Spectroscopic properties and electronic structure of the cycloheptatrienyl molybdenum alkynyl complexes [Mo(C≡CR)(Ph2PCH 2CH2PPh2)(η-C7H 7)]n+ (n = O or 1; R = But, Fc, CO 2Me, or C6H4-4-X, X = NH2, OMe, Me, H, CHO, CO2Me)

Article abstract of DOI:10.1021/om901074b

A series of molybdenum alkynyl complexes [Mo(C≡CR)(dppe)(η-C ₇H₇)] featuring a range of alkynyl substituents R with varying electron-donating and -withdrawing properties have been prepared. Oxidation of representative members of the series to the corresponding 17-electron radical cations has been achieved through both chemical oxidation and in situ spectroelectrochemical methods. The largely metal-centered character of the HOMO in this class of compounds has been established through a combination of experimental measurements (IR, UV-vis-NIR, EPR spectroscopies) and DFT-based calculations and rationalized in terms of the stabilization of the metal d_xy, d_yz, d_xz, and dX²-y ² through π- and δ-interactions with the C₇H ₇ ring and concomitant destablization of the d _Z2 orbital.

Full text of DOI:10.1021/om901074b

Organometallics 2010, 29, 1261–1276 1261
DOI: 10.1021/om901074b
Spectroscopic Properties and Electronic Structure of the
Cycloheptatrienyl Molybdenum Alkynyl Complexes
[Mo(CtCR)(Ph2PCH2CH2PPh2)(η-C7H7)]nþ (n = 0 or 1; R = But,
Fc, CO2Me, or C6H4-4-X, X = NH2, OMe, Me, H, CHO, CO2Me)
Neil J. Brown,† David Collison,*,‡ Ruth Edge,‡ Emma C. Fitzgerald,§ Madeleine Helliwell,§
Judith A. K. Howard,† Hannah N. Lancashire,§ Paul J. Low,*,† Joseph J. W. McDouall,*,§
James Raftery,§ Charlene A. Smith,§ Dmitry S. Yufit,† and Mark W. Whiteley*,§
†Department of Chemistry, Durham University, South Road, Durham, DH1 3LE, U.K., ‡EPSRC National
Service for EPR Spectroscopy, School of Chemistry, University of Manchester, Manchester, M13 9PL,
U.K., and §School of Chemistry, University of Manchester, Manchester, M13 9PL, U.K.
Received December 16, 2009
A series of molybdenum alkynyl complexes [Mo(CtCR)(dppe)(η-C7H7)] featuring a range of
alkynyl substituents R with varying electron-donating and -withdrawing properties have been
prepared. Oxidation of representative members of the series to the corresponding 17-electron radical
cations has been achieved through both chemical oxidation and in situ spectroelectrochemical
methods. The largely metal-centered character of the HOMO in this class of compounds has been
established through a combination of experimental measurements (IR, UV-vis-NIR, EPR spectro-
scopies) and DFT-based calculations and rationalized in terms of the stabilization of the metal dxy,
dyz, dxz, and dx2-y2 through π- and δ-interactions with the C7H7 ring and concomitant destablization
of the dz2 orbital.
Introduction
applications in electronic,3 magnetic,4 and optical5 devices.
Central to these investigations is an understanding of elec-
tronic structure and consequently the degree of electronic
interaction between the metal center and the alkynyl ligand;6
recent work on the group 8 metal systems [M(CtCR)-
(dppe)Cp*]nþ (M = Fe7-10 or Ru;11,12 Cp* = C5Me5; n =
0 or 1) demonstrates a strong dependence on the identity of
the metal. Thus the iron (3d) complexes exhibit substantial
metal character in the frontier orbitals, whereas the HOMOs
of the ruthenium (4d) analogues show more alkynyl ligand
character, leading to interest in heterometallic systems con-
taining these moieties.13
Metal alkynyl complexes have been the focus of intense
research activity for many decades.1,2 Much of the early
work on synthetic methodology and structural properties of
alkynyl complexes has developed into current studies of
alkynyl complexes as molecular materials with potential
*Corresponding authors. E-mail: p.j.low@durham.ac.uk; Mark.
Whiteley@manchester.ac.uk.
(1) Nast, R. Coord. Chem. Rev. 1982, 47, 89.
(2) Long, N. J.; Williams, C. K. Angew. Chem., Int. Ed. 2003, 42, 2586.
(3) (a) Gauthier, N.; Argouarch, G.; Paul, F.; Humphrey, M. G.;
Toupet, L.; Ababou-Girad, S.; Sabbah, H.; Hapiot, P.; Fabre, B. Adv.
Mater. 2008, 20, 1952. (b) Low, P. J. Dalton Trans. 2005, 2821. (c) Kim, B.;
Beebe, J. M.; Olivier, C.; Rigaut, S.; Touchard, D.; Kushmerick, J. G.; Zhu,
X. Y.; Frisbie, C. D. J. Phys. Chem. C 2007, 111, 7521.
The cycloheptatrienyl molybdenum auxiliary Mo(dppe)-
(η-C7H7) is well recognized as an isoelectronic analogue of
the widely explored group 8 systems M(dppe)Cp* (M = Fe
(4) (a) Lapinte, C. J. Organomet. Chem. 2008, 693, 793. (b) Paul, F.;
Lapinte, C. In Unusual Structures and Physical Properties in Organome-
tallic Chemistry; Gielen, M., Willem, R., Wrackmeyer, B., Eds.; John Wiley
and Sons Ltd.: New York, 2002.
(7) Connelly, N. G.; Gamasa, M. P.; Gimeno, G.; Lapinte, C.; Lastra,
E.; Maher, J. P.; Le Narvor, N.; Rieger, A. L.; Rieger, P. H. J. Chem.
Soc., Dalton Trans. 1993, 2575.
(8) Denis, R.; Toupet, L.; Paul, F.; Lapinte, C. Organometallics 2000,
19, 4240.
(9) Costuas, K.; Paul, F.; Toupet, L.; Halet, J.-F; Lapinte, C.
(5) (a) Samoc, M.; Dalton, G. T.; Gladysz, J. A.; Zheng, Q.; Velkov,
˚
Y.; Agren, H.; Norman, P.; Humphrey, M. G. Inorg. Chem. 2008, 47,
9946. (b) Samoc, M.; Gauthier, N.; Cifuentes, M. P.; Paul, F.; Lapinte, C.;
Humphrey, M. G.; Dalton, G. T. Angew. Chem., Int. Ed. 2008, 47, 629. (c)
Whittall, I. R.; McDonagh, A. M.; Humphrey, M. G.; Samoc, M. Adv.
Organomet. Chem. 1998, 42, 291. (d) Whittall, I. R.; McDonagh, A. M.;
Humphrey, M. G.; Samoc, M. Adv. Organomet. Chem. 1999, 43, 349. (e)
Morrall, J. P.; Dalton, G. T.; Humphrey, M. G.; Samoc, M. Adv. Organomet.
Chem. 2008, 55, 61. (f) Powell, C. E.; Humphrey, M. G. Coord. Chem. Rev.
2004, 248, 725.
(6) (a) Koentjoro, O. F.; Rousseau, R.; Low, P. J. Organometallics
2001, 20, 4502. (b) Lichtenberger, D. L.; Gruhn, N. E.; Renshaw, S. K. J.
Mol. Struct. 1997, 405, 79. (c) Lichtenberger, D. L.; Renshaw, S. K.; Wong,
A.; Tagge, C. D. Organometallics 1993, 12, 3522. (d) Delfs, C. D.; Stranger,
R.; Humphrey, M. G.; McDonagh, A. M. J. Organomet. Chem. 2000, 607,
208. (e) McGrady, J. E.; Lovell, T.; Stranger, R.; Humphrey, M. G.
Organometallics 1997, 16, 4004.
Organometallics 2004, 23, 2053.
(10) Paul, F.; Toupet, L.; Thepot, J.-Y.; Costuas, K.; Halet, J.-F.;
ꢀ
Lapinte, C. Organometallics 2005, 24, 5464.
(11) Paul, F.; Ellis, B. G.; Bruce, M. I.; Toupet, L.; Roisnel, T.;
Costuas, K.; Halet, J.-F.; Lapinte, C. Organometallics 2006, 25, 649.
(12) Fox, M. A.; Roberts, R. L.; Khairul, W. M.; Hartl, F.; Low, P. J.
J. Organomet. Chem. 2007, 692, 3277.
(13) (a) Gauthier, N.; Oliver, C.; Rigaut, S.; Touchard, D.; Roisnel,
T.; Humphrey, M. G.; Paul, F. Organometallics 2008, 27, 1063. (b) Bruce,
M. I.; Costuas, K.; Davin, T.; Ellis, B. G.; Halet, J.-F.; Lapinte, C.; Low, P. J.;
Smith, M. E.; Skelton, B. W.; Toupet, L.; White, A. H. Organometallics
2005, 44, 3261. (c) Szafert, S.; Paul, F.; Meyer, W. E.; Gladysz, J. A.; Lapinte,
C. C. R. Chem. 2008, 11, 693.
r
2010 American Chemical Society
Published on Web 02/04/2010
pubs.acs.org/Organometallics

1262 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
or Ru) and has an extensive organometallic chemistry as
a supporting group to alkynyl-based ligands;14-17 indeed
alkynyl derivatives of the type [Mo(CtCR)(dppe)(η-C7H7)]nþ
(R = Ph or But; n = 0 or 1) have been in the vanguard of
spectroscopic14 and structural investigations18 on the redox
chemistry of metal alkynyl complexes. Although formally
isoelectronic with M(dppe)Cp* (M=Fe, Ru), the Mo(dppe)-
(η-C7H7) auxiliary is distinctive in the combination of a 4d
metal system with very low formal potentials for one-electron
oxidation19 exemplified by a comparison of the alkynyl deri-
vatives {cf. E1/2 (V vs SCE, CH2Cl2/0.2 M [NBu4]BF4),
[M(CtCPh)(dppe)(η-L)], M = Ru, L = Cp*, 0.35;20 M =
Fe, L=Cp*, -0.04;20 M=Mo, L=η-C7H7, -0.1514}. The
Mo(dppe)(η-C7H7) auxiliary in combination with the alkynyl
fragment, which might be considered as the prototypical un-
saturated carbon-based ligand,21 therefore offers an interesting
system for further exploration of the M-CtC interaction.
In this work a series of alkynyl complexes [Mo(CtCR)(dppe)-
(η-C7H7)]nþ (n = 0 or 1; R = But, Fc, CO2Me, C6H4-4-X;
X = NH2, OMe, Me, H, CHO, CO2Me) have been studied
using a range of spectroscopic, including spectroelectro-
chemical, and theoretical techniques. Together these inves-
tigations demonstrate that the complexes [Mo(CtCR)-
(dppe)(η-C7H7)]nþ feature largely metal-localized occupied
frontier orbitals.
Scheme 1. Routes to [Mo(CtCR)(dppe)(η-C7H7)] Complexes
described a convenient route to [MoBr(dppe)(η-C7H7)] (1),17
which permits the ready syntheses of an extended range
of alkynyl derivatives. Thus, reaction of 1 with HCtCR
in refluxing methanol gives the vinylidene complexes [Mo-
{CdC(H)R}(dppe)(η-C7H7)]þ directly as the bromide salts,
which were not isolated but deprotonated in situ by base
(KOBut or NaOMe), resulting in precipitation of the alkynyl
complexes [Mo(CtCR)(dppe)(η-C7H7)]. Alternatively, fluor-
ide-mediated desilylation/metalation approaches23 have also
been employed in certain cases. For example, reaction of
Me3SiCtCC6H4-4-OMe with [MoBr(dppe)(η-C7H7)] and
KF in methanol afforded 2f, in yields comparable to those
obtained from the terminal alkyne. The new aryl-alkynyl com-
plexes 2e-2i were isolated as blue-black, brown, or green solids,
in yields in the range ca. 30-60%.
The previously reported 17-electron radical cations [Mo-
(CtCR)(dppe)(η-C7H7)][BF4] (R = Ph, [2a]þ; R = But,
[2b]þ)14 were also prepared by respective oxidation of 2a or
2b with [FcH][BF4] in CH2Cl2 and obtained as deep purple-
blue ([2a]þ) or orange ([2b]þ) solids. These isolated examples
of the radical cations facilitate structural and other investi-
gations, complemented by studies of the radicals generated in
situ by spectroelectrochemical techniques.
Spectroscopic Studies. The complexes 2a-2i were charac-
terized by microanalysis, mass spectrometry, infrared spec-
troscopy, and by 1H, 31P{1H}, and 13C{1H} NMR spectro-
scopies (see Experimental Section and Table 1). The electro-
nic properties of the aryl substituent X in [Mo(CtCC6H4-4-
X)(dppe)(η-C7H7)] correlate with trends in ν(CtC) data in
the infrared spectra and δCR in the 13C{1H} NMR data.
Thus in progressing from electron-releasing to electron-
withdrawing substituents X in the series 2e-2i, ν(CtC)
decreases from 2055 cm-1 (2e) to 2030, 1993(shoulder)
cm-1 (2i). In the case of the methyl propiolate complex 2d,
the combination of a strongly electron-donating metal sys-
tem and a highly electron-withdrawing alkynyl substituent
results in an exceptionally low value for ν(CtC) (2020
cm-1). In the 13C NMR spectra the CR resonances were
readily identified by the characteristic JCP coupling (ca. 26
Hz), the chemical shift of which tracked the electronic
properties of the alkynyl substitutent. Thus, in the case of
Results and Discussion
Synthetic Studies. An overview of the synthetic work is
given in Scheme 1. The key advances are (i) the development
of an improved synthetic route to the previously reported
derivatives [Mo(CtCR)(dppe)(η-C7H7)] [R = Ph, 2a;14
But, 2b;14 Fc, 2c22 {Fc = ferrocenyl}] starting from [MoBr-
(dppe)(η-C7H7)] (1) and (ii) the synthesis of a new series
of aryl-alkynyl complexes [Mo(CtCC6H4-4-X)(dppe)-
(η-C7H7)]0/þ1 featuring a range of X groups with varying
donor/acceptor properties [X = NH2, 2e; OMe, 2f; Me, 2g;
CHO, 2h; CO2Me, 2i]. The non-aryl alkynyl series (2b,c) was
also extended with the preparation of the methyl propiolate
derivative [Mo(CtCCO2Me)(dppe)(η-C7H7)], 2d.
In previous studies,14 alkynyl complexes [Mo(CtCR)-
(dppe)(η-C7H7)] were obtained by deprotonation of the corres-
ponding vinylidenes [Mo{CdC(H)R}(dppe)(η-C7H7)]PF6,
prepared in turn from the sandwich system [Mo(η-C6H5Me)-
(η-C7H7)]PF6 through sequential reaction with dppe and
the terminal alkyne HCtCR. However, we have recently
(14) Adams, J. S.; Bitcon, C.; Brown, J. R.; Collison, D.; Cunningham,
M.; Whiteley, M. W. J. Chem. Soc., Dalton Trans. 1987, 3049.
(15) Beddoes, R. L.; Bitcon, C.; Grime, R. W.; Ricalton, A.; Whiteley,
M. W. J. Chem. Soc., Dalton Trans. 1995, 2873.
(16) (a) Cambridge, J.; Choudhary, A.; Friend, J.; Garg, R.; Hill, G.;
Hussain, Z. I.; Lovett, S. M.; Whiteley, M. W. J. Organomet. Chem.
1999, 577, 249. (b) Lancashire, H. N.; Ahmed, R.; Hague, T. L.; Helliwell,
M.; Hopgood, G. A.; Sharp, L.; Whiteley, M. W. J. Organomet. Chem. 2006,
691, 3617.
(17) Grime, R. W.; Helliwell, M.; Hussain, Z. I.; Lancashire, H. N.;
Mason, C. R.; McDouall, J. J. W.; Mydlowski, C. M.; Whiteley, M. W.
Organometallics 2008, 27, 857.
(18) (a) Beddoes, R. L.; Bitcon, C.; Ricalton, A.; Whiteley, M. W. J.
Organomet. Chem. 1989, 367, C21. (b) Beddoes, R. L.; Bitcon, C.; Whiteley,
M. W. J. Organomet. Chem. 1991, 402, 85.
(19) (a) Petrovic, D.; Hrib, C. G.; Randoll, S.; Jones, P. G.; Tamm,
M. Organometallics 2008, 27, 778. (b) Tamm, M.; Bannenberg, T.; Dressel,
€
B.; Frohlich, R.; Holst, C. Inorg. Chem. 2002, 41, 47.
(20) Bitcon, C.; Whiteley, M. W. J. Organomet. Chem. 1987, 336, 385.
(21) Bruce, M. I.; Low, P. J. Adv. Organomet. Chem. 2004, 50, 179.
(22) Hussain, Z. I.; Whiteley, M. W.; McInnes, E. J. L. J. Organomet.
Chem. 1997, 543, 237.
(23) Bruce, M. I.; Hall, B. C.; Kelly, B. D.; Low, P. J.; Skelton, B. W.;
White, A. H. J. Chem. Soc., Dalton Trans. 1999, 3719.

Article
Organometallics, Vol. 29, No. 5, 2010 1263

1264 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
Figure 3. Plot of the molecular structure of 2g showing the
atom-labeling scheme.
Figure 1. Plot of the molecular structure of 2b showing the
atom-labeling scheme. In this and all subsequent structural
figures, thermal ellipsoids are plotted at 50% probability unless
stated otherwise. Hydrogen atoms have been omitted for clarity.
Figure 4. Plot of the molecular structure of 2h showing the
atom-labeling scheme.
Mo-CR distance exhibits a small dependence on the identity
of the alkynyl substituent and decreases from around 2.14 A
in 2a, 2b, 2d, and 2g to 2.11-2.12 A in 2h and 2i, which
˚
Figure 2. Plot of the molecular structure of 2d showing the
atom-labeling scheme.
˚
feature electron-withdrawing aryl-alkynyl substituents. The
Mo-P bond length is a sensitive probe of metal electron
density through M-P back-bonding effects,24 and the slight
elongation of the Mo-P distances in 2a, 2d, 2h, and 2i by
comparison with 2b and 2g is consistent with the electronic
parameters of the alkynyl substituents. Finally the longer
Cβ-C(substituent) bond length [C(9)-C(10)] in 2b probably
reflects the different hybridization at C(10) of the alkyl
substituent.
the phenyl and tolyl alkynyl complexes 2a and 2g, CR was
observed near 140 ppm (Table 1). The introduction of the
more electron-withdrawing groups (e.g., 2d, 2h) results in
significant shift of the CR resonance to nearer 160 ppm
(Table 1), while in the case of the electron-donating alkyl
derivative 2b, CR was observed at 113 ppm. Finally the
31P{1H} NMR resonances for 2a-2i all fall within a very
narrow range (δ 64-66 ppm) with little dependence upon the
identity of the alkynyl substituent. Similar trends are appar-
ent in equivalent 31P{1H} NMR data reported for [M(Ct
CC6H4-4-X)(dppe)Cp*] (M = Fe, δ 100-102 ppm;8 M =
Ru, δ 81-82.5 ppm11).
Structural Investigations. The X-ray crystal structures of
2b, 2d, 2g, 2h, and 2i and the 17-electron radical [2b][BF4]
have been determined (Figures 1 to 6, respectively), and
important bond lengths and angles (including those for the
previously reported structures of 2a and [2a]þ)18b are sum-
marized in Table 2. In the 18-electron systems, the key
The X-ray structures of the redox pairs 2a/[2a]þ and
2b/[2b]þ permit an examination of the structural effects of
one-electron oxidation within this series. The principal
structural modifications in each redox pair resulting from
removal of an electron are remarkably consistent and comprise
ꢀ
(24) (a) Cordiner, R. L.; Albesa-Jove, D.; Roberts, R. L.; Farmer, J.
D.; Puschmann, H.; Corcoran, D.; Goeta, A. E.; Howard, J. A. K.; Low,
P. J. J. Organomet. Chem. 2005, 690, 4908. (b) Manna, J.; John, K. D.;
Hopkins, M. D. Adv. Organomet. Chem. 1995, 38, 79.

Article
Organometallics, Vol. 29, No. 5, 2010 1265
one-electron oxidation with the separation between cathodic
and anodic peak potentials comparable to that determined
for the internal ferrocene standard. The anisole derivative 2f
also exhibits a second oxidation at a significantly more
positive potential, which is only partially chemically rever-
sible and likely due to oxidation of the anisole moiety. One-
electron oxidation of the methyl propiolate derivative 2d is
relatively unfavorable (in the thermodynamic sense) and
reflects the electron-withdrawing properties of the ester
group bonded directly to the alkynyl moiety. For compar-
ison, Table 3 also presents electrochemical data for one-
electron oxidation of the analogous group 8 complexes
[M(CtCC6H4-4-X)(dppe)Cp*] (M=Fe, 3a, 3e, 3f, 3g; Ru,
4a, 4e, 4f, 4g). One-electron oxidation of the various mem-
bers of the molybdenum series 2e-2i is clearly more thermo-
dynamically favorable than for analogous group 8 com-
plexes. However the E1/2 values of the molybdenum aryl-
alkynyl complexes 2a and 2e-2i span a range of only ca. 160
mV and are therefore less sensitive to the identity of the
alkynyl substituent than the analogous ruthenium com-
plexes, for which the equivalent range is in excess of 300 mV.
IR Spectroelectrochemical Analysis. Spectroelectrochem-
ical methods permit the convenient and rapid collection of
spectroscopic data from each member of a redox-related
family of compounds.25 In the case of metal alkynyl com-
plexes, the characteristic ν(CtC) band reflects a convolu-
tion of σ/π-donation, π-back-bonding, and physical/
kinematic effects. Nevertheless, the changes in ν(CtC)
that occur as a result of a redox change within a series of
closely related alkynyl complexes can be used to assess the
degree of involvement of the alkynyl ligand in the redox
orbital,8,11,12,26 especially when coupled with observation
of other IR-active bands elsewhere in the molecule.27
Given that π-back-bonding effects make a minimal con-
tribution to net M-C bond strength in d6 pseudo-octahedral
alkynyl complexes,11,12 the magnitude of the negative shift in
ν(CtC) resultant upon one-electron oxidation appears to be
a good first-order indicator of the extent of the character
of the redox-active frontier orbital within a series of similar
complexes. Thus the large negative shifts of ν(CtC)
observed upon oxidation of [Ru(CtCR)(dppe)Cp*] (-90
to -145 cm-1) are consistent with depopulation of an orbital
with a large amount of CtC bonding character and sig-
nificant radical-ligand character in the cations [Ru(Ct
CR)(dppe)Cp*]þ.11,12 By contrast the Fe complexes [Fe-
(CtCR)(dppe)Cp*] feature more metal-centered oxidation
processes, with less alkynyl ligand character in the resulting
semioccupied orbital, and consequently much smaller shifts
in ν(CtC), upon oxidation (ca. 0 to -70 cm-1).8,9 Several
factors may contribute to the marked differences between the
Fe and Ru series, but key considerations appear to be the
larger spatial extension (and better polarizability) of 4d
Figure 5. Plot of the molecular structure of 2i showing the
atom-labeling scheme.
Figure 6. Plot of the structure of 2b[BF4] (as one of four crystal-
lographically independent ion pairs in the asymmetric unit)
showing the atom-labeling scheme. Thermal ellipsoids are
plotted at 20% probability.
˚
a contraction in the Mo-CR bond length (0.07 A for both
˚
cases) and an increase in theþMo-P bond length (0.06 A for
þ
˚
2a/[2a] , 0.07 A for 2b/[2b] ); the latter change is readily
accounted for by Mo-P back-bonding effects, while the
contraction in the Mo-CR distance, although probably
largely electrostatic in origin, is further rationalized by a
DFT treatment (see below).
(25) (a) Krejeik, M.; Danek, M.; Hartl, F. J. Electroanal. Chem. 1991,
317, 179. (b) Spectroelectrochemistry; Kaim, W., Klein, A., Eds.; Royal
Society of Chemistry: Cambridge, 2008.
(26) (a) Bruce, M. I.; Low, P. J.; Hartl, F.; Humphrey, P. A.; de
Montigny, F.; Jevric, M.; Lapinte, C.; Perkins, G. J.; Roberts, R. L.;
Skelton, B. W.; White, A. H. Organometallics 2005, 24, 5241. (b) Bruce,
M. I.; Low, P. J.; Costuas, K.; Halet, J.-F.; Best, S. P.; Heath, G. A. J. Am.
Chem. Soc. 2000, 122, 1949.
Electrochemistry. A key objective of this investigation is to
explore the redox chemistry of the metal alkynyls [Mo-
(CtCR)(dppe)(η-C7H7)] to facilitate comparison with [M-
(CtCR)(dppe)Cp*] (M = Fe or Ru) analogues. To com-
mence these investigations, we examined the cyclic voltam-
metry of each of the complexes 2a-2i at a platinum electrode
in CH2Cl2 (Table 3).
(27) (a) Fox, M. A.; LeGuennic, B.; Roberts, R. L.; Baines, T. E.;
ꢀ
Yufit, D. S.; Albesa-Jove, D.; Halet, J.-F.; Hartl, F.; Howard, J. A. K.;
Low, P. J. J. Am. Chem. Soc. 2008, 130, 3566. (b) Brady, M.; Weng, W.;
Under these conditions each of the complexes 2a-2i
undergoes a diffusion-controlled, chemically reversible,
€
Zhou, Y.; Seyler, J. W.; Amoroso, A. J.; Arif, A. M.; Bohme, M.; Frenking,
G.; Gladysz, J. A. J. Am. Chem. Soc. 1997, 119, 775.

1266 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
Table 2. Selected Bond Length and Bond Angle Data for the Alkynyl Complexes [Mo(CrtCβR)(dppe)(η-C7H7)]nþ (n = 0 or 1)
2a
2b
2d
2g
2h
2i
[2a]þ
[2b]þ
˚
Bond Lengths (A)
Mo-CR
2.138(5)
1.205(6)
1.434(7)
2.477(1),
2.467(1)
2.1382(17)
1.216(2)
1.485(2)
2.4648(4),
2.4520(4)
2.146(2)
2.140(5)
1.196(6)
1.441(7)
2.4677(14),
2.4525(14)
2.1094(19)
1.212(3)
1.426(3)
2.4767(5),
2.4772(5)
2.122(7)
1.191(10)
1.453(10)
2.477(2),
2.455(2)
2.067(9)
1.196(11)
1.445(12)
2.538(2),
2.528(3)
2.070(11)
1.201(12)
1.507(14)
2.537(3),
2.528(3)
CR-Cβ
1.179(3)
1.436(3)
2.4731(6),
2.4686(6)
Cβ-C(substituent)
Mo-P
Bond Angles (deg)
Mo-CR-Cβ
CR-Cβ-C(substituent)
P-Mo-P
178.5(4)
177.9(5)
78.2(1)
83.8(2),
77.3(2)
175.17(14)
177.42(18)
78.312(14)
80.24(4),
176.33(18)
177.6(2)
78.38(2)
80.52(5),
79.18(5)
174.0(5)
174.6(6)
78.39(5)
81.59(13),
77.96(13)
178.54(16)
176.71(19)
78.558(18)
84.47(5),
174.17(6)
171.0(8)
78.63(7)
81.21(2),
77.43(2)
174.6(8)
175.0(10)
78.4(1)
83.5(3),
75.1(3)
175.1(10)
178.2(12)
78.70(10)
80.5(3),
P-Mo-CR
78.67(4)
76.73(5)
75.0(3)
Table 3. Cyclic Voltammetric Data for [M(CtCR)(dppe)(η-L)]
(M = Mo, L = C7H7; M = Fe, Ru, L = Cp*)h
Table 4. Infrared Spectroscopic Data for the Alkynyl Complexes
[Mo(CtCR)(dppe)(η-C7H7)]nþ
compound
E1/2 (V)
notes
n = 0
n = 1
Δν(CtC)
ref
a
2a (-30 °C)
2b (-30 °C)
2c
2d (-30 °C)
2e
2f (-30 °C)
2g (-30 °C)
2h
-0.72
2aa
2ba
2cb
2dc,d
2eb
2fb
2045
2057
2057
2020
2055
2055
2050
2047
2030, 1993
2032
2044
2009
not obsd
1996
2011
2017
2033
2031
-13
-13
-48
n/a
-59
-44
-33
-14
-1
14
14
a
-0.82
b
-0.79, þ0.07
this work
this work
this work
this work
this work
this work
this work
a
-0.50
a,f
a,f,g
a
-0.81
-0.74, þ0.60
-0.72
2gc
2hb
2ic,e
a
-0.70
a
2i
-0.65
c
3a, [Fe(CꢁCC6H5)(dppe)Cp*]
-0.61
a Data for n = 1 from isolated sample in CH2Cl2. b Spectroelectro-
chemical measurement in 0.1 M NBu4PF6/THF. c Spectroelectrochem-
ical measurement in 0.1 M NBu4PF6/CH2Cl2. d Also ν(CdO) 1659 (n =
c
3e, [Fe(CtCC6H4-4-NH2)(dppe)Cp*]
3f, [Fe(CtCC6H4-4-OMe)(dppe)Cp*]
3g, [Fe(CtCC6H4-4-Me)(dppe)Cp*]
4a, [Ru(CtCC6H5)(dppe)Cp*]
-0.71
c
-0.67
c
-0.64
0), 1694 (n = 1) cm-1 e Also ν(CdO) 1706 (n = 0), 1721 (n = 1) cm-1
. .
d
-0.16
e
4e, [Ru(CtCC6H4-4-NH2)(dppe)Cp*]
4f, [Ru(CtCC6H4-4-OMe)(dppe)Cp*]
4g, [Ru(CtCC6H4-4-Me)(dppe)Cp*]
-0.41
e
Table 5. Infrared Spectroscopic Data for the Alkynyl Complexes
[M(CtCR)(dppe)Cp*]nþ (M = Fe or Ru)
-0.31
d
-0.15
a This work. b Ref 22. c Ref 8. d Ref 12. e Ref 11. f 0.1 M NBu4PF6/THF
(FcH/FcHþ = 0.00 V). g Irreversible oxidation at þ0.60 V. h All
potentials are reported vs FcH/FcHþ (FcH/FcHþ = 0.00 V) from 0.1
M NBu4PF6/CH2Cl2 solutions at ambient temperature unless stated
otherwise. Data for 2c originally reported vs SCE in 0.2 M NBu4BF4
(FcH/FcHþ = 0.56 V)22 have been adjusted to the FcH/FcHþ reference
scale adopted in the current work.
M = Fe
M = Ru
Δν-
(CtC)
Δν-
n = 1 (CtC)
R
n = 0
n = 1
n = 0
e
C6H5
2053a 2021,
1988b
2072c
1930
1928
1929
1940
1942
-141
-145
-145
-135
-106
C6H4-4-
Me
C6H4-4-
OMe
C6H4-4-
NH2
C6H4-4-
NO2
2056b 1994b
-62
2073c
2074d
versus 3d metal orbitals and the energies of these metal
orbitals relative to the alkynyl π-system.
2058b 1988b
-70
Guided by the results of the cyclic voltammetry, an IR
spectroelectrochemical investigation was carried on each of the
molybdenum complexes 2a-2i. In each case, the stability of the
radical cation (and hence the validity of the ν(CtC) data) was
established by back-reduction, which resulted in full recovery of
the IR spectrum assigned to the 18-electron precursor. Table 4
summarizes the data obtained and the shifts in ν(CtC)
[Δν(CtC)] resulting from one-electron oxidation. The corre-
sponding data for [M(CtCC6H4-4-X)(dppe)Cp*] (M = Fe or
Ru) are given in Table 5. In some cases more than one band is
observed for the ν(CtC) resonance, generally ascribed to Fermi
coupling,28 and therefore the precise magnitude of Δν(CtC) is
somewhat debatable.
e
2060b 1988sh,
1962b
2075
2050shd
2048
2036, 2038b
2008b
e
2017d
a Ref 10. b Ref 8. c Ref 12. d Ref 11. e Δν(CtC) in Fe series not quoted
due to multiple ν(CtC) bands.
progressively shifted to lower wavenumber as the electron-
donating ability of the metal fragment increases (e.g., X =
Me, ν(CtC) (cm-1/CH2Cl2): M = Ru, 2073; M = Fe, 2056;
M = Mo, 2050). Furthermore it is evident that the three systems
are distinct in the magnitude of the shift in ν(CtC) following
one-electron oxidation, with the molybdenum complexes exhi-
biting the smallest negative shifts. Taking as an example X =
Comparing first IR data from across the series of neutral
alkynyl complexes [M(CtCC6H4-4-X)(dppe)(η-L)] (M =
Mo, L = C7H7; M = Fe or Ru, L = Cp*) (Tables 4 and 5)
reveals that for any given substituent X, the ν(CtC) values
reflect the nature of the metal fragment, with the band being
Me, the respective shifts in Δν(CtC) are M = Ru, -145 cm-1
;
M = Fe, -62 cm-1; and M = Mo, -33 cm-1
.
In summary, the electrochemical and spectroscopic prop-
erties of the [Mo(CtCR)(dppe)(η-C7H7)]nþ complexes more
closely resemble those of the Fe than the Ru series. The
apparently limited Mo-alkynyl mixing cannot be accounted
for by the simple analogies between the 3d and 4d transition
(28) Paul, F.; Mevellec, J.-Y.; Lapinte, C. J. Chem. Soc., Dalton
Trans. 2002, 1783.

Article
Organometallics, Vol. 29, No. 5, 2010 1267
Table 6. Selected Bond Lengths and Bond Angles for 2A and
[2A]þ, Obtained at the BP86/SVP and B3LYP/SVP Levels
2A
[2A]þ
B3LYP
BP86
B3LYP
˚
BP86
Bond Lengths (A)
Mo-CR
2.093
1.254
1.427
2.486
2.485
2.125
1.238
1.429
2.529
2.527
2.047
1.257
1.423
2.556
2.553
2.073
1.239
1.427
2.596
2.591
CR-Cβ
Cβ-C(substituent)
Mo-P
Bond Angles (deg)
Mo-CR-Cβ
CR-Cβ-C(substituent)
P-Mo-P
177.8
179.7
78.8
83.5
77.2
177.6
179.8
78.6
84.2
77.7
176.9
178.7
79.5
82.8
77.3
176.5
179.5
79.1
83.1
77.5
P-Mo-CR
a
Ct-Mo-CR
127.6
127.4
130.3
129.9
a Ct = centroid of the C7H7 ring.
series advanced to rationalize the differences between the
group 8 systems. Therefore to understand further the pre-
dominantly metal-localized redox character of the [Mo-
(CtCR)(dppe)(η-C7H7)] system, a DFT analysis of 2a/[2a]þ
was undertaken.
Figure 7. Energy (eV) of key molecular orbitals of 2A (B3LYP/
SVP including PCM solvation in CH2Cl2).
Electronic Structure Calculations. A computational study
of the electronic structure of the model system
[Mo(CtCPh)(dppe)(C7H7)]nþ (denoted 2A, [2A]þ to distin-
guish the computational and experimental systems) was
conducted at the DFT level. Starting from the crystallo-
graphic structure of 2a, full geometry optimizations were
performed on 2A and [2A]þ using both the BP86 and B3LYP
functionals and the Def2-SVP basis obtained from the
Turbomole library.29 For H, C, and P atoms, this constitutes
an all-electron split valence plus polarization basis, while for
Mo an effective core potential is applied to account for 28
core electrons (K, L, and M shells) with a split valence orbital
set including a set of f-type polarization functions. This basis
was also used to obtain IR frequencies and the UV/vis
spectra (via time-dependent density functional theory, TD-
DFT, as implemented in the Gaussian suite of programs30).
There is good agreement between the crystallographically
determined structures of 2a/[2a]þ and the DFT-optimized
geometries; key variables from the DFT-optimized geome-
tries are shown in Table 6. Comparing with the X-ray data
and averaging over these structures we find an average
˚
absolute error in the bond lengths of 0.033 A (BP86)/0.040
A (B3LYP) and 1.3° in the angles for both functionals. The
energies and composition of the key orbitals for 2A (LUMOþ1,
LUMO, HOMO, HOMO-1, HOMO-2, HOMO-18, and
HOMO-19) are summarized in Table 7 and Figure 7; Figure 8
illustrates the plots of these frontier orbitals.
˚
The structural and IR spectroscopic changes resulting
from one-electron oxidation can be satisfactorily rationa-
lized in terms of the properties of the HOMO of 2A
(Figure 8c). The antibonding character of the HOMO with
respect to the metal-alkynyl bond prescribes a shorter,
stronger Mo-CR bond on removal of an electron from 2A
to form [2A]þ (see Table 6). Conversely the alkynyl CtC
bond is modestly weakened by removal of an electron from a
HOMO that has alkynyl CtC bonding character but a
relatively small population on the C2 unit (see Table 7),
and this is reflected in the small reduction in the ν(CtC)
stretching frequency in the 17-electron radical [ν(CtC), cm-1
(CH2Cl2), calculated: 2A, 2048 (BP86)/2138 (B3LYP), [2A]þ,
2008 (BP86)/2101 (B3LYP); experimental: 2a, 2045, [2a]þ,
2032; the calculated values have not been scaled]. The
B3LYP level predicts a reduction of 37 cm-1, while the
BP86 functional gives a reduction of 40 cm-1; the absolute
values predicted by the BP86 functional are in better agree-
ment with experiment, but the trend is slightly better repro-
duced at the B3LYP level. A related DFT treatment of
[Ru(CtCPh)(PH3)2Cp] reveals that the HOMO population
at the C2 unit is 38% (and 62% in total on the C2Ph ligand, cf.
2A, 25%), and these differences can clearly account for the
much larger impact on ν(CtC) of removal of an electron
from the HOMO of the Ru system. The challenge however is
to explain the contrasting population distribution in the
HOMO of the Mo and Ru complexes, and this issue is
addressed below.
(29) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123. Weigend, F.; Ahlrichs, R. Phys. Chem.
Chem. Phys. 2005, 7, 3297. http://bases.turbo-forum.com/TURBOMO-
LE_BASISSET_LIBRARY/tbl.html.
(30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, ; Scalmani, M. G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
A key feature of the electronic structure of 2A is the strong
δ-interaction between the metal dxy and dx2-y2 orbitals and
the e2 level of the C7H7 ring, evident from HOMO-1 and

1268 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
Table 7. Population Analysis (%) of Key Molecular Orbitals of 2A at B3LYP/SVP (including PCM solvation in CH2Cl2)
fragment (excluding hydrogen atoms and Ph groups of dppe)
orbital
Mo
C7H7
C2
C6H5
P2
LUMOþ1
LUMO
HOMO
HOMO-1
HOMO-2
HOMO-18
HOMO-19
10
14
62
24
47
9
14
34
4
37
37
60
33
7
2
16
25
6
10
2
9
6
2
6
12
3
4
3
0
12
0
2
8
9
11
Figure 8. LUMOþ1, LUMO, HOMO, HOMO-1, HOMO-2, HOMO-18, and HOMO-19 of 2A at B3LYP/SVP (including PCM
solvation in CH2Cl2) plotted as an isosurface of 0.04 au.
HOMO-2 (Figure 8d and e, respectively), which are
strongly bonding between the metal and C7H7 ring. This
represents a significant difference between the electronic
structures of 2A and its ruthenium analogue [Ru(Ct
CPh)(dppe)Cp*] because δ-bonding has very limited impor-
tance in the ruthenium-cyclopentadienyl interaction. The
origin of this effect can be traced to a stabilization of the ring
orbitals of e1 and e2 symmetry as ring size increases, and this
principle has been applied with some success in rationalizing
the electronic structure and reactivity of sandwich complexes
of the type [MCp(η-C7H7)] (M = Ti, Zr, V, Cr, Mo) and
[M(η-C7H7)2]n (M = Th, Pa, U, Np, Pu, Am; n = -2, -1, 0,
þ1; M = Ti, V, Cr, n = 0).31 Essentially, as ring size
increases, the ring MOs move to lower energy,32 and there-
fore the principal frontier orbital metal-ring interactions
(31) (a) Clack, D. W.; Warren, K. D. Theor. Chim. Acta 1977, 46, 313.
(b) Evans, S.; Green, J. C.; Jackson, S. E.; Higginson, B. J. Chem. Soc.,
Dalton Trans. 1974, 304. (c) Menconi, G.; Kaltsoyannis, N. Organometal-
lics 2005, 24, 1189. (d) Tamm, M.; Kunst, A.; Bannenberg, T.; Herdtweck, E.;
€
Schmid, R. Organometallics 2005, 24, 3163. (e) Glockner, A.; Bannenberg,
T.; Tamm, M.; Arif, A. M.; Ernst, R. D. Organometallics 2009, 28, 5866. (f)
Li, J.; Bursten, B. E. J. Am. Chem. Soc. 1997, 119, 9021. (g) Wang, H.; Xie,
Y.; King, R. B.; Schaefer, H. F.III. Eur. J. Inorg. Chem. 2008, 3698.
(32) Elian, M.; Chen, M. M. L.; Mingos, D. M. P.; Hoffmann, R.
Inorg. Chem. 1976, 15, 1148.

Article
Organometallics, Vol. 29, No. 5, 2010 1269
Figure 9. HOMO and HOMO-1 of [Ru(CtCPh)(PH3)2Cp] at B3LYP/3-21G* plotted as an isosurface of 0.04 au.12
involve the e1 level for Cp but the e2 level for C7H7.
Confirmation that the e1 level of the C7H7 ligand is moved
to low energy in the half-sandwich system under investiga-
tion is provided by the identification of the π-interaction
between the C7H7 MOs of e1 symmetry and metal orbitals
with mainly dyz and dxz character; these are much lower in
the orbital manifold at HOMO-18 and HOMO-19
(Figure 8f and g). The perturbation in electronic structure
introduced by metal-ring δ-bonding leads to a reordering of
the metal d orbital manifold, resulting in a novel electronic
structure for the half-sandwich, cycloheptatrienyl molybde-
num system.
To understand the contrast between the electronic struc-
ture of 2A and that of [Ru(CtCPh)(dppe)Cp*], it is first
necessary to specify the coordinate systems employed. In the
familiar case of [Ru(CtCPh)(dppe)Cp*], DFT calculations
on the model complex [Ru(CtCPh)(PH3)2Cp] define the z
axis along the Ru-CR (alkynyl) bond, determined by dona-
tion from the filled alkynyl σ orbital to the dz2 orbital on
Ru.9,11,12 By contrast in 2A, the z axis is directed along the
line connecting the metal and the center of the C7H7 ring; the
location of the dz2 orbital along this axis is evident in
Figure 8c. For convenience, we have then assigned the plane
passing through the Mo-CR (alkynyl) bond, the metal, and
the center of the C7H7 ring as the xz plane.
Inspection of the MO energy level diagram for 2A
(Figure 7) reveals two key points. First the HOMO in 2A is
energetically discrete from both the LUMO and HOMO-1;
this contrasts with the electronic structure of [Ru(Ct
CPh)(PH3)2Cp], where the HOMO and HOMO-1 are se-
parated in energy by less than 0.1 eV and may interchange
depending on the identity of the substituent on the alkynyl
ligand.11 The lowering in energy of HOMO-1 in 2A might
be attributed to the additional δ-bonding between the
metal dxy orbital and the C7H7 ring. The second feature of
interest lies in the directional properties of the HOMO;
for [Ru(CtCPh)(PH3)2Cp] the HOMO/HOMO-1 (dxz or
dyz) include the Ru-CR (alkynyl) bond in the plane of the
orbital (Figure 9).11,12 However, in the case of 2A, the
HOMO, dz2, is approximately orthogonal to the Mo-CR
(alkynyl) bond.
Focusing now on the metal-alkynyl interaction, there are
three important frontier components: a bonding interaction
involving σ-donation from a filled orbital on the alkynyl
ligand to an acceptor metal orbital and two antibonding,
filled-filled π-interactions between the two orthogonal
CtC π-bonds on the alkynyl ligand and filled metal
d orbitals. For 2A these latter π-interactions are clearly
identified in the HOMO and HOMO-1 (Figure 8c and d)
involving metal dz2 and dxy orbitals. The σ-donation from the
alkynyl ligand is much harder to trace, but the best candidate
is shown in HOMO-19 (Figure 8g). In the ruthenium
complex [Ru(CtCPh)(PH3)2Cp] the filled-filled π-anti-
bonding interactions can also be found in the HOMO and
HOMO-1 (dxz or dyz in the coordinate system of the
cyclopentadienyl ruthenium complex). In both complexes,
the HOMO and HOMO-1 are antibonding with respect to
the metal-CR (alkynyl) interaction but conversely possess
alkynyl CtC bonding character.
The one-electron oxidation of the alkynyl complexes
[Mo(CtCPh)(dppe)(η-C7H7)] and [Ru(CtCPh)(dppe)Cp*]
will result in a depletion of electron density in the HOMO
(and to a lesser degree HOMO-1) and should therefore lead
to strengthening of the metal alkynyl bond but weakening of
the alkynyl CtC bond. However the extent of the change is
strongly dependent on the effectiveness of the metal to
alkynyl interaction in the HOMO and HOMO-1. In the
ruthenium system, there is good overlap between the metal
dxz and dyz with the orthogonal alkynyl π-bonds in the
almost degenerate HOMO and HOMO-1. Conversely for
[Mo(CtCPh)(dppe)(η-C7H7)], the metal-alkynyl interac-
tion in the HOMO is diminished by symmetry constraints;
the metal dz2 orbital of the HOMO must tilt to achieve
effective overlap with the alkynyl π orbital. While the dxy
orbital of the HOMO-1 of 2A does possess the correct
symmetry for an effective metal-alkynyl interaction, this
(in contrast to the Ru system) is significantly lower in energy
and is also involved in strong metal to ring δ-bonding.
Therefore, the redox-induced changes observed for [Mo-
(CtCPh)(dppe)(η-C7H7)] are moderated by the confine-
ment of the redox process essentially to a single orbital
in which symmetry constraints limit the efficiency of the
metal-alkynyl interaction.
In summary, the contrasting redox behavior of [Mo-
(CtCPh)(dppe)(η-C7H7)] and [Ru(CtCPh)(dppe)Cp*] can
be attributed to the very different electronic structures of
the two systems. The strongly bonding interactions between
molybdenum and the C7H7 ring lead to stabilization of the
metal dxy and dx2-y2 (e2) and dxz and dyz (e1) orbitals, leaving an
energetically discrete HOMO with significant dz2 character. The
redox process is therefore centered on an orbital with a
symmetry-constrained interaction between the metal and the
filled alkynyl π orbital such that redox-induced changes at the
alkynyl ligand are relatively small. No such symmetry con-
straints operate for [Ru(CtCPh)(dppe)Cp*], where the dxz
and dyz components of the almost degenerate HOMO and
HOMO-1 overlap directly with the filled, orthogonal alkynyl
π orbitals.
Finally we note that the novel electronic structure of the
Mo(dppe)(η-C7H7) system sets it apart in studies on the
redox chemistry of metal complexes of carbon chain systems.
DFT calculations on models for key, alternative half-sandwich

1270 Organometallics, Vol. 29, No. 5, 2010
Table 8. UV/Vis/NIR Data for Selected Members of the Series [2]nþ (n = 0, 1)a
Brown et al.
complex
nþ
2a
0
32 700 (21 000)
33 300 (15 600), br
33 900 (8400)
33 900 (7100), sh
33 800 (26 800)
30 100 (19 100), sh
32 500 (17 600)
27 000 (12 200)
28 300 (8400)
18 900 (1000), br
18 900 (2100), sh, 16 900 (4800)
þ1
12 300 (390), br
2b
2d
2f
0
þ1
0
20 000 (720)
28 300 (18800)
21 600 (8000)
27 900 (11 100)
28 200 (8600)
27 100 (12800)
28 250 (7800)
22 200 (16200)
27 850 (10 500), 24 200 (6500)
18 700 (890), br
18 600 (3200), br
17 000 (650)
16 000 (5900)
19 200 (800)
16 600 (5300)
17 300 (2200)
18 600 (2400), 16 800 (3500)
þ1
0
þ1
0
12 400 (770)
12 500 (150)
11 800 (200)
2g
2ib
32 680 (18 300)
31 700 (16 300)
þ1
0
þ1
a νmax/cm-1 (ε/M-1 cm-1) determined by spectroelectrochemical methods in 0.1 M NBu4PF6/CH2Cl2 unless stated otherwise. b 0.1 M NBu4PF6/
THF.
Figure 10. Experimental UV/visible spectra of 2a/[2a]þ in CH2-
Cl2/0.1 M NBu4PF6.
end-cap systems, Fe(dppe)Cp*,9,10 Re(NO)(PPh3)Cp*,27b,33
and Mn(Me2PCH2CH2PMe2)Cp,34 all reveal in-plane over-
lap between frontier metal d orbitals and filled carbon chain
π orbitals in the HOMO. The unique orthogonal overlap
between the dz2 orbital of the Mo(dppe)(η-C7H7) system and
the filled π orbitals of a carbon chain and the consequential
reduced metal-carbon interaction offers the potential to
enhance the stability of highly reactive radicals of extended
carbon chain complexes and to facilitate delineation of the
separate metal and carbon chain components of a redox
process.
UV/Visible Electronic Absorption Spectroscopy. The UV/
visible/NIR spectra of complexes (2a, 2b, 2d, 2f, 2g, 2i),
chosen as being representative of the series, and the corre-
sponding radicals, were recorded using spectroelectrochem-
ical methods; the experimental spectra for 2a and [2a]þ
are illustrated in Figure 10, and all data are summarized in
Table 8.
Figure 11. Simulated UV/vis spectrum of 2A (TD-DFT,
B3LYP/SVP).
intense blue-purple or emerald green color of these systems.
The energy of this visible transition is sensitive to the alkynyl
substituent, being found at higher energy as the electron-
withdrawing nature of the substituent is increased, consistent
with assignment as a LMCT (aryl-alkynyl to metal) transi-
tion. In addition, very weak, low-energy bands are also
observed in the spectra of the 17-electron radical cations
around 12 000 cm-1 (800 nm), which are likely d-d in origin.
In the case of the non-aryl-substituted derivatives [2b]þ and
[2d]þ, the LMCT bands are weaker in intensity and shifted to
higher energy, 18 600-20 000 cm-1 (537-500 nm), giving
these species a distinct orange or red coloration. These
differences between aryl- and alkyl-substituted alkynyls in-
dicate a significant contribution from the aryl group to the
LMCT processes in the former.
The neutral 18-electron complexes all show strong absorp-
tions below 30 000 cm-1/333 nm) and at ca. 25 000 cm-1/400
nm arising from transitions between the metal, phosphine,
and cycloheptatrienyl ligands. A series of lower intensity
bands around 17 000-19 200 cm-1 (590-520 nm) are re-
sponsible for the varying blue-black, brown, or green color
of the alkynyl complexes. The spectra of the aryl-alkynyl 17-
electron radical cations [2aþ], [2f]þ, [2g]þ, and [2i]þ were
characterized by a strong absorption band in the region
16 000-17 000 cm-1 (625-590 nm) responsible for the
To provide a more detailed assignment of the electronic
transitions that occur in these complexes, a TD-DFT study
was undertaken on 2A/[2A]þ. TD-DFT calculations were
performed for the first 100 excited states for 2A and [2A]þ,
and the effect of solvent (CH2Cl2) was included using
the polarizable continuum model. Calculated oscillator
strengths were then used to simulate35 spectra in the region
14 000-40 000 cm-1 (700-250 nm). Simulated spectra for
2A and [2A]þ are shown in Figures 11 and 12.
For the 18-electron complex 2A, both observed and
simulated spectra feature two significant absorption maxima
(33) Jiao, H.; Costuas, K.; Gladysz, J. A.; Halet, J.-F.; Guillemot, M.;
Toupet, L.; Paul, F.; Lapinte, C. J. Am. Chem. Soc. 2003, 125, 9511.
(34) Venkatesan, K.; Blacque, O.; Berke, H. Dalton Trans. 2007,
1091.
(35) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.
Chem. 2008, 29, 839.

Article
Organometallics, Vol. 29, No. 5, 2010 1271
radicals [MoX(dppe)(η-C7H7)]þ has been the focus of several
previous reports,14,36 and the DFT treatment, outlined be-
low, now permits a much clearer rationalization of these
results. Figure 15 illustrates the calculated unpaired spin
density distribution in [2A]þ together with Mulliken spin
densities. It is clear that unpaired spin density is located
principally at the Mo center in a metal-based orbital with a
significant dz2 contribution; this is consistent with the dis-
tinctly isotropic character of the experimental EPR spectra
and in agreement with previous predictions for these com-
plexes.37
Two new experimental investigations have been under-
taken in the current work. First, EPR spectra of a series of
17-electron alkynyl radicals [Mo(CtCR)(dppe)(η-C7H7)]þ,
selected to encompass a wide range of R substituents, were
recorded in fluid solution in dichloromethane at 243 K (on
an isolated sample, [2a][BF4], or by generation in situ on
addition of [FcH][PF6] to samples of 2d, 2f, 2g, and 2i). A
typical spectrum (as a second derivative) is presented in
Figure 16, and Table 9 summarizes experimental solution
isotropic g values and hyperfine coupling parameters
(confirmed by simulation); very little variation is observed
with change in the identity of R.
Second we have recorded the W-band, solid-state spec-
trum (Figure 17) of an isolated sample of [2a][BF4] and
obtained, for the first time, anisotropic g values for an
alkynyl complex of this type (g1 = 1.9992, g2 = 1.9958,
g3 = 1.9892; Δg = g1 - g3 = 0.010). These data, which
complement reports for the related systems [MoX(dppe)(η-
C7H7)]þ (X = F, NCO, NCS, CN) obtained at the Q-band,36
reveal very small g value anisotropy, in marked contrast to
complexes of the type [M(CtCR)(dppe)Cp*]þ (M = Fe or
Ru), which exhibit much larger g value anisotropy (R = Ph;
M = Fe, Δg = 0.489;10 M = Ru, Δg = 0.239,11 frozen
solution, 80 K).
An extended DFT investigation on the 17-electron radical
[2A]þ was undertaken, leading to calculation of the unpaired
spin density distribution as presented in Figure 15. For
calculations of EPR g tensors and hyperfine coupling ten-
sors, the Mo basis was changed to the all-electron TZVPP-
All-S2 basis, again from the Turbomole library.29 In the all-
electron calculations, s-type functions on all atoms were
uncontracted. Convergence of the EPR parameters was
checked by the inclusion of additional tight s functions (the
exponents being obtained in geometric progression) on all
atoms. Typically it was found that the inclusion of two
additional sets of s functions was sufficient, since any further
addition produced changes of less than 1 MHz (0.36 G) in the
computed hyperfine tensors. The g tensor of [2A]þ was
calculated using gauge-including atomic orbitals (GIAO)
and the B3LYP functional as implemented in the Gaussian
suite of programs.30 Although the g tensor is thought to show
only a weak gauge dependence, the GIAO calculations
reported here are rigorously gauge independent; however
in the implementation used the spin-orbit operator is ap-
proximated by a one-electron operator with a scaled effective
charge to account for the screened nuclear charge. The
calculated components of the g tensor obtained were g1 =
2.0078, g2 = 1.9996, g3 = 1.9792, giving giso = 1.9955
Figure 12. Simulated UV/vis spectrum of [2A]þ (TD-DFT,
B3LYP/SVP).
below 400 nm (25 000 cm-1). The dominant states that
contribute to these bands in 2A appear at 337 nm (29 700
cm-1) and 375 nm (26 670 cm-1), but there are close lying
adjacent states (with smaller but significant oscillator
strengths) that shift the band centers to lower wavelengths.
The state at 375 nm is composed of many orbital transitions,
but the principal component, accounting for 17% of the
wave function, is a transition from the HOMO to a high-
lying virtual orbital (LUMOþ10), which corresponds to the
transfer of electron density from the metal to the phenyl ring
of the CtCPh unit. The only other excitations that carry a
weight of >5% in the final wave function are a transition
from HOMO-2 to LUMO, with a weight of 7% (this
corresponds to a transfer of electron density from the Mo
to the P atoms of the dppe unit) and a transition (weight 6%)
from the HOMO to the phenyl rings of the dppe unit. The
state at 337 nm arises principally due to transfer of electron
density from the Mo(η-C7H7) unit to the dppe unit; there is
very little weight for any transitions corresponding to the
phenyl ring of the CtCPh unit. Finally, there is a band
centered at 546 nm (18 300 cm-1) responsible for the brown
color of the complex, 31% of which comes about from an
excitation from the HOMO (Figure 8c) to the LUMOþ1
(Figure 8a).
As noted above, the experimental UV/vis spectra of the
oxidized aryl-alkynyl complexes each feature an intense
absorption in the region 16 000-17 000 cm-1 (625-590
nm) (Figure 10). The simulated spectrum of [2A]þ
(Figure 12) shows a similar band centered on a transition
at 582 nm (2aþ, experimental value 592 nm). This state is
dominated by a single transition (weight 85%) that corre-
sponds to an excitation within the β-spin orbital manifold of
HOMO-2 f LUMO, but it should be noted that the spatial
form of the β-spin LUMO is effectively that of the SOMO
(R-spin). The orbitals involved are shown in Figure 13 and
suggest that this transition corresponds to a transfer from the
phenyl ring of the CtCPh unit to the Mo(η-C7H7)CtC-
moiety, thus supporting the experimental LMCT assignment
of the transition.
The band at 374 nm is dominated (30%) by an excitation
from HOMO-2 to LUMOþ1 within the R-spin orbital
manifold as shown in Figure 14 and appears to indicate
transfer from a Mo-C7H7 unit to the alkynyl ligand.
EPR Investigations. The well-resolved, isotropic character
of the X-band EPR solution spectra of the 17-electron
(36) Aston, G. M.; Badriya, S.; Farley, R. D.; Grime, R. W.; Ledger,
S. J.; Mabbs, F. E.; McInnes, E. J. L.; Morris, H. W.; Ricalton, A.;
Rowlands, C. C.; Wagner, K.; Whiteley, M. W. J. Chem. Soc., Dalton
Trans. 1999, 4379.
(37) Rieger, P. H. Coord. Chem. Rev. 1994, 135/136, 203.

1272 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
Figure 13. β-Spin orbital transition in [2A]þ that gives rise to the band calculated at 582 nm.
Figure 14. R-Spin orbital transition in [2A]þ that gives rise to the band calculated at 374 nm.
Figure 15. PBE0 spin density of [2A]þ shown as an isosurface of 0.004 au. For clarity the atoms of the dppe unit are shown as wireframe
only.
Table 9. EPR Data for the 17-Electron Radicals
[Mo(CtCR)(dppe)(η-C7H7)]þa
Aiso(Mo)
aiso
(
31P)
aiso(1H)
giso
[2a]þ
[2d]þ
[2f]þ
[2g]þ
[2i]þ
31.3b
31.3
31.3
31.3
31.3
22.6
23.3
22.1
22.1
22.8
4.3
4.5
4.2
4.3
4.3
1.996
1.996
1.995
1.995
1.996
a X-band solution spectra in CH2Cl2 at 243 K; hyperfine couplings in
gauss. b Aiso(Mo) value differs slightly from that reported (34.8) in ref 36.
The revised value gives an enhanced fit to the positions of the Mo
satellites.
Figure 16. Fluid solution (CH2Cl2, 243 K), second-derivative
X-band EPR spectrum of [Mo(CtCC6H4-4-Me)(dppe)(η-
C7H7)][PF6], [2g][PF6].
values obtained were g1 = 2.0052, g2 = 1.9997, g3 = 1.9881,
giving giso = 1.9977. The two schemes used give a difference
of some 2200 ppm in the value of giso. Table 10 compares the
calculated g values with those from experiment and indicates
that the use of the mean field spin-orbit operator is superior
in predicting the components of the g tensor even though the
isotropic value is less well reproduced.
The evaluation of the hyperfine tensors requires some
careful comment. For the Mo atom, in addition to the
isotropic Fermi contact (FC) term, it is important to include
the spin-orbit contributions and relativistic effects. We
performed calculations with the B3LYP and PBE0 func-
tionals. We found the PBE0 functional to be slightly superior,
(experimental: g1 = 1.9992, g2 = 1.9958, g3 = 1.9892, giso
=
1.9947). To assess the effect of the choice of spin-orbit
operator, we performed additional calculations of the g
tensor using a mean field spin-orbit operator, which in-
cludes the effect of the two-electron spin-orbit integrals in
an averaged way. Center of charge coordinates (which
should minimize the gauge dependence) were used without
the GIAO, as implemented in the ORCA program.38 The
(38) Neese, F. ORCA, An Ab Initio, Density Functional, and Semi-
empirical Program Package, version 2.6, Rev. 35; Univ. Bonn: Bonn, 2008.

Article
Organometallics, Vol. 29, No. 5, 2010 1273
Figure 17. Solid-state
W-band
EPR
spectrum
of
[Mo(CtCPh)(dppe)(η-C7H7)][BF4], [2a][BF4].
Table 10. Errors in the Components of g for [2A]þ Obtained with
the Two Computational Procedures Used
GIAO þ
center of charge
coordinates þ mean field
spin-orbit operator
scaled Z one-electron
spin-orbit operator
Figure 18. Calculated isotropic hyperfine coupling constants
(G) of [2A]þ.
Δg1
Δg2
Δg3
Δgiso
0.0086
0.0038
0.0100
0.0008
0.0060
0.0039
-0.0011
0.0030
in the small change in the alkynyl stretching frequency resulting
from one-electron oxidation, the relatively high stability of the
17-electron radical cations, [Mo(CtCR)(dppe)(η-C7H7)]þ, and
the excellent resolution of hyperfine coupling observed in the
solution EPR spectra. These experimental observations can be
rationalized by electronic structure calculations, which establish
a significant metal dz2 contribution both to the HOMO of
18-electron [Mo(CtCR)(dppe)(η-C7H7)] and to the location
of unpaired spin density in the corresponding radical. The
electronic structure of [Mo(CtCR)(dppe)(η-C7H7)], in which
the metal dz2-based HOMO is energetically well separated from
HOMO-1, is in marked contrast to the analogous group 8
complexes [Ru(CtCR)(dppe)Cp*], for which the HOMO and
HOMO-1 are almost degenerate. These differences in electronic
structure can be attributed to a diversity in metal-ring bonding.
Stabilization of the ring molecular orbitals with increase in ring
size facilitates a strong δ-interaction between molybdenum and
the C7H7 ligand, an effect not present in the RuCp* system. The
perturbation introduced by metal-ring δ-bonding leads to a
reordering of the metal d orbital manifold, resulting in very
different electronic structures for the two series of complexes
[Mo(CtCR)(dppe)(η-C7H7)] and [Ru(CtCR)(dppe)Cp*],
which, on first inspection, might appear to be closely analogous
systems of the 4d metals Mo and Ru.
overall, and the hyperfine couplings we report refer exclu-
sively to this functional. The experimentally assigned iso-
tropic hyperfine splitting for Mo (Table 9) is 31.3 G.
Calculating just the FC term gives 22.4 G; adding the
spin-orbit contribution of 3.2 G gives Aiso(Mo) = 25.6 G,
which is our best nonrelativistic value. Adding relativistic
effects via the ZORA approach changes the FC term to 26.8
G. Fully uncontracting all basis functions changes the FC
term only a small amount to 27.4 G and the spin-orbit term
changes to 3.0 G, indicating that our approach of uncon-
tracting only the s functions is valid. Our best estimate
becomes Aiso(Mo) = 30.4 G. While relativistic and
spin-orbit contributions are important for Aiso(Mo), they
contribute negligibly to the hyperfine interactions of the
other atoms. For example, taking the heaviest of the remain-
ing atom types, P, we find that the spin-orbit contribution to
aiso(P) is <0.1 G and the inclusion of relativistic effects
changes aiso(P) by <0.7 G. Thus in discussing the hyperfine
couplings of all atoms except Mo, our values refer to the FC
term alone evaluated in a nonrelativistic PBE0 calculation.
Figure 18 summarizes the principal hyperfine couplings. In
the X-band solution EPR experiment the hyperfine cou-
plings for the P atoms of dppe are averaged, whereas in
the calculations the static structure gives slightly inequiva-
lent P atom environments. The averaged calculated value is
aiso(P) = 23.8 G (experiment aiso(P) = 22.6 G). Similarly, the
protons of the C7H7 ring give an averaged aiso(H) = 3.9 G
(experiment aiso(H) = 4.3 G). The remaining aiso values
shown in Figure 18 are not resolved under current experi-
mental conditions. The values we are able to compare with
experiment are reproduced within about 1 G, and we expect
that the remaining aiso have a similar reliability.
Experimental Section
General Procedures. The preparation, purification, and reac-
tions of the complexes described were carried out under dry
nitrogen using standard Schlenk techniques. Methanol was best
available commercial grade or dried on an Innovative Technol-
ogies SPS-400 system, and deoxygenated thoroughly by spar-
ging with dry nitrogen before use. The complex [MoBr(dppe)(η-
C7H7)] 0.5CH2Cl2 was prepared by a published procedure.17
3
The alkynes HCtCFc, HCtCC6H4-4-CHO, HCtCC6H4-4-
OMe, and HCtCC6H4-4-CO2Me were prepared by standard
methods39,40 or purchased from commercial sources (HCtCPh,
Conclusions
The alkynyl complexes [Mo(CtCR)(dppe)(η-C7H7)]nþ
(n = 0 or 1) feature occupied frontier orbitals that are
essentially metal-localized. Experimentally this is manifest
(39) Polin, J.; Schottenberger, H. Org. Synth. 1996, 73, 262.
(40) Lavastre, O; Ollivier, L.; Dixneuf, P. H.; Sinbandhit, S. Tetra-
hedron 1996, 52, 5495.

1274 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
HCꢁCBut, HCtCCO2Me, HCtCC6H4-4-NH2, HCtCC6H4-
4-Me). NMR spectra were recorded from CD2Cl2 solutions,
unless otherwise stated, containing trace amounts of CoCp2 to
prevent accumulation of MoI species by aerial oxidation, on
Varian Inova 300 or 400 or Bruker DRX-400 spectrometers at
room temperature and referenced against solvent resonances
(1H, 13C) or external H3PO4 (31P). Infrared spectra were ob-
tained on Perkin-Elmer FT RX1 or Nicolet Avatar spectro-
meters. Mass spectra were recorded using Thermo Quest
Finnigan Trace GC/MS or Thermo Electron Finnigan LTQ
FT mass spectrometers. MALDI mass spectra were recorded
using a Micromass/Waters TOF Spec 2E instrument. Micro-
analyses were conducted by the staff of the Microanalytical
Services of the School of Chemistry, University of Manchester,
and the Department of Chemistry, Durham University. Cyclic
voltammograms were recorded (ν = 100 mV s-1) from 0.1 M
NBu4PF6, CH2Cl2 solutions ca. 1 ꢀ 10-4 M in analyte using a
gastight single-compartment three-electrode cell equipped with
a Pt disk working electrode, Pt wire counter electrode, and Pt
wire pseudoreference electrode, and data collected on an Auto-
lab PG-STAT 30 potentiostat. The working electrode was
polished with alumina paste before each scan. All redox poten-
tials are reported with reference to an internal standard of the
ferrocene/ferrocenium couple (FcH/FcHþ = 0.00 V). Electro-
chemical measurements at subambient temperatures were per-
formed in the same cell, cooled by immersion in an external
bath. UV/vis/NIR and IR spectroelectrochemical experiments
were performed at room temperature with an airtight OTTLE
cell equipped with Pt minigrid working and counter electrodes, a
Ag wire reference electrode, and CaF2 windows25 using either a
Nicolet Avatar spectrometer or a Perkin-Elmer Lambda 900
spectrophotometer. EPR experiments were conducted on a
Bruker BioSPin EMX microspectrometer at X-band (9 GHz);
spectra were recorded at 243 K and are the average of 16 scans.
Spectral analysis and simulation were carried out using Bruker
WinEPR software (Bruker Biospin Ltd.).
effervescence had ceased, and the solvent was removed in vacuo.
The solid residue was extracted with CH2Cl2 (5 cm3), the
extracts were filtered, and the filtrate was taken to dryness in
vacuo to give a dark-colored solid, which was washed with Et2O
(3 ꢀ 5 cm3) and hexane (3 ꢀ 5 cm3) or until the washings were
colorless and dried to afford 2c as a free-flowing brown powder;
yield: 10 mg (9%). 1H NMR: δ 2.17 (m, 2H, CH2), 2.64 (m, 2H,
CH2), 3.29 (t, JHH = 2 Hz, 2H, Cpβ), 3.60 (s, 5H, Cp), 3.69 (t,
JHH = 2 Hz, 2H, CpR), 4.71 (br, 7H, C7H7), 7.29-7.89 (m,
20H, Ph). 31P{1H} NMR: δ 65.6 (s, dppe). MALDI-MS (m/z):
796, [M]þ. Anal. Calcd (%) for C45H40FeMoP2: C, 68.0; H, 5.0.
Found: C, 68.4; H, 5.2.
Preparation of [Mo(CtCCO2Me)(dppe)(η-C7H7)] (2d). This
was prepared and purified in an identical fashion to 2a, using
methyl propiolate (288 mg, 3.43 mmol) in place of HCtCPh,
[MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (808 mg, 1.14 mmol), and
3
KOBut (211 mg, 1.88 mmol). The product was obtained as a
purple-red solid; yield: 240 mg (31%). 1H NMR: δ 1.89 (m, 2H,
CH2), 2.35 (m, 2H, CH2), 3.18 (s, 3H, Me), 4.71 (br, 7H, C7H7),
7.19-7.67 (m, 20H, Ph). 31P{1H} NMR (CDCl3): δ 62.2 (s,
dppe). IR (CH2Cl2): ν(CtC) 2020, ν(CdO) 1657 cm-1. MAL-
DI-MS (m/z): 669, [M]þ. Anal. Calcd (%) for C37H34MoP2: C,
66.5; H, 5.1. Found: C, 66.2; H, 5.1.
Preparation of [Mo(CtCC6H4-4-NH2)(dppe)(η-C7H7)] (2e).
The reaction between [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (100
3
mg 0.14 mmol) and HCtCC6H4-4-NH2 (27 mg, 0.23 mmol) was
carried out in a manner similar to that described for 2c, afford-
ing 2e as a deep green solid; yield 40 mg (38%). 1H NMR: δ 2.06
(m, 2H, CH2), 2.48 (m, 2H, CH2); 3.40 (s, 2H, NH2), 4.75 (t,
J
PH = 2 Hz, 7H, C7H7), 5.87 (d, (AB), 2H, JHH ≈ 8 Hz, C6H4),
6.20 (d, (AB) 2H, JHH ≈ 8 Hz, C6H4), 7.30-7.81 (m, 20H, Ph).
31P{1H} NMR: δ 65.7 (s, dppe). IR (CH2Cl2): ν(CtC) 2053.
ES(þ)-MS (m/z): 701 [M]þ. HR ES(þ)-MS (m/z): 703.14469
(MoNP2C41H37 requires 703.14552).
Preparation of [Mo(CtCC6H4-4-OMe)(dppe)(η-C7H7)] (2f).
A solution of Me3SiCtCC6H4-4-OMe (29 mg, 0.23 mmol) and
KF (26 mg, 0.452 mmol) in methanol (20 cm3) was treated with
Preparations. Preparation of [Mo(CtCPh)(dppe)(η-C7H7)]
[MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (100 mg, 0.14 mmol) and
(2a). A mixture of [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (813 mg,
3
3
1.15 mmol), HCtCPh (587 mg, 5.75 mmol), and KOBut (387
mg, 3.45 mmol) in methanol (50 cm3) was heated at reflux for
3 h. The resulting deep brown precipitate was collected, washed
with hexane, and dried in vacuo. The solid was dissolved in
CH2Cl2 (ca. 5 cm3), loaded onto a hexane/alumina column, and
eluted with CH2Cl2/hexane/acetone (45:45:10 v/v/v). The major
brown band was collected and evaporated to dryness to give a
deep brown solid, which was reprecipitated from CH2Cl2/
NaPF6 (25 mg, 0.151 mmol). The resulting solution was heated
at reflux point for 90 min before the solvent was removed. The
residue was washed with hexane (2 ꢀ 10 cm3) and Et2O (2 ꢀ 10
cm3) or until the washings were clear. The remaining solid was
extracted with CH2Cl2 (5 cm3) and cannula filtered into stirred
hexane (15 cm3) to give 2f as a dark British racing green colored
precipitate; yield: 40 mg (37%). 1H NMR: δ 2.03 (m, 2H, CH2),
2.50 (m, 2H, CH2), 3.63 (s, 3H, OMe), 4.78 (br, 7H, C7H7), 5.96
(d, (AB), 2H, JHH ≈ 9 Hz, C6H4), 6.40 (d, (AB), 2H, JHH ≈ 9 Hz,
C6H4), 7.29-7.83 (m, 20 H, Ph). 31P{1H} NMR: δ 64.1 (s, dppe).
IR (CH2Cl2): ν(CtC) 2051; ν(Ar) 1604 cm-1. ES(þ)-MS (m/z):
716, [M]þ. HR ES(þ)-MS (m/z): 718.1446 (MoOP2C42H38
requires 718.1466).
1
hexane; yield: 425 mg (54%). H NMR: δ 2.11 (m, 2H, CH2),
2.49 (m, 2H, CH2), 4.82 (t, 7H, JHP = 2.1 Hz, C7H7), 6.03 (d, 2H,
JHH 7 Hz, CtCPho), 6.74 (m, 1H, CtCPhp), 6.84 (m, 2H,
CtCPhm), 7.33-7.85 (m, 20H, Ph). 31P{1H} NMR (CDCl3): δ
64.6 (s, dppe). IR (CH2Cl2) ν(CtC) 2045 cm-1. MALDI-MS
(m/z): 687, [M]þ; 586 [(M - CtCPh)]þ. Anal. Calcd (%) for
C41H36MoP2: C, 71.7; H, 5.3. Found: C, 71.4; H, 5.2.
Preparation of [Mo(CtCC6H4-4-Me)(dppe)(η-C7H7)] (2g). A
mixture of [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (1.46 g, 2.06
3
Preparation of [Mo(CtCBut)(dppe)(η-C7H7)] (2b). This was
prepared and purified in an identical fashion to 2a, using
mmol), HCtCC6H4-4-Me (720 mg, 6.20 mmol), and KOBut
(485 mg, 4.32 mmol) in methanol (50 cm3) was heated at reflux
for 2 h. The reaction mixture was cooled and reduced in volume
to ca. 20 cm3. The deep brown precipitate was collected, washed
with hexane, and dried in vacuo; yield: 1.123 g (78%). 1H NMR:
δ 1.97 (m, 2H, CH2); 2.04 (s, 3H, C6H4CH3); 2.38 (m, 2H, CH2);
4.70 (br, 7H, C7H7); 5.81 (d, (AB), 2H, JHH ≈ 8 Hz, C6H4), 6.54
(d, (AB), 2H, JHH ≈ 8 Hz, C6H4); 7.20-7.71 (m, 20H, Ph).
31P{1H} NMR (CDCl3): δ 64.6 (s, dppe). IR (CH2Cl2) ν(CtC)
2048 cm-1; MALDI-MS (m/z): 703, [M]þ. Anal. Calcd (%) for
C42H38MoP2: C, 72.0; H, 5.5. Found: C, 72.0; H, 5.4.
HCtCBut (870 mg, 10.60 mmol), [MoBr(dppe)(η-C7H7)]
3
0.5CH2Cl2 (1.50 g, 2.12 mmol), and KOBut (0.71 g, 6.36 mmol).
The product was obtained as a dark brown solid; yield: 430 mg
(30%). 1H NMR: δ 0.40 (s, 9H, But), 2.20 (m, 2H, CH2), 2.60 (m,
2H, CH2), 4.66 (s, 7H, C7H7), 7.25-7.82 (m, 20H, Ph). 31P{1H}
NMR (CDCl3): δ 66.2 (s, dppe). IR (CH2Cl2): ν(CtC) 2057
cm-1. MALDI-MS (m/z): 668, [M]þ. Anal. Calcd (%) for
C39H40MoP2: C, 70.3; H, 6.1. Found: C, 70.3; H, 6.0.
Preparation of [Mo(CtCFc)(dppe)(η-C7H7)] (2c). A warm
(ca. 50 °C) solution of [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (100
Preparation of [Mo(CtCC6H4-4-CHO)(dppe)(η-C7H7)] (2h).
3
mg 0.14 mmol) in methanol (20 cm3) was treated with HCtCFc
(38 mg, 0.18 mmol), and the resulting mixture heated at reflux
point for 90 min. The solution was allowed to cool to room
temperature before addition of Na metal (100 mg, 4 mmol). The
solution was allowed to stir for 20 min, after which time
The reaction between [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2 (100
3
mg, 0.14 mmol), KF (0.013 g, 0.225 mmol), and SiMe3Ct
CC6H4-4-CHO (0.036 g, 0.180 mmol) in methanol (20 cm3)
was carried out in a manner similar to that described for 2f,
affording 2h as a dark purple solid; yield: 73 mg (68%). 1H

Article
Organometallics, Vol. 29, No. 5, 2010 1275
Table 11. Crystal Data and Refinement Parameters for Complexes 2b, 2d, 2g, 2h, 2i, and [2b]þ
2b
C39H40MoP2
2d
C37H34MoO2P2
2g
C42H38MoP2
2h
C42H36 MoOP2
2i
C43H38MoO2P2
[2b]þ
C39.38H41MoP2BF4Cl0.75
mass
666.59
293(2)
triclinic
P1
9.4042(5)
10.0595(5)
17.0911(9)
89.5460(10)
88.0610(10)
82.4370(10)
1601.85(14); 2
5.36
668.52
100(2)
700.60
100(2)
orthorhombic
Pbca
9.3678(7)
17.8334(14)
39.472(3)
90
90
90
714.59
120(2)
triclinic
P1
9.7255(3)
10.1112(3)
18.3099(5)
79.663(10)
83.21(2)
70.905(10)
1670.40(8); 2
5.22
744.61
120(2)
monoclinic
P21/c
21.834(2)
9.4090(10)
17.5983(19)
90
109.75(3)
90
785.50
100(2)
orthorhombic
Pbca
19.117(2)
27.208(4)
55.768(8)
90
90
90
temperature (K)
cryst syst
space group
monoclinic
P2(1)/n
8.4986(13)
18.382(3)
19.712(3)
90
95.361(2)
90
3066.1(18); 4
5.65
˚
a (A)
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
3
˚
V (A ); Z
6594.2(9); 8
5.25
3402.7(6); 4
4
29006(7); 32
5.54
absorp coeff (cm-1
)
θ range (deg)
limiting indices (h,k,l)
2.04-26.39
-11/11;
-12/12;
-21/21
12 830
6462
0.0243
0.0614
98.5
1.52-26.35
-10/10;
-22/22;
-24/24
24 108
6257
0.0276
0.0708
100
2.06-28.32
-12/12;
-22/23;
-51/52
54 369
1.13-30.00
-13/13;
-14/14;
-25/25
22 417
9720
0.0344
0.0882
99.6
2.32-27.50
-18/28;
-12/12;
-22/2
1.29-21.97
-20/19;
27/28;
-56/58
110 282
17 696
0.0601
0.0998
99.9
total reflns
indep reflns, I > 2σ(I)
R1
wR2
completeness to θ (%)
21 624
8042
7802
0.0585
0.0659
100
0.0699
0.1522
99.8
NMR: δ 2.13 (m, 2H, CH2), 2.42 (m, 2H, CH2), 4.83 (br, 7H,
C7H7), 5.94 (d, (AB), 2H, JHH ≈ 8 Hz, C6H4), 7.31 (d, (AB), 2H,
JHH ≈ 8 Hz, C6H4), 7.29-7.84 (m, 20H, Ph), 9.65 (s, 1H, CHO).
31P{1H} NMR: δ 63.6 (s, dppe). IR (CH2Cl2): ν(CtC) 2023;
ν(CdO) 1680 cm-1. HR ES(þ)-MS (m/z): 716.12860 (MoOP2-
C42H36 requires 716.12953).
SHELXL-9742 for the computing structure refinement, and an
absorption correction was applied with the aid of the SADABS
program.43 Diffraction data for 2h and 2i were collected on a Bruker
3-circle diffractometer with a SMART 6K area detector; reflection
intensities were integrated using the SAINT V6.45 program, and a
numerical absorption correction was applied. All structures were
solved by direct methods with refinement by full-matrix least-
squares based on F2 against all reflections. With the exception of
[2b][BF4], all non-hydrogen atoms were refined anisotropically and
hydrogen atoms were included in calculated positions. In the case of
[2b][BF4] the asymmetric unit contains four crystallographically
independent pairs of ions and 1.5 molecules of CH2Cl2, and the
Preparation of [Mo(CtCC6H4-4-CO2Me)(dppe)(η-C7H7)]
(2i). The reaction between [MoBr(dppe)(η-C7H7)] 0.5CH2Cl2
3
(500 mg, 0.70 mmol) and HCtCC6H4-4-CO2Me (300 mg, 1.88
mmol) in methanol (100 cm3) was carried out in a manner
similar to that described for 2c, with deprotonation being
effected by addition of Na (250 mg, 10 mmol), affording 2i as
1
˚
a black solid; yield: 280 mg (50%). H NMR: δ 2.14 (m, 2H,
crystal diffracted very weakly so the data were cut at 0.95 A
resolution. One phenyl ring was constrained to a regular hexagon
with restraints also applied to some geometric parameters. The key
CH2), 2.47 (m, 2H, CH2), 3.80 (s, 3H, Me), 4.85 (br, 7H, C7H7),
5.37 (d, (AB), 2H, JHH ≈ 5 Hz, C6H4), 7.50 (d, (AB), 2H, JHH ≈ 5
Hz, C6H4), 7.35-7.85 (m, 20H, Ph). 31P{1H} NMR: δ 64.9 (s,
dppe). IR (CH2Cl2): ν(CtC) 2027, 1991; ν(CdO) 1702. ES(þ)-
MS (m/z): 744, [M]þ. HR ES(þ)-MS (m/z): 746.13946 (MoO2-
P2C43H38 requires 746.14150).
Mo-CR distances (A) in the four independent cations [2b]þ in the
˚
lattice are 2.070(11), 2.076(11), 2.050(13), and 2.067(13).
Electronic Structure Calculations. Electronic structure calcula-
tions were carried out on 2A and the corresponding cation, 2Aþ.
Starting from the crystallographic structure of 2A, full geometry
optimizations were performed on 2A and 2Aþ using the Def2-SVP
basis obtained from the Turbomole library.29 ForH,C,andPatoms,
this constitutes an all-electron split valence plus polarization basis,
while for Mo an effective core potential is applied to account for 28
core electrons (K, L, and M shells) with a split valence orbital set
including a set of f-type polarization functions. This basis was also
used to obtain IR frequencies and the UV/vis spectra (via time-
dependent density functional theory). For calculations of EPR g
tensors and hyperfine coupling tensors, the Mo basis was changed to
the all-electron TZVPP-All-S2 basis, again from the Turbomole
library.29 In the all-electron calculations, s-type functions on all
atoms were uncontracted. Convergence of the EPR parameters
was checked by the inclusion of additional tight s functions (the
exponents being obtained in geometric progression) on all atoms.
Typically it was found that the inclusion of two additional sets of s
functions was sufficient, since any further addition produced changes
of less than 1 MHz (0.36 G) in the computed hyperfine tensors.
DFT calculations were performed in the generalized gradient
approximation (GGA) with the BP86 exchange-correlation
functional.44 For the TD-DFT, g tensor, and hyperfine tensor
Crystallography. X-ray crystal structures of [Mo(CtCBut)-
(dppe)(η-C7H7)], 2b, [Mo(CtCCO2Me)(dppe)(η-C7H7)], 2d,
[Mo(CtC-C6H4-4-Me)(dppe)(η-C7H7)], 2g, [Mo(CtCC6H4-
4-CHO)(dppe)(η-C7H7)], 2h, [Mo(CtCC6H4-4-CO2Me)(dppe)-
(η-C7H7)], 2i, and [Mo(CtCBut)(dppe)(η-C7H7)][BF4], [2b]þ,
were obtained.
The majority of details of the structure analyses carried out on
complexes 2b, 2d, 2g, 2h, 2i, and [2b]þ are given in Table 11.
Single crystals of the complexes were obtained as follows: 2b:
vapor diffusion of pentane into a CH2Cl2 solution of the
complex to give black, irregular shaped crystals; 2d: vapor
diffusion of pentane into a CH2Cl2 solution of the complex to
give red-brown plates; 2g: vapor diffusion of pentane into a
toluene solution of the complex to give brown plates; 2h: vapor
diffusion of diethyl ether into a CH2Cl2 solution of the complex
to give deep purple plates; 2i: vapor diffusion of diethyl ether
into a CH2Cl2 solution of the complex to give green plates; and
[2b]þ: vapor diffusion of diethyl ether into a CH2Cl2 solution of
the complex to give orange-brown plates. In all cases the X-ray
data were collected using graphite-monochromated Mo KR
˚
(λ = 0.7103 A) radiation. With the exception of 2h and 2i, data
collection, cell refinement, and data reduction were carried out
with Bruker SMART and Bruker SAINT software; SHELXS-
9741 was employed for the computing structure solution and
(42) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure
€
€
Refinement; Universitat Gottingen: Germany, 1997.
(43) Sheldrick, G. M. SADABS, an Empirical Absorption Corrections
€
€
Program; Universitat Gottingen: Germany, 1997.
(44) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098. (b) Perdew, J. P.
Phys. Rev. B 1986, 33, 8822.
(41) Sheldrick, G. M. SHELXS-97, Program for Crystal Structure
€
€
Solution; Universitat Gottingen, Germany, 1997.

1276 Organometallics, Vol. 29, No. 5, 2010
Brown et al.
calculations, the B3LYP hybrid exchange-correlation func-
tional45 was used. Hybrid functionals are generally found to
give better predictions for UV/vis spectra and EPR parameters.
For the latter, the PBE0 functional46 was also used since this
functional was designed for use in calculations of magnetic
properties.
The IR frequencies were calculated from analytic second
derivatives using the Gaussian suite of programs.30 For g tensor
calculations the gauge-including atomic orbitals (GIAO), as
implemented in ref 30, were employed with a one-electron
spin-orbit operator with a scaled charge. The ORCA pro-
gram38 was also used for g tensor calculation employing a mean
field spin-orbit operator. In this work we are only interested in
isotropic values of the hyperfine coupling constants. Accord-
ingly, the Fermi contact term of the hyperfine interaction was
evaluated and also the spin-orbit contribution. The ORCA
code38 was used for the latter. The spin-spin dipole contribu-
tion is a traceless tensor and does not affect the isotropic
hyperfine couplings. The influence of relativistic effects on the
Mo hyperfine interaction was investigated within the zero-order
regular approximation (ZORA),47 including the picture change
transformation necessary for the correct treatment of the FC
term.48
Acknowledgment. We thank the EPSRC for support of
this work through grants EP/E025544/1 and EP/
E02582X/1.
Supporting Information Available: This material is available
free of charge via the Internet at http://pubs.acs.org.
(47) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1993, 99, 4597.
(48) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem.
(45) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(46) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158.
Phys. 1998, 108, 4783.