Ultrahigh-field NMR spectroscopy of quadrupolar transition metals: 55Mn NMR of several solid manganese carbonyls

Article abstract of DOI:10.1021/ic0608445

⁵⁵Mn NMR spectra acquired at 21.14 T (ν_L( ⁵⁵Mn) = 223.1 MHz) are presented and demonstrate the advantages of using ultrahigh magnetic fields for characterizing the chemical shift tensors of several manganese carbonyls: η⁵-CpMn(CO)₃, Mn ₂(CO)₁₀, and (CO)₅MnMPh₃ (M = Ge, Sn, Pb). For the compounds investigated, the anisotropies of the manganese, chemical shift tensors are less than 250 ppm except for η⁵- CpMn(CO)₃, which has an anisotropy of 920 ppm. At 21.14 T, one can excite the entire m_l = 1/2 ? m_l = -1/2 central transition of η⁵-CpMn(CO)₃, which has a breadth of approximately 700 kHz. The breadth arises from second-order quadrupolar broadening due to the ⁵⁵Mn quadrupolar coupling constant of 64.3 MHz, as well as the anisotropic shielding. Subtle variations in the electric field gradient tensors at the manganese are observed for crystallographically unique sites in two of the solid pentacarbonyls, resulting in measurably different C_Q values. MQMAS experiments are able to distinguish four magnetically unique Mn sites in (CO)₅MnPbPh₃, each with slightly different values of δ_iso, C_Q, and η_Q.

Full text of DOI:10.1021/ic0608445

Inorg. Chem. 2006, 45, 8492−8499
Ultrahigh-Field NMR Spectroscopy of Quadrupolar Transition Metals:
55Mn NMR of Several Solid Manganese Carbonyls
Kristopher J. Ooms,† Kirk W. Feindel,† Victor V. Terskikh,‡ and Roderick E. Wasylishen*,†
Department of Chemistry, UniVersity of Alberta, Edmonton, Alberta T6G 2G2, Canada, and
Department of Chemistry, UniVersity of Ottawa, Ottawa, Ontario K1N 6N5, Canada
Received May 16, 2006
55Mn NMR spectra acquired at 21.14 T (νL
of using ultrahigh magnetic fields for characterizing the chemical shift tensors of several manganese carbonyls:
5-CpMn(CO)3, Mn2(CO)10, and (CO)5MnMPh3 (M
Ge, Sn, Pb). For the compounds investigated, the anisotropies
of the manganese chemical shift tensors are less than 250 ppm except for
5-CpMn(CO)3, which has an anisotropy
of 920 ppm. At 21.14 T, one can excite the entire mI 1/2 mI ) −1/2 central transition of
5-CpMn(CO)3,
( ) 223.1 MHz) are presented and demonstrate the advantages
55Mn)
η
)
η
)
T
η
which has a breadth of approximately 700 kHz. The breadth arises from second-order quadrupolar broadening due
to the 55Mn quadrupolar coupling constant of 64.3 MHz, as well as the anisotropic shielding. Subtle variations in
the electric field gradient tensors at the manganese are observed for crystallographically unique sites in two of the
solid pentacarbonyls, resulting in measurably different CQ values. MQMAS experiments are able to distinguish four
magnetically unique Mn sites in (CO)5MnPbPh3, each with slightly different values of
δiso, CQ, and ηQ.
Introduction
possibility of studying a wide range of inorganic and
organometallic compounds using transition metal NMR is
increasing.7
The latest developments in ultrahigh-field NMR magnets
have resulted in an increased ability to probe the environ-
ments of many transition metal nuclei via solid-state NMR,
SSNMR.1-6 Most NMR-active transition metal nuclei are
quadrupolar with noninteger spin and typically yield broad
NMR powder patterns in the solid state. Due to the inverse
magnetic field dependence of the second-order quadrupolar
interaction, broadening of the mI ) 1/2 T mI ) -1/2 central
transition can be reduced by performing experiments at the
highest applied magnetic fields available. With the advent
of commercially available magnets operating at 21.1 T, the
Manganese NMR spectroscopy has been limited due to
the relatively large quadrupole moment of 55Mn, Q ) 33
fm2,8 which can result in large 55Mn quadrupolar couplings.
55Mn nuclear quadrupolar coupling constants, CQ, have been
measured ranging from 1.58 MHz for KMnO4 to 85.8 MHz
for Ni(1,10-phen)3[(Ph3P)2Mn(CO)4]2.9,10 The larger CQ
values are typically determined by nuclear quadrupole
resonance, NQR, and Mo¨ssbauer spectroscopic investigations
of solid samples or by microwave spectroscopy of molecules
in the gas phase. 55Mn has a spin I ) 5/2, a natural abundance
of 100%, a Larmor frequency close to that of 13C,11 and a
chemical shift range of approximately 3500 ppm. A survey
of the 55Mn NMR literature indicates that the manganese of
* To whom correspondence should be addressed. E-mail:
roderick.wasylishen@ualberta.ca. Tel: (780) 492-4336. Fax: (780) 492-
8231.
† University of Alberta.
+
‡ University of Ottawa.
Mn(CO)3(MeCN)3 is the least shielded, δiso ) 490 ppm,
(1) Ooms, K. J.; Wasylishen, R. E. J. Am. Chem. Soc. 2004, 126, 10972-
while the manganese of MnH(PF3)5 is the most shielded,
10980.
-
δiso ) -2953 ppm, relative to MnO4
.
12,13 Early reviews
(2) Bryce, D. L.; Wasylishen, R. E. Phys. Chem. Chem. Phys. 2002, 4,
3591-3600.
(aq)
(3) Bryce, D. L.; Wasylishen, R. E. Phys. Chem. Chem. Phys. 2001, 3,
5154-5157.
(7) Can. Chem. News 2006, April, 5.
(8) Pyykko¨, P. Mol. Phys. 2001, 99, 1617-1629.
(9) Wadsworth, M.; France, P. W. J. Magn. Reson. 1983, 51, 424-429.
(10) Pribula, C. D.; Brown, T. L.; Mu¨nck, E. J. Am. Chem. Soc. 1974, 96,
4149-4154.
(11) Harris, R. K.; Becker, E. D.; De Menezes, S. M. C.; Goodfellow, R.;
Granger, P. Pure Appl. Chem. 2001, 73, 1795-1818.
(12) Rehder, D.; Bechthold, H.-Ch.; Kec¸eci, A.; Schmidt, H.; Siewing, M.
Z. Naturforsch. 1982, 37b, 631-645.
(4) Lipton, A. S.; Heck, R. W.; Ellis, P. D. J. Am. Chem. Soc. 2004, 126,
4735-4739.
(5) De Lacaillerie, J. B. D.; Barberon, F.; Romanenko, K. V.; Lapina, O.
B.; Le Polles, L.; Gautier, R.; Gan, Z. H. J. Phys. Chem. B 2005,
109, 14033-14042.
(6) Larsen, F. H.; Farnan, I.; Lipton, A. S. J. Magn. Reson. 2006, 178,
228-236.
8492 Inorganic Chemistry, Vol. 45, No. 21, 2006
10.1021/ic0608445 CCC: $33.50
© 2006 American Chemical Society
Published on Web 07/27/2006

Ultrahigh-Field NMR Spectroscopy of Transition Metals
55Mn NMR experiments were performed on Bruker Avance
spectrometers at 7.05 (74.5 MHz), 11.75 (124.0 MHz), and 21.14
T (223.1 MHz). The experiments at 21.14 T were performed at the
National Ultrahigh-field NMR Facility for Solids in Ottawa, Canada;
experiments at 11.75 and 7.05 T were performed at the University
of Alberta. Bruker double- and triple-resonance magic angle
spinning, MAS, probes (4, 3.2, and 2.5 mm o.d.) were used. All
of the 55Mn NMR literature have been presented by Kidd
and Goodfellow,14 as well as Rehder;15 more recent reviews
have been published by Granger16 and Pregosin.17 There have
been only a few 55Mn NMR studies of compounds in the
solid state, most of them performed on powders of perman-
ganate salts or single crystals.9,18-24 Particularly significant
are the early 55Mn single-crystal NMR investigations of
Sheline and co-workers,23,24 which provided evidence that
the anisotropy of the manganese shielding could be as large
as 1000 ppm.
Manganese chemistry is a growing area of research, owing
to the use of Mn in organic and inorganic synthesis;25
manganese is a relatively inexpensive transition metal that
can be used for a variety of oxidation, substitution, and
catalytic reactions.25 Additionally, methylcyclopentadieneyl
manganese tricarbonyl, MeCpMn(CO)3, is finding increased
use as an anti-knock fuel additive throughout North America.26
The bonding environment about manganese in organoman-
ganese compounds often dictates their reactivity. Since
SSNMR is an ideal technique for probing the subtle
variations in the electronic-molecular structure about a
nucleus, the ability to perform SSNMR of 55Mn compounds
cannot be overemphasized.
55Mn NMR chemical shifts were referenced with respect to a 0.82
55
m aqueous solution of KMnO4, δ ) (νsample - νref)/ νref.11 The
-
Mn π/2 pulse lengths were determined for the setup sample, 0.82
m KMnO4, and ranged from 1.6 to 2.5 µs depending on the
spectrometer used. An echo (π/2 - τ1 - π/2 - τ2 - ACQ or π/2
- τ1 - π - τ2 - ACQ) pulse sequence was used for acquiring
55Mn NMR spectra with pulse lengths, τp(sel), that selectively excited
the central transition, τp(sel) ) τp(non-sel)/(I + 1/2) ) τp(non-sel)/3; in
some cases, the pulse widths were optimized on the samples to
reduce line shape distortions.29 The quadrupolar Carr-Purcell
Meiboom-Gill, QCPMG,30,31 experiment was used to acquire
stepped-frequency spectra which were co-added using the sky-line
projection method.32 A 3-pulse sequence with z-filter was used to
acquire 3Q MAS spectra of (CO)5MnPbPh3; p1 - t1(3Q) - p2 -
t(ZQF) - p3 - ACQ.33 Excitation (p1 ) 1.95 µs) and conversion
(p2 ) 0.65 µs) pulses were applied with an RF field amplitude of
130 kHz; the weak central-transition selective pulse was p3 ) 17
µs, corresponding to an RF field of 5 kHz. The sample spinning
speed was 15 kHz using a 2.5 mm Bruker MAS probe. The
acquisition matrix was 2048(t2) × 128(t1) with 192 scans ac-
cumulated for each t1. Shearing transformation was performed
during data processing. The 1D spectra used for line shape analysis
were extracted from the 2D spectrum and simulated with WSolids,
a program developed in the Wasylishen laboratory.
In the present study, we demonstrate the advantages of
acquiring 55Mn NMR spectra at 21.14 T and illustrate some
of the valuable information that can be obtained at this field
strength. We have obtained 55Mn NMR spectra of five solid
diamagnetic compounds, η5-CpMn(CO)3, Mn2(CO)10, and
(CO)5MnMPh3 (M ) Ge, Sn, Pb) and have fully character-
ized both their 55Mn EFG and chemical shift tensors. The
possibility of resolving crystallographically nonequivalent
Mn sites has also been illustrated for two of the compounds.
EFG34 and magnetic shielding calculations of η5-CpMn(CO)3
were performed with the relativistic ZORA-DFT formalism in the
Amsterdam Density Functional (ADF 2004.01) package.35-41 The
ZORA valence triple-ú doubly polarized, TZ2P, basis set was used
for all nuclei. The local density approximation of VWN42 and
generalized gradient approximation of Becke8843 and Perdew8644,45
Experimental Section
η5-CpMn(CO)3 and Mn2(CO)10 were purchased from Aldrich and
used without further purification. The (CO)5MnMPh3 (M ) Ge,
Sn, Pb) compounds were prepared under nitrogen by reacting
NaMn(CO)5 with the appropriate ClMPh3 and subsequently recrys-
talized from hexane in the air.27,28
(28) NaMn(CO)5 was prepared by addition of Mn2(CO)10 to a Na amalgam
(0.5 g Na in 3 mL Hg) in dry THF. The NaMn(CO)5 solution was
then transferred to a flask containing the appropriate ClMPh3 and
allowed to react for 2 h. The solvent was reduced to half volume,
poured into cold water, and filtered, and then the precipitate was
extracted with hot hexane (yields 70-80%).
(29) Bodart, P. R.; Amoureux, J.-P.; Dumazy, Y.; Lefort, R. Mol. Phys.
2000, 98, 1545-1551.
(13) Miles, W. J.; Garrett, B. B.; Clark, R. J. Inorg. Chem. 1969, 8, 2817-
2818.
(14) Kidd, R. G.; Goodfellow, R. J. In NMR and the Periodic Table; Harris,
R. K., Mann, B. E., Eds.; Academic Press: London, 1979; pp 218-
222.
(15) Rehder, D. In Multinuclear NMR; Mason, J., Ed.; Plenum: New York,
1989; pp 507-511.
(16) Granger, P. In Encyclopedia of NMR; Grant, D. M., Harris, R. K.,
Eds.; John Wiley and Sons: Chichester, 2002; Vol. 6, pp 3889-3900.
(17) Pregosin, P. S. Transition Metal Nuclear Magnetic Resonance;
Elsevier: Amsterdam, 1991.
(30) Larsen, F. H.; Jakobsen, H. J.; Ellis, P. D.; Nielsen, N. C. J. Magn.
Reson. 1998, 131, 144-147.
(31) Larsen, F. H.; Jakobsen, H. J.; Ellis, P. D.; Nielsen, N. C. J. Phys.
Chem. A 1997, 101, 8597-8606.
(32) Lipton, A. S.; Wright, T. A.; Bowman, M. K.; Reger, D. L.; Ellis, P.
D. J. Am. Chem. Soc. 2002, 124, 5850-5860.
(33) Amoureux, J.-P.; Fernandez, C.; Steuernagel, S. J. Magn. Reson. A
1996, 123, 116-118.
(34) van Lenthe E.; Jan Baerends, E. J. Chem. Phys. 2000, 112, 8279-
8291.
(18) Segel, S. L. J. Chem. Phys. 1969, 51, 848-850.
(19) Mooberry, E. S.; Sheline, R. K. J. Chem. Phys. 1972, 56, 1852-1855.
(20) Mooberry, E. S.; Spiess, H. W.; Sheline, R. K. J. Chem. Phys. 1972,
57, 813-821.
(35) Schreckenbach, G.; Ziegler, T. J. Phys. Chem. 1995, 99, 606-611.
(36) Schreckenbach, G.; Ziegler, T. Int. J. Quantum Chem. 1996, 60, 753-
766.
(37) Schreckenbach, G.; Ziegler, T. Int. J. Quantum Chem. 1997, 61, 899-
(21) Burton, D. J.; Harris, R. K. Chem. Commun. 1982, 256-257.
(22) Tarasov, V. P.; Kirkakosyan, G. A.; Meladze, M. A.; German, K. EÄ .
Koordinats. Khim. 1993, 19, 269-283.
918.
(38) Wolff, S. K.; Ziegler, T. J. Chem. Phys. 1998, 109, 895-905.
(39) Wolff, S. K.; Ziegler, T.; van Lenthe, E.; Baerends, E. J. J. Chem.
Phys. 1999, 110 7689-7698.
(40) Guerra, C. F.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391-403.
(23) Spiess, H. W.; Sheline, R. K. J. Chem. Phys. 1971, 54, 1099-1103.
(24) Slater, J. L.; Pupp, M.; Sheline, R. K. J. Chem. Phys. 1972, 57, 2105-
2108.
(25) Sweigart, D. A.; Reingold, J. A. In Encyclopedia of Inorganic
Chemistry; King, R. B., Ed.; Wiley: Chichester, 2005; Vol. 5, p 2907.
(26) Toxicological Profile for Manganese; U.S. Department of Health and
Human Services: Washington, DC, 2000; p 329.
(41) ADF 2004.01, SCM [online], Theoretical Chemistry; Vrije Univer-
siteit: Amsterdam, Availaible from http://www.scm.com.
(42) Vosko, S. H.; Wilk, L.; Nusair, M.; Can. J. Phys. 1980, 58, 1200-
1211.
(27) Gorsich, R. D. J. Am. Chem. Soc. 1962, 84, 2486-2491.
(43) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100.
Inorganic Chemistry, Vol. 45, No. 21, 2006 8493

Ooms et al.
Table 1. Solid-State 55Mn NMR Parameters for the Five Compounds Studieda
CQ/MHz
ηQ
δ
iso/ppm
Ω/ppm
κ
R/°
â/°
γ/°
η5-CpMn(CO)3
calculated
Mn2(CO)10
64.3 ( 0.5
66.3
0.02 ( 0.02
0.01
-2200 ( 10
-2155
920 ( 10
567
0.95 ( 0.05
0.80
0 ( 10
0
0 ( 10
0
0 ( 10
0
3.28 ( 0.05
24.6 ( 0.2
23.4 ( 0.2
18.6 ( 0.1
11.9 ( 0.1
12.0 ( 0.1
10.7 ( 0.1
10.5 ( 0.1
0.35 ( 0.02
0.36 ( 0.05
0.30 ( 0.05
0.08 ( 0.02
0.05 ( 0.05
0.05 ( 0.05
0.10 ( 0.05
0.10 ( 0.05
-2288 ( 2
-2468 ( 3
-2465 ( 3
-2535 ( 5
-2395 ( 5
-2390 ( 5
-2381 ( 5
-2376 ( 5
105 ( 10
220 ( 20
220 ( 20
215 ( 10
100 ( 10
100 ( 10
90 ( 10
90 ( 10
0.95 ( 0.05
-0.9 ( 0.1
-0.9 ( 0.1
-0.95 ( 0.05
-0.7 ( 0.1
-0.7 ( 0.1
-0.7 ( 0.1
-0.7 ( 0.1
90 ( 10
0 ( 10
0 ( 10
40 ( 10
0 ( 10
0 ( 10
0 ( 10
0 ( 10
90 ( 5
90 ( 5
90 ( 5
87 ( 1
70 ( 5
70 ( 5
74 ( 5
74 ( 5
0 ( 10
0 ( 10
0 ( 10
0 ( 10
0 ( 10
0 ( 10
0 ( 10
0 ( 10
(CO)5MnGePh3
(CO)5MnSnPh3
(CO)5MnPbPh3
a The chemical shift parameters are defined as follows. δ33 eδ22 eδ11, δiso ) (δ11 + δ22 + δ33)/3, Ω ) δ11 - δ33, κ ) 3(δ22 - δiso)/Ω.66 The components
of the chemical shift tensor are δii ) (νii - νref)/νref (ii ) 11, 22, 33). The EFG parameters are CQ ) eQVZZ/h and ηQ ) (VXX - VYY)/VZZ where |VZZ| g |VYY
with respect to the EFG tensor.67
|
g |VXX|, e is the electron charge, and h is Planck’s constant. R, â and γ are the Euler angles that describe the relative orientation of the chemical shift tensor
were employed. The default accuracy of the integration grid was
modified to better describe the core orbitals, increasing accint to 5
and accsph to 6. Magnetic shielding values were referenced to the
-
calculated isotropic shielding value of an isolated tetrahedral MnO4
ion (Mn-O bond length ) 1.602 Å), δ ) (σref - σsample)/(1 -
σref). Recent computation studies by Bu¨hl indicate that the
manganese shielding of an isolated permanganate ion (equilibrium
structure) becomes deshielded by 35 ppm on dissolution in water.46
Results and Discussion
1. Analysis of 55Mn NMR Spectra. The 55Mn NMR
parameters obtained from the best-fit simulations of the NMR
spectra acquired for η5-CpMn(CO)3, Mn2(CO)10, and three
group-14 substituted manganese pentacarbonyls are presented
in Table 1. The CQ values range from 3.28 MHz for Mn2-
(CO)10 to 64.3 MHz for η5-CpMn(CO)3. The spans, Ω, of
the chemical shift tensors range from 105 to 920 ppm for
Mn2(CO)10 and η5-CpMn(CO)3, respectively. Where possible,
the EFG parameters available from early single-crystal 55Mn
NMR studies were used as a starting point for the spectral
simulations.
1.1. η5-CpMn(CO)3. The 55Mn NMR spectra (central
Figure 1. Central transition 55Mn NMR spectra of stationary samples of
transition) obtained of a powder sample of η5-CpMn(CO)3
η5-CpMn(CO)3 acquired at 7.05, 11.75, and 21.14 T. Spectra simulated
are shown in Figure 1. The spectrum acquired at 21.14 T
using the parameters contained in Table 1 are presented above experimental
was collected with a single transmitter offset. The spectrum
is ca. 700 kHz broad and is dominated by a quadrupolar
coupling constant of 64.3 MHz. For comparison, the 55Mn
NMR spectra obtained at 7.05 and 11.75 T are also presented.
The QCPMG spectrum at 7.05 T was acquired using 25
different, evenly spaced, steps of the transmitter while the
QCPMG, and echo spectra acquired at 11.75 T were collected
in 12 and 6 steps, respectively.32 The spectrum acquired at
21.14 T shows all the features necessary to accurately
determine both the 55Mn EFG and chemical shift parameters.
When both an EFG and anisotropic chemical shift are
present, the line shape of the central transition spectrum of
a half-integer quadrupolar nucleus can be used to obtain
information about the EFG and chemical shift tensors. If a
rotation axis (Cn, n g 3) passes through the nucleus of
interest, the EFG and chemical shift tensors must be axially
symmetric. In the case of η5-CpMn(CO)3, X-ray diffraction
spectra. The spectra acquired at 7.05 and 11.75 T were acquired using several
different transmitter offsets, while the spectrum at 21.14 T was collected
with a single transmitter offset.
indicates the molecule possesses C1 symmetry;47 however,
the deviations from D5h and C3V symmetries in the Cp ring
and the Mn(CO)3 fragments, respectively, are negligible.48
Furthermore, the Cp ring is thought to have a low barrier to
internal rotation. From consideration of this information
together with the NMR data, it is clear that the EFG tensor
is axially symmetric with the unique component, VZZ, along
the approximate C3 axis, Figure 2. This leaves two possible
orientations for the 55Mn chemical shift tensor; either the
skew, κ ) 1 and δ33 is parallel to VZZ, Figure 2e, or κ ) -1
and δ11 is parallel to VZZ, Figure 2f. Clearly, the experimental
spectrum obtained at 21.14 T is only consistent with the first
possibility.
(47) Fitzpatrick, P. J.; Le Page, Y.; Sedman, J.; Butler, I. S. Inorg Chem.
(44) Perdew, J. P. Phys. ReV. B 1986, 33, 8822-8824.
(45) Perdew, J. P. Phys. ReV. B 1986, 34, 7406.
(46) Bu¨hl M. J. Phys. Chem. A 2002, 106, 10505-10509.
1981, 20, 2852-2861.
(48) Full, J.; Gonza´lez, L.; Daniel, C. J. Phys. Chem. A 2001, 105, 184-
189.
8494 Inorganic Chemistry, Vol. 45, No. 21, 2006

Ultrahigh-Field NMR Spectroscopy of Transition Metals
Figure 3. The simplified orbital diagram for η5-CpMn(CO)3. The orbitals
are derived from the Mn(CO)3+ fragment as taken from ref 49. Below is a
section taken from Table 3 in ref 52 which indicates the effects that the
angular momentum operators, first column, have on the occupied orbitals,
first row.
Figure 2. Calculated 55Mn NMR spectra showing the interplay of the
EFG and chemical shift parameters on the powder line shape. (a) CQ ) 0
MHz, κ ) 1, Ω ) 920 ppm; (b) CQ ) 0 MHz, κ ) -1, Ω ) 920 ppm;
(c,d) CQ ) 64.3 MHz, Ω ) 0 ppm; (e) CQ ) 64.3 MHz, κ ) 1, Ω ) 920
ppm; (f) CQ ) 64.3 MHz, κ ) -1, Ω ) 920 ppm. For all simulations η )
0 and the unique components of the two tensors are coincident. The angles
indicate the orientation of the unique component of the tensors with respect
to B0. The dashed line shows the effect that Ω has on the position of the
shoulder in the center of the axially symmetric powder pattern.
molecular orbitals, MOs.51,52 Qualitatively, the magnitude of
the paramagnetic term is dependent both on the energy
difference between the two orbitals and the matrix elements
of the form ævirt|Lˆi|æocc , where ævirt and æocc are the virtual
and occupied orbital wave functions and Lˆi is the angular
momentum operator acting along i.51,52 On the basis of the
simple MO model proposed by Lichtenberger and Fenske
for η5-CpMn(CO)3, the highest occupied molecular orbitals
are localized on the metal and, to a good approximation,
The position of the small shoulder near the center of the
axially symmetric pattern strongly depends on the magnitude
of the chemical shift anisotropy; this is indicated by the
dashed line in Figure 2c and e for Ω ) 0 and 920 ppm,
respectively. The shoulder is not observable in the spectra
acquired at lower fields due to the stepwise acquisition
method used and lower signal-to-noise. For many quadru-
polar nuclei, large nuclear quadrupolar couplings typically
limit solid-state NMR investigations. The ability to acquire
NMR spectra of spin-5/2 nuclei with CQ values in excess of
50 MHz in a straightforward, single offset experiment opens
the door to characterizing EFG tensors in molecules with
55Mn and other noninteger spin transition metal nuclei with
moderate quadruple moments.
The values of CQ and ηQ obtained from simulation of the
powder line shapes, Table 1, are in agreement with an earlier
single-crystal 55Mn NMR study of Spiess and Sheline which
found CQ and ηQ to be 64.275 MHz and 0.0134, respec-
tively.23 The Ω and κ of the chemical shift tensor are
determined to be 920 ppm and 0.95, respectively. As already
indicated, the orientation of the two tensors is such that VZZ
and the unique component of the chemical shift tensor, δ33,
are along the pseudorotation axis of the molecule, Figure 2.
Large spans have been previously reported for other “piano
stool” compounds. For example, the molybdenum chemical
shift anisotropy for mesitylenetricarbonylmolybdenum is 775
ppm with κ ) 0.87.2 The origin of the large metal shielding
anisotropy in transition metal “piano stool” compounds lies
in the availability of symmetry appropriate virtual d orbit-
als.49,50
2
2
2
may be described as dz , dx -y , and dxy orbitals, as illustrated
in Figure 3 (the molecule is oriented such that the pseu-
dorotation axis is along the z axis).49 The lowest-lying virtual
orbitals are of dxz and dyz character.48,53 The assumption that
the Mn MOs in η5-CpMn(CO)3 can be described by the
atomic d orbitals is clearly an approximation; nevertheless,
the symmetry of the metal d orbitals provides a qualitative
understanding of the manganese shielding tensor. The matrix
in Figure 3 shows the result of using the angular momentum
operators (left column) on the occupied MOs (top row). By
applying the magnetic field in the xy plane, mixing can occur
between the occupied MOs and the virtual MOs; for example,
applying Lˆy to æocc ) dxy generates a dxz orbital which can
mix with ævirt ) dxz.52 Since there is a relatively small energy
gap between these orbitals, the paramagnetic term will be
substantial,48 and the result will be an overall deshielding of
the Mn nucleus in the xy plane, i.e., σ11 and σ22. When the
magnetic field is applied along the z direction, Lˆz operating
on the three highest occupied orbitals does not result in
efficient mixing with either ævirt, and therefore, the para-
magnetic contribution along z, i.e., σ33, is small.2,52
The 55Mn CQ value in η5-CpMn(CO)3 is much larger than
the 95Mo CQ in mesitylenetricarbonylmolybdenum, 0.96
MHz, despite their similar structures. Simply scaling the 95Mo
CQ by the ratio of Q(55Mn)/Q(95Mo), a 55Mn CQ of 14.4 MHz
would be expected, significantly smaller than that observed
for η5-CpMn(CO)3. The small 95Mo CQ value for mesityl-
enetricarbonylmolybdenum was rationalized on the basis of
the geometry of the molecule, in particular implying that
The paramagnetic contribution to shielding, which is
normally negative, is related to the magnetic-dipole-allowed
mixing between symmetry appropriate occupied and virtual
(51) Ramsey, N. F. Phys. ReV. 1950, 78, 699-703.
(52) Jameson, C. J.; Gutowsky, H. A. J. Chem. Phys. 1964, 40, 1714-
1724.
(49) Lichtenberger, D. L.; Fenske, R. F. J. Am. Chem. Soc. 1976, 98, 50-
(53) Daniel, C.; Full, J.; Gonza´lez, L.; Kaposta, C.; Krenz, M.; Lupulescu,
C.; Manz, J.; Minemoto, S.; Oppel, M.; Rosendo-Francisco, P.; Vajda,
Sˇ.; Wo¨ste, L. Chem. Phys. 2001, 267, 247-260.
63.
(50) Albright, T. A. Acc. Chem. Res. 1982, 15, 149-55.
Inorganic Chemistry, Vol. 45, No. 21, 2006 8495

Ooms et al.
Figure 5. 55Mn NMR spectra of stationary samples of (CO)5MnGePh3
acquired at 7.05, 11.75, and 21.14 T. Spectra simulated using the parameters
contained in Table 1 are presented above experimental spectra. The top
spectrum at 21.14 T is the calculated spectrum without including chemical
shift anisotropy (i.e., Ω ) 0 ppm). The inset shows the structure of
(CO)5MnGePh3 and the orientation of the unique components of the 55Mn
EFG and chemical shift tensors.
Figure 4. 55Mn NMR spectra of a stationary sample of Mn2(CO)10 acquired
at 21.14 T. (a) Central transition and (b) entire 55Mn NMR spectrum
including central and satellite transitions. The central transition spectrum
is dominated by the chemical shift anisotropy at the ultrahigh field. Upper
spectra were simulated using the parameters listed in Table 1, while the
bottom spectrum in (a) is the calculated spectrum without including 55Mn
chemical shift anisotropy (i.e., Ω ) 0 ppm). The inset shows the structure
of Mn2(CO)10 and the orientation of the unique components of the 55Mn
EFG and chemical shift tensors.
well to the value calculated on the basis of the Mn-Mn bond
length, 2.9042 Å, determined from X-ray diffraction.56 The
simulations also included the effects of 1J(55Mn,55Mn) which
has been reported as 65 Hz; this coupling constant had a
negligible effect on the calculated spectra.58 Figure 4b shows
the entire 55Mn NMR spectrum (central and satellite transi-
tions) of a stationary sample of Mn2(CO)10. The spectrum is
approximately 2.0 MHz broad and therefore could not be
completely excited using a single transmitter offset. Despite
the distortions, the discontinuities were found to be sensitive
to both the EFG and chemical shift parameters. The CQ and
ηQ values determined are consistent with previous measure-
ments.18,20,58
The manganese chemical shift tensor has not been previ-
ously characterized for Mn2(CO)10. Due to the presence of
the quadrupolar, dipolar, and indirect spin-spin coupling,
accurately determining the chemical shift tensor at lower
applied magnetic field strengths is difficult; for example, the
spectrum acquired at 7.05 T for a stationary sample is a
relatively featureless peak.59 The VZZ component of the EFG
tensor is oriented 90° from the Mn-Mn dipolar vector, in
agreement with the early single-crystal 55Mn NMR study.20
On the basis of the Euler angles that describe the relative
orientation of the EFG and chemical shift tensors, the unique
component of the chemical shift tensor is along the Mn-
Mn bond, Figure 4. The orientations of the tensors are
discussed further in a later section.
since the angles between the Mo-CO bond and the pseu-
dorotation axis (θ ) 53.9°, 52.8°, and 54.3°) were close to
the magic angle, 54.7356°, a small EFG would be ex-
pected.2,54 For η5-CpMn(CO)3, θ is also close to the magic
angle, 56.1°, 56.0°, and 56.3°, and yet a large CQ is observed.
Clearly factors other than θ must be responsible for the
significant EFG at Mn.
We have used the atomic coordinates available from X-ray
crystallography to perform quantum chemical calculations
of the Mn shielding and EFG tensors.47 The results presented
in Table 1 are in good agreement with experiment given that
some differences may result from the calculation being
performed on an isolated molecule.55 The agreement with
experiment is a strong indicator that in the case of η5-CpMn-
(CO)3 both the EFG at Mn and the chemical shift parameters
are predominantly determined by the local molecular struc-
ture.
1.2. Mn2(CO)10. The 55Mn NMR spectrum of the central
transition, mI ) 1/2 T mI ) -1/2, of a stationary sample of
Mn2(CO)10 acquired at 21.14 T is completely dominated by
anisotropic shielding, Figure 4a. The single-crystal X-ray
diffraction structure of Mn2(CO)10 reveals that there is a C2
axis through the midpoint of the Mn-Mn bond, resulting in
only one unique manganese site.56 When simulating the 55Mn
NMR spectra acquired at 11.75 T (not shown), a homo-
nuclear dipolar coupling of 305 Hz was required; the dipolar
coupling did not significantly influence the spectrum acquired
at 21.14 T.57 The dipolar coupling of 305 Hz corresponds
1.3.(CO)5MnGePh3.Thecrystalstructureof(CO)5MnGePh3
has not been reported, but previous single-crystal 55Mn NMR
studies indicate that there are two crystallographically unique
Mn sites.24 The presence of two sites was confirmed by MAS
measurements (spinning rate, 35 kHz) at 21.14 T. The 55Mn
NMR spectra of stationary samples acquired at the three
fields are presented in Figure 5. The spectra acquired at 7.05
and 11.75 T could be fit without including chemical shift
anisotropy. The quadrupolar coupling parameters determined
from the stationary sample spectra, Table 1, are in good
(54) Akitt, J. W.; McDonald, W. S. J. Magn. Reson. 1984, 58, 401-412.
(55) Ooms, K. J.; Feindel, K. W.; Willans, M. J.; Wasylishen, R. E.; Hanna,
J. V.; Pike, K. J.; Smith, M. E. Solid State Nucl. Magn. Reson. 2005,
28, 125-134.
(56) Churchill, M. R.; Amoh, K. N.; Wasserman, H. J. Inorg. Chem. 1981,
20, 1609-1611.
(57) The simulation program does not properly treat the residual dipolar
coupling to the neighboring quadrupolar nucleus which arises from
the second-order quadrupolar perturbation on the neighboring Mn.
Given that the CQ is small, this is not likely to significantly affect the
simulated line shapes.
(58) Wi, S.; Frydman, L. J. Chem. Phys. 2000, 112, 3248-3261.
(59) Siegel, R.; Nakashima, T. T.; Wasylishen, R. E. Chem. Phys. Lett.
2006, 421, 529-533.
8496 Inorganic Chemistry, Vol. 45, No. 21, 2006

Ultrahigh-Field NMR Spectroscopy of Transition Metals
Figure 7. 55Mn NMR spectra of stationary samples of (CO)5MnPbPh3
acquired at 11.75 and 21.14 T. Spectra simulated using the parameters
contained in Table 1 are presented above experimental spectra. The top
spectrum at 21.14 T is the calculated spectrum without including chemical
shift anisotropy (i.e., Ω ) 0 ppm). The inset shows the structure of
(CO)5MnPbPh3 and the orientation of VZZ and δ11.
Figure 6. 55Mn NMR spectra of stationary samples of (CO)5MnSnPh3
acquired at 7.05, 11.75, and 21.14 T. Spectra simulated using the parameters
contained in Table 1 are presented above experimental spectra. The top
spectrum at 21.14 T is the calculated spectrum without including chemical
shift anisotropy (i.e., Ω ) 0 ppm). The inset shows the structure of
(CO)5MnSnPh3 and the orientation of the unique components of the 55Mn
EFG and chemical shift tensors.
agreement with the previous single-crystal 55Mn NMR
results.24 The spectrum of a stationary sample acquired at
21.14 T requires the inclusion of chemical shift anisotropy
and reveals Ω ) 220 ppm and κ ) -0.95 ( 0.05.
1.4. (CO)5MnSnPh3. The 55Mn NMR line shapes of
stationary samples of (CO)5MnSnPh3 show dramatic differ-
ences at the different applied magnetic field strengths, Figure
6. The crystal structure determined by X-ray diffraction
indicates that there are four unique molecules in the unit
cell; however, it is not clear which solvent was used for the
crystallization.60 The 55Mn NMR spectra of stationary
samples, shown in Figure 6, and of an MAS sample spinning
at 30 kHz, not shown, and previous single-crystal 55Mn NMR
results all indicate that there is only one Mn site;24 hexane
was used to recrystallize (CO)5MnSnPh3 in this study. This
indicates that there are different polymorphs of (CO)5MnSnPh3;
polymorphism has previously been noted for (CO)5MnPbPh3.61
The quadrupolar parameters determined here, Table 1, are
in good agreement with the previous single-crystal 55Mn
NMR study.24 Accurately characterizing the chemical shift
tensor is difficult based on the spectra acquired at 7.05 and
11.75 T; however, at 21.14 T, the experimental NMR
spectrum cannot be fit without including a Ω of 215 ppm.
1.5.(CO)5MnPbPh3.Thecrystalstructureof(CO)5MnPbPh3
is unknown; therefore, any insight regarding the structure
of (CO)5MnPbPh3 is valuable. The breadth of the 55Mn NMR
spectra of stationary samples of (CO)5MnPbPh3, Figure 7,
reveal that the 55Mn CQ values are smaller than for either
the analogous Sn or Ge compounds. The 55Mn NMR
spectrum of the MAS sample could not be fit with the two
sites determined in an earlier single-crystal NMR study.24
To determine the CQ parameters more accurately, a 3Q-MAS
spectrum was acquired at 21.14 T, Figure 8. The high applied
magnetic field strength enabled four sites with slightly
different values for δiso, CQ, and ηQ to be distinguished, Table
1. Acquisition of this spectrum would be difficult at lower
applied magnetic field strengths, as the large quadrupolar
broadening would require much higher spinning rates and
Figure 8. 55Mn NMR 3Q MAS spectra of (CO)5MnPbPh3 acquired at
21.14 T, with νrot ) 15 kHz. On the left are the four anisotropic slices
taken through the four sites; simulated spectra are presented above
experimental spectra. The top spectrum on the left is the spectrum collected
under MAS conditions at 21.14 T.
the four sites may not be adequately resolved. The presence
of four sites indicates that a different polymorph of
(CO)5MnPbPh3 is obtained when recrystallized from hexane
than was previously obtained from methanol. In general,
when performing transition metal SSNMR on powder
samples, the presence of multiple crystallographic sites can
make accurate determination of the NMR parameters from
one-dimensional spectra difficult or impossible; however, the
MQMAS experiment allows for the separation of the
different sites into a second dimension.62 The ability to
differentiate sites with small differences in δiso and CQ at
ultrahigh-field strengths, as demonstrated here, is an impor-
tant step to applying SSNMR to the study of more complex
transition metal complexes.
2. Electric Field Gradients at Mn in (CO)5MnY
Compounds. On the basis of a point charge analysis of the
EFG in octahedral transition metal compounds, MX5Y
compounds with ideal geometries, i.e., equal bond lengths
and angles, should have a substantial, axially symmetric EFG
at the metal center.63 The magnitude of the EFG at M depends
on differences in the electronic properties of X and Y.
(60) Weber, H. P.; Bryan, R. F. Acta Crystallogr. 1967, 22, 822-836.
(61) Christendat, D.; Butler, I. S.; Gilson, F. R.; Morin, D. F. G. Can. J.
Chem. 1999, 77, 1892-1898.
(62) Medek, A.; Harwood: J. S.; Frydman, L. J. Am. Chem. Soc. 1995,
117, 12779-12787.
(63) Han, O. H.; Oldfield, E. Inorg. Chem. 1990, 29, 3666-3669.
Inorganic Chemistry, Vol. 45, No. 21, 2006 8497

Ooms et al.
Table 2. Principal Components of the 55Mn Chemical Shift Tensors for the Compounds Studied
δ11/ppm
δ22/ppm
δ33/ppm
δiso
Ω/ppm
η5-CpMn(CO)3
Mn2(CO)10
(CO)5MnGePh3
-1885 ( 10
-2252 ( 5
-2325 ( 10
-2322 ( 10
-2393 ( 5
-2333 ( 5
-2325 ( 5
-1908 ( 10
-2255 ( 5
-2534 ( 10
-2531 ( 10
-2603 ( 10
-2418 ( 5
-2402 ( 5
-2805 ( 5
-2357 ( 5
-2545 ( 10
-2542 ( 10
-2608 ( 5
-2433 ( 5
-2416 ( 5
-2200 ( 10
-2288 ( 2
-2468 ( 3
-2465 ( 3
-2535 ( 5
-2395 ( 5
-2381 ( 5
920 ( 10
105 ( 10
220 ( 20
220 ( 20
215 ( 10
100 ( 10
90 ( 10
(CO)5MnSnPh3
(CO)5MnPbPh3
a
a For (CO)5MnPbPh3, the differences in the Eulers angles, Ω, κ between sites 1 and 2, as well as 3 and 4, were not observable.
Deviations from ideal octahedral geometry, either in the local
molecular structure or the long-range crystal packing, can
also play a role in determining the magnitude and orientation
of the principal components of the EFG in the molecular
frame.
For (CO)5MnSnPh3 and (CO)5MnPbPh3, the EFG tensors
are nearly axially symmetric, ηQ e 0.10. The EFG tensor at
the Mn in (CO)5MnSnPh3 and (CO)5MnPbPh3 is dictated
primarily by the approximate C4 local symmetry and is
relatively unaffected by the three phenyl rings which break
the rotational symmetry. For (CO)5MnGePh3 the ηQ values
for the two sites are 0.30 and 0.36, indicating that eQVXX/h
and eQVYY/h differ by 8.8 and 5.8 MHz; these differences
are more than twice the CQ value of Mn2(CO)10.
For Mn2(CO)10, the 55Mn CQ is small compared to the
known range of 55Mn CQ values and the orientation of the
EFG tensor is surprising (i.e., VZZ is not along the Mn-Mn
bond, Figure 4). The nuclear quadrupolar coupling con-
stant,CQ ) eQVZZ/h, depends on the largest component of
the EFG, VZZ, and the nuclear quadrupole moment, Q. Since
Q is relatively large for 55Mn, a CQ of 3.28 MHz translates
into a very small VZZ; the charge distribution about the Mn
is nearly spherical, an unusual case for an MX5Y complex.
The orientation of the EFG tensor suggests that distortion
in the octahedral geometry, i.e., different ligand bond lengths
and bond angles and lattice effects, are likely responsible
for the EFG at the Mn nucleus rather than the MX5Y
coordination. The dominant role of small structural distor-
tions to the EFG is also emphasized by the value of ηQ, 0.35.
If an isolated Mn2(CO)10 molecule had a C4 symmetry axis
passing through the Mn site, the ηQ would be 0, but the
distortions in the structure,56 such as the fact that the OC-
Mn-Mn bond angle is not 180°, break the C4 symmetry and
result in the nonzero value of ηQ.
The 55Mn CQ values of (CO)5MnMPh3 (M ) Ge, Sn, Pb)
are an order of magnitude larger than for Mn2(CO)10 and
decrease as M becomes heavier, in agreement with previous
results.24 As has been previously noted, this conforms to the
trend in the electronegativities of the M atoms, øGe ) 2.02,
øSn ) 1.72, øPb ) 1.55.64 The larger CQ values suggest that
the EFG is likely dominated by the MX5Y coordination
geometry and that the distortions from octahedral geometry
introduced by the lattice will be a secondary contribution.
The orientation of the EFG tensor is expected to be such
that VZZ is nearly parallel to the Mn-M bond.24 The MPh3
groups destroy the C4 local symmetry at Mn, and conse-
quently, there are no symmetry-related reasons for the 55Mn
EFG or chemical shift tensors to be axially symmetric.
The ability to observe small variations in the EFG of MX5Y
complexes is an important step toward using SSNMR to
characterize transition metal solids. Structural distortions are
routinely investigated using SSNMR of MX6 and fac-MX3Y3
complexes because these complexes typically yield small or
zero EFGs.63 The ability to perform NMR of solids with large
CQ values will lead to greater sensitivity in detecting different
contributions to the EFG tensor, e.g., ligand charge distribu-
tion versus distortions in geometry.
3. Manganese Chemical Shift Anisotropies in (CO)5MnY
Compounds. The 55Mn chemical shift tensors for the
(CO)5MnMPh3 (M ) Ge, Sn, Pb) complexes and Mn2(CO)10
do not display an obvious trend in any of the components
of the chemical shift tensor as the M atom is changed, Table
2. The relative orientations of the chemical shift and EFG
tensors can usually be obtained from analysis of an NMR
powder pattern. If the orientation of the EFG tensor in the
molecular frame is available from single-crystal NMR
experiments, symmetry arguments, or computational quantum
chemistry, then the orientation of the chemical shift tensor
in the molecular frame can be obtained. As discussed
previously, the unique component of the chemical shift tensor
in the molecular frame for Mn2(CO)10, δ33, is along the Mn-
Mn bond, 90° from VZZ. Like Mn2(CO)10, (CO)5MnGePh3
and (CO)5MnSnPh3 have nearly axially symmetric chemical
shift tensors. In both cases, the unique component of the
chemical shift tensor, δ11, is parallel to VZZ and is therefore
along the M-M bond, Figures 5 and 6. In general, for MX5Y
complexes with ideal geometry, the unique component of
the chemical shift tensor is expected to be along the Mn-Y
bond. The chemical shift tensors for the four sites of
(CO)5MnPbPh3 deviate from axial symmetry, κ ) 0.7, and
do not appear to be coincident with the EFG tensor.
The four unique sites resolved in (CO)5MnPbPh3 demon-
strate that, even when the EFG is dominated by the charge
distribution created by the MX5Y geometry, changes in the
EFG due to subtle variations in molecular structure can be
detected. Despite being the same molecules, the 55Mn CQ
values of the four sites vary by up to 15%.
The component of the chemical shift tensor approximately
along the Mn-M bond arises from magnetic-dipole-allowed
mixing of occupied and virtual molecular orbitals in a plane
perpendicular to the Mn-M bond. Therefore, the shielding
along the bond should be constant if there is no change in
(64) Alred, A. L.; Rochow, E. G. J. Inorg. Nucl. Chem. 1958, 5, 264-
268.
8498 Inorganic Chemistry, Vol. 45, No. 21, 2006

Ultrahigh-Field NMR Spectroscopy of Transition Metals
theMn(CO)4 plane.65 Thisappearstobetruefor(CO)5MnGePh3
and (CO)5MnPbPh3, δ11 is approximately -2330 ppm, but
δ11 for (CO)5MnSnPh3 is different. Density functional theory
calculations on idealized structures did not reveal any obvious
reasons for the different chemical shift parameters. Without
crystal structures, further interpretation is impractical.
tensors, information about distortions in the MX5Y geometry
can be obtained. In addition, SSNMR studies at ultrahigh
fields offer the ability to perform routine, single-step
acquisitions of samples where the CQ is in excess of 60 MHz
for spin I g 5/2 nuclei, as shown for η5-CpMn(CO)3. Clearly,
the availability of ultrahigh-field magnets will facilitate NMR
investigation of many other transition metal nuclei in solid
inorganic and organometallic compounds.
Conclusions
We have demonstrated some of the advantages of per-
forming solid-state 55Mn NMR spectroscopy at 21.14 T. The
ability to extract small chemical shift anisotropies from NMR
spectra that are dominated by the quadrupolar interaction at
lower fields is established by studying Mn2(CO)10 and
(CO)5MnMPh3 (M ) Ge, Sn, Pb). The high resolution
attainable at ultrahigh fields makes it possible to distinguish
crystallographically nonequivalent sites in solid powder
samples using MQMAS, even when the CQ values are
relatively large and the differences in isotropic chemical shifts
are small. On the basis of small differences in the EFG
Acknowledgment. The research was funded by the
Natural Sciences and Engineering Research Council (NSERC)
of Canada, the Canada Foundation for Innovation (CFI), the
Alberta Ingenuity Fund, and the University of Alberta.
R.E.W. is a Canada Research Chair in Physical Chemistry
at the University of Alberta. Access to the 900 MHz NMR
spectrometer was provided by the National Ultrahigh-Field
NMR Facility for Solids (Ottawa, Canada), a national
research facility funded by CFI, the Ontario Innovation Trust,
Recherche Que´bec, the National Research Council of Canada,
and Bruker BioSpin and managed by the University of
Ottawa (www.nmr900.ca).
(65) Ooms, K. J.; Wasylishen, R. E. Can. J. Chem. 2006, 84, 300-308.
(66) Mason, J. Solid State Nucl. Magn. Reson. 1993, 2, 285-288.
(67) Willans, M. J.; Feindel, K. W.; Ooms, K. J.; Wasylishen, R. E. Chem.
Eur. J., 2006, 12, 159-168.
IC0608445
Inorganic Chemistry, Vol. 45, No. 21, 2006 8499