (Table
1, entry 10). With 2-iodoethylbenzene 4 the conversion
of [11C]carbon monoxide was markedly lower compared
to syntheses in which iodides 1 and 2 were used (Table
1, entry 10). The major unlabeled byproduct in the
synthesis of 21 was identified as styrene. The elimination
pathway leading to isobutene was probably significant
also in the synthesis of 23.
Phenyl iodide gave a rather low decay-corrected radio-
chemical yield in reaction with amine 10 (Table 1, entry
14). This was probably a result of photodehalogenation8
of phenyl iodide since the homolysis of aryl iodides is a
facile process9 and decarbonylation of PhCO radical is
slow.7
Several organobromides were compared with the cor-
responding iodides in the labeling reactions. Generally,
the consumption of [11C]carbon monoxide reached rela-
tively high levels while the purity of the crude product
was low with several side-products. For example, the
conversion of [11C]carbon monoxide in preparation of 15
from cyclohexyl bromide in DMI was 70% while LC purity
of the crude product was only 18%. When cyclohexyl
F IGURE 2. Amines used in the study.
for biological screening. The palladium-mediated syn-
theses worked satisfactorily for labeling of the analogues
possessing arylcarboxamide moieties,6 but there was a
need also to include alkylcarboxamides (e.g., cyclohexyl-
carboxamide as in the case of WAY-100365 itself).
Resu lts a n d Discu ssion
The scope and limitation study of radical carbonylation
using submicromolar amounts of [11C]carbon monoxide
was performed using the alkyl halides (Figure 1) and
amines (Figure 2), with the aim of obtaining the labeled
amides shown in Table 1.
The reactions were carried out in a 270 µL stainless
steel reaction vessel equipped with a sapphire window.
The reactor was filled with [11C]carbon monoxide in
helium. Then the solution of an alkyl halide together with
an amine was transferred to the reactor with a pressure
of 35 MPa and the reaction mixture irradiated with
focused light from a medium-pressure Hg lamp (400 W)
during 400 s. The amount of [11C]carbon monoxide used
in these reactions was approximately 10-8 mol (10-9
L
at 35 MPa), corresponding to a partial pressure of 200
Pa. After the synthesis, the crude mixture was evacuated
from the reactor into a collection vial. The amount of
[11C]carbon monoxide consumed in the reaction was
estimated by measurements of radioactivity before and
after purge of the gas phase in the collection vial. A
sample from the crude reaction mixture was withdrawn
for HPLC analysis, and the remaining material was
purified by HPLC.
The decay-corrected radiochemical yields of the target
compounds were found to depend on the solvent, the
nucleophilicity of the amine, and the structure of the
organo halide. When low polar solvents (e.g., n-hexane
and cyclohexane) were used, the conversion of [11C]carbon
monoxide did not exceed 5%. It was therefore anticipated
that more polar solvents were needed to stabilize the acyl
radicals to some extent and facilitate the acylation step,
thus pushing the reaction in the direction of the desired
path.7
(7) Chatgilialoglu, C.; Crich, D.; Komatsu, M.; Ryu, I. Chem. Rev.
1999, 99, 1991-2069.
(8) Wolf, W.; Kharasch, N. J . Org. Chem. 1965, 30, 2493-2498.
(9) Davidson, R. S.; Goodin, J . W.; Kemp, G. Adv. Phys. Org. Chem.
1984, 191-233.
(5) N-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinyl)-
cycloxexanecarboxamide: effective radioligand for imaging of 5-HT1A
receptor.
(6) Results to be published elsewhere.
J . Org. Chem, Vol. 69, No. 13, 2004 4357

Itsenko et al.
TABLE 1. Tr a p p in g Efficien cy a n d Ra d ioch em ica l Yield s of 11C-La beled Am id es (11C La belin g P osition Ma r k ed w ith *)
a
b
Decay-corrected, the fraction of radioactivity left in the crude product after purge with nitrogen. Radiochemical yield; decay-corrected,
calculated from the amount of radioactivity in the crude product before nitrogen purge, and the radioactivity of the LC purified product.
d
c Radiochemical yield; decay-corrected, calculated from analytical HPLC. Number of runs.
4358 J . Org. Chem., Vol. 69, No. 13, 2004

Photoinitiated 11C-Carbonylation
TABLE 2. Dep en d en ce of Con ver sion a n d
was synthesized with a specific radioactivity of 192 GBq/
µmol in a 35 min synthesis time after the end of
bombardment (33 µAh).
Ra d ioch em ica l Yield on th e Con cen tr a tion of Rea gen tsa
conversion of
concn of
1 (mM)
concn of
10 (mM)
[
11C]carbon
LC purity
(%)
yieldc
(%)
The basic identification of the labeled compounds was
performed by comparison of analytical LC retention times
for the labeled and reference compounds. The authentic-
ity of the labeled compounds was verified by LC-MS
analysis. LC-MS Chromatograms recorded at SIR M +
1 m/z for each of the labeled compounds 15-25 featured
peaks with the same retention time as the chromatogram
of the corresponding unlabeled reference substance. All
reference compounds were prepared via alternative
synthetic routes and characterized via MS and NMR. To
collect more solid evidence the synthesis of compound 15
was scaled up. The reaction was performed in the same
manner using 866 µL of carbon monoxide of natural
isotopic distribution. The amounts of the precursors were
correspondingly increased and the reaction time pro-
longed to 20 min. In this case, [11C]carbon monoxide was
monoxide (%)b
0
1
2
65
65
65
65
65
65
65
3
12
36
40
36
83
86
91
40
59
0
5
0
2
16
39
63
75
72
5
6
8
14
52
64
66
3
15
39
77
77
8
7
10
6
a
All reactions were performed in DMSO with a reaction time
b
400 s. Decay-corrected, the fraction of radioactivity left in the
crude product after purge with nitrogen. c Radiochemical yield;
decay-corrected, calculated from analytical HPLC.
iodide was used, the conversion of [11C]carbon monoxide
was 85% and the LC-determined purity was 69%.
In contrast to the corresponding palladium-mediated
reaction with [11C]carbon monoxide,10 the free-radical
reactions showed a notable dependence on efficient
stirring of the reaction mixture. In our previous works
on palladium- and selenium-mediated carbonylations,
nearly quantitative conversions of [11C]carbon monoxide
in several cases were obtained without stirring.11 The
conversion of [11C]carbon monoxide in the synthesis of
15 increased from 29 ( 4% to 85 ( 7% when stirring was
introduced. Likewise, a reduction in pressure from 35
MPa to 10 MPa resulted in a 3-fold decrease of the
conversion.
Proper selection of the photoirradiation conditions for
generation of radicals was important in optimization of
the synthesis. When visible light (a 250 W halogen lamp
with UV-stop glass) was used, no labeled product was
detected. A negative result was also obtained when the
Hg lamp was used with a filter absorbing the 190-420
nm spectral range. The photoirradiation intensity was
also important. Lower irradiation intensity gave lower
decay-corrected radiochemical yields.
Although the changes of solvent, temperature, stirring,
and type of organo halide all had a pronounced effect on
the decay-corrected radiochemical yield the basic product
pattern remained roughly the same.
The reduction of the amount of reagents used is
sometimes desirable in order to facilitate purification and
reduce costs. In Table 2, the dependence of the decay-
corrected radiochemical yield of compound 15 on this
factor is presented. The amounts of side products in-
creased when the concentration of substrates reached a
certain threshold.
1
used as a tracer to track the target compound. The H
NMR spectrum of the isolated compound was identical
with the spectrum of the reference compound prepared
from the appropriate amide and acid chloride.
To verify the labeling position, a 13C-substituted ana-
logue of 22 was synthesized. (13C)Carbon monoxide was
added to the [11C]carbon monoxide and the reaction was
performed as a standard 11C-labeling except that the
amounts of the alkyl iodide and amine were increased
and reaction time was extended to 31 min. Calculated
from carbon monoxide using radioactivity measurements,
the isolated yield of 13C-substituted 22 was 41%. The
conversion of (13C)carbon monoxide was 68%. 13C NMR
analysis was used to assign the labeling position. The
identity of the compound was further confirmed by the
1H NMR spectrum, which showed the characteristic
1
splittings due to H-13C coupling.
Con clu sion s
This work confirms the suitability of radical-mediated
carbonylation for the synthesis of labeled amides starting
from [11C]carbon monoxide, alkyl iodides, and amines.
This method may provide an important complement to
the previously described palladium-mediated reactions
using [11C]carbon monoxide and allows the preparation
of target structures incompatible with Grignard synthe-
sis. The labeled compounds had high specific radio-
activities, which is important in PET studies.
This method may also be valuable for synthesis with
(13C)carbon monoxide.
Exp er im en ta l Section
High specific radioactivity is an important issue in PET
studies where amount of tracer should be minimized to
keep the biological system unperturbed. Compound 15
[11C]Carbon dioxide production was performed using a
Scanditronix MC-17 cyclotron at Uppsala Imanet. The
14N(p,R)11C reaction was employed in a gas target containing
nitrogen (Nitrogen 6.0) and 0.1% oxygen (Oxygen 4.8), which
was bombarded with 17 MeV protons. [11C]Carbon monoxide
was obtained by reduction of [11C]carbon dioxide as described
previously.12 The syntheses with [11C]carbon dioxide were
performed with an automated module as part of the system
“Synthia 2000”. Liquid chromatographic analysis (LC) was
(10) Blaser, H.-U. J . Org. Chem. 2003, 68, 3725-3728.
(11) (a) Karimi, F.; Kihlberg, T.; Långstro¨m, B. J . Chem. Soc., Perkin
Trans. 1 2001, 1528-1531. (b) Karimi, F.; Långstro¨m, B. J . Chem.
Soc., Perkin Trans. 1 2002, 2111-2116. (c) Karimi, F.; Långstro¨m, B.
J . Chem. Soc., Perkin Trans. 1 2002, 2256-2259. (d) Rahman, O.;
Kihlberg, T.; Långstro¨m, B. J . Chem. Soc., Perkin Trans. 1 2002, 2699-
2703. (e) Rahman, O.; Kihlberg, T.; Långstro¨m, B. J . Org. Chem. 2003,
68, 3558-3562. (f) Karimi, F.; Lånstro¨m, B. Org. Biomol. Chem. 2003,
541-546. (g) Karimi, F.; Långstro¨m, B. Eur. J . Org. Chem. 2003, 68,
2132-2137. (h) Karimi, F.; Långstro¨m, B. J . Labelled Compd. Radio-
pharm. 2002, 423-434.
(12) Kihlberg, T.; Långstro¨m, B. Method and apparatus for produc-
tion and use of [11C]carbon monoxide in labeling synthesis. PCT
International Application PCT/SE02/01222.
J . Org. Chem, Vol. 69, No. 13, 2004 4359

Itsenko et al.
performed with a gradient pump and a variable-wavelength
UV detector in series with a â+-flow detector. The following
mobile phases were used: 25 mM potassium dihydrogenphos-
phate (A) and acetonitrile/H2O 50/7 (B). For analytical LC, a
mmol) in dichloromethane (5 mL) was added the acid chloride
(2 mmol) dropwise. The reaction mixture was allowed to warm
to room temperature and was then stirred for 1 h. Water was
added and the mixture extracted with dichloromethane three
times. The combined extract was dried over MgSO4, and the
solvent was then evaporated under reduced pressure. The
crude product was purified with flash chromatography or
recrystallized.
1-Cycloh exa n eca r bon yl-4-p h en ylp ip er a zin e (15). A col-
orless oil was obtained after flash chromatography, yield 92%.
N-Isop r op ylcycloh exa n eca r boxa m id e (16). Recrystal-
lization from dichloromethane/pentane gave white crystals,
yield 89%.
Cycloh exa n eca r boxa n ilid e (17). Recrystallization from
dichloromethane/pentane gave white crystals, yield 88%.
N-2-P yr id ylcycloh exa n eca r boxa m id e (18). Purified by
flash chromatography and recrystallized from dichloromethane/
pentane: pale yellow crystals; yield 52%.
1-Non a n oyl-4-p h en ylp ip er a zin e (19). Recrystallization
from dichloromethane/diethyl ether gave white crystals, yield
81%.
C
18, 3 µm, 50 × 4.6 mm i.d. column was used at a flow of 2
mL/min. For semipreparative LC, a C18, 4 µm, 250 × 10 mm
(i.d.) column was used at a flow of 4 mL/min. An automated
synthesis system, Synthia,13 was used for LC injection and
fraction collection. Radioactivity was measured in an ion
chamber. A Philips HOK 4/120SE mercury lamp was used as
photoirradiation source. In the analysis of the 11C-labeled
compounds, unlabeled reference substances were used for
comparison in all the LC runs. NMR spectra were recorded at
400 MHz for 1H and at 100 MHz for 13C, at 25 °C. Chemical
shifts were referenced to TMS via the solvent signals. LC-
MS analysis was performed with electrospray ionization.
THF was distilled under nitrogen from sodium/benzo-
phenone. DMSO was purged with helium for 5 min before use.
Other solvents were used as supplied without further purifica-
tion. Compound 7 was synthesized from the corresponding
bromide by the Finkelstein reaction. All other starting materi-
als were commercially available.
1-Acetyl-4-p h en ylp ip er a zin e (20). The crude product was
purified by column chromatography and recrystallized from
dichloromethane/pentane to yield 88% of white crystals.
N-(2-Hyd r oxyeth yl)-3-p h en ylp r op ion a m id e (21). The
amino alcohol 12 (15 mmol) was dissolved in dichloromethane
(20 mL). The solution was cooled in an ice bath, and a solution
of 3-phenylpropionyl chloride (3.4 mmol) in dichloromethane
(10 mL) was added dropwise. The reaction mixture was
allowed to warm to room temperature and then stirred
overnight. The dichloromethane phase was separated, added
to water (30 mL) acidified to pH 3 with concentrated hydro-
chloric acid, and extracted twice with dichloromethane. The
combined extract was dried over MgSO4 and evaporated at
reduced pressure. Recrystallization from dichloromethane/
pentane gave white crystals, yield 81%.
1-Isobu ta n oyl-4-p h en ylp ip er a zin e (22). The crude prod-
uct was purified by column chromatography and recrystallized
from diethyl ether/pentane to yield 82% of white crystals.
1-P iva loyl-4-p h en ylp ip er a zin e (23). The crude product
was purified by column chromatography and recrystallized
from diethyl ether/pentane to yield 78% of pale yellow crystals.
1-(4-Hyd r oxybu ta n oyl)-4-p h en ylp ip er a zin e (24). A neat
mixture of butyrolactone (1.5 mmol) and 1-phenylpiperazine
(1.5 mmol) was heated at 130 °C overnight with stirring. The
resulting crude product was cooled to ambient temperature,
partitioned between dichloromethane and sodium hydrogen
carbonate solution, and extracted thrice with dichloromethane.
The combined extract was dried over MgSO4 and evaporated
at reduced pressure. The crude product was purified by column
chromatography to yield 63% of a colorless oil.
All reactions were carried out in a high-pressure reactor.
The reactor comprises a 270 µL cylindrical cavity in a stainless
steel tube. One end of the cavity is sealed and outfitted with
two tubes of smaller diameter that serve as filling and
evacuation ports. A sapphire window is attached to the
opposite end of the cavity. A thermocouple is attached to the
outer body of the reactor and serves to control the reaction
temperature. An externally driven Teflon-coated magnet bar
was used to stir the reaction mixture.
P r ep a r a tion of 11C-La beled Am id es. A capped vial (1
mL) was flushed with nitrogen and was charged with an amine
(50 µmol), triethylamine (100 µmol), and a solvent (500 µL).
An organo iodide (50 µmol) was added to the solution roughly
7 min before synthesis. The resulting mixture was pressurized
(35 MPa) into the microautoclave (270 µL), precharged with
[11C]carbon monoxide in He at ambient temperature. The
autoclave was then irradiated with a mercury lamp for 400 s.
The crude reaction mixture was then transferred from the
autoclave to a capped vial (1 mL) held under reduced pressure.
After measurement of the radioactivity, the vial was purged
with nitrogen and the radioactivity was measured again. The
crude product was diluted with acetonitrile (0.55 to 1 mL) and
injected onto the semipreparative LC. Analytical LC and LC-
MS were used to assess the identity and radiochemical purity
of the collected fraction.
(
13C-Ca r bon yl)-1-isobu ta n oylp ip er a zin e (22). A capped
vial (1 mL) was flushed with nitrogen and charged with
1-phenylpiperazine (165 µmol), triethylamine (180 µmol), and
NMP (500 µL). Isopropyl iodide (250 µmol) was added to the
solution roughly 7 min before synthesis. The resulting mixture
was pressurized (35 MPa) into the microautoclave (270 µL),
precharged with [11C]carbon monoxide in He and (13C)carbon
monoxide (40 µmol) at ambient temperature. The autoclave
was then irradiated with a mercury lamp for 31 min. The crude
reaction mixture was then transferred from the autoclave to
a capped vial (1 mL) held under reduced pressure. After
measurement of the radioactivity, the vial was purged with
nitrogen and the radioactivity was measured again. The crude
product was diluted with acetonitrile (0.55 mL) and injected
onto the semipreparative LC.
1-Ben zoyl-4-p h en ylp ip er a zin e (25). The crude product
was purified by column chromatography and recrystallized
from diethyl ether/pentane to yield 85% of white crystals.
Ack n ow led gm en t. We thank Prof. Wyn Brown for
linguistic advice and Tommy Ferm for skillful technical
assistance. The Swedish Research Council is acknowl-
edged for its support through Grant No. 621-2003-2855
(B.L.).
Gen er a l P r oced u r e for P r ep a r a tion of Am id es 15-20,
22, 23, a n d 25 Used a s Refer en ce Com p ou n d s. To an ice-
cooled solution of an amine (2 mmol) and triethylamine (3
1
Su p p or tin g In for m a tion Ava ila ble: Melting points, H
and 13C NMR, and MS data for reference compounds 15-23,
(carbonyl-13C)15, and (carbonyl-13C)22. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
(13) Bjurling, P.; Reineck, R.; Westerberg, G.; Gee, A. D.; Sutcliffe,
J .; Långstro¨m, B. Proceedings of the VIth workshop on targetry and
target chemistry; TRIUMF: Vancouver, Canada, 1995, pp 282-284.
J O049934M
4360 J . Org. Chem., Vol. 69, No. 13, 2004