Effect of ligand modification on the reactivity of phosphinoamide-bridged heterobimetallic Zr/Co complexes

Article abstract of DOI:10.1039/c3dt52133h

The effect of modifying the N-aryl substituent (aryl = mesityl vs. m-xylyl) of the phosphinoamide ligands linking Zr and Co in tris(phosphinoamide)-linked heterobimetallic complexes has been investigated. Treatment of the metalloligand (ⁱPr₂PNXyl)₃ZrCl (2) (Xyl = m-xylyl) with CoI₂ affords the iodide-bridged product ICo(ⁱPr ₂PNXyl)₂(μ-I)Zr(η^2-iPr₂PNXyl) (3) rather than the C₃-symmetric isomer observed using the N-mesityl derivative, ICo(ⁱPr₂PNMes)₃ZrCl. Upon two-electron reduction of complex 3, ligand rearrangement occurs to generate the three-fold symmetric reduced complex N₂Co(ⁱPr ₂PNXyl)₃Zr(THF) (4). Comparison of 4 with the previously reported mesityl-substituted complex N₂Co(ⁱPr ₂PNMes)₃Zr(THF) (1) reveals similar structural features but with a less sterically hindered Zr apical site in complex 4. An obvious electronic difference between these two complexes is also present based on the drastically different infrared N₂ stretching frequencies of 1 and 4. These notable differences lend themselves to different reactivity in both stoichiometric and catalytic reactions. Alkyl halide addition to complex 4 results in homo-coupling products resulting from alkyl radicals rather than the alkyl-bridged or intramolecular C-H activation products formed upon addition of RX to 1. This difference in reactivity with alkyl halides renders complex 3 a less effective catalyst for the Kumada cross-coupling of alkyl halides with n-octylMgBr than ICo(ⁱPr₂PNMes)₃ZrCl, as a greater proportion of homocoupling products are formed under catalytic conditions.

Full text of DOI:10.1039/c3dt52133h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Effect of ligand modification on the reactivity of
phosphinoamide-bridged heterobimetallic Zr/Co
complexes†
Cite this: Dalton Trans., 2014, 43,
1984
Wen Zhou, Noam I. Saper, Jeremy P. Krogman, Bruce M. Foxman and
Christine M. Thomas*
The effect of modifying the N-aryl substituent (aryl = mesityl vs. m-xylyl) of the phosphinoamide ligands
linking Zr and Co in tris(phosphinoamide)-linked heterobimetallic complexes has been investigated.
Treatment of the metalloligand (iPr2PNXyl)3ZrCl (2) (Xyl = m-xylyl) with CoI2 affords the iodide-bridged
product ICo(iPr2PNXyl)2(μ-I)Zr(η2-iPr2PNXyl) (3) rather than the C3-symmetric isomer observed using the
N-mesityl derivative, ICo(iPr2PNMes)3ZrCl. Upon two-electron reduction of complex 3, ligand rearrange-
ment occurs to generate the three-fold symmetric reduced complex N2Co(iPr2PNXyl)3Zr(THF) (4). Com-
parison of 4 with the previously reported mesityl-substituted complex N2Co(iPr2PNMes)3Zr(THF) (1)
reveals similar structural features but with a less sterically hindered Zr apical site in complex 4. An obvious
electronic difference between these two complexes is also present based on the drastically different infra-
red N2 stretching frequencies of 1 and 4. These notable differences lend themselves to different reactivity
in both stoichiometric and catalytic reactions. Alkyl halide addition to complex 4 results in homo-coupling
products resulting from alkyl radicals rather than the alkyl-bridged or intramolecular C–H activation pro-
ducts formed upon addition of RX to 1. This difference in reactivity with alkyl halides renders complex 3 a
less effective catalyst for the Kumada cross-coupling of alkyl halides with n-octylMgBr than ICo(iPr2PN-
Mes)3ZrCl, as a greater proportion of homocoupling products are formed under catalytic conditions.
Received 5th August 2013,
Accepted 18th November 2013
DOI: 10.1039/c3dt52133h
www.rsc.org/dalton
working towards synthesizing a library of early/late hetero-
bimetallic complexes (Co/Zr,4 Co/Hf,5 Pt/Zr,6 V/Fe,7 Nb/Co,8
Introduction
In recent years, much attention has been devoted to the study Ta/Co8) featuring metal–metal interactions of varying
of early/late heterobimetallic complexes, under the premise strengths and polarities to better understand the effect of
that the distinct electronic nature of the two metals involved these interactions on reactivity.
will promote different reactivity from monometallics, homo-
The most well-studied metal–metal combination to date is
bimetallics, or polynuclear clusters.1–3 While a number of het- Co/Zr, and the Co–Zr multiply-bonded complex N2Co(iPr2PN-
erobimetallic complexes have been synthesized, there are few Mes)3Zr(THF) (1, Mes = 2,4,6-trimethylphenyl)9 has been
examples of legitimate catalytic applications that would not be shown to activate a variety of σ and π bonds such as those in
feasible with one of the metals alone.3 With this in mind, our H2,10 CO2,11 hydrazines,12 benzophenone,13 alkyl halides,10
group has been exploring flexible LX-type phosphinoamide thiols,14 and alcohols.14 Using these bond activation pro-
ligand scaffolds to promote strong interactions between two cesses, we have thus far uncovered two catalytic applications
different metal centers, with the ultimate goal of developing for tris(phosphinoamide)-linked Co/Zr heterobimetallic com-
homogeneous catalysts.4 Using these ligands we have been plexes, including Kumada coupling.15,16 Since reduced Co/Zr
complexes such as 1 were shown to react readily with alkyl
halides and Grignard reagents proved sufficiently reducing to
Department of Chemistry, Brandeis University, 415 South Street MS 015, Waltham,
generate 1 from CoI/ZrIV dihalide precursors, a catalytic cycle for
MA, USA. E-mail: thomasc@brandeis.edu; Fax: +1 (781)736-2516;
C–C bond-forming cross coupling could be envisioned. Indeed,
a series of Co/Zr complexes ICo(R2PNR′)3ZrCl (R = Ph, Pr, R′ =
Tel: +1 (781)736-2576
†Electronic supplementary information (ESI) available: Additional crystallo-
i
graphic details for complexes 4 and 5, 1H NMR spectra of 3 and 5, as well as the
1H NMR spectrum of the reaction of 4 with MeI, cyclic voltammetry data for
complex 3, and computational details. CCDC 953851 (4) and 953852 (5). For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c3dt52133h
iPr, Mes) with varying phosphine and amide ligand substituents
was shown to be an active family of precatalysts for the Kumada
coupling of organic Grignard reagents with alkyl halides.16
Monometallic tris(phosphinoamide) ZrIV complexes did not
1984 | Dalton Trans., 2014, 43, 1984–1989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
show any catalytic activity under identical conditions. Mono-
metallic CoI tris(phosphine) complexes were far less catalytically
active than the bimetallic complexes and, while the Co/Zr bi-
metallics showed moderate catalytic activity toward more
challenging alkyl chloride substrates, monometallic CoI com-
plexes were completely inactive towards alkyl chlorides.
While considerable time has been spent examining new
metal–metal combinations, little effort has been devoted to
probing the effects of small modifications of the phosphino-
amide ligand on electronic structure and/or stoichiometric or
catalytic activity. In this manuscript we have chosen to
examine the effects of changing the phosphinoamide nitrogen
substituent from mesityl(2,4,6-trimethylphenyl) to m-xylyl (3,5-
dimethylphenyl). The impetus for this seemingly minor
change stems from the following: in studies of stoichiometric
reactivity, it was found that the activation of alkyl halides RX
(where is R is anything larger than Me) by 1 is accompanied by
intramolecular ligand C–H activation at a mesityl–methyl
group to generate (η3-PiPr2N{(CH3)2C6H2(CH2)})Zr(MesNPiPr2)2-
CoX quantitatively.10 As this pathway may represent a catalyst
deactivation mechanism, we postulated that the performance
of the Zr/Co catalysts in Kumada cross coupling reactions may be
improved by removing the ortho-methyl substituents and shutting
down the intramolecular C–H activation process.17 Thus, the
phosphinoamide ligand [iPr2PNXyl]− (Xyl = 3,5-dimethylphenyl),
recently reported by Fryzuk et al.,18 was used to support the Zr/Co
system. Herein we report the synthesis of heterobimetallic Zr/Co
complexes supported by [iPr2PNXyl]−, their stoichiometric reactiv-
ity with alkyl halides, and their activity towards Kumada-type
cross coupling reactions in the context of comparisons to our pre-
vious work with the [iPr2PNMes]− analogue.
Scheme 1 Synthesis of heterobimetallic Zr/Co complexes.
stage, it is difficult to predict whether steric or electronic
factors dictate this product preference. Nonetheless, the solu-
tion magnetic moment of 3 is μeff = 3.10 B.M., consistent with
the S = 1 state expected for a ZrIVCoI complex.
The cyclic voltammogram (CV) of 3 reveals two irreversible
reductive processes at −1.6 V and −1.8 V vs. ferrocene/ferroce-
nium (Fig. S2 and S3†). Likewise, reduction of 3 with 2.5 equiv.
Na/Hg under N2 generated a dark red solution of the two-elec-
tron reduced complex N2Co(iPr2PNXyl)3Zr(THF) (4), as shown
in Scheme 1. Complex 4 is diamagnetic, with a 31P{1H} NMR
spectral shift at δ 47 ppm and a well-defined 1H NMR spec-
trum consistent with a C3-symmetric structure.
X-ray quality crystals of 4 were obtained, and the solid state
structure is shown in Fig. 1. The structure of 4 is largely
Results and discussion
i
The phosphinoamine Pr2PNHXyl was prepared using modifi-
cations of literature procedures.18 Deprotonation of Pr2PNH-
i
Xyl with 1 equiv. of nBuLi at −78 °C, followed by treatment
with 1/3 equiv. of ZrCl4 led to isolation of the zirconium metal-
loligand (iPr2PNXyl)3ZrCl (2) as shown in Scheme 1. Complex 2
was then treated with CoI2 to generate a Co/Zr heterobimetallic
complex 3 (Scheme 1). The 1H NMR spectrum of the sole
product of this reaction revealed 12 distinct broad paramagne-
tically shifted resonances between δ 35 and δ −9 ppm,
suggesting an asymmetric structure analogous to the pre-
viously reported iodide-bridged complex ICo(iPr2PNMes)2(μ-I)-
Zr(η2-iPr2PNMes).10 While a solid state structure of 3 was not
obtained, making it difficult to determine the identity of the
halide ligands in the terminal and bridging positions,
repeated combustion analysis results were consistent with a
Fig. 1 Displacement ellipsoid (50%) representation of complex 4 (left)
and 19 (right). For clarity, hydrogen atoms and isopropyl methyl groups
complex containing two iodide ligands. Thus, it was deter- have been omitted. Only one of the two chemically equivalent but crys-
tallographically independent molecules in the asymmetric unit of 4 is
shown. Relevant interatomic distances (Å) for 4 [second molecule in
asymmetric unit]: Zr–Co, 2.3778(5) [2.3854(5)]; Zr–O, 2.3303(18)
[2.3285(19)]; Zr–N1, 2.142(2) [2.140(2)]; Zr–N2, 2.079(2) [2.081(2)];
mined that the heterobimetallic product isolated upon metal-
lation of 2 with CoI2 is the asymmetric iodide-bridged isomer
ICo(iPr2PNXyl)2(μ-I)Zr(η2-iPr2PNXyl) (3). In contrast, a similar
synthetic procedure using the [iPr2PNMes]− derivative gener-
Zr–N3, 2.148(2) [2.162(2)]; Co–P1, 2.2331(8) [2.2326(8)]; Co–P2, 2.2464(8)
ates the C3-symmetric complex ICo(iPr2PNMes)3ZrCl.19 At this [2.2356(7)], Co–P3, 2.2290(8) [2.2221(7)].
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1984–1989 | 1985

View Article Online
Paper
Dalton Transactions
similar to that of 1, with a few minor, yet noticeable differ- halides such as 1-bromopentane and 2-iodopropane, the
ences. The Zr–Co bond distance of 2.382 0.004 Å (avg. of two halide-bridged species XCo(iPr2PNXyl)2(μ-X)Zr(η2-iPr2PNXyl)
molecules in asymmetric unit) in 4 is slightly longer than that (X = I (3), Br (5)) is formed exclusively. As expected, the yield of
of 1 (2.365 0.003 Å).9 This would seem to indicate a slightly these complexes is dramatically improved if two equivalents of
weaker Zr/Co interaction in 4, although the difference is small RX are added. In the case of 1-bromopentane and 2-iodopro-
enough that other factors such as crystal packing may play a pane, the reaction mixtures were examined by GC-MS, reveal-
role. More significantly, the Zr–O bond distance of 2.329(1) Å ing the R–R coupling products, decane and 2,3-
in 4 is 0.12 Å shorter than that in 1 (2.452(5) Å). This may be a dimethylbutane. Further, the reaction of 4 with MeI was moni-
1
consequence of the decreased steric bulk at the Zr center in tored by H NMR spectroscopy, and a resonance at δ 0.79 ppm
complex 4. Although the Zr–Co distances would suggest confirmed the generation of ethane as a byproduct (Fig. S5†).
similar metal–metal interactions in complexes 4 and 1 there is The structure of complex 5 (and by extension, complex 3) was
a marked increase in the ν(N2) from 2026 cm−1 in 19 to confirmed crystallographically and is shown in Fig. 2. The
2045 cm−1 in 4. This indication of decreased π-back-bonding structure is generally quite similar to that reported for the
from Co to N2 either suggests (1) less electron donation from iodide-bridged N-mesityl complex ICo(iPr2PNMes)2(μ-I)Zr-
the phosphines to Co, or (2) a stronger interaction between Co (η2-iPr2PNMes).10
and Zr in complex 4, as it has been shown that the same Co
While the absence of an ortho-methyl substituent on the
orbitals of π symmetry (i.e. dxz and dyz) are involved in bonding N-aryl ring does prevent intramolecular C–H activation, bridged
to both N2 and Zr in these heterobimetallic systems.9 Compu- alkyl species remain inaccessible with this ligand system,
tational examination of complex 4 using density functional perhaps because the diminished steric bulk permits inner
theory (DFT) and natural bond orbital (NBO) methods suggests sphere radical coupling events. Nonetheless, to compare the
that the Co–Zr bond in 4 is actually slightly weaker than in 1. C–C cross-coupling activity of Co/Zr complexes as a function of
The Co–Zr Wiberg bond index (WBI) in complex 4 is calculated N-aryl substituent, complex 3 was screened as a precatalyst for
to be 0.91, while in complex 1 the Co–Zr WBI is 0.95.12 Thus, the Kumada cross coupling of alkyl halides (R–X) with n-octyl-
we conclude that the Co center in 4 is markedly less electron magnesium bromide. The yields of these reactions are shown
rich than in complex 1 as a result of the less electron-donating in Table 1, alongside those previously reported using ICo
nature of the [iPr2PNXyl]− ligand.
(iPr2PNMes)3ZrCl (7) as a precatalyst.16 Based on the yields of
The stoichiometric reactivity of 4 towards alkyl halide sub- cross-coupled product, which are lower in all cases using the
strates was examined to compare with that previously reported xylyl-substituted precatalyst 3, complex 7 shows superior per-
for 1.10 As shown in Scheme 2, treatment of 1 with MeI affords formance as a precatalyst for this reaction. Upon examination
the methyl-bridged complex ICo(iPr2PNMes)2(μ-CH3)Zr- of the reaction mixtures using GC-MS, significantly more
(η2-iPr2PNMes), while other alkyl halides such as iodoethane, homo-coupling byproducts, R–R, are formed when complex 3
2-iodopropane, and chlorocyclohexane lead exclusively to the
intramolecular
C–H
activation
product
(η3-PiPr2N-
{(CH3)2C6H2(CH2)})Zr(MesNPiPr2)2CoX (X = I, Cl). In contrast,
upon treatment of 4 with stoichiometric MeI or other alkyl
Fig. 2 Displacement ellipsoid (50%) representation of complex 5.
Hydrogen atoms have been omitted for clarity. In the representation
above, a dashed line is used to indicate a weak intermetallic donor–
acceptor interaction. Relevant interatomic distances (Å) and angles (°):
Zr–Co, 2.7602(3); Zr–Br2, 2.6374(3); Co–Br2, 2.3603(3); Co–Br1, 2.5273(4);
Co–P1, 2.2807(6); Co–P2, 2.2872(6); Zr–N1, 2.0893(17); Zr–N2, 2.1067(17);
Zr–N3, 2.1365(17); Zr–P3, 2.6716(5); Zr–Br2–Co, 64.569(9).
Scheme 2 Contrasting reactivity of complexes 1 and 4 with alkyl halide
substrates.
1986 | Dalton Trans., 2014, 43, 1984–1989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
study has shown that minor differences in ligand substituents
can dramatically impact the electronic structure and reactivity
of heterobimetallic Co/Zr complexes, and this will likely apply
to other heterobimetallic metal–metal combinations syn-
thesized using similar ligand sets. Future studies will focus on
additional systematic ligand modifications to elucidate the
implications of steric and electronic effects, with the overall
goal of tuning these complexes for specific applications in
catalysis.
Table 1 Comparison of the catalytic activity of complexes 3 and 7
towards the Kumada cross-coupling of alkyl halides with n-octylmagne-
sium bromide
Entry
RX
Yield,a cat = 3
Yield,a cat = 716
1
2
55.8%
39.4%
81.9%
61.3%
3
48.2%
55.9%
Experimental section
General considerations
4
5
24.3%
19.7%
45.8%
33.4%
All manipulations were carried out under an inert atmosphere
using a nitrogen-filled glovebox or standard Schlenk tech-
niques unless otherwise noted. All glassware was oven- or
flame-dried immediately prior to use. Tetrahydrofuran, diethyl
ether, dichloromethane, and toluene were degassed and dried
by sparging with ultra-high-purity argon gas, followed by
passage through a series of drying columns using a Seca
Solvent System by Glass Contour. All solvents were stored over
3 Å molecular sieves and frequently tested using a standard
solution of sodium benzophenone ketyl in tetrahydrofuran
to confirm the absence of oxygen and moisture. Benzene-d6
was degassed and dried over 3 Å molecular sieves before use.
All other reagents were purchased from Aldrich, Acros
Organics, or Alfa Aesar and dried and degassed thoroughly
prior to use. iPr2PNHXyl was prepared using modifications of a
literature procedure,18 described below. Complex 1 was pre-
pared as described previously.9 NMR spectra were recorded at
ambient temperature unless otherwise stated on a Varian
Inova 400 MHz instrument. 1H and 13C NMR chemical
shifts were referenced to residual solvent and are reported in
parts per million (ppm). GC-MS data were collected on an
Agilent 7890A GC System and 5975C VL MSD with a Triple-Axis
Detector. GC-MS yields were reported via integration of start-
ing materials and products compared to an internal standard
(tetradecane). Infrared spectra were recorded on a Varian 640-
IR spectrometer controlled by Resolutions Pro software. UV–vis
spectra were recorded on a Cary 50 UV-vis spectrophotometer
using Cary WinUV software. Elemental micro-analyses
were performed by Complete Analysis Laboratories, Inc.,
Parsippany, NJ.
a Average of two values obtained via GC-MS analysis using tetradecane
as an internal integration standard.
is used. Thus, we find that while intramolecular C–H acti-
vation may be a competing pathway during catalytic turnover
with precatalyst 7, the key drawback for complex 3 is its pro-
pensity to promote homo-coupling of alkyl halides.
Conclusions
In conclusion, we have found remarkable differences in the
reactivity and catalytic activity of Zr/Co complexes as a result of
a seemingly minor modification of the substituents on the
phosphinoamide ligands bridging the Zr and Co centers. The
N-aryl group on the phosphinoamide ligand [ArNPiPr2]− was
changed from mesityl to m-xylyl, leading to an apparent prefer-
ence for the ZrIV/CoI dihalide bridged isomer XCo(iPr2P-
NAr)2(μ-X)Zr(η2-iPr2PNAr) over the C3-symmetric isomer XCo-
(iPr2PNAr)3ZrX. While two-electron reduction of both ZrIV/CoI
dihalide complexes leads to C3-symmetric dinitrogen com-
pounds, N2Co(iPr2PNAr)3Zr(THF), the ν(N2) stretches of
2026 cm−1 in 1 (Ar = Mes) and 2045 cm−1 in 4 (Ar = m-xylyl) are
telling indicators of the inherent differences in the electronic
structures of these two species. Steric differences are also
apparent, as N-mesityl substituents lead to significantly more
steric encumbrance around the Zr center than N-m-xylyl
groups.
X-Ray crystallography
All operations were performed on a Bruker-Nonius Kappa
The variation of the phosphinoamide ligand N-donor sub- Apex2 diffractometer, using graphite-monochromated MoKα
stituents results in dramatic differences in the reactivity of radiation. All diffractometer manipulations, including data col-
these complexes in both stoichiometric and catalytic reactions. lection, integration, scaling, and absorption corrections were
The m-xylyl-substituted complex displays a much greater pre- carried out using the Bruker Apex2 software.20 Preliminary cell
ference for radical homocoupling when treated with alkyl constants were obtained from three sets of 12 frames.
halide substrates, leading to diminished yields when used as a Additional crystallographic details and final cif data are
catalyst for Kumada cross-coupling reactions. In general, this included as ESI.†
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1984–1989 | 1987

View Article Online
Paper
Dalton Transactions
Cyclic voltammetry
filtered through a pad of Celite to remove LiCl. The volume of
the filtrate was reduced to 10 mL in vacuo. The resulting super-
saturated solution was layered with pentane and cooled to
−35 °C to yield analytically pure colorless crystals of 2. (Yield:
1.2 g, 71%) 31P NMR (C6D6, 162 MHz): δ 8.07 (s); 1H NMR
(C6D6, 400 MHz): δ 6.94 (s, 6H, o-Ar), 6.47 (s, 3H, p-Ar), 2.52
(m, 6H, CH(CH3)2), 2.22 (s, 18H, Ar–CH3), 1.18 (m, 36H,
CH(CH3)2); 13C NMR (C6D6, 293 K, 100.63 MHz): δ 137.1, 122.6,
121.0, 120.9, 27.2, 21.2, 20.4 (d), 19.5 (d). Anal. Calcd for
C42H69N3P3ZrCl: C, 60.37; H, 8.32; N, 5.03. Found: C, 60.35; H,
8.38; N, 4.93.
Cyclic voltammetry measurements were carried out in a glove-
box under a dinitrogen atmosphere in a one-compartment cell
using a CH Instruments electrochemical analyzer. A glassy
carbon electrode and platinum wire were used as the working
and auxiliary electrodes, respectively. The reference electrode
was Ag/AgNO3 in THF. Solutions (THF) of electrolyte (0.40 M
[nBu4N][PF6]) and analyte (2 mM) were also prepared in the
glovebox.
Computational details
Synthesis of ICo(iPr2PNXyl)2(μ-I)Zr(η2-iPr2PNXyl) (3)
All calculations were performed using Gaussian09, Revision
A.02 for the Linux operating system.21 Density functional
theory calculations were carried out using a combination of
Becke’s 1988 gradient-corrected exchange functional22 and
Perdew’s 1986 electron correlation functional23 (BP86). A
mixed-basis set was employed, using the LANL2TZ(f) triple
zeta basis set with effective core potentials for cobalt and
zirconium,24–26 Gaussian09’s internal 6-311+G(d) for hetero-
atoms (nitrogen, oxygen, phosphorus), and Gaussian09’s
internal LANL2DZ basis set (equivalent to D95V27) for carbon
and hydrogen. Using crystallographically determined geome-
tries as a starting point, the geometries were optimized to a
minimum, followed by analytical frequency calculations to
confirm that no imaginary frequencies were present.
Solid
2 (1.30 g, 1.56 mmol) and solid CoI2 (0.487 g,
1.56 mmol) were combined in CH2Cl2 (30 mL) and stirred for
48 h at room temperature. The resulting bright green solution
was filtered through Celite and solvent was removed from the
filtrate in vacuo. The remaining green solids were washed with
copious amounts of pentane and dried in vacuo to yield ana-
lytically pure 3 as a bright green solid. (Yield: 1.07 g, 67.1%)
1H NMR (400 MHz, C6D6): δ 35.0, 10.54, 8.47, 8.18, 7.49, 3.02,
2.38, 2.13, 1.87, 0.35, −0.82, −8.63. UV-vis (THF) λmax (nm)
(ε [mol−1 cm−1]): 250 (5000), 293 (1850), 370 (850), 450 (500),
510 (350). Evans’ method (C6D6): μeff = 3.10 μB. Anal. Calcd for
C42H69N3P3ZrCoI2: C, 45.33; H, 6.25; N, 3.78. Found: C, 45.29;
H, 6.31; N, 3.72.
Synthesis of N2Co(iPr2PNXyl)3Zr(THF) (4)
Synthesis of N-bis(diisopropylphosphino)-3,5-dimethylaniline,
iPr2PNHXyl
A 0.5% Na/Hg amalgam was prepared from 0.0057 g Na
(0.25 mmol) and 1.1 g Hg. To this vigorously stirred amalgam
in 5 mL of THF, a solution of 3 (106 mg, 0.0993 mmol) in THF
(5 mL) was added. The solution immediately changed color
from green to red. After 4 h, the resulting red solution was fil-
tered away from the amalgam, and the solvent was removed
from the filtrate in vacuo. The resulting solids were extracted
into pentane and filtered through Celite. The concentrated red
pentane solution was cooled to −35 °C, resulting in X-ray
quality red crystals of 4 (yield: 0.0879 g, 61.5%). 31P NMR
(C6D6, 162 MHz): δ 47.48 (very broad s); 1H NMR (C6D6,
400 MHz): δ 6.82 (s, 6H, o-Ar), 6.43 (s, 3H, p-Ar), 3.05 (br m,
4H, THF), 2.58 (m, 6H, CH(CH3)2), 2.10 (s, 18H, Ar–CH3),
1.64–1.38 (m, 36H, CH(CH3)2), 0.79 (br m, 4H, THF); 13C NMR
(C6D6, 100.63 MHz): δ 138.2, 124.1, 122.0, 120.3, 68.0, 27.2,
23.8, 22.2, 21.3, 20.5. UV-vis (THF) λmax (nm) (ε [mol−1 cm−1]):
250 (2590), 290 (640), 365 (155). IR (KBr solution cell, C6H6):
2045 cm−1. Anal. Calcd for C46H77N5P3ZrCoO: C, 57.60; H,
8.09; N, 7.30. Found: C, 57.63; H, 8.16, N, 7.19.
Triethylamine (17 mL, 0.12 mol) and 3,5-dimethylaniline
(7.5 mL, 0.060 mol) were added via syringe to pentane
(150 mL) in a 500 mL flask equipped with a stir bar. At room
temperature, chlorodiisopropylphosphine (9.5 mL, 0.060 mol)
was added dropwise by syringe over the course of 30 min. The
reaction mixture was stirred at room temperature for 15 h, and
filtered through a 1 inch pad of Celite. Volatiles were removed
from the filtrate in vacuo. The resulting residues were extracted
with pentane (3 × 50 mL). The pentane was removed in vacuo
and the resulting solids were recrystallized from 50 mL
pentane at −35 °C. The large colorless crystals were collected
on a glass frit and dried in vacuo. (Yield: 12 g, 86%) 31P NMR
1
(C6D6, 162 MHz): δ 47.1 (s); H NMR (C6D6, 400 MHz): δ 6.67
(s, 2H, o-Ar), 6.39 (s, 1H, p-Ar), 3.32 (d, 1H, N–H), 2.16 (s, 6H,
Ar–CH3), 1.48 (m, 2H, P–CH), 0.97 (m, 12H, CH(CH3)2). Spec-
troscopic data was identical to that reported in the literature.18
Synthesis of (iPr2PNXyl)3ZrCl (2)
A solution of iPr2PNHXyl (1.4 g, 6.0 mmol) in Et2O (50 mL) was
n
Synthesis of BrCo(iPr2PNXyl)2(μ-Br)Zr(η2-iPr2PNXyl) (5)
cooled to −78 °C. To this was added BuLi (4.0 mL, 1.6 M in
hexanes, 6.4 mmol) dropwise over 10 min. The resulting yellow To a solution of 4 (250 mg, 0.26 mmol) in pentane (5 mL),
solution was warmed to room temperature and stirred for 2 h. 1-bromopentane (78 mg, 0.52 mmol) was added. The solution
The solution was then cooled again to −78 °C, and ZrCl4 immediately changed color from red to yellow-green and a
(0.47 g, 2.0 mmol) was added portionwise as a solid. The reac- yellow-green precipitate formed. After 1 h, the precipitate was
tion mixture was warmed to room temperature and stirred for collected via filtration. (Yield: 213 mg, 80.8%) A solution of 5
12 h. Volatiles were removed from the solution in vacuo, and in Et2O–pentane was cooled to −35 °C, resulting in X-ray
the resulting solids were extracted with CH2Cl2 (30 mL) and quality yellow-green crystals of 5. 1H NMR (400 MHz, C6D6):
1988 | Dalton Trans., 2014, 43, 1984–1989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
δ 31.2, 10.60, 8.23, 6.66, 3.28, 2.04, 1.39, 0.42, −1.03, −1.96,
−7.46. UV-vis (THF) λmax (nm) (ε [mol−1 cm−1]): 250 (3600), 295
(660), 360 (180), 450 (60). Evans’ method (C6D6): μeff = 2.95 μB.
Anal. Calcd for C42H69N3P3ZrCoBr2: C, 49.51; H, 6.83; N, 4.12.
Found: C, 49.38; H, 6.93; N, 4.26.
8 D. A. Evers, A. H. Bluestein, B. M. Foxman and
C. M. Thomas, Dalton Trans., 2012, 41, 8111–8115.
9 B. P. Greenwood, G. T. Rowe, C.-H. Chen, B. M. Foxman
and C. M. Thomas, J. Am. Chem. Soc., 2010, 132, 44–45.
10 C. M. Thomas, J. W. Napoline, G. T. Rowe and
B. M. Foxman, Chem. Commun., 2010, 46, 5790–5792.
11 J. P. Krogman, B. M. Foxman and C. M. Thomas, J. Am.
Chem. Soc., 2011, 133, 14582–14585.
Typical procedure for Kumada coupling reactions16
In a nitrogen-filled glovebox, alkyl halide (0.47 mmol) and
TMEDA (17 mg, 0.14 mmol, 30 mol%) were weighed into a
20 mL tube. Dry THF (7 mL) and 3 (0.024 mmol, 5 mol%) were
added and the mixture was stirred for 10 min. To this mixture,
n-OctMgBr (2 M in Et2O, 0.28 mL, 0.56 mmol) was added by
syringe pump at a rate of 2 mL per hour. Upon completion of
this addition, the reaction was removed from the glovebox and
saturated aqueous NH4Cl was added to quench the reaction.
The organic layer was passed through a silica plug and the
resulting mixture was analyzed using GC-MS. Yields were
determined via comparison to an internal standard
(tetradecane).
12 J. W. Napoline, M. W. Bezpalko, B. M. Foxman and
C. M. Thomas, Chem. Commun., 2013, 49, 4388–4390.
13 S. L. Marquard, M. W. Bezpalko, B. M. Foxman and
C. M. Thomas, J. Am. Chem. Soc., 2013, 135, 6018–6021.
14 J. W. Napoline, J. P. Krogman, R. Shi, S. Kuppuswamy,
M. W. Bezpalko, B. M. Foxman and C. M. Thomas,
Eur. J. Inorg. Chem., 2013, 2013, 3874–3882.
15 W. Zhou, S. L. Marquard, M. W. Bezpalko, B. M. Foxman
and C. M. Thomas, Organometallics, 2013, 32, 1766–1772.
16 W. Zhou, J. W. Napoline and C. M. Thomas, Eur. J. Inorg.
Chem., 2011, 2011, 2029–2033.
17 Although aryl C–H activation is still a possibility, this
would generate a more-strained and sterically-congested
product.
18 N. R. Halcovitch and M. D. Fryzuk, Dalton Trans., 2012, 41,
1524–1528.
Acknowledgements
The authors acknowledge the U.S. Department of Energy
(under Award no. DE-SC0004019) and Brandeis University for
funding this work. C. M. T. is also grateful for a 2011 Alfred
P. Sloan Fellowship.
19 B. P. Greenwood, S. I. Forman, G. T. Rowe, C.-H. Chen,
B. M. Foxman and C. M. Thomas, Inorg. Chem., 2009, 48,
6251–6260.
20 Bruker Analytical X-ray Systems, Madison, WI, 2006.
21 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Braone,
B. Mennucci and G. A. Petersson, Gaussian 09, Revision
A. 2, Gaussian, Inc., Wallingford, CT, et al., 2009, (see ESI†
for full reference).
Notes and references
1 L. H. Gade, Angew. Chem., Int. Ed., 2000, 39, 2658–2678.
2 N. Wheatley and P. Kalck, Chem. Rev., 1999, 99, 3379–3420.
3 B. G. Cooper, J. W. Napoline and C. M. Thomas, Catal. Rev. 22 A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38,
Sci. Eng., 2012, 54, 1–40.
3098–3100.
4 C. M. Thomas, Comments Inorg. Chem., 2011, 32, 14–38.
5 V. N. Setty, W. Zhou, B. M. Foxman and C. M. Thomas,
Inorg. Chem., 2011, 50, 4647–4655.
23 J. P. Perdew, Phys. Rev. B: Condens. Matter, 1986, 33, 8822–
8824.
24 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299–310.
6 B. G. Cooper, C. M. Fafard, B. M. Foxman and 25 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
C. M. Thomas, Organometallics, 2010, 29, 5179–5186.
7 S. Kuppuswamy, T. M. Powers, J. P. Krogman,
26 L. E. Roy, P. J. Hay and R. L. Martin, J. Chem. Theor.
Comput., 2008, 4, 1029–1031.
M. W. Bezpalko, B. M. Foxman and C. M. Thomas, Chem. 27 T. H. Dunning and P. J. Hay, in Modern Theoretical Chem-
Sci., 2013, 4, 3557–3565.
istry, Plenum, New York, 1976.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1984–1989 | 1989