Complexation of diphenyl(phenylacetenyl)phosphine to rhodium(III) tetraphenyl porphyrins: Synthesis and structural, spectroscopic, and thermodynamic studies

Article abstract of DOI:10.1021/ic026257a

The coordination of diphenyl(phenylacetenyl)phosphine (DPAP, 1) to (X)Rh^IIITPP (X = 1 (2) or Me (3); TPP = tetraphenyl porphyrin) was studied in solution and in the solid state. The iodide is readily displaced by the phosphine, leading to the bis-phosphine complex [(DPAP)₂Rh(TPP)](I) (4). The methylide on rhodium in 3 is not displaced, leading selectively to the mono-phosphine complex (DPAP)(Me)Rh(TPP) (5). The first and second association constants, as determined by isothermal titration calorimetry and UV-vis titrations, are in the range 10⁴-10⁷ M^-1 (in CH₂Cl₂). Using LDI-TOF mass spectrometry, the mono-phosphine complexes can be detected but not the bisphosphine complexes. The electronic spectrum of 4 is similar to those previously reported with other tertiary phosphine ligands, whereas (DPAP)(I)Rh(TPP) (6) displays a low energy B-band absorption and a high energy Q-band absorption. In contrast to earlier reports, displacement of the methylide on rhodium in 5 could not be observed at any concentration, and the electronic spectra of 4 and 5 are almost identical. Isothermal titration calorimetry experiments showed that all binding events are exothermic, and all are enthalpy driven. The largest values of ΔG° are found for 6. The thermodynamic and UV-vis data reveal that the methylide and the phosphine ligand have an almost identical electronic trans-influence on the sixth ligand.

Full text of DOI:10.1021/ic026257a

Inorg. Chem. 2003, 42, 3086−3096
Complexation of Diphenyl(phenylacetenyl)phosphine to Rhodium(III)
Tetraphenyl Porphyrins: Synthesis and Structural, Spectroscopic, and
Thermodynamic Studies
,†
Eugen Stulz,* Sonya M. Scott, Andrew D. Bond, Sijbren Otto, and Jeremy K. M. Sanders*
UniVersity Chemical Laboratory, UniVersity of Cambridge, Lensfield Road,
Cambridge CB2 1EW, U.K.
Received December 11, 2002
III
The coordination of diphenyl(phenylacetenyl)phosphine (DPAP, 1) to (X)Rh TPP (X ) I (2) or Me (3); TPP )
tetraphenyl porphyrin) was studied in solution and in the solid state. The iodide is readily displaced by the phosphine,
leading to the bis-phosphine complex [(DPAP)2Rh(TPP)](I) (4). The methylide on rhodium in 3 is not displaced,
leading selectively to the mono-phosphine complex (DPAP)(Me)Rh(TPP) (5). The first and second association
constants, as determined by isothermal titration calorimetry and UV−vis titrations, are in the range 104−107 M-1 (in
CH Cl2). Using LDI-TOF mass spectrometry, the mono-phosphine complexes can be detected but not the bis-
2
phosphine complexes. The electronic spectrum of 4 is similar to those previously reported with other tertiary phosphine
ligands, whereas (DPAP)(I)Rh(TPP) (6) displays a low energy B-band absorption and a high energy Q-band
absorption. In contrast to earlier reports, displacement of the methylide on rhodium in 5 could not be observed at
any concentration, and the electronic spectra of 4 and 5 are almost identical. Isothermal titration calorimetry
experiments showed that all binding events are exothermic, and all are enthalpy driven. The largest values of ∆G°
are found for 6. The thermodynamic and UV−vis data reveal that the methylide and the phosphine ligand have an
almost identical electronic trans-influence on the sixth ligand.
Introduction
porphyrins into multiporphyrin arrays to obtain constructs
where the rhodium porphyrin could act as a redox or as a
Rhodium(III) porphyrin complexes are of interest because
of their ability to act as hydrogen activators in hydride
transfer reactions1,2 and as catalysts in hydrogen evolution.3
A variety of applications, where Rh(II) porphyrins act as
redox-activators of C-H bonds, have also been reported.4-6
In bioinorganic chemistry, rhodium(III) porphyrins have been
(6) (a) Wayland, B. B.; Sherry, A. E.; Bunn, A. G. J. Am. Chem. Soc.
1993, 115, 7675. (b) Wayland, B. B.; Ba, S.; Sherry, A. E. J. Am.
Chem. Soc. 1991, 113, 5305. (c) Zhang, X. X.; Wayland, B. B. J.
Am. Chem. Soc. 1994, 116, 7897. (d) Zhang, X. X.; Parks, G. F.;
Wayland, B. B. J. Am. Chem. Soc. 1997, 119, 7938. (e) Basickes, L.;
Bunn, A. G.; Wayland, B. B. Can. J. Chem. 2001, 79, 854. (f) Barrett,
A. G. M.; Braddock, D. C.; Lenoir, I.; Tone, H. J. Org. Chem. 2001,
66, 8260. (g) Kimata, S.; Aida, T. Tetrahedron Lett. 2001, 42, 4187.
(h) Lo Schiavo, S.; Serroni, S.; Puntoriero, F.; Tresoldi, G.; Piraino,
P. Eur. J. Inorg. Chem. 2002, 79. (i) Tagliatesta, P.; Pastorini, A. J.
Mol. Catal. A: Chem. 2002, 185, 127. (j) Mak, K. W.; Yeung, S. K.;
Chan, K. S. Organometallics 2002, 21, 2362.
7
used for biomimetic studies on vitamin B12 and on cyto-
chrome c.8 The redox chemistry and photophysical properties
of Rh porphyrins have extensively been investigated by
James,9 Kadish,4,10-13 Collman,1,8,14 Wayland,2,6,7 and
Save´ant.3,15 It thus seems promising to embed rhodium(III)
(7) Wayland, B. B.; Vanvoorhees, S. L.; Delrossi, K. J. J. Am. Chem.
Soc. 1987, 109, 6513.
(8) Collman, J. P.; Boulatov, R. J. Am. Chem. Soc. 2000, 122, 11812.
(9) Thackray, D. C.; Ariel, S.; Leung, T. W.; Menon, K.; James, B. R.;
Trotter, J. Can. J. Chem. 1986, 64, 2440.
(10) Anderson, J. E.; Yao, C.-L.; Kadish, K. M. Inorg. Chem. 1986, 25,
3224.
* To whom correspondence should be addressed. E-mail: Eugen.
Stulz@unibas.ch (E.S.); jkms@cam.ac.uk (J.K.M.S.).
† Present address: Department of Chemistry, University of Basel, St.
Johanns-Ring 19, 4056 Basel, Switzerland.
(1) Collman, J. P.; Ha, Y.; Guilard, R. Inorg. Chem. 1993, 32, 1788.
(2) Bosch, H. W.; Wayland, B. B. J. Chem. Soc., Chem. Commun. 1986,
900.
(3) Grass, V.; Lexa, D.; Save´ant, J. M. J. Am. Chem. Soc. 1997, 119,
7526.
(11) Anderson, J. E.; Yao, C.-L.; Kadish, K. M. Inorg. Chem. 1986, 25,
718.
(12) Kadish, K. M.; Araullo, C.; Yao, C.-L. Organometallics 1988, 7, 1583.
(13) Kadish, K. M.; Hu, Y.; Boschi, T.; Tagliatesta, P. Inorg. Chem. 1993,
32, 2996.
(4) Guilard, R.; Kadish, K. M. Chem. ReV. 1988, 88, 1121.
(5) Nelson, A. P.; DiMagno, S. G. J. Am. Chem. Soc. 2000, 122, 8569.
(14) Collman, J. P.; Ha, Y. Y.; Guilard, R.; Lopez, M. A. Inorg. Chem.
1993, 32, 1788.
3086 Inorganic Chemistry, Vol. 42, No. 9, 2003
10.1021/ic026257a CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/01/2003

Phosphine Complexation with Rh(III)TPP
Chart 1
Chart 2
photophysically active center.16 In previous work, we have
introduced Rh(III) porphyrins into supramolecular arrays17
and studied a variety of N-, S-, and Se-coordination com-
pounds both in solution and in the solid state,18 culminating
in the formation of a heterometallic porphyrin undecamer.19
Recently, we have used the coordination of a phosphine sub-
stituted porphyrin to RhIII(TPP) to prepare selectively a cyclic
porphyrin tetramer by amplification from a biased dynamic
combinatorial library20 using 4,4′-bipyridine as scaffold.21
Alkynyl substituted porphyrins provide versatile building
blocks for the construction of supramolecular assemblies.22
Attachment of a diphenyl phosphine group to porphyrins via
an acetylenic linker provides a simple route to phosphine
substituted porphyrins,23 which are ideal building blocks for
the construction of heterometallic porphyrin arrays. In order
to be able to predict electronic interactions in arrays such as
a [Zn/Rh/Zn] trimer (Chart 1), it is essential to have basic
knowledge about the structure and physical properties of
phosphorus rhodium porphyrin complexes. We have previ-
ously used diphenyl(phenylacetenyl)phosphine (DPAP, Chart
2) (1) to study the physical properties of phosphine com-
plexes of a ruthenium porphyrin.24,25 This phosphine ligand
serves as a model to mimic the substitution pattern in our
phosphine substituted porphyrins.
Here, we present a detailed study on the affinity of DPAP
(1) toward (X)(Y)RhIII(TPP) [X ) I, Y ) MeOH (2); X )
Me (3), Chart 2].26 The porphyrins have been chosen on the
basis that we already have demonstrated that they form bis-
phosphine complexes with 2;21 the methylide in 3 is thought
to be an inert ligand, blocking the sixth coordination site so
that mono-phosphine complexes can selectively be obtained.
All complexes have been studied in solution using 1H NMR
and 31P{1H} NMR spectroscopy, UV-vis spectroscopy, mass
spectrometry, and isothermal titration calorimetry (ITC). We
also report on the solid-state structures of 2, 3, [(DPAP)2RhIII-
(TPP)](I) (4), and (DPAP)(Me)RhIII(TPP) (5).
Experimental Section
General. Methylene chloride (CH2Cl2), chloroform (CHCl3), and
methanol (MeOH) were obtained from Bamford Laboratories (U.K.)
and used as received; CDCl3 (euriso-top, France) was filtered over
basic alumina prior to use. tBu4NI (Aldrich) was used as purchased.
DPAP (1), (MeOH)(I)Rh(TPP) (2), and (Me)Rh(TPP) (3) were
prepared according to literature procedures.27 NMR spectra were
recorded on a Bruker DPX400 NMR spectrometer at 161.98 MHz
(31P{1H}, H2PO4 external standard), or on a Bruker DRX500 NMR
spectrometer at 500.13 MHz (1H). Abbreviations for NMR spectra
used are the following: s, singlet; d, doublet; dd, doublet of
doublets; t, triplet; dt, doublet of triplets; q, quartet; m, multiplet.
UV-vis spectra were recorded on a Varian Cary 100 Bio spec-
trophotometer. LDI-TOF mass spectra were recorded on a Kompact
MALDI 4 mass spectrometer (Kratos Analytical Ltd), operated in
the linear positive mode, and using neat samples. Isothermal titration
calorimetry (ITC) was performed on a MICROCAL. INC micro-
calorimeter at 25 °C.
(15) Grass, V.; Lexa, D.; Momenteau, M.; Save´ant, J. M. J. Am. Chem.
Soc. 1997, 119, 3536.
(16) Vitols, S. E.; Friesen, D. A.; Williams, D. S.; Melamed, D.; Spiro, T.
G. J. Phys. Chem. 1996, 100, 207.
(17) Kim, H. J.; Redman, J. E.; Nakash, M.; Feeder, N.; Teat, S. J.; Sanders,
J. K. M. Inorg. Chem. 1999, 38, 5178.
(18) Redman, J. E.; Feeder, N.; Teat, S. J.; Sanders, J. K. M. Inorg. Chem.
2001, 40, 3217.
(19) Redman, J. E.; Feeder, N.; Teat, S. J.; Sanders, J. K. M. Inorg. Chem.
2001, 40, 2486.
(20) (a) Otto, S.; Furlan, R. L. E.; Sanders, J. K. M. Science 2002, 297,
590. (b) Roberts, S. L.; Furlan, R. L. E.; Cousins, G. R. L.; Sanders,
J. K. M. Chem. Commun. 2002, 938. (c) Furlan, R. L. E.; Ng, Y. F.;
Otto, S.; Sanders, J. K. M. J. Am. Chem. Soc. 2001, 123, 8876. (d)
Furlan, R. L. E.; Ng, Y. F.; Cousins, G. R. L.; Redman, J. E.; Sanders,
J. K. M. Tetrahedron 2002, 58, 771.
(21) Stulz, E.; Ng, Y.-F.; Scott, S. M.; Sanders, J. K. M. Chem. Commun.
2002, 524.
(22) (a) Mak, C. C.; Bampos, N.; Darling, S. L.; Montalti, M.; Prodi, L.;
Sanders, J. K. M. J. Org. Chem. 2001, 66, 4476. (b) Webb, S. J.;
Sanders, J. K. M. Inorg. Chem. 2000, 39, 5912. (c) Webb, S. J.;
Sanders, J. K. M. Inorg. Chem. 2000, 39, 5920. (d) Nakash, M.;
Sanders, J. K. M. J. Chem. Soc., Perkin Trans. 2 2001, 2189. (e) Mak,
C. C.; Pomeranc, D.; Montalti, M.; Prodi, L.; Sanders, J. K. M. Chem.
Commun. 1999, 1083. (f) Kim, H. J.; Bampos, N.; Sanders, J. K. M.
J. Am. Chem. Soc. 1999, 121, 8120. (g) Wilson, G. S.; Anderson, H.
L. Chem. Commun. 1999, 1539. (h) Taylor, P. N.; Anderson, H. L. J.
Am. Chem. Soc. 1999, 121, 11538. (i) Kodis, G.; Liddell, P. A.; de la
Garza, L.; Clausen, C.; Lindsey, J. L.; Moore, A. L.; Moore, T. A.;
Gust, D. J. Phys. Chem. A 2002, 106, 2036. (j) Ambroise, A.; Kirmaier,
C.; Wagner, R. W.; Loewe, R. S.; Bocian, D. F.; Holten, D.; Lindsey,
J. L. J. Org. Chem. 2002, 67, 3811. (k) Fletcher, J. T.; Therien, M. J.
Inorg. Chem. 2002, 41, 331. (l) Shediac, R.; Gray, M. H. B.; Uyeda,
H. T.; Johnson, R. C.; Hupp, J. T.; Angiolillo, P. J.; Therien, M. J. J.
Am. Chem. Soc. 2000, 122, 7017.
UV-vis titrations were performed at the following concentra-
tions: [2] ) 10-4 M ([1] ) 10-3 M), [2] ) 10-5 M ([1] ) 10-3
M), [2] ) 10-6 M ([1] ) 10-4 M); [3] ) 5 × 10-4 M ([1] ) 10-2
M), [3] ) 10-4 M ([1] ) 10-2 M), [3] ) 10-5 M ([1] ) 10-3 M),
[3] ) 10-6 M ([1] ) 10-5 M). Solutions were prepared from stock
solutions ([2] ) 10-3 M, [3] ) 10-3 M, [1] ) 10-2 M) in CHCl3.
(24) Stulz, E.; Sanders, J. K. M.; Montalti, M.; Prodi, L.; Zaccheroni, N.;
de Biani, F.; Grigiotti, E.; Zanello, P. Inorg. Chem. 2002, 41, 5269.
(25) Stulz, E.; Maue, M.; Feeder, N.; Teat, S. J.; Ng, Y. F.; Bond, A. D.;
Darling, S.; Sanders, J. K. M. Inorg. Chem. 2002, 41, 5255.
(26) On crystallization from CH2Cl2-MeOH, 2 always carries methanol
as a sixth ligand, which can be detected in the 1H NMR spectrum; in
3, no additional bound solvent could be detected. See ref 19.
(27) DPAP: Carty, A. J.; Hota, N. K.; Ng, T. W.; Patel, H. A.; O’Connor,
T. J. Can. J. Chem. 1971, 49, 2706. 2, 3: Kim, H.-J.; Bampos, N.;
Sanders, J. K. M. J. Am. Chem. Soc. 1999, 121, 8120.
(23) Darling, S. L.; Stulz, E.; Feeder, N.; Bampos, N.; Sanders, J. K. M.
New J. Chem. 2000, 24, 261.
Inorganic Chemistry, Vol. 42, No. 9, 2003 3087

Stulz et al.
Table 1. Crystallographic Data for (MeOH)(I)Rh(TPP) (2), (Me)Rh(TPP) (3), [(dpap)2Rh(TPP)](I) (4), and (dpap)(Me)Rh(TPP) (5)
2‚CHCl3
3
4‚2CHCl3
5
formula
M
T/K
radiation, λ/Å
cryst syst
space group
a/Å
b/Å
c/Å
R/deg
â/deg
γ/deg
V/Å3
C46H33Cl3IN4ORh
993.92
180(2)
Mo KR, 0.707
triclinic
P1h
9.7114(5)
11.1479(7)
21.6504(13)
80.451(3)
89.804(3)
65.549(3)
2098.6(2)
2
1.573
1.374
25.02
18266
7318
0.0673
0.0706
0.1841
C45H31N4Rh
730.65
180(2)
Mo KR, 0.707
tetragonal
I4/m
13.4741(5)
13.4741(5)
9.6462(5)
90
C86H60Cl6IN4P2Rh
1653.83
180(2)
Mo KR, 0.707
monoclinic
C2/c
25.9530(6)
11.6120(2)
23.8501(6)
90
91.116(1)
90
7186.3(3)
4
1.529
0.986
25.01
C65H46N4PRh
1016.94
180(2)
Mo KR, 0.707
triclinic
P1h
13.3949(7)
13.3979(8)
14.1945(9)
83.942(3)
86.237(3)
89.875(4)
2527.7(3)
2
90
90
1751.28(13)
2
1.386
0.526
24.68
3796
802
0.0739
0.0408
0.0753
1.15
Z
F
calcd/g cm-3
1.336
0.416
21.97
15705
µ/mm-1
max/deg
θ
total data
unique data
Rint
20541
6306
0.0408
0.0345
0.0828
1.04
6127
0.0815
0.1134
0.2873
1.10
R1 [F2 > 2σ(F2)]
wR2 (all data)
S
1.07
ITC experiments were done at the following concentrations (CHCl3,
298 K): [2] ) 2 × 10-5 M, [1] ) 4 × 10-4 M; [3] ) 10-3 M, [1]
) 10-2 M. For the ITC experiment in the presence of additional
iodide, both solutions of 1 (4 × 10-4 M) and 2 (2 × 10-5 M)
washed with methanol, and dried in vacuo to give 5 as a dark-
orange powder (15.0 mg, 14.7 mmol, 92%, calcd for C65H46N4P1-
Rh1). Crystals suitable for X-ray analysis were grown from a
concentrated CHCl3 solution (200 µL) of 5, layered with methanol
(500 µL). UV-vis (CHCl3, 298 K): λ (log ꢀ) 356 (3.62), 418
(shoulder), 444 (5.22), 516 (3.85), 556 (4.09), 597 (4.10), 620
t
1
contained 10-4 M Bu4NI. H NMR titrations of 3 with 1 were
performed in degassed (Ar saturated) CDCl3 at [3] ) 5 × 10-4 M,
using [1] ) 10-2 M.
1
(3.62). H NMR (400.13 MHz, CDCl3, 300 K): δ -6.52 ppm (d,
1
X-ray diffraction data were collected using a Nonius Kappa CCD
diffractometer. Structures were solved by direct methods using
either SHELXS-9728 or SIR-9229 and refined against all F2 data using
SHELXL-97.28 A summary of the crystallographic data is given in
Table 1.
1J ) 2.56 Hz), 4.51 [t, J ) 8.3 Hz, 4H, o-H of P(Ph)2], 6.58 [dt,
1
1
1Jo ) 9.8 Hz, Jm ) 2.0 Hz, 4H, m-H of P(Ph)2], 6.56 [dt, Jo )
8.4 Hz, 1Jm ) 1.1 Hz, 2H, p-H of P(Ph)2], 7.06 (dd, 1Jo ) 7.0 Hz,
1Jm ) 1.3 Hz, 2H, o-H of CtCsPPh), 7.19 (m, 1H, p-H of
CtCsPPh), 7.24 (m, 2H, m-H of CtCsPPh), 7.62 (m, 4H, m-H
of meso-Ph), 7.68 (m, 8H, m,p-H of meso-Ph), 7.71 (dt, 1Jo ) 7.0
[(DPAP)2Rh(TPP)](I) (4). Compound 2 (50.0 mg, 59 mmol)
was dissolved in CHCl3 (7 mL), and solid 1 (42.3 mg, 148 mmol)
was added. Once all the starting material had dissolved, the solvent
was evaporated, and the residues were dissolved in hot CHCl3 (1.5
mL). After careful layering of methanol (15 mL), the mixture was
left to crystallize overnight. The product was filtered off, washed
with methanol, and dried in vacuo. The product was obtained as
dark-orange crystals (71.4 mg, 55 mmol, 94%, calcd for C84H58I1N4P2-
Rh1). Crystals suitable for X-ray analysis were grown from a
concentrated CHCl3 solution (200 µL) of 4, layered with methanol
(500 µL). UV-vis (CHCl3, 298 K): λ (log ꢀ) 323 (4.63), 357 (4.56),
376 (shoulder), 445 (5.28), 557 (4.23), 596 (4.19). 1H NMR (500.13
MHz, CDCl3, 300 K): δ 4.12 [q, 1J ) 7.1 Hz, 8H, o-H of P(Ph)2],
1
1
Hz, Jm ) 1.5 Hz, 4H, o-H of meso-Ph), 8.19 (dt, Jo ) 6.9 Hz,
1Jm ) 1.5 Hz, 4H, o-H of meso-Ph), 8.58 (s, 8H, â-pyrrole). 31P
NMR (161.98 MHz, CDCl3, 300 K): δ -28 ppm. LDI-TOF MS:
m/z 1032.4, 1018.2, 1002.0, 745.6, 730.7, 715.7.
Results and Discussion
Synthesis and NMR Spectroscopy of (DPAP)2RhIIITPP.
Addition of 1 equiv of DPAP (1) to (MeOH)(I)Rh(TPP) (2)
in CDCl3 did not lead to clean formation of (DPAP)(I)Rh-
(TPP) (6).9 Instead, a mixture of both mono- and bis-
phosphine complexes was obtained, as judged from the
phosphorus NMR spectrum (Figure 1a). Two distinct dou-
blets at δ -9.5 ppm (1JRh,P ) 114 Hz) and at δ -10.1 ppm
(1JRh,P ) 87 Hz) were observed, whereas unbound 1 shows
a singlet at δ -32 ppm (not shown). The overall downfield
shift of the phosphorus nucleus on complexation to rhodium
can be rationalized in terms of electron donation of the ligand
to the Lewis acidic metal center upon σ-bond formation,
which outweighs the porphyrin ring current induced upfield
shift.25 The iodide is readily displaced by 1, in contrast to
the displacement of chloride by PPh3 and PEt3 on rhodium
porphyrins, which requires a large excess of phosphine.9
Addition of a second equivalent of 1 at ambient temperature
transforms the mixture within minutes to [(DPAP)2Rh(TPP)]-
(I) (4), (δ -10.1 ppm, 1JRh,P ) 87 Hz, Figure 1b). This allows
the mono- and the bis-phosphine complex to be distinguished
1
1
6.58 [dt, Jo ) 7.9 Hz, Jm ) 1.5 Hz, 8H, m-H of P(Ph)2], 6.89
(dd, 1Jo ) 8.3 Hz, 1Jm ) 1.4 Hz, 4H, o-H of CtCsPPh), 6.94 [t,
1J ) 10.8 Hz, 8H, p-H of P(Ph)2], 7.28 (m, 4H, m-H of CtCs
PPh), 7.35 (m, 2H, p-H of CtCsPPh), 7.62 (m, 12H, m,p-H of
meso-Ph), 7.67 (m, 8H, o-H of meso-Ph), 8.79 (s, 8H, â-pyrrole).
1
31P NMR (161.98 MHz, CDCl3, 300 K): δ -10.1 ppm (d, JRh,P
) 87 Hz). LDI-TOF MS: m/z 1002.0, 716.4.
(DPAP)(Me)Rh(TPP) (5). Compound 3 (12.0 mg, 16 mmol)
was dissolved in CHCl3 (4 mL), and solid 1 (5.9 mg, 20.5 mmol)
was added. Once all the starting material had dissolved (2-3 min),
the solvent was evaporated, and the residues were crystallized from
CHCl3-methanol (1.0 + 8.0 mL). The product was filtered off,
(28) Sheldrick, G. M. SHELXS-97; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(29) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
3088 Inorganic Chemistry, Vol. 42, No. 9, 2003

Phosphine Complexation with Rh(III)TPP
Figure 1. (a) 31P{1H} NMR spectrum of a 1:1 mixture of 2 and 1; (b)
31P{1H} NMR spectrum of 4; (c) part of the 1H NMR spectrum of 4. Spectra
are recorded in CDCl3 (500 MHz for 1H, 162 MHz for 31P) at 300 K, c )
10 mM.
Figure 2. (a) 1H NMR spectrum of 5; the inset shows, from top to bottom,
the gradually upfield shifted resonance of the methylide signal after addition
of 0, 0.2, 0.6, 0.9, and 1.5 equiv of 1 at [3] ) 5 mM; (b) 31P {1H} NMR
spectrum of 5. Conditions are as indicated in Figure 1.
on the basis of both the 31P{1H} NMR chemical shift and
the 1JRh,P coupling constant. The coupling constants, but not
the chemical shifts, are very close to those found in PPh3
and reveals a strong trans-influence of 1; just such a trans-
influence of phosphines was previously detected using the
Rh-C Raman frequencies in (Me)Rh(TTP) phosphine
complexes30 and agrees with our NMR results.
1
complexes of RhIII(OEP), where JRh,P ) 124 Hz (δ -2.6
ppm) and 86 Hz (δ 9.5 ppm) for the bis- and mono-phosphine
complex, respectively.9 Also, the observed 31P{1H} NMR
chemical shifts contrast with analogous ruthenium(II) por-
phyrin-DPAP complexes, where the mono-phosphine com-
plex (δ -12 ppm) and bis-phosphine complex (δ +3 ppm)
show a chemical shift difference of ∆δ ) +15 ppm.25 In
the Rh(III) binding, the bis-phosphine complexes resonate
at a slightly higher field than the mono-phosphine complexes,
whereas in the Ru(II) complexes, the order is reversed. This
can be attributed to a different trans-influence of the iodide
on rhodium compared to the strong π-acceptor properties of
the carbonyl on ruthenium.
In the presence of excess 1, the o-proton signal of the
phosphorus phenyl substituent at δ 4.49 ppm becomes
broadened and downfield shifted, which shows that fast
ligand exchange occurs. The two different ligands on the
central metal render the R- and â-sides of the porphyrin
nonequivalent, which is expressed in the splitting of the
meso-phenyl proton resonances; i.e., the o-protons appear
at δ 7.82 and 8.09 ppm. The 31P{1H} NMR of 5 is dis-
tinctively different with the resonance for the bound phos-
1
phorus at δ -29 ppm as a broadened signal, and no JRh,P
coupling can be detected (Figure 2b). The marginal shift,
the broadening, and the lack of a coupling to rhodium arises
from a dynamic exchange of unbound and bound ligand.
Solid-State Structures. Single crystals of rhodium por-
phyrins 2-5 suitable for X-ray analysis were grown from
concentrated CHCl3 solutions layered with methanol. The
molecular units are shown in Figure 3, and selected geo-
metrical data are summarized in Table 2. The crystals of 2
contain additional solvent molecules (one CHCl3 per por-
phyrin complex). The meso-phenyl substituent at C5 displays
disorder which was modeled in two orientations of equal
occupancy, with the phenyl rings constrained to be regular
hexagons and a single displacement parameter common to
all C atoms. Porphyrin 3 crystallizes in space group I4/m
with the complex positioned on a site of 4/m symmetry, such
that it displays gross D4h point symmetry. The methylide
carbon atom could be distinguished as 50% occupied on the
basis of its displacement parameter, but no other ligand could
be located on rhodium. Thus, rhodium is 5-coordinate with
the methylide carbon disordered equally above and below
the porphyrin plane.31 Porphyrin 3, together with previously
reported analogous structures,5,17,32 shows that rhodium(III)
porphyrins bearing a methylide ligand generally remain
5-coordinate. The Rh-C bond length for 3 (1.968(12) Å) is
The proton signals of the phenyl phosphorus substituents
are shifted upfield due to the proximity to the shielding region
of the aromatic porphyrin core (Figure 1c) and confirm
binding of 1 to the porphyrin central metal. Complex 4 can
easily be obtained in pure form by mixing a slight excess of
1 with 2 in CH2Cl2 or CHCl3, and crystallizing from a layered
CHCl3-methanol system.
Synthesis and NMR Spectroscopy of (DPAP)(Me)RhIII-
(TPP). (Me)RhTPP (3) selectively forms the mono-phos-
phine adduct (DPAP)(Me)Rh(TPP) (5) in solution (Figure
2a) and can be isolated in an analogous way to 4. The
resonance for the σ-bonded methylide in 5 appears as a
broadened signal at low concentrations ([5] < 10-3 M). Its
chemical shift is concentration dependent, indicating dynamic
ligand binding of 1 to 5. The limiting value at high concen-
tration ([5] > 10 mM), or upon addition of excess ligand
(0.1 mM 3, 10 mM 1), is at δ -6.52 ppm. The inset in Figure
2a shows the gradual upfield shift of the signal from the
Rh-CH3 group upon titrating 1 with 3. The final value
displays a significant upfield shift compared to that of the
parent (Me)RhTPP (3) (δ -5.80 ppm). The chemical shift
difference in 5 (compared to 3) is much larger than in the
analogous PPh3 complex in C6D6 (∆δ ) 0.40 ppm),12 where
the value for the (PPh3)(Me)RhTPP complex is δ -5.94 ppm,
indicated as a multiplet. The downfield shift indicates a larger
electron density on the methylide due to the σ-donation of 1
(30) Huang, J. K.; Haar, C. M.; Nolan, S. P.; Marshall, W. J.; Moloy, K.
G. J. Am. Chem. Soc. 1998, 120, 7806.
Inorganic Chemistry, Vol. 42, No. 9, 2003 3089

Stulz et al.
Figure 3. Molecular units of (MeOH)(I)Rh(TPP) (2), (Me)Rh(TPP) (3), (DPAP)2(Rh)(TPP) (4), and (DPAP)(Me)Rh(TPP) (5), showing displacement
ellipsoids at 50% probability. H atoms (except on the rhodium bound methyl carbon C10 in 3 and C45 in 5) and the iodide counterion in 4 have been omitted
for clarity. In 5, only one of the two orientations of the disordered DPAP moiety is shown.
Table 2. Selected Geometrical Data for 2-5 (Å, deg)
2‚CHCl3
0.029
87.1
89.8
2.210(6) (O1)
2.5719(8) (I1)
2.028(7)
2.026(7)
2.028(7)
3
4‚2CHCl3
0.034
88.9
88.9
2.3707(7) (P1′)
2.3707(7) (P1)
2.043(2)
2.041(2)
2.043(2)
5
σ/dega
0
0.043
89.6
87.3
â/degb
R/degb
90
Rh-L2/Å
2.059(11) (C45)
2.512(3) (P1)
2.033(6)
2.024(6)
2.032(6)
Rh-L1/Å
1.968(12) (C10)
2.018(3)
2.018(3)
2.018(3)
2.018(3)
90
Rh-N4/Å
Rh-N3/Å
Rh-N2/Å
Rh-N1/Å
2.025(7)
2.041(2)
61.2 (C5-Ph)
66.8 (C10-Ph)
2.028(7)
dihedral angles of
phenyl groups with
porphyrin plane/deg
73.0 (C5-Ph)
84.7 (C5-Ph′)
71.2 (C10-Ph)
75.5 (C15-Ph)
82.7 (C20-Ph)
-0.21 (C21)
0.32 (C21′)
-0.06 (C27)
0.26 (C33)
0.04 (C39)
79.9 (C5-Ph)
77.7 (C10-Ph)
78.6 (C15-Ph)
75.4 (C20-Ph)
deviation of ipso
carbons
perpendicular to
porphyrin plane/Å
0
0.17 (C11)
0.20 (C21)
0.17 (C27)
0.17 (C33)
0.22 (C39)
-0.17 (C17)
0.17 (C11′)
-0.17 (C17′)
a σ denotes the average perpendicular deviation of the porphyrin core atoms from the least-squares plane through all 24 core atoms. b R and â denote the
angle formed by the Rh-L1/L2 bond with the least-squares porphyrin plane.
slightly shorter than in CH3Rh(F28TPP) (2.027(4) Å),5 in
CH3Rh(OEP) (2.031(6) Å),32b or in CH3Rh(OETAP) (2.034-
(7) Å).32c Complex 4 crystallizes in space group C2/c with
two CHCl3 molecules per porphyrin complex. The complex
is sited on a center of symmetry (point symmetry Ci). The
crystals of 5 were small and relatively weakly diffracting,
and data were observed only to 2θ ) 22° (Mo KR radiation;
equivalent to 0.95 Å resolution). In this case, anisotropic
refinement was possible only for the heavy atoms (Rh, P),
the porphyrin core, and the methylide carbon atom. The
DPAP ligand is disordered and was modeled in two orienta-
tions of equal occupancy related by rotation about the Rh-P
bond, with the phenyl rings constrained to be regular
hexagons.
In all these structures, the Rh-N bond distances lie in
the expected range 2.018-2.043 Å. The overall deviations
(31) The single-crystal structure of Zn(TPP)(H2O) displays a notable
example of this type of disorder. In an early structure report, the
disorder of the water molecule led to the incorrect description of the
structure as a dihydrate containing 6-coordinate Zn: Fleischer, E. B.;
Miller, C. K.; Webb, L. E. J. Am. Chem. Soc. 1964, 86, 2342. The
correct disordered interpretation was distinguished later on the basis
of the displacement parameter of the oxygen atom: Glick, M. D.;
Cohen, G. H.; Hoard, J. L. J. Am. Chem. Soc. 1967, 89, 1996.
(32) (a) Whang, D. M.; Kim, K. M. Acta Crystallogr., Sect. C: Cryst.
Struct. Commun. 1991, 47, 2547. (b) Takenaka, A.; Syal, S. K.; Sasada,
Y.; Omura, T.; Ogoshi, H.; Yoshida, Z.-I. Acta Crystallogr., Sect. C:
Cryst. Struct. Commun. 1976, 32, 62. (c) Ni, Y.; Fitzgerald, J. P.;
Carroll, P.; Wayland, B. B. Inorg. Chem. 1994, 33, 2029-2035.
3090 Inorganic Chemistry, Vol. 42, No. 9, 2003

Phosphine Complexation with Rh(III)TPP
from planarity (σ) are small, but clearly dependent on the
ligands bound to rhodium. In 2 and 4, saddlelike conforma-
tions are adopted in which two opposite phenyl substituents
are bent respectively upward and downward from the
porphyrin plane while the two other phenyl substituents are
not significantly distorted from the plane (Table 2). The
porphyrin deformation is slightly more pronounced in 2 than
in 4. Disorder of the phenyl group at the meso-position C5
in 2 suggests that the porphyrin conformation may be
variable, with this phenyl group adopting either a syn- or
anti- conformation with respect to the phenyl group at the
opposite meso-position as in a variety of (X)(I)Rh(TPP)
complexes33,34 and in (Me)Rh(DPP).17 By contrast, porphyrin
5 shows an umbrella-like conformation, with the phenyl
substituents on the meso-carbons bowed toward the methylide
ligand. Similar distortions are observed in the mono-
phosphine complex (PPh3)(Cl)Rh(OEP)9 and in CH3Rh-
(F28TPP).5 The umbrella-like distortion appears to minimize
steric interactions between the phosphine ligand and the
phenyl substituents of the porphyrin core and arises where
the ligand in the opposite axial position is relatively small.
In 2 and 4, in which the larger iodo ligand occupies the axial
position opposite the phosphine, saddlelike conformations
are observed. In the structures of 2, 4, and 5, all phenyl
substituents are twisted significantly from orthogonality to
the porphyrin plane (Table 2). This twist is largest in 4 and
appears to result from reduction of steric interactions be-
tween the phenyl rings and the phosphine ligands. In 2, the
Rh-OCH3 and Rh-I bond distances of 2.210(6) and 2.5719-
(8) Å, respectively, compare well with values reported
previously for analogous Rh(III) iodo porphyrins with
alcohols in the sixth axial coordination site.19,33
Figure 4. Stereo representation of part of the crystal packing in 4. Solvent
molecules, iodides, and hydrogen atoms are omitted, except for the para-
hydrogen on the phosphine phenyl substituent, which is involved in the
edge-to-face interaction in the crystal. The edge-to-face interaction and face-
to-face offset stacking is indicated by dashed lines (- - -).
Figure 5. Ball and stick representation of part of the crystal packing in 5.
Hydrogen atoms have been omitted. Only one of the disordered geometries
of the DPAP unit is shown.
comparable, with the acetylenic groups located on either side
of the meso-phenyl group at C10.
Since there is rapid rotation about the Rh-P bond in
solution, the positions of the DPAP moieties with respect to
the porphyrin in the solid state are likely to be influenced
by intermolecular interactions. In 4, the complexes are
arranged into layers (in the crystallographic ab plane) in
which all porphyrin planes in a given layer are parallel. Each
DPAP ligand interacts with the DPAP on two adjacent
porphyrins within the layer (Figure 4). The interaction with
the first porphyrin is reminiscent of the “phenyl embrace”
commonly observed between PPh3 groups,36 in which the
phenyl rings bound directly to P form mutual edge-to-face
arrangements across a center of symmetry (interplane angle
44.2°, centroid-to-centroid distance 4.78 Å). The phenyl ring
bound to the acetylenic group interacts in a face-to-face offset
manner with the corresponding ring on a second porphyrin,
with an interplane separation of 3.45 Å and a centroid-to-
centroid distance of 3.69 Å. In 5, the porphyrin complexes
are arranged into “bilayers” in which the DPAP ligands
interact at the center of the bilayer and the methylide ligands
remain at the outside (Figure 5). The arrangement is clearly
different from that in 4, forming in this case a two-
dimensional network of interactions.37 The phenyl embrace
exists between the phenyl groups bound directly to P, similar
to that observed in 4, although the rings adopt a significantly
Of particular interest are the structural features of the
coordinated DPAP ligands. The Rh-P bond distances are
significantly different in 4 and 5, increasing by ca. 0.14 Å
in 5 compared with 4. This can be explained by a consider-
ably larger trans-influence exerted by the methylide ligand
compared with that for two phosphines trans to one another.
The Rh-P bond distances in both 4 and 5 are longer than
that reported for (PPh3)(Cl)Rh(OEP) (2.306(3) Å)9 and are
also longer than in other rhodium phosphine complexes.30,35
Introduction of the DPAP ligand in the sixth axial coordina-
tion site in 5 lengthens the bond to the trans-methylide
carbon by ca. 0.09 Å compared with that in 5-coordinate 3.
The acetylenic groups in 4 and 5 deviate slightly from
linearity, with bond angles 169.9(3)° in 4 (P1sC35tC36),
and 174(3)° and 172(4)° for the two orientations of
P1sC58tC59 in 5.25 The relative orientation of the DPAP
unit with respect to the porphyrin core is similar in both 4
and 5: to avoid steric interactions, the phenyl rings of the
DPAP unit are located over the pyrrole groups. The two
orientations of the disordered DPAP moiety in 5 are
(36) Dance, I.; Scudder, M. Chem. Commun. 1995, 1039.
(33) Simonato, J. P.; Pecaut, J.; Marchon, J. C. Inorg. Chim. Acta 2001,
315, 240.
(34) Jameson, G. B.; Collman, J. P.; Boulatov, R. Acta Crystallogr., Sect.
C: Cryst. Struct. Commun. 2001, 57, 406.
(35) Lai, W.; Lau, M. K.; Chong, V.; Wong, W. T.; Leung, W. H.; Yu, N.
T. J. Organomet. Chem. 2001, 634, 61.
(37) The disordered description of the DPAP ligand arises from the two-
dimensional network of the interactions between adjacent DPAP
moieties within a given bilayer: those in adjacent porphyrins must
adopt alternately the first and second orientation along one interaction
direction (roughly along the [11h0] direction), but the same orientation
along the second interaction direction (roughly along [110]).
Inorganic Chemistry, Vol. 42, No. 9, 2003 3091

Stulz et al.
Figure 6. LDI-TOF mass spectra (a) of 4 and (b) of 5, recorded in the
positive ion mode as neat samples.
larger separation compared with 4 (interplane angle 68.6°,
centroid-to-centroid distance 5.96 Å). This larger separation
between the DPAP ligands accommodates additional edge-
to-face interactions involving one of the phenyl groups at
the meso-position of the second porphyrin (interplane angle
31.5°, centroid-to-centroid distance 4.95 Å). In addition, the
phenyl ring attached to the acetylenic linkage interacts with
the phenyl groups bound directly to P on two adjacent
porphyrins, forming a 3-fold edge-to-face motif (Figure 5).
Face-to-face interactions between phenyl rings are not
observed in 5.
LDI-TOF Mass Spectrometry. Mass spectrometric analy-
ses of the complexes were performed using LDI-TOF MS
without the addition of any matrix, which can interfere with
the coordination of ligands.38 Porphyrin 2 showed only a
mass peak at m/z 716.4, which corresponds to the molecular
ion with loss of iodide,38 and represents the bare [RhTPP]+•
(calcd m/z 715.63). Similar behavior was reported when using
LSI ionization.14 Measurement of 3 showed three peaks in
the molecular ion region, separated by 15 mass units. The
peaks can be assigned to the molecular ion [M]+• at m/z 730.7
(calcd m/z 730.66), the parent ion with loss of the methylide
[M - CH3]+• at m/z 715.7 (corresponding to [Rh(TPP)]+•),
and an additional peak at m/z 745.6, corresponding to the
methylide adduct of 3 [(Me)2Rh(TPP)]+• (calcd m/z 745.69),
arising from intermolecular methyl transfer. As can be seen
in Figure 6b, the relative intensity is increasing with
subsequent loss of the methyl groups. We have observed
similar intramolecular transfers with iodo substituted por-
phyrins, and this was attributed to laser induced photolytic
bond cleavage in the excited state.38 Addition of a matrix
such as hydroxy cinnamic acid, p-nitroaniline, or other
porphyrins did not prevent the fragmentation of 3.
Figure 7. UV-vis spectra (a) of (MeOH)(I)Rh(TPP) (2) (s) and
[(DPAP)2Rh(TPP)](I) (4) (- - -), and (b) of (Me)Rh(TPP) (3) (s) and
(DPAP)(Me)Rh(TPP) (5) (- - -). The spectra are recorded in CHCl3 at 298
K, c ) 10-5 M. The insets show the expanded part of the Q-band absorption
region.
Table 3. Absorption Maxima of the UV-Vis Spectra of
(MeOH)(I)Rh(TPP) (2), (Me)Rh(TPP) (3), [(DPAP)2Rh(TPP)](I) (4),
(DPAP)(Me)Rh(TPP) (5), and (DPAP)(I)Rh(TPP) (6) in CH2Cl2, T )
298 K
λ/nm (log ꢀ)
(MeOH)(I)Rh(TPP) (2)
(Me)Rh(TPP) (3)
(DPAP)(I)Rh(TPP) (6)
[(DPAP)2Rh(TPP)](I) (4)
422 (5.27), 533 (4.37), 566 (3.68)
412 (5.29), 520 (4.34), 547 (3.73)
437, 504a
323 (4.63), 357 (4.56), 376 (shoulder),
445 (5.28), 557 (4.23), 596 (4.19)
356 (3.62), 418 (shoulder), 444 (5.22),
516 (3.85), 556 (4.09), 597 (4.10),
620 (3.62)
(DPAP)(Me)Rh(TPP) (5)
a λ values are calculated using a Gaussian peak deconvolution.
see preceding description). The fragmentation in both 3 and
5 also affected the signal intensity. The observation that laser
induced loss of the methylide can occur from the excited
state in the gas phase is consistent with the reported
photolysis of 3 to generate RhII(TPP) in solution upon
irradiation at 416 nm.16 Dimerization, which often occurs
with Rh(II)-porphyrins in solution,3,13,39 was not observed
in the gas phase.
UV-Vis Spectroscopy. The UV-vis absorption spectra
of 2-6, measured in CHCl3, are displayed in Figure 7, and
the data are summarized in Table 3. Comparison with the
electronic absorption data of 2, preciously reported in a range
of solvents,15 confirms a significant solvatochromicity of the
lower energy Q-band absorption, which is blue-shifted in
The parent mass peak of 4 could not be detected, but loss
of one DPAP ligand gave the mono-phosphine complex
[(DPAP)Rh(TPP)]+• at m/z ) 1002.0 (calcd m/z 1001.93).
When (DPAP)(Me)Rh(TPP) (5) was analyzed in the LDI-
TOF MS, the molecular ion was detected at m/z 1018.2,
together with a peak at m/z 1002.0 (loss of the methylide,
(38) Stulz, E.; Mak, C. C.; Sanders, J. K. M. J. Chem. Soc., Dalton Trans.
2001, 604.
(39) Zhang, X. X.; Wayland, B. B. Inorg. Chem. 2000, 39, 5318.
3092 Inorganic Chemistry, Vol. 42, No. 9, 2003

Phosphine Complexation with Rh(III)TPP
Figure 8. (a) Selected electronic spectrum of a mixture of 2, 4, and 6 obtained during the titration of 2 with 1; [2] ) 9.89 × 10-7 M, [1] ) 1.38 × 10-6
M, 293 K. Peak deconvolutions (after baseline correction) are shown for the high energy part (b) and for the low energy part (c) of the spectrum; the broken
lines show the calculated absorption based on the best fit.
CHCl3 (566 nm) compared to the absorptions obtained in
solvents with strong donor properties such as n-PrCN (570
nm), DMF (574 nm), or DMSO (578 nm). The absorption
maxima found in 3 also match the reported data.9,10,12,13,15,16
Upon titration of 1 into a CHCl3 solution of 2, the
intermediate complex (DPAP)(I)Rh(TPP) (6) can only be
detected at low concentrations ([2] ) 10-6 M). Since there
is always a mixture of all three species 2, 4, and 6 present,
the spectrum of 6 could not be measured in pure form. The
absorption maxima of 6 were calculated using a Gaussian
peak function to deconvolute the spectrum at two different
rhodium-to-phosphine ratios. Figure 8 shows the results for
one of the deconvolutions. The peaks found for 2 and 4
match the values obtained for the pure samples, confirming
the validity of the method. The B-band absorption of 6 with
a 437 nm is more red-shifted than the value previously
observed for (PPh3)(THF)Rh(TPP) (λ 425 nm)13 but is close
to that of (PPh3)(Me)Rh(TPP) (λ 439 nm).12 The red-shift
of the B-band might indicate partial displacement of the
weakly bound iodide.15 The single Q-band absorption, which
is normally found in mono-phosphine rhodium porphyrin
complexes around 525-540 nm,13 was located at 504 nm,
which is a significant blue-shift, and displays the highest
energy Q-band absorption for a phosphine rhodium complex
so far. At higher concentrations of porphyrin (>10-6 M),
the titrations always yielded 4 directly. The final complex
[(DPAP)2Rh(TPP)](I) (4) shows a bathochromic shift of the
Soret band absorption of ∆λ ) 23 nm compared to 2, and
the Q-bands are also red-shifted by 24 and 30 nm, respec-
tively (Table 3). These values are very similar to those of
the analogous PPh3 and the more basic PPh2Me and PPhMe2
complexes. It seems that, in contrast to the mono-phosphine
complex, the electronic spectra of the bis-phosphine com-
plexes are relatively insensitive to the nature of the phosphine
ligand; thus, the MO energy separations are affected mini-
mally by changing the basicity of the ligand.
As was found for 6, the electronic spectrum for (DPAP)-
(Me)Rh(TPP) (5) is very different from that for the corre-
sponding PPh3 complex. UV-vis measurements led Kadish
et al.12 to conclude loss of the methylide at low concentra-
tions (<10-5 M) to produce [(PPh3)2RhTPP]+, whereas at
higher concentrations (>10-3 M), the mono-complex (PPh3)-
(Me)RhTPP is obtained.40 However, upon titrating 1 into a
solution of 3, we always obtained a spectrum which
essentially matched that of 4, independently of the concen-
tration of 3, which ranged from 10-6 to 5 × 10-4 M. To
determine whether this spectrum corresponds to that of 5 or
4, we performed a 1H NMR titration at 5 × 10-4 M porphyrin
concentration. This experiment showed that, even when
adding a large excess of ligand, the methylide remained
bound to rhodium, as judged from the high field doublet
resonance at ∼-6.5 ppm in the spectrum (data not shown).
We therefore confidently assign the electronic spectrum
measured as that of 5. An additional low energy band at
620 nm is present in 5, which is clearly missing in the
spectrum of 4. Some differences in the high energy part of
(40) We have been able to reproduce Kadish’s unexpected results with
PPh3.
Inorganic Chemistry, Vol. 42, No. 9, 2003 3093

Stulz et al.
Table 4. Thermodynamic Data Obtained from the Isothermal
Calorimetry Titrations of Reactions 1, 2 and 3a
∆G°/ ∆H°/
T∆S°/
kJ mol-1 kJ mol-1 kJ mol-1
Ka/M-1
1, (DPAP)(I)Rh(TPP) (6)
2, [(DPAP)2Rh(TPP)](I) (4) -25.1
3, (DPAP)(Me)Rh(TPP) (5) -24.6
-42.7
-90.4
-54.0
-54.0
-47.7 3.0 ((0.6) × 107
-28.8 4.6 ((0.6) × 104
-29.3 2.1 ((0.03) × 104
a See text; CHCl3, T ) 298 K.
The thermodynamic data, summarized in Table 4, show
that complexation of 1 to any of the rhodium porphyrins is
exothermic, and all binding events are enthalpy driven. The
first binding to (MeOH)(I)Rh(TPP) (2) according to reaction
1, which has a Gibbs standard free energy change of ∆G1°
) -42.7 kJ mol-1, shows the largest change both in enthalpy
(∆H1° ) -90.4 kJ mol-1) and in entropy (T∆S1° ) -47.7
kJ mol-1). These values are much larger than those obtained
for the binding of pyridine to zinc tetraphenyl porphyrin in
chloroform (∆GPy° ) -20.6 kJ mol-1, ∆HPy° ) -35.8 kJ
mol-1, T∆SPy° ) -15.2 kJ mol-1).41 Reaction 1 involves
displacement of a weakly bound solvate molecule (methanol)
by 1, and the large value for ∆H1° indicates a high Rh-P
bond energy in 6. The relatively large entropic contribution
might arise from much stronger solvation of the released
polar methanol by CHCl3 compared to 1. Reactions 2 and 3
show almost identical Gibbs free energy changes: ∆G2° )
-25.1 kJ mol-1 and ∆G3° ) -24.6 kJ mol-1, respectively.
The enthalpy change ∆H2,3° is -54.0 kJ mol-1 for both
reactions, and the change in entropy T∆S2° and T∆S3° differ
only by 0.5 kJ mol-1. All values for reactions 2 and 3 are
about 0.59 that of the first binding of 1 to 2.
Figure 9. Isothermal titration calorimetry measurements (a) of 2 with 1,
and (b) of 3 with 1. The upper parts of the figures display the measured
heat effects, and the lower parts show the extracted enthalpy values. The
dashed lines (- - -) represent the nonlinear fitting procedures to the data for
the calculation of the thermodynamic parameters.
DPAP (1) + (MeOH)(I)Rh(TPP) (2) a
(DPAP)(I)Rh(TPP) (6) + MeOH Ka1 (1)
DPAP (1) + (DPAP)(I)Rh(TPP) a
[(DPAP)2Rh(TPP)](I) (4) Ka2 (2)
the spectra below 400 nm can be seen as well. Our UV-vis
and NMR spectroscopic data indicate that the methylide is
not displaced upon titrating 3 with 1 at any concentration. It
seems also in this case that the MO energies of the mono-
phosphine complex are highly sensitive to the nature of the
phosphorus ligand. Using a wider range of different σ-donor
and π-acceptor phosphines or phosphonates should therefore
lead to very different electronic properties of the complexes
obtained.
Isothermal Titration Calorimetry. Thermodynamic data
of ligand binding to metalloporphyrins are relatively scarce.
Solvent effects on the binding of pyridine to zinc porphyrins
have been described.41 Also, the values for the binding of
PPh3 to (Cl)Rh(OEP), obtained by variable temperature
equilibrium measurements, have been reported.9 In order
to gain insight into the thermodynamics of the binding of
ligand 1 to rhodium(III) porphyrins, we have performed
isothermal titration calorimetry (ITC) with both 2 and 3 at
298 K (Figure 9).
DPAP (1) + (Me)Rh(TPP) (3) a
(DPAP)(Me)Rh(TPP) (5) Ka3 (3)
The enthalpy change for reaction 2 is larger than for the
displacement of chloride from (PPh3)(Cl)Rh(OEP) to form
[(PPh3)2Rh(OEP)](Cl), where a value of ∆HPPh3° ) -32.6
kJ mol-1 was determined in CH2Cl2.9 The entropy change
T∆SPPh3° for the displacement of chloride by PPh3 is -43.4
kJ mol-1 (calcd at 298 K) and is considerably more
unfavorable than the entropy change associated with the
displacement of iodide in our system, which could be due
to solvation effects of the different ions released in the
reactions. In fact, the displacement of chloride by PPh3 is
overall an unfavorable reaction (∆GPPh3° ) 10.8 kJ mol-1,
Kassoc ) 0.0174).9
In reaction 3, no bonds are being broken, and to a first
approximation, the ∆H3° value can be regarded as reflecting
the Rh-P bond energy in 5. The value of ∆H3° ) -54.0 kJ
mol-1 is in the range of the reaction enthalpies found for
displacement of cyclooctene by PPh3 in other rhodium
complexes (∆HPPh3° ) -30 to -45 kJ mol-1).30 Compared
(41) (a) Zielenkiewicz, W.; Lebedeva, N. S.; Antina, E. V.; Vyugin, A. I.;
Kaminski, M. J. Solution Chem. 1998, 27, 879. (b) Zielenkiewicz,
W.; Lebedeva, N. S.; Kaminski, M.; Antina, E. V.; Vyugin, A. I. J.
Therm. Anal. 1999, 58, 741.
3094 Inorganic Chemistry, Vol. 42, No. 9, 2003

Phosphine Complexation with Rh(III)TPP
second, broad peak of much larger intensity, and showing a
maximum heat effect ca. 1.7 min after initial mixing. The
heat effect of the second process, which dominates in the
presence of excess iodide, arises from reaction 2. This
indicates that, at concentrations in the submillimolar range,
the iodide is partially displaced, and 6 actually exists largely
as a contact ion pair [(DPAP)Rh(TPP)]+‚‚‚I-. The coordina-
tion of 1 to the 5-coordinate [(DPAP)Rh(TPP)]+ is rapid, in
contrast to the displacement of iodide in reaction 2. The rate
determining step in reaction 2 therefore seems to be
dissociation of the iodide.
For the binding of 1 to 2, the data were fitted using a
model with two independent binding events, and the sto-
ichiometries were in good agreement with a 1:1 stoichiometry
for each binding site. For the binding of 1 to 3, the fitting
procedure was performed using a single binding model, again
in agreement with a stoichiometry of 1:1. No displacement
of the methyl group was observed. The thus obtained values
for Kassoc agree well with the values calculated from UV-
vis titrations (data not shown). Clearly, formation of 6 shows
the highest thermodynamic stability with Ka1 ) 3.0 ((0.6)
× 107 M-1. The second binding of 1 to 4 is less favorable
by a factor of about 1000 [Ka2 ) 4.6 ((0.6) × 104 M-1]. A
similar decrease in binding strength has been reported for
the complexation of PPh3 to (O2)Rh(TPP).10 Since the
basicity of iodide (pKa ) -11) is much lower than that of
1 (pKa ) 1.04),25 the much weaker second binding is due to
the much stronger σ/π-interactions of the phosphine ligand
compared to iodide. Phosphorus ligands are expected to have
a strong trans-influence,35 which is expressed in the ther-
modynamic parameters. However, a large trans-effect, which
would affect the kinetics of the displacement reactions, does
not seem to be the case, because the ligand exchange of
iodide with 1 is relatively slow.
Ka3 ) 2.1 ((0.03) × 104 M-1 is in the same order of
magnitude as Ka2. This value is about 10 times larger than
the binding of PPh3 to 3 (Kassoc ) 3.9 × 103 M-1).12 Since
PPh3 is a stronger base (pKa ) 2.73)42 than 1, the weaker
binding of PPh3 is most probably due to steric effects. We
have already observed an analogous difference in the binding
of PPh3 and 1 to ruthenium porphyrins.25 Strong binding of
a sixth ligand was not expected, since alkyl rhodium
porphyrins usually do not have additional solvent molecules
such as alcohols or ethers attached, both in solution and in
the solid state, contrary to the halide complexes. The
σ-bonded methylide does not have any significant π-acceptor
properties, and the high basicity (pKa ∼ 50) drastically
increases the electron density on the metal, thus reducing
the electrophilicity of the metal, hence the much lower Ka3
value.
Figure 10. Overlay of the ITC profiles recorded during titration of the
porphyrins with DPAP (1) in CHCl3. The peaks are normalized to the same
height. (a) (s) 2 with 1, (- - -) 3 with 1, (b) (s) 6 with 1, (- - -) 6 with 1
in the presence of 5 equiv of iodide.
to the binding of pyridine to Zn(TPP) in CHCl3,41 which is
similar in that no ligand displacement occurs, the bond
energy in 5 is stronger by ∆∆H° ) -18.2 kJ mol-1, but the
larger entropy cost of ∆∆S° ) 14.1 kJ mol-1 might be
due to the loss in rotational freedom of the substituents on
the phosphorus, caused by steric interactions with the
porphyrin core. In addition, desolvation must be taken into
account. For reaction 3, almost the same change in enthalpy
is observed as for reaction 2, and this could be explained by
two factors: (i) the iodide is only weakly bound, so that
displacement does not require significant enthalpy contribu-
tions, and (ii) the axially bound phosphine has an overall
similar electronic trans-influence as the σ-bonded methyl
group.
Figure 10 shows the overlay of some of the injections
recorded upon formation of 4, 5, and 6. The almost identical
profiles for the formation of 5 and 6 (Figure 10a) indicate
that the MeOH is only very weakly bound and is rapidly
displaced or even already dissociated in this concentration
range. On the other hand, the formation of 4 consists of two
different processes, as can be seen from the shape of the
recorded peak in Figure 10b (s). After an initial rapid
reaction, the peak shape changes, and the process follows a
much slower pathway. Addition of 5 equiv of iodide (as tert-
butylammonium salt) changes the heat emission profile,
and the two processes become clearly visible [Figure 10b
(- - -)]. A first small and sharp peak, identical in shape to
the ones obtained for reactions 1 and 3, is followed by a
Conclusions
We have presented the analytical data of new phosphine
rhodium porphyrin complexes, including spectroscopic stud-
ies (UV-vis, 1H NMR, and 31P NMR), mass spectrometry,
thermodynamic analysis, and the solid-state structures of all
(42) Henderson, W. A.; Streuli, C. A. J. Am. Chem. Soc. 1960, 82, 5791.
Inorganic Chemistry, Vol. 42, No. 9, 2003 3095

Stulz et al.
complexes. The spectroscopic results presented here show
that the properties of phosphine (P) complexes (P)(X)RhIII-
(TPP) are sensitive to the nature of the axial trans-ligand X,
where X ) I, Me, or P. The mono-complexes of DPAP (1)
display especially large differences in the thermodynamic
parameters and in the electronics of the porphyrin, i.e., in
the UV-vis and 31P NMR spectra, depending on whether
the trans-ligand is iodide as in 6, or methylide as in 5. In
the former case, binding is strong and is accompanied by
large changes in enthalpy and entropy; in the latter case,
binding is relatively weak, and the thermodynamic changes
are significantly smaller. The nature of the phosphine ligands
also seems to have a large influence on the electronics of
the complex formed. Even though the absorption spectrum
of 5 resembles the spectrum of 4 very closely, loss of the
σ-bonded methyl ligand is not observed, contrary to what
was found with the slightly more basic PPh3 as ligand.12
For the bis-phosphine complex 4, the electronic spectrum
obtained is very similar to those reported with other
phosphine ligands.8,11-13,15 The relative energies in the bis-
phosphine complexes therefore do not seem to be very
sensitive to the nature of the phosphorus ligand. The
phosphorus as trans-ligand weakens the second binding
considerably. The spectroscopic and thermodynamic data
suggest that 1 and Me have an almost identical trans-
influence on the opposite axial ligand, but these arise from
very different σ-donor and π-acceptor properties. The
electronic spectra are strikingly similar, despite the large
differences of the absorption spectra of the free porphyrins
2 and 3. 31P NMR spectroscopy has shown to be very
sensitive to the nature of the trans-ligand: in 4 and 6, the
resonances are sharp and downfield shifted compared to the
free phosphine. Since both complexes are almost isochro-
nous, the Rh-P coupling constants can be used to distinguish
the two complexes. In the case of a σ-bonded alkyl group
as sixth ligand, the 31P NMR spectrum displays a broad and
only marginally shifted resonance for the bound phosphine.
A recent density functional computational study found43
that metal-phosphine bond energies cannot be regarded as
intrinsic, universal, or transferable properties, and the synergy
between σ-donor and π-acceptor ligands is pivotal in the
interpretation of bond enthalpies. Our thermodynamic data
show that very similar values can be observed when ligands
exhibiting large differences in σ/π-properties are trans to
each other. Overall, the trans-influence of the σ-bonded
methyl group is comparable to 1, but in the former, it is due
to strong σ-donation, whereas in the latter the π-acceptor
properties seem to play an important role.
Since the phosphine ligand used in these studies serves
us as a model for our phosphine substituted porphyrins, the
results discussed here let us predict that incorporating
rhodium porphyrins in our supramolecular arrays will lead
to complexes exhibiting attractive physicochemical proper-
ties. The binding constants of alkynyl phosphines to Rh(III)
porphyrins are high enough to be useful in the construction
of a stable supramolecular structure (Chart 1). Both axial
sites on the porphyrin can be used for complexation;
however, the sixth coordination site can be blocked by
introducing a σ-bonded methyl group, allowing the selective
formation of mono-phosphine complexes without strongly
affecting the electronic properties. The results show that 1
is a very versatile ligand which forms more stable complexes
with rhodium(III) porphyrins than PPh3 and is a good model
for studying phosphine-rhodium interactions in larger arrays.
Acknowledgment. Financial support from the Swiss
National Science Foundation (E.S.), from the Royal Society
(S.O.), and from the EPSRC (E.S., S.M.S., and A.D.B.) is
gratefully acknowledged.
Supporting Information Available: X-ray crystallographic
tables for structures 2, 3, 4, and 5 (atomic coordinates, bonds lengths
and angles, anisotropic thermal factors, hydrogen atom positions
as CIF files). This material is available free of charge via the Internet
at http://pubs.acs.org.
1
IC026257A
Also, no JRh,P coupling could be observed. This behavior
can be explained by a dynamic ligand exchange of unbound
and bound species.
(43) Landis, C. R.; Feldgus, S.; Uddin, J.; Wozniak, C. E.; Moloy, K. G.
Organometallics 2000, 19, 4878.
3096 Inorganic Chemistry, Vol. 42, No. 9, 2003