Process optimization and synthesis of 3-(4-fluorophenyl)-4,5-dihydro-N-[4-(trifluoromethyl)phenyl]-4-[5- (trifluoromethyl)-2-pyridyl]-1H-pyrazole-1-carboxamide

Article abstract of DOI:10.1021/op025619v

A convenient, high-yielding synthesis of the insecticide 3-(4-fluorophenyl)-4,5-dihydro-N-[4-(trifluoromethyl)phenyl]-4-[5- (trifluoromethyl)-2-pyridyl]-1H-pyrazole-1-carboxamide is presented. Attempted methenylation of the 2-pyridylacetophenone, obtained

Full text of DOI:10.1021/op025619v

Organic Process Research & Development 2003, 7, 267−271
Process Optimization and Synthesis of 3-(4-Fluorophenyl)-4,5-dihydro-N-[4-(tri-
fluoromethyl)phenyl]-4-[5-(trifluoromethyl)-2-pyridyl]-1H-pyrazole-1-carboxamide
James M. Renga,* Kevin L. McLaren, and Michael J. Ricks
DiscoVery Process Research, Dow AgroSciences, 9410 ZionsVille Road, Indianapolis, Indiana 46268-1053, U.S.A.
Scheme 1
A convenient, high-yielding synthesis of the insecticide
3-(4-fluorophenyl)-4,5-dihydro-N-[4-(trifluoromethyl)phenyl]-
4-[5-(trifluoromethyl)-2-pyridyl]-1H-pyrazole-1-carboxam-
ide is presented. Attempted methenylation of the 2-pyridy-
lacetophenone, obtained from the alkylation of 4-fluoroaceto-
phenone with 2-chloro-5-trifluoromethylpyridine, under stan-
dard Mannich conditions resulted in the formation of a
methylene-bridged dimer. In addition, the required 2-pyridyl-
propenone underwent a rearrangement to an acylhydrazine
when subjected to hydrazine in a solvent-dependent process.
These side reactions were circumvented by the sequential
reaction of bis(dimethylamino)methane, trifluoroacetic acid,
hydrazine, and arylisocyanate with the 2-pyridylacetophenone
in methylene chloride. This one-pot process for preparing
the pyrazoline-1-carboxamide enabled rapid and economical
scale-up for field and toxicological studies.
Scheme 2
Introduction
3-(4-Fluorophenyl)-4,5-dihydro-N-[4-(trifluoromethyl)-
phenyl]-4-[5-(trifluoromethyl)-2-pyridyl]-1H-pyrazole-1-car-
boxamide (1) has potent contact and foliar activity against
both lepidoptera and orthoptera insects.1 Extensive SAR
studies focused on varying the heterocycle at position 4
resulted in the identification of pyrazoline-1-carboxamide 1
as a potential candidate for commercialization.2 While the
synthesis of 3,4,N-tri(substituted)aryl compounds (2) is well-
documented and proceeds in good yields (Scheme 1),3,4 the
inclusion of a electron-deficient heterocycle in the 4 position
brought with it the need to dramatically alter the conditions
and reagents used in the diaryl case.5 Our interest to expand
the SAR as well as to scale up active analogues for initial
field studies led us to explore in detail the reaction pathway.
The results of this process optimization directed toward the
large-scale synthesis of 1 are presented.
Results and Discussion
Process Optimization. Alkylation of the enolate of
4-fluoroacetophenone (3) with a 2-halopyridine 4 proved to
be the route of choice for the synthesis of ethanone 5a
(Scheme 2). We made numerous attempts to minimize the
production of the corresponding undesired enol ether 6 by
manipulation of the reaction conditions. The impact of
changing the rate, the order of addition, the countercation
of the enolate, or the temperature had only subtle effects
upon the product ratio. However, the choice of solvent had
a dramatic effect on the ratio of 5a to 6 (Table 1). Highly
solvating solvents, such as DMSO, promoted reaction at
oxygen, while THF enhanced reaction at carbon. Solvents
less polar than THF did not offer an advantage due to the
substantially longer reaction times required for acceptable
(2) (a) McLaren, K. L.; Hertlein, M. B.; Pechacek, J. T.; Ricks, M. J.; Tong,
Y. C.; Karr, L. L. Eur. Pat. Appl. 508469; Chem. Abstr. 1993, 118, 101947.
(b) Renga, J. M.; McLaren, K. L.; Pechacek, J. T.; Ricks, M. J.; Tong, Y.
C. U.S. Patent 5,324,837; Chem. Abstr. 1994, 121, 205339. (c) Ricks, M.
J.; Tong, Y. C. U.S. Patent 5,338,856; Chem. Abstr. 1994, 121, 300887.
(3) Neh, H.; Bu¨hmann, U.; Joppien, H.; Giles, D. Ger. Offen. DE 3638631;
Chem. Abstr. 1988, 109, 93002.
(4) Chan, D. M. T.; Stevenson, T. M.; Piotrowski, D. W.; Harrison, C. R.;
Fahmy, M. A. H.; Lowe, R. L.; Monaco, K. L.; Reeves, B. M.; Folgar, M.
P.; Esrey, E. G ACS Symp. Ser. 2002, 800(VI), 144.
* Author for correspondence. E-mail: jmrenga@dow.com. Telephone: 317-
337-3096. Fax: 317-337-3249.
(1) Leonard, P. K.; Hertlein, M. B.; Thompson, G. D.; Paroonagian, D. L. BCPC
Monograph 1994, 59, 67.
(5) Fuchs, R.; Gallenkamp, B.; Erdelen, C.; Maurer, F.; Wada, K.; Grosser,
R.; Born, L.; Goehrt, A. Pesticide Science 1997, 50, 333.
10.1021/op025619v CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/02/2003
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
267

Table 1. Formation of 5a from 4
While two equivalents of trifluoroacetic acid (TFA) in
methylene chloride as a cosolvent were typically sufficient
to favor 13a, neat TFA was required to optimize the
formation of 13b. The analogous rearrangement product was
not observed in either ethanol or DMF when a phenyl ring
replaced the pyridine ring.
solvent
5a:6a
DMSO
THF
THF
4a
4a
4b
4b
36:64
78:22
95:<5
98:<2
DME
Scheme 3 illustrates a possible explanation for the
observed selectivity combined with a proposed mechanism
by which the two products are formed. Initially, hydrazine
adds in a Michael sense to propenone 8 to give the
â-hydrazinopropanone 10, which readily cyclizes onto the
ketone to form the intermediate hydroxypyrazolidine 11.
Under acidic conditions protonation of the hydroxy moiety
present on 11 would cause rapid dehydration to the desired
pyrazoline 13. Alternatively, basic conditions would allow
the oxygen to undergo a retro aldol-type fragmentation to
form the undesired acyl hydrazide 12. It is believed that polar
and highly dielectric solvents favor side-product formation
via their ability to separate charge, thereby acting as a general
base. It is for this reason that the excessive DMA formed in
conjunction with propenone 8 must be thoroughly removed.
Finally, the successful use of TFA as a cosolvent to direct
the formation of 13a arises from the observation that
hydroxypyrazolidine 11a is stable in solution in chlorinated
a Determined by GC.
conversion. In general, using DME only improved the yield
of 5a slightly without evidence for substantial reaction at
oxygen and proved more useful for the synthesis of
analogues.2
A change in the leaving group from fluorine to chlorine
tended to favor reaction at carbon. Consistent isolated yields
of 60-65% of ethanone 5a were obtained on a 0.1-0.25-
mol scale by forming the enolate of 3 prior to adding 4b.
On a 2-mol scale the concurrent addition of 3 and 4b to a
suspension of NaH in THF gave the best results (65%
recrystallized yield).
Treatment of 5a under standard Mannich conditions (Me2-
NH‚HCl, (CH2O)n, EtOH) initially gave the desired Mannich
adduct 7a (Scheme 3). However, this intermediate was highly
activated by the presence of the pyridine ring and readily
eliminated dimethylamine to form propenone 8a. The ensuing
Michael reaction between 8a and the enol form of 5a is more
rapid than the Mannich reaction. Essentially only the
resulting methylene-bridged dimer 9a was obtained in 81%
yield. This dimerization has not been reported during the
formation of diaryl propenones (Scheme 1). While the
formation of dimer 9a is reversible,6 attempts to regenerate
propenone 8a in situ resulted in a low recovery of material.
Switching to the more reactive formaldehyde equivalent
bis(dimethylamino)methane (BDAM)7 in the presence of
acetic anhydride cleanly gave the â-dimethylaminopropanone
7a at -15 °C. While intermediate 7a is isolable, adding a
second equivalent of acetic anhydride effects elimination in
situ to the desired propenone 8a in 99% crude yield.
Unfortunately, the significant amount of dimethylacetamide
(DMA) byproduct must be rigorously removed since its
presence affects pyrazoline formation in the next step.
Attempts to purify 8a by chromatography resulted in the
formation of 9a presumably from the generation of 5a via
the Michael addition of water followed by a retro Aldol
reaction.
1
solvents and may be observed by H NMR in the absence
of TFA.8 There is no evidence that the two products acyl
hydrazide 12a and pyrazoline 13a are in equilibrium.9
Intermediate 13a was also unstable to ambient conditions10
and must be acylated immediately with 4-(trifluoromethyl)-
phenyl isocyanate to give pyrazoline-1-carboxamide 1.
One-Pot Synthesis. Although quite functional, the pre-
ceding synthesis of pyrazoline-1-carboxamide 1 from etha-
none 5a suffers from several limitations. Isolation and
purification of both the unstable propenone 8a and the
pyrazoline 13a were problematic and resulted in lower
overall yields. In addition, a large-scale preparation of
pyrazoline-1-carboxamide 1 would require excess reagents
to achieve competitive formation of propenone 8a, the need
to remove excess DMA, and the inconvenient use of TFA
as a cosolvent. We therefore looked to develop a more
efficient sequence that would address these limitations.
We found that the use of acetic anhydride was unneces-
sary to release the formaldehyde moiety in the Mannich
reaction of 5a with BDAM. Instead the addition of one
equivalent of TFA to a solution of 5a and BDAM in
methylene chloride at 0 °C gave â-dimethylaminopropanone
7a. Anhydrous hydrazine was added, and the reaction was
warmed to room temperature to give the pyrazoline 13a.
The one equivalent of TFA present from the formation
of the Mannich adduct 7a was sufficient to favor the
formation of 13a (Table 2, entry 8). However, the addition
of two equivalents of TFA to the TFA salt of 7b prior to the
addition of three equivalents of hydrazine did not fully
Propenones readily react with hydrazine to form pyrazo-
lines (Scheme 1).3 However, under typical polar conditions
(Table 2) the desired pyrazoline 13 was not observed; instead,
the major product was the acyl hydrazide 12 (Scheme 3).
The selectivity for 12 and 13 was strongly influenced by
the properties of the solvent. Polar solvents and basic solvents
favored the rearrangement to 12 (Table 2, entries 1-5), while
weakly basic solvents (Table 2, entry 6) and acidic solvents
(Table 2, entry 10) favored the formation of pyrazoline 13.
(8) The NMR of 11a in either CDCl3 or CD2Cl2 was highly complex.
(9) Isolated products 12a and 13a were subjected to the reaction conditions
which favored formation of the opposite product, (CH2Cl2, TFA, H2NNH2)
and (DMSO, H2O, H2NNH2), respectively. After extended reaction times,
neither 12a nor 13a showed any interconversion, as determined by 1H NMR.
(10) To avoid oxidative decomposition, the HCl salt can be made and stored at
0 °C.
(6) A dynamic equilibrium between dimer 9a and propenone 8a was demon-
strated using a crossover experiment, wherein 9a was mixed with a
differentially substituted ethanone. Scrambling of the substitution under both
acid and base conditions was observed by 1H NMR over 24 h.
(7) Takahashi, K.; Shimizu, S.; Ogata, M. Synth. Commun. 1987, 17, 809.
268
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development

Scheme 3
Table 2. Dependence of 12:13 upon solvent and pKa
carbamoylation step to the pyrazoline-1-carboxamide 1 was
accomplished after a simple extraction procedure, again
without requiring isolation. The process was used to rapidly
and economically prepare samples for field and toxicological
studies and to develop the SAR.2b,c
entry
solvent
pKaa 12a:13ab 12b:13bb
1
2
3
4
5
6
7
EtOH
DMSO
DMF
Et3N
-
-
95:<5
95:<5
82:18
95:<5
67:33
8:92
95:<5
95:<5
95:<5
-
-
11.0
8.3
5.3
-
morpholine
pyridine
-
Experimental Section
<5:95
71:29
General. All reagents and solvents were used directly as
purchased from commercial suppliers. All reactions were
conducted under nitrogen. Boiling points and melting points
7.AcOH
6:94
CH2Cl2
8
9
7.TFA
-
2:98
-
54:46
25:75
<5:95
CH2Cl2
1
(Pyrex capillary) were uncorrected. H NMR spectra were
7.TFA + 2 equiv TFA
CH2Cl2
-
recorded at 200.1 MHz using a Varian XL-200 spectropho-
tometer or at 400.1 MHz using a Bruker AM-400 spectro-
10
TFA
0.2
<5:95
photometer in CDCl3 as solvent, unless otherwise noted. 13
C
a The listed pKa values are for those solvents in H2O. b Determined by NMR.
NMR spectroscopy was performed on a Bruker AM-400
operating at 100.6 MHz in CDCl3 as solvent. Elemental
analyses and mass spectra (EI 70 meV) were furnished by
R. K. Hallberg at the former research facility of the Dow
Chemical Company in Walnut Creek, CA.
suppress the formation of 12b (Table 2, compare entries 8
and 9). While the reaction medium was actually basic due
to the relatively large amounts of dimethylamine and
hydrazine present, sufficient proton transfer occurred to
catalyze both the conversion of 7 to â-hydrazinopropanone
10 via propenone 8 and the elimination of water from
hydroxypyrazolidine 11.
On a 1-mol scale, the solution of 13a was washed, dried,
and treated with 4-(trifluoromethyl)phenyl isocyanate to a
give a 78% yield of 1 from 5a.
Preparation of 1-(4-Fluorophenyl)-2-(5-trifluoromethyl-
2-pyridyl)ethanone (5a). The mineral oil was removed from
60% NaH/oil (192 g, 4.8 mol) via treatment with hexanes
(3 × 400 mL). The resulting solid was slurried with THF (1
L) and heated to 50-55 °C with mechanical stirring. A
solution of 3 (304 g, 2.20 mol) and 4b (363 g, 2.00 mol) in
THF (400 mL) was added over 2 h. After an additional 10
h at reflux, the mixture was cooled to 0 °C, and excess NaH
was quenched via addition of a solution of MeOH (50 mL)
in Et2O (200 mL) over 1 h. The mixture was partitioned
between Et2O (4 L) and brine (4 L). The organic layer was
washed with brine (2 L). The combined aqueous layers were
washed with Et2O (2 × 1 L). The combined Et2O layers
were dried (Na2SO4), filtered, and evaporated to a tacky dark
Conclusions
Through process optimization a high-yielding synthesis
of pyrazoline 13a was achieved without dimerization of
ethanone 5a or significant formation of acylhydrazide 12a.
Five discrete chemical steps were achieved in one pot without
the need for isolation of any intermediates. The final
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
269

solid (622 g). Kugelrohr (85-95 °C/0.05 mmHg) distillation
gave a yellow solid (469 g). Recrystallization (hexane) gave
368 g (65% yield) of 5a in three crops as yellow needles,
mp 82-83 °C. Calcd for C14H9F4NO: C, 59.37; H, 3.20;
N, 4.95. Found: C, 59.43; H, 3.18; N, 4.83.
2H), meso: 2.60 (ddd, 1H, J ) 7.0, 7.4, 14.2), 3.09 (ddd,
1H, J ) 7.0, 7.4, 14.2), 4.87 (dd, 2H, J ) 7.4, 7.4), 7.05
(dd, 4H, JH-H ) 8.6, JH-F ) 8.6), 7.36 (d, 2H, J ) 8.2),
7.81 (dd, 2H, J ) 2.5, 8.3), 7.99 (dd, 4H, JH-H ) 8.9, JH-F
) 5.4), 8.7 (m, 2H). Calcd for C29H18F8N2O2: C, 60.21; H,
3.14; N, 4.84. Found: C, 59.97; H, 3.16; N, 4.90.
1
Ketone: H NMR δ 4.53 (s, 2H), 7.13 (dd, 2H, J ) 8.5,
8.5), 7.44 (d, 1H, J ) 8.0), 7.9 (m, 1H), 8.08 (dd, 2H, J )
5.4, 8.9), 8.8 (m, 1H).
Table 2. Entry 3 (Similar Experiments Were Done for
Entries 1, 2, 4-6, and 10). To a solution of 8a (0.55 g,
1.86 mmol) in 3 mL of DMF at room temperature was added
hydrazine monohydrate (0.16 mL, 3.3 mmol). After 15 min
the mixture was added to 50 mL of EtOAc and was washed
with a saturated NaCl solution (3 × 50 mL) and dried
(MgSO4) to afford 0.55 g of yellow powder, an 82:18 mixture
of 12a:13a by NMR. Recrystallization from EtOAc/hexane
gave 0.30 g (49% yield) of 12a as white plates, mp 101-
103 °C. MS (m/z) 497 (4), 496 (17), 309 (31), 308 (21), 282
Enol: 1H NMR δ 6.06 (s, 1H), 7.09 (dd, 2H, J ) 8.7,
8.7), 7.13 (d, 1H, J ) 8.5), 7.79 (dd, 1H, J ) 2.4, 7.9), 7.83
(dd, 2H, J ) 5.4, 9.0), 8.6 (m, 1H), 14.97 (s, 1H).
2-(5-Cyano-2-pyridyl)-1-(4-fluorophenyl)ethanone (5b).
Prepared according to 5a from 3 (10.0 mL, 11.4 g, 82.4
mmol), 60% NaH (8.2 g, 210 mmol), and 2-chloro-5-
cyanopyridine (11.4 g, 82.3 mmol) in THF (90 mL) to give
15.3 g (77%) of 5b as fine orange needles, mp 166-167
°C. Calcd for C14H9FN2O: C, 70.00; H, 3.78; N, 11.66.
Found: C, 69.62; H, 3.85; N, 11.36.
Enol: 1H NMR δ 6.05 (s, 1H), 7.1 (m, 3H), 7.76 (dd,
1H, J ) 2.1, 8.5), 7.82 (dd, 2H, J ) 5.4, 8.9), 8.57 (d, 1H,
J ) 1.8), 10.36 (br s, 1H).
Preparation of 1-(4-Fluorophenyl)-2-(5-trifluoromethyl-
2-pyridyl)propenone (8a). A solution of ethanone 5a (2.7
g, 9.5 mmol) and Ac2O (2.0 mL, 2.2 g, 21 mmol) in
methylene chloride (4 mL) was added to a stirred solution
of BDAM (2.2 mL, 1.6 g, 16 mmol) in methylene chloride
(4 mL) at -15 °C. After 5 min the mixture was partitioned
between 1:1 Et2O/pentane and H2O. The organic layer was
washed with water (2 × 50 mL), a 5% solution of HCl (50
mL), a saturated NaCl solution (50 mL), and dried (MgSO4)
to afford 3.0 g (99% yield) of 8a as an orange red oil. MS
(m/z) 295(1), 294 (1), 267 (46), 266 (100), 123 (89), 95 (86),
75 (38). 1H NMR δ 5.89 (s, 1H), 6.72 (s, 1H), 7.12 (dd, 2H,
J ) 8.6, 8.6), 7.62 (d, 1H, J ) 8.0), 7.90 (dd, 1H, J ) 2.5,
8.3), 7.94 (dd, 2H, J ) 5.4, 8.9), 8.8 (m, 1H).
Preparation of 2-(5-Cyano-2-pyridyl)-1-(4-fluorophe-
nyl)propenone (8b). Prepared according to 8a from ethanone
5b (3.04 g, 12.7 mmol) to give 2.54 g (80%) of 8b as a
yellow powder, mp 91-92 °C. 1H NMR δ 5.94 (s, 1H), 6.79
(s, 1H), 7.12 (dd, 2H, J ) 8.6, 8.6), 7.63 (dd, 1H, J ) 0.7,
8.3), 7.9 (m, 3H), 8.82 (dd, 1H, J ) 0.7, 2.1). Calcd for
C15H9FN2O: C, 71.42; H, 3.60; N, 11.11. Found: C, 71.79;
H, 3.86; N, 11.38. Calcd for C15H9F4NO: C, 61.02; H, 3.07;
N, 4.74. Found: C, 59.99, H, 3.20; N, 4.82.
1
(17), 281 (26), 280 (10), 187 (19), 163 (100), 147 (14). H
NMR δ 3.09 (t, 2H, J ) 5.0), 3.38 (t, 2H, J ) 5.0), 7.07 (t,
2H, J ) 6.5), 7.33 (d, 1H, J ) 6.1), 7.72 (m, 2H), 7.82 (dd,
2H, J ) 2.1, 6.0), 8.2 (br s, 1H), 8.77 (s, 1H). 13C NMR δ
36.703, 50.869, 115.757, 123.028, 123.872, 124.454, 128.897,
129.186, 133.613, 146.148, 163.929, 164.925, 166.227. Calcd
for C15H13F4N3O: C, 55.05; H, 4.00; N, 12.84. Found: C,
55.15; H, 4.20; N, 12.75.
Table 2. Entry 8 (Similar Experiments Were Done for
Entries 7 and 9). To an NMR tube containing 5a (0.14 g,
0.5 mmol) in 0.6 mL of CD2Cl2 cooled in an ice bath was
added BDAM (0.068 mL, 0.5 mmol) followed by TFA
(0.039 mL, 5 mmol). The tube was warmed to room
temperature over 10 min to cleanly give 7a (1H NMR δ 2.7
(s, 6H), 2.92 (dd, 1H, J ) 7.3, 12.6), 3.22 (dd, 1H, J ) 7.3,
12.6), 5.27 (t, 1H, J ) 6.4), 7.12 (t, 2H, J ) 8.5), 7.57 (d,
1H, J ) 8.2), 7.9 (d, 1H, J ) 8.2), 8.10 (dd, 2H, J ) 5.4,
8.8), 8.8 (br s, 1H)). Anhydrous hydrazine (0.047 mL, 1.5
mmol) was added, and after 1.5 h a 2:98 ratio of 12a:13a
(1H NMR δ 3.72 (dd, 1H, J ) 3.8, 9.8), 4.03 (dd, 1H, J )
9.9, 10.8), 4.80 (dd, 1H, J ) 3.7, 10.7), 5.5 (br s, 1H), 6.91
(dd, 2H, J ) 8.8, 8.8), 7.39 (dd, 1H, J ) 0.9, 8.2), 7.59 (dd,
2H, J ) 5.4, 9.0), 7.88 (dd, 1H, J ) 2.2, 8.2), 8.75 (dd, 1H,
J ) 0.9, 2.2)) was observed.
Preparation of 4,5-Dihydro-3-(4-fluorophenyl)-N-(4-
trifluoromethylphenyl)-4-(5-trifluoromethyl-2-pyridyl)-
1H-pyrazole-1-carboxamide (1). A solution of TFA (81 mL,
120 g, 1.05 mol) in methylene chloride (400 mL) was
added to solution of ethanone 5a (283 g, 1.00 mol) and of
BDAM (143 mL, 107 g, 1.05 mol) in methylene chloride (1
L) at -4 °C over 1 h while maintaining the temperature
below 0 °C. After stirring an additional 1 h, anhydrous
hydrazine (95.0 mL, 97.0 g, 3.03 mol) was added over 30
min while maintaining the temperature below 10 °C. The
mixture was allowed to warm to ambient temperature over
16 h and was added to ice/water (1.2 L). The organic layer
was washed with a saturated solution of NH4Cl (2 × 500
mL), dried (MgSO4), and filtered. A solution of 4-(trifluo-
romethyl)phenyl isocyanate (187 g, 1.00 mol) in methylene
chloride (200 mL) was added to the solution of pyrazoline
13a over 30 min while maintaining the temperature below
30 °C. After 1 h, the mixture was diluted with Et2O (1.5 L)
Preparation of (2RS,4RS)- and (2RS,4SR)-1,5-Bis(4-
fluorophenyl)-2,4-bis(5-trifluoromethyl-2-pyridyl)-1,5-
pentanedione (9a). A solution of ethanone 5a (2.83 g, 9.99
mmol) in absolute EtOH (20 mL) was treated sequentially
with Me2NH‚HCl (0.90 g, 11 mmol) and paraformaldehyde
(0.48 g, 15 mmol). The resulting yellow solution was heated
at reflux for 1 h causing a colorless solution. Upon cooling
a white solid formed. The solid was collected to afford 2.33
g (81% yield) of 9a as a 1.3:1 mixture of dl to meso
diasteromers. Recrystallization (acetone/hexanes) gave pure
dl as clear prisms, mp 159-160 °C. 1H NMR δ dl: 2.87 (t,
2H, J ) 7.1), 4.97 (t, 2H, J ) 7.1), 6.99 (dd, 4H, JH-H
)
8.6, JH-F ) 8.6), 7.47 (d, 2H, J ) 8.2), 7.85 (dd, 2H, J )
2.5, 8.3), 7.89 (dd, 4H, JH-H ) 9.0, JH-F ) 5.3), 8.7 (m,
270
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development

and cooled to 10 °C. The resulting solid was collected by
suction filtration to give a total of 389 g (78.4% yield) of
white needles, mp 214-215 °C. MS (m/z) 497 (4), 496 (17),
309 (31), 308 (21), 282 (17), 281 (26), 280 (10), 187 (19),
163 (100), 147 (14). 1H NMR δ 4.24 (dd, 1H, J ) 5.3, 11.6),
4.51 (dd, 1H, J ) 11.7, 11.7), 5.06 (dd, 1H, J ) 5.4, 11.7),
7.02 (dd, 2H, J ) 8.6, 8.6), 7.34 (d, 1H, J ) 8.2), 7.57 (d,
2H, J ) 8.7), 7.6-7.7 (m, 4H), 7.88 (dd, 1H, J ) 2.2, 8.2),
8.2 (br s, 1NH), 8.8 (m, 1H). Calcd for C23H15F7N4O: C,
55.65; H, 3.05; N, 11.29. Found: C, 55.35; H, 3.15; N, 11.34.
Acknowledgment
We thank Virginia W. Miner for conducting the two-
dimensional (2-D) NMR analysis and the Analytical Labora-
tory operated by The University of California, Berkeley for
providing the FAB H+ mass spectrum. We would also like
to thank John M. Owen for the optimization of conditions
used in the large-scale runs of 1.
Received for review December 18, 2002.
OP025619V
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
271