Role of cation-anion interactions in ionic complexes containing [Pd{C 6H3(CH2NMe)2-2,6}(OH 2)]+ and [{Pd(C6H3(CH 2NMe2)2-2,6)}2</

Article abstract of DOI:10.1021/om0303698

The reaction of [PdCl{C₆H₃(CH₂NMe ₂)₂-2,6}] ([PdCl(NCN)]) with either the sodium or silver salts of the borate anions [B{C₆H₄SiMe₃)-4} ₄]^- and [B

Full text of DOI:10.1021/om0303698

Organometallics 2004, 23, 2287-2294
2287
Role of Ca tion -An ion In ter a ction s in Ion ic Com p lexes
Con ta in in g [P d {C6H3(CH2NMe2)2-2,6}(OH2)]+ a n d
[{P d (C6H3(CH2NMe2)2-2,6)}2(µ-Cl)]+ Ca tion s
J oep van den Broeke,† J org J . H. Heeringa,† Alexey V. Chuchuryukin,†
Huub Kooijman,‡ Allison M. Mills,‡ Anthony L. Spek,‡,§ J oop H. van Lenthe,|
Paul J . A. Ruttink,#,| Berth-J an Deelman,*, and Gerard van Koten†
Department of Metal-Mediated Synthesis, Debye Institute, Utrecht University,
Padualaan 8, 3584 CH Utrecht, The Netherlands, Department of Crystal and Structural
Chemistry, Bijvoet Center for Biomolecular Research, Utrecht University, Padualaan 8,
3584 CH Utrecht, The Netherlands, Theoretical Chemistry Group, Debye Institute, Utrecht
University, Padualaan 14, 3584 CH Utrecht, The Netherlands, and ATOFINA Vlissingen B.V.,
P.O. Box 70, 4380 AB Vlissingen, The Netherlands
Received May 19, 2003
The reaction of [PdCl{C6H3(CH2NMe2)2-2,6}] ([PdCl(NCN)]) with either the sodium or silver
salts of the borate anions [B{C6H4(SiMe3)-4}4]- and [B(C6H4{SiMe2(CH2CH2C6F13)}-4)4]-
under various reaction conditions resulted in the exclusive formation of the respective
dinuclear [{Pd(NCN)}2(µ-Cl)][B{C6H4(SiMe2R)-4}4] salts 4a (R ) Me) and 4b (R ) CH2-
CH2C6F13). When using BF4-, both dinuclear [{Pd(NCN)}2(µ-Cl)]BF4 (4c) and mononuclear
[Pd(NCN)(OH2)]BF4 (5c) could be obtained. The molecular structure of 5c showed the
presence of hydrogen-bridging interactions between coordinated H2O molecules and the
fluorine atoms of BF4-. Attempts to isolate complexes of the type [Pd(NCN)(OH2)][B{C6H4-
(SiMe2R)-4}4] were unsuccessful. This has been ascribed to a difference in stabilization of
the respective mono- and dinuclear cations by the borate anions BF4- and [B{C6H4(SiMe2R)-
4}4]-, respectively. The conclusions are supported by DFT calculations.
In tr od u ction
paid to the use of the counterions as a site for the
introduction of a suitable functionality for catalyst
recovery.
Cationic palladium complexes of the monoanionic,
terdentate 2,6-bis(dimethylaminomethyl)phenyl pincer
ligand [C6H3(CH2NMe2)2-2,6]- (abbreviated as NCN) are
active catalysts in various C-C bond formation reac-
tions, such as the aldol-condensation1 and the Michael
reaction.2 To recover the catalysts after reaction, various
methods have been employed to immobilize the NCN-
Pd unit on a support. The majority of these efforts have
focused on attaching the ligand covalently to a soluble
carrier, e.g., carbosilane dendrimers,2c,3 fullerenes,2c,4
and cartwheel molecules.2b,c So far no attention has been
The immobilization of transition-metal catalysts in
perfluorinated solvents using highly fluorous anions is
interesting, as this facilitates catalyst recovery using
phase separation while allowing catalysis under homo-
geneous conditions and because it does not affect the
catalytically active center. Recently we reported the
introduction of perfluoroalkyl-substituted tetraphenyl-
borate anions for the fluorous phase immobilization of
cationic rhodium complexes.5 Herein, the reaction of
various tetraphenylborate salts with [PdCl(NCN)] is
described. This, surprisingly, resulted in the formation
of chloride-bridged dinuclear complexes, whereas a
chloride-free, mononuclear [Pd(NCN)(OH2)]+ cation was
* Corresponding author. Fax:
Berth-J an.Deelman@atofina.com.
+31 30-2523615. E-mail:
† Department of Metal-Mediated Synthesis, Debye Institute.
‡ Bijvoet Center for Biomolecular Research.
§ Address correspondence pertaining to crystallographic studies to
this author.
isolated when commonly employed BF4 was used.6
-
#
Resu lts a n d Discu ssion
Address correspondence pertaining to density functional calcula-
tions to this author.
| Theoretical Chemistry Group, Debye Institute.
Syn th esis. Reaction of [PdCl(NCN)] (1) with 1 equiv
of Na[B{C6H4(SiMe3)-4}4] (2a ), Na[B(C6H4{SiMe2(CH2-
CH2C6F13)}-4)4] (2b), Ag[B{C6H4(SiMe3)-4}4] (3a ), or Ag-
[B(C6H4{SiMe2(CH2CH2C6F13)}-4)4] (3b) in an acetone/
Atofina Vlissingen B.V.
(1) (a) Nesper, R.; Pregosin, P. S.; Pu¨ntener, K.; Wo¨rle, M. Helv.
Chim. Acta 1993, 76, 2239-2249. (b) Gorla, F.; Togni, A.; Venanzi, L.
M.; Albinati, A.; Lianza, F. Organometallics 1994, 13, 1607-1616.
(2) (a) Stark, M. A.; Richards, C. J . Tetrahedron Lett. 1997, 38, 5881.
(b) Dijkstra, H. P.; Meijer, M. D.; Patel, J .; Kreiter, R.; van Klink, G.
P. M.; Lutz, M.; Spek, A. L.; Canty, A. J .; van Koten, G. Organo-
metallics 2001, 20, 3159. (c) Albrecht, M.; van Koten, G. Angew. Chem.,
Int. Ed. 2001, 40, 3750.
(5) (a) van den Broeke, J .; Kooijman, H.; Lutz, M.; Spek, A. L.;
Deelman, B.-J .; van Koten, G. Organometallics 2001, 20, 2114. (b) van
den Broeke, J .; Deelman, B.-J .; van Koten, G. Tetrahedron Lett. 2001,
42, 8085. (c) van den Broeke, J .; de Wolf, E.; Deelman, B.-J .; van Koten,
G. Adv. Synth. Catal. 2003, 345, 625.
(3) Kreiter, R.; Kleij, A. W.; Klein Gebbink, R. J . M.; van Koten, G.
Top. Curr. Chem. 2001, 217, 164.
(4) Meijer, M. D.; Ronde, N.; Vogt, D.; van Klink, G. P. M.; van
Koten, G. Organometallics 2001, 20, 3993.
(6) Grove, D. M.; van Koten, G.; Louwen, J . N.; Noltes, J .; Spek, A.
L.; Ubbels, H. J . C. J . Am. Chem. Soc. 1982, 104, 6609.
10.1021/om0303698 CCC: $27.50 © 2004 American Chemical Society
Publication on Web 04/03/2004

2288 Organometallics, Vol. 23, No. 10, 2004
van den Broeke et al.
Sch em e 1. P r ep a r a tion of µ-Cl Br id ged Din u clea r NCN-P a lla d iu m Com p lexes
H2O mixture gave rise to the formation of the dinuclear
nature of these interactions was investigated more
thoroughly. In these studies only [B{C6H4(SiMe3)-4}4]-
was considered, as this anion was synthetically more
accessible than [B(C6H4{SiMe2(CH2CH2C6F13)}-4)4]- and
because the solid state structure of dinuclear 4a allowed
a direct comparison with those obtained for known
complexes 5c6 and 4c.7
Molecu lar Str u ctu r es of [{P d(NCN)}2(µ-Cl)][B{C6-
H4(SiMe3)-4}4] (4a ) a n d [P d (NCN)(OH 2)]BF 4 (5c).
The molecular structures of 4a and 5c were determined
by single-crystal X-ray diffraction studies. Off-white
crystals of 4a , suitable for X-ray crystal structure
determination, were obtained by slow diffusion of pen-
tane into a concentrated solution of 4a in methanol. The
molecular structure is represented in Figure 1, and
relevant bond angles and distances are collected in
Table 1.
In the chloride-bridged dinuclear cation, both pal-
ladium centers have a slightly distorted square-planar
geometry, with a maximum deviation from their least
squares NCN-Pd-Cl planes of only 0.016(1) Å for Pd-
(1) and 0.010(1) Å for Pd(2, major), respectively. The
boron has a slightly distorted tetrahedral coordination
sphere. These observations, together with the majority
of the values of bond angles and distances, correspond
with structural features of the earlier determined
structures for the dinuclear cation in [{Pd(NCN)}2(µ-
Cl)]BF4 (4c)7 (Pd-N ) 2.10 Å, Pd-C ) 1.92 Å, N-Pd-N
) 163°) and the borate anion in [(18-crown-6)Na(THF)2]-
[B{C6H4(SiMe3)-4}4]- (B-C ) 1.64-1.72 Å, C-B-C )
103-113°).5b However, significant deviations in the Pd-
Cl-Pd angle (121° in 4a vs 134° in 4c), the Pd‚‚‚Pd
distance (4.2902(11) Å in 4a vs 4.5438(15) Å in 4c), and
the angle between the planes of the NCN ligands (77°
in 4a vs 88° in 4c) were observed. It is most likely that
the steric bulk of [B{C6H4(SiMe3)-4}4]- in 4a enforces
a different crystal packing in comparison with 4c,11
which causes the reduction of the Pd-Cl-Pd angle and
the significant shorting of the Pd‚‚‚Pd distance in 4a .
Another noteworthy aspect in the structure of 4a is the
complexes [(Pd(NCN))2(µ-Cl)][B{C6H4(SiMe2R)-4}4] (4a ,
1
R ) Me; 4b, R ) CH2CH2C6F13) (Scheme 1). H NMR
and elemental analysis confirmed that these complexes
have an NCN-ligand to borate anion ratio of 2:1.
Changing the solvent to CH2Cl2 or the molar ratio of 1
and the borate salts during the synthesis did not affect
the nature of the product formed. Starting with an
excess of 1 (2:1 molar ratio) or an excess of sodium
borate (up to 1:4 molar ratio) yielded dinuclear 4 as the
only product isolated, the complexes [Pd(NCN)(OH2)]-
[B{C6H4(SiMe2R)-4}4] (5a , R ) Me; 5b, R ) CH2-
CH2C6F13) proving to be inaccessible. Also addition of
excess of the silver borate 3b did not afford 5b. When
AgBF4 (2 equiv) was added to acetone solutions of
dinuclear 4b, rapid formation of 5c was observed,
underlining the preference for this species. Also a 1
equiv reaction afforded only 5c along with some un-
reacted 4b but no indication for the tetraarylborate aqua
complex 5b.
The preference for the formation of µ-Cl-bridged
complexes was not expected on the basis of previous
-
reactions in which BF4 was used as borate anion.
Under similar conditions, reaction of 1 with AgBF4
resulted in the formation of a mononuclear complex with
the composition [Pd(NCN)(OH2)]BF4 (5c).6 The di-
nuclear complex [(Pd(NCN))2(µ-Cl)]BF4 (4c) could be
isolated only when thoroughly dried acetone was used.7
The reaction of 1 with 2a was also performed in the
presence of acetonitrile (Pd:MeCN ) 1:3), as the forma-
tion of dicationic [Pd2{C6(CH2NMe2)4-2,3,4,5}(MeCN)2]-
[BPh4]2 has been reported under similar conditions.8
However, the decomposition of the borate anion was
observed and no ionic palladium species containing a
tetraarylborate anion could be identified.9 The results
obtained suggest that in the case of the large tetra-
arylborate anions formation of the chloride-bridged
dinuclear palladium complex is preferred over that of
the chloride-free mononuclear aqua complex. This signi-
fies that a difference exists in the cation-anion interac-
-
tions in the presence of the BF4 anion in comparison
with those for the [B{C6H4(SiMe2R)-4}4]- anions (R )
Me or CH2CH2C6F13). As dinuclear palladium species
such as 4, with a single halogen bridge without sup-
porting bridging ligands, are relatively rare,10 the
(10) Only 8 unique Pd compounds were reported in the J une 2003
update of the Cambridge Structural Database. These compounds are
reported in 9 publications: (a) Reference 7b. (b) Clark, R. J . H.; Croud,
V. B.; Kurmoo, M. J . Chem. Soc., Dalton Trans. 1985, 815. (c) Clark,
R. J . H.; Croud, V. B.; Wills, R. J .; Bates, P. A.; Dawes, H. M.;
Hursthouse, M. B. Acta Cystallogr. Sect. B 1989, 45, 147. (d) Fornies,
J .; Navarro, R.; Sicilia, V.; Martinez, F.; Welch, A. J . J . Organomet.
Chem. 1991, 408, 425. (e) Kitano, Y.; Kajimoto, T.; Kashiwagi, M.;
Kinoshita, Y. J . Organomet. Chem. 1971, 33, 123. (f) Martin R.; Keller,
H. J .; Muller, B. Z. Naturforsch. B 1985, 40, 57. (g) Mazet, C.; Gade,
L. H. Organometallics 2001, 20, 4144. (h) Raper, E. S.; Kubiak, M.;
Glowiak, T. Acta Crystallogr. Sect. C 1997, 53, 1201. (i) Vicente, J .;
Abad, J . A.; Frankland, A. D.; Lopez-Serrano, J .; Ramirez de Arellano,
M. C.; J ones, P. G. Organometallics 2002, 21, 272.
(7) (a) Grove, D. M.; van Koten, G.; Ubbels, H. J . C. J . Am. Chem.
Soc. 1982, 104, 4285. (b) Terheijden, J .; van Koten, G.; Grove, D. M.;
Vrieze, K. J . Chem. Soc., Dalton Trans. 1987, 1359.
(8) Steenwinkel, P.; J ames, S. L.; Grove, D. M.; Kooijman, H.; Spek,
A. L.; van Koten, G. Organometallics 1997, 16, 513.
(9) The reaction mixture was analyzed using 1H and 11B{1H} NMR.
The former shows the disappearance of all resonances corresponding
with the aromatic protons of the anion, the latter that of the
characteristic resonance belonging to tetraphenylboron (11B{1H} NMR
(acetone-d6, 96.3 MHz): δ -1.1 (s)).
(11) The bulky tetraphenylborate anion is known to be a lattice-
dominating moiety: Bock, H.; Van, T. T. H.; Scho¨del, H. J . Prak. Chem.
1997, 339, 525.

Cation-Anion Interactions in Ionic Complexes
Organometallics, Vol. 23, No. 10, 2004 2289
F igu r e 2. ORTEP drawing of mononuclear 5c with
ellipsoids at the 50% probability level. Hydrogen atoms of
the NCN ligand are omitted for clarity.
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles
(d eg) of 5c
F igu r e 1. ORTEP drawing of dinuclear 4a with ellipsoids
at the 50% probability level. Hydrogen atoms and the
mirror disorder component are omitted for clarity.
bond lengths
bond angles
Ta ble 1. Selected Bon d Len gth s (Å) a n d An gles
(d eg) of 4a
Pd-O
2.192(2)
2.107(2)
2.108(2)
1.909(2)
1.368(3)
1.386(3)
1.401(4)
1.386(4)
1.92(4)
N(1)-Pd-N(2)
162.70(8)
81.64(9)
81.18(10)
176.84(9)
110.7(2)
108.9(3)
107.2(2)
108.5(2)
176(4)
Pd-N(1)
Pd-N(2)
Pd-C(1)
B-F(1)
N(1)-Pd-C(1)
N(2)-Pd-C(1)
C(1)-Pd-O
bond lengths
bond angles
F(1)-B-F(2)
F(2)-B-F(3)
F(2)-B-F(4)
F(3)-B-F(4)
O-H(1a)-F3
O-H(1b)-F4
Pd(1)-Cl
2.4527(17)
2.102(5)
2.091(5)
1.906(5)
2.4738(17)
2.062(8)
2.138(10)
1.912(8)
1.653(7)
1.644(7)
1.641(6)
1.660(6)
Pd(1)-Cl-Pd(2)
121.11(5)
162.97(19)
176.86(13)
B-F(2)
Pd(1)-N(1)
Pd(1)-N(2)
Pd(1)-C(9)
Pd(2)-Cl
N(1)-Pd(1)-N(2)
C(9)-Pd(1)-Cl
B-F(3)
B-F(4)
F3-H(1a)
F4-H(1b)
O-H(1a)
O-H(1b)
N(3)-Pd(2)-N(4)
C(21)-Pd(2)-Cl
164.5(3)
174.6(2)
2.04(4)
0.76(4)
0.84(4)
173(5)
Pd(2)-N(3)
Pd(2)-N(4)
Pd(2)-C(21)
B-C(25)
C(25)-B-C(34)
C(25)-B-C(43)
C(25)-B-C(52)
C(34)-B-C(52)
110.1(3)
110.4(4)
107.8(3)
111.7(4)
B-C(34)
B-C(43)
B-C(52)
puckering of the five-membered rings in the NCN-Pd-
Cl entities. Whereas these units have C2-symmetry in
4c, resulting in positioning of the axial N-methyl groups
at different sides of the coordination plane, in 4a these
units have C2v-symmetry and the axial N-methyl groups
are positioned on the same side of this plane. The latter
structural feature is possibly also the result of having
to accommodate the bulky anion in the crystal lattice.
No π‚‚‚π interactions or agostic interactions between
cations and anions were observed in the solid state, and
investigation of 4a in solution using 1H-1H NOESY
F igu r e 3. Crystal packing of 5c, showing the hydrogen-
bridging interactions of the BF4 anions and the water
ligands.
-
-
and 2.04(4) Å), with each water ligand and each BF4
12
NMR in CD2Cl2 also showed no indications for such
anion participating in one of the shorter and one of the
longer hydrogen bridges. Furthermore, the hydrogen
atoms that have the shortest O-H distance, 0.76(4) Å,
are connected to the fluorine atoms with the longest
B-F bond distances (1.401(4) Å) and vice versa (O-H
) 0.84(4) Å and B-F ) 1.386(4) Å).
Although hydrogen bonding of water to weakly coor-
dinating anions is frequently observed in organometallic
complexes,14 direct crystallographic evidence for hydro-
gen-bonding interactions in palladium aqua complexes
is rare. Previous examples have been encountered in
[Pd(AsPh3)(C6F5)(OH2)2][CF3SO3],15 [Pd{C6H3(CH2P-
specific cation-anion contacts.
White crystals of 5c, suitable for single-crystal X-ray
diffraction analysis, were obtained by slow diffusion of
hexane into a concentrated solution of [Pd(NCN)(OH2)]-
BF4 in dichloromethane. The molecular structure in the
crystal is represented in Figure 2, and relevant bond
angles and distances are collected in Table 2.
In 5c, the palladium and the boron centers have
slightly distorted square-planar and tetrahedral coor-
dination spheres, respectively. The NCN-Pd unit has
C2-symmetry, and the bond angles (e.g., N-Pd-N )
162.7°) and distances (e.g., Pd-N ) 2.11 Å and Pd-C
) 1.91 Å) correspond well with known structures for
comparable NCN-Pd complexes.13 In the crystal lattice
each BF4- anion is connected to two water ligands, and
(13) Ma, J .-F.; Kojima, Y.; Yamamoto, Y. J . Organomet. Chem. 2000,
616, 149. (b) Maassarani, F.; Davidson, M. F.; Wehman-Ooyevaar, I.
C. M.; Grove, D. M.; van Koten, M.; Smeets, W. J . J .; Spek, A. L.; van
Koten, G. Inorg. Chim. Acta 1995, 235, 327. (c) Castan, P.; J aud, J .;
Wimmer, S.; Wimmer, F. L. J . Chem. Soc., Dalton Trans. 1991, 1155.
(d) Barco, I. C.; Falvello, L. R.; Fernandez, S.; Navarro, R.; Urriola-
beitia, E. P. J . Chem. Soc., Dalton Trans. 1998, 3745. (e) Deeming, A.
J .; Rothwell, I. P.; Hursthouse, M. B.; New, L. J . Chem. Soc., Dalton
Trans. 1978, 1490. (f) Vicente, J .; Arcas, A.; Bautista, D.; J ones, P. G.
Organometallics 1997, 16, 2127.
(14) The tendency of H2O to act as a hydrogen-bond donor is the
result of its coordination to a metal, which makes H2O more acidic.
Previous studies on [NCN-Pt-(OH2)][OTf] reported a pKa value for the
coordinated water of 10.8, whereas the value is 15.7 for noncoordinated
water: Schmu¨lling, M.; Grove, D. M.; van Koten, G.; van Eldik, R.;
Veldman, N.; Spek, A. L. Organometallics 1996, 15, 1384.
-
each water molecule to two BF4 anions, through
hydrogen bonds (Figure 3). As a result, an intermolecu-
lar hydrogen-bonding network, with infinite chains of
cations and anions, exists. Within this network, two
distinct lengths of F‚‚‚H distances are observed (1.92(4)
(12) Macchioni, A.; Bellachioma, G.; Cardaci, G.; Travaglia, M.;
Zuccacia, C. Organometallics 1999, 18, 3061. (b) Zuccacia, C.; Mac-
chioni, A. Organometallics 1999, 18, 4367. (c) Macchioni, A.; Zuccacia,
C.; Clot, E.; Gruet, K.; Crabtree, R. H. Organometallics 2001, 20, 2367.

2290 Organometallics, Vol. 23, No. 10, 2004
van den Broeke et al.
tBu2)-2,6}(OH2)]BF4,16 and [Pd(5,8,11,14,17-pentaoxa-
2,20-dithia[21]-m-cyclophane)(OH2)]BF4,17 where F‚‚‚H
distances of 1.85-2.05 Å were reported. A more closely
related example of F‚‚‚H hydrogen bonding is found in
the structure of [Pd{C6H4(CH2NMe2)-2}(PPh3)(OH2)]-
PF6,13a where a F‚‚‚H distance of 2.087 Å was observed,
although the authors did not discuss the possibility of
hydrogen bonding. The F‚‚‚H distances observed in 5c
correspond with these previously reported examples.
Role of Hyd r ogen -Bon d in g In ter a ction s in th e
F or m a tion of Ca tion ic NCN-P d -a qu a Com p lexes.
Complex 4c is formed when [PdX(NCN)] (X ) Cl-, Br-,
I-, or CN-) is reacted with [Pd(NCN)(OH2)]BF4 in CH2-
18
Cl2 or by the reaction of [PdCl(NCN)] with AgBF4 in
dry acetone (vide supra). However, when the latter
reaction is carried out in the presence of water, [Pd-
(NCN)(OH2)]BF4 is formed. Thus, with the BF4- anion
both the mononuclear aqua complex and the dinuclear
chloride-bridged complexes are accessible.7b Therefore,
the observation that reaction of 1 with the tetraaryl-
borate salts 2a or 3a instead of AgBF4 results in the
selective formation of chloride-bridged 4a was surpris-
ing. From the crystallographic studies on 4a , 4c, and
5c it is clear that BF4- can easily form hydrogen bonds
with the Pd-aqua species, whereas [B{C6H4(SiMe3)-4}4]-
does not show this tendency. In fact, the large and
weakly coordinating [B{C6H4(SiMe3)-4}4]- anion has
little possibility for effective hydrogen bonding. Other
cationic palladium(II) aqua complexes that have been
spectroscopically characterized invariably contain hy-
drogen-bond acceptors in the form of small anions such
F igu r e 4. Calculated structures for 4a , 4c, 5a , and 5c.
Hydrogen atoms, except for H2O, are omitted for clarity.
of the ions was disregarded, because its assessment was
beyond the capability of the methods employed. Within
the limits of the basis set employed, the optimized
geometries of dinuclear 4a and 4c and mononuclear 5c
were good representations of the structures obtained
from the crystallographic studies on these complexes
(Figure 4).25 As 5a was experimentally inaccessible, its
geometry could not be compared with an experimentally
determined molecular structure. Calculated molecular
- 13b,19
- 13a
- 13c-f
- 13f,20
as BF4
,
PF6
,
ClO4
,
SO3CF3
,
and
phenoxide,21 or water15 and ether functionalities,17 in
close proximity to the water ligand. These observations
suggest that inter-ionic interactions, in the form of
hydrogen bonding, could be essential for the formation
of mononuclear palladium aqua species. However, given
the weakness of such interactions (vide infra), it remains
hard to understand why dinuclear complexes are pre-
ferred when the large tetraarylborate anions are used.
To make a more detailed evaluation of the effects of
BF4- and [B{C6H4(SiMe3)-4}4]- on the relative thermo-
dynamic stability of 4a , 4c, [Pd(NCN)(OH2)][B{C6H4-
(SiMe3)-4}4] (5a ), and 5c, these complexes were studied
using density functional theory (DFT) calculations. The
structures were preoptimized using semiempirical (PM3)
molecular modeling as implemented in the Spartan
5.1.1 (SGI) software package. For 4a , 4c, and 5c the
crystallographic data were used as a starting point for
these calculations. Geometry optimizations were final-
ized using the Gaussian 9822 and Gamess-UK SGI23
software. For these calculations, the LANL2DZ24 basis
set was chosen, as it offered a good compromise between
accuracy and computational power available. Solvation
-
models of 5c consistently showed the BF4 anion in a
F3BF‚‚‚H-O-Pd orientation (F‚‚‚H ) 1.60 Å, B-F‚‚‚H
) 113°, O-H‚‚‚F ) 171°, Scheme 2), which is charac-
teristic for hydrogen bonding and similar to that en-
countered in the crystal structure of [Pd(NCN)(OH2)]-
-
BF4. This propensity of BF4 for interacting with
(22) Frisch, M. J .; Trucks, G. W.; Schlegel, G. E.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zarzewski, V. G.; Montgomery, J . R.,
J r.; Stratman, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma,
K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J . B.;
Ciolowski, J .; Ortiz, J . V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaroni, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, G.; Head-Gordon, M.; Repogle, E. S.; Pople, J . A. Gaussian
98, revision A.9; Gaussian Inc.: Pittsburgh, PA, 1998.
(23) Guest, M. F.; van Lenthe, J . H.; Kendrick, J .; Scho¨ffel, K.;
Sherwood, P.; Harrison, R. J . GAMESS-UK, a package of ab initio
programs, 2000. With contributions from Amos, R. D.; Buenker, R. J .;
Dupuis, M.; Handy, N. C.; Hillier, I. H.; Knowles, P. J .; Bonacic-
Koutecky, V.; von Niessen, W.; Saunders, V. R.; Stone, A. J . It is derived
from the original GAMESS code due to Dupuis, M.; Spangler, D.;
Wendolowski, J . NRCC Software Catalog, 1980; Vol. 1, Program No.
QG01 (GAMESS).
(15) Ruiz, J .; Florenciano, F.; Vicente, C.; Ram´ırez de Arellano, M.;
Lo´pez, G. Inorg. Chem. Commun. 2000, 3, 73.
(16) Kimmich, B. F. M.; Marshall, W. J .; Fagan, P. J .; Hauptmann,
E.; Bullock, R. M. Inorg. Chim. Acta 2002, 330, 52.
(17) Kickham, J . E.; Loeb, S. J . Inorg. Chem. 1995, 34, 5656.
(18) [{Pd(NCN)}2(µ-X)]BF4 (X ) Cl-, Br-, I-, or CN-) are reported
in ref 7b.
(19) See Figure 3 for the hydrogen-bonding interactions in the
molecular structure of 5c.
(20) Maassarani, F.; Pfeffer, M.; Le Borgno, G. Organometallics
1987, 6, 2043.
(24) (a) Dunning, T. H.; Hay, P. J . In Modern Theoretical Chemistry;
Schaefer, H. J ., Ed.; Plenum: New York, 1976; p 1. (b) Hay, P. J .; Wadt,
W. R. J . Chem. Phys. 1985, 82, 270. (c) Wadt, W. R.; Hay, P. J . J .
Chem. Phys. 1985, 82, 284. (d) Hay, P. J .; Wadt, W. R. J . Chem. Phys.
1985, 82, 299.
(21) Canty, A. J .; J in, H.; Roberts, A. S.; Skelton, B. W.; White, A.
H. Organometallics 1996, 15, 5713.
(25) For comparison, a collection of bond lengths and angles is listed
in Table 4.

Cation-Anion Interactions in Ionic Complexes
Organometallics, Vol. 23, No. 10, 2004 2291
Sch em e 2. Hyd r ogen -Bon d in g In ter a ction s in th e
DF T-Op tim ized Str u ctu r es of 5a a n d 5c
4c) is higher (∆∆Ee ) 22 kcal/mol) than for [B{C6H4-
(SiMe3)-4}4]- (formation of 4a ). As no direct cation-
anion interactions are observed in the molecular models
of 4a and 4c, which is supported by the crystallographic
studies on both complexes,28 this difference in the
interaction with the anions must be electrostatic in
-
nature. As BF4 is smaller and has a higher charge
density than the silyl-substituted tetraarylborate anion,
stronger electrostatic interactions with the cation can
be easily envisioned. For the combinations of [Pd(NCN)-
Ta ble 3. Electr on ic En er gy Va lu es for th e
Com p on en ts P a r ticip a tin g in th e Rea ction s of
[P d Cl(NCN)] w ith AgBF 4 or Ag[B{C6H4(SiMe3)-4}4]
-
(OH2)]+ with either BF4 or [B{C6H4(SiMe3)-4}4]-
a
similar trend was observed (∆∆Ee ) 20 kcal/mol).
Furthermore, the calculations indicate that the effect
of hydrogen bonding on the ∆Ee of ion-pairing is
marginal. In 5c, which displayed extensive hydrogen-
bond formation in both the crystallographic and DFT
studies, the ∆Ee of ion-pairing is -90 kcal/mol. Calcula-
tions performed on H‚‚‚F bonding between BF3 and H2O
indicate that the energy associated with the formation
of this bond is only -3 kcal/mol, which is supported by
experimental determination of hydrogen-fluorine bonds
as well.29 Within the accuracy of the computational
methods, the contribution of the hydrogen bonds is not
significant and the energy related with the cation-anion
interactions is determined by the electrostatic contribu-
tion.
no.
component
Ee (hartree)a
H2O
BF3
-76.41432
-324.55458
-400.97410
-719.88593
-704.67694
-781.12747
-1424.62039
-14.99863
-424.56708
-1443.59058
-1205.83881
-2224.82984
-1849.29933
-2868.28808
F2BF‚‚‚HOH
[PdCl(NCN)]
[Pd(NCN)]+
1
[Pd(NCN)(OH2)]+
[{Pd(NCN)}2(µ-Cl)]+
Cl-
-
BF4
[B{C6H4(SiMe3)-4}4]-
5c
5a
4c
4a
[Pd(NCN)(OH2)]BF4
[Pd(NCN)(OH2)][B{C6H4(SiMe3)-4}4]
[{Pd(NCN)}2(µ-Cl)]BF4
[{Pd(NCN)}2(µ-Cl)][B{C6H4(SiMe3)-4}4]
a
1 hartree ) 627.52 kcal/mol.
The results of the calculations can be summarized by
the simplified model shown in Scheme 4, which allows
a comparison of relative stability of mononuclear and
dinuclear complexes.
hydrogen-bond donors26 was also reported recently in a
theoretical study on H3O+‚‚‚BF4 ion pairs.27 In the
-
structure of 5a the hydrogen atoms of the water ligand
are also in close contact with the anion, the close contact
distances with the phenyl ring carbons of the tet-
raarylborate anion being 2.18 and 2.26 Å (Scheme 2).
Subsequently, the cation-anion interactions in 4a ,
4c, 5a , and 5c were assessed by stepwise assembly of
these complexes starting from the [Pd(NCN)]+ cation
(Scheme 3). To arrive at the energies associated with
each step represented in Scheme 3, the electronic energy
difference between the product and the components
from which the product was composed was calculated;
see Table 3.
From Scheme 3, the differences in energy as a
consequence of the cation-anion interactions in the
mononuclear and dinuclear complexes can be clearly
established. A large drop in energy (∆Ee of -132 kcal/
mol) is associated with the formation of [PdCl(NCN)]
from [Pd(NCN)]+ and Cl-. Subsequent interaction of
neutral [PdCl(NCN)] with another [Pd(NCN)]+ cation
yields just an additional -36 kcal/mol, and when [Pd-
(NCN)]+ is combined with H2O to form [Pd(NCN)-
(OH2)]+, ∆Ee is only -23 kcal/mol. Apparently, H2O is
a much weaker Lewis base than [PdCl(NCN)]. As a
result, the formation of a dinuclear [{Pd(NCN)}2(µ-Cl]+
cation is more favorable than that of two mononuclear
[Pd(NCN)(OH2)]+ cations when no charge-compensating
anion is present.
As the formation of the dinuclear cation is energeti-
cally more favorable than that of the mononuclear
cation, strong cation-anion interactions are required
for the aqua complex to be stable. When these are
absent, as in the presence of [B{C6H4(SiMe3)-4}4]-, the
stability of the chloride-bridged dinuclear species is
much higher (Scheme 4). This corresponds with the
experimentally observed preference for the formation
of the dinuclear complex 4a when using the tetraarylbo-
rate anion. When employing BF4-, the inter-ionic in-
teractions are stronger, which enhances the stability of
5c to such an extent that its formation is almost equally
favorable to 4c (Scheme 4), in correspondence with the
experimental accessibility of both complexes. For
[B{C6H4(SiMe3)-4}4]- (5a ) the aqua complex is clearly
less favorable than for BF4- (5c) by as much as 19 kcal/
mol.
-
As Scheme 4 shows that even in the presence of BF4
a modest preference for the dinuclear species remains,
the experimentally observed preference for the aqua
complex suggests the existence of an additional stabiliz-
ing contribution not taken into account in our calcula-
tions. Most likely, the high formation enthalpy of AgCl
(or NaCl) provides an explanation for preference of the
formation of the mononuclear compound. In the prepa-
ration of 5c, 1 equiv of AgCl is formed per Pd against 1
equiv per two Pd centers for 4c. As a result of the
enthalpy associated with the formation of AgCl, the
equilibrium between the dinuclear and mononuclear
species will shift toward the latter, irrespective of the
Upon introduction of the charge-compensating anions,
a major difference between the small BF4- and the more
bulky [B{C6H4(SiMe3)-4}4]- could be discerned. In the
case of the dinuclear [{Pd(NCN)}2(µ-Cl)]+ cation the
energy gained from ion-pairing with BF4- (formation of
(28) For 4a vide supra; for 4c see ref 7b.
(29) Hydrogen bonds of the type OH2‚‚‚F are generally weak in
character, with binding energies of ca. 2.4 kcal/mol: Desiraju, G. R.;
Steiner, T. In The Weak Hydrogen Bond in Structural Chemistry and
Biology; Oxford University Press: Oxford, 1999.
(26) For reviews see: (a) Rosenthal, M. R. J . Chem. Educ. 1973,
50, 331. (b) Beck, W.; Su¨nkel, K. Chem. Rev. 1988, 88, 1405.
(27) Farcasiu, D.; Haˆncu, D. J . Phys. Chem. A 1999, 103, 754.

2292 Organometallics, Vol. 23, No. 10, 2004
van den Broeke et al.
Sch em e 3. En er gy of F or m a tion (k ca l/m ol) for 4a , 4c, 5a , a n d 5c a s Assem bled fr om Sep a r a te Com p on en ts
(fr om DF T ca lcu la tion s, see text)
Sch em e 4
this study it has been shown that the interactions
between the cation and the anion play a determining
role in the formation of these complexes. A small anion
that can participate in strong electrostatic interactions
with the cationic aqua species appears to be a prereq-
uisite for the formation of stable [Pd(NCN)(OH2)]+
complexes. Hydrogen-bonding interactions were also
observed when BF4- was used; however their enthalpic
contribution to the stability of palladium aqua species
appears to be marginal. When the possibility for strong
stabilizing interactions between the anion and [Pd-
(NCN)(OH2)]+ cation is lacking, as is the case with the
very weakly coordinating silyl-substituted tetraphen-
ylborate anions, formation of a bridged dinuclear pal-
ladium species is preferred.
anion. The small difference in Ee between the mono-
nuclear and dinuclear complexes with the BF4 anion
-
implies that 5c will be the energetically most favorable,
whereas for the [B{C6H4(SiMe3)-4}4]- anion 4a remains
the most favorable configuration. The presence of an
excess of water instead of stoichiometric amounts might
be an additional factor that shifts the equilibrium
toward mononuclear species.
With respect to the relative stabilities of other known
cationic Pd-aqua and corresponding chloride-bridged
complexes it can be said that there is very little
reference material. However, the few known aqua
complexes are invariably stabilized by small anions that
form H-bridges to the water ligand. Strong electrostatic
interactions no doubt add to the stability of these
complexes. Brookhart’s [(N∧N)PdMe(OEt2)][BArf] (N∧N
) diimine ligand) complexes containing a fluorinated
tetraarylborate anion30 need Et2O as Lewis base to fill
up the empty coordination site on the Pd center. [(dppp)-
Pd((C(O)Me)(CO)][TFPB]31 and [(N∧N)Pd(Me)(MeCN)]-
[TFPB] (N∧N ) 2,2′-bipyridine or 1,10-phenanthroline)32
also contain strongly coordinating ligands to keep them
mononuclear. The only somewhat comparable halide-
bridged complex, [{Pd[C,N-C(NXy)C6H4-NH2](CNXy)}2-
As pointed out by one of the referees of this paper,
although the calculations give an acceptable thermo-
dynamic explanation of the observations, kinetic factors
such as difference in the rates of attack of the different
borates might be important as well. Unfortunately,
quantitative kinetic studies are severly hindered by the
fact that the reaction mixtures are inhomogeneous due
to salt formation. One aspect that may well play a
crucial kinetic role is the H-F interaction to afford a
[BF4‚‚‚H2O]- fragment that makes water a better nu-
cleophile. In an associative mechanism where (NCN)-
PdCl has to be replaced from the [(NCN)Pd-Cl-
Pd(NCN)]+ fragment, the increased negative charge and
Lewis basicity of water might be important. A general
kinetic resistance of the dinuclear species toward chlo-
ride abstraction, irrespective of the type of borate, is
certainly not the problem, as was shown by the fast
reaction with AgBF4.
(30) (a) J ohnson, L. K.; Killian, C. M.; Brookhart, M. J . Am. Chem.
Soc. 1995, 117, 6414. (b) J ohnson, L. K.; Mecking, S.; Brookhart, M.
J . Am. Chem. Soc. 1996, 118, 267. (c) Tempel, D. J .; J ohnson, L. K.;
Huff, R. L.; White, P. S.; Brookhart, M. J . Am. Chem. Soc. 2000, 122,
6686. (d) Guan, Z.; Cotts, P. M.; McCord, E. F.; McLain, S. J . Science
1999, 283, 2059.
Con clu sion s
(31) Schultz, C. S.; Ledford, J .; DeSimone, J . M.; Brookhart, M. J .
Am. Chem. Soc. 2000, 122, 6351.
(32) Brookhart, M.; Rix, F. C.; DeSimone, J . M. J . Am. Chem. Soc.
1992, 114, 5894.
Whereas, previously, hardly any attention was given
to the nature of the weakly coordinating anion in the
formation of cationic [Pd(NCN)(OH2)]+ complexes, in

Cation-Anion Interactions in Ionic Complexes
Organometallics, Vol. 23, No. 10, 2004 2293
µ-Ch lor ob is[2,6-b is{(d im et h yla m in o)m et h yl}p h en yl-
p a lla d iu m (II)] Tetr a k is[4-(tr im eth ylsilyl)p h en yl]bor a te
(4a ) Usin g Ag[B{C6H4(SiMe3)-4}4] (3a ). Using a procedure
similar to the preparation of 4a from 1 and Na[B{C6H4(SiMe3)-
4}4], 0.25 g (0.76 mmol) of 1 and 0.54 g (0.76 mmol) of 3a
(µ-I)]OTf, is an isocyanide complex based on a triflate
anion,10i demonstrating that dinuclear complexes can
be obtained also in the presence of a small anion when
a strongly coordination Lewis base is absent. It thus
appears that mononuclear and monocationic aqua com-
plexes of the type [L3PdII(OH2)]+ can be stable only
when the water interacts efficiently with the anion. For
complexes containing tetraarylborate anions, where
these interactions are lacking, water becomes too weak
a base and more strongly donating ligands are needed
to stabilize these compounds in mononuclear form.
1
yielded 0.60 g (64%) of a light yellow solid. H NMR (acetone-
d6, 300.1 MHz): δ 0.17 (s, 36H), 2.94 (s, 24H), 4.08 (s, 8H),
6.80 (dd, 4H), 6.97 (dd, 2H), 7.15 (d, 8H), 7.43 (m, 8H).
µ-Ch lor ob is[2,6-b is{(d im et h yla m in o)m et h yl}p h en yl-
p a lla d iu m (II)] Tet r a k is[4-{d im et h yl(1H ,1H ,2H ,2H -p er -
flu or ooctyl)silyl}p h en yl]bor a te (4b). Using a procedure
similar to the preparation of 4a , 0.30 g (1.00 mmol) of [PdCl-
{C6H3(CH3NMe2)2-2,6}] and 2.25 g (1.00 mmol) of Na[B-
{C6H4(SiMe2CH2CH2C6F13)-4}4] yielded 0.90 g (80%) of a light
yellow solid. 1H NMR (acetone-d6, 300.1 MHz): δ 0.27 (s, 24H),
0.93 (m, 8H), 2.15 (m, 8H), 2.93 (s, 24H), 4.13 (s, 8H), 6.83
(dd, 4H), 7.00 (dd, 2H), 7.10 (d, 8H), 7.40 (m, 8H). 13C{1H}
NMR (acetone-d6, 75.5 MHz): δ -3.7 (s), 5.5 (s), 26.2 (t), 52.7
(s), 74.0 (s), 108.2 (m), 111.2 (m), 115.6 (m), 115.9 (m), 119.3
(m), 120.4 (s), 122.8 (m), 125.4 (s), 128.0 (s), 130.9 (m), 136.3
Exp er im en ta l Section
Gen er a l P r oced u r es. All operations were carried out
under a dry, oxygen-free, nitrogen atmosphere, using standard
Schlenk techniques. Acetone was distilled from CaCl2, CH2-
Cl2 was distilled from CaH2, and hexane was distilled from
sodium benzophenone ketyl. Na[B{C6H4(SiMe3)-4}4],5a Na-
[B(C6H4{SiMe2(CH2CH2C6F13)}-4)4],5a and [PdCl(NCN)] (1)6
were prepared according to reported procedures. AgBF4 was
dried in vacuo at 100 °C for 16 h before use. All other chemicals
were used as received. NMR spectra were recorded on a Varian
Inova 300 spectrometer and referenced externally against
TMS. Acetone-d6 and CD2Cl2 were dried over molecular sieves
(3 Å). Infrared spectra were recorded on a Mattson Galaxy FT-
IR 5000 spectrometer. Elemental analyses were carried out
by H. Kolbe, Mikroanalytisches Laboratorium, Mu¨lheim an der
Ruhr.
Silver Tetr a k is[4-(tr im eth ylsilyl)p h en yl]bor a te (3a ). A
solution of 0.23 g (1.40 mmol) of AgNO3 in 10 mL of deminer-
alized water was solidified upon cooling to -78 °C. A solution
of 1.00 g (1.40 mmol) of Na[B{C6H4(SiMe3)-4}4] in 25 mL of
EtOH was layered on top, and the mixture was allowed to
warm to room temperature. A precipitate formed, which was
filtered off and washed with water and ethanol. Drying in
vacuo yielded 0.70 g (70%) of a white solid. Anal. Calcd for
1
(s), 145.5 (s), 154.0 (s), 165.8 (q, J BC ) 49.3 Hz). Anal. Calcd
for C88H94BClF52N4Pd2Si4: C, 41.17; H, 3.69; Cl, 1.38; N, 2.18.
Found: C, 41.35; H, 3.62; Cl, 1.35; N, 2.11.
µ-Ch lor ob is[2,6-b is{(d im et h yla m in o)m et h yl}p h en yl-
p a lla d iu m (II)] Tet r a k is[4-{d im et h yl(1H ,1H ,2H ,2H -p er -
flu or ooctyl)silyl}p h en yl]bor a te (4b) Usin g Ag[B(C6H4-
{SiMe2(CH2CH2C6F 13)}-4)4] (3b). Using a procedure similar
to the preparation of 4b from 1 and Na[B(C6H4{SiMe2(CH2-
CH2C6F13)}-4)4], 0.13 g (0.40 mmol) of 1 and 0.82 g (0.40 mmol)
of 3b yielded 0.46 g (93%) of a light yellow solid. 1H NMR
(acetone-d6, 300.1 MHz): δ 0.27 (s, 24H), 0.93 (m, 8H), 2.15
(m, 8H), 2.93 (s, 24H), 4.13 (s, 8H), 6.83 (dd, 4H), 7.00 (dd,
2H), 7.10 (d, 8H), 7.40 (m, 8H).
µ-Ch lor ob is[2,6-b is{(d im et h yla m in o)m et h yl}p h en yl-
p a lla d iu m (II)]BF 4 (4c).7 To 0.14 g (0.70 mmol) of AgBF4 was
added 0.22 g (0.66 mmol) of 1 in acetone (10 mL). After stirring
the resulting mixture for 1 h, it was filtered over dried Celite
and all volatiles were removed in vacuo. The resulting waxy
solid was taken up in CH2Cl2 (20 mL), all solids were separated
by centrifugation, and the solution was concentrated to 5 mL.
Addition of hexane resulted in the precipitation of 0.05 g (10%)
C
36H52AgBSi4: C, 60.40; H, 7.32; Ag, 15.07. Found: C, 59.92;
H, 7.40; Ag, 14.18. IR (KBr, cm-1): 3045, 2953, 2895, 1575,
1359, 1248, 1143, 1087, 841, 804, 756.
1
of a white solid. H NMR (acetone-d6, 300.1 MHz): δ 2.83 (s,
Silver Tet r a k is[4-{d im et h yl(1H ,1H ,2H ,2H -p er flu or o-
octyl)silyl}p h en yl]bor a te (3b). A solution of 0.12 g (0.70
mmol) of AgNO3 in 5 mL of demineralized water and 5 mL of
EtOH was solidified at -78 °C. A solution of 1.55 g (0.80 mmol)
of Na[B(C6H4{SiMe2(CH2CH2C6F13)}-4)4] in 10 mL of EtOH
was layered on top, and the mixture was allowed to warm to
room temperature. The solid products isolated were separated
by centrifugation and washed with water and EtOH. Drying
in vacuo yielded 1.50 g (95%) of a white solid. Anal. Calcd for
C64H56AgBF52Si4: C, 37.61; H, 2.76; Ag, 5.28. Found: C, 37.64;
H, 2.94; Ag, 5.20. IR (KBr, cm-1): 3049, 2960, 1575, 1456, 1361,
1244, 1211, 1143, 1068, 1018, 896, 841, 810, 734, 707.
µ-Ch lor ob is[2,6-b is{(d im et h yla m in o)m et h yl}p h en yl-
p a lla d iu m (II)] Tetr a k is[4-(tr im eth ylsilyl)p h en yl]bor a te
(4a ). To a solution of 0.13 g (0.40 mmol) of 1 in acetone (20
mL) were added 0.30 g (0.40 mmol) of Na[B{C6H4(SiMe3)-4}4]
and H2O (0.25 mL, 13.8 mmol). The resulting suspension was
stirred for 16 h, filtered over Celite, and evaporated to dryness.
The solid product was dissolved in acetone (1 mL), and pentane
was added (20 mL). Centrifugation was used to remove small
amounts of insolubles, and the resulting solution was dried
in vacuo, yielding 0.51 g (95%) of a light yellow solid. Slow
diffusion of pentane into a concentrated solution of 4a in
MeOH yielded crystals suitable for X-ray analysis. 1H NMR
(acetone-d6, 300.1 MHz): δ 0.17 (s, 36H), 2.94 (s, 24H), 4.08
(s, 8H), 6.80 (dd, 4H), 6.97 (dd, 2H), 7.15 (d, 8H), 7.43 (m, 8H).
13C{1H} NMR (acetone-d6, 75.5 MHz): δ -1.1 (s), 52.8 (s), 74.4
(s), 119.9 (s), 124.8 (s), 130.7 (s), 136.0 (s), 145.7 (s), 165.4 (q,
1J BC ) 49.3 Hz). Anal. Calcd for C60H90BClN4Pd2Si4: C, 58.17;
H, 7.32; N, 4.52. Found: C, 56.50; H, 7.21; N, 3.89.
24H), 4.13 (s, 8H), 6.84 (dd, 4H), 7.03 (dd, 2H). 13C{1H} NMR
(acetone-d6, 75.5 MHz): δ 51.9 (s), 73.3 (s), 120.1 (s), 124.4
(s), 146.2 (s), 150.9 (s). Anal. Calcd for C24H38BClF4N4Pd2: C,
40.17; H, 5.54. Found: C, 39.94; H, 5.54.
[2,6-Bis{(d im eth yla m in o)m eth yl}p h en ylp a lla d iu m (II)-
a qu a ]BF 4 (5c).6 Using a procedure similar to the preparation
of 4a , 0.20 g (0.66 mmol) of 1 and 0.14 g (0.71 mmol) of AgBF4
yielded 0.27 g (97%) of a white solid. Crystals suitable for X-ray
analysis were grown by slow diffusion of hexane into a CH2-
1
Cl2 solution of 5c in the presence of water. H NMR (acetone-
d6, 300.1 MHz): δ 2.86 (s, 12H), 3.30 (s, H2O), 4.18 (s, 4H),
6.85 (dd, 2H), 7.04 (dd, 1H). 13C{1H} NMR (acetone-d6, 75.5
MHz): δ 51.8 (s), 73.4 (s), 120.5 (s), 125.9 (s), 145.6 (s), 150.9
(s).
Reaction of µ-Ch lor obis[{2,6-bis[(dim eth ylam in o)m eth -
yl]p h en yl-N,C,N′}p a lla d iu m (II)] Tet r a k is[4-{d im et h yl-
(1H ,1H ,2H ,2H -p er flu or ooct yl)silyl}p h en yl]b or a t e (4b )
w ith AgBF 4. To a solution of 0.05 g (0.02 mmol) of 4b in
acetone (5 mL) was added 3.8 mg (0.02 mmol) of AgBF4. The
formed suspension was filtered over Celite and evaporated to
dryness. A light yellow solid was obtained that was analyzed
1
by H and 11B NMR to be a 2:1 molar mixture of 5c and 4b.
The material was redissolved, and an additional equivalent
of AgBF4 was added following the above procedure to afford
5c. 1H NMR (acetone-d6, 300.1 MHz): δ 7.04 (dd, 1H), 6.85
(dd, 2H), 4.18 (s, 4H), 2.86 (s, 12H). 11B NMR (acetone-d6, 96.3
MHz): δ 4.37 (s, BF4).
Cr yst a l St r u ct u r e Det er m in a t ion of 4a . [C24H38ClN4-
Pd2]+‚[C36H52BSi4]-, Mr ) 1238.82. A colorless, plate-shaped

2294 Organometallics, Vol. 23, No. 10, 2004
van den Broeke et al.
crystal (0.05 × 0.15 × 0.35 mm3) was fixed to a glass capillary
and transferred into the cold nitrogen stream on a Nonius
KappaCCD area detector diffractometer with rotating anode.
The measured crystal was monoclinic, space group P21/c (no.
14) with a ) 16.8686(15), b ) 20.674(3), c ) 25.349(5) Å, â )
Ta ble 4. Com p a r a tive Da ta for th e Geom etr y of
4a , 4c, a n d 5c Obta in ed fr om Cr ysta llogr a p h ic
Stu d ies a n d DF T Stu d ies
bond distancesa
bond anglesa
X-ray DFT
X-ray DFT
129.298(12)°, V ) 6841.3(17) Å3, Z ) 4, Dx ) 1.203 g cm-3
,
F(000) ) 2592, µ(Mo KR) ) 0.671 mm-1. A total of 116 887
reflections were measured, 15 709 of which were independent,
(Rσ ) 0.0468, Rint ) 0.0874, 1.0° < θ < 27.5°, T ) 150 K,
Mo KR radiation, graphite monochromator, λ ) 0.71073 Å,
φ scan and ω scans with κ offset, distance crystal to detector
50 mm, data reduction by DENZO,33 refined mosaicity 0.381-
(1)°, no absorption correction). The structure was solved by
direct methods (SHELXS-86)34 and refined on F2 using
SHELXL-97-2.35 The disorder in one of the NCN ligands was
described with a two-site disorder model. The occupancy factor
of the major component refined to 0.710(7). A total of 327
distance restraints were introduced to ensure equal geometries
in chemically equivalent bonds and 1,3-distances in the NCN
ligands. Hydrogen atoms were included in the refinement on
calculated positions riding on their carrier atoms. All ordered
non-hydrogen atoms were refined with anisotropic displace-
ment parameters. The non-hydrogen atoms in the major
disorder component were refined with isotropic displacement
parameters; the displacement parameters of the minor com-
ponent atoms were equated to the parameters of the corre-
sponding major component atoms. Hydrogen atoms were
refined with a fixed isotropic displacement parameter related
to the value of the equivalent isotropic displacement parameter
of their carrier atoms. Refinement of 618 parameters con-
verged at wR2 ) 0.1548, w ) 1/[σ2(F2) + (0.0568P)2 + 21.76P],
where P ) (max(Fo2,0) + 2Fc2)/3, R1 ) 0.0643 (for 12 285 I >
2σ(I)), S ) 1.045, and -0.97 < ∆F < 1.76 e Å-3 (near a
disordered SiMe3 group).
4a
Pd(1)-C(9) 1.906 1.947 Pd(1)-Cl(1)-Pd(2) 121.11 137.48
Pd(1)-N(1) 2.102 2.149 N(1)-Pd(1)-N(2) 162.97 162.82
Pd(1)-N(2) 2.091 2.174 C(9)-Pd(1)-Cl(1) 176.87 176.12
Pd(1)-Cl(1) 2.453 2.630
B(1)-C(25) 1.653 1.657 C(25)-B(1)-C(34) 110.1 109.1
B(1)-C(34) 1.644 1.654 C(25)-B(1)-C(43) 110.4 110.7
4c
Pd(1)-C(9) 1.929 1.946 Pd(1)-Cl(1)-Pd(2) 134.8 139.9
Pd(1)-N(1) 2.105 2.174 N(1)-Pd(1)-N(2) 162.6 161.4
Pd(1)-N(2) 2.100 2.168 C(9)-Pd(1)-Cl(1) 174.5 173.3
Pd(1)-Cl(1) 2.463 2.656
B(1)-F(1)
B(1)-F(2)
1.318 1.467 F(1)-B(1)-F(2)
1.356 1.452 F(1)-B(1)-F(3)
114.6 110.5
107.2 106.9
5c
Pd(1)-C(1) 1.909 1.940 N(1)-Pd(1)-N(2) 162.7 162.0
Pd(1)-N(1) 2.107 2.155 C(1)-Pd(1)-O(1)
Pd(1)-N(2) 2.108 2.170
176.8 175.0
Pd(1)-O(1) 2.192 2.220 F(1)-B(1)-F(2)
110.7 112.7
107.2 105.7
B(1)-F(3)
B(1)-F(4)
1.401 1.495 F(2)-B(1)-F(4)
1.386 1.441
F(3)-H(1a) 1.92
1.602 O(1)-H(1a)-F(3) 176
170.7
a
Numbering according to the scheme applied in the crystal-
lographic studies.
functional calculations (DFT). Initial molecular conformations
were calculated using the semiempirical (PM3(tm)) method as
implemented in the Spartan 5.1.1 (SGI) molecular modeling
software package. The geometry of the thus obtained struc-
tures was further optimized using Becke’s three-parameter
hybrid functional and the Lee, Yang, and Parr correlation
functional (B3LYP)38 as implemented in both the Gamess-UK
SGI (version 6.3)23 and the Gaussian 9822 programs. All atoms
are described with the double-ú basis set LANL2DZ,24 with
effective core potentials being used for the atoms Si, Cl, and
Pd. Due to the size of the calculations on the structures
containing the [B{C6H4(SiMe3)-4}4]- anion, frequency analyses
could not be completed within an acceptable time span.
Therefore only the electronic energies were taken into account
in this study. Absolute energies and atomic coordinates for the
calculated molecules can be supplied on request from the
author.
Cr yst a l St r u ct u r e Det er m in a t ion of 5c. [C12H21N2-
OPd]+‚[BF4]-, Mr ) 402.52. Intensity data were collected for
a colorless, needle-shaped crystal (0.30 × 0.12 × 0.12 mm3)
using the same procedure as for the structure determination
of 4a . Of the 19 397 reflections measured, 3820 were unique,
3179 with I > 2σ(I) (Rint ) 0.0460). An absorption correction
was considered unnecessary (µ ) 1.150 mm-1). The structure
was solved by automated Patterson methods using DIRDIF99.36
Refinement of 202 parameters was performed as for 4a , except
that the hydrogen atoms of the coordinated water molecule
were refined with isotropic displacement parameters. Struc-
ture validation and molecular graphics preparation were
performed with the PLATON package.37 The measured crystal
was monoclinic, space group P21/c (no. 14) with a ) 9.0800(2),
b ) 18.5226(4), c ) 12.1707(2) Å, â ) 125.423(1)°, V )
1668.03(6) Å3, Z ) 4, Dx ) 1.603 g cm-3, R(F) [I > 2σ(I)] )
0.0273, wR(F2) ) 0.0634, S ) 1.084, and -0.86 < ∆F < 0.70
Ack n ow led gm en t. This work was financially sup-
ported by ATOFINA Vlissingen B.V. and the Dutch
Ministry of Economic Affairs (Senter/BTS) (J .v.d.B.,
B.J .D.), the Council for Chemical Sciences of The
Netherlands Organisation for Scientific Research (CW-
NWO) (A.L.S.), and The Netherlands Organisation for
Scientific Research by making available the SGI TERAS
Computer of SARA (Amsterdam) (J .H.v.L., P.J .A.R.).
e Å-3
.
Th eor etica l Stu d ies. The relative energies of complexes
and their separate components were evaluated using density
(33) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
(34) Sheldrick, G. M. SHELXS-86, Program for Crystal Structure
Determination; University of Go¨ttingen: Germany, 1986.
(35) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure
Refinement; University of Go¨ttingen: Germany, 1997.
(36) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garc´ıa-Granda,
S.; Gould, O.; Israel, R.; Smits, J . M. M. The DIRDIF99 program
system; Crystallography Laboratory, University of Nijmegen: The
Netherlands, 1999.
Su p p or tin g In for m a tion Ava ila ble: Crystal data and
structure refinement details for 4a and 5c. This material is
available free of charge via the Internet at http://pubs.acs.org.
OM0303698
(37) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
Utrecht University: The Netherlands, 2002.
(38) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.