Synthesis, characterization, interionic structure, and self-aggregation tendency of zirconaaziridinium salts bearing long alkyl chains

Article abstract of DOI:10.1021/om100827z

Zirconaaziridinium inner sphere ion pairs (ISIPs) bearing alkyl chains of different lengths, [Cp₂Zr(η²-CH₂NR ¹R²)⁺...X^-] (R¹ = Me, R² = C₂H₅

Full text of DOI:10.1021/om100827z

1
00 Organometallics 2011, 30, 100–114
DOI: 10.1021/om100827z
Synthesis, Characterization, Interionic Structure, and Self-Aggregation
Tendency of Zirconaaziridinium Salts Bearing Long Alkyl Chains
Luca Rocchigiani, Gianfranco Bellachioma, Gianluca Ciancaleoni, Alceo Macchioni,*
Daniele Zuccaccia, and Cristiano Zuccaccia*
Dipartimento di Chimica, Universit aꢀ degli Studi di Perugia, via Elce di Sotto 8, I-06123 Perugia, Italy
Received August 25, 2010
Zirconaaziridinium inner sphere ion pairs (ISIPs) bearing alkyl chains of different lengths,
NC2
2
1
2 þ
-
1
2
-
-
1
[
R =C H , X =MeB(C F ) , Zr
Cp Zr(η -CH NR R )
X ] (R = Me, R = C H , X = MeB(C F ) , Zr BN; R = Me,
2 5 6 5 3
2
2
3 3 3
2
-
-
NC16
1
2
BN; R =R =C H , X =MeB(C F ) , Zr
-
-
NC18
BN;
NC18
BT;
1
6
33
2
6
5 3
18 37
6 5 3
1
-
-
NPh
1
2
R =Me, R =Ph, X =MeB(C F ) , Zr BN; R =R =C H , X =B(C F ) , Zr
R = Me, R = Ph, X = B(C F ) , Zr BT), have been synthesized by the reaction of
6
-
-
6
5 3
18 37
6 5 4
1
2
-
-
NPh
5 4
þ
-
þ
-
[
Cp ZrMe
2
MeB(C F ) ] or [Cp ZrMe
6
B(C F ) ] with the suitable tertiary amine followed
6 5 4
3 3 3
5 3
2
3 3 3
by selective C-H activation of one Me group of the coordinated amine and methane elimination.
Subsequent reaction of ISIPs with THF afforded the corresponding outer-sphere ion pairs (OSIPs)
NC2
[Zr THF][BN], [Zr
NC16
NC18
NPh
THF][BN], [Zr THF][BN], [Zr
NC18
(
[
THF][BN], [Zr
Zr THF][BT]) in which the anion has been displaced into the second coordination sphere. The
THF][BT],
NPh
interionic structure in solution (i.e., the relative cation-anion position) and the self-aggregation level
1
19
1
of both ISIPs and OSIPs have been investigated by means of H NOESY, F, H HOESY, and
diffusion PGSE NMR methods, in low polar solvents. It is found that, independent of the nature or
length of the alkyl chains, the anion prefers to pair with the cation from the side of the nitrogen atom
and THF in ISIPs and OSIPs, respectively. Self-aggregation of ion pairs into higher aggregates is also
weakly influenced by the nature of the alkyl chains, but it is strictly connected with the nature of the
-
counterion: ISIPs and OSIPs bearing B(C F ) showed an increased tendency to form higher
6
5 4
-
aggregates with respect to those containing MeB(C F ) . The level of the self-aggregation in
6
5 3
benzene-d and cyclohexane-d has been quantified by fitting the hydrodynamic volumes obtained
6
12
from NMR diffusion experiments with several models of indefinite self-association. The IK model, in
which the equilibrium constants slightly increase on increasing the aggregation step, best describes
0
the experimental trends in all cases. The standard Gibbs free energy of self-association at 297 K (ΔG )
NC18
-1
-1
for Zr
BN is -6.7 ( 0.2 kJ mol and -12.6 ( 0.7 kJ mol in benzene-d and cyclohexane-d ,
6
12
-1
0
-1
respectively. In the same solvents, the values of ΔG are -17.9 ( 0.5 kJ mol and -25.0 ( 0.3 kJ mol
NC18
NC18
for [Zr
form of ion pairs in cyclohexane-d at analytical concentrations of 10 and 10 M, respectively.
THF][BT]. At room temperature, 80% and 4% of [Zr
THF][BT] are present in the
-
5
-3
1
2
NC18
The self-aggregation of [Zr
temperature.
THF][BN] in cyclohexane-d12 is strongly depressed by increasing the
Introduction
cation-anion interaction in these ion pairs is of great
importance for the activity, stability, and selectivity of the
þ
-
2
It is generally accepted that [Cp MR X ] metallo-
cenium ion pairs are the active catalysts in homogeneous
Ziegler-Natta olefin polymerization. The strength of the
catalytic system because the nature of the anion may
profoundly alter the relative rates of the elementary catalytic
2
3 3 3
1
3
steps. Since the ion pair symmetrization process (i.e., the
process that exchanges the anion from one side of the
zirconocenium cation to the other) is sensitive to the strength
of the cation-anion interaction, the dynamic of ion pair
symmetrization has been extensively studied by NMR meth-
ods, in an attempt to correlate anion mobility with catalytic
*
To whom correspondence should be addressed. E-mail: cristiano.
zuccaccia@unipg.it.
1) (a) Gibson, V. C.; Spitzmesser, S. K. Chem. Rev. 2003, 103, 283.
b) Pedeutour, J.-N.; Radhakrishnan, K.; Cramail, H.; Deffieux, A. Macro-
mol. Rapid Commun. 2001, 22, 1095. (c) Gladysz, J. A., Ed. Chem. Rev.
000, 100 (special issue on “Frontiers in Metal-Catalyzed Poly-
merization”). (d) Marks, T. J., Stevens, J. C., Eds. Top. Catal. 1999,
5 and references therein. (e) Britovsek, G. J. P.; Gibson, V. C.; Wass,
(
(
2
(2) (a) Chen, E. Y. X.; Marks, T. J. Chem. Rev. 2000, 100, 1391.
(b) Bochmann, M. J. Organomet. Chem. 2004, 689, 3982.
(3) (a) Chen, M.-C.; Roberts, J. A.; Marks, T. J. J. Am. Chem. Soc.
2004, 126, 4605and references therein. (b) Roberts, J. A. S.; Chen, M.-C.;
Seyam, A. M.; Li, L.; Zuccaccia, C.; Stahl, N. G.; Marks, T. J. J. Am. Chem.
Soc. 2007, 129, 12713.
1
D. F. Angew. Chem., Int. Ed. 1999, 38, 428. (f) Kaminsky, W.; Arndt, M.
Adv. Polym. Sci. 1997, 127, 144. (g) Bochmann, M. J. Chem. Soc., Dalton
Trans. 1996, 255. (h) Brintzinger, H. H.; Fischer, D.; M €u lhaupt, R..; Rieger,
B.; Waymouth, R. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143.
pubs.acs.org/Organometallics
Published on Web 12/14/2010
r 2010 American Chemical Society

Article
Organometallics, Vol. 30, No. 1, 2011 101
4
performances. In particular, the experimental observations
that an increased metallocenium concentration or addition
of an external salt caused an entropic acceleration of the ion
function of the isotactic propylene chain length, in a silicon-
bridged zirconocene ion pair, was predicted by Klesing and
Bettonville through DFT (density functional theory) calcu-
5
14
pair symmetrization process led Brintzinger and co-workers
to propose that ionic aggregates larger than ion pairs can be
formed in appreciable amounts under catalytic conditions
lations. An additional layer of complexity is faced when
solvents with relative permittivity (ε ) lower than that of
r
benzene are taken into account. These solvents, as hexanes or
isoparaffins, are by far the most used reaction media in in-
dustrial applications, but structural and dynamic investiga-
tions of ion pairs in these solvents are particularly chal-
6
and may take part in the polymerization process. The latter
hypothesis was further reinforced by diffusion NMR experi-
ments, which showed the formation of ion quadruples from
zirconocenium ion pairs in benzene-d at concentrations
around 10 M. Successively, in collaboration with Marks
and co-workers, we demonstrated that ion pairs bearing
1
5
lenging. With the goal of obtaining information about the
position of the counterion during the growth of the polymer
chain and the tendency of ion pairs bearing a polymer chain
to self-aggregate, we decided to synthesize model zircono-
cenium ion pairs with alkylic chains of different length and
6
-
3
7
-
weakly coordinating anions, such as MeB(C F ) and FAl-
6
5 3
-
[
inner-sphere ion pairs) exhibit a negligible tendency to self-
2-(C F )(C F )] , in the first coordination sphere (ISIPs=
6 5 6 4 3
8
16
investigate them through advanced NMR techniques. We
focused on two commonly used anions, namely MeB-
9
associate in benzene-d . In contrast, ion pairs in which the
counterion resides in the second coordination sphere (OSIPs=
6
-
17
- 18
(C F )
and B(C F ) ,
6 5 4
which are representative of
weakly and very weakly coordinating anions, respectively. In
6
5 3
8
outer-sphere ion pairs) showed a marked tendency to self-
aggregate, but a negligible percentage of higher aggregates
-
the absence of a stabilizing donor ligand, MeB(C F ) is
6
5 3
-4
are generated at catalytic concentration values (<10 M) in
usually found to form loosely associated ISIPs with highly
electrophilic early-transition-metal centers by way of M-Me
agostic interactions. More recently it has been observed that
1
0
benzene-d6. In general, it seems that the higher the charge
2
separation in the ion pairs, the higher their tendency to form
11
-
higher aggregates. Thus, if the search for large counter-
anions with a dispersed charge, which are less and less
the MeB(C F ) may also form OSIPs in the absence of a
6
5 3
9
1
nucleophile. The presence of long alkyl chains on the cation
rendered zirconocenium ion pairs soluble in cyclohexane,
providing us the unique possibility to explore their self-
aggregation tendency in a reaction medium with character-
istics very similar to those used in industrial plants.
1
2
coordinating, reduces the possibility that the anion com-
petes with the substrate for the coordination at the metal, on
the one hand, it increases the charge separation and the
possibility of forming larger aggregates. Clearly, also an
1
3
enhanced dimension of the cation may contribute to reduce
the ion-pair strength, thereby increasing the charge separa-
tion and the tendency to self-aggregate. In this respect, it
should be noted that, during the polymerization process, a
rapid growth of the polymer chain in the proximity of the
catalytic site necessarily occurs. Consequently, the enhanced
dimension of the cation in the active species may lead to an
increase of ion pair separation and self-aggregation ten-
dency. For example, an increase of the dipole moment as a
The Results and Discussion section of the paper is divided
into three parts. First, a description of the synthesis and char-
acterization in solution of the ion pairs, from both the intra-
molecular and interionic points of view, is presented. Second,
a detailed examination of the self-aggregation behavior is dis-
cussed. Finally, a quantitative analysis of the self-aggregation
process, including relevant thermodynamic parameters and
catalyst speciation in terms of ion pairs or higher order
aggregates, is offered. When possible, the experimental
results based on NMR are contrasted and analyzed with
respect to the outcome of DFT calculations.
(4) For a few examples see: (a) Yang, X.; Stern, C. L.; Marks, T. J.
J. Am. Chem. Soc. 1991, 113, 3623. (b) Siedle, A. R.; Newmark, R. A.
J. Organomet. Chem. 1995, 497, 119. (c) Deck, P. A.; Beswick, C. L.;
Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1772. (d) Jia, L.; Yang, X.; Stern,
C. L.; Marks, T. J. Organometallics 1997, 16, 842. (e) Mohammed, M.; Nele,
M.; Al-Humydi, A.; Xin, S.; Stapleton, R. A.; Collins, S. J. Am. Chem. Soc.
Results and Discussion
1
. Synthesis and Characterization of Zirconaaziridinium
2003, 125, 7930. (f) Song, F.; Lancaster, S. J.; Cannon, R. D.; Schormann, M.;
Ion Pairs. Schemes 1 and 2 illustrate the strategy for the
synthesis of zirconaaziridinium ion pairs, which incorporate
alkyl chains of different length into the cation. The initial
step of the synthesis is the methide abstraction (or proto-
nolysis) from the complex Cp ZrMe (Zr , Cp = cyclo-
2 2
pentadienyl anion) using a suitable Lewis (B(C F ) and
6
Humphrey, S. M.; Zuccaccia, C.; Macchioni, A.; Bochmann, M. Organo-
metallics 2005, 24, 1315. (g) Alonso-Moreno, C.; Lancaster, S. J.; Zuccaccia,
C.; Macchioni, A.; Bochmann, M. J. Am. Chem. Soc. 2007, 129, 9282.
(
h) Alonso-Moreno, C.; Lancaster, S. J.; Wright, J. A.; Hughes, D. L.;
Me
2
Zuccaccia, C.; Correa, A.; Macchioni, A.; Cavallo, L.; Bochmann, M.
Organometallics 2008, 27, 5474.
5 3
(
5) Beck, S.; Lieber, S.; Schaper, F.; Geyer, A.; Brintzinger, H. H.
J. Am. Chem. Soc. 2001, 123, 1483.
6) (a) Song, F.; Cannon, R. D.; Bochmann, M. J. Am. Chem. Soc.
003, 125, 7641. (b) Song, F.; Cannon, R. D.; Lancaster, S. J.; Bochmann, M.
J. Mol. Catal. A 2004, 218, 21.
7) Beck, S.; Geyer, A.; Brintzinger, H. H. Chem. Commun. 1999,
477.
þ
-
þ
-
-
CPh3 B(C F ) ) or Brønsted (HN(Me) Ph B(C F )
6
)
5 4
2
6
Me
5 4
þ
(
acid. This leads to [Cp ZrMe
2
MeB(C F ) ] (Zr BN,
6 5 3
3 3 3
and [Cp ZrMe
-
- 17a
þ
-
2
BN = MeB(C F )
Me
)
Zr BT, BT =B(C F ) ) ISIPs. The latter are treated
B(C F ) ]
6 5 4
6
5 3
2
20
3 3 3
-
-
(
6
5 4
(
2
(
(
8) Macchioni, A. Chem. Rev. 2005, 105, 2039.
9) Stahl, N. G.; Zuccaccia, C.; Jensen, T. R.; Marks, T. J. J. Am.
(14) Klesing, A.; Bettonville, S. Phys. Chem. Chem. Phys. 1999, 1,
2373.
(15) (a) Beswick, C. L.; Marks, T. J. Organometallics 1999, 18, 2410.
(b) Beswick, C. L.; Marks, T. J. J. Am. Chem. Soc. 2000, 122, 10358.
(16) Rocchigiani, L.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A.
Chem. Eur. J. 2008, 14, 6589.
(17) Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116,
10015.
(18) Chien, J. C. W.; Tsai, W.-M.; Rausch, M. D. J. Am. Chem. Soc.
Chem. Soc. 2003, 125, 5256.
10) Zuccaccia, C.; Stahl, N. G.; Macchioni, A.; Chen, M.-C.;
Roberts, J. A.; Marks, T. J. J. Am. Chem. Soc. 2004, 126, 1448.
11) (a) Zuccaccia, D.; Bellachioma, G.; Cardaci, G.; Ciancaleoni,
G.; Zuccaccia, C.; Clot, E.; Macchioni, A. Organometallics 2007, 26,
930. (b) Bellachioma, G.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.;
Macchioni, A. Coord. Chem. Rev. 2008, 252, 2224.
12) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066and
references therein.
13) Price, C. J.; Chen, H.-Y.; Launer, L. M.; Miller, S. A. Angew.
Chem., Int. Ed. 2009, 48, 956.
(
(
3
(
1991, 113, 8570.
(
(19) Ciancaleoni, G.; Fraldi, N.; Budzelaar, P. H. M.; Busico, V.;
Macchioni, A. Dalton Trans. 2009, 41, 8824.

1
02 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
a
Scheme 1. Syntheses of ISIPs
a
6
Unless otherwise stated, the syntheses were performed in benzene-d .
Scheme 2. Syntheses of OSIPs
The disappearance of IBN and formation of the final prod-
uct are accompanied by the release of CH (δH 0.15 ppm).
4
The reaction is complete in a few hours at room temperature.
All the NMR data of the final complex are consistent with
2
þ
the general formula [Cp Zr(η -CH NMeC H )
MeB-
NC16
2
2
16 33 3 3 3
-
NC16
25
BN; Scheme 1). In the complex Zr
(
C F ) ] (Zr
-
6
5 3
BN the resonances due to Cp rings (δH 5.26 and 5.24 ppm)
and methylene hydrogen atoms of both the N-CH -R (δH
2
2
.43 and 2.04 ppm) and Zr-CH -N groups (δ 1.68 and
2
H
2
.37 ppm, J = 8.4 Hz) are not magnetically equivalent.
1
These observations indicate a slow nitrogen inversion pro-
cess on the H NMR time scale. Reasonably, this is due to
the interaction between the nitrogen lone pair and the acidic
zirconium center, which in principle may be intramolecular
HH
1
26
1
with 1 equiv of the suitable tertiary amine NMeR R (R =
2
1
2
Me, R = Et (NC2); R = Me, R = C H (NC16); R =
1
2
1
or intermolecular. The two possibilities are discriminated by
1
analyzing the values of J coupling constants relative to
1
6
33
2
R =C H (NC18); R =Me, R =Ph (NPh)) to afford the
1
2
CH
1
8
37
the carbon atoms in R-positions with respect to the nitrogen.
intermediates IBN and IBT OSIPs (Scheme 1), where the
amine is coordinated to the metal center via the nitrogen lone
1
3
The C NMR resonance of the Zr-CH -N moiety is ob-
2
1
served at 55.4 ppm, while the related J coupling constant
2
1
CH
pair. At room temperature, IBN and IBT spontaneously
transform into the final ISIPs by selective C-H activation of
one methyl group on the tertiary amine and methane
amounts to 150.3 Hz. This value is about 13 Hz larger than
1
those observed for the N-CH ( J
= 137.4 Hz) and
CH
N-CH -R ( J = 137.2 Hz) groups and is diagnostic
3
1
27
2
2
2
CH
elimination. The full course of the reaction was followed
Me
by NMR in the case of the reaction of Zr BN with NC16.
The displacement of the anion into the second coordination
sphere is established by the small difference in chemical shift
(
25) For other examples of metalla-aza-cyclopropanes see: (a) Dreos,
R.; Felluga, A.; Nardin, G.; Randaccio, L.; Tauzher, G. Organometallics
2003, 22, 2486and references therein. (b) Hartshorn, R. M.; Telfer, S. G.
Dalton Trans. 2000, 2801. (c) Tonei, D. M.; Baker, L.-J.; Brothers, P. J.;
Clark, G. R.; Ware, D. C. Chem. Commun. 1998, 2593. (d) Marzilli, L. G.;
Polson, S. M.; Hansen, L.; Moore, S. J.; Marzilli, P. A. Inorg. Chem. 1997,
36, 3854. (e) Lu, C. C.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 15818.
f) Maire, P.; Sreekanth, A.; B €u ttner, T.; Harmer, J.; Gromov, I.; R €u egger, H.;
Breher, F.; Schweiger, A.; Gr €u tzmacher, H. Angew. Chem., Int. Ed. 2006,
5, 3265. (g) Adams, R. D.; Babin, J. E.; Kim, H.-S. Organometallics 1987,
6, 749. (h) Booij, M.; Kiers, N. H.; Meetsma, A.; Teuben, J. H. Organome-
tallics 1989, 8, 2454. (i) Abbenhuis, H. C.L.; Feiken, N.; Haarman, H. F.;
Grove, D. M.; Horn, E.; Spek, A. L.; Pfeffer, M.; van Koten, G. Organome-
tallics 1993, 12, 2227. (j) Liu, G.; Beetstra, D. J.; Meetsma, A.; Hessen, B.
Organometallics 2004, 23, 3914. (k) Knight, P. D.; Munslow, I.; O'Shaughnessy,
P. N.; Scott, P. Chem. Commun. 2004, 894. (l) Rankin, M. A.; McDonald, R.;
Ferguson, M. J.; Stradiotto, M. Organometallics 2005, 24, 4981. (m) Liptau,
P.; Carmona, D.; Oro, L. A.; Lahoz, F. J.; Kehr, G.; Erker, G. Eur. J. Inorg.
Chem. 2004, 4586. (n) Krut'ko, D. M.; Borzov, M. V.; Kirsanov, R. S.;
Churakov, A. V.; Kuz'mina, L. G. J. Organomet. Chem. 2005, 690, 3243.
(26) Belostoskii, A. M.; Aped, P.; Hassner, A. J. Mol. Struct.
(THEOCHEM) 1997, 398-399, 427–434.
1
value between m-F and p-F resonances in the F NMR
9
2
3
spectrum (Δδ(m,p-F)= 2.9 ppm). Amine coordination in
IBN is indicated by the low-frequency shift of the resonance
1
13
due to the N-Me group(s) in both H and C NMR
spectra.
2
4
(
4
3þ
benzene, a weakly coordinating solvent, to afford the intermediate
6
5
4
-
(
20) The activation using CPh B(C
F
)
was carried out in chloro-
þ
-
Me(solv)
[
Cp
A. K.; Jordan, R. F. J. Am. Chem. Soc. 2004, 126, 15360.
21) Schaper, F.; Geyer, A.; Brintzinger, H. H. Organometallics 2002,
1, 473.
2
ZrMe(Cl-C
6
H
5
)
B(C
6
F
5
)
4
] (Zr
BT): Wu, F.; Dash,
3
3 3
(
2
(22) (a) Turner, H. W.; Hlatky, G. G.; Eckman, R. R. U.S. Patent
5
,198,401, 1993. (b) Bertuleit, A.; Fritze, C.; Erker, G.; Fr €o hlich, R.
Organometallics 1997, 16, 2891. (c) Wrobel, O.; Schaper, F.; Wieser, U.;
Gregorius, H.; Brintzinger, H. H. Organometallics 2003, 22, 1320.
1
9
(
23) The value of Δδ(m,p-F) ( F NMR) is a good probe of coordina-
- 0
tion of MeB(C
6
F
5
)
3
to cationic d metals (values of 3-6 ppm indicate
coordination; <3 ppm indicates noncoordination). See: Horton, A. D.;
(27) (a) Bertuleit, A.; Fritze, C.; Erker, G.; Fr o€ hlich, R. Organome-
tallics 1997, 16, 2891. (b) Pflug, J.; Bertuleit, A.; Kehr, G.; Fr €o hlich, R.;
Erker, G. Organometallics 1999, 16, 3818. (c) Pflug, J.; Erker, G.; Kehr, G.;
Fr €o hlich, R. Eur. J. Inorg. Chem. 2000, 1795.
de With, J. Organometallics 1997, 16, 5424and references therein.
(24) Wilson, P. A.; Wright, J. A.; Oganesyan, V. S.; Lancaster, S. J.;
Bochmann, M. Organometallics 2008, 27, 6371and references therein.

Article
Organometallics, Vol. 30, No. 1, 2011 103
Scheme 3
apply to all ion pairs containing the BN anion. However, for
NC18
Zr
BN the distinction between the “up” and “down”
sides is no longer possible. As for the nature of cation-anion
interaction in solution, the situation is more complicated in
the case of complexes containing the BT anion, since no
diagnostic tool has been reported that unambiguously allows
us to discriminate between ISIPs and OSIPs.
Addition of THF to the zirconaaziridinium ion pairs
described above results in a fast and complete anion dis-
placement, leading to the corresponding OSIPs as depicted
in Scheme 2. Only one molecule of THF is incorporated into
the cation even when a large excess of THF is used. The
coordination of THF is indicated by the chemical shift values
of the OCH moiety. In particular, the proton resonance is
2
observed around δH ∼3.3-3.1 ppm, while the carbon reso-
1
Figure 1. Two sections of the H NOESY NMR spectrum of
nance appears at δ ∼78-77 ppm. These values are about 0.4
C
NC16
and 10 ppm lower and higher than in “free” THF, respec-
Zr
6
BN (400.13 MHz, benzene-d , 297 K) showing the strong
3
tively. The zirconaaziridinium skeleton remains intact, as
1
intramolecular interactions between the N-Me moiety and
1
u
2
u
Zr-CH
-N and Cp groups.
demonstrated by the J value of the Zr-CH -N moiety
CH
2
(
∼154 Hz), while the anion displacement is clearly seen for
of the heterocyclopropane character of the zirconaaziridi-
2
ion pairs containing the BN anion by a decrease of the Δδ(m,
p-F) values.
nium framework. In analogy with η -imine complexes of
2
8
zirconium two extreme resonance structures may in prin-
ciple contribute to the description of the iminium-Zr bond,
but that involving the π coordination of the iminium double
bond to a formally Zr(II) center is certainly less important
due to the high π basicity of Zr(II). Full resonance assign-
ment is easily accomplished by considering the plane defined
by the azazirconacyclopropane ring (see Figure 1). For
2
. Structure in Solution of ISIPs and OSIPs. Two struc-
tural arrangements are possible within the zirconaaziridi-
nium cations described above: the outside structure with the
carbon of the Zr-CH -N group interposed between the
anion (or THF) and the nitrogen atom, and the inside struc-
ture with the nitrogen atom occupying the central position
Scheme 3).
In solution, the two possibilities may be discriminated by
means of homo- and heteronuclear NOE NMR measure-
ments. For ISIPs containing the BN anion, strong dipolar
interionic interactions are observed between B-Me and the
o-F atoms of the anion and N-Me, N-CH -R, and both
Cp moieties of the cation in H NOESY and F, H HOESY
2
2
9
32
(
example, starting from the resonance due to the N-Me
u
moiety (δH 1.90 ppm), the signals due to Zr-CH -N and
2
u
Cp (arbitrarily we label as “up” (u) the groups that are
positioned on the same side as the N-Me moiety with
respect to the above-defined plane) are easily recognized by
their intense NOE interaction with the N-Me group
2
19
1
1
(
Figure 1).
NC16
spectra. Figure 2 shows the case of Zr
BN as an example.
u
u
Having identified Zr-CH -N and Cp , the remaining
2
Interionic dipolar interactions are not observed for m-F and
p-F nuclei, indicating that the anion is preferentially oriented
in such a way that the B-Me group steadily points toward
the cation. Notably, we were unable to detect any interionic
contacts involving the Zr-CH -N moiety, indicating that
the inside structure is preferentially present in solution.
Analogous results are obtained for all the ISIPs, including
those containing the BT anion. In the latter case, interionic
interactions involving the Cp(s), N-Me, and N-CH -R
groups are observed for all the anion resonances, those due
to m-F atoms being the strongest.
resonances are assigned from the homo- and/or heteronu-
clear scalar and dipolar connectivities. The formation of the
zirconaaziridinium metallacycle is accompanied by the re-
entering of the anion into the first coordination sphere of the
cation. The chemical shift difference between m-F and p-F
10
2
resonances is again used as a diagnostic tool. The Δδ(m,p-F)
NC16
observed in Zr
BN amounts to 4.4 ppm, indicating that
2
3
an ISIP is formed. In agreement, the proton resonance due
to B-Me appears at 0.07 ppm, a value which is about 1.3 ppm
lower in frequency with respect to that observed for the
2
3
0
“
free” anion. The same NMR arguments described above
The inside coordination mode is also preferred for OSIPs,
as demonstrated by the strong dipolar interactions present in
(
28) Cummings, S. A.; Tunge, J. A.; Norton, J. R. Top. Organomet.
Chem. 2005, 10, 1.
29) Crabtree, R. H. The Organometallic Chemistry of Transition
Metals; Wiley: New York, 2001.
30) (a) Beck, S.; Prosenc, M. H.; Brintzinger, H.-H.; Goretzki, R.;
(
(31) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997,
62, 7512.
(32) The inside and outside nomenclature was first introduced by:
Tatsumi, K.; Nakamura, A.; Hofmann, P.; Staufert, P.; Hofmann, R.
J. Am. Chem. Soc. 1985, 107, 4440.
(
Herfert, N.; Fink, G. J. Mol. Catal. A: Chem. 1996, 111, 67. (b) Beck, S.;
Prosenc, M. H.; Brintzinger, H.-H. J. Mol. Catal. A: Chem. 1998, 128, 41.

1
04 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
1
Figure 2. (left) Two sections of the H NOESY NMR spectrum of Zr
NC16
BN (400.13 MHz, benzene-d , 297 K) showing the interionic
6
d
u
19
1
interactions between the B-Me moiety and N-Me, Cp , and Cp groups. (right) Two sections of the F, H HOESY spectrum of
NC16
Zr
BN (376.47 MHz, benzene-d , 297 K) showing the interionic interactions between the fluorine atoms in ortho positions and the
d
6
u
NC16
N-Me, Cp , Cp (strong), and N-CH
2
-R (weak) groups. The spectra shown here refer to an in situ generation of Zr
BN rather
than to the purified complex. The asterisk denotes the resonance of CH . The small peak close to the B-Me resonance is due to an
4
unknown impurity.
1
the H NOESY spectrum between the O-CH moiety of the
rings adopt a quasi-eclipsed conformation and are partly
tilted toward the side where the THF ligand is coordinated.
Similar considerations apply to the analysis of the inside
2
coordinated THF and both the N-Me and N-CH -R
groups. In agreement, dipolar contacts could not be detected
between THF protons and the Zr-CH -N group.
In order to validate the conclusions based on NOE NMR
measurements, we performed DFT quantum mechanical
geometry optimizations for the inside and outside structures.
2
NC2
NPh
2
and outside optimized geometries of Zr BN and Zr BN.
3
The inside geometry always is the most stable, with the
3
-
1
computed energy differences amounting to 23.8 and 41 kJ mol
NC2
NPh
for Zr
BN and Zr
observed in the case of Zr
BN, respectively. The larger gap
NPh
NC2
Initially, the cation of [Zr THF][BN] was optimized. The
-1
inside
BN most likely reflects the
NC2
inside geometry ([Zr THF]
þ
) is about 7.5 kJ mol
increased steric demand of the phenyl ring with respect to the
ethyl group.
The interionic structures of OSIPs were investigated in
NC2
more stable than the outside geometry ([Zr THF] outside).
þ
NC2
The higher stability of [Zr THF]
þ
seems mainly due
inside
1
19
1
to steric factors, since the sterically demanding NMeEt
moiety is oriented toward a relatively free region of the
space, in front of the bent metallocene (Figure 3, top). In
solution by means of H NOESY and F, H HOESY NMR
experiments. As an example, four relevant sections of the
1
9
1
NC16
F, H HOESY of [Zr
are shown in Figure 4.
THF][BN] recorded in benzene-d6
NC2
þ
contrast, in [Zr THF]
the bulky NMeEt moiety is
outside
oriented toward a more hindered lateral position (Figure 3,
bottom). To partially release such a steric congestion, the Cp
In addition to the strong intramolecular NOE interaction
between B-Me and o-F nuclei, NOE interionic interactions
between o-F, m-F, and, in some cases, p-F nuclei and Cp,
THF, and N-Me protons are observed. As in the case of the
corresponding ISIPs, interionic contacts between the anion
(33) The thermodynamic preference for the inside coordination mode
seems rather general and independent from the nature of the additional
donor ligand. For related zirconaaziridines see for example: (a) Buchwald,
S. L.; Watson, B. T.; Wannamaker, M. W.; Dewan, J. C. J. Am. Chem.
Soc. 1989, 111, 4486. (b) Gately, D. A.; Norton, J. R.; Goodson, P. A. J. Am.
Chem. Soc. 1995, 117, 986. (c) Lubben, T. V.; Pl €o ssl, K.; Norton, J. R.;
Miller, M. M.; Anderson, O. P. Organometallics 1992, 11, 122. On the other
hand, cases have been reported where the ouside coordination geometry is
observed in the solid state. See: (d) Cheung, M.-S.; Chan, H.-S.; Xie, Z.
Organometallics 2005, 24, 5217.
and Zr-CH -N are below detectability. Despite the overlap
2
of the central CH groups of the aliphatic chain with the
protons of THF in β-positions, a careful quantification of the
NOE contacts indicates that they only weakly interact with
2
34
the o-F and m-F nuclei of the counterion. These observations
suggest that the counterion is prevalently located in front of

Article
Organometallics, Vol. 30, No. 1, 2011 105
Figure 3. Optimized geometries of inside (top) and outside
(
NC2
bottom) structures of the [Zr THF] cation.
þ
the cation, partially shifted toward the side of THF. As for
3
the anion orientation, NOE quantification carried out on
4
NPh
the complex [Zr THF][BN] indicates that the average
interionic distance between the B-Me moiety of the anion
˚
and the R-THF protons of the cation is about 4.1 A. In agree-
1
0,35
ment with previous results,
this relatively large average
19 1
Figure 4. Four sections of the F, H HOESY NMR spectrum
interionic distance indicates that the orientation in which the
B-Me moiety of the BN anion points away from the metal
center is favored. Similar interionic interactions are observed
for OSIPs containing the BT anion, indicating that the
interionic structures of all the OSIPs are the same, indepen-
dent of the nature of the R substituents on the nitrogen atom
and the nature of the counterion.
NC16
of [Zr
the relevant interionic interactions. Note the absence of dipolar
interactions between the anion and the Zr-CH -N protons.
6
THF][BN] (376.47 MHz, benzene-d , 297 K), showing
2
NC2
charge separation. The computed dipole moments of [Zr
-
NC2
THF][BN]front and [Zr THF][BN]lateral are 22.6 and 29.6 D,
respectively. This suggests that the positive charge in the
cation is partially shifted toward the nitrogen atom and the
THF ligand.
The conclusions from NMR studies are in good agreement
with the results of DFT calculations carried out on
NC2
Zr THF][BN]. Initially, two ion pairs with different rela-
[
3. Self-Aggregation by PGSE NMR. Self-aggregation was
investigated for all ion pairs shown in Scheme 2 by means of
PGSE NMR measurements (Tables 1 and 2). This technique
allows a fast and reliable estimation of the translational self-
tive cation-anion positions were optimized. The two mini-
ma correspond to [Zr THF][BN]front, where the anion is
NC2
NC2
positioned close to the THF ligand, and [Zr THF]-
[
BN]lateral, where the anion is shifted toward the side of the
diffusion coefficient (D
namic radius (r ) of the diffusing species in solution is de-
rived by means of the modified Stokes-Einstein equation:
) from which the average hydrody-
t
Zr-CH -N group as shown in Figure 5. In both cases the
methyl group of the anion is oriented far away from the metal
center.
H
2
36
In agreement with NOE results, the optimized structure of
NC2
kT
Dt ¼
-1
[
Zr
THF][BN]front is more stable by 10.0 kJ mol , despite
cπηrH
˚
the fact that the Zr-B distance is about 0.25 A shorter in
NC2
Zr THF][BN]lateral than in [Zr THF][BN]front. A rea-
NC2
[
where k is the Boltzmann constant, T is the temperature, c is a
numerical factor that depends on the ratio between r and
the hydrodymanic volume of the solvent (rsolv), and η is the
solution viscosity. From r , the average hydrodynamic
sonable explanation can be found by considering the dipole
moments of the ion pairs. In fact, in low polar solvents, a
higher stability is expected for species having a smaller
H
H
volume (V ) of the diffusing particle is obtained, assuming
H
that the latter has a spherical shape. Finally, the ratio
(34) (a) Macchioni, A. Eur. J. Inorg. Chem. 2003, 195. (b) Ciancaleoni,
G.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A. In Techniques in Inorganic
Chemistry; Fackler, J. P., Jr., Falvello, L., Eds.; CRC Press Taylor & Francis
Group: Boca Raton, FL, 2010; pp 129-180. (c) Neuhaus, D.; Williamson, M.
The Nuclear Overhauser Effect in Structural and Conformational Anal-
ysis; Wiley-VCH: New York, 2000. (c) Macura, S.; Ernst, R. R. Mol. Phys.
(36) (a) Gierer, A.; Wirtz, K. Z. Z. Naturforsch., A: Astrophys., Phys.,
Phys. Chem. 1953, 8, 522. (b) Gierer, A.; Wirtz, K. Z. Z. Naturforsch., A:
Astrophys., Phys., Phys. Chem. 1953, 8, 532. (c) Chen, H.-C.; Chen, S.-H.
J. Phys. Chem. 1984, 88, 5118. (d) Zuccaccia, D.; Macchioni, A. Organo-
metallics 2005, 24, 3476.
1980, 41, 95.
(35) Correa, A.; Cavallo, L. J. Am. Chem. Soc. 2006, 128, 10952.

1
06 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
-
10
2 -1 a
Table 1. Diffusion Coefficients (D
˚
t
, 10
namic Radii (r , A), Hydrodynamic Volumes (V , A ), and
m s ), Hydrody-
3
˚
H
H
Aggregation Numbers (N) for ISIPs and OSIPs in Benzene-d
˚
6
(
rsolv = 2.6 A, ε = 2.28) at 297 K and Different Concentrations
r
-
3
C, 10 M)
(
0
)
compd (V
H
C
D
t
r
H
V
H
N
NC2
Zr BN (829)
0.86
6.29
6.21
6.17
5.82
5.82
5.88
5.91
6.20
825
851
864
998
1.0
1.0
1.0
1.2
2
6
.67
.37
1
3.56
NC16
Zr
BN (1284)
0.6
5.26
5.09
5.11
4.89
6.75
6.95
6.92
7.18
1 290
1 410
1 390
1 550
1.0
1.1
1.1
1.2
2
7
.7
.4
1
3.4
NC18
Zr
BN (2088)
0.56
4.40
4.37
4.34
4.11
7.94
7.94
7.98
8.39
2 100
2 100
2 130
2 470
1.0
1.0
1.0
1.2
1
2
.11
.63
2
5.0
NPh
Zr BN (870)
3.8
5.1
6.13
5.21
5.94
6.81
877
1 320
1.0
1.5
3
NPh
Zr BT (985)
0.1
5.80
5.61
5.35
5.18
4.84
4.34
6.23
6.40
6.66
6.85
7.26
8.0
1 010
1 100
1 240
1 350
1 600
2 140
1.0
1.1
1.3
1.4
1.6
2.2
0
0
1
1
3
.31
.65
.1
.8
.5
b
NC18
Zr
BT (2203)
0.14
4.36
3.81
3.17
2.93
2.55
2.17
1.87
8.0
9.0
10.7
11.5
13.1
15.1
17.6
2 170
3 010
5 060
6 350
9 470
15 050
23 060
1.0
1.4
2.3
2.9
4.3
6.8
10.5
0
2
4
9
.86
.8
.4
.9
1
3
9.8
2.4
NC2
Zr THF][BN] (899)
[
0.09
0.23
0.42
5.98
5.85
5.23
4.86
6.06
932
988
1 300
1 580
1.0
1.1
1.5
1.8
-
1
Figure 5. Optimized geometries and relative energies (in kJ mol
NC2
)
6.18
6.78
7.23
for two different cation-anion orientations of [Zr THF][BN].
b
0
.98
between V and the hydrodynamic volume of the elemental
H
building block that form the supramolecular aggregate (VH )
affords the aggregation number N. Taking into account the
NC16
[Zr
THF][BN] (1354)
0.12
0.32
4.77
4.34
4.30
3.85
3.83
3.44
3.15
2.89
7.34
7.81
8.04
8.90
8.95
9.89
1 660
1 990
2 180
2 950
3 000
4 050
5 190
6 570
1.2
1.5
1.6
2.2
2.2
3.0
3.8
4.9
0
0
1
1
3
.59
.6
.8
.88
low polarity of the solvents (benzene ε =2.28, cyclohexane
r
ε = 2.04), dissociation of ion pairs into free ions can be
r
3
neglected. Consequently, we can reasonably assume that
7
6.2
8.45
10.7
11.6
b
the elemental aggregating unit is the whole ion pair for all the
0
ISIPs and OSIPs. VH values for ISIPs containing the BN
NC18
[
Zr
THF][BN] (2158)
0.11
0
4.17
3.84
3.50
3.10
2.74
2.68
2.37
8.28
8.91
9.73
2 380
2 960
3 860
5 420
7 795
8 160
11 600
1.1
1.4
1.8
2.5
3.6
3.8
5.4
anion were obtained from the VH values at different con-
0
.49
38,39
centrations by extrapolation at infinite dilution.
VH
1.09
3.02
values for ISIPs containing the BT anion were estimated
by subtracting the hydrodynamic volume of the BN anion
from the corresponding ISIP and then adding the hydro-
10.9
6
7
.63
.07
12.3
12.5
14.0
3
40
11.7
˚
dynamic volume of the BT anion (605 A ). The hydro-
dynamic volume of the BN anion was estimated by con-
sidering that the ratio between the van der Waals volumes of
NPh
[Zr THF][BN] (940)
0.67
5.01
7.05
1 470
1.6
NPh
Zr THF][BT] (1055)
[
[
0.56
.3
4.79
4.26
7.2
8.1
1 540
2 240
1.5
2.1
1
-
5
37) For example, it is known that, at concentrations below 10 M,
(
the fraction of ammonium salt existing as free ions amounts to one part
per million or less: Copenhafer, D. T.; Kraus, C. A. J. Am. Chem. Soc.
NC18
Zr
THF][BT] (2273)
0.26
3.76
3.33
3.20
2.65
2.18
9.1
10.2
10.5
12.6
15.2
3 130
4 390
4 920
8 390
14 760
1.4
1.9
2.2
3.7
6.5
0
1
.67
.55
1
951, 73, 4557.
38) Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A.
Organometallics 2007, 26, 3624and references therein.
39) Macchioni, A.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.
Chem. Soc. Rev. 2008, 37, 479.
40) Rocchigiani, L.; Bellachioma, G.; Ciancaleoni, G.; Crocchianti,
(
6.74
13.9
(
a
The reported values have been corrected for changes in solu-
tion viscosity due to the increased ion pair concentration using the
(
3
6d,50 b
S.; Lagan ꢀa , A.; Zuccaccia, C.; Zuccaccia, D.; Macchioni, A. ChemPhys-
Chem 2010, 11, 3243.
residual solvent resonance as an internal standard.
solution.
Saturated

Article
Organometallics, Vol. 30, No. 1, 2011 107
-
10
2 -1
Table 2. Diffusion Coefficients (D
˚
t
, 10
Radii (r , A), Hydrodynamic Volumes (V , A ), and Aggre-
m s ), Hydrodynamic
3
˚
H
H
gation Numbers (N) for ISIPs and OSIPs in Cyclohexane-d12
˚
a
(
rsolv = 2.8 A, ε = 2.04) at 297 K and Different Concentrations
r
3
-
(
C, 10 M)
0
H
compd (V
)
C
D
t
r
H
V
H
N
NC16
Zr
BN (1490)
0.083
3.36
3.37
3.31
3.22
3.00
2.77
7.08
7.08
7.18
7.35
7.81
8.38
1 490
1 490
1 550
1 660
1 990
2 460
1.0
0
0
3
5
8
.26
.95
.25
.42
.3
1.0
1.0
1.1
1.3
1.7
NC18
Zr
BN (2620)
0.076
2.67
2.71
2.69
2.62
2.35
8.66
8.55
8.61
8.82
9.72
2 720
2 620
2 670
2 870
3 840
5 560
1.0
1.0
1.0
1.1
1.5
2.1
0
1
3
8
.5
.45
.36
.2
Figure 6. Trends of the aggregation number (N) as a function of
the analytical concentration (C) for BN-containing ISIPs (solid
symbols) and OSIPs (open symbols) in benzene-d at 297 K.
1
5.8
2.06 11.0
6
NC16
NC18
b
[
[
Zr
THF][BN] (2620)
0.036
2.62
8.82
2 870
1.8
is higher than 11.7 mM in benzene-d . A coherent picture is
6
Zr
THF][BN] (2690)
0.05
2.01 11.2
1.44 15.4
1.22 18.2
1.11 19.8
0.97 22.8
0.78 28.3
0.68 32.3
0.65 33.8
5 910
15 400
25 200
32 700
49 800
94 700
141 000
162 000
2.2
5.7
9.4
12.2
18.5
35.2
52.4
60.2
derived from NOE and PGSE measurements in benzene-d6.
NOE experiments indicate that the position and distance of
the counterion from the metal center is little affected by the
nature of the R substituent; consequently the latter has no
appreciable effect on the overall dipole moment and self-
aggregation tendencies of ion pairs observed by means of
diffusion experiments.
0
1
2
4
.90
.60
.47
.75
1
1
2
1.8
9.0
3.2
NC18
3
.1.2. Ion Pairs Containing the BT Anion. The data of N at
different concentrations in benzene-d for ion pairs bearing
[
Zr
THF][BT] (2760)
0.26
1.76 12.7
1.32 16.8
1.02 22.0
0.78 28.0
0.72 30.7
0.64 34.4
0.54 40.9
8 640
19 890
44 100
91 610
121 020
171 170
3.1
7.2
16.0
33.2
43.9
62.0
0
4
7
.8
.27
.9
6
the BT anion are reported in Table 1. The variation of
NPh
the aggregation number of OSIPs ([Zr THF][BT] and
1
1
2
0.33
7.26
5.23
NC18
Zr
[
THF][BT]) is very similar to that observed for all
287 000 104.0
the other OSIPs containing the BN anion. It thus appears
that the nature of the anion has only a marginal effect on the
self-aggregation of OSIPs. This is in agreement with the
NOE data discussed above, where it was noted that the inter-
ionic structures of OSIPs are substantially the same for both
the BN and BT anions. As discussed above, the BN anion
possesses an increased rotational freedom when it is forced
into the second coordination sphere. Consequently, only
small differences are expected between the dipole moment
of corresponding OSIPs having either the BT or the BN
anion (for example, the computed dipole moments of
a
b
Benzene (80 mM) was added as PGSE internal standard. Saturated
solution.
BN and BT anions (VvdW(BT)/VvdW(BN)=1.236) is equal to
the ratio between their corresponding hydrodynamic vol-
0
umes. Finally, VH values for all the OSIPs were computed
3
by adding the hydrodynamic volume of THF (70 A ) to the
˚
0
VH values of the corresponding ISIPs.
3.1. Effect of Anion and R Substituents on Nitro-
gen. 3.1.1. Ion Pairs Containing the BN Anion. The trends
of N vs C (concentration) in benzene-d for ion pairs bearing
6
the BN anion (Figure 6) clearly show different behaviors of
4
ISIPs and OSIPs, as previously found.
For ISIPs, N deviates very little from 1, indicating a
negligible tendency to self-aggregate. In solution these spe-
cies are mainly present as ion pairs and a small amount of
higher aggregates start to be present only at fairly high
NC2
NC2
[Zr THF][BN] and [Zr THF][BT] are 22.3 and 25.9 D,
respectively) and a similar tendency to form larger ionic
aggregates is observed.
d,9,10
NPh
NC18
The ion pairs Zr BT and Zr
BT appear to behave
NPh
differently from either ISIP or OSIP. N values of Zr BT
substantially deviate from 1 when the concentration is
increased, even if the aggregation is slightly reduced with
NPh
NPh
NPh
concentration (N = 1.5 at 35.1 mM for Zr BN). In con-
respect to those of [Zr THF][BN] and [Zr THF][BT].
NPh
For example, the N value of Zr BT is 2.2 at 3.5 mM, while
trast, N values of OSIPs rapidly increase when the concen-
tration is increased, approaching 2 already at concentrations
around 1 mM. This indicates an increased tendency to form
higher order ionic aggregates. Nevertheless, N values for
NPh
that of [Zr THF][BT] is 2.1 at 1.3 mM (Table 1). A similar
NC18
trend is observed for Zr
is much higher than that of Zr
BT: its tendency to self-aggregate
NC18
BN and close to those of
THF][BT]. This can be ex-
-
5
NC18
[Zr
NC18
OSIPs approaches 1 at high dilution (ca. 10 M), indicating
the prevalence of ion pairs at very low concentration. Con-
cerning the effect of the R substituents on the nitrogen atom,
the experimental data show that increasing the length of the
alkyl chain or substituting an alkyl with an aryl group has no
effect on self-aggregation. On the other hand, the nature of
the alkyl chain dramatically changes the solubility of
OSIPs: for example, concentration of a saturated solution of
THF][BN] and [Zr
plained by considering that the dipole moments of BT-
containing ISIPs are much higher than those of the corre-
sponding BN-containing ISIPs and are similar to the dipole
moments of OSIPs. DFT calculations carried out on ion
pairs bearing the NC2 moiety are consistent with the pre-
vious hypothesis. In fact, two minima were found for
NC2
Zr BT. The ion pair in which the anion interacts with
the metal center through one fluorine atom in a meta
NC2
NC18
[
Zr THF][BN] is around 1 mM and that of [Zr
THF][BN]

1
08 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
Table 3. Equations To Calculate the Weight Fraction of the
Monomer (Α) and Aggregation Number (N) for EK, AK, and IK
Models
a
model
R
N
2
2
EK
AK
IK
R L - R(2L þ 1) þ 1 = 0
(1 - RL)
L/[(1/R) - 1]
L/[ln(L þ 1)]
R - e- = 0
R(L þ 1) - 1 = 0
L is defined as L = KC, where C is the analytical molar concentra-
tion of the monomer.
RL
a
(1.6 mM) in cyclohexane-d , N amounts to 9.4. Aggregates
containing an average of 104 ion pairs are present for
12
NC18
[
Zr
tration, while, at submillimolar concentration (0.05 mM),
THF][BT] in cyclohexane-d12 at the highest concen-
NC18
NC16
Figure 7. Trends of the aggregation number (N) as a function of
the analytical concentration (C) for ISIPs (filled symbols) and
OSIPs (open symbols) in cyclohexane-d12 at 297 K.
N is still 2.2 for [Zr
THF][BN]. Although [Zr
THF][BN]
was not soluble enough in cyclohexane-d12 to allow a com-
plete PGSE study as a function of the concentration, the
N value measured for its saturated solution (0.04 mM)
amounts to 1.8 and fits rather well in the N versus C trends
NC2
m
position (Zr BT ) is about 8.4 kJ mol more stable than
-1
NC18
NC18
that where the anion approaches the cation with one fluorine
of [Zr
THF][BN] and [Zr
THF][BT] (Figure 7, inset).
NC2
p
atom in a para position (Zr BT ). In agreement with
We conclude that also in cyclohexane-d12 the chain length
has a negligible effect on the self-aggregation tendency of
OSIPs.
NC2
diffusion data, the computed dipole moment of Zr BT
m
NC2
is 24.4 D. This value is very similar to that of [Zr THF]-
NC2
BN]front (22.6 D) and much higher than that of Zr BN
[
3.3. Thermodynamic Parameters of the Self-Aggregation
Process. From the discussion above, it is clear that, as the
total concentration is increased, N steadily increases to
values much larger than 2 for OSIPs. Thus, the aggregation
process proceeds beyond the formation of ion quadruples.
The tendency of ISIPs and OSIPs to self-aggregate has been
(17.1 D). On a closer look, it also appears that the aggrega-
tion tendency of Zr
NPh
NC18
BT is slightly larger than that of
Zr BT, suggesting an higher degree of charge separation
NC18
in Zr
BT. This can just be due to the higher steric requi-
rements of the two C18 tails, but an electronic contribution is
also possible. It may be hypothesized, for example, that one
of the two C18 tails folds back toward the metal center and
establishes a weak agostic interaction between one of the
42
quantified using models of indefinite self-association. Ion
pairs were considered as the elemental aggregating unit
(A) that may self-aggregate forming ion quadruples (A2),
CH moieties of the chain and the zirconium. This would
2
ion hextuples (A ), etc... as represented by the following
3
result in the displacement of the anion into the second
coordination sphere with a consequent increase of charge
separation within the ion pair. A weak support for this
hypothesis comes from NMR spectroscopy. In fact, the
equations:
½
A2ꢀ
A þ A ¼ A2 K2 ¼
A þ A ¼ A K ¼
2
½Aꢀ
two N-CH -R moieties of the two C18 tails appear as
2
1
separate signals in the H NMR spectrum (δ 2.65 and 2.50
H
½
A3ꢀ
3
ppm), suggesting that the plane defined by the azazircona-
cyclopropane ring no longer represents a plane of symmetry
2
3
3
K2½Aꢀ
NC18
for Zr
3
BT.
.2. Effect of Solvent Polarity. The enhanced solubility of
:
::
zirconaaziridinium ion pairs bearing long alkyl chains
prompted us to explore their self-aggregation behavior in a
solvent having a relative permittivity smaller than that of
benzene. We chose cyclohexane, because its features (aliphatic
½
Aiꢀ
A
þ A ¼ Ai K ¼
i - 1
i
i - 1
Q
i
½
Aꢀ
Ki
character and dielectric constant, ε =2.04) are very similar
r
i ¼ 2
4
1
to those of commercial ISOPAR commonly used in in-
dustrial plants for olefin polymerizations. Diffusional and
hydrodynamic data in cyclohexane-d12 for ion pairs contain-
ing C16 and C18 tails are reported in Table 2 and Figure 7.
As can be seen from Table 2, the self-aggregation tendency
of ISIPs is only slightly increased on passing from benzene-d6
to cyclohexane-d . For example, for a 7.4 mM solution of
Three different indefinite self-association models were taken
43
into account. In the EK (equal K) model each step is char-
acterized by the same equilibrium constant (K =K =...=
2
3
K =K ), while, in the AK (attenuated K) model, the equili-
E
i
brium constants decrease following the relation 2K =3K =
2
3
1
2
NC16
Zr
same concentration (5.4 mM) in cyclohexane-d , N amounts
BN in benzene-d N is equal to 1.1, while, at almost the
6
(
42) Martin, R. B. Chem. Rev. 1996, 96, 3043and references therein.
12
(43) For a review concerning the application of the indefinite self-
association models in supramolecular polymerizations see: (a) De Greef,
T. M. A.; Smulders, M. M. J.; Wolffs, M.; Schenning, A. P. H. J.;
Sijbesma, R. P.; Meijer, E. W. Chem. Rev. 2009, 109, 5687and references
therein. For other examples see: (b) Rymden, R.; Stilbs, P. Biophys. Chem.
1985, 21, 145. (c) Stokkeland, I.; Stilbs, P. Biophys. Chem. 1985, 22, 65.
to 1.3. In sharp contrast, the self-aggregation of OSIPs is
remarkably higher in cyclohexane-d12 than in benzene-d .
6
NC18
While for a 1.1 mM solution of [Zr THF][BN] in
benzene-d N is equal to 1.8, at almost the same concentration
6
(
4
d) Regan, D. G.; Chapman, B. E.; Kuchel, P. W. Magn. Reson. Chem. 2002,
0, S115. (e) Zhao, D.; Moore, J. S. Chem. Commun. 2003, 807. (f) Chen, Z.;
(41) ISOPAR is a trademark of Exxon Mobil Corporation. For more
Stepanenko, V.; Dehm, V.; Prins, P.; Siebbeles, L. D. A.; Seibt, J.; Marquetand, P.;
information see: http://www.exxonmobilchemical.com
Engel, V.; W €u rthner, F. Chem. Eur. J. 2007, 13, 436.

Article
Organometallics, Vol. 30, No. 1, 2011 109
Table 4. Limiting Equilibrium Constants and Corresponding Standard Gibbs Free Energies at 297 K for the Indefinite Self-Aggregation
(
EK, AK, and IK Models) of ISIPs and OSIPs Bearing C18 Tails in Benzene-d and Cyclohexane-d
6
12
model
EK
AK
IK
a
E
0
b
a
0
b
a
0
b
compd
K
-ΔG (EK)
K
A
-ΔG (AK)
K
I
-ΔG (IK)
Benzene-d
6
NC18
c
d
[
[
[
[
Zr
Zr
Zr
Zr
][BN]
][BT]
8.49 ( 0.55 (0.005 623)
2 250 ( 783 (2.175)
1 628 ( 302 (0.639)
2 156 ( 743 (0.865)
5.3 ( 0.2
19.1 ( 0.9
18.3 ( 0.5
19.0 ( 0.9
19.09 ( 4.19 (0.006 286)
73 613 ( 62 205 (5.65)
17 124 ( 12 500 (1.77)
32 084 ( 27 375 (2.27)
7.3 ( 0.5
27.8 ( 2.1
24.2 ( 1.8
25.7 ( 2.1
15.2 ( 0.96 (0.005 955)
1 145 ( 61 (0.514 7)
1 248 ( 72 (0.285 6)
1 392 ( 268 (0.708)
6.7 ( 0.2
NC18
NC18
NC18
17.5 ( 0.2
17.7 ( 0.2
17.9 ( 0.5
THF][BN]
THF][BT]
Cyclohexane-d12
334 ( 182 (0.43)
NC18
NC18
NC18
[
[
[
Zr
Zr
Zr
][BN]
114 ( 53 (0.327)
12 ( 1
29.0 ( 0.7
30.8 ( 1.3
14.4 ( 1.4
160 ( 42 (0.23)
15 589 ( 595 (3.18)
24 286 ( 2 941 (13.44)
12.6 ( 0.7
23.9 ( 0.1
25.0 ( 0.3
THF][BN] 122 991 ( 34 358 (13.82)
THF][BT] 250 751 ( 133 438 (34.22)
a
-1
b
-1
c
d
In M . In kJ mol . Error intervals are given at 95% confidence. The numbers in parentheses represent the Euclidean norm of the residuals with
respect to the experimental N values.
4K = ... = iK = K . Finally, in the IK model, successive
4 i A
equilibrium constants mildly increase according to the rela-
tion 2K =3K /2=4K /3=... = iK /(i - 1)=K . From these
2
3
4
i
I
relationships and the equations above, the weight fraction of
the monomer (R) and the aggregation number (N) can be
analytically formulated. Related expressions are given in
Table 3. In each formulation, R is given as a function of
the analytical concentration C and the corresponding equi-
librium constant K via the variable L=KC.
All three models of indefinite self-aggregation described
above have been used for the single-parameter (K , K , or K )
nonlinear least-squares fit of N data as a function of the
E
A
I
analytical concentration using a Fortran code reported in a
4
0
recent paper. Initially, ion pairs bearing C18 tails were taken
into account, because, thanks to the high solubility of these
complexes, the experimental data extend over a relatively large
concentration range in both benzene and cyclohexane. The
results of the fitting procedure are reported in Table 4. As an
example, Figure 8 shows the best-fit curves obtained for
NC18
NC18
Zr
BN and [Zr
We found that the IK model always gives the best fit to the
THF][BN] in cyclohexane-d .
12
NC18
experimental data points: only in the case of Zr
benzene-d is the best fit obtained with the EK model. The
BN in
6
fact that better fits are obtained with the IK model suggests a
sort of mild cooperativity, where the addition of an ion pair
to a given aggregate is somewhat progressively more exer-
gonic as the average size of the aggregate increases. From the
computed limiting equilibrium constants (K ), the corre-
I
sponding standard Gibbs free energies of self-association
0
Figure 8. Trends of the aggregation number (N) as a function of
NC18
(
quantitative analysis is in good agreement with the qualita-
ΔG ) at 297 K are calculated and shown in Table 4. The
the analytical concentration (C) for [Zr
THF][BN] in benzene-
d6 (top) and cyclohexane-d12 (bottom) at 297 K. Solid lines
represent best fits to the data using the EK (red), AK (green),
and IK (blue) indefinite self-aggregation models.
0
tive discussion reported above. In benzene-d , ΔG of
6
NC18
-1
0
Zr
proximately -17.7 kJ mol for Zr
BN amounts to -6.7 kJ mol . ΔG decreases to ap-
-1 NC18 NC18
BT, [Zr THF]-
THF][BT] as a consequence of an increased
NC18
[BN], and [Zr
charge separation. Changing the solvent from benzene-d to
given analytical concentration using the equation
6
i
i - 1
cyclohexane-d12 resulted in a remarkable increase of the aggre-
NC18
Ri ¼ R L
43b,44
gation tendency.
0
For instance, the ΔG of [Zr
THF]-
-
1
[BT] is -25 kJ mol in cyclohexane-d .
Figure 9 shows the distributions of i-mers at four different
-
1
2
6
analytical concentrations ranging from 10 M to 10 M for
-3
From the values of R and K , the distribution of i-mers in
I
NC18
terms of their weight fractions (R ) can be computed at a
[Zr
THF][BT] in benzene-d (top) and cyclohexane-d
6 12
i
(bottom). The distributions are plotted to an oligomerization
order (i) up to 10, even though the model accounts for an
indefinite self-association.
(
44) (a) Haino, T.; Tanaka, M.; Fukazawa, Y. Chem. Commun. 2008,
4
68. (b) Kraft, A.; Osterod, F.; Frh €o lich, R. J. Org. Chem. 1999, 64, 6425.

1
10 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
-
4
In benzene-d the total amount of aggregates larger than
6
simple ion pairs becomes appreciable only at concentrations
larger than 10 M. In contrast, the relative concentrations
of quadruples and larger aggregates become substantial
-
5
around 20%) already at 10 M in cyclohexane-d . At
(
the analytical concentration of 10 M, only 40% and 4% of
12
-
3
NC18
[
Zr
cyclohexane-d , respectively.
THF][BT] is present as ion pairs, in benzene-d and
6
1
2
To gain further insight into the thermodynamics of the
aggregation process, temperature-dependent NMR diffu-
NC18
sion measurements were performed for [Zr
in cyclohexane-d . Table 5 summarizes D , r , V , N, K ,
THF][BN]
1
2
t
H
H
I
0
and ΔG values at 297, 300.2, 313.6, and 328.3 K, as well as
0
an estimation of the standard enthalpy (ΔH ) and stan-
0
dard entropy (ΔS ) obtained from the van’t Hoff plot
(
Figure 10). Increasing the temperature results in a drastic
reduction of the aggregation tendency. At 11.8 mM, N for
NC18
[
Zr
THF][BN] decreases from 35.2 to 11.9 on passing
from 297 to 328.3 K. Quantitative analysis using the van’t
Hoff approximation shows that the limiting aggregation step
0
-1
is characterized by a ΔH value of about -33 ( 10 kJ mol
.
Figure 9. Estimated weight fractions (R ) of i-mers (1-mer = ion
i
pair, 2-mer = ion quadruple, 3-mer = ion hextuples, ...) at
-
6
C equal to 10 M (black), 10 M (red), 10 M (green), and
-5
-4
-
3
NC18
1
cyclohexane-d12 (bottom).
0
M (blue) for [Zr
THF][BT] in benzene-d
6
(top) and
Figure 10. van’t Hoff plot for the equilibrium A þ A = Ai
i-1
(i f ¥) according to the IK model.
-
10
2
-1
˚ ˚
t H H
Table 5. Diffusion Coefficients (D , 10 m s ), Hydrodynamic Radii (r , A), Hydrodynamic Volumes (V , A ), Aggregation Numbers (N),
3
-
1
0
-1
0
-1
0
-1 -1
a
) Values for [Zr
NC18
b
and K
I
(M ), ΔG (kJ mol ), ΔH (kJ mol ), and ΔS (J mol
K
at Different Temperatures (T, K) and Concentrations (C, 10 M)
r
THF][BN] in Cyclohexane-d12 (ε = 2.04)
3
-
0
0
0
T
C
D
t
r
H
V
H
N
K
I
-ΔG
-ΔH
-ΔS
297.0
0.05
2.01
1.44
1.22
1.11
0.97
0.78
0.68
0.65
11.2
15.4
18.2
19.8
22.8
28.3
32.3
33.8
5 910
15 400
25 200
32 700
49 800
94 700
141 000
162 000
2.2
5.7
9.4
12.2
18.5
35.2
52.4
60.2
15 589 ( 595
23.9 ( 0.1
33 ( 10
31 ( 33
0
1
2
4
.90
.60
.47
.75
1
1
2
1.8
9.0
3.2
3
3
3
00.2
0.9
1.44
1.22
1.10
0.98
0.79
15.5
18.2
20.2
22.5
27.8
15 600
25 200
34 400
47 600
89 900
5.8
9.4
12.8
17.7
33.4
15 368 ( 2 274
7 024 ( 352
3 764 ( 639
23.9 ( 0.4
21.9 ( 0.1
20.4 ( 0.4
1
2
4
.6
.47
.75
1
1
1
1.8
13.6
0.9
1.84
1.61
1.42
1.21
0.96
12.2
13.8
15.7
18.2
23.0
7 510
10 900
16 000
25 200
50 800
2.8
4.1
6.0
9.4
1
2
4
.6
.47
.75
1.8
18.8
28.3
0.9
2.19
2.01
1.70
1.38
1.13
10.4
11.2
13.1
16.1
19.7
4 680
5 850
9 500
17 300
31 900
1.7
2.2
3.5
6.4
1
2
4
.6
.47
.75
1.8
11.9
a
b
Error intervals are given at 95% confidence. Benzene (80 mM) was added as PGSE internal standard.

Article
At the 95% confidence level, the error associated with ΔS is
Organometallics, Vol. 30, No. 1, 2011 111
0
chlorobenzene-d
5
over CaH
rage Schlenkwith aPTFEvalve. Cp
2
) and vacuum-transferred to a sto-
ZrCl (g98%) and LiMe (1.6
-1
-1
rather large, but the average negative value of -31 J mol
K
2
2
M solution in Et O) were purchased from Aldrich and used
2
without further purifications. The liquid amines (Aldrich)
strongly suggests an enthalpy-driven self-aggregation pro-
cess. Most likely, electrostatic interactions give a larger
contribution to the overall interaction enthalpy, but, con-
sidering the large differences observed on passing from
benzene-d to cyclohexane-d , other weak forces, including
NMe
thaw degassed after 12 h of stirring over CaH
transferred to a storage Schlenk. The solid amine NMe(C H )
37 2
2
Et, NMe
2
C
H33, and NMe C H
16 2 6 5
were freeze-pump-
2
and vacuum-
18
6
12
(
CPh
6 5 3
g99%, Aldrich) was used without further purification. B(C F ) ,
van der Waals and solvophobic interactions, may also be
important.
þ
-
þ
-
3
B(C F )
6 5 4
, and HNMe C H
2 6 5
6 5
B(C F )
4
were obtained
as a gift from Dow Chemical. B(C F ) was purified by vacu-
5 3
6
5
-
um sublimation (60 ꢀC, 5 ꢁ 10 mmHg), while the other
2
Me 45
activators were used as received. Cp
2
ZrMe
2
(Zr
were prepared ac-
)
and
Conclusions
þ
-
Me
] (Zr BN)
17a
[Cp
cording to the literature.
2
ZrMe
MeB(C
6
5
F )
3
3
3 3
Up to now, the interionic structure and level of aggrega-
tion of metallocenium ion pairs have been studied mostly by
varying the nature of the anion and/or the metallocene
1
13
19
One- and two-dimensional H, C, and F NMR spectra
were measured on a Bruker DRX 400 spectrometer. Referencing
1
13
19
1
3
is relative to external TMS ( H and C) and CCl F ( F). H
4
NOESY NMR experiments were acquired by the standard
6
skeleton, whereas only Me or CH TMS groups were used
2
4
f-h,7,9,10
47
three-pulse sequence or by the PFG version. Two-dimensional
as models for the polymer chain.
In this work, we
1
9
1
F, H HOESY experiments were acquired using the standard
have investigated zirconaaziridinium ion pairs bearing long
alkyl chains of different lengths that would better mimic the
growing polymer chain. The azazirconacyclopropane struc-
ture imparts high stability to the cation, allowing us to
compare the behavior of ion pairs containing either the BN
and BT counterions under strictly similar conditions. The
results indicate that for BN-containing ISIPs and for all
OSIPs increasing the length of the alkyl chain or substituting an
alkyl with an aryl group has no effect on the self-aggregation
tendency, since the anion position and, consequently, the
dipole moment of the ion pair are almost unaffected by the
length of the chain. Instead, it seems that there is a small
increment of the self-aggregation tendency on passing from
4
8
four-pulse sequence or its modified version. The number of
transients and data points were chosen according to the sample
concentration and desired final digital resolution. Qualitative or
semiquantitative two-dimensional Overhauser spectra were re-
corded using a 1 s relaxation delay and 800 ms mixing time.
NMR samples for standard NMR characterization were pre-
pared inside the glovebox by dissolving the suitable amount of
compound in approximately 0.7 mL of the suitable deuterated
solvent.
2
þ
Synthesis and Characterization. [Cp
2
Zr(η -CH
2
NMeEt)
3
3 3
-
NC2
Me
6 5 3
MeB(C F ) ] (Zr BN). In the glovebox, Zr BN (25 mg, 32.6
mmol) was loaded into a J. Young NMR tube and dissolved in
approximately 0.7 mL of benzene-d . A 1.5 equiv amount of
6
NPh
NC18
NMe Et (5.3 μL) was added by a micrometric syringe, and the
mixture was strongly shaken. In a few minutes, a dense oily phase
2
Zr BT to Zr
BT, suggesting that the steric/electronic
properties of the different R substituents on the nitrogen
atom may weakly influence the strength of the cation-anion
interaction in the case of ISIP containing the weakly coordi-
nating BT anion. The self-aggregation is strongly affected by
the nature of the anion only in the case of ISIPs. The bulkier
and less coordinating BT anion affords ion pairs displaying
aggregation tendencies much higher than those of BN-
containing ISIPs and similar to those of OSIPs. The effect of
the solvent is dramatic: the level of aggregation increases
settled to the bottom of the tube and gas evolution was observed.
The supernatant solution was removed, and the residue was
rinsed with benzene (3 ꢁ 1 mL) to remove the small excess of
the amine. After 12 h, the J. Young NMR tube was interfaced to
the high-vacuum line and the solvent removed in vacuo. Finally,
the residue was rinsed with pentane (3 ꢁ 1 mL) and dried under
high vacuum. The reaction product was an oil. A 2.7 mM solution
NC2
of Zr BN in benzene-d
6
showed no sign of decomposition after
7
days.
1
H NMR (400.13 MHz, benzene-d
6
, 297 K, J in Hz): δ 5.28
about 10 times on passing from benzene (ε = 2.28) to
r
u
s, Cp ), 5.25 (s, Cp ), 2.40 (m, NCH
d
b
a
(
(
2
), 1.93 (m, NCH
2
), 1.87
d=8.3,
cyclohexane (ε = 2.02). As a consequence, the percentage
2
s, NMe), 1.61 (d, J
d
2
r
=8.3, ZrCH ), 1.34 (d, J
2 H
u
H
u
-H
d
-H
of ion quadruples or higher ionic aggregates in cyclohexane
is not negligible even at very low concentration. For instance,
the percentage of ion quadruples and higher aggregates is
u
3
ZrCH ), 0.57 (br t, J
ω 2
H -NCH
= 7.0, ωMe), 0.11 (s, BMe).
2
13
1
u
C{ H} NMR (100.55 MHz, benzene-d
6
, 297 K): δ 110.3 (Cp ),
), 54.6 (ZrCH ), 48.0 (NMe), 9.1
(ωMe), -6.3 (BMe). F NMR (376.65 MHz, benzene-d , 297
d
1
10.2 (Cp ), 64.8 (NCH
2
2
-5
19
about 20% for a 10 M solution of OSIPs in cyclohexane.
Finally, the self-aggregation is strongly depressed by increasing
6
3
K, J in Hz): δ -132.8 (brd
J
oF-mF = 21.8, o-F), -161.6
0
-1
0
3
the temperature (ΔH =-33 ( 10 kJ mol and ΔS =-31 (
(t JpF-mF=20.7, p-F), -165.8 (m, m-F).
2
þ
-
NC16
] (Zr
-
1
-1
NC18
[
Cp
2
Zr(η -CH
2
NMe(C16
H
33))
MeB(C
6
F
5
)
3
BN).
In the glovebox, Zr BN (100 mg, 130.5 mmol) was loaded into
3
3 J mol
K
for [Zr
THF][BN] in cyclohexane). This
3 3 3
Me
dependence on temperature suggests that the presence of
ionic aggregates larger than ion pairs may be negligible in
high-temperature solution polymerization process usually
used in industrial plants.
a screw-top vial and 10 mL of benzene was added. The solution
was lightly shaken, and 1.5 equiv of NMe (C16H33) (60 μL) was
2
added with a micrometric syringe. After an intense shake, a
dense oily phase was allowed to settle to the bottom of the vial.
The supernantant solution was removed, and the residue was
rinsed with benzene (3 ꢁ 2 mL). After 1 day, the oil was trans-
ferred to a flip frit apparatus interfaced to the high-vacuum line,
washed with pentane, and finally dried. The reaction product
was an oil.
Experimental Section
General Procedures.All manipulations of air-sensitive materials
were performed in flamed Schlenk glassware on a Schlenk
-
5
line, interfaced to a high-vacuum pump (10 mmHg), or in
a nitrogen-filled glovebox with a high-capacity recirculator
(
(
45) Samuel, E.; Raush, M. D. J. Am. Chem. Soc. 1973, 95, 6263.
46) Jeener, J.; Meier, B. H.; Bachmann, P.; Ernst, R. R. J. Chem.
2 2
(<1 ppm of O and H O). All solvents were preventively distilled
Phys. 1979, 71, 4546.
after 12 h of reflux over Na and freeze-pump-thaw degassed
over Na/K alloy. Deuterated solvents were freeze-pump-thaw
(
(
47) Wagner, R.; Berger, S. J. Magn. Reson. A 1996, 123, 119.
48) Lix, B.; Sonnichsen, F. D.; Sykes, B. D. J. Magn. Reson. A 1996,
€
degassed (benzene-d
6
and cyclohexane-d12 over Na/K alloy,
121, 83.

1
12 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
1
2
þ
-
NC18
H NMR (400.13 MHz, benzene-d
6
, 297 K, J in Hz): δ 5.26
[Cp
20 mg portion of Zr
2
Zr(η -CH
2
N(C18
H
37
)
(0.079 mmol) and 0.95 equiv of CPh
2
)
B(C
6
F
5
)
4
] (Zr
BT). A
B-
3
3 3
d
s, Cp ), 5.24 (s, Cp ), 2.43 (m, NCH
s, NMe), 1.68 (d JH -Hd =8.43, ZrCH
u
b
a
Me
2
þ
(
(
8
7
2
), 2.04 (m, NCH
2
), 1.90
3
2
d
2
-
2
), 1.37 (d, JH -Hd
u
=
=
(C
in approximately 0.7 mL of chlorobenzene-d
solution was added to a solution of 1.2 equiv of NMe(C18
(51.2 mg) in benzene-d , and immediately the color changed to
F
6
)
5
4
were loaded into a J. Young NMR tube and dissolved
u
u
3
.43 Hz, ZrCH
2
), 1.30-1.39 (b, chain), 0.92 (brt, JH -H
5
. The obtained red
ω
ω-1
1
3
1
.05, ωMe), 0.07 (br s, BMe). C{ H} NMR (100.55 MHz,
37 2
H )
u
d
benzene-d
6
, 297 K): δ 110.3 (Cp ), 110.2 (Cp ), 64.8 (NCH
2
),
), 49.0 (NMe), 14.1 (ωMe), -8.4 (BMe). F NMR
376.65 MHz, benzene-d , 297 K, J in Hz): δ -132.9 (br d,
6
19
yellow and gas evolution was observed. After 5 h, the J. Young
tube was interfaced to the high-vacuum line and the mixture of
solvents was removed. The resultant residue was rinsed with
5
5.4 (ZrCH
2
(
6
3
3
J
oF-mF = 22.3, o-F), -160.9 (t, JpF-mF = 20.7, p-F), -165.3
m, m-F).
Cp Zr(η -CH N(C H ) )
pentane (3 ꢁ 1 mL) and dried for 2 h. The reaction product was
(
NC18
2
þ
-
NC18
oily. A 42 mM solution of Zr
marginal signs of decomposition in 48 h.
BT in benzene-d showed only
6
[
MeB(C F ) ] (Zr
BN).
In the glovebox, Zr BN (20 mg, 26.1 mmol) and NMe-
(14 mg, 26.1 mmol) were loaded into a J. Young
2
2
18 37 2 3 3 3
6 5 3
Me
1
H NMR (400.13 MHz, benzene-d
6
, 297 K): δ 5.58 (s, Cp),
), 2.61 (s, ZrCH ), 1.55-1.01
b, chains), 0.93 (br t, ωMe). C{ H} NMR (100.55 MHz, 297
18 37 2
(C H )
a
b
2
.65 (m, NCH
2
), 2.50 (m, NCH
2
2
NMR tube and dissolved in approximately 0.7 mL of benzene-
d6. After 2 h, the reaction product was obtained. This com-
pound is soluble in pentane; consequently it was not further
1
3
1
(
1
9
K): δ 112.3 (Cp), 61.6 (ZrCH
2
), 60.2 (NCH
, 297 K, J in Hz): δ -131.9 (br d,
o-F), -162.3 (t, JpF-mF=20.8, p-F), -167.4 (m, m-F).
2
), 14.1 (ωMe).
F
NMR (376.65 MHz, benzene-d
6
purified.
3
1
H NMR (400.13 MHz, benzene-d , 297 K, J in Hz) δ 5.43
6
2
NC2
[
Cp Zr(η -CH NMeEt)(THF)][MeB(C F ) ] ([ZrNC2THF]-
(
s, Cp), 2.43 (m, NCH
0.93 (brt, JH -H = 7.0, ωMe), 0.25 (br s, BMe). C{ H}
NMR (100.55 MHz, benzene-d
NCH ), 56.6 (ZrCH ), 14.5 (ωMe), -5.4 (BMe). F NMR
376.65 MHz, benzene-d , 297 K, J in Hz): δ -132.7 (br d,
2
), 1.82 (s, ZrCH
2
), 1.07-1.43 (b, chains),
2
2
6 5 3
3
13
1
[
approximately 0.7 mL of benzene-d was connected to the high-
BN]). A J. Young NMR tube with a solution of Zr BN in
ω
ω-1
, 297 K): δ 111.0 (Cp), 60.8
6
6
1
9
vacuum line, and an excess of THF was condensed under
vacuum at -78 ꢀC into the tube. Immediately, a brightening
of the solution was observed. The solvent mixture THF/
(
(
2
2
6
3
3
J
oF-mF = 21.2, o-F), -161.3 (t, JpF-mF = 20.8, p-F), -165.6
m, m-F).
Cp Zr(η -CH NMe(C H ))
6
benzene-d was removed in vacuo. The resultant residue was
(
2
þ
-
NPh
rinsed with pentane (3 ꢁ 1 mL) and dried. The reaction product
[
MeB(C F ) ] (Zr BN).
6 5 3
2
2
6
5
3 3 3
NC2
was an oil. A 0.23 mM solution of [Zr THF][BN] in benzene-
Me
In the glovebox, 25 mg of Zr BN (0.033 mmol) was loaded
into a J. Young NMR tube and dissolved in approximately 0.7
mL of benzene. With a micrometric syringe, 1.2 equiv of N,N-
dimethylaniline (∼5 μL) was added and the mixture was vigor-
ously shaken. The solution immediately became dark yellow,
and gas evolution was observed. After a few minutes, a denser
oily phase settled to the bottom of the tube. The supernatant
solution was removed, the oil was rinsed with benzene (3 ꢁ 1
mL), and fresh benzene was added. After 5 h the tube was
interfaced to the high-vacuum line and the solvent was removed.
The resultant residue was rinsed with pentane (3 ꢁ 1 mL) and
d
6
is stable at least for 1 week.
H NMR (400.13 MHz, benzene-d , 297 K, J in Hz): δ 5.21
1
6
u
d
b
(
2
s, Cp ), 5.16 (s, Cp ), 3.04 (m, RTHF), 2.16 (m, NCH ), 1.87
a
2
d
(
m, NCH
u d
2 2
), 1.78 (s, NMe), 1.68 (d, JH -H =8.6, ZrCH ), 1.38
2
u
(
0
m, βTHF), 1.32 (br s, BMe), 1.31 (d, JH -H =8.6, ZrCH
2
),
.74 (brt, JH -NCH2 = 6.96, ωMe). C{ H} NMR (100.55
ω
u
d
3
13
1
u
d
MHz, benzene-d
NCH ), 77.6 (RTHF), 59.4 (ZrCH
.4 (ωMe). F NMR (376.65 MHz, benzene-d
6
, 297 K): δ 110.6 (Cp ), 110.5 (Cp ), 78.2
), 49.6 (NMe), 25.8 (βTHF),
, 297 K, J in Hz):
(
2
2
19
9
6
3
3
δ -132.3 (br d, JoF-mF = 20.0, o-F), -164.5 (t, JpF-mF = 20.7,
p-F), -167.1 (m, m-F).
dried again. The reaction product was an oil.
2
NC16
1
[Cp
2
Zr(η -CH
NMe(C16
2
H
33))(THF)][MeB(C
6
F
5
)
] ([Zr
3
-
BN
H NMR (400.13 MHz, benzene-d , 297 K, J in Hz): δ 7.01
6
NC16
3
m, m-H), 6.82 (br d, J
s, Cp ), 5.10 (s, Cp ), 2.70 (d, JH -Hd = 8.8, ZrCH
s, NCH ), 1.75 (d, JH -Hd =8.8, ZrCH
C{ H} NMR (100.55 MHz, benzene-d
29.7 (m-C), 127.4 (p-C), 121.5 (o-C), 111.5 (Cp ), 111.1 (Cp ),
THF][BN]). A J. Young NMR tube with a solution of Zr
(
(
(
=7.9, o-H), 6.78 (m, p-H), 5.37
oH-mH
u
d
2
d
2
in approximately 0.7 mL of benzene was connected to the high-
vacuum line, and an excess of THF was condensed in. A bright-
ening of the solution was observed. The mixture of solvents was
removed in vacuo. The resultant residue was rinsed with pentane
), 2.25
), 0.41 (br d, BMe).
6
, 297 K): δ156.5 (ipso-C),
u
2
u
3
2
u
1
3
1
d
u
1
5
(
3 ꢁ 1 mL) and finally dried. The product was an oil. An 8.45 mM
19
7.8 (NMe), 55.6 (ZrCH
, 297 K, Jin Hz): δ-132.7 (d JoF-mF=21.8, o-F), -161.0
t JpF-mF=20.8, p-F), -165.8 (m, m-F).
2
), -5.5 (BMe). F NMR (376.65 MHz,
NC16
solution of [Zr
month.
THF][BN] in benzene-d is stable at least for
6
3
benzene-d
6
1
3
(
1
H NMR (400.13 MHz, benzene-d , 297 K, J in Hz): δ 5.28
6
2
þ
-
NPh
[
Cp
the glovebox, 20 mg of Zr
metric amount of HNMe C H B(C F )
6 5 4
2
Zr(η -CH
2
NMe(C
6
H
5
))
B(C
6
F
5
)
4
] (Zr BT). In
u
d
b
3
3 3
(s, Cp ), 5.26 (s, Cp ), 3.10 (m, RTHF), 2.30 (m, NCH ), 2.03
Me
2
2
(0.0785 mmol) and a stoichio-
a
2
2
), 1.92 (s, NMe), 1.87 (d, JH -Hd =8.92, ZrCH
d
),
(
1
m, NCH
2
þ
-
u
(63 mg) were
2
u
2
6
5
.49 (d, JH -Hd =8.92, ZrCH
2
), 1.37-1.43 (b, chain), 1.37 (b,
u
loaded into a J. Young NMR tube and dissolved in approxi-
mately 0.7 mL of benzene-d . The mixture was vigorously
3
βTHF), 1.32 (br s, BMe), 0.94 (br t, JH -H = 6.97, ωMe).
ω
ω-1
13
1
C{ H} NMR (100.55 MHz, benzene-d
u
6
6
, 297 K): δ 110.3 (Cp ),
), 55.1 (ZrCH ), 49.6
shaken, and in a few minutes a red dense oily phase settled to
the bottom of the tube. Gas evolution was observed. The
supernatant solution was removed, and the oil was rinsed with
benzene (3 ꢁ 1 mL). After 5 h, the tube was interfaced to the
d
1
10.0 (Cp ), 77.6 (RTHF), 65.1 (NCH
NMe), 25.8 (βTHF), 14.5 (ωMe), 11.5 (BMe). F NMR
376.65 MHz, benzene-d , 297 K, J in Hz): δ -132.2 (br d,
2
2
19
(
(
6
3
3
J
oF-mF = 23.3, o-F), -164.4 (t, JmF-pF = 20.8, p-F), -167.0
m, m-F).
Cp Zr(η -CH
THF][BN]). In the glovebox, into a J. Young NMR tube, a
high-vacuum line and the solvent was removed. A 3.5 mM
(
NPh
2
NC18
solution of Zr
position after 7 days.
6
BT in benzene-d showed no sign of decom-
[
2
2
N(C18
H
37
)
2
)(THF)][MeB(C
6
F
5
)
3
]
([Zr
-
1
NC18
H NMR (400.13 MHz, benzene-d
6
/o-difluorobenzene 10:1,
solution of Zr
was prepared using a slight excess of NMe(C18
BN in approximately 0.7 mL of benzene-d
(5%). The
6
3
2
6
(
97 K, J in Hz): δ 7.18 (m, m-H), 7.08 (t, JmH-pH=7.3, p-H),
37 2
H )
3
u
d
.88 (d, JoH-mH =7.7, o-H), 5.75 (s, Cp ), 5.46 (s, Cp ), 3.22
tube was interfaced to the high-vacuum line, and an excess of
THF was condensed in at -78 ꢀC. Then the solvent was removed
in vacuo. The resultant oily residue was rinsed with pentane (3 ꢁ
2
d
2
d, J
=8.8 Hz, ZrCH ), 2.62 (s, NMe), 2.48 (d, J d=
u
H -H
H
u
-H
.8, ZrCH
d
2
u
13
). C{ H} NMR (100.55 MHz, benzene-d
1
8
2
6
/o-
difluorobenzene 10:1, 297 K): δ 156.4 (ipso-C), 130.6 (m-C),
1 mL) and dried.
d
28.1 (p-C), 121.6 (o-C), 113.5 (Cp ), 113.05 (Cp ), 60.0
u
1
1
(
H NMR (400.13 MHz, benzene-d , 297 K, J in Hz): δ 5.39
6
1
ZrCH ), 57.79 (NMe). F NMR (376.65 MHz, benzene-d6/
9
2
(s, Cp), 3.2 (m, RTHF), 2.39 (m, NCH ), 1.88 (s, ZrCH ), 1.36
2
2
o-difluorobenzene 10:1, 297 K, J in Hz): δ -131.98 (br d,
o-F), -162.66 (t, JpF-mF=20.8, p-F), -168.23 (m, m-F).
(s, βTHF), 1.33 (br s, BMe), 1.10-1.50 (b, chains), 0.95 (br t,
3
3 13 1
ω ω-1
6
JH -H =7.0, ωMe). C{ H} NMR (100.55 MHz, benzene-d ,

Article
Organometallics, Vol. 30, No. 1, 2011 113
2
97 K): δ 111.3 (Cp), 78.0 (RTHF), 62.2 (ZrCH
2
), 53.8 (NCH
2
),
self-diffusion coefficient, Δ=delay between the midpoints
of the gradients, δ=length of the gradient pulse, and γ=
magnetogyric ratio. The shape of the gradients was re-
ctangular, their lengths δ were 4-5 ms, and their strengths
G were varied during the experiments. All spectra were
19
25.4 (βTHF), 14.7 (ωMe), 12.4 (BMe). F NMR (376.65 MHz,
3
benzene-d , 297 K, J in Hz): δ -132.3 (br d, J
o-F), -164.4 (t, JmF-pF=20.8, p-F), -167.0 (m, m-F).
= 20.3,
6
oF-mF
3
2
NPh
[
BN]). A J. Young NMR tube containing a solution of
Cp
2 2 6 5 6 5 3
Zr(η -CH NMe(C H ))(THF)][MeB(C F ) ] ([Zr THF]-
[
Zr
1
NPh
acquired using 32K points and a spectral width of 5000 ( H)
19
and 18 000 ( F) Hz, and they were processed with a line
BN in benzene-d was interfaced to the high-vacuum
6
line, and an excess of THF was condensed in at -78 ꢀC. The
solution immediately changed from dark to bright yellow.
The mixture of solvents was quickly removed in vacuo, and
the residue was rinsed with pentane (3 ꢁ 1 mL) and dried. The
1 19
broadening of 1.0 ( H) and 1.5 ( F). The semilogari-
2
thmic plots of ln(I/I ) versus G were fitted using a stan-
0
dard linear regression algorithm, and a correlation factor
better than 0.99 was always obtained. Different values of
Δ, G, and numbers of transients were used for different
samples.
reaction product was an oil.
1
H NMR (400.13 MHz, benzene-d /chlorobenzene-d
6
5
1:1,
=7.7, p-H),
3
2
6
97 K, J in Hz): δ 7.11 (m, m-H), 6.99 (t, J
pH-mH
2 u
d
.74 (d, JoH-mH =7.7, o-H), 5.39 (s, Cp ), 5.10 (s, Cp ), 3.28
The self-diffusion coefficient D , which is directly propor-
tional to the slope (m) of the regression line obtained by plotting
ln(I/I ) versus G , was estimated by evaluating the proportion-
t
2
d
(
m, RTHF), 2.56 (d, JH -Hd=7.9, ZrCH
2
), 2.27 (s, NMe), 2.15
), 1.46 (m, βTHF), 1.27 (br d, BMe).
C{ H} NMR (100.55 MHz, benzene-d /chlorobenzene-d 1:1,
u
2
u
2
2
(
d, JH -H =7.9, ZrCH
u
d
0
1
3
1
6
5
ality constant using a sample of HDO (5%) in D O (known
2
4
9
2
97 K): δ 156.7 (ipso-C), 129.9 (m-C), 128.1 (p-C), 123.4 (o-C),
diffusion coefficients in the range 274-318 K) under the same
exact conditions as for the sample of interest. The solvent was
d u
1
10.6 (Cp ), 109.6 (Cp ), 77.5 (RTHF), 59.0 (NMe), 57.6
1
9
(
ZrCH ), 25.5 (βTHF), 11.2 (ωMe). F NMR (376.65 MHz,
2
taken as an internal standard. D data were treated as described
t
benzene-d /chlorobenzene-d 1:1, 297 K, J in Hz): δ -132.3
6
5
in the literature in order to derive the hydrodynamic dimen-
sions.
3
3
3
6d,39,50
(
(
d, JoF-mF=21.8, o-F), -164.3 (t, JpF-mF=20.8, p-F), -166.9
br d, m-F).
The measurement uncertainty was estimated by
determining the standard deviation of m when performing
experiments with different Δ values. Error propagation analysis
yielded a standard deviation of approximately 3-4% in the
hydrodynamic radius and 10-15% in the aggregation number
N. NMR samples for PGSE NMR measurements were prepared
inside the glovebox by successive dilutions. The actual concen-
tration was estimated from integration relative to an external
standard.
2
NPh
[
Cp Zr(η -CH NMe(C H ))(THF)][B(C F ) ] ([Zr THF]-
2 2 6 5 6 5 4
NPh
[
BT]). A J. Young NMR tube containing a solution of Zr BT
in benzene-d /chlorobenzene-d was interfaced to the high-
6
5
vacuum line, and an excess of THF was condensed under
vacuum at -78 ꢀC. The mixture of solvents was removed in
vacuo, and the resultant residue was rinsed with pentane (3 ꢁ 1mL)
and dried again.
1
H NMR (400.13 MHz, benzene-d /chlorobenzene-d 1:1,
6
5
Computational Details. Density fuctional calculation were
3
5
1
2
97 K, J in Hz): δ 7.10 (m, m-H), 7.02 (t J
=7.3, p-H),
pH-mH
performed with the Turbomole program (version 5.8) in
combination with the OPTIMIZE routine of Baker and co-
workers. All geometries were fully optimized at the B3LYP
2
u
d
6
.77 (d, JoH-mH=7.7, 2H, o-H), 5.43 (s, Cp ), 5.13 (s, Cp ), 3.33
2
d
5
2
(
(
(
(
2
m, R THF), 2.61 (d, JH -Hd=8.3, ZrCH ), 2.31 (s, NMe), 2.19
u
2
d, J
u
13
1
5
3
54
= 8.3, ZrCH ), 1.49 (m, βTHF). C{ H} NMR
H
u
-H
d
2
level, using the single-ζ SV(P) basis set (small-core pseu-
55
6 5
400 MHz, benzene-d /chlorobenzene-d 1:1, 297 K): δ 156.7
d
ipso-C), 129.7 (m-C), 127.5 (p-C), 123.0 (o-C), 110.3 (Cp ),
dopotential on Zr ). All the structures were characterized by
vibrational analyses using analytic or numerical second deri-
vatives, in order to check that no imaginary frequencies were
present. Improved electronic energies were calculated at the
SV(P) optimized geometries using the TZVP basis set on all
u
1
09.3 (Cp ), 77.2 (RTHF), 58.7 (NMe), 57.4 (ZrCH
2
), 25.3
19
(
βTHF). F NMR (376.65 MHz, benzene-d /chlorobenzene-
6
3
d
5
1:1, 297 K, J in Hz): δ -132.5 (br d, o-F), -162.7 (t, JoF-pF
=
5
6
2
0.8, p-F), -166.7 (m, m-F).
atoms. Single-point solvent corrections were calculated using
2
NC18
5
7
[
Cp
BT]). A J. Young NMR tube containing a solution of Zr
in benzene-d /chlorobenzene-d was interfaced to the high-vacuum
2
Zr(η -CH
2 37
N(C18H )
2 6 5
)(THF)][B(C F )
4
] ([Zr
THF]-
the conductor-like screening model (COSMO) with ε
to model an apolar solvent. If not otherwise noted, all the
r
=2.3
NC18
[
BT
6
5
line, and an excess of dry THF was condensed in at -78 ꢀC. A
brightening of the solution was observed. Then, the mixture of
solventswas removedinvacuoandthe resultantresiduewas dried
(
49) (a) Tyrrell, H. J. W.; Harris, K. R. Diffusion in Liquids;
Butterworth: London, 1984. (b) Mills, R. J. Phys. Chem. 1973, 77,
85.
(50) Cabrita, E. J.; Berger, S. Magn. Reson. Chem. 2001, 39, S142.
(51) (a) Ahlrichs, R.; B €a r, M.; Baron, H.-P.; Bauernschmitt, R.;
6
for 2 h. The reaction product was oily.
1
6
H NMR (400 MHz, benzene-d , 297 K): δ 5.38 (s, Cp),
B o€ cker, S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.;
H €a ser, M.; H €a ttig, C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek,
M.; K o€ hn, A.; K o€ lmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.;
Ohm, H.; Sch €a fer, A.; Schneider, U.; Treutler, O.; Tsereteli, K.;
Unterreiner, B.; von Arnim, M.; Weigend, F.; Weis, P.; Weiss, H.
Turbomole, Version 5; Theoretical Chemistry Group, University of
Karlsruhe, Karlruhe, Germany, 2002. (b) Treutler, O.; Ahlrichs, R.
J. Chem. Phys. 1995, 102, 346.
3
.19 (m, RTHF), 2.39 (m, NCH ), 1.88 (s, ZrCH ), 1.6-1.0
2
2
13 1
(
b, chains/βTHF), 0.94 (br t, ωMe). C{ H} NMR (100.55
),
0.2 (NCH
MHz, benzene-d6, 297 K): δ 110.1 (Cp), 77.3 (R), 61.6 (ZrCH
6
2
1
9
), 14.1 (ωMe). F NMR (376.65 MHz, benzene-d ,
2 6
3
2
97 K, J in Hz): δ -132.2 (br d, o-F), -162.5 (t, JmF-pF=20.8
Hz, p-F), -166.5 (m, m-F).
1
19
PGSE Measurements. H and F PGSE NMR measurements
were performed by using the standard stimulated echo pulse
sequence on a Bruker AVANCE DRX 400 spectrometer
equipped with a GREAT 1/10 gradient unit and a QNP probe
with a Z-gradient coil, at 297 K without spinning. The depen-
dence of the resonance intensity (I) on a constant waiting time
and on a varied gradient strength (G) is described by the
equation
(
52) (a) PQS, version 2.4; Parallel Quantum Solutions:, Fayetteville,
AK, 2001 (the Baker optimizer is available separately from PQS upon
request). (b) Baker, J. J. Comput. Chem. 1986, 7, 385.
(53) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(b) Becke, A. D. J. Chem. Phys. 1993, 98, 1372. (c) Becke, A. D. J. Chem.
Phys. 1993, 98, 5648. (d) Calculations were performed using the Turbomole
functional “b3-lyp”, which is not identical with the Gaussian “B3LYP”
functional.
(
54) Sch €a fer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
571.
(55) Andrae, D.; H €a ussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
56) Sch €a fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
829.
(57) Sch €a fer, A.; Klamt, A.; Sattel, D.; Lohrenz, J. C. W.; Eckert, F.
Phys. Chem. Chem. Phys. 2000, 2, 2187.
ꢀ
ꢁ
2
I
δ
2
2
ln
¼ - ðγδÞ Dt Δ -
G
I0
3
(
5
where I = intensity of the observed spin echo, I = inten-
0
sity of the spin echo in the absence of a gradient, D =
t

1
14 Organometallics, Vol. 30, No. 1, 2011
Rocchigiani et al.
energies are calculated with solvation correction. Initial geo-
metries and reasonable starting Hessians were obtained from
PM3 computations with the Spartan package from Wavefunc-
tion Inc.
5
8
(
58) Shao, Y.; Molnar, L. F.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.;
Acknowledgment. This work was supported by grants
from the Ministero dell’Istruzione, dell’Universit ꢀa e della
Ricerca through the PRIN 2007 (2007X2RLL2) pro-
gram. We thank Prof. Giuseppe Cardaci for helpful
discussions.
Brown, S. T.; Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.; O’Neill,
D. P.; DiStasio, R. A., Jr.; Lochan, R. C.; Wang, T.; Beran, G. J. O.; Besley,
N. A.; Herbert, J. M.; Lin, C. Y.; Van Voorhis, T.; Chien, S. H.; Sodt, A.;
Steele, R. P.; Rassolov, V. A.; Maslen, P. E.; Korambath, P. P.; Adamson,
R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.; Dachsel, H.; Doerksen, R. J.;
Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney, S. R.;
Heyden, A.; Hirata, S.; Hsu, C.-P.; Kedziora, G.; Khalliulin, R. Z.;
Klunzinger, P.; Lee, A. M.; Lee, M. S.; Liang, W. Z.; Lotan, I.; Nair, N.;
Peters, B.; Proynov, E. I.; Pieniazek, P. A.; Rhee, Y. M.; Ritchie, J.; Rosta,
E.; Sherrill, C. D.; Simmonett, A. C.; Subotnik, J. E.; Woodcock, H. L., III;
Zhang, W.; Bell, A. T.; Chakraborty, A. K.; Chipman, D. M.; Keil, F. J.;
Warshel, A.; Hehre, W. J.; Schaefer, H. F.; Kong, J.; Krylov, A. I.; Gill,
P. M. W.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2006, 8, 3172.
Supporting Information Available: Figures and tables
giving selected PGSE NMR data in benzene-d6 and
cyclohexane-d12 and additional structures of the complexes.
This material is available free of charge via the Internet at
http://pubs.acs.org.