Unusual electronic structure of first row transition metal complexes featuring redox-active dipyrromethane ligands

Article abstract of DOI:10.1021/ja903997a

Transition metal complexes (Mn → Zn) of the dipyrromethane ligand, 1,9-dimesityl-5,5-dimethyldipyrromethane (dpm), have been prepared. Arylation of the dpm ligand α to the pyrrolic nitrogen donors limits the accessibility of the pyrrole π-electrons for tr

Full text of DOI:10.1021/ja903997a

Published on Web 09/15/2009
Unusual Electronic Structure of First Row Transition Metal
Complexes Featuring Redox-Active Dipyrromethane Ligands
Evan R. King and Theodore A. Betley*
Department of Chemistry and Chemical Biology, HarVard UniVersity, 12 Oxford Street, 306E
Mallinckrodt, Cambridge, Massachusetts 02138
Received May 16, 2009; E-mail: betley@chemistry.harvard.edu
Abstract: Transition metal complexes (Mn f Zn) of the dipyrromethane ligand, 1,9-dimesityl-5,5-
dimethyldipyrromethane (dpm), have been prepared. Arylation of the dpm ligand R to the pyrrolic nitrogen
donors limits the accessibility of the pyrrole π-electrons for transition metal coordination, instead forcing
η1,η1 coordination to the divalent metal series as revealed by X-ray diffraction studies. Structural and magnetic
characterization (SQUID, EPR) of the bis-pyridine adducts of (dpm)MnII(py)2, (dpm)FeII(py)2, and (dpm)-
CoII(py)2 reveal each divalent ion to be high-spin and pseudotetrahedral in the solid state, whereas the
(dpm)NiII(py)2 is low-spin and adopts a square-planar geometry. Differential pulse voltammetry on the
(dpm)MII(py)2 series reveals a common two-electron oxidation pathway that is entirely ligand-based, invariant
to the divalent metal-bound, its geometry or spin state within the dpm framework. This latter observation
indicates that fully populated ligand-based orbitals from the dpm construct lie above partially filled metal
3d orbitals without intramolecular redox chemistry or spin-state tautomerism occurring. DFT analysis on
this family of complexes corroborates this electronic structure assignment, revealing that the highest lying
molecular orbitals are completely ligand-based. Chemical oxidation of the deprotonated dpm framework
results in the four-electron oxidation of the dipyrrolide framework, although this oxidation product was not
observed either in the electrochemical or chemical oxidation of the (dpm)MII(py)2 complexes.
1. Introduction
ligands have most notably enriched the field of organometallic
research with impacts ranging from bestowing multielectron
reactivity to d0 metals,7 facilitating electronic interplay between
metal and ligand during catalytic transformations,6 and enabling
dual-site reactivity during reaction sequences.8
Our own studies have focused on the coordination chemistry
of dipyrromethane9,10 and dipyrromethene11 ligand frameworks
as analogues to porphyrinogen and porphyrin scaffolds, respec-
tively. The pyrrole aromaticity both attenuates the binding N
π-basicity and makes them good candidates for ligand-centered
redox activity by virtue of the high-lying π orbitals.12 This
Multielectron redox activity has long been associated with
transition metal complexes with two or more accessible oxida-
tion state changes (e.g., Mn f Mn+1 f Mn+2). When the ligands
that bind the transition metal ions can themselves operate as
electron (or hole) reservoirs, the notion of accessible molecular
redox states expands beyond that dictated by the metal’s
d-orbital electron configuration.1 Redox activity has been
attributed to a variety of ligand platforms (e.g., catechols,2
dithiolates,3 phenolates,4 porphyrins,5 bisimino-pyridines,6 amido-
phenolates,7 etc.) via their observed coordination chemistry and
spectrochemical properties. The redox properties of the ligand
platforms are being increasingly parlayed into driving new
stoichiometric and catalytic reaction sequences. Redox-active
(7) (a) Herebian, D.; Bothe, E.; Bill, E.; Weyhermuller, T.; Wieghardt,
K. J. Am. Chem. Soc. 2001, 123, 10012–10023. (b) Blackmore, K. J.;
Ziller, J. W.; Heyduk, A. F. Inorg. Chem. 2005, 44, 5559–5561. (c)
Haneline, M. R.; Heyduk, A. F. J. Am. Chem. Soc. 2006, 128, 8410–
8411. (d) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. J. Am.
Chem. Soc. 2008, 130, 2728–2730.
(1) Chaudhuri, P.; Verani, C. N.; Bill, E.; Bothe, E.; Weyhermu¨ller, T.;
Wieghardt, K. J. Am. Chem. Soc. 2001, 123, 2213–2223.
(2) Masui, H.; Lever, A. B. P.; Auburn, P. R. Inorg. Chem. 1991, 30,
2402–2410.
(8) (a) Gunanathan, C.; Milstein, D. Angew. Chem., Int. Ed. 2008, 47,
8661–8664. (b) Gunanathan, C.; Shimon, L. J. W.; Milstein, D. J. Am.
Chem. Soc. 2009, 131, 3146–3147. (c) Kohl, S. W.; Weiner, L.;
Schwartsburd, L.; Konstantinovski, L.; Shimon, L. J. W.; Ben-David,
Y.; Iron, M. A.; Milstein, D. Science 2009, 324, 74–77.
(9) (a) Jones, D. J.; Gibson, V. C. Heterocycles 2006, 68, 1121–1127.
(b) Swartz, D. L.; Odom, A. L. Organometallics 2006, 25, 6125–
6133.
(3) (a) Schrauzer, G. N.; Mayweg, V. J. Am. Chem. Soc. 1962, 84, 3221.
(b) Stiefel, E. I.; Waters, J. H.; Billig, E.; Gray, H. B. J. Am. Chem.
Soc. 1965, 87, 3016–3017.
(4) (a) Hockertz, J.; Steenken, S.; Wieghardt, K.; Hildebrandt, P. J. Am.
Chem. Soc. 1993, 115, 11222–11230. (b) Chaudhuri, P.; Hess, M.;
Muller, J.; Hildenbrand, K.; Bill, E.; Weyhermuller, T.; Wieghardt,
K. J. Am. Chem. Soc. 1999, 121, 9599–9610.
(10) (a) Novak, A.; Blake, A. J.; Wilson, C.; Love, J. B. Chem. Commun.
2002, 2796–2797. (b) Love, J. B.; Salyer, P. A.; Bailey, A. S.; Wilson,
C.; Blake, A. J.; Davies, E. S.; Evans, D. J. Chem. Commun. 2003,
1390–1391.
(5) Kadish, K. M., Smith, K. M., Guilard, R., Eds. The Porphyrin
Handbook; Academic Press: San Diego, CA, 2003; Vols. 15-20.
(6) (a) Bouwkamp, M. W.; Bowman, A. C.; Lobkovsky, E.; Chirik, P. J.
J. Am. Chem. Soc. 2006, 128, 13340–13341. (b) Bart, S. C.; Chlopek,
K.; Bill, E.; Bouwkamp, M. W.; Lobkovsky, E.; Neese, F.; Wieghardt,
K. J. Am. Chem. Soc. 2006, 128, 13901–13912. (c) Bart, S. C.;
Lobkovsky, E.; Bill, E.; Wieghardt, K.; Chirik, P. J. Inorg. Chem.
2007, 46, 7055–7063.
(11) King, E. R.; Betley, T. A. Inorg. Chem. 2009, 48, 2361–2363.
(12) (a) Floriani, C.; Floriani-Moro, R. In The Porphyrin Handbook; Kadish,
K. M., Smith, K. M., Guilard, R., Eds.;Academic Press: San Diego,
CA, 2000; Vol. 3, pp 405-420. (b) Bachmann, J.; Nocera, D. G. J. Am.
Chem. Soc. 2004, 126, 2829–2837. (c) Bachmann, J.; Nocera, D. G.
J. Am. Chem. Soc. 2005, 127, 4730–4743.
9
14374 J. AM. CHEM. SOC. 2009, 131, 14374–14380
10.1021/ja903997a CCC: $40.75  2009 American Chemical Society

Redox-Active Ligands of (dpm)MII(py)2 Complexes
A R T I C L E S
Figure 1. Solid-state molecular structures for complexes 3, 4, 5, and 7 with the thermal ellipsoids set at the 50% probability level (hydrogen atoms and
minor structural disorder in 3 and 4 are omitted for clarity).
Figure 2. Solid-state molecular structures for complexes 6, 8, and 9 with the thermal ellipsoids set at the 50% probability level (hydrogen atoms, one
solvent molecule for 6, and two solvent molecules for 9 are omitted for clarity). The Li+ in 8 is coordinated to an adjacent molecule’s pyrrole ring.
contribution reports the synthesis and characterization of a series
of divalent metal complexes of a dipyrromethane ligand. The
complexes display surprisingly uniform electrochemical behav-
ior despite differences in metal valence electron configuration
as assessed by magnetic and spectroscopic methods, suggesting
an interesting electronic structure which may involve intramo-
lecular redox between the ligand and metal. Specifically, we
consider three possible scenarios: (1) high-energy pyrrole-based
electrons could reduce the bound metal ion filling the partially
filled 3d orbitals with n e-, leaving n holes on the ligand
platform; (2) the ligand and metal-based orbitals are close
enough in energy such that spin-state tautomerism is possible;13
(3) no intramolecular redox occurs, leaving each metal in the
divalent state with the ligand fully reduced. Our data lead us to
favor the third scenario, wherein several fully populated ligand-
based orbitals reside higher in energy than partially filled metal
3d orbitals, giving rise to a non-Aufbau electronic arrangement.
(dpm)M(py)2 upon isolation via crystallization (M ) pale-yellow
Mn (3), bright orange Fe (4), maroon Co (5), crimson Ni (6),
and yellow Zn (7)).
In the absence of strongly coordinating solvents, the formation
of ate-complexes was observed (e.g., [(dpm)CoCl(TH-
F)][Li(THF)2] (8)). Substitution of the two pyridine ligands in
4 for 2,2′-bipyridine (bpy) results in the clean formation of
(dpm)FeII(bpy) (9). The dpm ligand was found to uniformly
coordinate in an η1,η1-coordination mode, accommodating both
tetrahedral (observed for complexes 3, 4, 5, 7, and 8) and square-
planar coordination modes at the metal ion (observed for the
diamagnetic Ni complex 6) as confirmed by X-ray diffraction
studies. The solid-state molecular structures for the tetrahedral
Mn (3), Fe (4 and 9), Co (5 and 8), Ni (6), and Zn (7) complexes
are shown in Figures 1 and 2.
2.2. Structural Characterization of Complexes 3-7. Com-
plexes 3-5 and 7 are isostructural, featuring a pseudotetrahedral
geometry at the metal center. The Ni complex 6 is distinct from
this series, featuring a square-planar geometry at Ni. The metal
pyrrolide nitrogen bond lengths decrease across the series of
pseudotetrahedral complexes (davg (Å): Mn · · · Ndpm ) 2.053,
Fe· · ·Ndpm ) 1.982, Co · · ·Ndpm ) 1.961, Zn · · ·Ndpm ) 1.965;
see Table SI-1, Supporting Information for a more comprehen-
sive comparison of bond lengths). A similar trend is found when
2. Results
2.1. Synthesis and Characterization of dpm and Its Metal
Complexes. To assess the steric and electronic properties the
1,9-dimesityl-5,5-dimethyldipyrromethane dpmH2 (1) ligand
imparts, transition metal complexes were prepared in the
following manner: Reaction of (dpm)Li2(OEt2) (2, prepared by
deprotonation of 1 with two equivalents of PhLi) with a
stoichiometric amount of a divalent metal precursor (i.e.,
MCl2(py)2; M ) Mn, Co, Fe, Ni, Zn; py ) pyridine) in thawing
THF solutions afforded the metalated species of the type
(13) Storr, T.; Verma, P.; Pratt, R. C.; Wasinger, E. C.; Shimazaki, Y.;
Stack, T. D. P. J. Am. Chem. Soc. 2008, 130, 15448–15459.
9
J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009 14375

A R T I C L E S
King and Betley
Table 1. Magnetic and Spectral Properties of Complexes 3, 4, 5, and 9
complex
S
µeff (BM)
λ/nm (ε/M-1 ·cm-1
)
δ (mm/s)
∆Eq (mm/s)
5.97,a 6.1(1)b
5.00,a 5.2(1)b
5.1(1)b
-
-
0.80 0.86
0.72
-
-
3.46 2.63
3.00
-
5
(dpm)Mn(py)2 (3)
(dpm)Fe(py)2 (4)
(dpm)Fe(bpy) (9)
(dpm)Co(py)2 (5)
(dpm)Ni(py)2 (6)
/
2
2
420 (1500), 1359 (70), 1675 (50) (sh)
2
535 (410), 585 (620), 613 (870), 1125 (110), 1260 (105) (sh)
/
2
4.20,a 4.5(1)b
-
395 (1100)
503 (190)
3
0
-
-
a Average moment over T range of 200-300 K from SQUID. b Room temperature (295 K) moment in solution by Evans’s method.
then be attributed to the limited motion of the mesityl aryl groups
when accommodating the square-planar geometry. The relatively
weak electronic transition observed for 6 (λmax ) 503 nm, ε )
190 M-1 cm-1) is consistent with a Laporte forbidden d-d
transition.
To confirm the high-spin character of complexes 3-5 at low
temperatures, each complex was also characterized by SQUID
magnetometry from 2-300 K and simulated with the julX
package.14 As can be seen from Figure 4, complexes 3-5
maintain their hextet (Mn 3), quintet (Fe 4), and quartet (Co 5)
ground states throughout this temperature range. The Curie Law
observed for each complex in this temperature range (indicated
-1
by the linear plot ꢀm versus T, Figure 4b) indicates that the
ascribed spin states are the only states thermally populated.14
The powder EPR spectrum of 3 at 77 K reveals a complex
multiline pattern consistent with multiple absorbances occurring
for the high-spin MnII in a weak field ligand environment (Figure
5a). The powder EPR spectrum for the CoII complex 5 exhibits
an isotropic signal in the region of g ≈ 3.48, consistent with
high-spin CoII (S ) 3/2), though the signal persists even to 298
K.15 The EPR data for 5 was simulated using large zero-field
splitting parameter D ) 7.855 cm-1 and E/D ) 0 as determined
from fitting the SQUID data, with g⊥ ) 2.02, g| ) 2.31.16
The Mo¨ssbauer spectrum of 4 shows multiple components
modeled as two subspectra (Figure 6a). The two component
subspectra have nearly identical isomer shifts (δ ) 0.80, 0.86
mm·s-1) with different quadrupole splittings (component A:
∆EQ ) 3.46 mm ·s-1 (84%); component B: ∆EQ ) 2.63 mm ·s-1
(16%)), both consistent with FeII nuclei. The minor component
grows in intensity the later the spectrum is acquired from the
time of material preparation, even though 1H NMR analysis does
not reveal any change for the bulk material over a similar time
frame. Analysis of 4 by 1H NMR following Mo¨ssbauer
measurements only reveals the (dpm)Fe(py)2 complex without
any observable decomposition (e.g., pyridine loss or dpm ligand
dissociation). Neither subspectra are representative of the
FeCl2(pyridine)2 starting material.17 One possibility for the
presence of two components could be structural isomers due to
variation in pyridine binding to the Fe center.17 The Mo¨ssbauer
spectrum for 9 (Figure 6b), where the two pyridine ligands have
been substituted for a single 2,2′-bipyridine ligand, reveals a
single species featuring an isomer shift of 0.72 mm·s-1 and a
quadrapole splitting of 3.00 mm ·s-1, consistent with the major
component observed in the spectrum of 4. The bipyridine ligand
in 9 enforces a more acute ∠(N3-Fe-N4) bond angle (77.38°)
Figure 3. UV/vis/NIR molar absorptivity spectra of 3, 4, 5, and 6. Spectra
were taken in dichloromethane, molar absorptivities are based on measure-
ments at four concentrations.
comparing the metal pyridine nitrogen bond lengths (davg (Å):
Mn· · ·Npy ) 2.197, Fe · · ·Npy ) 2.136, Co · · ·Npy ) 2.045,
Zn· · ·Npy ) 2.087). In order to adopt the square-planar geometry
in 6, the Ni(py)2 unit is brought out of the ligand plane (dihedral
angle of 37° with respect to the pyrrolide subunits), further
evidenced by the splaying of the two flanking mesityl aryl
groups. The mesityl units of the dpm ligand carve a cleft about
the Ni of 80.5°, defined as the angle formed by the two mesityl
ipso-C atoms and the dpm bridgehead carbon atom (∠Cipso
-CMe2-Cipso). This angle is consistent throughout the series,
where angles of 77.6 (Mn), 86.6 (Fe), and 81.9° (Co) are
observed.
2.3. Electronic and Magnetic Characterization of Complexes
3-7. The magnetic and spectral data for complexes 3-6 and 9
are provided in Table 1. The nearly colorless complex 3 has no
transitions in the visible or near IR (NIR) regions (Figure 3) as
1
5
expected for the H NMR silent, high spin d5 (S ) /2) MnII
center (solution magnetic moment, µeff (295 K) ) 6.1(1) µB).
Based on the magnitude of their molar absorptivities, the visible
absorptions for orange 4 (λmax/nm (ε/M-1 cm-1): 420 (1500))
and maroon 5 (λmax/nm (ε/M-1 cm-1): 613 (870)) are of the
appropriate magnitude for spin-allowed d-d transitions for the
high spin FeII (S ) 2, µeff (295 K) ) 5.2(1) µB) and CoII (S )
3/2, µeff (295 K) ) 4.5(1) µB) centers in pseudotetrahedral ligand
fields, respectively. Only very weak transitions are found for 4
(λmax/nm (ε/M-1 cm-1): 1359 (70), 1675 (50)) and 5 (λmax/nm
(ε/M-1 cm-1): 1125 (110), 1260 (105)) in the near-IR region,
which are too low in intensity to correspond to charge transfer
bands (e.g., LMCT). Despite the paramagnetism of 4 and 5, all
1
(14) Bill, E. julX: Simulation of Molecular Magnetic Data; available from
http://www.mpi-muelheim.mpg.de/bac/logins/bill/julX_en.php, 2008.
(15) See Supporting Information for further physical characterization
data.
nine proton resonances are discernible by H NMR at 295 K
ranging from δ 148 to -15 ppm. 1H NMR analysis of 6 reveals
several significantly broadened (∆δ ) 1-2 ppm) resonances,
yet no solution magnetic moment was measurable at room
temperature, thus signifying the square-planar geometry ob-
served in the low-temperature solid-state structure is preserved
in solution. The observed broadening of the 1H resonances can
(16) A similar modeling scheme was employed to model the EPR spectrum
for a series of isoelectronic Cr(III) complexes. Pedersen, E.; Toftlund,
H. Inorg. Chem. 1974, 13, 1603–1612.
(17) Long, G. J.; Whitney, D. L.; Kennedy, J. E. Inorg. Chem. 1971, 10,
1406–1410.
9
14376 J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009

Redox-Active Ligands of (dpm)MII(py)2 Complexes
A R T I C L E S
Figure 4. (a) SQUID magnetization data for complexes 3 (green O), 4 (red O), and 5 (blue O) shown as a plot of µeff (BM) versus T (K) and (b) as a plot
19
-1
of ꢀm (mol/cm3) versus T (K). Dotted lines are the simulations of each data set.
Figure 5. Powder EPR spectrum of (a) (dpm)Mn(py)2 (3) and (b) (dpm)Co(py)2 (5) (dotted line is data simulation: g⊥ 2.02, g| 2.31; A⊥ 0, A| 0.0112 cm-1
D 7.855 cm-1, E/D 0) (77 K, X-band, 9.397 GHz).
;
Figure 6. Zero-field Mo¨ssbauer spectrum of (a) (dpm)Fe(py)2 (4) and (b) (dpm)Fe(bpy) (9) at 4.2 K.
2.4. Electrochemical Behavior of Dipyrromethane Complexes.
Cyclic voltammetry on the bis-pyridine complexes, 3-7, showed
no reversible oxidative or reductive events, and instead showed
a rather complex series of irreversible oxidations. In order to
probe their electrochemical behavior further, differential pulse
voltammetry was used to assess the electrochemical potentials
of these irreversible events. Using oxidative scans, all five
as compared to the ∠(N3-Fe-N4) bond angle in 4 (97.32°).
The large ∆EQ for 9 is consistent with the greater deviation of
9 from an ideal tetrahedral geometry due to the bpy ligand.
While the connectivity for 4 may not change, reorientation of
the pyridine ligands in a more spherically symmetric fashion
about Fe may produce the minor species detected in Figure
6a.
9
J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009 14377

A R T I C L E S
King and Betley
Figure 7. Differential pulse voltammograms of (a) solutions in THF of 10-4 M (dpm)M, where M ) Mn(py)2 (3), Fe(py)2 (4), Co(py)2 (5), Ni(py)2 (6),
Zn(py)2 (7), H2 (1); Bu4NPF6 (0.3 M); scan rate 20 mV/s on glassy C electrode), and (b) solutions in acetonitrile of 10-4 M (dpm)M, where M ) Co(py)2
(5), Ni(py)2 (6), Zn(py)2 (7), Fe(bpy) (9), H2 (1); Bu4NPF6 (0.1 M); scan rate 20 mV/s on Pt electrode.
Figure 8. Calculated R (2) and ꢁ (1) orbital energies for Mn (3), Fe (4), and Co (5). Calculated molecular orbitals for Ni (6) and Zn (7). Metal-based
orbitals are gray, dipyrromethane ligand-based orbitals are highlighted red, and pyridine-based orbitals are black.
species exhibited three major anodic peaks between -800 and
+600 mV relative to Fc/Fc+ in THF (0.3 M Bu4NPF6, Figure
7a). The potential of the initial peak corresponds well with the
onset of current response in the cyclic voltammetry experiments,
yet it was indeterminable if the oxidations represented one or
two electron events. In acetonitrile solutions, the large oxidative
waves for complexes 5-7 and 9 in THF separated into two
well-defined one-electron events (Figure 7b). For 1 mM
solutions of (dpm)M(py)2 {M ) Co (5), Ni (6), Zn (7)} and
(dpm)Fe(bpy) (9) in acetonitrile (0.1 M Bu4NPF6, Pt working
electrode) the initial peak was consistently observed at Ep )
-240 mV. The Mn and Fe bis-pyridine complexes 3 and 4,
respectively, proved too labile in acetonitrile solutions to give
reliable data. The Fe(bpy) analogue 9, however, was stable
enough in acetonitrile solutions to provide reproducible data,
included in Figure 7b. The onset of oxidation for 9 is consistent
with the Co (5), Ni (6), and Zn (7) species, although it does
not exhibit a second oxidation event within a similar range.
Subsequent peaks were observed at different potentials but
followed similar patterns for all four metals. The Co and Zn
voltammograms overlay with little deviation of the observed
potentials, while the Ni voltammogram features the second one-
electron oxidation at a slightly lower potential. We interpret
the redox inactive ZnII data to imply the observed sequential
one-electron oxidations are predominantly ligand-centered.
Interestingly, although the oxidations appear to be ligand-
centered and vary little with divalent cation substitution, the
onset of oxidation for metalated dpm does shift nearly 700 mV
from the free dpmH2 (Ep ) 475 mV).
2.5. DFT Analysis. To further probe the electronic structure
of the (dpm)M(py)2 complexes, geometry-optimized calculations
on the (dpm)M(py)2 complexes 3-7 were performed (B3LYP/
TZVP/SV(P)).18 The orbital energies for complexes 3-7 are
presented in Figure 8. Dipyrromethane orbitals are highlighted
in red, metal 3d orbitals in gray, and pyridine orbitals are
highlighted in black. In every instance the calculations indicate
the highest lying occupied molecular orbitals (HOMO,
HOMO-1) exclusively feature ligand π-character on the pyrrole
9
14378 J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009

Redox-Active Ligands of (dpm)MII(py)2 Complexes
A R T I C L E S
Figure 9. (a) HOMO to HOMO-4 from molecular orbital diagrams for 5 by DFT (B3LYP/TZVP/SV(P); Gaussian 03). (b) Calculated spin density for
3
complex 5 (Co, S ) /2).
subunits, illustrated for complex 5 in Figure 9. The metal 3d-
orbitals do not significantly contribute to the frontier orbitals
until HOMO-4, often 1-2 eV lower in energy than the
dipyrromethane-based HOMO. For the high-spin ion complexes
3-5, the spin density plot (R-ꢁ) reveals all of the spin localized
on the metal ion (Figure 9b shows spin-plot for 5). These results
indicate that electron pairing within the higher-energy, delo-
calized dipyrromethane ligand π-orbitals is energetically favored
over electron pairing within lower-energy, but localized metal
ion 3d orbitals. This idea is consistent with encountering a
significantly larger Coulombic repulsion for pairing electrons
in 3d orbitals in the intramolecular redox state (dpmn+)M2-n
(scenario 1), than pairing electrons in energetically higher-lying,
though diffuse dipyrromethane-based π-orbitals (scenario 3).
The electronic structure suggested by the DFT results is
consistent with the electrochemical behavior exhibited by
complexes 3-9. The pyrrolide π-electrons are the most acces-
sible energetically; thus, we observe sequential pyrrole oxidation
events with little perturbation by the chelated divalent ion.
2.6. Chemical Oxidation. Metal porphyrinogen complexes are
known to undergo similar pyrrole-based oxidations (chemical
and electrochemical), as shown by Floriani and Nocera.12b,c
Sequential one-electron oxidations led to cyclopropane formation
in the porphyrinogen’s meso position via radical coupling of
adjacent pyrrolic carbons in the 2 and 2′ position. Sequential
one-electron oxidation of the two pyrrolide subunits should lead
to a transient diradical, illustrated in eq 2.
formation from the C4 and C6 positions, likely arising from
carbocation stabilization at the pyrrolic C4 and C6 positions
and radical localization at the pyrrolic C3 and C7 positions.
Cyclic voltammetry on 10 in THF reveals only two reversible
reduction events (E1/2 ) -1.14 V, -2.28 V vs Fc/Fc+), but no
oxidation (Figure S1, Supporting Information). Chemical oxida-
-
tion of complexes 3-6 with two equivalents of Fc+PF6 did
not produce detectable quantities of 10; only free ligand (1)
and one equivalent of ferrocene were observed as the assignable
products of oxidation. Likewise, attempts to directly synthesize
a (dpm)MIII complex were unsuccessful. For example, metalation
of MIII synthons (i.e., MnIII(acac)3, FeCl3, FeIII(acac)3; acac )
acetylacetonate) with 2 only yielded complex mixtures of
(dpm)MII(L)n species (L ) pyridine, THF), free ligand, and
unidentified metal salt mixtures, but no isolable (dpm)MIII
materials.
3. Discussion
The most unusual feature of the complexes reported is the
atypical electronic structure suggested by the electrochemical
experiments. More precisely, the highest occupied molecular
orbitals appear to be ligand based (see section 2.4), even for
the complexes featuring high-spin electron configurations (3-5).
The dipyrromethane pyrrole π-electrons are thus higher in
energy than the frontier orbitals of the 3d transition metal (Mn
f Zn) inVestigated, making intramolecular redox between the
ligand and metal possible. Having fully populated ligand-based
orbitals higher in energy than partially filled metal-based orbitals
suggests that intramolecular redox is possible and might occur.
We presented three possibilities for what types of intramolecular
redox could be present, namely: (1) ligand-based electrons could
reduce the bound metal ion, filling the partially filled 3d orbitals
with n e-, leaving n holes on the ligand platform; (2) the ligand
and metal-based orbitals are closely matched in energy, and spin-
state tautomerism is possible;13 (3) no intramolecular redox
occurs, leaving each metal in the divalent state with the ligand
fully reduced.
Should electronic structure scenario 1 be operative, the
dipyrromethane ligand structure should manifest stepwise oxida-
tion with concomitant bond elongation within the pyrrole
subunits,12b,c and the onset of oxidation in the electrochemical
experiments should vary with metal ion substitution. Close
inspection of all the bond lengths from the series of complexes
3-9 (Table SI-1, Supporting Information) does not reveal any
perturbation of the dipyrromethane σ-bonds or π-bonds within
the ligand architecture. The only significant variations discern-
ible within the series are the bond lengths of M-Ndpm and
M-Npyridine (bonds 1-2 and 3-4, respectively, Table SI-1,
Supporting Information) which vary due to the variable ionic
size for the metals in different spin states. The invariant
electrochemical behavior of metals bound by the dipyrromethane
Whether the dipyrromethane ligands can undergo a radical
coupling similar to the porphyrinogen platform to form a cyclo-
propane unit has yet to be confirmed. Chemical oxidation of the
dilithio species 2 with two equivalents of AgCl in thawing THF
solution produced one equivalent of free dpmH2 1 and an equimolar
amount of a new diamagnetic species. 1H NMR, HRMS, and X-ray
crystallography confirm the new species obtained is the four-
electron oxidized product diazacyclopentapentalene (10) (eq 3),
which is inconsistent with two-electron oxidized, cyclopropane
formation observed in porphyrinogen systems.
Presumably this process occurs from radical coupling of the
ligand C3 and C7 positions followed by further two-electron
oxidation and deprotonation. Coupling the pyrrole C3 and C7
positions, in this case, occurs preferentially to cyclopropane
9
J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009 14379

A R T I C L E S
King and Betley
ligand offer the most compelling evidence against scenario 1.
Namely, the DPV experiments on the (dpm)MII(py)2 series
reveal a common two-electron oxidation pathway that is entirely
ligand-based, invariant to the divalent metal-bound (or its spin
state) within the dpm framework. Moreover, the UV-vis/NIR
results do not reveal a significant LMCT band, suggesting there
are not even any excited states featuring charge transfer from
the dpm to the MII ions.
been prepared. Arylation of the dpm ligand R to the pyrrolic
nitrogen donors limits the accessibility of the pyrrole π-electrons
for transition metal coordination, instead forcing η1,η1 coordina-
tion to the divalent metal ion. Electrochemical studies on the
(dpm)MII(py)2 series reveal a common two-electron oxidation
pathway that is entirely ligand-based, invariant to the divalent
metal-bound (or its spin state) within the dpm framework. The
energetically high-lying π-electrons from the dipyrromethane
framework almost exclusively account for the observed redox
behavior of the metal complexes studied. This latter observation
indicates that fully populated ligand-based orbitals from the dpm
construct lie above partially filled metal 3d orbitals without
intramolecular redox chemistry or spin-state tautomerism oc-
curring. This unusual electronic structure is corroborated by DFT
studies, revealing fully occupied ligand-based π-orbitals at
higher energies than half-filled metal 3d orbitals for Mn-Zn.
While chemical oxidation of the (dpm)MII species did not yield
a stable ligand-based diradical/dication, further experimental and
computational studies are currently underway to elucidate the
exact nature of the biradical generated as a result of both
electrochemical and chemical oxidations.
The SQUID magnetometry results cannot rule out any of the
electronic structure postulates, as all three can be consistent with
the total spin for each complex being preserved over the
temperature range investigated. The EPR for both complex 3
5
3
and 5 are consistent for S ) /2 Mn and S ) /2 Co nuclei,
respectively, with observed g values inconsistent for organic
radicals, where g ≈ 2. The EPR signals for both 3 and 5 persist
over a broad temperature range (77-298 K). While this
observation is unusual for typical CoII ions in tetrahedral
environments where fast spin-lattice relaxation times of the
high-spin CoII nucleus usually limit observation of transitions
to below ∼30 K,19 it also is inconsistent with spin-state
tautomerism as a function of temperature.13 The Mo¨ssbauer data
for 4 and 9 suggest a single major nuclear environment for each
of the Fe(II) ions and do not show any variance between 4 and
77 K. On the basis of the structural, electrochemical, and ma-
gnetic data, we rule out electronic structure scenarios 1 and 2
in favor of scenario 3 wherein no intramolecular redox is
observed. This assignment is corroborated by the DFT analysis.
Acknowledgment. We thank the American Chemical Society
(PRF Grant Type G) and Harvard University for financial support,
and Prof. R. H. Holm for the generous use of his Mo¨ssbauer
spectrometer.
Supporting Information Available: Experimental procedures
and details for complexes 1-10. Cyclic voltammetry of 10
(Figure S1), ꢀm vs T and ꢀm-1 vs T for complexes 3-5 (Figure
S2), X-band EPR of 3 and 5 at room temperature (Figure S3),
DPV of complexes 3-7 and 9 in acetonitrile (Figure S4),
dipyrromethane bond length comparisons (Table SI-1), X-ray
diffraction experimental details for 3-10 (Table SI-2), complete
reference 18 (SI endnote 9), and the geometry-optimized
coordinates for complexes 3-7. This material is available free
of charge via the Internet at http://pubs.acs.org.
4. Conclusions
Transition metal complexes (Mn f Zn) of the dipyrromethane
ligand 1,9-dimesityl-5,5-dimethyldipyrromethane (dpm) have
(18) (a) Frisch, M. J.; et al. Gaussian 03, Revision E.01; Gaussian, Inc.:
Wallingford CT, 2004. (b) Weigend, F.; Ahlrichs, R. Phys. Chem.
Chem. Phys. 2005, 7, 3297-3305.
(19) (a) Pilbrow, J. R. Transition Ion Electron Paramagnetic Resonance;
Clarendon Press: Oxford, 1990. (b) Aasa, R.; Va¨nngård, T. J. Magn.
Reson. 1975, 19, 308–315.
JA903997A
9
14380 J. AM. CHEM. SOC. VOL. 131, NO. 40, 2009