(Z)- or (E)-selective hydrogenation of potassium (3,3,3-trifluoroprop-1-yn-1-yl)trifluoroborate: Route to either isomer of β-trifluoromethylstyrenes

Article abstract of DOI:10.1021/acs.orglett.5b00235

A Pd-catalyzed hydrogenation of potassium (3,3,3-trifluoroprop-1-yn-1-yl)trifluoroborate providing either the (Z)- or (E)-isomer of the vinylborate in >98% purity is described. The initially formed (Z)-isomer of the alkene is transformed to the (E)-isomer with time, irrespective of the catalyst used; coupling with bromo- and iodoarenes provides a variety of (Z)- or (E)-β-trifluoromethylstyrenes. Also, a safe synthesis of the alkynyltrifluoroborate from HFC-245fa and BF₃·OEt₂ has been described.

Full text of DOI:10.1021/acs.orglett.5b00235

Letter
pubs.acs.org/OrgLett
(Z)- or (E)‑Selective Hydrogenation of Potassium (3,3,3-Trifluoroprop-
1-yn-1-yl)trifluoroborate: Route to Either Isomer of
β‑Trifluoromethylstyrenes
P. Veeraraghavan Ramachandran* and Wataru Mitsuhashi
Herbert C. Brown Center for Borane Research, Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette,
Indiana 47907, United States
S
* Supporting Information
ABSTRACT: A Pd-catalyzed hydrogenation of potassium
(3,3,3-trifluoroprop-1-yn-1-yl)trifluoroborate providing either
the (Z)- or (E)-isomer of the vinylborate in >98% purity is
described. The initially formed (Z)-isomer of the alkene is
transformed to the (E)-isomer with time, irrespective of the
catalyst used; coupling with bromo- and iodoarenes provides a
variety of (Z)- or (E)-β-trifluoromethylstyrenes. Also, a safe synthesis of the alkynyltrifluoroborate from HFC-245fa and BF3·
OEt2 has been described.
he unique effects of the trifluoromethyl moiety in the
T
pharmacology of bioactive molecules attract medicinal
chemists,1 while the moiety’s high thermal and oxidative
stabilities appeal to materials chemists.2 Direct insertion of a
CF3 group is preferred over the cumbersome regiospecific
fluorination for their synthesis.3 The reaction of CF3-olefins has
also been employed as a surrogate to introduce a CF3 group.
Accordingly, the stereospecific synthesis of CF3-olefins has
attained importance. Recent literature has witnessed an increase
in nucleophilic, electrophilic, and radical trifluoromethylations,3
but a corresponding vinylogous reaction to prepare CF3-olefins
has trailed.
Olefination of aldehydes via a Wittig-type reaction of CF3-
ylides results in either pure (E)-olefins or mixtures.4 Photo-
redox-catalyzed hydrotrifluoromethylation of alkynes,5 decar-
boxylative trifluoromethylation of cinnamic acids,6 Heck-type
coupling of in situ generated trifluoropropene,7 and Suzuki
coupling of (E)-CF3-vinyldiphenylsulfonium salt8 with arylbor-
onic acids also provide (E)-olefins. Transition-metal-catalyzed
coupling processes of (E)-CF3-vinylmetal partners retained
their stereochemistry and resulted in (E)-olefins. For example,
Hiyama coupling of (E)-CF3-vinylsilanes (eq 1)9 and photo- or
Fe(II)-catalyzed coupling of (E)-alkenylBF3K with Togni’s
reagent (eq 2)10 generated (E)-alkenes either selectively or
predominantly. All of the aforementioned methodologies
provided only the (E)-isomer.
As part of our program on fluoro-organic synthesis via
boranes,11 we required both stereoisomers of CF3-styrenes. The
Suzuki coupling of (Z)-CF3-vinylboron seemed an obvious
choice to prepare the (Z)-olefins. However, the steric/
electronic effects of the CF3 group on the coupling were a
concern because isomerization of (Z)-alkenyl-Pd with bulky
groups such as t-butyl has been reported.12 We report herein
the preparation of either isomer of pure β-CF3-styrenes
involving a facile synthesis of CF3CCBF3K (2) from
CF3CH2CHF2 (4), a stereoselective hydrogenation, and Suzuki
coupling (eq 3).
The preparation of the vinyl trifluoroborate ((Z)-1) was the
initial step in our project. In the hydrocarbon series,
hydroboration of 1-haloalkynes, followed by reductive rear-
rangement, is a routine process to prepare the corresponding
(Z)-vinylboron.13 However, the instability of CF3CCBr14
and the potential for Markovnikov hydroboration of CF3-
alkynes11,15 dissuaded us from adopting this pathway.
Received: January 23, 2015
Published: February 23, 2015
© 2015 American Chemical Society
1252
DOI: 10.1021/acs.orglett.5b00235
Org. Lett. 2015, 17, 1252−1255

Organic Letters
Letter
Fortunately, the preparation of 2 has been reported,16 and we
envisaged a stereoselective hydrogenation to achieve our goal.
Frohn’s synthesis of 2 involved the alkynylation of B(OMe)3,
followed by treatment with aqueous KHF2 in the critical
presence of HF.17 Due to safety concerns,18 we opted to
explore other protocols for the preparation of 2. Utilization of
BF3·Et2O as a source of boron to prepare RFBF3K has been
reported in the literature. For example, Chambers et al.
reported the formation of CF3BF3K by trapping the CF3 anion
Table 2. Optimization of Semihydrogenation
c
product ratio
time
(h)
solvent
(M)
a
b
entry
Pd
Q
2
(Z)-1 (E)-1
5
1
2
1.5
1.5
DME (0.3) L (5)
10
5
0
2
3
80
73
18
24
19
from Me3SnCF3 with BF3 and converting the ate complex
MeOH
(0.3)
L (1)
0
formed to the desired salt using KF. Recently, Stefani et al.
reported the formation of KRBF3 from lithiated 2-substituted
1,3-dithianes using B(OMe)3 and KHF2.20 When necessary,
they used BF3·OEt2 to improve yields. Adopting this for a
possible safe synthesis of 2 from 3,3,3-trifluoropropyne (3), we
performed a protocol standardization by varying the solvents,
concentration, and equiv of KHF2 (Table 1; entry 5 is the
optimal condition).
3
4
1.5
1.0
1.0
1.5
1.0
3.0
1.0
3.0
3.9
2.5
DME (1.0) L (0.5)
DME (1.0) L (0.5)
DME (0.3) B (5)
DME (0.3) B (5)
DME (0.3) R (5)
DME (0.3) R (5)
5
5
0
9
44
79
57
0
42
9
14
3
5
10
10
10
10
10
10
10
10
34
0
5
4
6
56
14
60
8
44
6
7
1
79
8
0
6
d
34
3
9
THF (0.5)
THF (0.5)
R (5)
R (5)
0
89
10
11
12
0
6
63
31
19
3
e
DME (0.3) L (5)
THF (0.5) R (5)
0
1
80
Table 1. Optimization of the Formation of 2
f
0
92
5
a
Pd catalyst (mol %): L = Pd/CaCO3/Pb, B = Pd/BaCO3, R = Pd/
b
c
d
BaSO4. Quinoline (mol %). Determined by 1H and 19F NMR. Yield
43% (Z:E:alkane = 96:3:1) by recrystallization from EtOAc/CHCl3.
e
f
Yield 56%. Yield 59% (Z:E:alkane = 98:1.5:0.5) by recrystallization
a
b
entry
solvent
concn (M)
equiv of KHF2
yield (%)
from iPrOH.
1
2
3
4
5
THF
Et2O
THF
THF
Et2O
1.0
1.0
0.5
0.5
0.5
4.0
4.0
4.0
1.2
4.0
67
79
83
49
89
Et3N to yield pure (E)-7a (Table 3). To the best of our
knowledge, this may be the first example of the formation of
essentially pure (E)-alkene via Lindlar catalyst-mediated
hydrogenation.27−29
a
b
Concentration of 3. Isolated yield.
Returning to our original goal, we probed the necessary
conditions for the isolation of the initially formed (Z)-1. The
influence of solvent was examined, and the reaction was carried
out in MeOH when we noticed a similar ratio (Table 2, entry
2). The catalyst loading was then examined when we obtained
∼1:1 Z/E mixture (entry 3). When hydrogenation was
terminated within 1 h, the ratio was improved in favor of
(Z)-1 (entry 4).
The cost of 3 became a concern for the scale up of 2. The
preparation of CF3CCLi from 2-bromo-3,3,3-trifluoropro-
pene21 and HFC-245fa (CF3CH2CHF2, 4)16,22 has been
reported. The latter commodity chemical was converted to 2
by adopting our optimized protocol in a moderate 51% yield
for a 0.7 mol scale reaction (Scheme 1).
We also noticed that replacing hydrogen with a nitrogen
atmosphere after the initial reaction (1.5 h) and continued
stirring for an additional 30 min did not change the product
ratio any further (Table 2, entry 3). It appears that the
hydrogen atmosphere plays an important role in the isomer-
ization from (Z)- to (E)-1. We believe that the isomerization
should proceed via a palladated intermediate that can proceed
to yield (E)-1 via β-hydride elimination or 5 via a reductive
elimination. The fact that >18% of 5 is produced always with
(E)-1 supports this assumption.
Due to the difficulty in separating the product, we examined
other poisoned Pd catalysts, such as Pd/BaCO3 or Pd/BaSO4
(Rosenmund catalyst). Again, both of these catalysts revealed
atypical stereoselection in that the initial (Z)-product trans-
formed (entries 5 and 7, respectively) to the (E)-product with
time (entries 6 and 8, respectively), along with the formation of
5. Arresting the isomerization was targeted and the amounts of
catalyst, additives, solvent, concentration, temperature, and
time (see Supporting Information for details) were fine-tuned
to achieve the optimal condition for the (Z)-isomer (entry 9).
Prolonging the reaction provides the (E)-isomer predominantly
(entry 10). The selectivity improved slightly during the gram-
scale preparation of both (E)- (>98%, entry 11) and (Z)-
isomers (93%, entry 12). Crystallization from iPrOH yielded
Scheme 1. Synthesis of 2 from HFC-245fa
Attention was now turned to a selective hydrogenation to
prepare (Z)-1. Brown et al. standardized the hydrogenation of
alkynyl-B(OiPr)2 in dioxane to yield >95% (Z)-isomer using
Lindlar catalyst in the presence of quinoline.23 Soderquist et al.
reported a hydroboration−protodeboronation approach for
similar (Z)-alkenylboron species.24 Due to stability concerns for
2 in the presence of acids, Brown’s protocol was adopted.
Monitored by 19F NMR spectroscopy, 2 was semihydrogenated
in the presence of Lindlar catalyst with added quinoline in 1,2-
dimethoxyethane (DME) (Table 2, entry 1).25 To our surprise,
we observed that the peak at δ −61 ppm, corresponding to the
initial product, shifted with time to δ −66 ppm, along with the
formation of an additional triplet at δ −70 ppm, corresponding
to the over-reduced product 5. The (Z)-isomer gradually
converted almost entirely (2:80) to the thermodynamically
stable (E)-isomer (along with 18% of 5), as confirmed by
coupling26 with 1-bromonaphthalene (6a) in the presence of
1253
DOI: 10.1021/acs.orglett.5b00235
Org. Lett. 2015, 17, 1252−1255

Organic Letters
Letter
The generality of this procedure was demonstrated by
preparing a series of (Z)-β-CF3-styrenes (Table 3).
Table 3. Suzuki Coupling of (E)-1 and (Z)-1 with Aryl
Halides
An electron-withdrawing group (EWG) at the p-position
(entry 2) or sterically encumbered o-position (entry 3) did not
affect the yield or selectivity. p-Bromoanisole (6d) was
converted to the desired styrene ((Z)-7d) along with a
byproduct that was difficult to separate (entry 4). This was
avoided by replacing Br with I,26 although some isomerization
occurred during the coupling (entry 5). Unlike entry 5, iodide
with a p-acetyl EWG did not affect the E/Z ratio (entry 6). A
heteroaromatic, p-tosyl-5-bromoindole (6e) gave good yield
and isomeric selectivity (entry 7). We also examined the CF3-
alkene synthesis to prepare an enyne by coupling with an aryl
bromide bearing an internal alkyne (6f), which resulted in
moderate yield and good selectivity (entry 8). A slight excess of
(Z)-1 and longer reaction time were necessary to achieve good
yield of the product.
The generality for the (E)-styrenes was then probed,
standardizing the preparation of (E)-7a described earlier
(vide supra). The stoichiometry of (E)-1 and the amine were
adjusted to account for the presence of 5 in the reaction
medium, and the coupling was conducted with a selected series
of aryl halides (6a, 6c, 6d′). Although the diastereomeric ratio
remained excellent, the yield (42−66%) was slightly lower
compared to the (Z)-isomers. Increasing the equivalent of (E)-
1 (condition B) helped improve the yield and achieved the
synthesis of (E)-CF3-styrenes in yields matching that of the
(Z)-isomer.
In conclusion, we have accomplished the preparation of (Z)-
and (E)-β-CF3-styrenes in excellent yields.30 The correspond-
ing Suzuki coupling partners, (Z)- or (E)-CF3-vinylBF3K, were
prepared via a stereoselective semihydrogenation using
poisoned Pd catalysts. The precursor, CF3CCBF3K, was
successfully prepared in quantity avoiding the use of HF. We
are further examining the details of the stereoselectivity control
during the Pd-catalyzed hydrogenation, which will be disclosed
in due course.
ASSOCIATED CONTENT
* Supporting Information
Experimental details, characterization, and 1H, 11B, 13C, and 19
■
S
F
NMR spectra of compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: chandran@purdue.edu.
a
b
c
Notes
Isolated yield. Determined by 1H and 19F NMR. Yield ∼50%
d
(mixture with unidentified p-methoxyphenyl byproduct). Calculated
for the mixture of the (Z)- and (E)-product (76% combined). With
The authors declare no competing financial interest.
e
1.5 equiv of (Z)-1; the reaction time was 6 h.
ACKNOWLEDGMENTS
We thank the Herbert C. Brown Center for Borane Research
for financial assistance with this project.
■
>98% isomeric purity for (Z)-1. Thus, we achieved a
stereoselective synthesis of (Z)- and (E)-1 via a time-dependent
hydrogenation.
The Suzuki coupling of the (Z)-isomer was examined first
under conditions similar to those of the (E)-isomer (vide
supra). Unfortunately, the product was isolated in only 17:3
diastereoselectivity. Replacing the base with iPrNH2 suppressed
the isomerization, and (Z)-7a was isolated in 99% isomeric
purity. The standardized condition (A) yielded 85% of (Z)-7a.
REFERENCES
■
(1) Muller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
̈
(2) Shimizu, M.; Takeda, Y.; Higashi, M.; Hiyama, T. Chem. Asian J.
2011, 6, 2536.
(3) Ma, J.-A.; Cahard, D. J. Fluorine Chem. 2007, 128, 975.
(4) (a) Kobayashi, T.; Eda, T.; Tamura, O.; Ishibashi, H. J. Org.
Chem. 2002, 67, 3156. (b) Hanamoto, T.; Morita, N.; Shindo, K. Eur.
J. Org. Chem. 2003, 4279.
1254
DOI: 10.1021/acs.orglett.5b00235
Org. Lett. 2015, 17, 1252−1255

Organic Letters
Letter
(5) Iqbal, N.; Jung, J.; Park, S.; Cho, E. J. Angew. Chem., Int. Ed. 2014,
53, 539.
1). THF was added to 2 (10.0 g, 50.0 mmol), Pd/BaSO4 (5.32 g,
2.5 mmol, 5 wt %), and quinoline (0.59 mL, 5.0 mmol). The reaction
mixture was stirred under H2 at rt and monitored by 19F NMR. The
reaction was completed after being stirred for 2.5 h; the reaction
mixture was filtered through Celite, which was washed with EtOAc.
The product was precipitated with minimal amount of CH2Cl2, after
removing the solvents, and recrystallized from iPrOH (250 mL) to
obtain (Z)-1 (6.07 g, 59%, Z/E/alkane = 98:1.5:0.5). Preparation of
potassium [(E)-3,3,3-trifluoroprop-1-en-1-yl]trifuoroborate ((E)-
1). A similar reaction as above for 4 h in DME using Pd/CaCO3/
Pb (5.32 g, 2.5 mmol, 5 wt %) as the catalyst to 2 (10.0 g, 50.0 mmol)
yielded (E)-1 (7.06 g, 56%, Z/E/alkane = 1:80:19). General
procedure for Suzuki coupling (Conditions A/B). To a solution
of (Z)-1/(E)-1 (A: 208.0 mg, 1.03 mmol/B: 343.3 mg, 1.70 mmol),
PdCl2(dppf)·CH2Cl2 (16.3 mg, 0.02 mmol), aryl halide (1.00 mmol)
in degassed (freeze−pump−thaw) iPrOH/H2O (2/1 = v/v, 10 mL)
was added iPrNH2 (A: 0.27 mL, 3.09 mmol/B: 0.44 mL, 5.1 mmol)
under N2. The reaction mixture was stirred at reflux for 2.5 h, diluted
with H2O (15 mL), and extracted with CH2Cl2 (3 × 10 mL). The
combined organic layers were washed with brine (15 mL), filtered
through Na2SO4, concentrated, and purified by column chromatog-
raphy to obtain (Z)/(E)-β-trifluoromethylstyrenes.
(6) (a) He, Z.; Luo, T.; Hu, M.; Cao, Y.; Hu, J. Angew. Chem., Int. Ed.
2012, 51, 3944. (b) He, Z.; Hu, M.; Luo, T.; Li, L.; Hu, J. Angew.
Chem., Int. Ed. 2012, 51, 11545. (c) Xu, P.; Abdukader, A.; Hu, K.;
Cheng, Y.; Zhu, C. Chem. Commun. 2014, 50, 2308.
(7) Prakash, G. K. S.; Krishnan, H. S.; Jog, P. V.; Iyer, A. P.; Olah, G.
A. Org. Lett. 2012, 14, 1146.
(8) Lin, H.; Dong, X.; Li, Y.; Shen, Q.; Lu, L. Eur. J. Org. Chem. 2012,
4675.
(9) Omote, M.; Tanaka, M.; Ikeda, A.; Nomura, S.; Tarui, A.; Sato,
K.; Ando, A. Org. Lett. 2012, 14, 2286.
(10) Parsons, A. T.; Senecal, T. D.; Buchwald, S. L. Angew. Chem., Int.
Ed. 2012, 51, 2947.
(11) Brown, H. C.; Chen, G.-M.; Jennings, M. P.; Ramachandran, P.
V. Angew. Chem., Int. Ed. 1999, 38, 2052.
(12) Miyaura, N.; Satoh, M.; Suzuki, A. Tetrahedron Lett. 1986, 27,
3745.
(13) Negishi, E.; Williams, R. M.; Lew, G.; Yoshida, T. J. Organomet.
Chem. 1975, 92, C4.
(14) Trofimenko, S.; Johnson, R. W.; Doty, J. K. J. Org. Chem. 1978,
43, 43.
(15) Konno, T.; Chae, J.; Tanaka, T.; Ishihara, T.; Yamanaka, H.
Chem. Commun. 2004, 690.
(16) Bardin, V. V.; Adonin, N. Y.; Frohn, H.-J. Organometallics 2005,
24, 5311.
(17) HF is extremely crucial for the synthesis, without which none of
the salt is formed. See: (a) Bardin, V. V.; Frohn, H.-J. Main Group Met.
Chem. 2002, 25, 589. (b) Molander, G. A.; Hoag, B. P. Organometallics
2003, 22, 3313.
(18) HF, MSDS: Acute toxicity, Inhalation (Category 2), Acute
toxicity, Dermal (Category 1); Skin corrosion (Category 1A).
(19) Chambers, R. D.; Clark, H. C.; Willis, C. J. J. Am. Chem. Soc.
1960, 82, 5298.
(20) Vieira, A. S.; Fiorante, P. F.; Zukerman-Schepector, J.; Alves, D.;
Botteselle, G. V.; Stefani, H. A. Tetrahedron 2008, 64, 7234.
(21) Katritzky, R.; Qi, M.; Wells, A. P. J. Fluorine Chem. 1996, 80,
145.
(22) Brisdon, A. K.; Crossley, I. R. Chem. Commun. 2002, 2420.
(23) Srebnik, M.; Bhat, N. G.; Brown, H. C. Tetrahedron Lett. 1988,
29, 2635.
(24) Soderquist, J. A.; Rane, A. M.; Matos, K.; Ramos, J. Tetrahedron
Lett. 1995, 36, 6847. See also: Molander, G. A.; Ellis, N. M. J. Org.
Chem. 2008, 73, 6841.
(25) DME was chosen due to the favorable boiling point.
(26) Molander, G. A.; Bernardi, C. R. J. Org. Chem. 2002, 67, 8424.
(27) (a) Burch, R. R.; Muetterties, E. L.; Teller, R. G.; Williams, J. M.
J. Am. Chem. Soc. 1982, 104, 4257. (b) Radkowski, K.; Sundararaju, B.;
Furstner, A. Angew. Chem., Int. Ed. 2013, 52, 355. (c) Chen, Z.; Luo,
̈
M.; Wen, Y.; Luo, G.; Liu, L. Org. Lett. 2014, 16, 3020.
(28) Partial isomerization is known. However, the preparation of (E)-
alkene in high isomeric purity using Lindlar catalyst is not common.
See: Drost, R. M.; Bouwens, T.; van Leest, N. P.; de Bruin, B.; Elsevier,
C. J. ACS Catal. 2014, 4, 1349.
(29) For a report on Pd-catalyzed isomerization of some olefins
bearing strong EWGs, see: Canovese, L.; Santo, C.; Visentin, F.
Organometallics 2008, 27, 3577.
(30) Preparation of potassium (3,3,3-trifluroprop-1-yn-1-yl)
trifluoroborate (2). nBuLi (800 mL, 2.5 M in hexane) was added
to a solution of 4 (67.2 mL, 0.702 mol) in Et2O (1.4 L) in a 5 L RB
flask at −35 °C. After being stirred for 1 h, BF3·OEt2 (86.6 mL, 0.702
mol) was added at −78 °C, and the mixture was stirred for 10 min,
followed by stirring at 0 °C for an additional 0.5 h. Aqueous KHF2
(219 g, 2.81 mol, in 600 mL of H2O) was added to the mixture and
stirred for 20 min at 0 °C, 1 h at rt, and opened to air. The residue,
after removal of solvents, was extracted with acetone, filtered,
concentrated, and precipitated with CH2Cl2. The solid was filtered
and washed with CH2Cl2 to yield 68.3 g (51%) of 2. Preparation of
potassium [(Z)-3,3,3-trifluoroprop-1-en-1-yl]trifuoroborate ((Z)-
1255
DOI: 10.1021/acs.orglett.5b00235
Org. Lett. 2015, 17, 1252−1255