Phosphine and phosphonite complexes of a ruthenium(II) porphyrin. 1. Synthesis, structure, and solution state studies

Article abstract of DOI:10.1021/ic025727y

We have investigated the effect of complexation of different phosphorus ligands on the stability, solid state structure, and spectroscopic properties (NMR, IR, UV-vis) of a 5,15-diphenyl-substituted ruthenium porphyrin, (MeOH)Ru^II-(CO)(DPP) 2 [DPP = 5, 15-bis(3′,5′-di-tert-butyl)phenyl-2,8,12,18- tetraethyl-3,7,13,17-tetramethylporphyrin]. The ligands used are PPh₃, diphenyl(phenylacetenyl)phosphine (DPAP), bis(diphenylphosphino)acetylene (DPPA), tris- (phenylacetenyl)phosphine [(PA)₃P], and diethyl (phenylacetenyl)phosphonite [PAP(OEt)₂]. The mono-phosphine complexes (PR₃)Ru^II(CO)(DPP) are readily formed in solution in quantitative yields. The complexes display association constants ranging from 1.2 x 10⁴ M^-1 for PPh₃ to 4.8 x 10⁶ M^-1 for PAP(OEt)₂. The weak association of PPh₃ does not correlate with its pK_a, δ(³¹P), or cone angle value and is attributed to steric effects. Due to their kinetic lability, which is shown by 2D NMR spectroscopy, and the weakening of the carbonyl ligand via a trans effect, the mono-phosphine complexes could not be isolated. IR spectroscopy gives the relative order of π-acceptor strength as PPh₃ < DPAP, DPPA < (PA)₃P < PAP(OEt)₂, whereas the relative order of the σ-donor strength is PPh₃ < (PA)₃P < DPAP, DPPA < PAP(OEt)₂, based on the calculated pK_a values and on the ³¹P{¹H} NMR chemical shifts of the ligands. The chemical shift differences in the ³¹P{¹H} NMR spectra upon ligand binding display a linear correlation with the calculated pK_a values of the protonated ligands HPR₃⁺; we propose that the pK_a, and probably other electronic properties, of a specific phosphorus ligand can be estimated on the basis of the chemical shift difference Δδ(³¹P) upon complexation to a metalloporphyrin. The bis-phosphine complexes can be isolated in pure form by crystallization from CHCl₃-MeOH solutions using excess ligand. Association of the second ligand is in the same order of magnitude as the first binding for the phosphines, but the second phosphonite binding is decreased by a factor of about 100. The solid state structures show only marginal differences in the geometrical parameters. The calculated and the crystallographic cone angles of the ligands generally do not match, apart from the values obtained for PAP(OEt)₂.

Full text of DOI:10.1021/ic025727y

Inorg. Chem. 2002, 41, 5255−5268
Phosphine and Phosphonite Complexes of a Ruthenium(II) Porphyrin.
1. Synthesis, Structure, and Solution State Studies
Eugen Stulz,* Michael Maue, Neil Feeder, Simon J. Teat, Yiu-Fai Ng, Andrew D. Bond,
Scott L. Darling, and Jeremy K. M. Sanders*
UniVersity Chemical Laboratory, UniVersity of Cambridge, Lensfield Road,
Cambridge, CB2 1EW, U.K.
Received May 17, 2002
We have investigated the effect of complexation of different phosphorus ligands on the stability, solid state structure,
II
and spectroscopic properties (NMR, IR, UV−vis) of a 5,15-diphenyl-substituted ruthenium porphyrin, (MeOH)Ru -
(CO)(DPP) 2 [DPP ) 5,15-bis(3′,5′-di-tert-butyl)phenyl-2,8,12,18-tetraethyl-3,7,13,17-tetramethylporphyrin]. The ligands
used are PPh3, diphenyl(phenylacetenyl)phosphine (DPAP), bis(diphenylphosphino)acetylene (DPPA), tris-
(phenylacetenyl)phosphine [(PA)3P], and diethyl (phenylacetenyl)phosphonite [PAP(OEt)2]. The mono-phosphine
II
complexes (PR )Ru (CO)(DPP) are readily formed in solution in quantitative yields. The complexes display association
3
constants ranging from 1.2 × 104 M-1 for PPh3 to 4.8 × 106 M-1 for PAP(OEt)2. The weak association of PPh3
does not correlate with its pK , δ(31P), or cone angle value and is attributed to steric effects. Due to their kinetic
a
lability, which is shown by 2D NMR spectroscopy, and the weakening of the carbonyl ligand via a trans effect, the
mono-phosphine complexes could not be isolated. IR spectroscopy gives the relative order of π-acceptor strength
as PPh3 < DPAP, DPPA < (PA)3P < PAP(OEt)2, whereas the relative order of the σ-donor strength is PPh3 <
(PA)3P < DPAP, DPPA < PAP(OEt)2, based on the calculated pK values and on the 31P{1H} NMR chemical shifts
a
of the ligands. The chemical shift differences in the 31P{1H} NMR spectra upon ligand binding display a linear
correlation with the calculated pK values of the protonated ligands HPR +; we propose that the pK , and probably
a
3
a
other electronic properties, of a specific phosphorus ligand can be estimated on the basis of the chemical shift
difference ∆δ(31P) upon complexation to a metalloporphyrin. The bis-phosphine complexes can be isolated in pure
form by crystallization from CHCl3−MeOH solutions using excess ligand. Association of the second ligand is in the
same order of magnitude as the first binding for the phosphines, but the second phosphonite binding is decreased
by a factor of about 100. The solid state structures show only marginal differences in the geometrical parameters.
The calculated and the crystallographic cone angles of the ligands generally do not match, apart from the values
obtained for PAP(OEt)2.
Introduction
recognition. So far, we have made use of the known
preferences of Zn(II), Ru(II), and Rh(III) for nitrogen and
Sn(IV) for oxygen. In this way, we have been able to
synthesize assemblies of heterometallic oligoporphyrins using
cooperative zinc-nitrogen, ruthenium-nitrogen, and tin-
oxygen coordination chemistry,5 and recently an undecamer,
containing porphyrins in four different metalation states.6
Multiporphyrinic arrays constructed through covalent or
noncovalent connections have potential in acting as photo-
chemical devices,1,2 electronic multibit storage units,3 or
mimics of the natural photosynthetic system.4 One promising
approach is the self-assembly of heterometallic oligomers
via orthogonal metal-ligand binding according to Pearson’s
HSAB principle.5 The key to our strategy is to attach
covalently a potential ligand onto a first metalloporphyrin,
which is specifically chosen to recognize selectively a second
metalloporphyrin with a different metal, thus preventing self-
(1) (a) Yamaguchi, H.; Kamachi, M.; Harada, A. Angew. Chem., Int. Ed.
2000, 39, 3829. (b) Norsten, T. B.; Branda, N. R. J. Am. Chem. Soc.
2001, 123, 1784. (c) Imahori, H.; Arimura, M.; Hanada, T.; Nishimura,
Y.; Yamazaki, I.; Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001,
123, 335. (d) Cho, H. S.; Jeong, D. H.; Yoon, M. C.; Kim, Y. H.;
Kim, Y. R.; Kim, D.; Jeoung, S. C.; Kim, S. K.; Aratani, N.; Shinmori,
H.; Osuka, A. J. Phys. Chem. A 2001, 105, 4200. (e) 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.
* To whom correspondence should be addressed. E-mail: es255@
cam.ac.uk (E.S.); jkms@cam.ac.uk (J.K.M.S.).
10.1021/ic025727y CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/12/2002
Inorganic Chemistry, Vol. 41, No. 20, 2002 5255

Stulz et al.
Chart 1
Chart 2
To expand our repertoire of orthogonal binding modes,
we have become interested in the use of phosphorus as ligand
for ruthenium porphyrins. We have shown that phosphine-
substituted porphyrins are versatile building blocks for the
construction of supramolecular assemblies,7 including selec-
tion and virtually quantitative amplification of a thermody-
namically stabilized, heterometallic, tetraporphyrinic con-
struct from a dynamic combinatorial library.8 In view of
possible photophysical applications, the use of phosphorus
to connect porphyrins via coordination chemistry to ruthe-
nium(II), together with the immense diversity accessible by
the variation of the substituents on the coordinating phos-
phorus, should offer a convenient means for fine-tuning the
physical properties of assemblies. In order to predict
electronic interactions in arrays such as the [Zn/Ru/Zn] trimer
1 (Chart 1),7 it is advantageous to have detailed knowledge
about the structure and physical properties of phosphorus
metalloporphyrin complexes.
Trivalent phosphorus has been used as ligand for metal-
loporphyrins for several decades. Investigations have mainly
been focused on ferrous,9,10 ferric,11-13 Co(II),14 Ru(II),15-18
and Rh(II)19 porphyrins. Some phosphorus ruthenium(II)
porphyrin complexes have been isolated and characterized,
namely, RuII(TPP) (TPP ) tetraphenylporphyrin) and RuII-
(OEP) (OEP ) octaethylporphyrin) complexes of PPh3,15,16
of P(p-MeOPh)3,15 of PBu3,16,20 and of PF3.21 Cheng et al.18
have described a (PBu3)RuII(TPTBP) complex, the porphyrin
being a structural hybrid of TPP and tetrabenzoporphyrin
(TBP). To date no systematic study has been performed
where different phosphorus ligands were complexed to a
structurally fixed Ru(II) porphyrin.
We now have investigated the effect of different phos-
phorus ligands on stability, solid state structure, and spec-
troscopic properties (NMR, IR, UV-vis) upon complexation
to (MeOH)RuII(CO)(DPP) 2 [DPP ) 5,15-bis(3′,5′-di-tert-
butyl)phenyl-2,8,12,18-tetraethyl-3,7,13,17-tetramethylpor-
phyrin, Chart 2]. Photophysical and electrochemical studies
as well as EHMO calculations on these model complexes
will be described in the following paper.61 In our series of
phosphorus ligands, one phenylalkynyl substituent was held
constant [except for bis(diphenylphosphino)acetylene, DPPA
3] matching the substitution pattern of our phosphorus-
(9) (a) Simonneaux, G.; Sodano, P. Inorg. Chem. 1988, 27, 3956. (b)
Simonneaux, G.; Sodano, P. J. Organomet. Chem. 1988, 349, C11.
(c) Guillemot, M.; Simonneaux, G. J. Chem. Soc., Chem. Commun.
1995, 2093.
(10) Pilard, M. A.; Guillemot, M.; Toupet, L.; Jordanov, J.; Simonneaux,
G. Inorg. Chem. 1997, 36, 6307.
(11) Sodano, P.; Simonneaux, G.; Toupet, L. J. Chem. Soc., Dalton Trans.
1988, 2615.
(2) Mak, C. C.; Bampos, N.; Darling, S. L.; Montalti, M.; Prodi, L.;
Sanders, J. K. M. J. Org. Chem. 2001, 66, 4476.
(12) (a) Rieger, P. H. Coord. Chem. ReV. 1994, 135, 203. (b) Ohya, T.;
Morohoshi, H.; Sato, M. J. Pharmacobio-Dyn. 1985, 8, S14.
(13) Ohya, T.; Morohoshi, H.; Sato, M. Inorg. Chem. 1984, 23, 1303.
(14) Wayland, B. B.; Sherry, A. E.; Bunn, A. G. J. Am. Chem. Soc. 1993,
115, 7675.
(15) Ariel, S.; Dolphin, D.; Domazetis, G.; James, B. R.; Leung, T. W.;
Rettig, S. J.; Trotter, J.; Williams, G. M. Can. J. Chem. 1984, 62,
755.
(3) (a) Gryko, D. T.; Clausen, C.; Roth, K. M.; Dontha, N.; Bocian, D.
F.; Kuhr, W. G.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7345. (b) Li,
J. Z.; Gryko, D.; Dabke, R. B.; Diers, J. R.; Bocian, D. F.; Kuhr, W.
G.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7379. (c) Collman, J. P.;
Kendall, J. L.; Chen, J. L.; Collins, K. A.; Marchon, J. C. Inorg. Chem.
2000, 39, 1661.
(4) (a) Haycock, R. A.; Hunter, C. A.; James, D. A.; Michelsen, U.; Sutton,
L. R. Org. Lett. 2000, 2, 2435. (b) Gust, D.; Moore, T. A.; Moore, A.
L. Acc. Chem. Res. 2001, 34, 40. (c) Imahori, H.; Norieda, H.; Yamada,
H.; Nishimura, Y.; Yamazaki, I.; Sakata, Y.; Fukuzumi, S. J. Am.
Chem. Soc. 2001, 123, 100. (d) Sun, L.; Hammarstrom, L.; Akermark,
B.; Styring, S. Chem. Soc. ReV. 2001, 30, 36.
(16) Chow, B. C.; Cohen, I. A. Bioinorg. Chem. 1971, 1, 57.
(17) James, B. R.; Dolphin, D.; Leung, T. W.; Einstein, F. W. B.; Willis,
A. C. Can. J. Chem. 1984, 62, 1238.
(18) Cheng, R. J.; Lin, S. H.; Mo, H. M. Organometallics 1997, 16, 2121.
(19) (a) Grass, V.; Lexa, D.; Momenteau, M.; Saveant, J. M. J. Am. Chem.
Soc. 1997, 119, 3536. (b) Collman, J. P.; Boulatov, R. J. Am. Chem.
Soc. 2000, 122, 11812.
(20) (a) Tsutsui, M.; Osterfeld, D.; Hoffman, L. M. J. Am. Chem. Soc.
1971, 93, 1820. (b) Tsutsui, M.; Osterfeld, D.; Francis, J. N.; Hoffman,
L. M. J. Coord. Chem. 1971, 115. (c) Barley, M.; Becker, J. Y.;
Domazetis, G.; Dolphin, D.; James, B. R. Can. J. Chem. 1983, 61,
2389.
(5) Kim, H. J.; Bampos, N.; Sanders, J. K. M. J. Am. Chem. Soc. 1999,
121, 8120.
(6) Redman, J. E.; Feeder, N.; Teat, S. J.; Sanders, J. K. M. Inorg. Chem.
2001, 40, 2486.
(7) Darling, S. L.; Stulz, E.; Feeder, N.; Bampos, N.; Sanders, J. K. M.
New J. Chem. 2000, 24, 261.
(8) Stulz, E.; Ng, Y.-F.; Scott, S. M.; Sanders, J. K. M. Chem. Commun.
2002, 524.
(21) Kadish, K. M.; Hu, Y.; Tagliatesta, P.; Boschi, T. J. Chem. Soc., Dalton
Trans. 1993, 1167.
5256 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
a Varian Cary 100 Bio spectrophotometer. IR spectra were obtained
on a Perkin-Elmer Paragon 1000 FTIR spectrometer using CH2Cl2
as solvent (NaCl cell, 0.5 mm). The positive ESI-HRMS spectrum
of 6 was recorded on a Micromass Q-TOF1 MS instrument.
X-ray diffraction data were collected for (DPAP)2RuII(DPP)
(triclinic form) and [PAP(OEt2)]2RuII(DPP) using a Nonius Kappa
CCD diffractometer. Crystals of [(PA)3P]2RuII(DPP), (DPPA)2-
RuII(DPP), and (DPAP)2RuII(DPP) (monoclinic form) were small
and weakly diffracting, and data were collected for these at Station
9.8, Daresbury SRS, U.K., using a Bruker SMART CCD diffrac-
tometer. Structures were solved by direct methods using either
SHELXS-9732 or SIR-9233 and refined against all F 2 data using
SHELXL-97.32 A summary of the crystallographic data is given in
Table 1.
substituted porphyrins, as shown in 1, to enable direct
comparisons to be made. This should allow direct application
of the results obtained from this study to porphyrin-
substituted phosphines in supramolecular structures. PPh3 4
was added to the series for comparison with known literature
data for TPP and OEP.15,16
Generally, the physicochemical properties of phosphorus
complexes are influenced by both steric and electronic
effects. Steric effects are most commonly expressed in terms
of Tolman’s cone angle θ,22 obtainable from crystallographic
data.23 Electronic effects are controlled by a combination of
σ-donor and π-acceptor properties. σ-Donation usually
decreases in the order phosphine > phosphonite, e.g., in
going from diphenyl(phenylacetenyl)phosphine (DPAP 5)
to diethyl (phenylacetenyl)phosphonite [PAP(OEt)2 6].13
π-Acidity, on the other hand, is expected to be increased
for phosphonites10,24,25 compared to phosphines due to
P(3d)-O(2pπ/2pπ*) orbital interactions. Substituting phenyl
for alkyne, as in tris(phenylacetenyl)phosphine [(PA)3P 7],
has an influence on both σ-donor26 and π-acceptor27 proper-
ties, and an increased electronic interaction between the
phosphorus and the three acetylene substituents compared
to 5 may be expected.28 The same interactions should also
lead to differences in the properties of the bis(diphenylphos-
phino)acetylene (DPPA 3), where the acetylene unit bears
two phosphorus atoms compared to DPAP. Overall, the
ligands in our series were chosen to cover a wide range of
substitution patterns and of electronic properties.
PAP(OEt)2 (6).34 Phenylacetylene (2.00 g, 19.6 mmol) was
dissolved in THF (20 mL), and, after cooling to -78 °C (dry ice/
acetone bath), a 1 M EtMgBr solution (19.6 mL) was slowly added
via syringe. After 15 min of stirring at -78 °C, ClP(OEt)2 (3.07 g,
19.6 mmol) in THF (10 mL) was added dropwise. The yellowish
mixture was stirred for 30 min at -78 °C and warmed to room
temperature, and the reaction was quenched by addition of 20 mL
of saturated Na2CO3 aqueous solution. The mixture was poured
into 1:1 CH2Cl2/H2O (400 mL, purged with N2). The organic phase
was separated, washed once with H2O (100 mL, N2), dried
(MgSO4), and evaporated in vacuo. Column chromatography (silica,
5:1 hexane/EtOAc, N2) gave 6 as a colorless oil. Yield: 3.68 g
(16.6 mmol, 84%). The product is air sensitive and smelly; it should
be stored under an inert atmosphere at -20 °C. 1H NMR (400 MHz,
CDCl3): δ 7.48 (t, 2 H, o-H), 7.32 (m, 3 H, m/p-H), 4.04 (q, J )
6 Hz, 4 H, OCH2CH3), 1.31 (t, J ) 6 Hz, 6 H, OCH2CH3) ppm.
13C NMR (125.70 MHz, CDCl3): δ 132.1, 129.3, 128.3, 121.6,
2
1
103.7 (d, JP-C ) 3.4 Hz, P-C), 89.2 (d, JP-C ) 42.6 Hz), 63.5
(d, 2JP-C ) 6.8 Hz), 17.0 ppm. 31P{1H} NMR (162 MHz, CDCl3):
δ 131 ppm. ESI-HRMS: calcd for C12H16O2P [M + H+] 223.0888,
found 223.0881 [M + H+].
Experimental Section
All manipulations were performed using standard inert atmos-
phere techniques and freshly distilled and degassed solvents:
methylene chloride (CH2Cl2), chloroform (CHCl3), and tetrahy-
drofuran (THF) from CaH, and methanol (MeOH) from Mg; CDCl3
was filtered over basic alumina and degassed by purging with Ar
prior to use. (MeOH)RuII(CO)(DPP) (2),2,29 DPAP (5),30 and (PA)3P
(7)31 were prepared using literature procedures. DPPA (3) was
purchased from Strem Chemicals, and phenylacetylene, EtMgBr
(1.0 M THF), and ClP(OEt)2 were purchased from Aldrich (reagent
grade). All chemicals were used as obtained.
NMR spectra were recorded on a Bruker DPX400 NMR
spectrometer at 400.13 MHz (1H, solvent as internal standard) or
161.98 MHz (31P{1H}, H2PO4 external standard), or on a Bruker
DRX500 NMR spectrometer at 500.13 MHz (1H) or 125.70 MHz
(13C, solvent as internal standard). Abbrevations for 1H NMR spectra
used are as follows: s, singlet; d, doublet; t, triplet; dt, doublet of
triplets; m, multiplet; b, broad. UV-vis spectra were recorded on
Syntheses of the Mono-Phosphino Complexes (PR3)RuII(CO)-
(DPP). General Procedure. 2 (5.0 mg, 4.93 µmol) was suspended
in 0.5 mL of solvent (CDCl3 or CD2Cl2 for NMR studies, CH2Cl2
for UV-vis and IR measurement). Phosphine (10 equiv) was
dissolved in 1.0 mL of solvent, and 100 µL was added via syringe
to the porphyrin suspension. The solution was stirred at room
temperature under Ar atmosphere until all starting material had
dissolved (usually 15-30 min), and the orange solution was directly
used for spectroscopic studies.
(DPAP)RuII(CO)(DPP) (8). IR (CH2Cl2): 1943 (νCtO) cm-1
.
UV-vis (CH2Cl2): λ (log ꢀ) 413 (5.12), 533 (4.08), 560 (3.52)
nm. 1H NMR (CDCl3, 400 MHz): δ 9.72 (s, 2 H, por-mesoH; por
) porphyrinic), 7.82 (bs, 2 H, por-ArH), 7.70 (bs, 2 H, por-ArH),
7.33 (bs, 2 H, por-ArH), 7.23 (m, 2 H, CtCArH), 7.21 (m, 2 H,
CtCArH), 7.00 (dd, J ) 6, 2 Hz, 4 H, CtCArH), 6.78 (dt, J )
7, 1 Hz, 4 H, PAr-pH), 6.47 (dt, J ) 9, 2 Hz, 4 H, PAr-mH), 4.42
(m, 4 H, PAr-oH), 3.83 (q, J ) 7 Hz, 8 H, por-CH2CH3), 1.57 (t,
J ) 7 Hz, 12 H, por-CH2CH3), 2.21 (bs, 12 H, por-CH3), 1.48 (s,
36 H, Ar-tBuH) ppm. 31P{1H} NMR (162 MHz, CDCl3): δ -13
ppm.
(22) Tolman, C. A. Chem. ReV. 1977, 77, 313.
(23) Mu¨ller, T. E.; Mingos, D. M. P. Transition Met. Chem. 1995, 20, 533.
(24) Fernandez, A. L.; Reyes, C.; Prock, A.; Giering, W. P. J. Chem. Soc.,
Perkin Trans. 2 2000, 5, 1033.
(25) Mathew, N.; Jagirdar, B. R. Organometallics 2000, 19, 4506.
(26) (a) Taft, R. W. Steric Effects in Organic Chemistry; John Wiley &
Sons: New York, 1956. (b) Taft, R. W. J. Am. Chem. Soc. 1952, 74,
3120.
(27) Hengefeld, A.; Kopf, J.; Rehder, D. Organometallics 1983, 2, 114.
(28) Carty, A. J.; Hota, N. K.; Ng, T. W.; Patel, H. A.; O’Connor, T. J.
Can. J. Chem. 1971, 49, 2706.
(29) Twyman, L. J.; Sanders, J. K. M. Tetrahedron Lett. 1999, 40, 6681.
(30) Carty, A. J.; Hota, N. K.; Ng, T. W.; Patel, H. A.; O’Connor, T. J.
Can. J. Chem. 1971, 49, 2706.
(31) Bharathi, P.; Periasamy, M. Organometallics 2000, 19, 5511. Lang,
H.; Zsolnai, L. J. Organomet. Chem. 1989, 369, 131.
(DPPA)RuII(CO)(DPP) (9). IR (CH2Cl2): 1943 (νCtO) cm-1
.
UV-vis (CH2Cl2): λ (log ꢀ) 413 (5.09), 533 (4.04), 562 (3.27)
(32) Sheldrick, G. M. SHELXS-97; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(33) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(34) Gil, J. M.; Sung, J. W.; Park, C. P.; Oh, D. Y. Synth. Commun. 1997,
27, 3171.
Inorganic Chemistry, Vol. 41, No. 20, 2002 5257

Stulz et al.
Table 1. Crystallographic Data for (DPAP)2Ru(DPP) (12), Triclinic and Monoclinic Form, [(PA)3P]2Ru(DPP) (14), and [PAP(OEt)2]2Ru(DPP) (15)
(DPAP)2Ru(DPP) (12)
triclinic
monoclinic
[(PA)3P]2Ru(DPP)‚4CHCl3 (14)
[PAP(OEt)2]2Ru(DPP) (15)
formula
M
T/K
C
100H106N4P2Ru
C100H106N4P2Ru
1526.90
150(2)
C112H110Cl12N4P2Ru
2100.45
150(2)
C84H106N4O4P2Ru
1398.74
180(2)
1526.90
180(2)
radiation, λ/Å
cryst size/mm
cryst syst
space group
a/Å
Mo KR, 0.707
0.18 × 0.18 × 0.09
triclinic
P1h
11.1136(5)
12.4931(5)
16.7167(7)
71.554(2)
73.964(2)
73.214(2)
2063.8(2)
1
1.299
0.279
25.0
11339
7184
0.032
0.055
0.136
synchrotron, 0.6929
0.06 × 0.04 × 0.02
monoclinic
C2/c
22.920(5)
20.254(5)
18.366(5)
90
102.44(1)
90
8326(4)
4
1.218
0.276
20.3
13361
4335
0.207
0.204
0.506
synchrotron, 0.6923
0.08 × 0.08 × 0.06
triclinic
Mo KR, 0.707
0.30 × 0.23 × 0.18
triclinic
P1h
9.8043(2)
10.7826(3)
18.6293(6)
87.309(2)
77.381(2)
84.103(2)
1911.1(1)
1
1.215
0.298
27.5
21731
8637
0.061
0.054
0.149
P1h
13.8041(7)
13.8388(8)
14.7719(8)
87.730(2)
83.010(2)
67.105(2)
2580.2(2)
1
b/Å
c/Å
R/deg
â/deg
γ/deg
V/Å3
Z
F
calcd/g cm-3
1.352
0.544
29.4
25065
13243
0.026
0.051
0.137
µ/mm-1
max/deg
θ
total data
indep data
Rint
R1 [F2 > 2σ(F2)]
wR2 (all data)
S
1.20
1.82
1.04
1.04
1
nm. H NMR (CDCl3, 400 MHz): δ 9.97 (s, 2 H, por-mesoH),
7.81 (bs, 2 H, por-ArH), 7.70 (bs, 2 H, por-ArH), 7.58 (bs, 2 H,
por-ArH), 7.17 (bs, 2 H, CtCPArH), 7.09 (m, 4 H, CtCPArH),
6.94 (m, 4 H, CtCPArH), 6.76 (bs, 2 H, Ru-PAr-pH), 6.39 (bs, 4
H, Ru-PAr-mH), 4.12 (bs, 4 H, PAr-oH), 3.80 (m, 8 H, por-CH2-
CH3), 2.20 (bs, 12 H, por-CH3), 1.76 (t, J ) 7.1 Hz, 12 H, por-
CH2CH3), 1.50 (s, 18 H, Ar-tBuH), 1.47 (s, 18 H, Ar-tBuH) ppm.
31P{1H} NMR (162 MHz, CDCl3): δ -12 (Ru-PCtC), -32
(CtPPh2) ppm.
(DPAP)2RuII(DPP) (12). UV-vis (CH2Cl2): λ (log ꢀ) 423 (5.44)
nm. H NMR (CDCl3, 400 MHz): δ 9.13 (s, 2 H, por-mesoH),
1
7.55 (s, 2 H, por-ArH), 7.41 (s, 4 H, por-ArH), 7.24 (m, 4 H,
CtCArH), 7.18 (t, J ) 7 Hz, 4 H, CtCArH), 6.95 (d, J ) 7 Hz,
4 H, CtCArH), 6.69 (t, J ) 8 Hz, 4 H, PAr-pH), 6.39 (t, J ) 8
Hz, 8 H, PAr-mH), 4.38 (m, 8 H, PAr-oH), 3.45 (q, J ) 7 Hz, 8
H, por-CH2CH3), 2.13 (s, 12 H, por-CH3), 1.54 (t, J ) 7 Hz, 12 H,
por-CH2CH3), 1.41 (s, 36 H, Ar-tBuH). 31P{1H} NMR (162 MHz,
CDCl3): δ 3 ppm.
[(PA)3P]RuII(CO)(DPP) (10). IR (CH2Cl2): 1955 (νCtO) cm-1
.
(DPPA)2RuII(DPP) (13). UV-vis (CH2Cl2): λ (log ꢀ) 423 (5.25)
1
UV-vis (CH2Cl2): λ (log ꢀ) 412 (4.99), 532 (4.03), 560 (3.52)
nm. 1H NMR (CD2Cl2, 400 MHz): δ 10.10 (bs, 2 H, por-mesoH),
7.99 (s, 2 H, por-ArH), 7.90 (s, 2 H, por-ArH), 7.85 (s, 2 H, por-
ArH), 7.27 (m, 3 H, CtCArH), 7.20 (m, 6 H, CtCArH), 6.96 (d,
J ) 8 Hz, 6 H, CtCArH), 3.73 (bs, 8 H, por-CH2CH3), 2.53 (bs,
12 H, por-CH3), 1.85 (t, J ) 8 Hz, 12 H, por-CH2CH3), 1.56 (s, 36
H, Ar-tBuH) ppm. 31P{1H} NMR (162 MHz, CD2Cl2): δ -59 ppm.
nm. H NMR (CDCl3, 500 MHz): δ 9.54 (s, 2 H, por-mesoH),
7.55 (t, J ) 2 Hz, 2 H, por-ArH), 7.14 (d, J ) 2 Hz, 4 H, por-
ArH), 7.08 (t, J ) 9 Hz, 2 H, CtCPAr-pH), 7.02 (dt, J ) 9, 1 Hz,
4 H, CtCPAr-mH), 6.93 (dt, J ) 9, 1 Hz, 4 H, CtCPAr-oH),
6.69 (t, J ) 9 Hz, 4 H, Ru-PAr-pH), 6.32 (t, J ) 9 Hz, 8 H,
Ru-PAr-mH), 4.31 (m, 8 H, Ru-PAr-oH), 3.54 (q, J ) 9 Hz, 8 H,
por-CH2CH3), 1.94 (s, 12 H, por-CH3), 1.65 (t, J ) 9 Hz, 12 H,
por-CH2CH3), 1.42 (s, 36 H, Ar-tBuH). 13C NMR (125.70 MHz,
CDCl3): δ 148.1, 144.5, 143.0, 140.9, 140.3, 137.2, 135.6, 135.5,
138.1, 131.9, 129.2, 128.3, 128.1, 128.0, 127.0, 125.8, 119.7, 119.4,
104.3, 100.0, 35.4, 32.2, 20.1, 18.4, 15.0. 31P{1H} NMR (162 MHz,
CDCl3): δ 3 (Ru-PCtCPPh2), -30 ((Ru-PCtCPPh2) ppm.
[(PA)3P]2RuII(DPP) (14). UV-vis (CH2Cl2): λ (log ꢀ) 418
[PAP(OEt)2]RuII(CO)(DPP) (11). IR (CH2Cl2): 1958 (νCtO
)
cm-1. UV-vis (CH2Cl2): λ (log ꢀ) 412 (5.01), 532 (3.97), 561
(3.22) nm. 1H NMR (CDCl3, 400 MHz): δ 9.74 (s, 2 H,
por-mesoH), 7.99 (s, 2 H, por-ArH), 7.74 (s, 2 H, por-ArH), 7.70
(s, 2 H, por-ArH), 7.16 (t, J ) 8 Hz, 2 H, CtCArH), 7.24 (m, 2
H, CtCArH), 6.80 (d, J ) 7 Hz, 2 H, CtCArH), 3.80 (m, 8 H,
por-CH2CH3), 2.16 (s, 12 H, por-CH3), 1.67 (t, J ) 8 Hz, 12 H,
por-CH2CH3), 1.51 (s, 18 H, Ar-tBuH) ppm, 1.41 (s, 18 H,
Ar-tBuH), 0.92/0.78 (m, 2 × 2 H, PO-CH2CH3), -0.18 (t, J ) 7
Hz, PO-CH2CH3). 31P{1H} NMR (162 MHz, CDCl3): δ 108 ppm.
Synthesis of Bis-Phosphino Complexes (PR3)2RuII(DPP).
General Procedure. 2 (50 mg, 49.3 µmol) was suspended in CHCl3
(5 mL), and neat phosphine (5 equiv) was added. The solution was
stirred at room temperature under Ar atmosphere for 15 min, and
the solvent was removed on a rotary evaporator. The residue was
redissolved in 5 mL of CHCl3 and stirred at room temperature for
10 min., and the solvent was removed in vacuo. The orange solid
was dissolved in a minimum amount of hot CHCl3 (ca. 2 mL), and
10 mL of MeOH was carefully layered over the solution. After
standing overnight, orange to bronze colored crystals were collected
on a sintered funnel, washed with MeOH, and dried in vacuo. Yields
are in the range 90-98%.
1
(5.37) nm. H NMR (CD2Cl2, 500 MHz): δ 9.53 (s, 2 H, por-
mesoH), 7.60 (s, 4 H, por-ArH), 7.58 (s, 2 H, por-ArH), 7.16 (t, J
) 7 Hz, 2 H, CtCAr-pH), 7.10 (t, J ) 8 Hz, 4 H, CtCAr-mH),
6.89 (d, J ) 7 Hz, 4 H, CtCAr-oH), 3.52 (q, J ) 7 Hz, 8 H,
por-CH2CH3), 2.24 (s, 12 H, por-CH3), 1.44 (t, J ) 7 Hz, 12 H,
por-CH2CH3), 1.17 (s, 36 H, Ar-tBuH). 13C NMR (125.70 MHz,
CDCl3): δ 148.9, 144.7, 143.2, 141.2, 140.3, 140.0, 137.3, 137.2,
131.8, 128.9, 127.8, 120.9, 120.7, 100.8, 98.6, 34.5, 31.1, 19.8,
16.9, 15.0. 31P{1H} NMR (162 MHz, CDCl3): δ -50 ppm. This
complex is soluble in CH2Cl2, but not in CHCl3.
[PAP(OEt)2]2RuII(DPP) (15). UV-vis (CH2Cl2): λ (log ꢀ) 419
1
(5.64) nm. H NMR (CDCl3, 400 MHz): δ 9.33 (bs, 2 H, por-
mesoH), 7.75 (s, 4 H, por-ArH), 7.66 (s, 2 H, por-ArH), 7.20 (t, J
) 7 Hz, 2 H, CtCAr-pH), 7.12 (t, J ) 7 Hz, 4 H, CtCAr-mH),
6.83 (d, J ) 7 Hz, 4 H, CtCAr-oH), 3.63 (q, J ) 7 Hz, 8 H,
por-CH2CH3), 2.23 (s, 12 H, por-CH3), 1.55 (t, J ) 7 Hz, 12 H,
5258 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
Chart 3
por-CH2CH3), 1.40 (s, 36 H, Ar-tBuH), 0.96 (t, J ) 7 Hz, 2 H,
PO-CH2CH3), 0.84 (t, J ) 7 Hz, 2 H, PO-CH2CH3), -0.18 (t, J )
7 Hz, PO-CH2CH3).13C NMR (125.70 MHz, CDCl3): δ 148.7,
144.2, 144.0, 143.2, 140.7, 139.9, 136.5, 131.7, 131.6, 128.4, 127.6,
121.4, 121.1, 119.9, 98.5 (d), 58.4, 34.8, 31.5, 19.6, 17.7, 15.1,
14.4. 31P{1H} NMR (162 MHz, CDCl3): δ 114 ppm.
UV-Vis Titration of (L)RuII(CO)(DPP). The following stock
solutions were prepared in degassed methylene chloride: 2, 10-4
M; phosphine, 1.5 mM. The titrations were performed in a 0.1 mm
quarz cuvette, using 300 µL of porphyrin solution. The phosphine
was added in 2 µL aliquots (0.1 equiv) using a gastight syringe,
until 1.5 equiv were added. The absorbance was measured in the
range 350-650 nm after 2 min of equilibration. The association
constants were determined using the changes in the molar extinction
coefficient at 400, 411, 520, 534, and 551 nm; KA and xSL were
determined by a least-squares fitting of the data according to eq
1,35
Figure 1. 1H NMR spectra of (a) PAP(OEt)2 (6), (b) [PAP(OEt)2]RuII-
(CO)(DPP) (11), and (c) [PAP(OEt2)]2RuII(DPP) (15). Dotted lines indicate
characteristic chemical shift differences upon ligand binding. Solvent peaks
are marked by asterisks (*). Spectra are recorded in CDCl3 (400 MHz) at
25 °C.
solution, simply by mixing equimolar amounts of phosphine
and (MeOH)RuII(CO)(DPP) (Chart 3). All compounds are
soluble in chloroform up to 10 mM concentrations, except
for [(PA)3P]RuII(CO)(DPP). This complex precipitated im-
mediately after addition of 7 to (MeOH)RuII(CO)(DPP) in
CHCl3, but a freshly prepared sample retained its solubility
xobsd ) xS + 0.5(xSL - xS)([S]t + [L]t +
KD - {([S]t + [L]t + KD)2} - 4[S]t[L]t)1/2)/[S]t (1)
1
in CH2Cl2. The H NMR spectra showed the expected 1:1
where x ) molar extinction coefficient (xobsd ) Abs/[S]), S ) 2,
SL ) mono-phosphine complex, [X]t ) total concentration, and
KD ) dissociation constant. For the calculations of xobsd, an
appropriate dilution factor was taken into account.
ratio of the ligand to (MeOH)RuII(CO)(DPP). Figure 1
1
displays the H NMR spectra of PAP(OEt)2, [PAP(OEt)2]-
RuII(CO)(DPP), and [PAP(OEt2)]2RuII(DPP) as representative
examples. Characteristic upfield shifts of the phosphorus
phenyl substituents [∆δ ) -0.4 ppm (p-H), -0.8 ppm (m-
H), -2.8 ppm (o-H)] and phosphorus ethoxy substituents
(∆δ ) -3.19 ppm for Hd, -1.49 ppm for He, Figure 1)
confirm the proximity of these groups to the shielding
magnetic field of the porphyrinic macrocycle. For the
acetylenic phenyl protons, upfield shifting is much less
pronounced (∆δ -0.1 to -0.5 ppm, Ha,b,c in Figure 1), as
would be expected from the larger distance to the porphyrin
core. Lack of a mirror plane in the porphyrin macrocycle
renders the R- and â-side of the porphyrin inequivalent, and
is expressed in the non-isochronic shifts of the 3′- and 5′-
tert-butyl groups (δ ∼ 1.5 ppm) and of the 2′- and 4′-ortho-
protons on the meso-phenyl substituents (δ ∼ 7.7 ppm). In
[PAP(OEt)2]RuII(CO)(DPP) (Figure 1), the ethoxy side
chains showed additional splitting of the diastereotopic
methylene protons into two signals of equal integral value
at δ ) 0.92 and 0.78 ppm.
UV-Vis Titration of (L)2RuII(DPP). The bis-phosphine com-
plexes were prepared as 1 mM stock solutions in oxygen-free
methylene chloride. Using a 0.1 mm quartz cuvette, 12 µL of the
stock solution was added in 2 µL aliquots to 300 µL of methylene
chloride, and the absorption spectra were measured after each
addition (2 min of equilibration). The compositions of the mixtures
were evaluated using baseline correction and peak deconvolution
(Gaussian peak function) at ∼400, ∼411, and ∼423 nm, according
to the complex measured. Using the known total concentration, and
the extinction coefficients for 2 and the mono-phosphine complexes,
the distribution was calculated at each concentration according to
equilibria 2 and 3, using [2] ) 0.5[phosphine]. The determined
dissociation constants are an average of all individual concentra-
tions.
RuII(DPP) + PR3 a (PR3)RuII(DPP)
(2)
(3)
(PR3)RuII(DPP) + PR3 a (PR3)2RuII(DPP)
Results
Further characteristic chemical shifts were observed in the
31P{1H} NMR spectra (Table 2). Binding of the first
phosphorus to ruthenium induces a chemical shift difference
∆δ, which is +26 ppm for [(PA)3P]RuII(CO)(DPP), +20
ppm for (DPPA)RuII(CO)(DPP) and (DPAP)RuII(CO)(DPP),
and -22 ppm for [PAP(OEt)2]RuII(CO)(DPP) (Figure 2). The
chemical shift difference for PPh3 upon binding is only
Synthesis and NMR Spectroscopy of Mono-Phosphine
Complexes. The mono-phosphine complexes (DPAP)RuII-
(CO)(DPP), (DPPA)RuII(CO)(DPP), [(PA)3P]RuII(CO)(DPP),
and [PAP(OEt)2]RuII(CO)(DPP) are readily synthesized in
(35) Schneider, H.-J.; Yatsimirsky, A. Principles and Methods in Supramo-
lecular Chemistry; Wiley-VCH: Weinheim, Germany, 2000.
Inorganic Chemistry, Vol. 41, No. 20, 2002 5259

Stulz et al.
Table 2. Spectroscopic Data for Ru(CO)(DPP) and the Mono- and Bis-Phosphorus Complexes
IR νCtO/cm-1 a
UV-vis/λmax (log ꢀ)a
Ru(CO)(DPP) 2
1919
400 (5.30); 520 (4.29); 551 (4.25)
31P NMR/δ ppm:b
Ka/10-5 M-1
free ligand
UV-vis/λmax (log ꢀ):a
(PR3)Ru(CO)(DPP)
(PR3)2Ru(DPP)
(PR3)Ru(CO)(DPP)
(PR3)2Ru(DPP)
IR νCtO/cm-1 a
:
c
d
e
f
ligand PR3
(PR3)Ru(CO)(DPP)b
1943
pKa
Ka1
Ka2
Ka3
DPPA 3
-32
-12/-32
3/-32
-4
413 (5.09); 533 (4.04); 562 (3.27)
423 (5.25)
12 ( 4
0.12 ( 0.01
12 ( 3
2.5 ( 0.8
37 ( 7
PPh3 4
1937
1943
1958
1955
2.73
1.04
411 (5.24); 532 (4.31); 559 (3.99)
-7
ndg
DPAP 5
-32
-12
3
130
108
413 (5.12); 533 (4.08); 560 (3.52)
423 (5.44)
1.5 ( 0.7
10 ( 3
19 ( 4
PAP(OEt)2 6
(PA)3P 7
6.11
412 (5.01); 532 (3.97); 561 (3.22)
419 (5.64)
48 ( 2
0.95 ( 0.05
1.5 ( 0.5
114
-85
-59h
-50h
-2.96
412 (4.99); 532 (4.03); 560 (3.52)
418 (5.37)
1.3 ( 0.5
a In CH2Cl2, c ) 2.0 mM. b In CDCl3. c pKa values are calculated according to literature procedures.53 d Ru(CO)(DPP) + PR3 / (PR3)Ru(CO)(DPP).
e Ru(DPP) + PR3 / (PR3)Ru(DPP). f (PR3)Ru(DPP) + PR3 / (PR3)2Ru(DPP). g Not determined. h In CD2Cl2.
marginal (∆δ ∼ -3 ppm), accompanied with substantial
broadening of the resonance which indicates rapid ligand
exchange on the NMR time scale. Most of the phosphorus
resonances are also broadened upon complexation, except
for [PAP(OEt)2]RuII(CO)(DPP). For (DPPA)RuII(CO)(DPP),
two resonances can be found in the 31P{1H} NMR spectrum
at characteristic values for bound and free phosphine. Using
only 0.5 equiv of DPPA to ruthenium porphyrin produced a
31P{1H} NMR spectrum identical to the 1:1 mixture; thus
DPPA does not act as a bridging ligand to complex two
ruthenium porphyrins.
between the para protons Hc T Hf, which are more than 5
Å apart, are not expected. Thus the two phosphines in DPPA
are exchangeable upon complexation to ruthenium at a rate
which is slow on the chemical shift time scale, but fast on
the NOE and relaxation (R1) time scale.36 Whether this
exchange is a simple dissociation-reassociation process37,38
or if an alkyne bound transition state is present39 remains
unclear. A similar kinetic lability was also observed in other
ruthenium(II)25 and platinum(II) phosphine complexes on the
basis of 31P{1H} NMR spectroscopy40 and explains the
difficulties encountered in isolating the mono-phosphine
complexes. To our knowledge, such an exchange has not
previously been directly observed using 2D NMR spectros-
copy.
IR Spectroscopy. Also summarized in Table 2 are
ruthenium CtO triple bond IR stretching frequencies. The
spectra were recorded using 2.0 mM solutions in methylene
chloride. νCtO in (PR3)RuII(CO)(DPP) is shifted to higher
wavenumbers compared to (MeOH)RuII(CO)(DPP), the
differences varying from 18 cm-1 for PPh3 to 39 cm-1 for
PAP(OEt)2 (Figure 4). The relative intensities of the IR
absorptions decrease with increasing ν, and the half-widths
of the peaks increase with the same trend. The νCtO band in
(MeOH)RuII(CO)(DPP) shows a half-width of 19 cm-1,
while the band corresponding to [PAP(OEt)2]RuII(CO)(DPP)
has a half-width of 31 cm-1 and a 50% decreased intensity.
(DPAP)RuII(CO)(DPP) and (DPPA)RuII(CO)(DPP) displayed
the same IR spectra in the νCtO region. νCtO for the parent
RuII(CO)(DPP) is independent on whether the bound ligand
The mono-phosphine complexes could not be isolated.
NMR spectroscopic analysis showed that evaporation of the
solvent using a rotary evaporator, or crystallization from the
reaction mixture, always yielded the bis-phosphine com-
plexes together with unreacted (MeOH)RuII(CO)(DPP) in
equimolar amounts. The use of carbon monoxide saturated
solvents for crystallization gave mainly the mono adducts,
but these were always contaminated with 5-15% of the bis-
phosphine complexes. All complexes are stable in solution
under argon or CO for 1 week. Upon bubbling air through
the solutions, significant amounts of phosphine oxides were
observed after 2 days, as judged by 31P{1H} NMR spectros-
copy. The resonances of the phosphine oxides show a
characteristic downfield shift as compared to the phosphine,
e.g., ∆δ ) +41 ppm for the oxidized DPAP; contrarily, the
phosphonate resonance is shifted upfield by 135 ppm
compared to the phosphonite.
1
Broadening of the H and 31P{1H} NMR signals of the
ligands in (PR3)RuII(CO)(DPP) and UV-vis spectroscopy
indicates a dynamic ligand exchange in both mono- and bis-
phosphorus complexes. This was confirmed first by exchange
experiments involving mixtures of different phosphines in
solution, e.g., exchange of (PA)3P with DPAP on (PR3)RuII-
(CO)(DPP) (data not shown), and second by diagnostic
NOESY exchange peaks of the phenyl proton signals in
(DPPA)RuII(CO)(DPP) (Figure 3.) Through-space NOEs
(36) Sanders, J. K. M.; Hunter, B. K. Modern NMR Spectroscopy, 2nd
ed.; Oxford University Press: Oxford, 1993.
(37) James, S. L.; Lozano, E.; Nieuwenhuyzen, M. Chem. Commun. 2000,
617.
(38) Clark, H. C.; Ferguson, G.; Kapoor, P. N.; Parvez, M. Inorg. Chem.
1985, 24, 3924.
(39) (a) Lee, H. M.; Bianchini, C.; Jia, G. C.; Barbaro, P. Organometallics
1999, 18, 1961. (b) Ihmels, K.; Rehder, D. Organometallics 1985, 4,
1334.
(40) Romeo, R.; Alibrandi, G. Inorg. Chem. 1997, 36, 4822.
5260 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
Figure 3. Aromatic region of the NOESY spectrum of (DPPA)RuII(CO)-
(DPP) in CDCl3. The signal of Hg in the one-dimensional spectrum is
masked by the CHCl3 resonance.
Figure 4. IR spectra of the mono-phosphine complexes (MeOH)RuII(CO)-
(DPP) (-‚-‚), (PPh3)2RuII(DPP) (- - -), [(PA)3P]2RuII(DPP) (-‚‚-‚‚),
(DPAP)2RuII(DPP) and (DPPA)2RuII(DPP) (s), and [PAP(OEt2)]2RuII(DPP)
(---). Shown is the wavenumber region of the carbonyl stretching absorption.
Spectra are recorded in CH2Cl2 at 2.0 mM concentration of (MeOH)RuII-
(CO)(DPP) and ligand.
Figure 2. 31P{1H} NMR spectra of the free ligands (top), mono-phosphine
complexes (middle), and bis-phosphine complexes (bottom): (a) (PA)3P;
(b) DPAP; (c) PPh3; (d) PAP(OEt)2. Spectra are recorded in CDCl3 (162
MHz) at 25 °C.
is MeOH or THF. As the ruthenium porphyrin is crystallized
from CHCl3-MeOH after synthesis, and obtained as the
MeOH solvate, this additional ligand is therefore readily
displaced, and does not interfere with the binding of the
phosphorus ligand. Analogous observations were made
previously when pyridine ligands were complexed to Ru(II)
porphyrins.41
Synthesis and NMR Spectroscopy of Bis-Phosphine
Complexes. The bis-phosphine complexes (DPAP)2RuII(DPP),
(DPPA)2RuII(DPP), [(PA)3P]2RuII(DPP), and [PAP(OEt2)]2-
RuII(DPP) (Chart 3) are easily accessible by mixing 5 equiv
of the phosphine with (MeOH)RuII(CO)(DPP) and evaporat-
ing the solvent. Extensive boiling in CHCl3 or toluene as
for the PPh3 complexes is not required. Recrystallization from
hot chloroform-methanol mixtures gave the complexes in
high yield and purity.
1
The H NMR shifts of the ligands are similar to those of
the mono-phosphine complexes, but are generally sharper
and better resolved (Figure 1). On the porphyrin moiety, shift
differences are observed for the meso-protons (∆δ 0.41 to
0.59 ppm, Hm in Figure 1), and for the meso-phenyl
substituents (Ho,p in Figure 1). Due to the higher symmetry
compared to the mono-phosphine complexes, the resonances
of the 3′- and 5′-tert-butyl groups and of the 2′- and 4′-ortho-
protons on the meso-phenyl substituents become isochronous.
The phosphorus chemical shifts are significantly different
in the bis-phosphine complexes compared to the mono
complexes. The shift differences (Table 2), compared to the
free ligands, are +35 ppm for [(PA)3P]2RuII(DPP), (DPAP)2-
RuII(DPP), and (DPPA)2RuII(DPP), but -16 ppm for [PAP-
(OEt2)]2RuII(DPP) (Figure 2). Also, the signals are sharper
than for the monomeric counterparts, except for [PAP-
(OEt2)]2RuII(DPP), which showed a broadened signal.
(41) (a) Webb, S. J.; Sanders, J. K. M. Inorg. Chem. 2000, 39, 5920. (b)
Webb, S. J.; Sanders, J. K. M. Inorg. Chem. 2000, 39, 5912.
Inorganic Chemistry, Vol. 41, No. 20, 2002 5261

Stulz et al.
Figure 5. Molecular units in the bis-phosphine complexes showing displacement ellipsoids at the 50% probability level. For (DPAP)2RuII(DPP), only the
triclinic structure of the polymorphic structures is shown. For (DPPA)2RuII(DPP), only one of the disordered geometries of the unbound phosphine group
is displayed. H atoms are omitted for clarity.
X-ray Structures. The bis-phosphine complexes formed
deep orange to bronze colored single crystals overnight from
chloroform solutions layered with methanol. All complexes
crystallize with the Ru atom sited at a crystallographic center
of symmetry such that the molecular units are centrosym-
metric. [(PA)3P]2RuII(DPP), [PAP(OEt)2]2RuII(DPP), and
(DPAP)2RuII(DPP) crystallize in space group P1h. The crystals
of [(PA)3P]2RuII(DPP) also contain chloroform molecules
(four per molecular unit). Two of the three structures,
[(PA)3P]2RuII(DPP) and [PAP(OEt)2]2RuII(DPP), display
disorder of the tert-butyl groups, and these were modeled in
two orientations with a single isotropic displacement pa-
rameter common to all C atoms. (DPPA)2RuII(DPP) crystal-
lizes in the orthorhombic space group Pccn with four
molecules in the unit cell.62 The phenyl substituents of the
DPPA unit display some disorder, and these were modeled
in two orientations. The four molecular units of the com-
plexes are shown in Figure 5.
For (DPAP)2RuII(DPP), the majority of crystals were large
orange blocks (the triclinic form), but small orange plates
of a second polymorph (the monoclinic form) were also
Figure 6. Disordered molecular units of the monoclinic crystal structure
observed in the bulk material. The crystals of this monoclinic
form were small and weakly diffracting, but it was possible
to determine their structure using a synchrotron radiation
source. This polymorph crystallizes in the space group C2/c
with four molecules in the unit cell. The data are relatively
poor, however (Rint ca. 20%), and refinement of the structure
was problematic; it was possible only to refine all atoms
with isotropic displacement parameters. The alkynylphenyl
of (DPAP)2RuII(DPP). Isotropic spheres are shown at the 30% probability
level. H atoms are omitted.
substituents of the DPAP unit display some disorder and were
modeled in two orientations with the rings constrained to
be true hexagons (Figure 6).
Calculated geometrical parameters are summarized in
Table 3. The geometrical parameters (Figure 7) of the
5262 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
Table 3. Selected Geometrical Data for the Solid State Structures of the Bis-Phosphorus Complexesa
(DPAP)2RuII(DPP) (12)
(DPPA)2RuII(DPP) (13)
monoclinic
P unbound
[(PA)3P]2-
[PAP(OEt2)]2-
triclinic
0.026
2.061(3)
2.063(3)
81.9
2.3623(10)
89.50
132.1
isomer 1 isomer 2 P bound conformer 1 conformer 2
RuII(DPP) (14)
RuII(DPP) (15)
σ/Å
0.067
2.067(14)
2.079(13)
73.8
2.340(5)
77.4
0.047
0.042
2.0642(17)
2.0621(16)
86.2
2.297(5)
87.64
144.9
145
0.038
Ru-N1/Å
2.063(4)
2.053(2)
79.2
2.334(2)
84.75
2.067(2)
2.068(2)
82.8
2.3140(8)
85.47
127.9
121
1.774(4)
34.8
1.592(2)
Ru-N2/Å
Φ(Por-Armeso)/deg
d(Ru-P)/Å
γ(RuP-por)/deg
Ω/deg
θ/deg
l(P-Ct)/Å
Φ(RuPC-ArCt)/deg
l(P-CPh1/O1)/Å
Φ(RuPC-Ph1)/deg
Φ(O2-P-O1-C)/deg
l(P-CPh2/O2)/Å
Φ(RuPC-Ph2)/deg
Φ(O1-P-O2-C)/deg
æ(P-CtC)/deg
127.7
145
126.3
124.6
145
1.779(3)
121.7
1.757(4)
72.9
1.837(4)
1.758(18)
48.2
1.820(11)
1.776(3)
1.827(5)
1.759(2); 1.759(2); 1.755(2)
7.8; 13.1; 46.7
71.1
1.834(1)
55.2
1.955(4)
86.4
1.870(2)
74.3
76.3
74.6
1.862(14)
68.3
-55.6
1.643(3)
61.3
1.830(4)
58.2
1.827(2)
74.9
1.859(3)
57.3
1.849(2)
63.5
169.1(4)
170.1(17)
172.9
173.6
177.10(19); 176.4(2); 175.1(2)
173.5
a For assignments, see text and Figure 7.
(MeOH)RuII(CO)(DPP) + PR3 a (PR3)RuII(CO)(DPP) (4)
where PR3 is the corresponding phosphine. Using eq 1 (see
Experimental Section), a nonlinear least-squares fitting
procedure was applied to the changes in the molar extinction
coefficient at 400, 411, 520, 534, and 551 nm. Figure 8
shows the curves obtained for the decrease at 520 nm. Due
to possible partial loss of CO during the titrations, the
standard deviations shown in Table 2 may underestimate the
uncertainty in the binding constant determination. An excep-
tion was found when titrating PPh3 to (MeOH)RuII(CO)-
(DPP): the spectra showed isosbestic points when using 10-5
M (MeOH)RuII(CO)(DPP) solutions, and the first binding
constant was determined via standard calculations (Table 2).
Figure 7. Definition of selected geometrical parameters. Φ indicates the
dihedral angle between the least-squares planes (σ1, σ2) schematically shown
as rectangles. σ1 is defined by the atoms Ru, P, and C(acetylene); σ2 is
defined by the aryl group attached to the acetylene unit.
phosphines such as the Ru-P bond lengths (d), the angle
between the Ru-P bond and the porphyrin plane (γ), the
crystallographic cone angles (Ω), and dihedral angles (Φ)
between the plane Ru-P-C (σ1) and the alkynylphenyl mean
plane (σ2) show characteristic values for the individual
structures; these are discussed below. Ω values were
determined according to Mingos’s procedure,23 but using the
true crystallographic Ru-P bond lengths (d) instead of a
normalized value.
The second binding constants were determined using
dilution experiments of the bis-phosphine complexes. After
a baseline correction, a Gaussian deconvolution of the
absorption bands was possible when assuming that three
different species with absorption maxima at 400 nm [RuII-
(DPP)], ∼411 nm [(PR3)RuII(DPP)], and ∼420 nm [(PR3)2-
RuII(DPP)] are present (Figure 9). For [(PA)3P]RuII(CO)-
(DPP), the UV-vis spectra could not be deconvoluted using
the three-component model, but assuming the presence of
only the mono- and bis-phosphine complexes. Only the
B-band absorptions could be used in the calculations, because
the low-energy absorption of the bis-complexes is very broad,
and no distinct Q-band absorptions could be observed. This
is discussed in the following paper.61
The maxima assigned to RuII(DPP) and (PR3)RuII(DPP)
were found at the same values as were obtained by direct
measurement of the corresponding CO-containing complexes.
Under the assumption that the dissociated species have
similar extinction coefficients as the CO-containing com-
plexes (based on the same λmax values), and from the known
total concentration of porphyrin, the dissociation constants
were calculated according to equilibria 2 and 3 (see
Experimental Section).
UV-Vis Spectroscopy and Binding Constants. Standard
UV-vis titration methods (10-6 M solutions) to determine
directly the first and second binding constants using the
B-band absorption (∼400 nm) as well as the Q-band
absorptions (500-600 nm) failed for these systems, because
no isosbestic points could be observed, and the data-fitting
was not reproducible. The reason for this behavior is
irreversible loss of CO upon addition of a large excess of
ligand, leading to a complex equilibrium system according
to Scheme 1 (vide infra). To determine the first association
constant, the use of rather concentrated solutions (10-4
M
2), and titrating 0.1 equiv of phosphine, showed a hypo-
chromic shift in the absorption maxima, in both the B-band
and Q-band regions (see Table 2). Isosbestic points were
observed when no more than 1.2 equiv of phosphine was
added. The data were fitted to a model assuming a single
binding event, according to eq 4,
Inorganic Chemistry, Vol. 41, No. 20, 2002 5263

Stulz et al.
Scheme 1
Discussion
Despite their lability, the mono-phosphine complexes are
the kinetic products, as addition of up to 1.2 equiv of
phosphine did not yield detectable amounts of higher
complexes after 2 days (31P{1H} NMR.) Only after heating
was formation of the bis-complex detected. Extensive
bubbling of CO through solutions of (PR3)2RuII(DPP) in
chloroform did not result in a clean back-conversion to (PR3)-
RuII(CO)(DPP) as occurs with ferrous porphyrins.11 We
therefore conclude that, in solution, all complexes exist in a
dynamic equilibrium according to Scheme 1.
Solid State Structures and Cone Angles. The crystal
structures of the bis-phosphine complexes, as shown in
Figure 5, allow direct comparison of the differences in
geometry upon variation of the phosphorus substituents.
Common to all structures is the centrosymmetry (i ) Ru).
Even though significant ruffling could occur because of the
strong MfP π-back-bonding,10 the porphyrin cores display
relatively little ruffling. The porphyrin planes are essentially
flat, the mean deviation σ from planarity being in the range
of 0.026 Å for the triclinic form of (DPAP)2RuII(DPP) to
0.067 Å for the monoclinic form of (DPAP)2RuII(DPP). The
mean plane was defined using all 24 atoms of the porphyrin
core and the Ru2+ center. Tilting of the meso-phenyl groups
in all structures is within a small range of about 10° from
orthogonality. Such an arrangement minimizes steric interac-
tions with the methyl groups at the â-pyrrole positions in
the solid state.
The acetylene substituent of the ligands has a tendency to
adopt an approximately eclipsed orientation (ligand over
nitrogen), especially in (DPAP)2RuII(DPP) (both monoclinic
and triclinic forms), [(PA)3P]2RuII(DPP), and [PAP(OEt2)]2-
RuII(DPP). One of the remaining substituents (phenyl,
alkynylphenyl, ethoxy) in these structures is also eclipsed.
The ligands are oriented such that the alkynylphenyl group
is located toward the unsubstituted meso-carbon. Only in
(DPPA)2RuII(DPP) does the phosphine adopt a clear stag-
gered conformation over the free meso position, the phos-
phine phenyl groups being located directly above the pyrrole
nitrogens. The configurations of the ethoxy groups in [PAP-
(OEt2)]2RuII(DPP) are similar to those found in Fe(III)
porphyrin complexes of PhP(OEt)2.10 As the 1H NMR spectra
indicate, however, the ligands have no preferred geometry
in solution.
Synthesis and Reactivity in Solution. The rapid com-
plexation of phosphines and phosphonates to RuII(CO)
porphyrins has been reported earlier.15,17,18,21,42 Upon mixing
equimolar amounts of alkynyl phosphorus ligands with
(MeOH)RuII(CO)(DPP), 1H and 31P{1H} NMR spectra
showed rapid and complete formation of the mono-
phosphorus complexes in the millimoles per liter range. The
spectra are significantly different from those obtained with
PPh3; this ligand does not bind strongly to (MeOH)RuII(CO)-
(DPP) (Ka ) 1.2 × 104 M-1), as is clear from the broadened
31P{1H} NMR resonances (Figure 2). PPh3 complexes of RuII-
(OEP) (Ka ) 8.3 × 104 M-1) and RuII(TPP) (Ka ) 7.4 ×
103 M-1) were also reported to dissociate in solution.15 The
bis-phosphine complexes are readily accessible using an
excess of ligand and evaporating the solvent. A single
crystallization step gives the pure product in high yield. The
ligands therefore weaken the trans-CO sufficiently that it
can be removed simply by applying low pressure. This is in
contrast to triaryl phosphines, for which prolonged heating
in CHCl3 or toluene is required to form the bis-complexes.
1
The H NMR spectra for (PR3)RuII(CO)(DPP) and (PR3)2-
RuII(DPP) were very similar with respect to the chemical
shifts, but the 31P{1H} NMR spectra allowed unambiguous
assignment of the complexes due to their distinctive δ values.
Rotational flexibility around the Ru-P bond and P-C
bonds in solution, judged from the equivalence of the
â-pyrrole substituents and the phosphorus phenyl substituents
in the 1H NMR spectra, leads the complexes to have higher
C2V symmetry (mono-P complex) or D2h symmetry (bis-P
complex) than the crystallographic C1 or Ci symmetry,
respectively. This higher symmetry has no effect on the
inequivalency of the R- and â-sides of the porphyrin in the
(PR3)RuII(CO)(DPP) complexes, which is expressed in
different chemical shifts of the meso-phenyl protons.7,43
NOESY exchange peaks Hg T Hi (Figure 3) indicate,
however, that slow spinning with respect to the NMR
chemical shift time scale around the meso-aryl bond is
occurring in solution.
(42) Cheng, P. C.; Liu, I. C.; Hong, T. N.; Chen, J. H.; Wang, S. S.; Wang,
S. L.; Lin, J. C. Polyhedron 1996, 15, 2733.
(43) Abraham, R. J.; Marsden, I. Tetrahedron 1992, 48, 7489.
5264 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
Even though variations in Ω of about 30° for a specific
phosphine ligand in variable environments are common, the
mean values of all Ω’s usually coincide with θ.23 The θ
values of our ligands show that introducing an alkyne spacer
between the phosphorus and the aryl substituent does not
alter the half-angle of the ligand. DPAP, DPPA, and (PA)3P
display the same calculated steric demand as PPh3 (θ )
145°), which clearly is not reflected in the thermodynamic
stability of the complexes, because the alkyne phosphines
display a much higher association constant compared to PPh3.
Only PAP(OEt)2 has a smaller θ and a correspondingly larger
Ka1 value.
Systematic Cambridge structural database searches47 re-
vealed that crystal packing forces may lead to distortions,
and standard deviations (σ) of M-P bond lengths and torsion
angles Φ(M-P-Ph) (Figure 7) can be as high as 0.02 Å
and >40°, respectively. The Ru-P bond lengths d in our
structures are within a range of (0.03 Å. Inspection of the
dihedral angles Φ in the different structures (Table 3) shows
that Φ for the phosphine phenyl substituents varies from
55.2° to 86.4°. The alkynylphenyl substituents have an even
greater flexibility, showing values from 7.76° to 72.94°. As
θ is influenced by both Φ(PPh) and Φ(PCtCPh), and as Φ
has an apparently independent variation from the overall
relative conformation, Ω varies accordingly. For (DPAP)2-
RuII(DPP) and (DPPA)2RuII(DPP), Ω values are about 15-
20° smaller than θ; for [(PA)3P]2RuII(DPP) and [PAP(OEt2)]2-
RuII(DPP) the calculated and measured values match rea-
sonably well.
Figure 8. Decrease of porphyrin molar extinction coefficient at λ ) 520
nm, obtained from titration experiments; CH2Cl2, 25 °C, [2] ) 10-4 M.
Figure 9. B-band absorption region of (DPAP)2RuII(DPP) (s) and
deconvolution (- - -) after baseline correction; CH2Cl2, 25 °C, c ) 2.6
× 10-5 M.
Stereoelectronic Effects. Stereoelectronic effects have
strong influence on stability and reactivity of transition metal
phosphorus complexes, and a fine balance between σ-dona-
tion and π-acceptance defines the overall physical properties.
Stereoelectronic effects are influenced by both the type of
the substituents on the phosphorus (electronegativity, π-de-
localization) and the dihedral angles C-P-C, reflecting
steric effects of the ligand (θ). The qualitative influence of
σ/π- and θ-parameters on the reactivity of phosphorus ligands
remains hotly debated.46,48-50 To date, there is little informa-
tion available from spectroscopic studies involving acetylenic
phosphines,27,28 and only DPPA has gained significant
attention because of its ability to act as a bridging ligand to
form cage type structures.37,44,51,52
The acetylene units show deviation from linearity with
torsion angles æ(P-CtC) varying from 169.1(4)° to 177.10-
(19)° (Table 3). Such deviation is usually observed in µ2-
DPPA complexes and stems from strain or steric crowd-
ing.37,38,44 In the complexes considered here, the deviation
might arise from crystal packing forces, as the CtCPh units
are bent toward the aromatic macrocycle. The Ru-P bond
lengths d are clearly shorter [2.297(5)-2.3623(10) Å] than
those found for Ru-PPh3 complexes, which have an average
value of 2.428 Å in TPP and OEP.15 This is in agreement
with the higher binding constants obtained for the alkynyl
phosphorus ligands.
The calculated cone angles (Tolman’s cone angle θ)22 and
the crystallographically determined cone angles (solid angle
Ω)23 do not match for most of the complexes (Table 3). This
trend has been observed previously in systematic analyses
of phosphine complexes using crystallographic databases.23,45,46
Since the pKa of HPR3+ is free of π-back-bonding effects,
the basicity of the phosphorus is an attractive measure of
the σ-donor strength.24 The pKa values of the ligands in our
series were calculated using the substituent parameters
(47) Martin, A.; Orpen, A. G. J. Am. Chem. Soc. 1996, 118, 1464.
(48) (a) Bubel, R. J.; Douglass, W.; White, D. P. J. Comput. Chem. 2000,
21, 239. (b) Fernandez, A.; Reyes, C.; Prock, A.; Giering, W. P.
Organometallics 1998, 17, 2503. (c) Fernandez, A. L.; Lee, T. Y.;
Reyes, C.; Prock, A.; Giering, W. P.; Haar, C. M.; Nolan, S. P. J.
Chem. Soc., Perkin Trans. 2 1999, 2631. (d) Woska, D.; Prock, A.;
Giering, W. P. Organometallics 2000, 19, 4629. (e) Joerg, S.; Drago,
R. S.; Sales, J. Organometallics 1998, 17, 589.
(44) (a) Semmelmann, M.; Fenske, D.; Corrigan, J. F. J. Chem. Soc., Dalton
Trans. 1998, 2541. (b) Amoroso, A. J.; Johnson, B. F. G.; Lewis, J.;
Massey, A. D.; Raithby, P. R.; Wong, W. T. J. Organomet. Chem.
1992, 440, 219.
(45) (a) Smith, J. M.; Coville, N. J.; Cook, L. M.; Boeyens, J. C. A.
Organometallics 2000, 19, 5273. (b) White, D.; Taverner, B. C.;
Coville, N. J.; Wade, P. W. J. Organomet. Chem. 1995, 495, 41. (c)
Smith, J. M.; Taverner, B. C.; Coville, N. J. J. Organomet. Chem.
1997, 530, 131.
(49) Joerg, S.; Webster, C. E.; Drago, R. S.; Sales, J. Polyhedron 1999,
18, 1097.
(50) Bosque, R.; Sales, J. J. Chem. Inf. Comput. Sci. 2001, 41, 225.
(51) Hui, B. K. M.; Wong, W. T. J. Chem. Soc., Dalton Trans. 1998, 447.
(52) Layer, T. M.; Lewis, J.; Martin, A.; Raithby, P. R.; Wong, W. T. J.
Chem. Soc., Dalton Trans. 1992, 3411.
(46) Smith, J. M.; Coville, N. J. Organometallics 2001, 20, 1210.
Inorganic Chemistry, Vol. 41, No. 20, 2002 5265

Stulz et al.
to CO, and lowers the energy of the carbonyl triple bond,
which is expressed in an overall decreased stretching
frequency. π-Acceptance has an opposite effect and weakens
the Ru-CO bond. In the following section, both σ- and
π-effects are discussed individually for each type of ligand
(triarylphosphine, alkynylphosphine, phosphonite).
PPh3, being the second strongest base (pKa ) 2.73) among
the ligands in our series and consequently a relatively strong
σ-donor, displays the smallest IR shift of all complexes (∆ν
) 18 cm-1), which suggests that the π-acceptor ability is
weak.27 The 31P{1H} NMR resonance of the mono-complex
is very broad (vide supra), indicating rapid exchange between
free and bound species at a rate comparable to the NMR
chemical shift time scale. These observations are consistent
with the lowest association constant measured (Ka1 ) 1.2 ×
105 M-1), and with the long Ru-P bonds of porphyrin
phosphine complexes determined from the related TPP and
OEP systems. A threshold of d(Ru-P) ) 2.40 Å was
suggested for stable phosphine complexes,15 which is ex-
ceeded in PPh3 porphyrin complexes. Existence of the
(PPh3)2RuII(DPP) complex in solution could not be confirmed
by NMR spectroscopy. The association constant Ka1 is
between the values found for RuII(TPP) (7.4 × 103 M-1)
and RuII(OEP) (8.3 × 104 M-1)15 and suggests hybrid
properties of DPP compared to TPP and OEP. Since axial
binding to metalloporphyrins is also influenced by the
basicity of the porphyrin core, Ka1(PPh3) allows an estimation
of the pKa of DPP to be in the range of the values found for
OEP (pKa ∼3) and TPP (pKa ∼6).15,18
Figure 10. Correlation between the calculated53 pKa values of the ligands
and (a) the phosphorus chemical shift resonance δ(31P) or (b) the phosphorus
chemical shift differences ∆δ(31P) ) δ(31P)mono-phosphine - δ(31P)free phosphine
Straight lines are linear regressions [R(δ) ) 0.994, R(∆δ) ) 0.992].
.
reported in the literature (except for PPh3).53 The electronic
and steric properties of phosphines have been correlated with
31P{1H} NMR chemical shifts in order to predict the chemical
shifts of phosphines.50,54 Our data show a linear dependence
of the pKa vs δ(31P) (Figure 10), even if the primary
(phenylacetenyl)phosphine (PAPH2)55 is included in the
analysis.63 (PA)3P was excluded from the calculation of the
linear regression. This outlier indicates that the δ(31P) value
might also be influenced by electronic factors other than the
(calculated) pKa. More interestingly, the ∆δ(31P) values of
(PR3)RuII(CO)(DPP) compared to the free ligands also
correlate linearly with the pKa of the phosphine,64 and (PA)3P
again is an outlier. Based on the ∆δ-regression, the calculated
and measured ∆δ(31P) values show a linear correlation with
a slope of 1.06, confirming the validity of the regression.
Therefore we propose that the pKa of a specific phosphorus
ligand can be estimated on the basis of the chemical shift
difference ∆δ(31P) upon complexation to (MeOH)RuII(CO)-
(DPP), or to any other metalloporphyrin. The analyzed
system spans a very large range both on the phosphorus
NMR scale (δ ) -178 to +130 ppm) and on the pKa scale
(-3.96 to +6.11) and seems representative. This approach
is the reverse of the previously reported correlations, because
we are suggesting the use of the 31P{1H} NMR chemical
shift to probe the electronic properties of the phosphorus.
π-Back-bonding in phosphorus complexes is assumed to
increase with decreasing σ-donation.49 The IR shifts of the
carbonyl stretching frequency (νCtO) are among the most
diagnostic probes to elucidate the bonding in metal com-
plexes.18,51,56 νCtO is correlated to both σ- and π-effects:
σ-donation increases the electron density on the metal,
strengthens the Ru-CO bond by increased π-back-bonding
Generally, the substitution of phenyl groups in PPh3 by
phenylacetylene groups induces changes in the ligand
properties which have a large influence on the reactivity of
the phosphines. Most remarkable are the incrementally
decreased basicity [pKa(DPAP) ) 1.04, pKa({PA}3P) )
-2.96], higher association constants (Table 2), and relative
ease of formation of the bis-phosphine complex (weakening
of the Ru-CO bond through enhanced π-back-bonding).51
The larger ∆νCtO values for DPAP, DPPA, and (PA)3P
compared to PPh3 are due to an increased π-acidity as well
as to a decreased σ-donor strength of the ligands.27,28 The
increase in π-acceptor ability can be explained either by
delocalization into π*(CtC) orbitals or by electronic com-
munication via the ethynyl linker to the phenyl ring, leading
to enhanced delocalization. The stronger π-back-bonding also
leads to a decrease in the Ru-P bond lengths27 compared to
the PPh3 complexes of TPP and OEP, consistent with the
higher association constants. The 31P{1H} NMR shifts from
the free to the bound ligands upon first binding indicate a
relatively large decrease in electron density around the
phosphorus in the complexes, which is not compensated by
simultaneous RufPR3 π-back-bonding. Upon decarbony-
lation, the DPP is expected to have enhanced metal-to-
porphyrin (Mfpor) π-back-bonding, increasing its charge
density at the expense of the PR3fM σ-donation.57 The metal
therefore becomes a stronger Lewis acid, reducing the net
(53) (a) Henderson, W. A.; Streuli, C. A. J. Am. Chem. Soc. 1960, 82,
5791. (b) Allman, T.; Goel, R. G. Can. J. Chem. 1982, 60, 716.
(54) Abdur-Rashid, K.; Fong, T. P.; Greaves, B.; Gusev, D. G.; Hinman,
J. G.; Landau, S. E.; Lough, A. J.; Morris, R. H. J. Am. Chem. Soc.
2000, 122, 9155.
(55) Guillemin, J. C.; Savignac, P.; Denis, J. M. Inorg. Chem. 1991, 30,
2170.
(56) Vogel, K. M.; Kozlowski, P. M.; Zgierski, M. Z.; Spiro, T. G. Inorg.
Chim. Acta 2000, 297, 11.
(57) (a) Buchler, J. W.; Kokisch, W.; Smith, P. D. Struct. Bonding 1978,
34, 79. (b) Gross, Z.; Mahammed, A.; Barzilay, C. M. Chem. Commun.
1998, 1105.
5266 Inorganic Chemistry, Vol. 41, No. 20, 2002

Complexes of a Ru(II) Porphyrin. 1
charge density on the phosphorus in (PR3)2RuII(DPP) com-
pared to (PR3)RuII(CO)(DPP), and resulting overall in a larger
downfield shift of the 31P{1H} NMR resonance. This
demonstrates the large trans influence of ligands on both
stability and chemical shifts.11
of the phosphorus resonances in the mono-complex (sharp)
and in the bis-complex (broad) are reversed compared to the
phosphines. Accordingly, the values of Ka1 and Ka2 are the
largest, but Ka3 displays the smallest value of the complexes.
The first binding of PAP(OEt)2, which leads to a more stable
complex than with the phosphines but weakens the trans-
CO considerably, has an overall negative effect on the second
binding. The decrease of the Ka3 by about 2 orders of
magnitude compared to DPAP and DPPA is not reflected in
the Ru-P bond length, which would be expected to be the
longest (apart from PPh3), and also contradicts the expecta-
tions based on the π-acceptor ability (judged from the IR
shift) and on the cone angle θ.25,46 Additionally, d(Ru-P) is
not significantly decreased due to strong RufP π-back-
bonding.10
Attachment of a phosphorus at the alkynyl substituent in
DPPA was expected to alter the overall affinity to (MeOH)-
RuII(CO)(DPP). We intended to detect possible π-interactions
between the bound and the free phosphorus by spectroscopic
differences to DPAP.28 As the 31P{1H} NMR chemical shifts
and ∆νCtO show, however, there is no alteration in the
overall stereoelectronic properties. The Ka1 value does not
show a 2-fold increase in value which would be expected if
two phosphines were acting independently, doubling the
effective ratio kassoc/kdiss. The solid state structure shows that
the second phosphine orients its lone pair toward the
porphyrin core, in accordance with the conformation neces-
sary for intramolecular exchange in solution. This preferred
conformation might explain why DPPA does not act as a
bridging ligand, as is observed with other transition metal
complexes, and 31P{1H} NMR spectroscopy shows signals
for one bound and one free phosphorus.58 Ka3 increases about
2-fold on changing from DPAP to DPPA, which is not
consistent with the observations for the mono-phosphine
complex. The spectroscopic data suggest that there is indeed
no electronic communication between the two phosphines
through the acetylenic spacer.
The phosphonite PAP(OEt)2 has the highest 31P{1H} NMR
chemical shift (δ ) +130 ppm); thus the nucleus of the
phosphorus is strongly deshielded, which is attributed to the
polarized P-O bond. PAP(OEt)2 is the strongest base (pKa
) +6.11) of all our investigated ligands, contrary to the
expected decrease in basicity due to the electronegative
ethoxy substituents.13 Despite this high basicity, the largest
shift in the IR spectrum is observed (∆ν ) +39 cm-1),
consistent with the known π-acidity of phosphonites.10,11,25
This is a common feature of the P-O substitution pattern
and can best be seen in the relative bond lengths of the related
phosphate group, where the P-O(C) bonds have partial
double bond character, attributed to P(3d)-O(2pπ/2pπ*)
orbital interactions.59 Whereas all phosphinoalkyne ligands
display a downfield shift upon complexation, in the case of
PAP(OEt)2 the electron density around the phosphorus
nucleus increases, leading to a net upfield shift of the
resonance. This indicates that the energetic contribution of
the π-bonding is significantly larger compared to the
σ-bonding, in line with the theory that the strength of
σ-donation is inversely correlated to the strength of π-ac-
ceptance. Still the phosphonite seems to be a weaker σ-donor
compared to CO, reducing the π-basicity of the ruthenium
and leading to a net upfield shift of the 31P{1H} NMR
resonance in [PAP(OEt2)]2RuII(DPP) compared to [PAP-
(OEt)2]RuII(CO)(DPP). Significantly, the relative line widths
To summarize the observations, the following series were
obtained:
Ka1
Ka3
PPh3 < (PA)3P < DPAP, DPPA < PAP(OEt)2
PAP(OEt)2 < (PA)3P < DPAP, DPPA
σ-donor strength
π-acceptor strength
(PA)3P < DPAP < PPh3 < PAP(OEt)2
PPh3 < DPAP, DPPA < (PA)3P < PAP(OEt)2
Conclusions
We have investigated the effect of complexation of
different phosphorus ligands on the stability, solid state
structure, and spectroscopic properties of (MeOH)RuII(CO)-
(DPP) 2. The mono-phosphine complexes (PR3)RuII(CO)-
(DPP) are readily formed in solution in quantitative yields.
Due to their kinetic lability and the weakening of the
carbonyl ligand via a trans effect, the mono-phosphine
complexes could not be isolated. The bis-phosphine com-
plexes can be isolated in pure form by crystallization from
CHCl3-MeOH solutions using excess ligand. The X-ray
structure solutions of the bis-phosphine complexes show only
marginal differences in the geometrical parameters in the
solid state.
(61) Stulz, E.; Sanders, J. K. M.; Montalti, M.; Prodi, L.; Zaccheroni, N.;
Fabrizi de Biani, F.; Grigiotti, E.; Zanello, P. Inorg. Chem. 2002, 41,
5269.
(62) This structure has been communicated previously to the CCDC (CSD
refcode: HOJBAJ) but is described here in detail for the first time.60
(63) Originally, we had also included the primary phosphine PhCtC-
PH2 in our series. However, this ligand has proven to be very reactive
toward protic solvents and unstable with respect to oxidation, and
therefore unsuitable for the studies presented here; detailed studies
will be presented elsewhere. The 31P{1H} NMR chemical shifts were
included into the correlation to the pKa to improve the range of the
regression. pKa ) -3.69;53 δ(31P): -180 ppm (free ligand), -121
ppm (mono-phosphine complex).
(64) Diagnostic NMR shifts were obtained from the 31P nucleus, but
theoretically the chemical shift differences upon complexation cannot
be assigned purely to electronic effects upon formation of the Ru-P
bond, because the influence of the ring current of the aromatic
porphyrin system also has to be taken into account. The binding site
of a ligand to a metalloporphyrin is located in the strongly shielding
region of the π-system of the porphyrin, leading to a significant upfield
shift of the resonance of the respective nuclei. As judged from proton
NMR spectroscopy, however, the solution state geometries of the
complexes are almost identical in all cases. The relative positions of
the phosphorus atoms vary by 0.06 Å along the Ru-P axis orthogonal
to the porphyrin plane. We assume the influence of the bond length
differences to be negligible, and the observed chemical shift differences
to arise from the electronic effects of the different bond types and a
constant upfield shift caused by the ring current of the aromatic ring
system. This problem has not been addressed as such in previous
studies.
(58) Xu, D. F.; Murfee, H. J.; van der Veer, W. E.; Hong, B. J. Organomet.
Chem. 2000, 596, 53.
(59) Saenger, W. Principles of Nucleic Acid Structure; Springer-Verlag:
New York, 1984.
(60) Darling, S.; Feeder, N.; Sanders, J. K. M. 1999, private communication
to the CCDC.
Inorganic Chemistry, Vol. 41, No. 20, 2002 5267

Stulz et al.
The cone angle results demonstrate that mathematically
derived steric ligand parameters alone do not necessarily
represent an accurate model to predict solution stability of
phosphorus complexes. Also, it is not clear why an acetylenic
spacer, which introduces a structural gap between the
phosphorus and the phenyl ring when the phosphine is put
into an energetically minimized propeller twist conformation,
should not alter the steric demand of the ligand. But in our
system the Ka values span a range of 2 orders of magnitude,
which may be too small to detect a significant effect arising
purely from θ. Other comparisons of θ with stability
constants utilize systems where the association constants vary
by more than 5 orders of magnitude.22
The association constant Ka1 for PPh3 to (MeOH)RuII(CO)-
(DPP) is the smallest value measured, which is not in line
with the pKa, δ(31P), or θ values. Replacement of one phenyl
substituent for an alkynylphenyl substituent on the phosphine
has a large influence on the overall binding, and the
thermodynamic stability increases by a factor of ∼100. On
the other hand, if all phenyl groups are replaced, the Ka1
value drops back by a factor of 10, attributed to electronic
factors. It is also noteworthy that DPAP and DPPA seem to
have very similar stereoelectronic properties despite the
attachment of an additional phosphorus at the acetylene
substituent. The Ka2 values show that the first binding of
the ligands is decreased 4-fold upon decarbonylation, which
can be attributed to a trans effect of the CO ligand.18
Association to form the complexes (PR3)2RuII(DPP) for the
phosphines is in a range similar to that for the first binding
to (MeOH)RuII(CO)(DPP), but for the phosphonite PAP-
(OEt)2, binding is decreased 100-fold.
In summary, there is no simple correlation between the
individual spectroscopic values, the electronic and steric
effects, and the thermodynamic stability. The separation of
the electronic effects into pure σ-donor and π-acceptor
properties is generally oversimplified, and the overall
electronic density on the phosphorus has to be taken into
account. It remains to be evaluated whether the data can be
used to improve further the attempted calculations (ECW or
QALE model) on the stereoelectronic factors influencing the
properties of phosphines. We propose, however, that 31P-
{1H} NMR spectroscopy can be used to estimate the pKa,
and probably other stereoelectronic parameters, of a phos-
phorus ligand when ∆δ(31P) is determined for (PR3)RuII(CO)-
(DPP). The influence of the different ligands on the
photophysical and electrochemical behavior of RuII(DPP) will
be discussed in the following paper.61
Acknowledgment. The diffraction data of [PAP(OEt2)]2-
RuII(DPP) were collected by John E. Davies, Cambridge.
Financial support from the Swiss National Science Founda-
tion (Grant 8220-061293 to E.S.) and from the EPSRC is
gratefully acknowledged. We thank Amy L. Kieran for
careful revision of the manuscript. Helpful comments of
referees are gratefully acknowledged.
Supporting Information Available: X-ray crystallographic
tables for structures 12, 13, 14, and 15 (atomic coordinates, bonds
lengths and angles, anisotropic thermal factors, hydrogen atom
positions as CIFs); nonlinear least-squares fitting data for the UV-
vis titrations of 2 with 3, 5, 6, and 7; Gaussian deconvolution data
for the UV-vis dilution series of 12, 13, 14, and 15. This material
is available free of charge via the Internet at http://pubs.acs.org.
IC025727Y
5268 Inorganic Chemistry, Vol. 41, No. 20, 2002