Regioselective ortho palladation of stabilized iminophosphoranes in exo positions: Scope, limitations, and mechanistic insights

Article abstract of DOI:10.1021/om701174v

The reactivity of the stabilized iminophosphoranes R₃P=NC(O)Ph (R = p-tolyl, m-tolyl), Ph₃P=NC(O)CH₂Ph, and Ph ₂P(CH₂)_nPPh₂=NC(O)Ph (n = 1, 2) toward Pd(OAc)₂ has been examined. These substrates undergo palladation at the ortho C(sp²)-H bond of the benzamide ring, giving exo complexes with complete regioselectivity. The mechanism of this reaction has been studied using DFT calculations and the regioselectivity explained on a kinetic basis, since the activation barrier is lower for the exo pathway than for the endo. The origin of this lower energy for the exo barrier seems to reside mainly on the interaction energy between the metal center and the ligand at the transition state. On the other hand, the palladation of the related keto-stabilized systems Ph₃P=NC(O)FG (FG = NC₄H ₈, NC₄H₈O) gives selectively the endo isomers through C-H bond activation at the Ph ring of the PPh₃ unit.

Full text of DOI:10.1021/om701174v

Organometallics 2008, 27, 2929–2936
2929
Regioselective Ortho Palladation of Stabilized Iminophosphoranes in
Exo Positions: Scope, Limitations, and Mechanistic Insights
David Aguilar,† Raquel Bielsa,† Mar´ıa Contel,‡ Agust´ı Lledo´s,*,§ Rafael Navarro,†
Tatiana Soler,| and Esteban P. Urriolabeitia*,†
Departmento de Compuestos Organometa´licos, ICMA, UniVersidad de Zaragoza-CSIC, Plaza S. Francisco
s/n, E-50009 Zaragoza, Spain, Chemistry Department, Brooklyn College and The Graduate Center
(CUNY), 2900 Bedford AVenue, Brooklyn, New York 11210, Departament de Qu´ımica, Edifici C.n,
UniVersitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain, and SerVicios Te´cnicos de
InVestigacio´n, Facultad de Ciencias Fase II, UniVersidad de Alicante, 03690 San Vicente de Raspeig,
Alicante, Spain
ReceiVed NoVember 23, 2007
The reactivity of the stabilized iminophosphoranes R3PdNC(O)Ph (R ) p-tolyl, m-tolyl),
Ph3PdNC(O)CH2Ph, and Ph2P(CH2)nPPh2)NC(O)Ph (n ) 1, 2) toward Pd(OAc)2 has been examined.
These substrates undergo palladation at the ortho C(sp2)-H bond of the benzamide ring, giving exo
complexes with complete regioselectivity. The mechanism of this reaction has been studied using DFT
calculations and the regioselectivity explained on a kinetic basis, since the activation barrier is lower for
the exo pathway than for the endo. The origin of this lower energy for the exo barrier seems to reside
mainly on the interaction energy between the metal center and the ligand at the transition state. On the
other hand, the palladation of the related keto-stabilized systems Ph3PdNC(O)FG (FG ) NC4H8, NC4H8O)
gives selectively the endo isomers through CsH bond activation at the Ph ring of the PPh3 unit.
nophosphoranes MenR3-nPdNCH2Aryl can be metalated: (i) at
the R ring (phenyl or tolyl), giving endo derivatives; (ii) at the
Aryl ring, giving exo complexes; (iii) even at both rings at the
same time, this behavior being a function of the different
substituents present in the molecule. The behavior of the
stabilized iminophosphoranes Ph3PdNC(O)Aryl is quite more
selective than that shown by the other substrates, since in all
cases studied the palladation has been produced at the aryl ring
directly bonded to the deactivating carbonyl group.5 This
selective exo metalation is remarkable, since endo metalation
products seem to be more stable than the exo products on
electronic grounds (metalloaromaticity),7 although some con-
troversy have appeared recently around this concept, on the basis
Introduction
The metal-mediated CH bond activation reaction is recognized
at present as being one of the most important research subjects,
mainly due to its deep implications for the functionalization of
organic molecules.1 In this context, a selective activationswhen
several possibilities existsis highly desirable, and synthetic
strategies must be developed. For instance, when dealing with
aromatic substrates, ortho functionalization is easily achieved
through the introduction of an ancillary coordinating group on
the starting substrate.2 These reactions occur through the
formation of ortho-metalated complexes, which are very valu-
able tools in stoichiometric and/or catalytic processes.3 Follow-
ing our current research in this field,4 we have recently reported
the selective ortho palladation of phosphorus ylides5 and
iminophosphoranes.5,6 The results showed that both the stabi-
lized P ylides Ph3PdCHC(O)Aryl and the nonstabilized imi-
(3) (a) Dupont, J.; Pfeffer, M.; Spencer, J. Eur. J. Inorg. Chem. 2001,
1917. (b) Kakiuchi, F.; Murai, S. Acc. Chem. Res. 2002, 35, 826. (c)
Kakiuchi, F.; Chatani, N. AdV. Synth. Catal. 2003, 345, 1077. (d) Bedford,
R. B. Chem. Commun. 2003, 1787. (e) van der Boom, M. E.; Milstein, D.
Chem. ReV. 2003, 103, 1759. (f) Omae, I. Coord. Chem. ReV. 2004, 248,
995. (g) Farina, V. AdV. Synth. Catal. 2004, 346, 1553. (h) Schlummer, B.;
Scholz, U. AdV. Synth. Catal. 2004, 346, 1599. (i) Beletskaya, I. P.;
Cheprakov, A. V. J. Organomet. Chem. 2004, 689, 4055. (j) Dupont, J.;
Consorti, C.; Spencer, J. Chem. ReV. 2005, 105, 2527. (k) Kalyani, D.;
Deprez, N. R.; Desai, L. V.; Sanford, M. S. J. Am. Chem. Soc. 2005, 127,
7330. (l) Zapf, A.; Beller, M. Chem. Commun. 2005, 431. (m) Daugulis,
O.; Zaitsev, V. G.; Shavashov, D.; Pham, Q. N.; Lazareva, A. Synlett 2006,
3382. (n) Vicente, J.; Abad, J. A.; Lo´pez-Sa´ez, M. J.; Jones, P. G.
Organometallics 2006, 25, 1851. (o) Hull, K. L.; Lanni, E. L.; Sanford,
M. S. J. Am. Chem. Soc. 2006, 128, 14047. (p) Dick, A. R.; Sanford, M. S.
Tetrahedron 2006, 62, 2439. (q) Daugulis, O.; Chiong, H. A. Org. Lett.
2007, 9, 1449.
* To whom correspondence should be addressed. Fax: (+34) 976761187
(E.P.U.). E-mail: esteban@unizar.es (E.P.U.); agusti@klingon.uab.es (A.L.).
† Universidad de Zaragoza-CSIC.
‡ Brooklyn College and The Graduate Center (CUNY).
§ Universitat Auto`noma de Barcelona.
| Universidad de Alicante.
(1) (a) Ryabov, A. D. Synthesis 1985, 233. (b) Murai, S.; Kakiuchi, F.;
Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda, M.; Chatani, N. Nature 1993,
366, 529. (c) Dyker, G. Angew. Chem., Int. Ed. 1999, 38, 1698. (d) Rietling,
V.; Sirlin, C.; Pfeffer, M. Chem. ReV. 2002, 102, 1731. (e) Goldberg, K. I.;
Goldman, A. S., Eds. Organometallic C-H Bond ActiVation; ActiVation
and Functionalization of C-H Bonds; American Chemical Society:
Washington, DC, 2004; ACS Symposium Series 885. (f) Jones, W. D. Inorg.
Chem. 2005, 44, 4475. (g) Godula, K.; Sames, D. Science 2006, 312, 67.
(h) Deprez, N. R.; Sanford, M. S. Inorg. Chem. 2007, 46, 1924. (i) Bercaw,
J. E.; Labinger, J. A. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 6899. (j)
Chen, G. S.; Labinger, J. A.; Bercaw, J. E. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 6915.
(4) (a) Bielsa, R.; Larrea, A.; Navarro, R.; Soler, T.; Urriolabeitia, E. P.
Eur. J. Inorg. Chem. 2005, 1724. (b) Aguilar, D.; Contel, M.; Navarro, R.;
Urriolabeitia, E. P. Organometallics 2007, 26, 4604.
(5) Aguilar, D.; Aragu¨e´s, M. A.; Bielsa, R.; Serrano, E.; Navarro, R.;
Urriolabeitia, E. P. Organometallics 2007, 26, 3541.
(6) Bielsa, R.; Navarro, R.; Urriolabeitia, E. P.; Lledo´s, A. Inorg. Chem.
2007, 46, 10133.
(2) (a) Clegg, W.; Dale, S. H.; Hevia, E.; Honeyman, G. W.; Mulvey,
R. E. Angew. Chem., Int. Ed. 2006, 45, 2370. (b) Clegg, W.; Dale, S. H.;
Harrington, R. W.; Hevia, E.; Honeyman, G. W.; Mulvey, R. E. Angew.
Chem., Int. Ed. 2006, 45, 2374.
(7) (a) Masui, H. Coord. Chem. ReV. 2001, 219-221, 957. (b) Ghedini,
M.; Aiello, I.; Crispini, A.; Golemme, A.; La Deda, M.; Pucci, D. Coord.
Chem. ReV. 2006, 250, 1373.
10.1021/om701174v CCC: $40.75
2008 American Chemical Society
Publication on Web 05/31/2008

2930 Organometallics, Vol. 27, No. 13, 2008
Aguilar et al.
Scheme 1a
a
Legend: (i) Pd(OAc)2/CH2Cl2/∆; (ii) LiCl/MeOH; (iii) Tl(acac).
of NICS (nucleus-independent chemical shift) measurements.8
On the other hand, the CsH bond activation produced at the
phenyl ring of the benzamide group is also worth noting, due
to the deactivating nature of the carbonyl substituent. A plausible
explanation has been given, taking into account the resonant
forms of the free ligand, since the presence of a negative charge
delocalized around the NCO system could produce a more
charged phenyl ring, in comparison with those at the PPh3
phosphonium unit, and then a more favorable electrophilic
substitution at the former.5
Figure 1. Thermal ellipsoid plot of complex 2b. Non-hydrogen
atoms are drawn at the 50% probability level.
Aiming to obtain more insight about this very selective
metalation, we have performed a detailed study of the ortho
palladation of different stabilized iminophosphoranes. More
specifically we have (a) changed the substituents at the phos-
phine group, (b) replaced the phosphine by a diphosphine ligand,
and/or (c) introduced functional groups other than the aryl unit
at the carbonyl moiety. Moreover, in order to shed light about
on the ultimate reasons for the aforementioned selectivity, we
have studied the mechanism of the ortho palladation of the
model iminophosphorane PhH2PdNC(O)C6H5, induced by
Pd(OAc)2, through DFT methods.
Figure 2. Thermal ellipsoid plot of complex 3a. Non-hydrogen
atoms are drawn at the 50% probability level.
Results and Discussion
as single isomers (presumably trans), while 2c was characterized
as a 4/1 mixture of trans and cis isomers. The acetylacetonate
derivatives 3a-c have been prepared from reaction of 2a-c
with Tl(acac) (1:2 molar ratio) in order to increase the solubility
of the compounds, to avoid the presence of isomers, and to
properly characterize the metalated fragments. The spectroscopic
data of 2a-c and 3a-c show unequivocally that the metalation
has been produced regioselectively at the benzamide ring, giving
exo derivatives. The 31P NMR spectra show peaks in the 28
(2a,b and 3a,b)-36 (2c and 3c) ppm region: that is, showing a
moderate downfield shift of the resonances, as expected for exo
complexes.5
Ortho Palladation of R3PdNC(O)Ph and Ph3PdNC-
(O)CH2Ph. The starting iminophosphoranes R3PdNC(O)Ph
(R3P ) P(p-tol)3 (1a), P(m-tol)3 (1b), PPh2Me (1c)) have been
prepared by following procedures reported by Pomerantz et al.,9
while Ph3PdNC(O)CH2Ph (1d) has been synthesized through
the Staudinger method,10 by reaction of PPh3 with
N3C(O)CH2Ph.11 Attempts to prepare iminophosphoranes de-
rived from P(o-tol)3 failed, probably due to the particular steric
requirements of this phosphine.
The reaction of 1a-c with Pd(OAc)2 (1:1 molar ratio) in
refluxing CH2Cl2 and subsequent treatment of the acetate
intermediate (not isolated) with excess LiCl in MeOH at room
temperature give the dinuclear ortho-metalated derivatives 2a-c
as yellow solids (Scheme 1). Using different solvents or
increasing the temperature of the reaction does not result in the
formation of the endo isomer. The reaction of Pd(OAc)2 with
1a-c in refluxing toluene occurs with extensive decomposition,
and only black Pd0 was obtained. Complexes 2a,b were obtained
The X-ray crystal structures of 2b and 3a have been
determined. Figure 1 shows a molecular drawing of the dimeric
compound 2b, while that of 3a is represented in Figure 2.
Selected bond distances and angles are collected in Tables 1
(2b) and 2 (3a), and parameters concerning data solution and
refinement are given in Table 3. The structure of 2b shows a
dinuclear complex, with two metalated fragments [Pd{C6H4-
(C(O)NdP(m-tol)3)-2}] bridged by two chlorine ligands, which
is structurally analogous to examples previously reported.5 The
Pd1-C1 bond distance [1.979(2) Å] is slightly longer than those
found in [Pd(µ-Cl){C6H3(Me-3)(C(O)NdPPh3-2)}]2 (1.962(3)
Å)5 and [Pd(C6H4(PPh2dNC6H4Me-4′)-2)(µ-OAc)]2 (1.964(3)
(8) Milci, M. K.; Ostojic, B. D.; Zaric, S. D. Inorg. Chem. 2007, 46,
7109.
(9) Bittner, S.; Assaf, Y.; Krief, P.; Pomerantz, M.; Ziemmnicka, B. T.;
Smith, C. G. J. Org. Chem. 1985, 50, 1712.
(10) Staudinger, H.; Meyer, J. J. HelV. Chim. Acta 1919, 2, 635.
(11) Bra¨se, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew. Chem.,
Int. Ed. 2005, 44, 5188.

Ortho Palladation of Stabilized Iminophosphoranes
Organometallics, Vol. 27, No. 13, 2008 2931
Table 1. Selected Bond Distances (Å) and Angles (deg) for
Compound 2b
expected activation does not seem to be enough to change the
selectivity on the orientation of the reaction, since exo metalated
compounds are also obtained. In our hands, the reasons for the
observed behavior of 1a-c could be the same as those invoked
for the exo palladation of Ph3PdNC(O)Aryl.5
Pd1-C1
Pd1-Cl1
P1-N1
N1-C7
C1-C2
C2-C3
C4-C5
1.979(2)
2.3233(7)
1.644(2)
1.390(3)
1.387(3)
1.385(4)
1.384(4)
Pd1-N1
Pd1-Cl1A
O1-C7
C6-C7
C1-C6
C3-C4
C5-C6
2.085(2)
2.4504(8)
1.222(3)
1.478(3)
1.392(3)
1.382(4)
1.387(3)
More examples have been evaluated. Ph3PdNC(O)CH2Ph
(1d) contains a CH2 spacer between the amide function and
the phenyl ring. In this way, the transmission of electronic effects
from the amide group to the aryl ring becomes difficult and the
size of the plausible palladacycles also becomes modified. The
reaction of Pd(OAc)2 with 1d affords trans-[PdCl2{N(PPh3)-
C(O)CH2Ph}2] (4) (Scheme 2). However, the reaction of
Li2[PdCl4]sanother classical metalating reagent17swith 1d (1:1
molar ratio) in MeOH/CH2Cl2 affords the cyclopalladated [Pd(µ-
Cl){C6H4(CH2C(O)NdPPh3-κC,N)-2}]2 (2d) via activation of
the ortho CH bond of the phenylacetamide group (Scheme 2).
The NMR characterization of 2d shows that this complex is
obtained as a mixture of cis and trans geometric isomers in a
1/7 molar ratio. The 1H NMR spectrum shows peaks that belong
to the PPh3 fragment, as well as three signals (1/1/2 intensity)
that may be assigned to the four protons of the palladated
Pd(C6H4CH2) phenylacetamide group. Moreover, an AB spin
system is observed for the diastereotopic CH2 protons. The shape
of these signals indicates that the molecular plane does not
behave in 2d as a symmetry plane. A detailed explanation should
take into account that (i) a six-membered palladacycle is formed,
(ii) the palladacycle must be notably warped, and (iii) it should
be conformationally stable on the NMR time scale.3j,18 The
metalation of the benzyl ring in 1d to give 2d occurs easily
under the usual ortho-palladation conditions. Complex 2d
displays a somewhat unusual six-membered-ring arrangement.3f
This metalation is preferred over that involving the phenyl rings
of the PPh3 unit, which, in principle, should be a more likely
option due to (i) the formation of a more stable five-membered
ring, (ii) the endo effect, and (iii) the involvement of the same
type of C(sp2)-H bonds. These observations suggest that the
metalation of a given ring comes from a delicate balance of
several factorssendo effect, type of C(spn)-H bond, size of
metallacycle, type, position, and nature of substituentssand that
the consideration of only one of them is not a valid criterion.
C1-Pd1-N1
N1-Pd1-Cl1A
Pd1-Cl1-Pd1A
C7-N1-Pd1
O1-C7-N1
81.19(9)
102.79(6)
96.64(3)
113.22(15)
123.7(2)
112.3(2)
C1-Pd1-Cl1
Cl1-Pd1-Cl1A
C7-N1-P1
P1-N1-Pd1
O1-C7-C6
93.22(8)
83.36(3)
115.83(17)
129.49(11)
124.0(2)
N1-C7-C6
Table 2. Selected Bond Distances (Å) and Angles (deg) for
Compound 3a
Pd(1)-C(7)
Pd(1)-N(1)
N(1)-C(1)
C(1)-O(1)
C(2)-C(3)
C(3)-C(4)
C(5)-C(6)
C(8)-C(9)
C(9)-C(10)
C(11)-O(3)
1.961(2)
2.0695(16)
1.392(3)
1.227(3)
1.393(3)
1.384(3)
1.395(3)
1.502(3)
1.392(3)
1.262(3)
Pd(1)-O(2)
Pd(1)-O(3)
N(1)-P(1)
C(1)-C(2)
C(2)-C(7)
C(4)-C(5)
C(6)-C(7)
O(2)-C(9)
C(10)-C(11)
C(11)-C(12)
2.0304(15)
2.0787(15)
1.6457(18)
1.482(3)
1.400(3)
1.394(3)
1.390(3)
1.279(3)
1.398(3)
1.502(3)
C(7)-Pd(1)-O(2)
O(2)-Pd(1)-O(3)
C(1)-N(1)-P(1)
P(1)-N(1)-Pd(1)
O(1)-C(1)-C(2)
91.69(7)
91.16(6)
117.62(14)
125.16(9)
123.82(19)
C(7)-Pd(1)-N(1)
N(1)-Pd(1)-O(3)
C(1)-N(1)-Pd(1)
O(1)-C(1)-N(1)
N(1)-C(1)-C(2)
81.43(8)
95.71(6)
112.04(13)
124.7(2)
111.47(17)
and 1.959(3) Å),12 while it is statistically identical with that
found in [PdCl(C6H4(PPh2dNC(O)-2-py)-2)] (1.976(3) Å).4a In
the same way, the Pd1-N1 bond distance (2.085(2) Å) is also
longer than those found in the aforementioned complexes (range
1.997(2)-2.055(2) Å]).4a,5,12 The two Pd1-Cl1 bond distances
(2.3233(7) and 2.4504(8) Å) are different, indicating the
different trans influences of the Caryl and the iminic N atom,13
but fall in the usual range of distances found for these types of
complexes.14 The P1sN1 bond distance (1.644(2) Å) is longer
than that found in Ph3PdNC(O)Ph (1.626(3) Å),15 most
probably due to the N bonding to the Pd center. A similar
discussion can be carried out about internal parameters of the
palladated ligand in [Pd{C6H4(C(O)NdP(p-tol)3)-2}] (3a;
Pd1-C7 ) 1.961(2) Å; Pd1-N1 ) 2.0695(16) Å, and P1-N1
) 1.6457(18) Å). The Pd1-O2 (2.0304(15) Å) and Pd1-O3
(2.0787(15) Å) bond distances fall in the usual range of distances
found for chelating acac ligands.14
The palladation of 1a-c has been produced regioselectively
at the benzamide ring, as was observed in Ph3PdNC(O)Aryl,5
suggesting that the formation of exo isomers in this type of
substrates could be a general process. The presence of methyl
groups in the tolyl rings of the phosphine (1a,b) or, better,
directly bonded to the phosphonium P atom (1c) should activate
these rings toward an electrophilic substitution.16 However, the
Ortho Palladation of Ph2P(CH2)nPPh2dNC(O)Ph (n )
1, 2). The functionalized iminophosphoranes Ph2P(CH2)n-
PPh2dNC(O)Ph (n ) 1 (1e), 2 (1f)), derived from the
bis-phosphines dppm (n ) 1) and dppe (n ) 2), have been
obtained by reaction of PhC(O)N3 with the corresponding bis-
phosphine. The reaction was performed by careful dropwise
addition of a solution of the azide to a solution of the
bis-phosphine, ensuring an excess of phosphine with respect to
the azide in the reaction medium during the addition to favor
the formation of the monosubstituted derivative. Even with these
precautions, some bis-iminophosphorane (CH2)n[Ph2PdNC(O)-
Ph]2 (n ) 1, 2) was detected in the respective crude products
(almost 10% by NMR in the best cases). Fortunately, 1e,f could
be obtained as pure crystalline materials by column chromato-
graphic purification of the crude products, using Et2O/hexane
mixtures as eluents (9/1 for 1e and 7/3 for 1f). The 31P{1H}
NMR spectra show the presence of two doublets, one at about
22-25 ppm, assigned to the P nuclei of the PdN functional
group, and the other one at very high field (from -12 to -25
(12) Vicente, J.; Abad, J. A.; Clemente, R.; Lo´pez-Serrano, J.; Ram´ırez
de Arellano, M. C.; Jones, P. G.; Bautista, D. Organometallics 2003, 22,
4248.
(13) (a) Coe, B.; Clenwright, S. J. Coord. Chem. ReV. 2000, 203, 5. (b)
Zanini, M. L.; Meneghetti, M. R.; Ebeling, G.; Livotto, P. R.; Rominger,
F.; Dupont, J. Inorg. Chim. Acta 2003, 350, 527.
(14) Orpen, A. G.; Brammer, L.; Allen, F. H.; Kennard, O.; Watson,
D. G.; Taylor, R. J. Chem. Soc., Dalton Trans. 1989, S1.
(15) Bar, I.; Bernstein, J. Acta Crystallogr., Sect. B 1980, 36, 1962.
(16) (a) Ryabov, A. D. Chem. ReV. 1990, 90, 403. (b) Canty, A. J.; van
Koten, G. Acc. Chem. Res. 1995, 28, 406.
(17) Cope, A. C.; Friedrich, E. C. J. Am. Chem. Soc. 1968, 90, 909.
(18) Hiraki, K.; Fuchita, Y.; Takechi, K. Inorg. Chem. 1981, 20, 4316.

2932 Organometallics, Vol. 27, No. 13, 2008
Aguilar et al.
Table 3. Crystal Data and Structure Refinement Details for Compounds 2b, 2g · CHCl3, and 3a
2b
2g · CHCl3
3a
empirical formula
fw
C56H50Cl2N2O2P2Pd2
1128.62
C47H45Cl5N4O2P2Pd2
1149.86
C33H32NO3PPd
627.97
temp (K)
radiation (λ, Å)
cryst syst
space group
a (Å)
293(1)
293(1)
123(1)
Mo KR (0.710 73)
monoclinic
P21/n
10.074(2)
11.086(2)
Mo KR (0.710 73)
Mo KR (0.710 73)
monoclinic
P21/c
10.62824(12)
17.8907(2)
15.85581(14)
triclinic
j
P1
10.6915(6)
10.9268(6)
20.5707(12)
101.123(1)
91.864(1)
96.757(1)
2337.9(2)
2
b (Å)
c (Å)
21.677(4)
R (deg)
ꢀ (deg)
99.54(3)
105.5717(10)
γ (deg)
V (Å3)
2387.5(8)
2
1.570
0.978
2904.25(5)
4
1.436
Z
Dcalcd (Mg/m3)
1.633
1.167
µ (mm-1
)
0.728
cryst size (mm3)
0.19 × 0.07 × 0.02
0.18 × 0.14 × 0.08
0.38 × 0.33 × 0.25
no. of rflns collected
no. of indep rflns
no. of data/restraints/params
goodness of fit on F2
final R indices (I > 2σ(I))
20 904
18 164
52 246
5221 (Rint ) 0.0222)
5221/0/301
1.080
R1 ) 0.0262
wR2 ) 0.0748
R1 ) 0.0325,
wR2 ) 0.0760
0.611, -0.291
8192 (Rint ) 0.0829)
8192/0/559
1.019
R1 ) 0.0565
wR2 ) 0.1566
R1 ) 0.0617
wR2 ) 0.1603
1.807, -1.032
8428 (Rint ) 0.0226)
8428/0/357
1.043
R1 ) 0.0359
wR2 ) 0.0855
R1 ) 0.0443
wR2 ) 0.0894
0.855, -0.515
R indices (all data)
largest diff peak, hole (e Å-3
)
Scheme 2a
seems that, if a Ph ring is linked to the NC(O) or NC(O)CH2
groups, the metalation is regioselectively oriented to this ring,
giving exo derivatives. These results prompted us to investigate
other types of substrates.
Ortho Palladation of Ph3PdNC(O)FG (C(sp2)/C(sp3)).
Two types of starting iminophosphoranes have been synthesized,
by modification of the functional group linked to the C(O) group.
One of them contains functional groups like N-pyrrolidine
(NC4H8) (1g) or N-morpholine (NC4H8O) (1h) (eq 2). These
are substrates in which a plausible metalation at the PPh3 unit
(C(sp2)-H bond) would compete with that at the new hetero-
cyclic unit (C(sp3)-H bond). In the two cases five-membered
metallacycles would be obtained. The other one contains
functional groups such as 2-furan (1i) or 2-thiophene (1j), for
which a C(sp2)-H bond activation has been described previ-
ously.21 Unfortunately, the palladation of 1i,j was not successful,
in spite of many modifications of the reaction conditions and/
or the Pd source. In light of these results, 1i,j were not further
investigated.
a
Legend: (i) Pd(OAc)2/CH2Cl2/∆; (ii) LiCl/MeOH; (iii) Li2[PdCl4]/
MeOH; (iv) Tl(acac).
ppm), assigned to the free PPh2 moiety. These values are similar
to those reported previously.19
Compounds 1e,f react cleanly with Pd(OAc)2 (1:1 molar ratio)
and with an excess of LiCl (MeOH) to give the mononuclear
species [Pd(Cl){C6H4(C(O)NdPPh2(CH2)nPPh2-κC,N,P)-2}] (n
) 1 (2e), 2 (2f)) (eq 1). As expected, one of the ortho CsH
bonds of the benzamide ring has been palladated and, in addition
to the N bonding, the P coordination of the terminal PPh2 moiety
has taken place. In these compounds, this type of metalation
was expected, since the competitive C-H bond activation at
the phenyl rings of the NdPPh2 group is very unlikely, due to
steric constraints and the bonding of the terminal PPh2 group.
The trans arrangement of the P atom of the terminal PPh2 unit
and the palladated carbon atom is worth noting, evident from
2
the observation of a large JPC coupling constant (about 140
Hz), since it should be very unfavorable due to the transphobia
between these two groups.20 In light of the preceding results it
(19) (a) Katti, K. V.; Santarsiero, B. D.; Pinkerton, A. A.; Cavell, R. G.
Inorg. Chem. 1993, 32, 5919. (b) Cadierno, V.; Crochet, P.; D´ıez, J.; Garc´ıa-
Alvarez, J.; Garc´ıa-Garrido, S. E.; Garc´ıa-Granda, S.; Gimeno, J.; Rodr´ıguez,
M. A. Dalton Trans. 2003, 3240. (c) Boubekeur, L.; Ricard, L.; Me´zailles,
N.; Le Floch, P. Organometallics 2005, 24, 1065.
The synthesis of 1g-j has been carried out through the
Staudinger method, by reaction of the corresponding azide with

Ortho Palladation of Stabilized Iminophosphoranes
Organometallics, Vol. 27, No. 13, 2008 2933
Table 4. Selected Bond Distances (Å) and Angles (deg) for
Compound 2g · CHCl3
Pd(2)-Cl(2A)
Pd(2)-Cl(2)
Pd(2)-C(24)
Pd(2)-N(3)
N(3)-P(2)
P(2)-C(29)
C(29)-C(24)
C(24)-C(25)
C(25)-C(26)
C(26)-C(27)
2.3240(11)
2.4594(12)
1.963(4)
2.038(4)
1.635(4)
1.782(5)
1.420(6)
1.404(6)
1.384(7)
1.385(7)
C(27)-C(28)
C(28)-C(29)
N(3)-C(42)
C(42)-N(4)
C(42)-O(2)
N(4)-C(43)
C(43)-C(44)
C(44)-C(45)
C(45)-C(46)
C(46)-N(4)
1.380(7)
1.396(6)
1.386(6)
1.346(6)
1.234(5)
1.467(6)
1.534(7)
1.522(8)
1.511(7)
1.470(6)
Figure 3. Thermal ellipsoid plot of complex 2g. Non-hydrogen
atoms are drawn at the 50% probability level.
Pd(2)-Cl(2)-Pd(2A)
Cl(2A)-Pd(2)-Cl(2)
N(3)-Pd(2)-Cl(2)
N(3)-Pd(2)-C(24)
C(24)-Pd(2)-Cl(2A)
C(25)-C(24)-Pd(2)
Pd(2)-C(24)-C(29)
92.96(4)
87.04(4)
C(24)-C(29)-P(2) 113.8(3)
C(29)-P(2)-N(3)
99.6(2)
PPh3, and they have been characterized through the usual
methods. 1g,h react with Pd(OAc)2 (1:1 molar ratio), in refluxing
CH2Cl2, and with excess LiCl in MeOH to give the dinuclear
[Pd(µ-Cl){C6H4(PPh2dNC(O)NC4H8-κC,N)-2}]2 (2g) and [Pd(µ-
Cl){C6H4(PPh2dNC(O)NC4H8O-κC,N)-2}]2 (2h) derivatives,
respectively (eq 2). In this case, the pattern of reactivity has
been inverted, and one phenyl of the PPh3 unit has undergone
a CH bond activation process. While the presence of substituents
on the phosphine moiety (1a-c) or the benzamide ring,5 the
size of the final palladacycle (1d), and the type of phosphine
(1e,f) do not have a decisive influence over the outcome of the
reaction, a change in the nature of the CH bond to be activated
(C(sp2) vs C(sp3)) seems crucial to decide the final orientation
of the palladation. The structures of 2g,h (endo) can be proposed
on the basis of their NMR data and the X-ray structure of
complex 2g. The 31P{1H} NMR spectrum of 2g shows a single
peak at 42.53 ppm, while that of 2h shows two peaks at 45.42
and 46.92 ppm. These chemical shifts are strongly shifted to
low field with respect to those found for exo derivatives (range
29-32 ppm)5 and are in good agreement with previous data
reported for endo metalations.4a In addition, the number of peaks
suggests that while 2g is obtained as a single isomer, 2h is
probably a mixture of cis and trans geometric isomers.
The X-ray crystal structure of 2g has been determined. A
molecular drawing of 2g is shown in Figure 3, selected bond
distances and angles are collected in Table 4, and relevant
parameters concerning data collection and structure solution are
given in Table 3. There are two independent half-molecules in
the asymmetric part of the unit cell, which are structurally
analogous. Each independent molecule is a dinuclear complex,
in which two units [Pd{C6H4(PPh2dNC(O)NC4H8)-2}] are
bridged by two chlorine atoms. Within the dinuclear complex,
each Pd atom is located in a distorted-square-planar environ-
ment, surrounded by the palladated carbon atom, the iminic N
atom, and the two chlorine atoms. These facts show unambigu-
ously the endo metalation of 1g. In spite of the different ortho-
palladation positionswith respect to the exo complexes 2b and
3asthe Pd2-C24 bond distance (1.963(4) Å) is identical, within
experimental error, with those found in these complexes
(1.979(2) and 1.961(2) Å). The P2-N3 bond distance (1.6457(18)
Å) is also identical (1.635(4) Å (2g); 1.644(2) Å (2b)). Only
the Pd2-N3 bond distance (2.038(4) Å) appears to be slightly
shorter than those found in 2b and 3a (Tables 1 and 2).
Theoretical Study of the Reaction Mechanism. In order to
shed light on the regioselective metalation of iminophosphoranes
95.27(11) P(2)-N(3)-Pd(2)
109.77(19)
84.09(16) Pd(2)-N(3)-C(42) 126.7(3)
93.38(13) N(3)-C(42)-O(2)
122.6(4)
115.4(4)
118.0(3)
126.2(3)
115.7(3)
N(3)-C(42)-N(4)
C(42)-N(3)-P(2)
1a-f, we have performed theoretical calculations of the
mechanism of the palladation reaction using simplified model
complexes. Therefore, all stabilized iminophosphoranes
R3PdNC(O)Aryl have been represented by the ligand
PhH2PdNC(O)C6H5. The complete exploration of the potential
energy surface was performed using quantum mechanics (QM)
calculations (DFT at the B3LYP level).22,23 Test calculations
using the quantum mechanics/molecular mechanics (QM/MM)
methodology ONIOM (B3LYP/UFF) were also performed using
the actual ligand Ph3PdNC(O)Ph to check the accuracy of the
model (Supporting Information). Solvent effects were taken into
account by means of a continuum description of the solvent
with the CPCM method.24 The mechanism of the cyclopalla-
dation of amines has been studied by DFT methods.25 We have
successfully used this mechanism to explain the behavior of
nonstabilized iminophosphoranes R3PdNCH2Aryl,6 and we
have assumed here an identical pathway. It includes a six-
membered transition state, analogous to other structures previ-
ously proposed on the basis of experimental results16,26 and also
similar to other TS described for related systems.27
The energy profile for the exo palladation of the model ligand
PhH2PdNC(O)C6H5 (Figure 4) shows the same steps as those
described previously.6,25a After the initial N bonding of the
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonza´lez, C.; Pople, J. A. Gaussian 03,
Revision D.01; Gaussian, Inc., Wallingford, CT, 2004.
(23) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (c) Stephens, P. J.; Delvin,
F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623.
(24) (a) Tomasi, J.; Persico, M. Chem. ReV. 1994, 94, 2027. (b) Amovilli,
C.; Barone, V.; Cammi, R.; Cance`s, E.; Cossi, M.; Mennucci, B.; Pomelli,
C. S.; Tomasi, J. AdV. Quantum Chem. 1998, 32, 227. (c) Tomasi, J.;
Mennucci, B.; Cammi, R. Chem. ReV. 2005, 105, 2999.
(20) (a) Pearson, R. G. Inorg. Chem. 1973, 12, 712. (b) Vicente, J.;
Arcas, A.; Bautista, D.; Jones, P. G. Organometallics 1997, 16, 2127. (c)
Vicente, J.; Abad, J. A.; Frankland, A. D.; Ram´ırez de Arellano, M. C.
Chem. Eur. J. 1999, 5, 3066. (d) Vicente, J.; Abad, J. A.; Mart´ınez-Viviente,
E.; Jones, P. G. Organometallics 2002, 21, 4454.
(25) (a) Davies, D. L.; Donald, S. M. A.; Macgregor, S. A. J. Am. Chem.
Soc. 2005, 127, 13754. (b) Davies, D. L.; Donald, S. M. A.; Al-Duaij, O.;
Macgregor, S. A.; Po¨lleth, M. J. Am. Chem. Soc. 2006, 128, 4210. (c)
Ziatdinov, V. R.; Oxgaard, J.; Mironov, O. A.; Young, K. J. H.; Goddard,
W. A., III; Periana, R. A. J. Am. Chem. Soc. 2006, 128, 7404.
(21) (a) Anderson, C.; Crespo, M. J. Organomet. Chem. 2004, 689, 1496.
(b) Anderson, C.; Crespo, M.; Font-Bard´ıa, M.; Klein, A.; Solans, X. J.
Organomet. Chem. 2000, 601, 22.

2934 Organometallics, Vol. 27, No. 13, 2008
Aguilar et al.
Scheme 3. Decomposition of the Activation Barrier (AB)
(CH2Cl2 or toluene). The second, more interesting, fact is that
the activation barrier, determined by the ∆Eq value (Table 5),
shows clearly that the reaction is kinetically controlled, since
the values for the exo palladation are always substantially lower
than those calculated for the endo process. The values of ∆Eq
in CH2Cl2 (20-22 kcal/mol) are similar to those previously
found in related systems6 and seem to be easily reachable. In
fact, the reactions take place in refluxing CH2Cl2. However, the
values calculated for the endo pathway are notably higher (up
to 7 kcal/mol) and seem to be unreachable. All attempts at ortho
palladation in refluxing toluene were unsuccessful, and only
decomposition was observed in all cases. The good agreement
between the experimental facts and the calculated values (∆Eq
and ∆Er) for the mechanism described here allow us to propose
that the kinetic control of the reaction is the main reason for
the regioselective ortho palladation of stabilized iminophos-
phoranes.
Figure 4. Computed reaction profiles (kcal/mol) for the exo and
endo C-H bond activation of 1a in CH2Cl2 as solvent. The zero
energy point has been taken at the [Pd(κ2-OAc)(κ1-OAc)(κ1-N-
ligand)] species.
We have also performed calculations considering the complete
model. The palladation of Ph3PdNC(O)Ph with Pd(OAc)2 has
been studied using the QM/MM methodology. Complete details
are given in the Supporting Information. The calculated gas-
phase barriers for the model and real ligands differ by less than
1 kcal/mol. The QM/MM study on the real system provides
the same conclusions as those described for the simplified
model: (i) the reaction is kinetically controlled and favors the
exo metalation; (ii) the endo products are thermodynamically
more stable. Therefore, we will refer in the following paragraphs
to the simplified model.
Figure 5. Computed structures for the TS2 of exo (left) and endo
(right) palladations of PhH2P ) NC(O)C6H5. Distances in Å.
Table 5. Activation Barriers (∆Eq) and Reaction Energies (∆Er) for
the Exo and Endo Reaction Pathways in the Gas Phase and in
Different Solventsa
gas
dcm
tol
exo
endo
exo
endo
exo
endo
∆Eq
∆Er
24.0
-7.9
32.8
-11.5
21.9
-10.4
29.1
-12.1
22.2
-10.4
36.6
-13.3
The small differences obseved in the stabilization energies
(∆Er) between the endo and the exo metalations could be
attributed to the endo effect. This effect may be responsible for
the higher stabilization of the endo metalated derivatives with
respect to the exo complexes.7,8 However, the ultimate reasons
that could account for the ∆Eq values at the TS2 are not so
straightforward and, since this seems to be the key step of the
cyclopalladation reaction, we have attempted an analysis of the
different components of the energy barrier. Similar analyses have
been recently reported in discussions of the reactivity of the Pd
center toward σ(C-H) and σ(C-X) bonds.28 The formation of
the TS2-exo and TS-endo states could be attained from the
interaction of the Pd(OAc)2 moiety and the PhH2PdNC(O)C6H5
ligand when they are infinitely apart (Scheme 3), and this is
the activation barrier of the process (AB in Scheme 3). This
energy can be decomposed into three energies, two of them
accounting for the distortions of the ligand PhH2PdNC(O)C6H5
and the metallic fragment Pd(OAc)2 from their isolated geom-
etries to their distorted geometries in TS2 (∆EqdistL for the ligand
a Energies are given in kcal/mol.
iminophosphorane (R_exo, Supporting Information) an agostic
intermediate (I_exo) is reached through a first transition state
(TS1_exo). The actual activation barrier ∆Eq is that defined by
the transition state TS2 (Figure 5, Table 5), which accounts for
the proton migration from the aryl carbon to one oxygen atom
of an acetate ligand. The ortho-palladated product P_exo
(Supporting Information) is obtained after complete proton
transfer and stabilization through the formation of an hydrogen
bond. The energy profile of the endo palladation of
PhH2PdNC(O)C6H5 is also shown in Figure 4. In this case only
the TS2 is found (Figure 4, Table 5), in spite of intense research
on the region where the plausible agostic intermediate is thought
to be found, and thus the energy profile is simpler than that
observed for the exo metalation.
The values shown in Table 5 (those in Figure 4 are calculated
in CH2Cl2 as solvent) imply two significant facts. The first one
is that the endo palladation is thermodynamically favored (∆Er)
over the exo process, either in the gas phase or in a solvent
and ∆Eq
for the Pd fragment) and the other one being the
distM
interaction energy of the distorted fragments in the transition
state (∆EqintML). The values are collected in Table 6.
These values show that the distortion energies are quite
different in the two cases: 54.0 kcal/mol for the endo process
and 63.5 kcal/mol for the exo reaction. The endo palladation
(26) (a) Ryabov, A. D.; Sakodinskaya, A. K.; Yatsimirsky, A. K.
J. Chem. Soc., Dalton Trans. 1985, 2629. (b) Kurzeev, S. A.; Kazankov,
G. M.; Ryabov, A. D. Inorg. Chim. Acta 2002, 340, 192.
(27) (a) Garc´ıa-Cuadrado, D.; Braga, A. A. C.; Maseras, F.; Echavarren,
A. M. J. Am. Chem. Soc. 2006, 128, 1066. (b) Lafrance, M.; Fagnou, K.
J. Am. Chem. Soc. 2006, 128, 16496. (c) Lafrance, M.; Rowley, C. N.;
Woo, T. K.; Fagnou, K. J. Am. Chem. Soc. 2006, 128, 8754.
(28) (a) Bickelhaupt, F. M. Comput. Chem. 1999, 20, 114. (b) Bickel-
haupt, F. M.; Diefenbach, A. J. Phys. Chem. A 2004, 108, 8460. (c)
Pierrefixe, S. C. A. H.; Bickelhaupt, F. M. Chem. Eur. J. 2007, 13, 6321.

Ortho Palladation of Stabilized Iminophosphoranes
Organometallics, Vol. 27, No. 13, 2008 2935
Table 6. Decomposition of the Solvent-Phase Energy Barriers
Calculated Following Scheme 3a
be carried out with maximum caution. They must be stored at low
temperature (T ≈ 0 °C) and dissolved in an inert solvent.11
General Methods. General methods are as reported elsewhere.4–6
1a-c were prepared following reported procedures by Pomerantz
et al.9 1d-I were obtained through the Staudinger method (see
the Supporting Information).10 The azides N3C(O)CH2C6H5
N3C(O)C6H5, N3C(O)NC4H8, N3C(O)NC4H8O, N3C(O)-2-C4H3O,
and N3C(O)-2-C4H3S were synthesized according to published
methods.11
∆Eq
∆Eq
∆Eq
distL
distM
intML
endo
exo
38.6
36.0
15.4
27.5
-24.9
-41.6
a The solvent is CH2Cl2. Energies are given in kcal/mol.
implies a more flexible NP(H)2Ph fragment, while the exo
process implies a delocalized planar, more rigid NC(O)Ph
moiety. Nevertheless, endo palladation should be favored, since
less energy would be implied. Where is the difference? The
critical point comes from the calculated values of the interaction
energies. In fact, there is a much higher release of energy as a
consequence of the interaction of the distorted fragments in the
exo than in the endo case (-41.6 vs -24.9 kcal/mol). This large
difference (16.7 kcal/mol) counterbalances the differences in
distortion energies (9.5 kcal/mol) and gives a net balance of
7.2 kcal/mol in favor of the exo activation barrier. We believe
therefore that the ultimate reason for the regioselectivity found
is the strength of the metal-ligand interaction at the transition
state TS2, and we have demonstrated that in the exo pathway
this interaction is stronger. A careful inspection of the calculated
TS2 structures (Figure 5) provides additional information. The
exo structure shows a Pd-C bond distance (2.176 Å) shorter
than that found in the endo structure (2.214 Å), pointing again
to a stronger bond in the former case. In addition, the exo
structure shows the presence of a strong hydrogen bond (O · · · H
) 2.136 Å) between the nonbonded oxygen of the acetate ligand
cis to the iminic N atom and one ortho H atom of the C6H5
group linked to the P atom. In the endo structure there is also
a hydrogen bond between the same oxygen and the ortho proton
of the C(O)C6H5 group, but this H bond should be weaker, since
the O · · · H distance is 2.390 Å. These two factssthe different
strengths of the Pd-C interaction and the presence of H bonds
of different intensity involving one acetate and the phenyl unit
not involved in the palladationsseem to be at the origin of the
higher release of energy in the exo as compared to that in the
endo processes and, consequently, at the origin of the lower
activation barrier for the exo metalation.
[Pd(µ-Cl){C6H4(C(O)NdP(p-tol)3-KC,N)-2}]2 (2a). A solution
of Pd(OAc)2 (0.076 g, 0.342 mmol) and 1a (0.145 g, 0.342 mmol)
in CH2Cl2 (20 mL) was refluxed for 2.5 h. After this time, some
decomposition was evident. The black suspension was treated with
charcoal (15 min) and then filtered through a Celite pad. The orange
solution, which contained the acetate intermediate, was evaporated
to dryness and the residue dissolved in 20 mL of MeOH. The
resulting clear solution was treated with an excess of anhydrous
LiCl (0.206 g, 4.86 mmol), resulting in the precipitation of 2a (0.037
g, 19.2%) as a yellow solid, which was filtered, washed with MeOH
(5 mL) and Et2O (20 mL), and dried under vacuum. Anal. Calcd
for C56H50Cl2N2O2P2Pd2 (1128.2): C, 59.57; H, 4.46; N, 2.48.
Found: C, 59.27; H, 4.41; N, 1.95. IR: ν 1644 (νCdO), 1295 (νPdN
)
cm-1. 1H NMR (400 MHz, CD2Cl2): δ 2.28 (s, 9H, Me, P(p-tol)3),
6.62 (m, 1H, C6H4), 6.76 (m, 1H, C6H4), 6.86 (m, 1H, C6H4), 7.01
(m, 1H, C6H4), 7.21 (m, 6H, Hm, P(p-tol)3), 7.78 (m, 6H, Ho, P(p-
tol)3). 13C{1H} NMR (CDCl3): δ 21.75 (s, Me, P(p-tol)3), 122.31
1
(d, JPC ) 104.5 Hz, Ci, P(p-tol)3), 123.78 (s, C6H4), 127.80 (d,
3
3JPC ) 2.8 Hz, C6H4), 129.39 (d, JPC ) 13.4 Hz, Cm, P(p-tol)3),
129.54 (s, C6H4), 133.58 (s, C6H4), 133.76 (d, 2JPC ) 10.5 Hz, Co,
3
P(p-tol)3), 139.74 (d, JPC ) 14.0 Hz, C2, C6H4), 143.28 (s, C1,
C6H4), 143.44 (d, 3JPC ) 2.4 Hz, Cp, P(p-tol)3), 180.89 (d, 2JPC
)
5.4 Hz, CO). 31P{1H} NMR (CD2Cl2): δ 29.31. MS (FAB+): m/z
(%) 1130 (92) [M + H]+.
The syntheses of complexes 2b,c,e-h are similar to that
described for 2a and (see the Supporting Information).
[Pd(µ-Cl){C6H4(CH2C(O)NdPPh3-KC,N)-2}]2 (2d). To a solu-
tion of Li2[PdCl4] (0.132 g, 0.506 mmol) in a MeOH/CH2Cl2
mixture (1/10, 22 mL) was added compound 1d (0.200 g, 0.506
mmol), and the resulting mixture was stirred overnight (12 h) at
25 °C. After the reaction had taken place, the orange solution was
evaporated to a small volume (∼2 mL). Further stirring gave 2d
(0.212, 73.3%) as a yellow solid, which was filtered, washed with
MeOH (2 mL) and Et2O (20 mL), and dried under vacuum.
Complex 2d was characterized (NMR) as a mixture of trans and
cis isomers in a 7/1 molar ratio. Anal. Calcd for C52H42-
Cl2N2O2P2Pd2 (1072.1): C, 58.20; H, 3.94; N, 2.61. Found: C, 58.39;
Conclusion
The ortho metalation of stabilized iminophosphoranes
R3PdNC(O)Aryl with Pd(OAc)2 takes place regioselectively
at the aryl ring of the benzamide group, giving five-membered
exo palladacycles of high stability, through a C(sp2)-H bond
activation process. The presence of different substituents does
not alter the reaction orientation. Similar observations apply in
the formation of six-membered palladacycles or when bis-
phosphine derivatives Ph2P(CH2)nPPh2 are used. This general
trend is only inverted when a C(sp3)-H bond activation is
considered, and endo palladations are obtained instead. DFT
calculations (B3LYP level) on the mechanism of this cyclo-
palladation reaction show that the process is kinetically con-
trolled. The lower energy of the transition states of the exo
pathways are due to a stronger interaction of the aryl ring to be
palladated with the Pd center and also to the presence of stronger
H bonds between the acetate ligand trans to the palladated
position and the phenyl ring not involved in the palladation
process.
1
H, 3.84; N, 2.76. IR: ν 1629 (νCdO), 1333 (νPdN) cm-1. H NMR
(400 MHz, CDCl3): δ 5.01 (d, major, 2JHH ) 16.4 Hz, CH2), 5.12
2
(d, major, CH2), 5.30 (d, minor, JHH ) 16.6 Hz, CH2), 5.43 (d,
minor, 2JHH ) 16.6 Hz, CH2) 6.97 (d, major, 3JHH ) 7.1 Hz, C6H4),
3
7.09-7.13 (m, C6H4), 7.19 (t, major, JHH ) 7.6 Hz, C6H4),
7.32-7.39 (m, both, Hm, PPh3), 7.60 (m, both, Hp, PPh3), 7.83-7.88
(m, both, Ho, PPh3). 31P{1H} NMR (CDCl3): δ 37.54 (major), 37.87
(minor). MS (FAB+): m/z (%) 500 (30) [M/2 - Cl]+.
[Pd{C6H4(C(O)NdP(p-tol)3-KC,N)-2}(acac-O,O′)] (3a). To a
solution of complex 2a (0.045 g, 0.040 mmol) in CH2Cl2 (20 mL)
was added Tl(acac) (0.024 g, 0.080 mmol), and the resulting
suspension was stirred at 25 °C for 1 h. After the reaction had
taken place, the suspension was filtered through a Celite pad, and
the resulting solution was evaporated to dryness. Treatment of the
oily residue with cold n-hexane (15 mL) gave 3a (0.026 g, 52.5%)
as a white solid. Anal. Calcd for C33H32NO3PPd (627.7): C, 63.09;
H, 5.13; N, 2.23. Found: C, 62.83; H, 5.29; N, 2.03. IR: ν 1635
(νCdO), 1589 (νacac), 1516 (νacac), 1291 (νPdN) cm-1. 1H NMR (400
MHz, CDCl3): δ 1.07 (s, 3H, Me, acac), 1.97 (s, 3H, Me, acac),
2.39 (s, 9H, Me, P(p-tol)3), 5.02 (s, 1H, CH, acac), 7.01 (t, 1H,
Experimental Section
Caution! The organic azides are highly hazardous materials
which can explode and whose preparation and manipulation must
3
4
3JHH ) 7.3 Hz, H4, C6H4), 7.16 (td, 1H, JHH ) 7.3, JHH ) 1.1

2936 Organometallics, Vol. 27, No. 13, 2008
Hz, H5, C6H4), 7.23 (m, 6H, Hm, P(p-tol)3), 7.31 (dd, 1H, JHH
Aguilar et al.
3
)
through vibrational analysis. Solvent effects were taken into account
by means of C-PCM11 using standard options. Free energies of
solvation were calculated with toluene (ꢀ ) 2.38) and CH2Cl2 (ꢀ
) 8.93) as solvents, keeping the geometry optimized for the gas-
phase species.
4
3
7.5, JHH ) 1.6 Hz, H3, C6H4), 7.59 (d, 1H, JHH ) 7.6 Hz, H6,
C6H4), 7.88 (m, 6H, Ho, P(p-tol)3). 13C{1H} NMR (CDCl3): δ 21.69
(s, Me, P(p-tol)3), 26.39 (s, Me, acac), 27.47 (s, Me, acac), 99.17
1
(s, CH, acac), 122.94 (d, JPC ) 104.7 Hz, Ci, P(p-tol)3), 124.00
(s, C6H4), 127.37 (d, JPC ) 3.2 Hz, C6H4), 129.00 (s, C6H4), 129.12
X-ray Crystallography. Crystals of 2b,g and 3a of quality
sufficient for X-ray measurements were grown by vapor diffusion
of Et2O into CH2Cl2 (2b, 3a) or CHCl3 (2g) solutions of the crude
products at 25 °C. In each case, a single crystal was mounted at
the end of a quartz fiber in a random orientation, covered with
perfluorinated oil, and placed under a cold stream of N2 gas. Data
collections were performed on Bruker Smart Apex CCD and Oxford
Diffraction Xcalibur2 diffractometers using graphite-monocromated
Mo KR radiation (λ ) 0.710 73 Å). In all cases, a hemisphere of
data was collected on the basis of ω-scan and φ-scan runs. The
diffraction frames were integrated using the programs SAINT31
and CrysAlis RED,32 and the integrated intensities were corrected
for absorption with SADABS.33 The structures were solved and
developed by Fourier methods.34 All non-hydrogen atoms were
refined with anisotropic displacement parameters. The H atoms were
placed at idealized positions and treated as riding atoms. Each H
atom was assigned an isotropic displacement parameter equal to
1.2 times the equivalent isotropic displacement parameter of its
3
(d, JPC ) 13.4 Hz, Cm, P(p-tol)3), 130.10 (s, C6H4), 133.52 (d,
2JPC ) 10.6 Hz, Co, P(p-tol)3), 140.55 (d, 3JPC ) 13.6 Hz, C2, C6H4),
143.08 (d, 4JPC ) 2.9 Hz, Cp, P(p-tol)3), 146.50 (s, C1, C6H4), 181.06
(d, 2JPC ) 5.3 Hz, CO), 186.40 (s, CO, acac), 186.27 (s, CO, acac).
31P{1H} NMR (CDCl3): δ 28.04. MS (FAB+): m/z (%) 628 (30)
[M]+.
The syntheses of complexes 3b-d are similar to that described
for 3a (see the Supporting Information).
trans-[PdCl2{N(PPh3)C(O)CH2Ph}2] (4). To a solution of 1d
(0.178 g, 0.452 mmol) in CH2Cl2 (10 mL) was added Pd(OAc)2
(0.101 g, 0.452 mmol), and the resulting mixture was refluxed for
2 h. The resulting orange yellowish solution was evaporated to
dryness, the yellow residue was dissolved in MeOH (15 mL), and
this solution was treated with LiCl (0.153 g, 3.610 mmol). Further
stirring at 25 °C gave 4 (0.098 g, 44.7%) as a yellow solid. Anal.
Calcd for C52H44Cl2N2O2P2Pd (967.7): C, 64.48; H, 4.58; N, 2.89.
Found: C, 64.38; H, 4.69; N 2.85. IR: ν 1621 (νCdO), 1339 (νPdN
)
2
cm-1. 1H NMR (400 MHz, CDCl3): δ 3.82 (s, 2H, CH2), 6.90 (d,
parent atom. The structures were refined to Fo , and all reflections
were used in the least-squares calculations.35
3
2H, JHH ) 6.8 Hz, Ho, Ph), 7.05 (m, 1H, Hp, Ph), 7.12 (m, 2H,
Hm, Ph), 7.19-7.32 (m, 9H, Ho, Hp, PPh3), 7.79 (m, 6H, Hm, PPh3).
13C{1H} NMR (CDCl3): δ 45.97 (d, 2JPC ) 3.4 Hz, CH2), 125.98
Acknowledgment. Funding by the Ministerio de Educa-
cio´n y Ciencia (MEC) (Spain, Projects CTQ2005-01037 and
CTQ2005-09000-C02-01) is gratefully acknowledged. R.B.
(BES2003-0296) and D.A. thank the MEC (Spain) and
Diputacio´n General de Arago´n (Spain) for respective Ph.D.
research grants.
2
(s, Cp, Ph), 127.89 (s, Cm, Ph), 128.61 (d, JPC ) 13.3 Hz, Cm,
4
PPh3), 130.43 (s, Co, Ph), 131.96 (d, JPC ) 2.7 Hz, Cp, PPh3),
4
132.64 (s, broad, Co, PPh3), 136.39 (d, JPC ) 2.3 Hz, Ci, Ph),
2
183.13 (d, JPC ) 5.2 Hz, CO) (signals assigned to the Ci of the
PPh3 group were not detected). 31P{1H} NMR (CDCl3): δ 35.38
MS (FAB+): m/z (%) 967 (3) [M]+, 931 (75) [M - Cl]+.
Computational Details. Calculations were performed using the
Gaussian03 series of programs.22 The reaction mechanism of the
palladation of PhH2PdNC(O)C6H5 (simplified model for reactants
R, transition states TS, and products P of the endo and exo
pathways) was studied using density functional theory (DFT) with
the B3LYP functional.23 In all cases, effective core potentials (ECP)
were used to represent the innermost electrons of the palladium
atom.23,29 The basis set for Pd was that associated with the
pseudopotential with a standard double-ꢁ LANL2DZ contraction.22
The C, N, O, and P atoms were represented by means of the
6-31G(d) basis set, whereas the 6-31G basis set was employed for
the H atoms.30 All geometry optimizations were full, with no
restrictions. Stationary points located in the potential energy
hypersurface were characterized as true minima or transition states
Supporting Information Available: Text giving the complete
experimental section with all preparative data and figures and tables
giving complete data collection parameters for 2b, 3a, and
2g · CHCl3 and Cartesian coordinates of reagents (R), transition
states (TS), and ortho-metalated products (P) for the endo and exo
palladation of PhH2PdNC(O)Ph and Ph3PdNC(O)Ph. This material
is available free of charge via the Internet at http://pubs.acs.org.
OM701174V
(31) SAINT, version 5.0(31); Bruker Analytical X-ray Systems, Madi-
son, WI.
(32) CrysAlis RED, version 1.171.27p8; Oxford Diffraction Ltd., Oxford,
U.K., 2005.
(33) Sheldrick, G. M. SADABS, Program for Absorption and Other
Corrections; University of Go¨ttingen, Go¨ttingen, Germany, 1996.
(34) Sheldrick, G. M. SHELXS-86. Acta Crystallogr. 1990, A46, 467.
(35) Sheldrick, G. M. SHELXL-97: FORTRAN Program for the
Refinement of Crystal Structures from Diffraction Data; University of
Go¨ttingen, Go¨ttingen, Germany, 1997. Molecular graphics were done using
the following commercial package: SHELXTL-PLUS, Release 5.05/V;
Siemens Analytical X-ray Instruments, Inc., Madison, WI, 1996.
(29) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(30) (a) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Phys. Chem. 1972,
56, 2257. (b) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213. (c) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon,
M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654.