Journal of Organometallic Chemistry 695 (2010) 2789e2793
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
CeC coupling reaction of pyridine derivatives at the dimethyl rare-earth metal
cation [YMe2(THF)5]þ: A DFT investigationq
Ahmed Yahia a, b,1, 2, Mathias U. Kramer c, 3, Jun Okuda c, **, Laurent Maron b, *
a
Institut de Chimie Séparative de Marcoule, CEA, CNRS, UM2, ENSCM, Site de Marcoule, BP17171, F-30207 Banogls sur Cèze, France
Université de Toulouse et CNRS, INSA, UPS, CNRS; UMR 5215 LPCNO, 135 avenue de Rangueil, F-31077 Toulouse, France
c
Institute of Inorganic Chemistry, RWTH Aachen University, Landoltweg 1, D-52056 Aachen, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 April 2010
Received in revised form
29 August 2010
Accepted 1 September 2010
Available online 15 September 2010
Applying DFT methods, the reductive elimination reaction of bis(pyridyl) rare-earth metal cation [Ln(h2-C,
N-pyridyl)2(THF)3]þ (Ln ¼ Y, La) was studied. The effect of both electron-donating and electron-withdrawing
substituents in the para-position of the pyridine ring was considered. An alternative mechanism for the CeC
coupling reaction between pyridine derivatives (pyridine, DMAP and 4-trifluoromethyl-pyridine) and
[LnMe2(THF)5]þ (Ln ¼ Y or La) is suggested. The reaction involves a single electron reductive CeC coupling to
form of Ln(II) complexes with a bipyridine radical anion.
Ó 2010 Elsevier B.V. All rights reserved.
Keywords:
Group 3 metal
CeC coupling
Reductive elimination
Single electron reduction
Computational chemistry
1. Introduction
In contrast to late transition metals, reductive elimination at d0
early transition metal centers involving two hydrocarbyl ligands under
CeC bond formation is not common. Early transition metal complexes
often react with N-heterocycles to lead to ortho-metalated complexes
[1e11]. Two examples for reductive coupling of pyridyl functions were
reported in the literature: Teuben et al. [2] have shown that thermolysis of Cp*2Y(h2-C,N-pyridyl) with excess pyridine led to the formation
of 2,20 -bipyridine under H2 elimination. Cummins et al. [12] observed
a CeC coupling between two pyridine ligands in the presence of H2 in
tantalum(V) chemistry, leading to a bipyridine complex. This reaction
clearly involved a twoelectron CeC reductive elimination of the two
pyridyl ligands, resulting in a tantalum(III) complex. This reaction was
possible, since tantalum(III) is an accessible oxidation state. Recently,
Carver and Diaconescu [13] reported a similar observation in scandium
(III) chemistry. A CeC coupling between two pyridine ligands was
q In memoriam Herbert Schumann
* Corresponding author. Fax: þ33 561 559 697.
** Corresponding author. Tel.: þ49 0 241 894645; fax: þ49 241 80 92 644.
E-mail addresses: jun.okuda@ac.rwth-aachen.de (J. Okuda), laurent.maron@
irsamc.ups-tlse.fr (L. Maron).
1
Fax: þ33 466 797 611.
2
Fax: þ33 561 559 697.
3
Fax: þ49 241 80 92 644.
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.09.003
reported to give a dearomatized bipyridine ligand. These authors also
repeated experiments similar to that reported by Teuben et al. and
confirmed the formation of bipyridine. Since oxidation state þ I is
highly unstable for group 3 metals, Carver and Diaconescu proposed
that the above reactions occur via the formation a dearomatized
bipyridine. Schumann et al. [14] suggested the possible formation of an
ytterbium complex, exhibiting a dianionic bipyridine ligand. Moreover,
a recent study by Booth et al. [15] have shown that coordination of
bipyridine derivatived to Cp*2Yb was leading to a single electron
transfer between divalent ytterbium and the bipyridine ligand to
trivalent ytterbium. Here the þII and þIII oxidation states of ytterbium
are accessible, but the low-lying p* orbital of the bipyridine ligand
appears to be important in the CeC coupling reaction in group 3
complexes. We have studied the reductive and non-reductive type of
reactions and proposed an alternative mechanism for CeC coupling
reaction between pyridine derivatives (pyridine, DMAP, and 4-trifluoromethyl-pyridine) and [LnMe2(THF)5]þ (Ln ¼ Y or La). The reaction appear to involve a single electron reductive CeC coupling to form
a Ln(II) complex with a bipyridine radical anion.
2. Computational details
Based on our previous work, it has been shown that lanthanide center could be treated with f-in-core ECPs [16e21]. Thus, Y
and La were treated with a Stuttgart-Dresden pseudo-potential in
A. Yahia et al. / Journal of Organometallic Chemistry 695 (2010) 2789e2793
2790
combination with its adapted basis set [22,23]. In both cases, the
basis set has been augmented by a set of polarization function
f. Carbon, oxygen and hydrogen atoms have been described with
a 6e31G(d,p) double-z basis set [24]. Calculations were carried
out at the DFT level of theory using the hybrid functional B3PW91
[25,26]. Geometry optimizations were carried out without any
symmetry restrictions, and the nature of the extrema (minima)
was verified with analytical frequency calculations. All these
calculations were performed with the Gaussian 03 [27] suite of
programs. The electronic density has been analyzed using the
Natural Bond Orbital (NBO) technique [28].
3. Results and discussion
Cationic dimethyl complexes [LnMe2(THF)5]þ undergo a variety
of reactions, such as ortho CeH activation of pyridines derivatives
[29], benzophenone [30] as well as olefin and 1,3-diene insertions
[31]. A previous theoretical/experimental study on CeH activations
of pyridine derivatives [29] demonstrated that the reaction is
exergonic and the activation barriers were accessible with around
20e25 kcal mol 1. The cationic complex was thus able to undergo
two successive ortho-metalation of two pyridine derivatives to
form [Ln(h2-C,N-pyridyl)2(THF)3]þ. A similar intermediate was
postulated in the two electrons reductive CeC coupling reported by
Cummins et al. [12] Coupling of two anions is not a trivial task, as
electroneelectron repulsion between the two lone pairs has to be
overcome. Moreover, after forming the CeC bond, two electrons
remains that have to occupy molecular orbitals. In the case of group
3 or lanthanide complexes where oxidation state þ I is not stable,
two possibilities are conceivable (Scheme 1).
Either there is a single reduction of the metal center and formation
of bipyridine radical anion (Scheme 1, left), or there is no reduction of
the metal center (Scheme 1, right). In the former case, triplet spin state
will be formed, whereas a singlet spin state is formed in the later case.
The two electronic distributions as well as the thermodynamics of the
reaction, are defined by the relative energetic positions of the empty
orbitals (namely the d on the metal and p* on the bipyridine ligand).
This relative position of the empty orbitals is a parameter that can be
adjusted by changing the substituents on the pyridine. A donor
substituent such as in the case of DMAP will increase the p* orbital
energy, whereas an electron-withdrawing ligand such as in the case of
4-(trifluoromethyl)pyridine, will decrease the p* orbital energy. The
reference is pyridine. From the experimental point of view, only
results with DMAP are well-behaved and no CeC coupling product
was isolated. For pyridine, the results are not clear. Side reactions
occurred so that CeC coupling products could not be identified. Finally
for 4-(trifluoromethyl)pyridine, the CeH bond activation reaction was
slower than for pyridine or DMAP, so that no attempts were made to
identify CeC coupling products. In the case of DMAP, the mechanism
for a non-reductive CeC coupling, leading to a dearomatized bipyridine complex, will be detailed for La (the profile for pyridine is given
in Fig. 1). Starting from the bis(ortho-metalated) product 1, the coordination of a pyridine molecule is favorable (roughly 10 kcal mol 1 for
π∗
4d
4d
π∗
Ln
bipyridine
Ln
bipyridine
Scheme 1. Possible distributions of two electrons during the CeC coupling reactions
between two anions.
DMAP 2a and up to 23 kcal mol 1 for pyridine 2b). The activation
barrier for the non-reductive CeC coupling is predicted to be
23.2 kcal mol 1 for DMAP 3a and 27.4 kcal mol 1 for pyridine 3b,
which is in line with kinetically accessible reactions. From the pyridine
derivatives adduct 2, the reaction is endergonic by 10.8 kcal mol 1 for
DMAP 4a and 5.4 kcal mol 1 for pyridine 4b. This is in agreement with
the lack of CeC coupling reactivity observed with DMAP.
However, from the dearomatized bipyridine product 4, a ring
opening reaction can occur and has to be investigated. The reaction
is endergonic by 5.9 kcal mol 1 for DMAP 7a (16.7 kcal mol 1 from
the bipyridine derivatives adduct) and 14.2 kcal mol 1 for pyridine
7b (19.6 kcal mol 1 from the adduct). Thus, this ring opening
reaction is predicted to be not favorable. The reaction of nonreductive CeC coupling reaction is unlikely to occur with excess of
pyridine derivatives. The non-reductive CeC coupling reaction was
computed with only two equivalents of pyridine derivatives (from
the mixed pyridylemethyl complex). The activation barrier for the
CeC coupling 5 is predicted to be 2.0 kcal mol 1 higher than the
one for CeH bond activation by the methyl. The reaction is also
predicted to be athermic, whereas CeH bond activation is exergonic by 20 kcal mol 1 [29]. Thus, the non-reductive CeC coupling
is clearly not a possible reaction path. This is different from the
mechanism proposed by Carver and Diaconescu [13]. An alternative
reductive reaction mechanism was considered.
The reactions are predicted to be endergonic by more than
10 kcal mol 1 for DMAP and athermic for pyridine (Fig. 2). This is in
agreement with the lack of reaction for DMAP. The reactions
energies are in line with the energetic position of the p* orbital of
the bipyridine ligand formed. In the case of the DMAP, the amido
substituent is an electron donor increasing the energetic level of
the p* orbital. This disfavors all complexes where this p* orbital is
occupied. As mentioned earlier, in both cases a singlet and a triplet
spin state CeC coupling products have been located on the
Potential Energy Surface (PES). The singlet state can be described as
a Ln(III) complex of the bipyridyl anion (two electrons located in
the p* orbital), whereas the triplet state is a Ln(II) complex of the
bipyridine radical anion (one electron located in the p* orbital). The
singlet state is h4-coordinated (coordination by two N and the two
C to the metal center) and the triplet state is only h2-coordinated
(coordination only by the two N) (Fig. 3). In the case of DMAP, the
triplet spin state is found to be lower than the singlet state, as
expected (Fig. 2). It is noteworthy that the singlet-triplet gap is
larger for La than for Y. This may be due to the fact that the 5d
orbitals are higher in energy than the 4d orbitals due to relativistic
effects, so that the 5d orbitals are closer in energy to the p* orbital
than the 4d orbitals. For pyridine, the p* orbital is lower in energy
than that of DMAP, so that the singletetriplet gap is decreased. At
this stage, it is hard to predict for pyridine if the triplet state would
be lower than the singlet state. The calculations indicate that the
singlet state is in fact lower than the triplet state. As expected, the
gap is lower for La than for Y.
From the kinetic point of view, the reactions are predicted to
proceed with a relatively high activation barrier 10 (around
40 kcal mol 1 for DMAP and around 37 kcal mol 1 for pyridine).
The barriers are high so that the reactions are predicted to be
kinetically difficult. Including solvent effects (THF), by means of
continuum model (CPCM) [32,33], leads to a decrease of the barrier
by 5e6 kcal mol 1, indicating a kinetically accessible reaction for
pyridine. Thus, the CeC coupling reaction could occur for pyridine,
but it is of interest to test whether the reaction is more favorable
with an electron-withdrawing substituent on the pyridine ring. The
reaction with 4-(trifluoromethyl)pyridine was investigated (Fig. 4).
Contrary to the pyridine and DMAP cases, the CeH bond activation
reactions were not discussed in a previous publication for 4-(trifluoromethyl)pyridine and will be described here.
A. Yahia et al. / Journal of Organometallic Chemistry 695 (2010) 2789e2793
X
X=
2791
X
NMe2 (a)
H (b)
N
N
La(thf)
N
N
8
X
X
X
X
X
X
N
N
X
N
-21.2
-36.3
La(thf)2
La(thf)3
N
N
3
-23.8
-38.6
X
1
-26.7
-43.9
X
-50.0
-46.6
N
N
La(thf)2
TS
N
N
X
5
TS
X
X
TS
N
-32.3
-50.6
-49.1
-66.1
X
X
X
X
N
7
-43.2
-51.9
X
N
(thf)2La
N
N
La(thf)2
N
N
N
X
N
-59.9
-71.3
La(thf)3
6
X
N
La(thf)2
4
X
N
2
X
Fig. 1. Gibbs free energy profile (in kcal mol 1) for the non-reductive CeC coupling reaction for pyridine (lower value) and DMAP (upper value). The reference energy is taken from
[LnMe2(THF)5]þ þ 2 pyridine derivatives [29].
The CeH bond activation of the first molecule of 4-(trifluoromethyl)pyridine is predicted to be thermodynamically favorable (Fig. 4), as in pyridine and DMAP cases [29]. From the kinetic
point of view, the activation barriers are found to be accessible and
rather close to what found for the two other substrates. Thus, the
first CeH bond activation is found to be thermodynamically (16) and
kinetically (14) accessible. Similarly to the other substrates, the CeH
activation of a second molecule is predicted to be also kinetically (18)
Fig. 2. Free energy profile (in kcal$mol 1) for the C-C coupling reaction for pyridine
(X¼H) and DMAP (X¼NMe2). The red values account for La and in blue for Y.
and thermodynamically (20) favorable. The formation of the bis
(ortho-metalated) product is predicted to be favorable. As for the two
other substrates, the non-reductive CeC coupling is endergonic and
only the reductive pathway is reported.
As expected, the reaction is found to be slightly exergonic or
endergonic, depending on the spin state of the product (Fig. 4). This
is in agreement with a low-lying p* orbital. A difference between
the two metals is also reported. Indeed, for yttrium, the singlet state
22 is the most stable, whereas the triplet state 23 is the most stable
for lanthanum. This is related to the difference of energetic positions of the 5d and 4d orbitals. From the kinetic point of view, the
calculated barriers to form 21 are high in the gas phase (around
35 kcal mol 1 for both metals), which is slightly lower than for
Fig. 3. Optimized structures of the two possible CeC coupling products.
A. Yahia et al. / Journal of Organometallic Chemistry 695 (2010) 2789e2793
2792
CF3
∆rG° (kcal/mol)
H
M= Y
La
N
Me
thf
thf
M
thf
+31.6
+25.3
CF3
14
thf
thf
Me
TS
13
F3C
N
Me
18
thf
thf
N
M
thf
M
thf
thf
Me
+9.8
+4.5
0.0
17
F3C
M
thf
M
-13.7
-15.2
+
F3C
thf
Me
TS
N
-11.9
-10.4
CF3
N
thf
CH4
thf
thf
thf
thf
N
N
TS
thf
thf
M thf
thf
Me
CF3
N
H
-0.2
-0.5
N
N
CF3
F3C
CF3
Me
thf
21
thf
thf
Me
thf
thf
thf
thf
M
-20.0
-21.7
thf
thf
thf
Me
15
-26.2
-27.8
F3C
N
thf
16
thf
M
thf
thf
Me
+
thf + CH4
F3C
N
thf
CH4
-37.6
-37.0
thf
M
thf
N
19
thf
-46.7
-47.2
CF3
22
23
Singlet
Triplet
-51.2
-44.0
-41.4
-48.7
F3C
N
M
CF3
F3C
thf
N
N
thf
M
thf
20
N
thf
CF3
+ CH4 +
thf
thf
thf
þ
Fig. 4. Gibbs free energy profile for the reaction of [LnMe2(THF)5] with two molecules of 4-(trifluoromethyl)pyridine. Numbers for the reaction with yttrium is in above and
lanthanum below.
pyridine. Moreover, solvent effects decrease the barrier by
5e6 kcal mol 1, leading to kinetically accessible reactions.
For lanthanum, a single electron reductive CeC coupling reaction can take place, which is a rather unique type of reactivity. The
non-reductive CeC coupling leading to dearomatized bipyridine
complexes, as reported by Carver and Diaconescu [13], is not
competitive in the case of methyl cations [LnMe2(THF)5]þ. The
reactivity of such cations with other substrates such as N-
A. Yahia et al. / Journal of Organometallic Chemistry 695 (2010) 2789e2793
methylimidazole is currently under investigation, to see if reductive
coupling is still preferred over the non-reductive pathway.
Acknowledgment
This work was supported by the Fonds der Chemischen Industrie and the Deutsche Forschungsgemeinschaft. A. Y. and L.M. thank
the CINES, CCRT and CALMIP for generous grant of computing time.
L.M. is member of the Institut Universitaire de France.
References
[1] S. Arndt, B.R. Elvidge, P.M. Zeimentz, T.P. Spaniol, J. Okuda, Organometallics 25
(2006) 793e795.
[2] B.J. Deelman, W.M. Stevels, J.H. Teuben, M.T. Lakin, A.L. Spek, Organometallics
13 (1994) 3881e3891.
[3] K.H. Denhaan, Y. Wielstra, J.H. Teuben, Organometallics 6 (1987) 2053e2060.
[4] R. Duchateau, E.A.C. Brussee, A. Meetsma, J.H. Teuben, Organometallics 16
(1997) 5506e5516.
[5] R. Duchateau, C.T. vanWee, J.H. Teuben, Organometallics 15 (1996) 2291e2302.
[6] B.R. Elvidge, S. Arndt, P.M. Zeimentz, T.P. Spaniol, J. Okuda, Inorg. Chem. 44
(2005) 6777e6788.
[7] W.J. Evans, J.H. Meadows, W.E. Hunter, J.L. Atwood, J. Am. Chem. Soc. 106
(1984) 1291e1300.
[8] K.C. Jantunen, B.L. Scott, J.C. Gordon, J.L. Kiplinger, Organometallics 26 (2007)
2777e2781.
[9] M.E. Thompson, S.M. Baxter, A.R. Bulls, B.J. Burger, M.C. Nolan, B.D. Santarsiero,
W.P. Schaefer, J.E. Bercaw, J. Am. Chem. Soc. 109 (1987) 203e219.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
2793
P.L. Watson, J. Chem. Soc. Chem. Commun. (1983) 276e277.
P.M. Zeimentz, J. Okuda, Organometallics 26 (2007) 6388e6396.
H. Sen Soo, P.L. Diaconescu, C.C. Cummins, Organometallics 23 (2004) 498e503.
C.T. Carver, P.L. Diaconescu, J. Am. Chem. Soc. 130 (2008) 7558.
I.L. Fedushkin, T.V. Petrovskaya, F. Girgsdies, R.D. Köhn, M.N. Bochkarev,
H. Schumann, Angew. Chem. Int. Ed. 38 (1999) 2262e2264.
C.H. Booth, M.D. Walter, D. Kazhdan, Y.J. Hu, W.W. Lukens, E.D. Bauer, L. Maron,
O. Eisenstein, R.A. Andersen, J. Am. Chem. Soc. 131 (2009) 6480e6491.
L. Maron, O. Eisenstein, J. Phys. Chem. A 104 (2000) 7140e7143.
L. Maron, O. Eisenstein, J. Am. Chem. Soc. 123 (2001) 1036e1039.
L. Maron, L. Perrin, O. Eisenstein, J. Chem. Soc. Dalton Trans. (2002) 534e539.
L. Maron, E.L. Werkema, L. Perrin, O. Eisenstein, R.A. Andersen, J. Am. Chem.
Soc. 127 (2005) 279e292.
L. Perrin, L. Maron, O. Eisenstein, Inorg. Chem. 41 (2002) 4355e4362.
E.L. Werkema, E. Messines, L. Perrin, L. Maron, O. Eisenstein, R.A. Andersen,
J. Am. Chem. Soc. 127 (2005) 7781e7795.
M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 85 (1993) 441e450.
M. Dolg, H. Stoll, A. Savin, H. Preuss, Theor. Chim. Acta 75 (1989) 173e194.
W.J. Hehre, R. Ditchfie., J.A. Pople, J. Chem. Phys. 56 (1972) 2257.
A.D. Becke, J. Chem. Phys. 98 (1993) 5648e5652.
J.P. Perdew, Y. Wang, Phys. Rev. B. 45 (1992) 13244e13249.
M.J. Frisch, et al., Gaussian 03, Revision B.05, Gaussian, Inc., Wallingford CT, 2004.
A.E. Reed, L.A. Curtiss, F. Weinhold, Chem. Rev. 88 (1988) 899e926.
M.U. Kramer, A. Yahia, L. Maron, J. Okuda, Comptes Rendus Chim. 13 (2010)
626e632.
M.U. Kramer, D. Robert, S. Arndt, P.M. Zeimentz, T.P. Spaniol, A. Yahia,
L. Maron, O. Eisenstein, J. Okuda, Inorg. Chem. 47 (2008) 9265e9278.
S. Arndt, K. Beckerle, P.M. Zeimentz, T.P. Spaniol, J. Okuda, Angew. Chem. Int.
Ed. 44 (2005) 7473e7477.
V. Barone, M. Cossi, J. Phys. Chem. A 102 (1998) 1995e2001.
M. Cossi, N. Rega, G. Scabland, V. Barone, J. Comput. Chem. 24 (2003) 669e681.