Effects of 5,5-substitution on dipyrrolylmethane ligand isomerization

Article abstract of DOI:10.1039/b806947f

The difference in steric profiles for η⁵ and η¹-pyrrolyls can be used to modify barriers for pyrrolyl exchange. The method investigated incorporated 1,3-diaxial interactions in a cyclohexyl backbone of the dipyrrolylmethane. Some of the enthalpic barrier created with the 1,3-steric interactions appears to be mediated by increased entropy in the ground state, but noticeable changes are produced in the pyrrolyl exchange rates by ¹H NMR without steric crowding at the metal center.

Full text of DOI:10.1039/b806947f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Effects of 5,5-substitution on dipyrrolylmethane ligand isomerization†
Douglas L. Swartz II and Aaron L. Odom*
Received 24th April 2008, Accepted 22nd May 2008
First published as an Advance Article on the web 11th July 2008
DOI: 10.1039/b806947f
5
1
The difference in steric profiles for g and g -pyrrolyls can be used to modify barriers for pyrrolyl
exchange. The method investigated incorporated 1,3-diaxial interactions in a cyclohexyl backbone of
the dipyrrolylmethane. Some of the enthalpic barrier created with the 1,3-steric interactions appears to
be mediated by increased entropy in the ground state, but noticeable changes are produced in the
pyrrolyl exchange rates by 1H NMR without steric crowding at the metal center.
Previously,6 we have reported that altering the pyrrolyl ligand by
Introduction
placing aryl-substituents in the 2-position of the dpm framework
lowers the barriers associated with pyrrolyl exchange, presumably
Dipyrrolylmethane ligands on titanium provide catalysts capa-
ble of extremely fast hydroamination of alkynes with primary
amines1,2 and very efficient catalysts for iminoamination.3 One
of the most commonly employed dipyrrolylmethane derivatives is
prepared from acetone and pyrrole (Scheme 1) in the presence of
trifluoroacetic acid.4 The product 5,5-dimethyldipyrrolylmethane
(H2dpm) can be placed on titanium (Scheme 1) by transamination
with Ti(NMe2)4 to provide Ti(NMe2)2(dpm) (1).1,2
5
by sterically destabilizing the g -bonding mode.7 In fact placing a
3,5-bis(trifluoromethyl)phenyl or mesityl group in the 2-position
of the pyrrole lowers the pyrrolyl exchange barriers below what
can be measured by variable temperature NMR (<5 kcal mol−1).
However, the addition of sterically hindered groups in these
positions hinders access to the metal center as well, affecting
catalysis rates.6
In an attempt to alter the pyrrolyl isomerization rates without
blocking access to the metal we explored the use of substituents in
the dpm backbone. The strategy was to use 1,3-diaxial interactions
5
within a cyclohexyl ring to stabilize the g -bonding mode by
sterically inhibiting pyrrolyl-ring rotation (Fig. 1). The steric
interaction will be with the axial pyrrolyl substituent, which will
5
prefer to be in the g -bonding mode, allowing it to present one side
of the p-system to the cyclohexyl substituents rather than the larger
ring edge. By using gem-disubstitution, the 1,3-diaxial interactions
are required regardless of which ring is bound through the p-face.
5
The size of R should affect the ease with which the g -bound ring
1
can rotate to obtain the g -bonding mode.
Scheme 1 Synthesis of H2dpm and Ti(dpm)(NMe2)2 (1).
In the solid state and in solution, complex 1 has a structure
where the two pyrrolyl substituents are inequivalent. One of the
5
1
pyrrolyl rings adopts an g -geometry, and the other is g -bound.
However, in room temperature NMR spectra the two pyrroles
appear equivalent due to fast exchange on the timescale of this
spectroscopy. Our research group has previously reported the
barrier for pyrrolyl exchange as DG‡ ∼ 10 kcal mol−1.1
Fig. 1 Use of 1,3-diaxial interactions to affect dipyrrolylmethane isomer-
ization barriers.
The most likely mechanism for pyrrole exchange is conversion
1
of both rings to an g -geometry.5 Methods for altering this barrier
may provide clues to the active species in catalysis and allow the
control of complex structures.
In this paper, we report the synthesis of 1,1-bis(a-
pyrrolyl)cyclohexane ligands, their titanium complexes, and the
barriers for pyrrolyl exchange. The barriers were determined
using the Eyring method with kinetics from variable temperature
spin saturation transfer NMR spectroscopy. Consequently, the
enthalpic and entropic effects of this substitution were determined.
All of this data will be compared with the simple dimethyl
compound 1.
Michigan State University, Department of Chemistry, East Lansing, MI
48824, U. S. A. E-mail: odom@chemistry.msu.edu
† Electronic supplementary information (ESI) available: Tables for the X-
ray diffraction experiment and variable temperature 1H NMR spectra of
2 and 3. CCDC reference number 686141. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/b806947f
4254 | Dalton Trans., 2008, 4254–4258
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Results and discussion
We examined two cyclohexyl-dpm derivatives, which were eas-
ily prepared using commercially available cyclohexanone and
3,3,5,5-tetramethylcyclohexanone as shown in Scheme 2. The
1,1-bis(a-pyrrolyl)cyclohexane (H2cpm) and 1,1-bis(a-pyrrolyl)-
3,3,5,5-tetramethylcyclohexane (H2tmcpm) compounds were pre-
pared in 43% and 24% isolated yields. The transamination
reactions with Ti(NMe2)4 and both of these cyclohexyl derivatives
occurred in high yields to provide R = H Ti(NMe2)2(cpm) and
R = Me Ti(NMe2)2(tmcpm) (3).
Fig.
2
ORTEP diagram of the X-ray diffraction model for
Ti(tmcpm)(NMe2)2 (3). Ellipsoids at the 50% probability lev◦el. Hydrogens
˚
omitted for clarity. Selected bond distances (A) and angles ( ): Ti(1)–N(1)
2.292(2), Ti(1)–N(2) 2.023(1), Ti(1)–N(3) 1.901(1), Ti(1)–N(4) 1.895(2);
N(4)–Ti(1)–N(3) 103.49(6), N(4)–Ti(1)–N(2) 107.01(6), N(3)–Ti(1)–N(2)
101.47(6).
Scheme 2 Synthesis of H2cpm, H2tmcpm, Ti(cpm)(NMe2)2 (2), and
Ti(tmcpm)(NMe2)2 (3).
Table 1 Parameters for pyrrolyl exchange of compounds 1–3
The structure of the tetramethylcyclohexyl complex has been
determined by X-ray diffraction. An ORTEP diagram is shown in
Fig. 2. The metric parameters around titanium are quite similar
to previously reported derivatives.1,2 The cyclohexyl group has
DH‡/kcal
DS‡/cal
DG‡/kcal
mol−1
mol−1 K−1
mol−1
Ti(NMe2)2(dpm) (1)
Ti(NMe2)2(cpm) (2)
Ti(NMe2)2(tmcpm) (3)
11.8
14.1
17.2
5.2
5.1
12.7
10.3
12.6
13.4
5
the expected chair-conformation with the g -pyrrolyl in the more
sterically favorable axial position. This conformation puts the two
5
axial methyl groups directly over the g -pyrrolyl. The interaction
between these groups can be seen in the space-filling model from
the X-ray coordinates (Fig. 3).
from approximately 12 kcal mol−1 to over 17 kcal mol−1 for
1–3, reflective of the greater steric interactions on attempting
Variable temperature spin-saturation transfer was used to deter-
mine the barriers for pyrrolyl equilibration for Ti(NMe2)2(dpm)
(1), Ti(NMe2)2(cpm) (2), and Ti(NMe2)2(tmcpm) (3). As an
example, the Eyring plot for Ti(NMe2)2(tmcpm) (3) is shown in
Fig. 4.
The plot in Fig. 4 is of ln(k/T) versus 1/T and is used with the
Eyring Equation (Eqn 1), where kB = Boltzmann constant and R
is the gas constant, to determine the activation parameters under
the usual assumptions of Transition State Theory.8 Consequently,
the slope (m) can be used to determine DH‡ using Eqn 2, and the
intercept (b) can be used to determine DS‡ using Eqn 3.
5
to rotate the g -pyrrolyl substituent during the exchange in the
tetramethylcyclohexyl complex.
The entropic parameters are the same, small and positive, for
the 5,5-dimethyldipyrrolyl 1 and cyclohexyldipyrrolyl 2 during the
exchange.
The tetramethyl complex 3 shows a substantial increase in DS‡
to 13 cal mol−1 K−1, which is quite a large positive value for a
unimolecular process. The increase in entropy in the transition
state suggests that the ground state is highly ordered. This may be
due to the single preferred conformation when one of the pyrroles
5
is in the g -conformation as opposed to the more conformationally
1
flexible g -conformation where neither chair conformer will be
(1)
1
5
strongly favored and any C–C rotations difficult in the g ,g -
conformer will be more readily allowed.
DH‡ = −mR
(2)
As shown in Table 1, the difference in exchange barriers between
1 and 2 are largely enthalpic and presumably caused by the 1,3-
diaxial interactions. In contrast, the difference in barriers between,
1 and 3, is more complex. The enthalpic barrier to reach the
transition state for exchange is ∼5 kcal mol−1 higher for 3, but
the entropic term reduces the overall barrier to 13.4 kcal mol−1.
(3)
The parameters for the pyrrolyl exchanges in compounds 1–
3 can be found in Table 1. The enthalpic barriers increase
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4254–4258 | 4255
©

View Article Online
to 1), the entropic factor for sterically hindered 3 cancels some of
the enthalpic increases. Consequently, the free energy barrier only
increased by a modest 3 kcal mol−1 using this strategy.
All three of these compounds were evaluated for catalysis
using the same hydroamination test conditions used previously.6
All three exhibited the same rate for catalysis, indicating that
altering the isomerization barriers to the level possible here did
not affect the catalysis rates, which is consistent with the rate
limiting step not being associated with this isomerization. This is
not unexpected as barriers associated with the isomerization are
relatively small for this titanium system.
The same methodology used here may prove useful to study
pyrrolyl coordination effects in systems where the isomerization
has a much larger enthalpic barrier. In addition, this strategy using
cyclohexane-based dipyrrolylmethanes could potentially be used
to control complex chirality by using asymmetric cyclohexane
1
5
substituents in systems with a larger tendency to stay in the g ,g -
coordination mode.
Experimental
General considerations
Anhydrous ether was purchased from Columbus Chemical In-
dustries Inc., and pentane and toluene were purchased from
Spectrum Chemical Mfg. Corp., purified by sparging with dry
N2, then water was removed by running them through activated
alumina systems purchased from Solv-Tek. Hexanes and ethyl
acetate were purchased from Mallinckodt-Baker Inc. Reagent
grade cyclohexanone and 3,3,5,5-tetramethylcyclohexanone were
purchased from Acros Organics and used as received. Pyrrole
was purchased from TCI, refluxed with sodium, distilled under
nitrogen, and then stored in a purified nitrogen glove box.
Fig. 3 Two views of a space filling model of Ti(tmcpm)(NMe2)2 (3)
from the X-ray diffraction coordinates showing the g -pyrrolyl and
cyclohexylmethyl close contacts.
5
9
Ti(NMe2)4 was prepared using a modification of the literature
procedure. Deuterated solvents were dried over purple sodium
benzophenone ketyl (C6D6) or phosphoric anhydride (CDCl3)
1
and distilled under a nitrogen atmosphere. H and 13C spectra
were recorded on a VXR-500 spectrometer. All spectra were
referenced internally to residual protiosolvent (1H) or solvent (13C)
resonances. Chemical shifts are reported in ppm, and coupling
constants are reported in Hz. Typical coupling constants are not
reported.
Procedure for spin saturation transfer experiments
The spin saturation transfer experiments were carried out using the
method described by Kresge and co-workers.10 The dimethylamido
resonances were observed in the spin saturation transfer experi-
ments. In order to correct for decoupler spill-over, off-resonance
irradiation was carried out on the opposite side of the observation
peak relative to the dimethylamido resonance being irradiated.
Fig. 4 Eyring plot for the pyrrolyl exchange in Ti(NMe2)2(tmcpm) (3).
Conclusions
X-Ray diffraction
Cyclohexyl substituents in the backbone of the dipyrrolylmethane
ligand can be used to affect pyrrolyl exchange barriers. The
difference in sterics for the g -pyrrolyl and g -pyrrolyl groups can
be used to alter this barrier. While placing gem-dimethyl groups
on the 3-carbon of the cyclohexyl ring increases the enthalpic
cost to pyrrolyl exchange substantially (increased ∼50% relative
Crystals grown from concentrated toluene solutions at −35 ◦C
were moved quickly from a scintillation vial to a microscope slide
containing Paratone N. Samples were selected and mounted on a
glass fiber in wax and Paratone. The data collection was carried out
at a sample temperature of 173 K on a Bruker AXS platform three-
circle goniometer with a CCD detector. The data were processed
5
1
4256 | Dalton Trans., 2008, 4254–4258
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
and reduced using the program SAINTPLUS supplied by Bruker
AXS. The structures were solved by direct methods (SHELXTL
v5.1, Bruker AXS) in conjunction with standard difference Fourier
techniques.
Synthesis of Ti(NMe2)2(cpm) (2)
All manipulations were carried out in an inert atmosphere dry
box filled with purified dinitrogen. A 20 mL scintillation vial was
loaded with Ti(NMe2)4 (0.222 g, 0.991 mmol) and 2 mL of ether.
In a separate vial was loaded H2cpm (0.212 g., 0.991 mmol) in
2.5 mL of ether. The two vials were placed in a liquid nitrogen
cooled cold well where they sat until frozen. To a thawing solution
of Ti(NMe2)4, H2cpm was added. The solution was allowed to
warm to room temperature where it was left to react for 18 h. The
volatiles were removed under reduced pressure, which provided the
compound in◦pure form as judge by NMR and elemental analysis.
Mp 126–130 C (dec.). 1H NMR (CDCl3, 500 MHz, 25 ◦C) d 7.01
(br s, 2 H), 6.41 (br s, 2 H), 6.30 (br s, 2 H), 3.24 (s, 12 H), 2.19 (br s,
4 H), 1.51 (br s, 6 H). 1H NMR (CDCl3, 500 MHz, −40 ◦C): 7.25
(app s, 1 H), 6.83 (app s, 1 H), 6.78 (app s, 1 H), 6.73 (app s, 1 H),
6.11 (s, 1 H), 5.93 (s, 1 H), 3.34 (s, 6 H), 3.15 (s, 6 H) 2.76 (d, 1 H,
J = 12), 2.29 (d, 1 H, J = 14), 1.87 (app t, 1 H, J = 15), 1.79 (app t,
1 H, J = 8.5), 1.56–1.72 (m, 3 H), 1.42–1.56 (m, 1 H), 1.28–1.40 (m,
1 H), 1.18–1.26 (m, 1 H). 13C NMR (CDCl3, 126 MHz, −40 ◦C):
d 163.35, 160.79, 126.54, 123.97, 118.13, 115.33, 106.27, 100.67,
48.12, 47.32, 43.93, 38.81, 37.62, 25.60, 23.37, 23.26. Anal. Found
(Calcd.) C: 61.91 (62.07); H: 8.49 (8.10); N: 15.62 (16.08)%.
Crystal data for Ti(NMe2)2(tmcpm) (3): C22H36N4Ti, M =
˚
˚
˚
404.45, monoclinic, a = 21.3114(2) A, b = 8.5814(1) A, c =
◦
◦
◦
3
˚
12.2398(1) A, a = 90 , b = 98.185(1) , c = 90 , 2215.63(4) A , T =
173(2) K, space group P21/c, Z = 4, 35590 reflections measured,
4471 independent (Rint = 0.0313), Final R indices [I > 2r(I)] R1 =
0.0357, wR2 = 0.0960.
Synthesis of 1,1-bis(a-pyrrolyl)cyclohexane (H2cpm)
An oven dried 100 mL round bottom flask was charged with
cyclohexanone (1 g, 10 mmol) and pyrrole (17 g, 253 mmol)
and capped with a septum. The solution was then degassed with
argon for 10 min. Trifluoroacetic acid (0.116 g, 1 mmol) was
added via syringe. The reaction mixture was stirred for 15 min
under an argon atmosphere before being quenched with a 0.1 M
NaOH (30 mL) solution. The solution was then transferred to a
separatory funnel. It was extracted with OEt2, and the aqueous
layer was washed with OEt2 (2 × 30 mL). The combined organic
layers were dried with MgSO4 and subjected to rotary evaporation
to yield a viscous brown oil. The excess pyrrole was removed by
distillation under vacuum (∼1 torr). The product was purified by
column chromatography on silica gel with an eluant of hexanes–
ethyl acetate (7 : 3) to yield a white solid, (0.928 g, 43%). Mp
Synthesis of Ti(NMe2)2(tmcpm) (3)
All manipulations were carried out in an inert atmosphere dry
box filled with purified dinitrogen. In a 20 mL scintillation vial
was loaded Ti(NMe2)4 (0.195 g, 0.869 mmol) and 2 mL of ether.
In a separate vial was loaded H2tmcpm (0.270 g, 0.870 mmol)
and 2.5 mL of ether. The two vials were placed in the cold well
where they sat until frozen. To a thawing solution of Ti(NMe2)4
was added H2tmcpm. The solution was allowed to warm to room
temperature, where it was left to react for 18 h. The volatiles were
removed under reduced pressure, which provided the compound
in pure form as judged by NMR and elemental◦analysis. Mp 174–
177 ◦C (dec.). 1H NMR (CDCl3, 500 MHz, 25 C): d 6.97 (br s, 2
H), 6.37 (br s, 4 H), 3.24 (br s, 12 H), 2.6–1.6 (br s, 4 H), 1◦.25 (s,
◦
1
104–106 C. H NMR (CDCl3, 500 MHz): d 7.61 (br s, 2 H),
6.58–6.55 (m, 2 H), 6.18-6.13 (m, 2 H), 6.13-6.11 (m, 2 H), 2.12–
2.07 (m, 4 H), 1.62–1.53 (m, 4 H), 1.52–1.45 (m, 2 H). 13C NMR
(126 MHz, CDCl3): d 137.81, 116.67, 107.78, 104.30, 39.77, 37.23,
25.91, 22.74. Anal. Found (Calc.) C: 78.21 (78.46); H: 8.91 (8.47);
N: 12.83 (13.07)%.
1
2 H), 0.92 (br s, 12 H). H NMR (CDCl3, 500 MHz, −40 C): d
Synthesis of 1,1-bis(a-pyrrolyl)-3,3,5,5-tetramethylcyclohexane
(H2tmcpm)
7.19 (app s, 1 H), 6.77–6.75 (m, 3 H), 6.11 (t, 1 H, J = 2.6), 5.88
(dd, 1 H, J = 1.83 and 1.09), 3.35 (s, 6 H), 3.12 (s, 6 H), 2.85 (d,
1 H, J = 14.0), 2.29 (d, 1 H, J = 15.0), 1.73 (d, 1 H, J = 15.0),
1.63 (d, 1 H, J = 14.0), 1.23 (br s, 2 H), 1.00 (s, 3 H), 0.93 (s, 3 H),
0.90 (s, 3 H), 0.63 (s, 3 H). 13C NMR (CDCl3, 126 MHz) (−40 ◦C)
d 166.21, 162.59, 126.09, 123.93, 117.15, 115.16, 106.31, 101.06,
51.89, 50.05, 49.45, 48.16, 47.25, 43.43, 37.10, 35.82, 32.14, 31.91,
29.97, 27.40. Anal. Found (Calcd.) C: 65.13 (65.34); H: 8.91 (9.23);
N: 13.62 (13.85)%.
An oven dried 100 mL round bottom flask was charged with
3,3,5,5-tetramethylcyclohexanone (1.0 g, 6.5 mmol) and pyrrole
(9.3 g, 138 mmol) and capped with a septum. The solution was
then degassed with argon for 10 min. Trifluoroacetic acid (0.074 g,
0.65 mmol) was added via syringe. The reaction mixture was
stirred for 15 min under an atmosphere of argon before being
quenched with 0.1 M NaOH (30 mL) solution. The solution was
then transferred to a separatory funnel and was extracted with
OEt2. The aqueous layer was washed with OEt2 (2 × 30 mL). The
combined organic layers were dried with MgSO4 and subjected
to rotary evaporation to yield a viscous brown oil. The excess
pyrrole was removed by distillation under vacuum (∼1 torr). The
product was purified by column chromatography on silica gel with
an eluant of hexanes–ethyl acetate (7 : 3) to yield a white solid,
0.426 g (24%). Mp 83–85 ◦C. 1H NMR (CDCl3, 500 MHz): 7.65
(br s, 2 H), 6.55 (dd, 2 H, J = 2.5 and 1.5), 6.13–6.07 (m, 4 H),
2.00 (s, 4 H), 1.27 (s, 2 H), 0.94 (s, 12 H). 13C NMR (CDCl3 126
MHz): 138.7, 116.4, 107.7, 104.0, 51.9, 47.9, 39.5, 32.7, 31.7. Anal.
Found (Calc.) C: 80.17 (79.95); H: 9.94 (9.69); N: 10.46 (10.36)%.
Acknowledgements
The authors thank the National Science Foundation of the United
States and the Petroleum Research Fund administered by the
American Chemical Society for funding. The authors also thank
Dan Holmes for help with the spin saturation transfer experiments.
References
1 Y. Shi, C. Hall, J. T. Ciszewski, C. Cao and A. L. Odom, Chem.
Commun., 2003, 586.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4254–4258 | 4257
©

View Article Online
2 For some additional reports of dipyrrolylmethane ligands on transition
metals see: A. Novak, A. J. Blake, C. Wilson and J. B. Love, Chem.
Commun., 2002, 2796; S. Majumder and A. L. Odom, Organometallics,
2008, 1174; J. B. Love, P. A. Salyer, A. S. Bailey, C. Wilson,
A. J. Blake, E. S. Davies and D. J. Evans, Chem. Commun., 2003,
1390.
3 S. Majumder and A. L. Odom, manuscripts in preparation.
4 B. J. Littler, M. A. Miller, C. H. Hung, R. W. Wagner, D. F.
O’Shea, P. D. Boyle and J. S. Lindsey, J. Org. Chem., 1999, 64,
1391.
5 For a recent computational study on pyrrolyl isomerization see: A. R.
Dias, A. P. Ferreira and L. F. Veiros, Compt. Rend. Chim., 2005, 8,
1444.
6 D. L. Swartz II and A. L. Odom, Organometallics, 2006, 25, 6125.
7 J. M. Tanski and G. Parkin, Organometallics, 2002, 21, 587.
8 B. K. Carpenter, Determination of Organic Reaction Mechanisms, John
Wiley & Sons, New York, 1984.
9 D. C. Bradley and I. M. Thomas, J. Chem. Soc., 1960, 3859.
10 C. L. Perrin, J. D. Thoburn and A. J. Kresge, J. Am. Chem. Soc., 1992,
114, 8800.
4258 | Dalton Trans., 2008, 4254–4258
This journal is
The Royal Society of Chemistry 2008
©