Safe preparation and purification of sodium tetrakis[(3,5-trifluoromethyl) phenyl]borate (NaBArF24): Reliable and sensitive analysis of water in solutions of fluorinated tetraarylborates

Article abstract of DOI:10.1021/om0501428

A safe, convenient preparation of the reagent sodium tetrakis[(3,5- trifluoromethyl)phenyl]borate (NaBArF₂₄) has been devised by utilizing a magnesium-bromine exchange reaction in the absence of metallic magnesium. Purified material was then rigorously dried over P₂O ₅ (NaBArF₂₄ with <500 ppm H₂O by mass) or recrystallized as a hydrate (NaBArF₂₄·(2.6 ± 0.1)H ₂O). Accurate analysis of the water content of these samples by ¹H NMR was accomplished by using dimethylzirconocene (Cp ₂Zr(CH₃)₂).

Full text of DOI:10.1021/om0501428

Organometallics 2005, 24, 3579-3581
3579
Safe Preparation and Purification of Sodium
Tetrakis[(3,5-trifluoromethyl)phenyl]borate (NaBArF ):
2
4
Reliable and Sensitive Analysis of Water in Solutions of
Fluorinated Tetraarylborates
Neal A. Yakelis and Robert G. Bergman*
Division of Chemical Sciences, Lawrence Berkeley National Laboratory, and Department of
Chemistry, University of California, Berkeley, California 94720
Received February 28, 2005
,7,8,11-14
Summary: A safe, convenient preparation of the reagent
sodium tetrakis[(3,5-trifluoromethyl)phenyl]borate (Na-
BArF24) has been devised by utilizing a magnesium-
bromine exchange reaction in the absence of metallic
magnesium. Purified material was then rigorously dried
over P2O5 (NaBArF24 with <500 ppm H2O by mass) or
recrystallized as a hydrate (NaBArF24‚(2.6 ( 0.1)H2O).
Accurate analysis of the water content of these samples
borate, and (3) aqueous cation exchange.3
While
moderate to good yields have been reported for the
synthesis of NaBArF , safety remains a major concern.
The recent report by Leazer et al. confirms that while
3,5-bis(trifluoromethyl)phenyl]magnesium bromide is
not a hazard by itself, explosive exothermic decomposi-
tion of (trifluoromethyl)aryl Grignard reagents occurs
in the presence of any excess magnesium metal. To
24
[
15
1
by H NMR was accomplished by using dimethylzir-
date, every published procedure for the synthesis of
NaBArF24 generates the arylmagnesium halide inter-
conocene (Cp2Zr(CH3)2).
mediate in the presence of magnesium turnings or dust,
Introduction
often in refluxing ethereal solvent.3
,7,8,11-14
The hazards
associated with these methods prompted us to develop
a new, safe protocol for preparation of pure, anhydrous
NaBArF24 on a multigram scale that, in place of
magnesium metal, employs an alkyl Grignard reagent
to effect metal-halogen exchange. Discrepancies in the
reported appearance of the product as well as the varied
procedures for its purification and drying inspired us
to formulate a standardized protocol that would consis-
tently provide material at a high level of purity.
1
Weakly coordinating anions such as the fluorinated
2,3
tetraarylborates have enabled the study of very reac-
tive cationic transition-metal complexes and their ap-
plications in polymerization, C-H bond activation, and
Lewis acid catalysis. Sodium tetrakis[(3,5-trifluoro-
methyl)phenyl]borate, NaBArF24 (1), the most com-
monly used salt and precursor for other reagents, was
first utilized by Kobayashi in 1981. The numerous
synthetic protocols that have been developed since that
initial report all employ three basic steps: (1) formation
of [3,5-bis(trifluoromethyl)phenyl]magnesium bromide,
4
5
6
7-9
1
0
Results and Discussion
(
2) quenching with NaBF4 or BF3‚Et2O to form the
By the reported procedure for magnesium-halogen
exchange,1
5,16
isopropylmagnesium chloride (conve-
*
To whom correspondence should be addressed. E-mail:
17
niently purchased as a solution in THF) was added to
a solution of 1-bromo-3,5-bis(trifluoromethyl)benzene (2)
in THF at -20 °C (Scheme 1). Vacuum-dried sodium
tetrafluoroborate was then added to the newly formed
aryl Grignard reagent. After aqueous workup and
extraction with diethyl ether, the crude NaBArF24
mixed etherate (3) (soluble in CH2Cl2) was dried in
vacuo at 100 °C for approximately 10 h. The crude
product (5) then appeared as a yellowish-tan solid (as
has been noted in several literature preparations and
for the commercially available material). It was critical
that this crude material be washed with CH2Cl2 (ini-
rbergman@berkeley.edu.
(
1) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066.
2) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1964, 2, 245.
(
Fujiki, K.; Ichikawa, J.; Kobayashi, H.; Sonoda, A.; Sonoda, T. J.
Fluorine Chem. 2000, 102, 293.
(3) Fujiki, K.; Kashiwagi, M.; Miyamoto, H.; Sonoda, A.; Ichikawa,
J.; Kobayashi, H.; Sonoda, T. J. Fluorine Chem. 1992, 57, 307.
(4) Recent examples include: (a) Albietz, P. J., Jr.; Cleary, B. P.;
Paw, W.; Eisenberg, R. Inorg. Chem. 2002, 41, 2095. (b) Leatherman,
M. D.; Svejda, S. A.; Johnson, L. K.; Brookhart, M. J. Am. Chem. Soc.
2
003, 125, 3068. (c) Chen, M.-C.; Roberts, J. A. S.; Marks, T. J. J. Am.
Chem. Soc. 2004, 126, 4605. (d) Liu, W.; Brookhart, M. Organometallics
004, 23, 6099.
5) (a) Arndtsen, B. A.; Bergman, R. G. Science 1995, 270, 1970. (b)
Holtcamp, M. W.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
997, 119, 848. (c) Konze, W.; Scott, B. L.; Kubas, G. J. J. Am. Chem.
Soc. 2002, 124, 12550.
2
(
1
(6) (a) Fujiki, K.; Ikeda, S.; Kobayashi, H.; Mori, A.; Nagira, A.; Nie,
(11) Nishida, H.; Takada, N.; Yoshimura, M.; Sonoda, T.; Kobayashi,
H. Bull. Chem. Soc. Jpn. 1984, 57, 2600.
J.; Sonoda, T.; Yagupolskii, Y. Chem. Lett. 2000, 29, 66. (b) Janka, M.;
He, W.; Frontier, A. J.; Eisenberg, R. J. Am. Chem. Soc. 2004, 126,
6
(12) Taube, R.; Wache, S. J. Organomet. Chem. 1992, 428, 431.
(13) Reger, D. L.; Little, C. A.; Lamba, J. J. S.; Brown, K. J. Inorg.
Synth. 2004, 34, 5.
864.
(
7) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992,
1
1, 3920.
(14) Zuccaccia, C.; Bellachioma, G.; Cardaci, G.; Macchioni, A. J.
Am. Chem. Soc. 2001, 123, 11020.
(
(
8) Bahr, S. R.; Boudjouk, P. J. Org. Chem. 1992, 57, 5545.
9) (a) Buschmann, W. E.; Miller, J. S.; Bowman-James, K.; Miller,
(15) Leazer, J. L., Jr.; Cvetovich, R.; Tsay, F.-R.; Dolling, U.; Vickery,
T.; Bachert, D. J. Org. Chem. 2003, 68, 3695.
(16) Abarbri, M.; Dehmel, F.; Knochel, P. Tetrahedron Lett. 1999,
40, 7449.
C. N. Inorg. Synth. 2002, 33, 83 (AgBArF24). (b) Tellers, D. M.; Yung,
C. M.; Arndtsen, B. A.; Adamson, D. R.; Bergman, R. G. J. Am. Chem.
Soc. 2002, 124, 1400 ([n-hex N]BArF24).
4
(10) Kobayashi, H.; Sonoda, A.; Iwamoto, H.; Yoshimura, M. Chem.
(17) Available from Aldrich Chemical, Acros Organics, TCI America,
Alfa Aesar, Strem Chemical, and others.
Lett. 1981, 10, 579.
1
0.1021/om0501428 CCC: $30.25 © 2005 American Chemical Society
Publication on Web 06/01/2005

3580 Organometallics, Vol. 24, No. 14, 2005
Notes
2 3 2
Zr(CH )
1H NMR Analysis of
Scheme 1. Magnesium Metal Free Synthesis of
NaBArF24 (1)
Table 1. Cp
NaBArF24 Hydration in THF-d
8
sample purification conditions
Cp2Zr(CH3)2 anal.
1
2
3
recrystallized from
C6H5F in air
NaBArF24‚(2.6 ( 1)H2O
dried at 100 °C, 0.1 Torr, NaBArF24‚(0.034 ( 0.009)H2O
8 h
dried at 120 °C, 0.1 Torr, NaBArF24‚(0.021 ( 0.005)H2O
8 h, over P2O5
4
1
stoichiometric trap for water enabled accurate analysis
of the samples by 1H NMR spectroscopy (Table 1).
Dimethylzirconocene is known to react rapidly with 0.5
equiv of water to produce [Cp2Zr(CH3)]2O (7).21 The
liberation of methane was complete within 1 h in THF-
d8, and the results were reproducible in multiple trials.
The amount of water in the sample was then calculated
from the increase in integration for the signals of 7
relative to the integrals of the aryl protons of the added
NaBArF24 against an internal standard of hexamethyl-
benzene. Measuring the decrease in integration of the
signals for 6 proved to be a less reliable method, because
a large excess of dimethylzirconocene is employed
relative to the small amount of water present in the
samples.
tially chilled to -30 °C) to remove the colored impurity
and furnish a white powder, as prescribed by
4
7
8
Brookhart and Boudjouk. The identity of the contami-
nant 4 has not been previously addressed in the
literature, but by NMR analysis, it appears to be a
triarylborate species (with a 11B chemical shift distinct
from NaBArF24) arising from incomplete arylation.
While the dichloromethane wash does remove a small
portion of NaBArF24 from the crude mixture, the overall
yield (55-70% over several runs) is still comparable to
that reported for other syntheses, and this procedure
can be easily scaled for the safe production of larger
amounts of pure material.
A series of control experiments was performed to
determine the error inherent in the detection of water
1
using this H NMR protocol. It was found that J. Young
resealable NMR tubes furnished the best results, due
to the sensitivity of Cp Zr(CH ) to both air and mois-
2
3 2
ture. The concentration of the hexamethylbenzene
internal standard was set at a level (approximately 30
µM) so that its integration would be comparable, within
an order of magnitude, to the peaks of interest. At very
low concentrations of water (0.5-1.0 µmol in 0.5 mL of
THF-d8), the method could reliably detect 36 ( 9 ppm
of H2O in THF-d8. Higher concentrations of water could
be detected with lower percent error, 180 ( 10 ppm, due
to more precise integration of the correspondingly larger
peaks of interest.
It has previously been noted that NaBArF24 is a
hygroscopic salt, and the reported elemental analyses
of hydrated samples had determined the presence of
between 0.5 and 3 molar equiv of H2O per mole of
1
0,12,18
NaBArF24.
The hydrate crystallizes from fluo-
14,19
3
robenzene, CH2Cl2,
or CHCl3 as large cubic, color-
Drying this purified material had been
previously accomplished by azeotropic removal of water
14,19
less crystals.
As shown in Table 1, the amount of water calculated
in the recrystallized sample of NaBArF24 was 2.6 ( 0.1
mol equiv. This value was corroborated by elemental
analysis, and it also correlated closely with the molec-
ular formula (NaBArF24‚2.5H2O) given in a footnote of
the original 1981 Kobayashi communication.10 The
material that was dried in the absence of additional
desiccant still showed a residual 0.034 ( 0.009 mol
equiv of water, an amount not reliably detectable using
current elemental analysis methods. However, use of
material dried in this way occasionally gave products
that appeared to result from somewhat higher H2O
levels. The sample dried under vacuum over P2O5 was
found to contain a smaller amount of water: 0.021 (
1
3
7,8,12,19
with benzene or heating under vacuum.
After
drying the product using reasonable conditions de-
scribed in earlier preparations (0.1 Torr, 100 °C, 48 h),
we found that adventitious water was sometimes still
present in the material, causing unsuspected reactions
to occur in certain experiments. Thus, a somewhat more
rigorous, yet operationally simple, drying protocol was
successfully executedsheating the sample to 120 °C at
0
.1 Torr over P2O5 for 18 h.
Given the degree of error inherent in elemental
analysis measurements, a more accurate method was
required to quantify the small amounts of water present
in the NaBArF24 samples. Using a known amount of
0
.005 mol equiv. The quantity of water present in the
2
0
dimethylzirconocene [Cp2Zr(CH3)2] (6) as a kinetic,
(
21) (a) Hunter, W. E.; Hrncir, D. C.; Vann Bynum, R.; Pentilla, R.
A.; Atwood, J. L. Organometallics 1983, 2, 750. (b) Marsella, J. A.;
Huffman, J. C.; Folting, K.; Caulton, K. G. Inorg. Chim. Acta 1985,
(
18) Reger, D. L.; Wright, T. D.; Little, C. A.; Lamba, J. J. S.; Smith,
M. D. Inorg. Chem. 2001, 40, 3810.
19) Hennis, A. D.; Polley, J. D.; Long, G. S.; Sen, A.; Yandulov, D.;
Lipian, J.; Benedikt, G. M.; Rhodes, L. F.; Huffman, J. Organometallics
2 3 2
96, 161. We have utilized Cp Zr(CH ) as an efficient desiccant in
(
previous water-sensitive experiments; see, for example: (c) Dobbs, D.
A.; Bergman, R. G. Organometallics 1994, 13, 4594. (d) Proulx, G.;
Bergman, R. G. Organometallics 1996, 15, 684. (e) Johnson, J. S.;
Bergman, R. G. J. Am. Chem. Soc. 2001, 123, 2923. (f) Holland, A. W.;
Bergman, R. G. J. Am. Chem. Soc. 2002, 124, 9010.
2
001, 20, 2802.
20) Wailes, P. C.; Weigold, H.; Bell, A. P. J. Organomet. Chem.
972, 34, 155.
(
1

Notes
Organometallics, Vol. 24, No. 14, 2005 3581
latter sample is less than 500 ppm by mass, and
material dried in this way gave more reproducible
results in water-sensitive reactions.
°C. After the reaction mixture was warmed from -20 to 0 °C
over 1 h, NaBF
added as a solid under a stream of N
stirred for 48 h at 23 °C (under N ). (This long reaction time
ensures high conversion to the tetraarylborate). The contents
were then poured into a solution of Na CO (44 g) and NaHCO
20 g) in water (600 mL) (all workup procedures are not air-
4
(2.68 g, 0.024 mol, 1.0 equiv) was quickly
2
. The mixture then was
2
In summary, a modified synthesis of NaBArF24 has
been devised that ensures the absence of metallic
magnesium by utilizing the magnesium-bromine ex-
change reaction with commercially available isopropyl-
magnesium chloride in THF. Using the newly developed
procedure, the crude material could then be rigorously
purified and dried over a desiccant, consistently result-
ing in NaBArF24 of identical appearance and composi-
tion. Analysis of the water content at the parts per
million level by use of dimethylzirconocene proved to
2
3
3
(
sensitive). This mixture was stirred vigorously for 1 h and then
extracted with diethyl ether (4 × 200 mL). The combined
organic layers were washed with brine and dried over anhy-
drous Na
2
SO
4
. After filtration of the mixture and rotary
O (3)
evaporation of the filtrate, the crude NaBArF24‚xTHF/Et
2
was dried at 100 °C/0.1 Torr for 10 h to yield a tan-yellow solid.
Dichloromethane, chilled to -30 °C (3 × 20 mL), was used to
wash all colored contaminant and filter it away from the
product, a fine white powder. The product was then heated in
1
be a valuable H NMR based technique for the rapid
quantification of the water present in small NaBArF24
samples. Expanded application of this procedure to
other salts of weakly coordinating anions would be
appropriate. The preparation and use of extremely pure,
dry NaBArF24 by this method may improve the activity
of known catalysts and also may extend the use of
NaBArF24 to more water-sensitive, cationic transition-
metal systems.
2 5
vacuo over P O (0.1 Torr, 115 °C, 20 h) to yield pure,
anhydrous NaBArF24 (1) as a hygroscopic white powder (12.2
g, 57% yield). Mp: 330-335 °C dec. Anal. Calcd for C32
12
H -
1
BF24Na: C, 43.37; H, 1.36. Found: C, 43.37; H, 1.27. H NMR
(400 MHz, acetone-d ): δ 7.79 (br s, 8 H), 7.66 (br s, 4 H). 13C-
6
1
1
{ H} NMR (100 MHz, acetone-d
134.6 (s), 129.1 (q, JC-F ) 31.5 Hz), 124.5 Hz (q, JC-F ) 270
Hz), 117.5 (s). F NMR (376 MHz, acetone-d
B NMR (160 MHz, acetone-d
romethane filtrate was concentrated to a tacky brown-yellow
6
): δ 161.5 (q, JB-C ) 50.0 Hz),
2
1
1
9
6
): δ -62.46 (s).
1
1
6
): δ -7.18 (s). The dichlo-
Experimental Section
solid that contained both NaBArF24 and the contaminant 4:
1
General Procedures. Unless otherwise noted, reactions
and manipulations were performed at 23 °C in an inert-
6
H NMR (400 MHz, acetone-d ) δ 7.98 (br s, 2 H), 7.56 (br s,
1
9
11
1 H); F NMR (376 MHz, acetone-d
(160 MHz, acetone-d ) δ -2.26 (s). A sample of NaBArF24
product was recrystallized from fluorobenzene in air on the
benchtop to yield NaBArF24‚2.6H O as large, cubic colorless
crystals: mp 300-302 °C dec. Anal. Calcd for C32
C, 41.19; H, 1.86. Found: C, 41.39; H, 1.76. A
powdered sample of NaBArF24‚2.6H O was dried in vacuo in
the absence of desiccant (100 °C, 0.1 Torr, 48 h). Anal. Calcd
for C32 12BF24Na (NaBArF24): C, 43.37; H, 1.36; Found: C,
6
) δ -62.12 (s); B NMR
atmosphere (N
vacuum techniques. Glassware was dried for a minimum of
2 h at a temperature of 180 °C or greater. All NMR spectra
2
) glovebox or using standard Schlenk and high-
6
1
2
were obtained using Bruker AV-400 and DRX-500 MHz
spectrometers at 22 °C at the indicated frequency. Chemical
shifts are reported in parts per million (δ), coupling constants
are reported in hertz (Hz), and integrations are reported as
H17.2BF24-
NaO2.6
:
2
1
number of protons. H NMR spectra were recorded at 400
H
MHz, and chemical shifts were referenced to the residual
43.09; H, 1.16.
Typical Procedure for Analysis of H O Content by
2
1
3
protons in the solvent. C NMR spectra are proton decoupled
and were recorded at 100 MHz; chemical shifts were referenced
Cp
2 3 2
Zr(CH ) NMR Titration. Dimethylzirconocene (6; 20.7
1
9
to the solvent. F NMR spectra were recorded at 376 MHz,
and chemical shifts are reported relative to external CFCl
mg, 82.3 µmol) and hexamethylbenzene (1.8 mg, 11 µmol) were
1
.
dissolved in THF-d
NMR single-pulse spectrum was recorded: H NMR (400 MHz,
THF-d ) δ 6.13 (s, 10 H, Cp of 6), 2.21 (s, internal std), -0.38
(s, 6H, Zr-CH of 6). Next, the hydrated NaBArF24 sample
(6-20 mg) was added to the tube under an inert atmosphere,
resulting in the immediate evolution of CH (g). After 1 h, a
second H NMR single-pulse spectrum was obtained: H NMR
(400 MHz, THF-d ) δ 7.81 (br s, BArF), 7.58 (br s, BArF), 6.13
(s, Cp of 6), 5.97 (s, Cp of 7), 2.22 (s, internal std.), 0.21 (s,
CH ), -0.02 (s, Zr-CH of 7), -0.38 (s, 6H, Zr-CH of 6). No
8
(0.50 mL) in a J. Young NMR tube. A H
3
11
1
B NMR spectra were recorded at 160 MHz; chemical shifts
were referenced to an external standard of BF ‚Et O. Melting
point values were obtained in capillary tubes sealed under N
3
2
8
2
3
and are uncorrected. Elemental analyses were performed at
the University of California, Berkeley, Microanalytical Facility
on a Perkin-Elmer 2400 Series II CHNO/S analyzer.
Materials. Unless otherwise noted, reagents were pur-
chased from commercial suppliers and used without further
purification. Tetrahydrofuran was passed through a column
4
1
1
8
4
3
3
of activated alumina under N
use. Acetone-d and THF-d (Cambridge Isotope Laboratories)
were dried using standard procedures (CaSO (s) for acetone-
; sodium/benzophenone ketyl for THF-d ) and vacuum-
transferred prior to use. Dimethylzirconocene (stored at -35
2
and sparged with N
2
prior to
further change in integration was evident after longer periods
of time. The resulting increase in integration for the signals
of 7 could then be compared to the peaks attributed to BArF24,
integrated against the internal standard, to determine the
amount of water present in the original NaBArF24 sample.
6
8
4
d
6
8
2
0
°
C) was prepared according to a standard procedure, but it
is also commercially available from Aldrich Chemical, Strem
Chemical, and Acros Organics. Sodium tetrafluoroborate was
dried under 0.1 Torr at 120 °C for 18 h.
Acknowledgment. This work was supported by the
Director, Office of Energy Research, Office of Basic
Energy Sciences, Chemical Sciences Division, U.S.
Department of Energy, under Contract No. DE-AC03-
Sodium tetrakis[(3,5-trifluoromethyl)phenyl]borate
(
0
NaBArF24) (1). A 2.0 M solution of i-PrMgCl in THF (80 mL,
.16 mol, 6.6 equiv) was added by addition funnel over 45 min
to a stirred solution of 1-bromo-3,5-bis(trifluoromethyl)benzene
2; 25 mL, 0.14 mol, 5.8 equiv) in THF (120 mL) chilled to -20
7
6SF00098 and by the NIH postdoctoral fellowship GM-
66582 (N.A.Y.).
0
(
OM0501428