Amine- and dimeric amino-borane complexes of the {Rh(PiPr 3)2}+ fragment and their relevance to the transition-metal-mediated dehydrocoupling of amine-boranes

Article abstract of DOI:10.1021/ic9020542

Complexes formed between {Rh(PⁱPr₃)₂} ⁺ or {Rh(H)₂(PⁱPr₃) ₂}⁺ fragments and the amine- and dimeric aminoborane σ ligands H₃B·NMe₃ and [H ₂BNMe₂]₂ have been prepared and their solution and solid-state structures determined: [Rh(fⁱPr₃) ₂(η²-H₃B·NMe₃)][BAr ^F₂ (1), [Rh(PⁱPr₃) ₂{η²-(H₂BNMe₂) ₂}][BAr^F₄] (2), [Rh(H)₂(P ⁱPr₃)₂(η²H₃B · NMe₃)][BAr^F₄] (3), and [Rh(H) ₂(PⁱPr₃)₂{η²-(H ₂BNMe₂)₂}][BAr^F₄] (4) [Ar^F = C₆H₃(CF₃)₂] The last compound was only observed in the solid state, as in solution it dissociates to give [Rh(H)₂(PⁱPr₃) ₂][BAr^F ₄] and [H₂BNMe ₂]₂ due to steric pressure between the ligand and the metal fragment. The structures and reactivities of these new complexes are compared with the previously reported tri-isobutyl congeners. On the basis of ¹¹B and ¹H NMR spectroscopy In solution and the Rh ... B distances measured in the solid state, the PⁱPr ₃ complexes show tighter interactions with the σ ligands compared to the PⁱBu₃ complexes forthe Rh(I) species and a greater stability toward H₂ loss for the Rh(III) salts. For the Rh(I) species (1 and 2), this Is suggested to be due to electronic factors associated with the bending of the ML₂ fragment. For the Rh(III) complexes (3 and 4), the underlying reasons for Increased stability toward H₂ loss are not as clear, but steric factors are suggested to influence the relative stability toward a loss of dihydrogen, although other factors, such as supporting agostic interactions, might also play a part. These tighter interactions and a slower H2 loss are reflected in a catalyst that turns over more slowly in the dehydrocoupling of H₃B · NHMe ₂ to give the dimeric amino-borane [H₂BNMe ₂]₂, when compared with the PⁱBu ₃-ligated catalyst (ToF 4 h^-1, cf., 15 h^-1, respectively). The addition of excess MeCN to 1,2, or 3 results in the displacement of the σ-ligand and the formation of the adduct species trans-[Rh(PⁱPr₃)₂(NCMe)₂][BAr ^F ₄] (with 1 and 2) and the previously reported [Rh(H)₂(PⁱPr₃)₂(NCMe) ₂][BAr^F₄(with 3).

Full text of DOI:10.1021/ic9020542

Inorg. Chem. 2010, 49, 1111–1121 1111
DOI: 10.1021/ic9020542
Amine- and Dimeric Amino-Borane Complexes of the {Rh(PiPr3)2}þ Fragment
and Their Relevance to the Transition-Metal-Mediated Dehydrocoupling of
Amine-Boranes
Adrian B. Chaplin and Andrew S. Weller*
Department of Inorganic Chemistry, University of Oxford, Oxford, OX1 3QR, United Kingdom
Received October 16, 2009
Complexes formed between {Rh(PiPr3)2}þ or {Rh(H)2(PiPr3)2}þ fragments and the amine- and dimeric amino-
borane σ ligands H3B NMe3 and [H2BNMe2]2 have been prepared and their solution and solid-state structures
3
determined: [Rh(PiPr3)2(η2-H3B NMe3)][BArF ] (1), [Rh(PiPr3)2{η2-(H2BNMe2)2}][BArF ] (2), [Rh(H)2(PiPr3)2(η2-
4
4
3
H3B NMe3)][BArF ] (3), and [Rh(H)2(PiPr3)2{η2-(H2BNMe2)2}][BArF ] (4) [ArF = C6H3(CF3)2]. The last compound was
4
4
3
only observed in the solid state, as in solution it dissociates to give [Rh(H)2(PiPr3)2][BArF ] and [H2BNMe2]2 due to steric
pressure between the ligand and the metal fragment. The structures and reactivities of these new complexes are compared
4
with the previously reported tri-isobutyl congeners. On the basis of 11B and 1H NMR spectroscopy in solution and the
Rh B distances measured in the solid state, the PiPr3 complexes show tighter interactions with the σ ligands compared
3 3 3
to the PiBu3 complexes for the Rh(I) species and a greater stability toward H2 loss for the Rh(III) salts. For the Rh(I) species
(1 and 2), this is suggested to be due to electronic factors associated with the bending of the ML2 fragment. For the Rh(III)
complexes (3 and 4), the underlying reasons for increased stability toward H2 loss are not as clear, but steric factors are
suggested to influence the relative stability toward a loss of dihydrogen, although other factors, such as supporting agostic
interactions, might also play a part. These tighter interactions and a slower H2 loss are reflected in a catalyst that turns over
more slowly in the dehydrocoupling of H3B NHMe2 to give the dimeric amino-borane [H2BNMe2]2, when compared with
3
the PiBu3-ligated catalyst (ToF 4 h-1, c.f., 15 h-1, respectively). The addition of excess MeCN to 1, 2, or 3 results in the
displacement of the σ-ligand and the formation of the adduct species trans-[Rh(PiPr3)2(NCMe)2][BArF ] (with 1and 2) and
4
the previously reported [Rh(H)2(PiPr3)2(NCMe)2][BArF ] (with 3).
4
1. Introduction
boranes,6-20 and impressive rates of hydrogen release13,17
and control over polymer formation11 have been achieved.
The transition-metal-mediated dehydrocoupling of phos-
phine- and amine-boranes is of considerable current interest
due to the potential to control (a) the kinetics of hydrogen re-
lease and (b) the product distributions for the resulting group
13 and 15 materials (e.g., linear oligomeric, cyclic, or poly-
meric materials).1 The high gravimetric hydrogen content of
(7) Sloan, M. E.; Clark, T. J.; Manners, I. Inorg. Chem. 2009, 48, 2429–2435.
(8) Rousseau, R.; Schenter, G. K.; Fulton, J. L.; Linehan, J. C.;
Engelhard, M. H.; Autrey, T. J. Am. Chem. Soc. 2009, 131, 10516–10524.
(9) Jiang, Y.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H. Organome-
tallics 2009, 28, 5493-5504.
(10) Dallanegra, R.; Chaplin, A. B.; Weller, A. S. Angew. Chem., Int. Ed.
2009, 48, 6875–6878.
(11) Staubitz, A.; Soto, A. P.; Manners, I. Angew. Chem., Int. Ed. 2008,
47, 6212–6215.
(12) Douglas, T. M.; Chaplin, A. B.; Weller, A. S. J. Am. Chem. Soc. 2008,
130, 14432–14433.
(13) Blaquiere, N.; Diallo-Garcia, S.; Gorelsky, S. I.; Black, D. A.;
Fagnou, K. J. Am. Chem. Soc. 2008, 130, 14034–14034.
(14) Pun, D.; Lobkovsky, E.; Chirik, P. J. Chem. Commun. 2007, 3297–3299.
(15) Jiang, Y.; Berke, H. Chem. Commun. 2007, 3571–3573.
(16) Fulton, J. L.; Linehan, J. C.; Autrey, T.; Balasubramanian, M.;
Chen, Y.; Szymczak, N. K. J. Am. Chem. Soc. 2007, 129, 11936–11949.
(17) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey, D. M.; Goldberg,
K. I. J. Am. Chem. Soc. 2006, 128, 12048–12049.
(18) Clark, T. J.; Russell, C. A.; Manners, I. J. Am. Chem. Soc. 2006, 128,
9582–9583.
(19) Kawano, Y.; Uruichi, M.; Shimoi, M.; Taki, S.; Kawaguchi, T.;
Kakizawa, T.; Ogino, H. J. Am. Chem. Soc. 2009, 131, 14946-14957.
(20) Keaton, R. J.; Blacquiere, J. M.; Baker, R. T. J. Am. Chem. Soc.
2007, 129, 1844–1845.
H3B NH3 has made it an attractive target as a hydrogen-
3
transport vector for future energy requirements,2-4 and new
developments in the regeneration of spent materials offer
encouraging strategies for the reuse of these materials.5 Cat-
alysts from across the transition series of the periodic table
have proved effective for the dehydrocoupling of amine-
*To whom correspondence should be addressed. E-mail: andrew.weller@
chem.ox.ac.uk.
(1) Clark, T. J.; Lee, K.; Manners, I. Chem.—Eur. J. 2006, 12, 8634–8648.
(2) Marder, T. B. Angew. Chem., Int. Ed. 2007, 46, 8116–8118.
(3) Hamilton, C. W.; Baker, R. T.; Staubitz, A.; Manners, I. Chem. Soc.
Rev. 2009, 38, 279–293.
(4) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613–2626.
(5) Davis, B. L.; Dixon, D. A.; Garner, E. B.; Gordon, J. C.; Matus,
M. H.; Scott, B.; Stephens, F. H. Angew. Chem., Int. Ed. 2009, 48, 6812–6816.
€
(6) Zahmakiran, M.; Ozkar, S. Inorg. Chem. 2009, 48, 8955–8964.
r
2010 American Chemical Society
Published on Web 01/06/2010
pubs.acs.org/IC

1112 Inorganic Chemistry, Vol. 49, No. 3, 2010
Chaplin and Weller
Chart 1.a
aR = H (a), Me (b). [BArF - anions are not shown.
]
4
Mechanistic investigations using computational19,21-26 and
empirical studies27-29 indicate a complex process that first
Scheme 1. Dehydrocoupling of Amine-Boranes H3B NMeRH
3
involves dehydrogenation of H3B NR2H at a metal center to
3
form an amino-borane H2BdNR2. This process may be
inner-sphere (i.e., involving BH/NH transfer pathways at the
metal), that is, single-site, colloidal,27,30,31 or small cluster
mediated.8,16 It can also be outer-sphere, which is related to
alcohol oxidation, in which a ligand is also involved in
hydrogen transfer.13,19,32 Subsequent oligomerization, po-
lymerization, or dimerization of H2BdNMe2 can then occur,
either on or off the metal (all with the possible further
mechansim.22,23,35,36 Hydrolytic processes catalyzed by tran-
sition metal complexes have also been reported.37-40
In an effort to understand the mechanistic scenario for
amine-borane dehydrogenation in more detail, we have
recently reported intermediate and model species that result
from cationic {Rh(PiBu3)2}þ fragments partnered with H3B
involvement of H3B NR2H). Although the details of this
3
3
NRMe2 (R = H, Me), H2BdNMe2, H3B NMe2BH2 NH-
and further steps toward the formation of oligomeric and
polymeric materials are considerably less-well resolved for
metal-based processes,24,27-29,33,34 off-metal dehydrocou-
pling has received more attention with regard to the overall
3
3
Me2, and [H2BNMe2]2 ligands.12,24 Examples of these “un-
stretched” σ complexes41 of Rh(I) and Rh(III) are shown in
Chart 1 (I-V). Bis-amine-borane motifs (VI)10 and pro-
ducts of B-H activation (VII)42 have also recently been
described by us. These complexes build upon the pioneering
work by Shimoi and co-workers on the synthesis of ami-
ne-borane and phosphine-borane σ complexes with a
variety of transition metal fragments.43-46 With regard to
(21) Yang, X.; Hall, M. B. J. Organomet. Chem. 2009, 694, 2831–2838.
(22) Zimmerman, P. M.; Paul, A.; Zhang, Z. Y.; Musgrave, C. B. Angew.
Chem., Int. Ed. 2009, 48, 2201–2205.
(23) Zimmerman, P. M.; Paul, A.; Zhang, Z. Y.; Musgrave, C. B. Inorg.
Chem. 2009, 48, 1069–1081.
(24) Douglas, T. M.; Chaplin, A. B.; Weller, A. S.; Yang, X. Z.; Hall,
M. B. J. Am. Chem. Soc. 2009, 131, 15440–15456.
(25) Paul, A.; Musgrave, C. B. Angew. Chem., Int. Ed. 2007, 46, 8153–
8156.
(26) Luo, Y.; Ohno, K. Organometallics 2007, 26, 3597–3600.
(27) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. J. Am. Chem.
Soc. 2003, 125, 9424–9434.
(28) Pons, V.; Baker, R. T.; Szymczak, N. K.; Heldebrant, D. J.; Linehan,
J. C.; Matus, M. H.; Grant, D. J.; Dixon, D. A. Chem. Commun. 2008, 6597–
6599.
the dehydrocoupling of H3B NHMe2 to cyclic [H2BNMe2]2
3
(Scheme 1) catalyzed by [Rh(PiBu3)2][BArF4] [ArF = C6H3-
(CF3)2], complexes Ia, IIa, IV, and V all potentially sit on the
catalytic cycle. Calculations24 indicate a complex multipath-
way process in which the likely first steps are sequential
BH/NH or NH/BH activation at the Rh(I) center, follow-
ing initial formation of a simple amine-borane σ complex
(29) Friedrich, A.; Drees, M.; Schneider, S. Chem.—Eur. J. 2009, 15,
10339–10342.
(30) Jaska, C. A.; Clark, T. J.; Clendenning, S. B.; Grozea, D.; Turak, A.;
Lu, Z.-H.; Manners, I. J. Am. Chem. Soc. 2005, 127, 5116–5124.
(31) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776–9785.
(32) Kass, M.; Friedrich, A.; Drees, M.; Schneider, S. Angew. Chem., Int.
Ed. 2009, 48, 905–907.
(33) Dietrich, B. L.; Goldberg, K. I.; Heinekey, D. M.; Autrey, T.;
Linehan, J. C. Inorg. Chem. 2008, 47, 8583–8585.
(34) Hebden, T. J.; Denney, M. C.; Pons, V.; Piccoli, P. M. B.; Koetzle, T.
F.; Schultz, A. J.; Kaminsky, W.; Goldberg, K. I.; Heinekey, D. M. J. Am.
Chem. Soc. 2008, 130, 10812–10820.
(38) Yoon, C. W.; Sneddon, L. G. J. Am. Chem. Soc. 2006, 128, 13992–
13993.
(39) Clark, T. J.; Whittell, G. R.; Manners, I. Inorg. Chem. 2007, 46, 7522–
7527.
(40) Li, Y.; Xie, L.; Li, Y.; Zheng, J.; Li, X. Chem.—Eur. J. 2010, 15,
8951–8954.
(41) Kubas, G. J. Metal Dihydrogen and o-Bond Complexes; Kluwer
Academic/Plenum Publishers: New York, 2001.
(42) Chaplin, A. B.; Weller, A. S. Angew. Chem., Int. Ed. 2009, DOI:
10.1002/anie.200905185.
(43) Kawano, Y.; Yamaguchi, K.; Miyake, S. Y.; Kakizawa, T.; Shimoi,
M. Chem.—Eur. J. 2007, 13, 6920–6931.
(44) Kawano, Y.; Hashiva, M.; Shimoi, M. Organometallics 2006, 25,
4420–4426.
(35) Shaw, W. J.; Linehan, J. C.; Szymczak, N. K.; Heldebrant, D. J.;
Yonker, C.; Camaioni, D. M.; Baker, R. T.; Autrey, T. Angew. Chem., Int.
Ed. 2008, 47, 7493–7496.
(36) Staubitz, A.; Besora, M.; Harvey, J. N.; Manners, I. Inorg. Chem.
2008, 47, 5910–5918.
(37) Yan, J. M.; Zhang, X. B.; Han, S.; Shioyama, H.; Xu, Q. Angew.
Chem., Int. Ed. 2008, 47, 2287–2289.
(45) Kakizawa, T.; Kawano, Y.; Shimoi, M. Organometallics 2001, 20,
3211–3213.
(46) Shimoi, M.; Nagai, S.; Ichikawa, M.; Kawano, Y.; Katoh, K.;
Uruichi, M.; Ogino, H. J. Am. Chem. Soc. 1999, 121, 11704–11712.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 1113
Scheme 2. Synthesis of 1 and 2a
a [BArF - anions not shown.
]
4
(e.g., Ia). Consistent with their observation during and at the
end of catalysis, calculations indicate that IIa and IV sit in
relatively deep energy wells.
Given that these well-defined complexes of amine-boranes
could be prepared using the {Rh(PiBu3)2}þ fragment, we were
interested in varying the steric and electronic profile of the
phosphine to probe the consequences (if any) on observed
ground state structures and the rate of catalysis for the
dehydrocoupling of H3B NHMe2. For our first attempt at
3
this, we turned to model complexes based upon the
{Rh(PiPr3)2}þ and {Rh(H)2(PiPr3)2}þ fragments partnered
with the amine-borane H3B NMe3 or, the ultimate product
3
of H3B NMe2H dehydrogenation, the cyclic amino-borane
3
[H2BNMe2]2. Solid-state and solution studies allow a compar-
ison between these complexes and those of PiBu3. Despite the
steric and electronic similarity between PiBu3 and PiPr3,47
there is a subtle, but distinct, change in the strength of
interaction between these ligands and the metal centers in
the Rh(I) complexes, as measured by NMR spectroscopy and
X-ray crystallography, as well as the ease of reductive elimina-
tion of H2 from the Rh(III) dihydride complexes. We also
demonstrate an attenuation in the rate of dehydrocoupling of
Figure 1. Cationic portion of 1. Thermal ellipsoids shown at the 50%
probability level. Phosphine H atoms omitted for clarity. Selected bond
lengths (A) and angles (deg): Rh1-B1, 2.1376(3); B1-H1A, 1.22(4); B1-
H1B, 1.22(4); B1-H1C, 1.13(6); B1-N1, 1.594(7); Rh1-P1, 2.2478(12);
Rh1-P2, 2.2679(12); P1-Rh1-P2, 106.08(5); Rh1-B1-N1, 131.5(4);
H1A-Rh1-H1B, 68(3).
˚
H3B NHMe2 to form cyclic [H2BNMe2]2 when using the iPr-
3
substituted catalyst.
Studies of single-site transition-metal catalyzed-amine-
borane dehydrogenation where the ligand set or metal has
been varied have been reported previously; these have gen-
erally focused upon the relative rates of catalysis rather
than the isolation, per se, of likely intermediates or models
thereof.7,9,15,19,20 Systematic studies on the coordination
properties of H2RB L (R = Cl, H, Me, Ph; L = PMe3,
3
NMe3, PPh3, for example) have also been reported.43
Some of these complexes have been briefly described in a
recent communication.42
2. Results
2.1. Synthesis of Rh(I) Complexes of H3B NMe3 and
[H2BNMe2]2. The substrate H3B NMe3 does not contain
3
3
a NH group, and thus dehydrocoupling at a transi-
tion metal center cannot proceed. This enables the isola-
tion of relatively stable and long-lived complexes.
Reaction of H3B NMe3 with [Rh(PiPr3)2(η6-C6H5F)]-
3
48
[BArF
]
in 1,2-F2C6H4 solution led to the immediate
4
formation of the new complex [Rh(PiPr3)2(η2-H3B
Figure 2. Cationic portion of one of the crystallographically indepen-
dent cations in the unit cell for 2. Thermal ellipsoids shown at the 50%
probability level. Phosphine H atoms omitted for clarity. Selected bond
3
NMe3)][BArF4] (1) and elimination of C6H5F. Simi-
27
larly, reaction of the cyclic dimer [H2BNMe2]2 with
˚
lengths (A) and angles (deg): Rh1-B1, 2.124(4); B1-H1A, 1.23(2); B1-
[Rh(PiPr3)2(η6-C6H5F)][BArF4] afforded [Rh(PiPr3)2-
{η2-(H2BNMe2)2}][BArF4] (2; Scheme 2). Both 1 and 2
are isolated as purple, air-sensitive, crystalline solids in
H1B, 1.24(2); Rh1-P1, 2.2681(8); Rh1-P2, 2.2883(9); P1-Rh1-P2,
105.61(3). Second independent molecule: Rh2-B11, 2.118(4); P3-Rh2-
P4, 107.14(3).
good yields. They are direct analogs of the PiBu3 ligated
complexes Ib and III (Chart 1). Their solid-state struc-
tures are shown in Figures 1 and 2, respectively. For 2,
there are two independent ion pairs in the unit cell,
(47) Tolman, C. A. Chem. Rev. 1977, 77, 313–348.
(48) Chaplin, A. B.; Poblador-Bahamonde, A. I.; Sparkes, H. A.;
Howard, J. A. K.; Macgregor, S. A.; Weller, A. S. Chem. Commun. 2009,
244–246.

1114 Inorganic Chemistry, Vol. 49, No. 3, 2010
Chaplin and Weller
Table 1. Comparison of Selected Structural and Spectroscopic Dataa for the PiPr3 (1-5) and PiBu3 Complexes (Ia, IIa/b, III)24
δ(11B)
Δδ(11B)b
J(HB)/Hzc
δ(1H) Rh-H-B, Rh-Hd
˚
d(Rh-B)/A
P-Rh-P/deg
1
Ib
2
2.137(6)
2.180(4)
2.124(4)
2.118(4)
2.161(6)
2.140(7)e
2.325(4)
f
106.08(5)
97.35(4)
105.61(3)
107.14(3)e
98.31(6)
94.42(5)e
157.05(3)
f
28.8
23.1
35.3
þ36.1
þ30.4
þ29.8
80
d
89
-3.09, n/a (-6.91)
-2.12, n/a (-5.55)
-6.50, n/a
III
31.1
þ25.6
89
-5.07, n/a
3
IIb
4
5
IIa
5.0
5.1
h
3.2
2.2
þ12.3
þ12.4
h
80
g
h
g
g
-0.57, -19.75
-0.47, -18.81
h
-1.0, -18.15 (-3.56)
-0.77, -17.42 (-3.15)
2.386(15)
f
2.318(8)
151.11(15)
f
163.65(7)
þ16.0
þ15.0
a Data for 1-5 were collected in a CD2Cl2 solution, and for I-III in 1,2-F2C6H4, except for low-temperature data (CD2Cl2). Comparison of 11B and
1H chemical shifts for the new complexes in both solvents show only small chemical shift differences (∼0.4 ppm, 11B; ∼0.1 ppm, 1H). b Difference in
chemical shift from free ligand: δ(11B) H3B NMe3, -7.3 [J(HB) 98 Hz]; H3B NHMe2, -12.8 [J(HB) 96 Hz]; [H2BNMe2]2, 5.5 [J(HB) 112 Hz], as
3
3
measured in 1,2-C6H4F2.24 c Measured from the 11B NMR spectrum at 298 K. d 298 K, Rh-H-B in parentheses if low-temperature limit reached, see
text. e Data for second crystallographically independent molecule in the unit cell. f Structural data not available. g 11B coupling not resolved;broad
peaks. h Data not available, complex dissociates in solution, see text.
although the structural metrics are not grossly different
between the two.
Both the cations in salts 1 and 2 have pseudo-square-
planar Rh(I) centers with cis-phosphines and an η2-cor-
dinated amine-borane or dimeric amino-borane, each
bonding through two three-center, two-electron bonds.
˚
The Rh B distances [2.137(6) and 2.124(4)/2.118(4) A,
respectively] are similar to, but shorter than, those repor-
3 3 3
ted for the PiBu3 analogs [2.180(4) and 2.161(6)/2.140(7)
A for Ib and III, respectively]. The P1-Rh1-P2 angle is
˚
similar for 1 and 2 [106.08(5)° and 105.61(3)/107.14(3)°,
respectively], and both are larger than found for Ib
[97.35(4)°] and III [98.31(6)/94.42(5)°]. The hydrogen
atoms associated with the borane/metal were located in
all cases, and although within error there is no lengthen-
ing of the B-H bonds on coordination, absolute values
show a trend for a longer B-H bond on interaction with
the metal—as expected.41 The shorter Rh B distances
3 3 3
Figure 3. Selected 1H NMR data for 2 for the Rh-H-B resonance
suggest a tighter interaction for 1 and 2 compared with
that for Ib and III. Table 1 summarizes these structural
centered at δ -6.50 and a simulated (gNMR)49 spectrum at the bottom.
metrics. Interestingly, a short Rh B distance is accom-
3 3 3
89; J(RhH), 31; J(PH), 30 Hz. Experimentally, decoupling
panied by a wider P-Rh-P bond angle.
31P resolves this multiplet into a quartet of doublets (11B and
In solution, 1 and 2 retain the Rh-H-B interactions. The
103Rh coupling), while decoupling 11B resolves it into an
1H NMR spectrum for 1 shows, in addition to the ex-
apparent 1:2:1 triplet (coupling to 1 ꢀ 31P and 103Rh). Pre-
pected signals due to anion and iPr groups, a broad signal
1
sumably, it is the trans H-31P coupling that is observed,
at δ -3.09 (relative integral 3-H) that shows coupling to 11B
[J(BH) 80 Hz], which collapses into a sharper signal in the
with the cis coupling small and unresolved. The 11B NMR
spectrum of 1 shows a signal at δ 28.8 [J(BH) 80 Hz] shifted
36.1 ppm downfield from the free ligand (Table 1) and also
shows a reduced coupling constant compared to the free
ligand [98 Hz]. That of 2 shows two signals, δ 35.3 [t, J(BH)
89 Hz] and 5.3 [t, J(BH) 112 Hz], the former also showing a
1H{11B} NMR spectrum. The observation of a 3-H relative
integral signal indicates rapid site exchange between bound
and terminal B-H, resulting in a frequency-averaged che-
mical shift.46 This is frozen out at 190 K, so that an integral
2-H signal is observed at δ -6.91 for the Rh-H-B hydro-
reduced coupling constant, allowing it to be assigned to the
gens at this temperature. The terminal BH group is observed
boron atom involved in the three-center, two-electron bond
with the metal. This signal at δ 35.3 is also shifted 29.8 ppm
as a broad, integral 1-H signal at δ 4.43. For 2, site exchange
does not occur at room temperature, and a 2-H relative
downfield compared to [H2BNMe2]2, while that at δ 5.3 is
essentially unshifted. These chemical shifts and the reduction
in coupling coupling constant are as expected on coordina-
integral resonance is observed at δ -6.50 that is assigned to
the Rh-H-B hydrogens, while those due to uncoordinated
BH2 are observed at δ2.70. The high-field hydride signal in 2
tion of the amine-borane to the metal center.19,41,46,50,51 In
comparison with Ib and III, 1and 2show larger upfield shifts
is beautifully resolved in the 1H NMR spectrum into a multi-
plet that shows coupling to 11B, 31P, and 103Rh (Figure 3)—
which can be simulated49 as an AA0MNXX0 system: J(BH),
(50) Alcaraz, G.; Sabo-Etienne, S. Coord. Chem. Rev. 2008, 252, 2395–
2409.
(49) Budzelaar, P. gNMR, 4.0; Cherwell Scientific Publishing, Oxford,
1997.
(51) Merle, N.; Koicok-Kohn, G.; Mahon, M. F.; Frost, C. G.; Ruggerio,
G. D.; Weller, A. S.; Willis, M. C. Dalton Trans. 2004, 3883–3892.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 1115
it to be isolated and the solid-state structure to be deter-
mined.
The solid-state structure of 3 is shown in Figure 5 and
reveals a pseudo-octahedral Rh(III) center, coordinated
with trans-phosphines, cis-hydrides, and a σ-H3B NMe3
ligand. All hydrogen atoms associated with the metal
3
center were located. The structure is very similar to that
]
reported for [Rh(H)2(PiBu3)2(η2-H3B NHMe2)][BArF
(IIa). Although the difference in the amine-borane
ligand discourages detailed structural comparisons, the
4
3
Rh B distance is the same within error between the two
3 3 3
˚
[2.325(4) versus 2.318(8) A], while the P-Rh-P angle is
slightly more acute in 3 [157.05(3) versus 163.65(7)°].
These markers suggest a similar strength of the Rh
B
3 3 3
interaction. The solution NMR data for 3 are virtually
identical to those of IIb (Table 1), showing an upfield-
shifted resonance in the 11B NMR spectrum [δ 5.0, shifted
12.3 ppm from the free ligand] on coordination of the
borane and a relative integral 3-H signal in the 1H NMR
spectrum at δ -0.57, demonstrating site exchange for the
H3B hydrogen atoms. This site exchange cannot be frozen
out at 190 K, similar to IIb. The hydrides are observed as a
sharp relative integral 2-H doublet of triplets at δ -19.75,
showing coupling to 103Rh and 31P, which collapse to a
doublet in the 1H{31P} NMR spectrum. The observation
of only one 31P environment in the 31P{1H} NMR spec-
trum [δ 64.5; J(RhP), 111 Hz] indicates that the BH site
exchange is accompanied by a spin of the amine-borane
around the Rh-B vector. These NMR and structural
data all point to a weaker Rh-H-B interaction in 3
when compared with the Rh(I) complexes 1 and 2. This
is no doubt due to the increased steric pressure between
the borane and the trans phosphine ligands as well as
the high-trans-influence hydride ligands. We have
noted this previously in complexes I and II.24 The
addition of MeCN (excess) to 3 results in the immediate
formation of the previously reported adduct species
Figure 4. Cationic portion of one of the crystallographically indepen-
dent cations in the unit cell for trans-[Rh(PiPr3)2(NCMe)2][BArF4].
Thermal ellipsoids shown at the 50% probability level. Phosphine H
atoms omitted for clarity. Selected bond lengths (A) and angles (deg):
Rh1-N1, 1.966(3); Rh1-N2, 1.964(4); Rh1-P1, 2.3353(11); Rh1-P2,
2.3347(10); P1-Rh1-P2, 179.18(4); N1-Rh1-N2, 179.74(14).
˚
for the Rh-H-B hydrogens and greater downfield shifts in
the 11B NMR spectrum on coordination to the metal
fragment. Both are indicative of a stronger interaction in
the PiPr3 Rh(I) complexes compared to the PiBu3 analogs
and also correspond with shorter Rh B distances in the
3 3 3
solid state.
The addition of MeCN to a CD2Cl2 solution of 1 or 2
results in the displacement of the amine-borane/dimeric
amino-borane ligand and the formation of the bis-MeCN
adduct [Rh(PiPr3)2(NCMe)2][BArF ] as the cis isomer,
4
which converts to the trans over 30 min. The observation
of CHMe2 protons in the 1H NMR spectrum as an apparent
quartet, that collapses to a doublet in the 1H{31P} spectrum,
is consistent with this assignment. The solid-state structure
is given in Figure 4 and is similar to that reported for
trans-[Rh(PCy3)2(NCMe)2][BF4].52 The Ir analog has also
been reported (as the [PF6]- salt).53
[Rh(H)2(PiPr3)2(NCMe)2][BArF ]54 via displacement of
4
H3B NMe3 (observed).
3
We have previously reported24 that complexes of [H2B-
NMe2]2 could not be observed for the PiBu3-ligated com-
plexes of Rh(III)—presumably due to steric pressure bet-
ween the bulky trans-orientated phosphines and this
dimeric amino-borane. Thus, the addition of H2 to III
resulted in the formation of VIII alongside free [H2B-
NMe2]2, while in the absence of H2, the addition of
[H2BNMe2]2 to VIII resulted in the reductive-elimination
of H2 and the isolation of III. Similar reactivity is ob-
served for the PiPr3 complexes reported here, Scheme 4:
the addition of H2 (1 atm) to 2 results in the formation of
IX alongside free [H2BNMe2]2, and the reaction of
[H2BNMe2]2 with IX resulted in the rapid reductive
elimination of H2 and the formation of 2. The presence
2.2. Synthesis of Rh(III)-Dihydride Complexes of H3B
3
NMe3 and [H2BNMe2]2. We have previously reported24 that
the addition of H2 to Ib results in the formation of [Rh(H)2-
(PiBu3)2(η2-H3B NMe3)][BArF ] (IIb). This complex could
not be isolated, as it readily lost H2 to reform Ib, even by
simply placing it under an Ar atmosphere. The addition of
4
3
H3B NMe3 to [Rh(H)2(PiBu3)2][BArF ], VIII, under a H2
4
3
atmosphere also afforded IIb. In a similar manner, the addi-
tion of H2 to purple 1resulted in the immediate formation of
a colorless material that was initially characterized spectro-
scopically as [Rh(H)2(PiPr3)2(η2-H3B NMe3)][BArF ] (3),
4
3
by comparison with IIb, while combining [Rh(H)2(PiPr3)2-
(L)2][BArF ] (IX, L=agostic interaction or solvent)54 also
4
afforded 3 (Scheme 3). Interestingly, even though 3 has the
bulkier PiPr3 phosphine, it does not lose H2, even when
placed under a vacuum for an extended period of time (72 h,
5 ꢀ 10-3 Torr), instead requiring a hydrogen acceptor, tert-
butylethene, to regenerate 1. This lack of H2 lability allowed
1
of liberated H2 from the latter reaction, observed by H
NMR spectroscopy, results in equilibrium between IX
and 2 being established—the yield of 2 increasing with
added [H2BNMe2]2. For example, the reaction of IX with
2 equiv of [H2BNMe2]2 affords 2 in 80% yield, while with
50 equiv the yield is essentially quantitative. Although this
means that the Rh(III)-dihydride complex ligated with
(52) Chisholm, M. H.; Huffman, J. C.; Iyer, S. S. J. Chem. Soc., Dalton
Trans. 2000, 9, 1483–1489.
(53) Dorta, R.; Goikhman, R.; Milstein, D. Organometallics 2003, 22,
2806–2809.
(54) Ingleson, M. J.; Brayshaw, S. K.; Mahon, M. F.; Ruggiero, G. D.;
Weller, A. S. Inorg. Chem. 2005, 44, 3162–3171.
[H2BNMe2]2, [Rh(H)2(PiPr3)2{η2-(H2BNMe2)2}][BArF ]
4
(4), could not be observed in solution, slow crystallization
under a H2 atmosphere of a mixture of IX and [H2BNMe2]2

1116 Inorganic Chemistry, Vol. 49, No. 3, 2010
Scheme 3. Synthesis of 3a
Chaplin and Weller
a [BArF - anions are not shown.
]
4
a
Scheme 4. Reactivity of 2 and IX with H2/[H2BNMe2]2
a [BArF - anions not shown.
]
4
Figure 5. Cationic portion of 3. Thermal ellipsoids shown at the 50%
probability level. Phosphine H atoms and minor disordered component
omitted for clarity. Selected bond lengths (A) and angles (deg): Rh1-B1,
2.325(4); Rh1-H0A, 1.50(3); Rh1-H0B, 1.50(3); B1-H1A, 1.25(3); B1-
H1B, 1.25(3); B1-H1C, 1.16(4); Rh1-P1, 2.3259(9); Rh1-P2, 2.3194(9);
B1-N1, 1.604(5); P1-Rh1-P2, 157.05(3); Rh1-B1-N1, 137.2(2); H1A-
Rh1-H1B, 63(2).
Figure 6. Cationic portion of 4. Thermal ellipsoids at the 30% prob-
ability level. Phosphine H atoms omitted for clarity. Symmetry equivalent
atoms (/) generated by the operation: y þ 1, x - 1, -z. Selected bond
˚
˚
lengths (A) and angles (deg): Rh1-B1, 2.386(15); Rh1-P1, 2.343(2); Rh1-
H0, 1.44(6); B1-H1, 1.18(8); P1-Rh1-P1, 151.10(14); H1*-Rh1-H1,
59(5).
gave colorless crystals of 4.55 This suggests under a H2
atmosphere an equilibrium exists between 4 and IX/[H2B-
NMe2]2, albeit rather one-sided; presumably a lower relative
solubility of 4 helps drive the equilibrium to give crystalline
material. Redissolving crystalline 4 in CD2Cl2 at a low
temperature (225 K) afforded a 1:1 mixture of IX and
[H2BNMe2]2 (by NMR spectroscopy). The solid-state
structure of 4 is shown in Figure 6 and displays an
overall motif similar to that observed for 3—with trans-
phosphine ligands and cis-hydrides. The Rh B dis-
3 3 3
˚
tance, 2.386(15) A, is the longest of all the complexes
discussed here, while the P-Rh-P angle, 151.10(14)°,
is the smallest for the Rh(III) complexes. These metrics
point to a relatively crowded molecule—consistent
with the fact that it can only be observed in the solid
state. The observation that H2 reductive elimination
only occurs rapidly when [H2BNMe2]2 is added to the
Rh(III)-dihydrides VIII or IX indicates that this
occurs from a σ-borane complex, i.e., 4.
(55) Complex 4 is unstable in the solid state, slowly losing H2 to form
purple 2. Similarly, the addition of H2 to 2 in the solid state results in the
formation of 4 by a gas/solid reaction (Figure S-1 in the Supporting
Information). The addition of H2 to metal complexes in the solid state has
been reported previously: (a) Olivan, M.; Marchenko, A. V.; Coalter, J. N.;
Caulton, K. G. J. Am. Chem. Soc. 1997, 119, 8389–8390. (b) Kubas, G. J.;
Unkefer, C. J.; Swanson, B. I.; Fukushima, E. J. Am. Chem. Soc. 1986, 108,
7000–7009. (c) Matthes, J.; Pery, T.; Grundemann, S.; Buntkowsky, G.;
Sabo-Etienne, S.; Chaudret, B.; Limbach, H. H. J. Am. Chem. Soc. 2004,
126, 8366–8367. (d) Douglas, T. M.; Weller, A. S. N. J. Chem. 2008, 32, 966–
969.
2.3. Catalysis: Dehydrocoupling of H3B NHMe2 to
3
[H2BNMe2]2. [Rh(PiPr3)2(η6-C6H5F)][BArF ] is a com-
petent precatalyst for the dehydrocoupling of H3B NHMe2
4
3
[0.072 M initial concentration], to ultimately give [H2B-
NMe2]2. Monitoring this reaction in an open system under

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 1117
Figure 7. Plot of 11B concentration for the dehydrocoupling of H3B NHMe2 (initial concentration = 0.072 M) using [Rh(PR3)2(η6-C6H5F)][BArF4]:
3
5 mol %, 1,2-F2C6H4, 298 K. R = iPr, blue; R = iBu, red. b = H3B NHMe2, 9 = [H2BNMe2]2, ( = H3B NMe2BH2 NHMe2.
3
3
3
a slow flow of argon (i.e., not in a sealed tube) by periodic
sampling and 11B NMR spectroscopy revealed that, at a
loading of 5 mol %, catalysis was complete in 300 min, ToF
4 h-1. This is considerably slower than found for
σ ligand (H3B NMe3, [H2BNMe2]2) are varied. This allows
for the resulting structures (solution and solid state), reac-
tivity (loss of H2), and efficacy in the dehydrocoupling of
3
H3B NHMe2 to be compared. From this, three general
3
[Rh(PiBu3)2(η6-C6H5F)][BArF ] under the same conditions
observations can be drawn: (i) that the σ ligands in the Rh(I)
4
(80 min for 100% completion, ToF 15 h-1),56 although both
are much slower than the current best-reported systems
for amine borane dehydrocoupling.6,14,17 Inspection of a
time-11B concentration plot (Figure 7) also demonstrates
the growth and disappearance of an intermediate boron-
containing species, identified as the linear diborazine
H3B NMe2BH2 NHMe2.27,57,58 This has recently been ob-
complexes appear to be more tightly bound when L = PiPr3
over L = PiBu3 and all Rh B distances are shorter than
3 3 3
found in the Rh(III)-dihydride complexes; (ii) when L =
PiPr3, the Rh(III) dihydride amine-borane σ complexes are
relatively more stable towards H2 loss than when L = PiBu3;
(iii) the L = PiPr3 ligated complexes are considerably slower
in mediating the dehydrocoupling of H3B NHMe2 than their
3
3
3
served by us in the PiBu3 catalyst system,24 shown by
Manners and co-workers to be a plausible intermediate in
the dehydrocoupling of H3B NHMe2 as catalyzed by col-
PiBu3 analogs. We discuss these observations in turn.
For the Rh(I) complexes, the larger cone angle of PiPr3
(160°) versus PiBu3 (143°)47 is consistent with a larger P-
Rh-P angle in 1 and 2 when compared with Ib and III
(Table 1) but is, at first glance, at odds with the tighter
Rh-H-B interaction for the PiPr3 complexes in both solu-
tion and the solid state. This suggests that steric pressure
between the phosphine and amine-borane has little to do
with modulating this interaction in these particular Rh(I)
species. Considering electronic factors, a qualitative Walsh
diagramfor a linear ML2 fragment that undergoes bending to
a C2v geometry is shown in Scheme 5. As it is established that
3
loidal rhodium,27 and most recently observed by Schneider
and co-workers in the dehydrocoupling of H3B NHMe2 by
3
bifunctional Ru catalysts.29 Under the conditions of a sealed
NMR tube, after ca. 30% (10 h) and 70% (48 h) conver-
sion during catalysis, there is only one organometallic
species observed by 31P{1H} NMR spectroscopy (∼98%
detection limit), the chemical shift of which is very close to 3
[δ 67.4, J(RhP) 109 Hz], suggesting its formulation as
[Rh(H)2(PiPr3)2(η2-H3B NHMe2)][BArF ] (5) and that this
4
3
1
is a resting state during the catalytic cycle. Other H and
H3B L complexes are σ donors only, the B-H σ* orbitals
3
11B NMR data are consistent with this assignment, as is ESI-
being too high in energy,46 the relative energies of the metal-
based LUMO (1b1) and LUMOþ1 (2a1) would be expected
to be important in determining the strength of the M-H-B
interaction on combination with the filled e set46 (B-H bon-
MS (obsd, 484.2832; calcd, 484.2880 for [Rh(H)2(PiPr3)2(η2-
H3B NHMe2)]þ, Figure S-4, Supporting Information).
3
This differs for the PiBu3 system in which the analogous
complex, IIa, is observed but at only ca. 20% composition in
an otherwise complex mixture during catalysis.24
ding) of the C3v H3B NMe3 fragment. As Scheme 5 shows,
3
the 1b1 orbital drops in energy with increasing L-M-L
angle, and in this qualitative approach, a wider P-Rh-P
angle would thus lead to a stronger Rh-H-B interaction.
Thisappears to bethe casefor 1 and 2 whencomparedwith Ib
and III (Table 1). A wider P-Rh-P angle in the PiPr3
complexes that leads to a shorter Rh-B distance would also
induce a larger change in chemical shift on coordination of
the borane to the metal fragment in both 1H and 11B NMR
spectra—as is observed.
3. Discussion
In this paper and the previous one,24 a number of amine-
and dimeric amino-borane σ complexes of Rh(I), {RhL2}þ,
and Rh(III), {Rh(H)2L2}þ, fragments have been prepared
where the phosphine ligand (L = PiBu3 versus PiPr3) and
(56) Under the same loading, but approximately 3 times the concentra-
tion (0.2 M), this catalyst gives complete conversion in 35 min, ref 24.
(57) Hahn, G. A.; Schaeffer, R. J. Am. Chem. Soc. 1964, 86, 1503–1504.
(58) Noth, H.; Thomas, S. Eur. J. Inorg. Chem. 1999, 1373–1379.
The observations regarding the relative stabilities for the
Rh(III)-dihydrides 3, 4, and IIb also are at first inspection
counterintuitive: the bulkier trans-orientated PiPr3 ligands

1118 Inorganic Chemistry, Vol. 49, No. 3, 2010
Chaplin and Weller
Scheme 5. Adapted59,60 Qualitative Walsh Diagram Showing the
to be fully determined, although stabilization of the PiPr3
ligated complexes (as observed for 1 and 3) that lead to an
increased barrier to BH/NH activation or H2 loss would be
expected to slow the rate of catalysis if these steps were either
rate-determining or pooled species outside the catalytic cycle.
An analysis of relative contributions from these and other
factors is complicated significantly by the fact that the
Frontier Metal Orbitals for the Bending of a ML2 d8 Fragment and
the HOMO-1 Orbitals of a C3v H3B NMe3 Ligand46a
3
dehydrogenation pathway of H3B NHMe2, to initially give
3
H2BdNMe2, has been calculated to proceed via a number of
energetically similar pathways,24 and the onward reactivity of
this unsaturated fragment to ultimately give dimeric
[H2BNMe2]2 can also proceed via a number of pathways.29
Nevertheless, the empirical observation that changing the
phosphine alters the rate of dehydrocoupling (albeit slower)
is an encouraging one for future rational catalyst design.
aEnergies are approximate, and the specific ordering is for illustration
only.
4. Experimental Section
All manipulations, unless otherwise stated, were per-
formed under an atmosphere of argon, using standard
Schlenk and glovebox techniques. Glassware was oven-dried
at 130 °C overnight and flamed under a vacuum prior to use.
CH2Cl2, MeCN, and pentane weredried using a Grubbs-type
solvent purification system (MBraun SPS-800) and degassed
by successive freeze-pump-thaw cycles.66 CD2Cl2 and 1,2-
C6H4F2 were distilled under a vacuum from CaH2 and stored
lead to a more stable Rh(III)-dihydride, that is, 3 versus IIb,
and 4 can be isolated (albeit only in the solid state) whereas
the PiBu3 analog cannot, although relative solubilites might
also play a part in the latter. It is well established that
L-M-L bite angles exert a significant influence on reduc-
tive elimination, and oxidative addition reactions at metal
centers61,62 and PiBu3, with its smaller cone angle, allows a
tighter P-Rh-P angle in the Rh(III) dihydride that could
encourage H2 reductive elimination. Dubois and co-workers
have reported that phosphine bite angles play an important
role in determining the position of equilibrium on the addi-
tion of H2 to [Rh(L2)2]þ cations (L2 = chelating diphos-
phine), with a small bite angle favoring the reductive elimina-
tion of H2, while those with a larger bite angle favor the
formation of [Rh(H)2(L2)2]þ.63 Although a direct compar-
ison between the DuBois system and ours is not appropriate
given the difference in ligand sets, consideration of the role of
the P-Rh-P angle does thus appear to be important when
determining relative stabilities of dihydride species bound
with these σ ligands. In addition, agostic interactions from
the phosphines could lower the energy of the transition state
for H2 elimnation from the Rh(III) intermediates by stabili-
zation of a low-coordinate structure formed during H2 loss.
In support of this, we have previously reported that VIII—
˚
over 3 A molecular sieves. H3B NMe3 and H3B NMe2H
3
3
were purchased from Aldrich and sublimed before use. [Rh-
(C6H5F)(PiPr3)2][BArF4],48 [Rh(C6H5F)(PiBu3)2][BArF4],64
[Rh(H)2(PiPr3)2][BArF ],54 [Rh(NBD)(PiPr3)2][BArF ],54 and
4
4
[H2BNMe2]227 were prepared according to literature methods.
All other chemicals are commercial products and were used as
received. NMR spectra were recorded on a Varian Unity or a
Bruker AVC 500 MHz spectrometer at room temperature,
unless otherwise stated. Chemical shifts are quoted in parts per
million and coupling constants in hertz. ESI-MS were recorded
on a Bruker MicroOTOF-Q instrument.67 Microanalyses were
performed by Elemental Microanalysis Ltd.
[Rh(η2-H3B NMe3)(PiPr3)2][BArF4] (1). To a Schlenk flask
3
charged with [Rh(C6H5F)(PiPr3)2][BArF4] (0.091 g, 0.066 mmol)
and H3B NMe3 (0.005 g, 0.069 mmol) was added 1,2-C6H4F2
3
(2 mL). The resulting purple solution was stirred at room
temperature for 5 min and then layered with pentane and held at
5 °C for 72 h to afford the product as purple crystals. Yield:
0.058 g (65%). Complex 1 is prepared quantitatively in situ by
the addition of a slight excess of H3B NMe3 to [Rh(C6H5F)-
3
(PiPr3)2][BArF4] in 1,2-C6H4F2.
1H NMR (CD2Cl2, 500 MHz): δ 7.70-7.74 (m, 8H, BArF4),
which in the solid state shows Rh HC agostic interac-
3 3 3
7.56 (br, 4H, BArF4), 2.81 (s, 9H, NMe), 1.99 (apparent octet,†
tions—readily loses H2 in solution, whereas IX does not,
although we do not have a solid-state structure for the
latter.64 Caulton and co-workers have previously commented
on the role that agostic interactions from phosphine ligands
might play in promoting H2 loss from iridium(III)-dihy-
dride species.65
3
J ∼ 7, 6H, PCH; {31P@δ = 73.9} sept, JHH = 7.2), 1.30
(apparent dd,† J ∼ 14, J ∼ 7, 36H, PCMe; {31P@δ = 73.9} d,
3JHH = 7.2), -3.09 (partially collapsed q,‡ fwhm = 290 Hz,
1JBH = 80,‡ 3H, BH; {11B} δ -3.09, br s, fwhm = 60). [†These
signals are observed as complicated second-order multiplets
‡
that collapse on 31P decoupling. This resonance appears as a
The reasons behind the slower catalysis for H3B NHMe2
broad doublet with J ∼ 100. 1JBH from the 11B NMR spectrum.]
1H NMR (CD2Cl2, 500 MHz, 200 K): δ 7.71-7.76 (m, 8H,
BArF4), 7.56 (br, 4H, BArF4), 2.76 (s, 9H, NMe), 1.89 (apparent
octet, J ∼ 7, 6H, PCH), 1.23 (apparent dd, J ∼ 13, J ∼ 7, 36H,
PCMe), -6.85 (vbr, fwhm = 460 Hz, 2H, RhHB). The remain-
ing BH signal was not observed, presumably as it was very
broad. 1H NMR (CD2Cl2, 500 MHz, 190 K): δ 7.71-7.76 (m,
8H, BArF4), 7.55 (br, 4H, BArF4), 4.43 (vbr, fwhm = 350 Hz,
1H, BH), 2.75 (s, 9H, NMe), 1.87 (apparent octet, J ∼ 7, 6H,
3
dehydrogenation when compared with PiBu3 analogs remain
(59) Albright, T. A.; Burdett, J. K.; Whangbo, M. H. Orbital Interactions
in Chemistry; Wiley: New York, 1985.
(60) Su, M. D.; Chu, S. Y. Inorg. Chem. 1998, 37, 3400–3406.
(61) Brown, J. M.; Guiry, P. J. Inorg. Chim. Acta 1994, 220, 249–259.
(62) Freixa, Z.; van Leeuwen, P. Dalton Trans. 2003, 1890–1901.
(63) DuBois, D. L.; Blake, D. M.; Miedaner, A.; Curtis, C. J.; DuBois,
M. R.; Franz, J. A.; Linehan, J. C. Organometallics 2006, 25, 4414-4419 and
references therin.
(64) Douglas, T. M.; Chaplin, A. B.; Weller, A. S. Organometallics 2008,
27, 2918–2921.
(65) Cooper, A. C.; Huffman, J. C.; Caulton, K. G. Organometallics 1997,
16, 1974–1978.
(66) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518–1520.
(67) Lubben, A. T.; McIndoe, J. S.; Weller, A. S. Organometallics 2008,
27, 3303–3306.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 1119
PCH), 1.22 (apparent dd, J ∼ 13, J ∼ 7, 36H, PCMe), -6.91
(vbr, fwhm = 220 Hz, 2H, RhHB). 13C{1H} NMR (CD2Cl2, 126
MHz): δ 162.3 (q, JBC = 50, BArF4), 135.4 (s, BArF4), 129.4 (qq,
C54H60BF24N2P2Rh (1368.71 g mol-1): C, 47.39; H, 4.42; N,
2.05. Found: C, 47.21; H, 3.94; N, 2.38.
[Rh(H)2(η2-H3B NMe3)(PiPr3)2][BArF4] (3). To a Schlenk
3
J
FC=32, JBC=3, BArF4), 125.1 (q, JFC=272, BArF4), 118.0
flask charged with [Rh(H)2(PiPr3)2][BArF4] {prepared by the
hydrogenation (1 atm) of [Rh(NBD)(PiPr3)2][BArF4] (0.100 g,
0.073 mmol) in CH2Cl2, followed by removal of the solvent in
(apparent sept, J=4, BArF4), 53.3 (s, NMe), 27.3-27.9 (second-
order m, PCH), 21.0 (s, PCMe). 11B NMR (CD2Cl2, 160 MHz):
δ 28.8 (br q, 1JBH = 80, 1B, H3B NMe3), -6.6 (s, 1B, BArF4).
vacuo}54 and H3B NMe3 (0.0056 g, 0.077 mmol) was added 1,2-
3
3
11B NMR (CD2Cl2, 160 MHz, 200 K): δ 28.9 (vbr, fwhm =
C6H4F2 (2 mL). The resulting yellow solution was stirred at
room temperature for 5 min, then layered with pentane and held
at 5 °C for 72 h to afford the product as colorless crystals. Yield:
0.065 g (65%). Compound 3 is prepared quantitatively in situ by
400 Hz, 1B, H3B NMe3), -6.8 (s, 1B, BArF4). 31P{1H} NMR
3
(CD2Cl2, 202 MHz): δ 73.9 (d, JRhP = 180). 31P{1H} NMR
1
(CD2Cl2, 202 MHz, 200 K): δ 71.8 (d, JRhP = 176). 31P{1H}
1
NMR (1,2-C6H4F2, 202 MHz): δ 73.8 (d, 1JRhP = 180). ESI-MS
(CH2Cl2, 60 °C, 4.5 kV) positive ion: m/z 496.2823 [M]þ (calcd,
496.2880). Anal. Calcd for C53H66B2F24NP2Rh (1359.55 g
mol-1): C, 46.82; H, 4.89; N, 1.03. Found: C, 46.82; H, 5.49;
N, 1.25.
the hydrogenation (1 atm) of [Rh(η2-H3B NMe3)(PiPr3)2]-
3
[BArF4] in 1,2-C6H4F2.
1H NMR (CD2Cl2, 500 MHz): δ 7.70-7.74 (m, 8H, BArF4),
7.56 (br, 4H, BArF4), 2.76 (s, 9H, NMe), 2.14-2.26 (second-
order m,† 6H, PCH; {31P@δ = 64.5} sept, JHH = 7.1), 1.30
=
3
[Rh{η2-(H2BNMe2)2}(PiPr3)2][BArF4] (2). To a Schlenk flask
charged with [Rh(C6H5F)(PiPr3)2][BArF4] (0.060 g, 0.043 mmol)
and [H2BNMe2]2 (0.0052 g, 0.046 mmol) was added 1,2-C6H4F2
(2 mL). The resulting purple solution was stirred at room
temperature for 5 min and then layered with pentane and held
at 5 °C for 72 h to afford the product as purple crystals. Yield:
0.051 g (85%). Complound 2 is prepared quantitatively in situ
by the addition of a slight excess of [H2BNMe2]2 to [Rh(C6H5F)-
(PiPr3)2][BArF4] in 1,2-C6H4F2.
(apparent q,† J ∼ 7, 36H, PCMe; {31P@δ = 64.5} d, JHH
3
7.1), -0.57 (partially collapsed q,‡ fwhm = 260 Hz, 1JBH = 80,‡
3H, BH; {11B} δ -0.57, d, J = 11, fwhm = 24), -19.75 (dt,
=
1JRhH=21.6, 2JPH=14.6, 2H, RhH; {31P@δ=64.5} d, 2JRhH
21.8). [†These signals are observed as complicated second-order
multiplets that collapse on 31P decoupling. ‡This resonance
appears as a broad doublet with J ∼ 110. JBH from the 11B
1
NMR spectrum.] 1H NMR (CD2Cl2, 500 MHz, 200 K): δ
7.70-7.75 (m, 8H, BArF4), 7.55 (br, 4H, BArF4), 2.71 (s, 9H,
NMe), 2.06-2.18 (second-order m, 6H, PCH), 1.23 (apparent q,
J ∼ 7, 36H, PCMe), -0.79 (vbr, fwhm = 150 Hz, 3H, BH),
-19.56 (dt, 1JRhH=21.5, 2JPH=14.2, 2H, RhH). No significant
change is observed on cooling further to 190 K. 13C{1H} NMR
(CD2Cl2, 126 MHz): δ 162.3 (q, JBC = 50, BArF4), 135.4 (br,
1H NMR (CD2Cl2, 500 MHz): δ 7.70-7.74 (m, 8H, BArF4),
7.56 (br, 4H, BArF4), 2.70 (vbr, fwhm ∼ 300 Hz, 2H, BH; {11B}
br), 2.63 (s, 12H, NMe), 2.10 (apparent octet,† J ∼ 7, 6H, PCH;
{31P@δ = 68.2} sept, 3JHH = 7.2), 1.37 (apparent dd,† J ∼ 14,
3
J∼7, 36H, PCMe; {31P@δ=68.2} d, JHH=7.2), -6.50 (ap-
parent qt,‡1JBH=89.0, 1JRhH = 30.7, 2JPH = 30.4, 2H, RhHB;
BArF4), 129.4 (qq, JFC = 32, JBC = 3, BArF4), 125.1 (q, JFC
272, BArF4), 118.0 (apparent sept, J = 4, BArF4), 53.9
=
1
{11B} apparent t, J ∼ 30; {31P@δ = 68.2} qd, JBH = 89.0,
1JRhH = 30.7). [†These signals are observed as complicated
1,3
2
=12, JRhC=1, PCH), 20.8
(obscured, NMe), 26.7 (td,
J
PC
‡
second-order multiplets that collapse on 31P decoupling. This
(dd, 2JPC=2, J =1, PCMe). 11B NMR (CD2Cl2, 160 MHz): δ 5.0
complex AA0MNXX0 signal was simulated using gNMR49 to
extract the corresponding spectral parameters.] 13C{1H} NMR
(CD2Cl2, 126 MHz): δ 162.3 (q, JBC = 50, BArF4), 135.4 (s,
(br q, JBH = 80, 1B, H3B NMe3), -6.6 (s, 1B, BArF4). 11B
1
3
NMR (CD2Cl2, 160 MHz, 200 K): δ 5.4 (vbr, fwhm = 380 Hz,
1B, H3B NMe3), -6.7 (s, 1B, BArF4). 31P{1H} NMR (CD2Cl2,
3
202 MHz): δ 64.5 (d, 1JRhP = 111). 31P{1H} NMR (CD2Cl2, 202
BArF4), 129.4 (qq, JFC = 32, JBC = 3, BArF4), 125.1 (q, JFC
=
MHz, 200 K): δ 63.7 (d, JRhP = 111). 31P{1H} NMR (1,2-
1
273, BArF4), 118.0 (apparent sept, J = 4, BArF4), 51.2 (s, NMe),
27.9-28.5 (second-order m, PCH), 21.1 (s, PCMe). 11B NMR
(CD2Cl2, 160 MHz): δ 35.3 (t, 1JBH = 89, 1B, RhH2B), 5.3 (t,
1JBH = 112, 1B, BH), -6.6 (s, 1B, BArF4). 31P{1H} NMR
C6H4F2, 202 MHz): δ 64.6 (d, 1JRhP=111). ESI-MS (CH2Cl2,
60 °C, 4.5 kV) positive ion: m/z 425.1902 [M - H3B NMe3]þ
3
(calcd, 425.1973). Anal. Calcd for C53H68B2F24NP2Rh (1361.57
g mol-1): C, 46.75; H, 5.03; N, 1.03. Found: C, 46.62; H, 5.01;
N, 1.12.
1
(CD2Cl2, 202 MHz): δ 68.2 (d, JRhP = 177). 31P{1H} NMR
1
(1,2-C6H4F2, 202 MHz): δ 68.2 (d, JRhP = 178). ESI-MS
[Rh(H)2{η2-(H2BNMe2)2}(PiPr3)2][BArF4] (4). A solution of
[Rh{η2-(H2BNMe2)2}(PiPr3)2][BArF4] (0.040 g, 0.029 mmol) in
1,2-C6H4F2 (2 mL) was placed under hydrogen (4 atm). The
resulting pale yellow solution was left to stand for 5 min and then
layered with pentane under hydrogen (1 atm) and held at 5 °C
for 72 h to afford the product as colorless crystals. Yield ∼ 0.01 g
(25%). NMR spectra of this material dissolved in CD2Cl2 at
225 K indicate complete dissociation of [H2BNMe2]2; only
[Rh(H)2(PiPr3)2][BArF4] and free [H2BNMe2]2 are seen. Com-
plex 4 is unstable in the absence of hydrogen and converts to 2
over days in the solid state under an argon atmosphere. This
process is accelerated by placing the sample under a vacuum.
Characterization was carried by X-ray crystallography on this
material. Compound 4 can alternatively be prepared by the
hydrogenation of 2 in the solid state (4 atm, 48 h).
(CH2Cl2, 60 °C, 4.5 kV) positive ion: m/z 537.3316 [M]þ (weak
peak, calcd, 537.3316). Anal. Calcd for C54H70B3F24N2P2Rh
(1400.41 g mol-1): C, 46.31; H, 5.04; N, 2.00. Found: C, 46.30;
H, 4.88; N, 2.06.
trans-[Rh(PiPr3)2(MeCN)2][BArF4]2. To a Schlenk flask
charged with [Rh(C6H5F)(PiPr3)2][BArF4] (0.055 g, 0.043 mmol)
was added MeCN (0.011 mL, 0.22 mmol) followed by CH2Cl2 (2
mL). The resulting yellow solution was stirred at room tem-
perature for 48 h and recrystallized from CH2Cl2/pentane.
Yield: 0.040 g (68%). Alternatively prepared by the addition
of excess MeCN to 1 or 2.
1H NMR (CD2Cl2, 500 MHz): δ 7.70-7.74 (m, 8H,
BArF4), 7.56 (br, 4H, BArF4), 2.11-2.22 (second-order multi-
plet,† 6H, PCH; {31P@δ=42.7} δ 2.17, sept, 3JHH=7.2), 2.17
5
4
(td, JPH = 1.7, JRhH = 0.7, 6H, MeCN), 1.33 (apparent q,†
[Rh(H)2(PiPr3)2][BArF4] (IX). Prepared as previously de-
scribed.54
J ∼ 7, 36H, PCMe; {31P@δ = 42.7} d, JHH = 7.2). [†These
3
signals are observed as complicated second-order multiplets
that collapse on 31P decoupling.] 13C{1H} NMR (CD2Cl2,
126 MHz): δ 163.9 (q, JBC =50, BArF4), 136.6 (br, BArF4),
130.6 (qq, JFC = 32, JBC = 3, BArF4), 126.4 (br d, J = 13.4,
1H NMR (1,2-C6H4F2, 500 MHz): δ 8.33 (br, 8H, BArF4),
7.69 (br, 4H, BArF4), 2.28 (br, 6H, PCH), 1.24 (apparent q, J ∼
7, 36H, PCMe), -26.91 (br d, 1JRhH = 40, 2H, RhH). 31P{1H}
NMR (1,2-C6H4F2, 202 MHz): δ 60.2 (d, 1JRhP = 112).
NMR Experiments.
MeCN), 126.2 (q, JFC = 275, BArF4), 119.0 (apparent sept,
J = 4, BArF4), 23.3 (t,
J
= 10, PCH), 19.4 (s, PCMe),
1,3
PC
3.9 (s, MeCN). 31P{1H} NMR (CD2Cl2, 202 MHz): δ 42.6
1. Reaction of 1, 2, and 3 with MeCN.
a. To a solution of 1/2 (0.008 g, 0.0059/0.0057 mmol) in
CD2Cl2 (0.4 mL) in a J. Young NMR tube was added MeCN
1
(d, JRhP = 128). ESI-MS (CH2Cl2, 60 °C, 4.5 kV) positive
ion: m/z 505.2347 [M]þ (calcd, 505.2348). Anal. Calcd for

1120 Inorganic Chemistry, Vol. 49, No. 3, 2010
Chaplin and Weller
Table 2. Crystallographic Data
1
2
trans-[Rh(PiPr3)2(NCMe)2][BArF
]
3
4
4
formula
C
53H66B2-
F24NP2Rh
C
54H70B3F24
N2P2Rh
-
C
54H60BF24
N2P2Rh
-
C
53H68B2-
F24NP2Rh
C
54H72B3F24
N2P2Rh
-
M
1359.54
triclinic
P1
150(2)
12.4906(2)
13.4472(2)
19.3634(3)
77.0895(6)
86.5573(6)
82.4775(5)
3141.25(8)
2
1400.40
triclinic
P1
1368.70
monoclinic
P21/c
1361.55
triclinic
P1
150(2)
12.4135(2)
13.8862(3)
19.2604(4)
105.4259(10)
104.2746(10)
90.2780(8)
3092.79(10)
2
1402.42
tetragonal
P41212
220(2)
13.6939(7)
cryst syst
space group
T [K]
150(2)
150(2)
˚
a [A]
12.76990(10)
19.72410(10)
27.1519(2)
73.7772(2)
86.7370(3)
73.2126(3)
6284.63(7)
4
1.480
0.431
5.11 e θ e 26.37°
44379
20.75950(10)
29.7418(2)
21.3100(2)
˚
b [A]
˚
c [A]
34.408(3)
R [deg]
β [deg]
γ [deg]
108.5254(5)
3
˚
V [A ]
Z
12475.6(2)
8
1.457
0.433
5.09 e θ e 26.37°
46573
6452.3(7)
4
1.444
0.420
5.11 e θ e 25.03°
6046
density [g cm-3
]
1.437
0.428
5.14 e θ e 26.37°
19950
1.462
0.435
5.10 e θ e 26.37°
20886
μ (mm-1
)
θ range [deg]
reflns collected
[Rint = 0.0182]
97.3%
12518/715/946
[Rint = 0.0189]
99.1%
25462/1074/1818
[Rint = 0.0476]
99.1%
25306/1773/1913
[Rint = 0.0244]
98.7%
12505/592/978
[Rint = 0.0345]
88.1%
4435/980/517
completeness
no. of data/restraints/
params
R1 [I > 2σ(I)]
wR2 [all data]
GoF
largest diff. peak
˚
and hole [e A
0.0428
0.1087
1.028
1.165 and -0.742a
0.0408
0.1037
1.021
0.764 and -0.855
0.0518
0.1360
1.024
0.624 and -0.569
0.0443
0.1142
1.020
0.709 and -0.555
0.0692
0.1702
1.057
0.596 and -0.476
-3
]
a
The presence of a relatively large Fourier peak (1.16 e A-3) near the metal center in this structure is attributed to Fourier truncation errors. If this
˚
peak is excluded, the max/min residual density is 0.73/-0.74.
(5 μL, excess), resulting in an immediate change in color from
purple to bright yellow. An intermediate species, cis-[Rh-
placed under argon (no change of color was observed). The
addition of CH2dCHtBu (5 μL, 0.039 mmol) resulted in an
immediate change from pale yellow to deep purple, and the
quantitative formation of 1 as determined by 1H and 31P NMR
1
(PiPr3)2(NCMe)2][BArF4] {δ31P 53.4, d, JRhP = 179}, was
observed initially by 1H and 31P NMR spectroscopy, which
t
subsequently converted to trans-[Rh(PiPr3)2(NCMe)2][BArF
(pale yellow solution) quantitatively over ca. 30 min. Free
]
spectroscopy (MeCH2 Bu was also observed).
4
c. A Schlenk flask changed with a crystalline sample of 4 (ca.
0.01 g, prepared as described above by crystallization) under an
atmosphere of hydrogen was cooled to 195 K and then placed
under argon. The solid was extracted with cold CD2Cl2 (195 K)
into a J. Young NMR tube, warmed to 225 K, and analyzed by
NMR spectroscopy. The quantitative formation of IX, con-
taminant with free [H2BNMe2]2, was observed. On warming
further to 298 K, a mixture of IX and 2 was observed (1X/2 =
1:0.2).
H3B NMe3 and [H2BNMe2]2, respectively, were the only spe-
3
cies observed by 11B NMR spectroscopy (with the exception of
the anion resonance, [BArF4]-).
b. To a solution of 3 in (0.008 g, 0.0059 mmol) in CD2Cl2
(0.4 mL) in a J. Young NMR tube was added MeCN (5 μL,
excess) and was immediately analyzed—the formation of
[Rh(H)2(PiPr3)2(NCMe)2][BArF4], concomitant with the forma-
tion of H3B NMe3, was quantitative by 1H, 11B, and 31P NMR
3
4. Reaction of IX with H3B NMe3 and [H2BNMe2]2. Com-
spectroscopy.
2. Reaction of 1 and 2 with H2.
3
pound IX was prepared by the hydrogenation (1 atm) of
[Rh(NBD)(PiPr3)2][BArF4] (0.010 g, 0.073 mmol) in CH2Cl2
(0.4 mL) in a J. Young NMR tube, followed by the removal of
solvent in vacuo.54
a. A solution of 1 (0.010 g, 0.0074 mmol) in 1,2-C6H4F2
(0.4 mL) was hydrogenated (1 atm) in a J. Young NMR tube and
immediately analyzed—the formation of 3 was quantitative by
1H and 31P NMR spectroscopy.
a. To a solution of IX (0.0073 mmol, prepared as described
above) in 1,2-C6H4F2 (0.4 mL) in a J. Young NMR tube was
added H3B NMe3 (0.001 g, 0.014 mmol) and immediately
b. A solution of 2 (0.010 g, 0.0071 mmol) in 1,2-C6H4F2
(0.4 mL) was hydrogenated (1 atm) in a J. Young NMR tube and
then immediately analyzed—the formation of IX, concomitant
with the formation of [H2BNMe2]2, was quantitative as deter-
3
analyzed—the formation of [Rh(H)2(η2-H3B NMe3)(PiPr3)2]-
3
[BArF4] was quantitative by 1H and 31P NMR spectroscopy.
b. To a J. Young NMR tube charged with IX (0.0073 mmol,
prepared as described above) and [H2BNMe2]2 (2, 10, or
50 equiv) was added 1,2-C6H4F2 (0.4 mL), resulting in a mixture
of IX and 2 by 1H and 31P NMR spectroscopy (2 equiv 1X/2 =
1:4, 10 equiv 1X/2 = 1:10, 50 equiv 1X/2 = essentially only 2). A
tiny amount of dissolved H2 (δ 4.7) was observed in each case.
1
mined by H, 11B, and 31P NMR spectroscopy. The hydride
signal for IX was shifted ca. 8 ppm downfield and broadened
(fwhm = 600 Hz) in comparison to isolated material, presum-
ably due to exchange with dissolved H2.
3. Stability of 3 and 4.
a. A solution of 3 (0.009 g, 0.0066 mmol) in CD2Cl2 (0.4 mL)
in a J. Young NMR tube was degassed by three successive
freeze, pump, and thaw cycles and left to stand for 72 h under a
vacuum (5 ꢀ 10-3 Torr). No change could be detected by 1H and
31P NMR spectroscopy.
b. A solution of 3 (0.0074 mmol) in 1,2-C6H4F2 (0.4 mL) in a
J. Young NMR tube, prepared as described above (1a), was
degassed by three successive freeze, pump, and thaw cycles and
5. Reaction of [Rh(C6H5F)(PiPr3)2][BArF
] with H3B
NMe2H. To a J. Young NMR tube charged with [Rh(C6H5F)-
4
3
(PiPr3)2][BArF4] (0.010 g, 0.0072 mmol) and H3B NMe2H
3
(0.0043 g, 0.0730 mmol, 10 equiv) was added 1,2-C6H4F2
(0.4 mL) to give an initially purple solution, which very rapidly
(<10 s) became colorless. The reaction was then monitored by
NMR spectroscopy. After 10 h, conversion of H3B NMe2H
3

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 1121
was ca. 30%; after 48 h, conversion was ca. 70%. [H2BNMe2]2
was the major product at both times, and the only observed
metal-containing species at these points was [Rh(H)2(η2-
1JBH = 94, 1B, BH3)}27 was observed as an intermediate
species—no other significant species were observed using 11B
NMR spectroscopy under these conditions.
H3B NMe2H)(PiPr3)2][BArF4]—characterized in situ by NMR
Crystallography. Relevant details about the structure refine-
ments are given in Table 2, and the structures are depicted
in Figures 1, 2, 4, 5, and 6. Data were collected on an Enraf
Nonius Kappa CCD diffractometer using graphite mono-
3
spectroscopy and ESI-MS. Furthermore, on the basis of 1H
integrals, and within the implied errors associated with their
measurement, this rhodium complex is formed in a quantitative
amount from the precatalyst (comparison to [BArF - integral).
]
˚
chromated Mo KR radiation (λ = 0.71073 A) and a low-
4
[Rh(H)2(η2-H3B NMe2H)(PiPr3)2][BArF
]
4
(5). 1H NMR
temperature device.68 Data were collected using COLLECT;
reduction and cell refinement were performed using DENZO/
SCALEPACK.69 The structures were solved by direct meth-
ods using SIR2004 (1, 2, trans-[Rh(PiPr3)2(NCMe)2][BArF4],
3)70 and SHELXS-97 (4)71 and refined full-matrix least-
squares on F2 using SHELXL-97. All non-hydrogen atoms
were refined anisotropically. All Rh-H and B-H hydrogen
atoms were located on the Fourier difference map (with the
exception of H2A and H2B in 4); their isotropic displacement
parameters were fixed to ride on the parent atoms. All other
hydrogen atoms were placed in calculated positions using the
riding model. The following restraints were applied: B1-
H1A=B1-H1B in 1; B1-H1A=B1-H1B=B11-H11A=
B11-H11B in 2; B1-H1A=B1-H1B in 3; H0-P1=H0-P1*;
and H1-N1 = H1-N1* in 4 (where * refers to sym-
metry-related atoms). Disorder of the phosphine ligands in one
of the independent molecules in the structure trans-[Rh(PiPr3)2-
(NCMe)2][BArF4] and in one of the phosphine ligands in the
structure of 3 was treated by modeling the appropriate substitu-
ents over two sites and restraining their geometry. Rotational
disorder of the CF3 groups of the anions was treated by modeling
the fluorine atoms over two sites and restraining their geometry.
Restraints to thermal parameters were applied where necessary in
order to maintain sensible values. Graphical representations of
the structures were made with ORTEP3.72
3
(1,2-C6H4F2, 500 MHz): δ 8.33 (br, 8H, BArF4), 7.68 (br, 4H,
BArF4), 3.74 (br, 1H, NH), 2.87 (s, 6H, NMe), 2.05-2.15
(second-order m, 6H, PCH), 1.30 (apparent q, J ∼ 7, 36H,
PCMe), -0.97 (vbr, fwhm = 350 Hz, 3H, BH), -18.03 (dt,
1JRhH = 21.0, JPH = 13.5, 2H, RhH). 1H NMR (CD2Cl2,
2
500 MHz, 200 K, selected data): δ -3.56 (br, 2H, RhHB),
-18.01 (m, 2H, RhH). 11B NMR (1,2-C6H4F2, 160 MHz): δ
3.4 (br, 1B, H3B NMe2H), -6.1 (s, 1B, BArF4). 31P{1H} NMR
3
(1,2-C6H4F2, 202 MHz): δ 67.4 (d, 1JRhP = 109). 31P{1H} NMR
(CD2Cl2, 202 MHz, 200 K): δ 69.1 (dd, 2JPP = 300, 1JRhP = 109,
2
1
1P), 64.9 (dd, JPP = 300, JRhP = 110, 1P). ESI-MS (1,2-
C6H4F2/CH2Cl2, 60 °C, 4.5 kV) positive ion: m/z 484.2832 [M]þ
(calcd, 484.2880).
Dehydrocoupling of H3B NMe2H. To a Schlenk flask, con-
3
nected to an external mineral oil bubbler and charged with a
magnetic stirrer bar and the rhodium precursor [Rh(C6H5F)-
(PR3)][BArF4] (R = iBu, 0.0265 g, 0.018 mmol; R = iPr, 0.0250
g, 0.018 mmol), was added a solution of H3B NMe2H (0.072 M,
3
5.0 mL, 0.36 mmol, 20 equiv) in 1,2-C6H4F2. Reaction progress
was monitored by analyzing regular aliquots of the reaction
solution (0.1 mL; immediately quenched by the addition of 0.4
mL of MeCN) by 11B NMR spectroscopy. Under these condi-
tions, complete conversion of H3B NMe2H {δ -13.0, q, 1JBH
=
3
1
96} to [H2BNMe2]2 {δ 6.0, t, JBH = 112}27 occurred after
80 min with [Rh(C6H5F)(PiBu3)2][BArF4] and 300 min with
[Rh(C6H5F)(PiPr3)2][BArF4]. The diborazane [H3B NMe2-
Acknowledgment. The EPSRC and the University of
Oxford for support.
3
1
BH2 NMe2H] {δ 2.9 (t, JBH = 109, 1B, BH2), δ -12.1 (q,
3 3
(68) Cosier, J.; Glazer, A. M. J. Appl. Crystallogr. 1986, 19, 105–107.
(69) Otwinowski, Z.; Minor, W. Methods in Enzymology. In Macro-
molecular Crystallography, part A; Carter, C. W., Jr., Sweet, R. M., Eds.; 1997;
Vol. 276, pp 307-326.
(70) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano,
G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl.
Crystallogr. 2005, 38, 381–388.
Supporting Information Available: Figure S-1 to S-4 (selected
NMR spectra and ESI-MS). This material is available free of
charge via the Internet at http://pubs.acs.org.
(71) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112–122.
(72) Farrugia, L. J. Appl. Crystallogr. 1997, 30, 565–565.