Tridentate Ligand Effects on Enthalpies of Protonation of (L3)M(CO)3 Complexes (M = W, Mo)

Article abstract of DOI:10.1021/ic970670e

Titration calorimetry has been used to determine the enthalpies of protonation (ΔH_HM) for the reaction of (L₃)M-(CO)₃ complexes, where M = W and Mo and L₃ = cyclic and noncyclic tridentate ligands of the types CH₂-CH₂-X-CH₂CH₂-Y-CH ₂CH₂-Z and RX-CH₂CH₂-Y-CH₂CH₂-XR with N, S, and P donor atoms, with CF₃SO₃H in 1,2-dichloroethane solution at 25°C to give (L₃)M(CO)₃(H)⁺CF₃SO₃ ^-. The basicities (-ΔH_HM) increase with the ligand donor groups (X, Y, or Z) in the order S ≤ PPh ? NR (R = Me, Et) for both cyclic and noncyclic ligand complexes that have the same structure of the protonated product. Although the metal basicity (-ΔH_HM) generally increases as the ligand donor group basicities (pK_a's of the conjugate acids) increase, the large difference between the pK_a values of thioethers (-6,8) and phosphines (6.25) suggests that thioether donor groups should be much weaker donors than phosphines. The observation that thioether groups contribute nearly as much as phosphine groups to the basicity of the metal in the (L₃)M(CO)₃ complexes may be explained by suggesting that repulsion between the π-symmetry lone electron pair on sulfur and the filled metal d orbitals increases the energies of the d orbitals thereby making the metal more basic than expected from only the σ-donor ability of the sulfur. There is a good correlation (r = 0.973) between -ΔH_HM and average v(CO) values of the eight (L₃)W(CO)₃ complexes that have the same structure of their protonated forms. A plot of the average of the three v(CO) frequencies for the (L₃)W(CO)₃ complexes vs the average v(CO) frequencies for the analogous Mo complexes is linear (r = 0.9996), and the slope of 1.07 indicates that the tridentate ligands have nearly the same electronic effects on both W and Mo complexes. Noncyclic ligands make the metal more basic by 1.6 ± 0.3 kcal/mol than cyclic ligands with the same donor atoms. The tungsten complexes are 2.8 ± 0.1 kcal/mol more basic than their molybdenum analogs. Determinations of ΔH_HM values for both fac- and mer-(PNP)M(CO)₃ complexes (M = W, Mo; PNP = MeN(C₂H₄PPh₂)₂) allowed the calculation of enthalpies of mer-to-fac isomerization for both the tungsten (-2.0 kcal/mol) and molybdenum (-4,8 kcal/mol) complexes. These studies demonstrate that the metal, ligands, and geometry of the protonated products all substantially affect the heats of protonation (ΔH_HM) of (L₃)M(CO)₃ complexes.

Full text of DOI:10.1021/ic970670e

432
Inorg. Chem. 1998, 37, 432-444
Tridentate Ligand Effects on Enthalpies of Protonation of (L3)M(CO)3 Complexes (M ) W,
Mo)
Oltea P. Siclovan and Robert J. Angelici*
Department of Chemistry, Iowa State University, Ames, Iowa 50011
ReceiVed May 30, 1997
Titration calorimetry has been used to determine the enthalpies of protonation (∆HHM) for the reaction of (L3)M-
(CO)3 complexes, where M ) W and Mo and L3 ) cyclic and noncyclic tridentate ligands of the types CH2-
CH2-X-CH2CH2-Y-CH2CH2-Z and RX-CH2CH2-Y-CH2CH2-XR with N, S, and P donor atoms, with
CF3SO3H in 1,2-dichloroethane solution at 25 °C to give (L3)M(CO)3(H)+CF3SO3-. The basicities (-∆HHM
)
increase with the ligand donor groups (X, Y, or Z) in the order S e PPh , NR (R ) Me, Et) for both cyclic and
noncyclic ligand complexes that have the same structure of the protonated product. Although the metal basicity
(-∆HHM) generally increases as the ligand donor group basicities (pKa’s of the conjugate acids) increase, the
large difference between the pKa values of thioethers (-6.8) and phosphines (6.25) suggests that thioether donor
groups should be much weaker donors than phosphines. The observation that thioether groups contribute nearly
as much as phosphine groups to the basicity of the metal in the (L3)M(CO)3 complexes may be explained by
suggesting that repulsion between the π-symmetry lone electron pair on sulfur and the filled metal d orbitals
increases the energies of the d orbitals thereby making the metal more basic than expected from only the σ-donor
ability of the sulfur. There is a good correlation (r ) 0.973) between -∆HHM and average ν(CO) values of the
eight (L3)W(CO)3 complexes that have the same structure of their protonated forms. A plot of the average of the
three ν(CO) frequencies for the (L3)W(CO)3 complexes Vs the average ν(CO) frequencies for the analogous Mo
complexes is linear (r ) 0.9996), and the slope of 1.07 indicates that the tridentate ligands have nearly the same
electronic effects on both W and Mo complexes. Noncyclic ligands make the metal more basic by 1.6 ( 0.3
kcal/mol than cyclic ligands with the same donor atoms. The tungsten complexes are 2.8 ( 0.1 kcal/mol more
basic than their molybdenum analogs. Determinations of ∆HHM values for both fac- and mer-(PNP)M(CO)3
complexes (M ) W, Mo; PNP ) MeN(C2H4PPh2)2) allowed the calculation of enthalpies of mer-to-fac
isomerization for both the tungsten (-2.0 kcal/mol) and molybdenum (-4.8 kcal/mol) complexes. These studies
demonstrate that the metal, ligands, and geometry of the protonated products all substantially affect the heats of
protonation (∆HHM) of (L3)M(CO)3 complexes.
Introduction
with CF3SO3H in 1,2-dichloroethane (DCE) solution at 25 °C
(eq 1), and phosphine basicity (-∆HHP, eq 2, or pKa of the
Basicities of metal centers1,2 in transition metal complexes
are of particular interest because they can be used as a guide to
predict other types of reactivity that depend upon electron
richness of the metal center.3 Properties of transition metal
complexes are greatly influenced by their ligands, and several
studies4 of ligand effects on spectroscopic, electrochemical, and
kinetic properties of complexes have been reported. In previous
studies in our laboratories, excellent correlations between metal
basicity (-∆HHM), as measured by the enthalpies of protonation
conjugate acid) have been reported for the following series of
phosphine complexes: CpIr(CO)(PR3),5,3d Fe(CO)3(PR3)2,5
W(CO)3(PR3)3,6 and CpOs(PR3)2Br.7a
DCE
+
-
MLn + CF3SO3H 25 °C8 HMLn CF3SO3 ∆HHM (1)
DCE
+
-
PR3 + CF3SO3H 25 °C8 HPR3 CF3SO3 ∆HHP (2)
Effects of chelating phosphines on the basicities of metal
centers were investigated in several systems;7 it was established
(1) (a) Pearson, R. G. Chem. ReV. 1985, 85, 41. (b) Martinho Simo˜es, J.
A.; Beauchamp, J. L. Chem. ReV. 1990, 90, 629. (c) Kristja´nsdo´ttir,
S. S.; Norton, J. R. In Transition Metal Hydrides: Recent AdVances
in Theory and Experiment; Dedieu, A., Ed.; VCH: New York, 1991;
Chapter 10. (d) Bullock, R. M. Comments Inorg. Chem. 1991, 12, 1.
(e) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987. (f) Jessop, P. G.;
Morris, R. H. Coord. Chem. ReV. 1992, 121, 155. (g) Heinekey, D.
M.; Oldham, W. J., Jr. Chem. ReV. 1993, 93, 913. (h) Kramarz, K.;
Norton, J. R. Prog. Inorg. Chem. 1994, 42, 1.
(4) (a) Feltham, R. D.; Brant, P. J. Am. Chem. Soc. 1982, 104, 641. (b)
Chatt, J. Coord. Chem. ReV. 1982, 43, 337. (c) Bursten, B. E.; Green,
M. R. Prog. Inorg. Chem. 1988, 36, 393. (d) Lever, A. B. P. Inorg.
Chem. 1990, 29, 1271. (e) Poe¨, A. J. Pure Appl. Chem. 1988, 60,
1209. (f) Liu, H.-Y.; Eriks, K.; Prock, A.; Giering, W. P. Organo-
metallics 1990, 9, 1758. (g) Nolan, S. P. Comments Inorg. Chem. 1995,
17, 131. (h) Lu, S.; Strelets, V. V.; Ryan, M. F.; Pietro, W. J.; Lever,
A. B. P. Inorg. Chem. 1996, 35, 1013.
(5) Sowa, J. R., Jr.; Zanotti, V.; Facchin, G.; Angelici, R. J. J. Am. Chem.
Soc. 1991, 113, 9185.
(6) Sowa, J. R., Jr.; Zanotti, V.; Angelici, R. J. Inorg. Chem. 1993, 32,
848.
(2) Angelici, R. J. Acc. Chem. Res. 1995, 28, 51.
(3) (a) Shubina, E. S.; Epshtein, L. M. J. Mol. Struct. 1992, 265, 367. (b)
Kazarian, S. G.; Hamley, P. A.; Poliakoff, M. J. Am. Chem. Soc. 1993,
115, 9069. (c) Hamley, P. A.; Kazarian, S. G.; Poliakoff, M.
Organometallics 1994, 13, 1767. (d) Wang, D.; Angelici, R. J. Inorg.
Chem. 1996, 35, 1321.
(7) (a) Rottink, M. K.; Angelici, R. J. J. Am. Chem. Soc. 1993, 115, 7267.
(b) Sowa, J. R., Jr.; Bonanno, J. B.; Zanotti, V.; Angelici, R. J. Inorg.
Chem. 1992, 31, 1370. (c) Sowa, J. R., Jr.; Zanotti, V.; Facchin, G.;
Angelici, R. J. J. Am. Chem. Soc. 1992, 114, 160.
S0020-1669(97)00670-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/22/1998

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 433
that the metal basicity increases as the size of the chelate ring
decreases for cis-M(CO)2(LkL) complexes, where M ) Mo and
W.7b Small chelate ring phosphines also increase the basicities
of (LkL)Fe(CO)3 complexes by distorting their structures from
the most stable diaxial P-donor arrangement that is found in
the corresponding monodentate phosphine (L)2Fe(CO)37c com-
plexes. Structural effects of tridentate ligands also influence
the -∆HHM value of fac-[PhP(C2H4PPh2)2]W(CO)3 (16.7 kcal/
mol) (eq 3), which is much more basic than fac-[MeC(CH2-
PPh2)3]W(CO)3 (10.5 kcal/mol) (eq 4).6,8
Chart 1
mentioned. Solvents were purified under nitrogen using standard
methods.11 Hexanes, methylene chloride, and acetonitrile were refluxed
over CaH2 and then distilled. Ethanol (100%) was used as purchased.
Tetrahydrofuran (THF), diethyl ether, and dioxane were distilled from
sodium benzophenone. CD2Cl2 and CDCl3 were stored over molecular
sieves under nitrogen. 1,2-Dichloroethane (DCE) was purified by
washing with concentrated sulfuric acid, distilled deionized water, 5%
NaOH, and again with water. The solvent was then predried over
anhydrous MgSO4 and stored in amber bottles over molecular sieves
(4 Å). The DCE was distilled from P4O10 under argon immediately
before use. Triflic acid (CF3SO3H) was purchased from 3M Co. and
purified by fractional distillation under argon prior to use. Neutral
alumina (Brockmann, activity I) used for chromatography was deoxy-
genated at room temperature under vacuum for 12 h, deactivated with
5% (w/w) N2-saturated water, and stored under nitrogen. Silica gel
(40 µm) used for chromatography was deoxygenated at room temper-
ature under vacuum for 12 h and stored under N2. 2-Mercaptoethyl
sulfide and 1,4,7-trimethyltriazacyclononane (Aldrich) and N-ethyldi-
ethanolamine (Acros) were purchased and used as received.
Both complexes have facial structures, but upon protonation,
fac-[PhP(C2H4PPh2)2]W(CO)3 rearranges9 to a structure in which
the ligand donor atoms are nearly coplanar with the W atom.
The flexibility of the PhP(C2H4PPh2)2 ligand allows the complex
to achieve this stable structure, which is also the structure of
the complex (PMePh2)3W(CO)3(H)+ with monodentate ligands.9
On the other hand, the MeC(CH2PPh2)3 ligand does not allow
the protonated product to achieve this lower energy structure
which makes its -∆HHM (10.5 kcal/mol) much less favorable
for the reaction in eq 4 than that in eq 3.
In order to expand our understanding of tridentate ligand
effects on metal basicity, we studied basicities (-∆HHM, eq 5)
DCE
fac-(L3)M(CO)3 + CF3SO3H
8
25 °C
(L3)M(CO)3(H)+CF3SO3 ∆HHM (5)
-
The 1H NMR spectra were obtained on samples dissolved in CDCl3
or CD2Cl2 on a Varian VXR 300-MHz NMR spectrometer using TMS
(δ ) 0.00 ppm) as the internal reference. The 31P{H} NMR spectra
of samples in CD2Cl2 or CDCl3 were recorded on a Bruker AC 200-
MHz spectrometer using 85% phosphoric acid (δ ) 0.00) as the external
reference. Solution spectra were recorded on a Nicolet 710 FT-IR
spectrometer using sodium chloride cells with 0.1 mm spacers. Electron
ionization mass spectra (EIMS at 70 eV) and chemical ionization mass
spectra (CIMS) were run on a Finnigan 4000 spectrometer. Elemental
microanalyses were performed on a Perkin-Elmer 2400 series II
CHNS/O analyzer.
The compounds M(CO)3(CH3CN)3, where M ) W and Mo, are
starting materials for the syntheses of all metal complexes and were
prepared according to literature procedures by refluxing W(CO)6 (14
h) or Mo(CO)6(4 h) in CH3CN.12,13 The reactions were checked by IR
spectroscopy for completeness (ν(CO) (CH3CN): for W(CO)3(CH3-
CN)3, 1911 (s), 1791 (s, br) cm-1; for Mo(CO)3(CH3CN)3, 1920 (s),
1796 (s, br) cm-1). The solutions were then used as such in subsequent
reactions.
L3 ) SSSc (1W, 1Mo); L3 ) SPSc (2W, 2Mo);
L3 ) SSS (3W, 3Mo); L3 ) SPS (4W, 4Mo);
L3 ) SNSc (5W, 5Mo); L3 ) SNS (6W, 6Mo);
L3 ) NSNc (7W, 7Mo); L3 ) NNNc (8W, 8Mo);
L3 ) PNP (9W-fac, 9Mo-fac)
of a series of complexes (L3)M(CO)3 (M ) W, Mo; L3 ) cyclic
or noncyclic tridentate ligand with S, N, and/or P donor atoms).
The ligands and their abbreviations are shown in Chart 1. The
abbreviations are based on the ligand donor atoms, and a small
case c distinguishes the cyclic ligands from their noncyclic
counterparts. We report here the effects of the donor atoms,
the metal, and the structural differences imposed by the ligands
on the basicities of these complexes.
Experimental Section
Preparation of (SSSc)M(CO)3, M ) W (1W) and Mo (1Mo).
These compounds were prepared by a method described for the
synthesis of carbon-substituted 1,4,7 trithiacyclononanes.14 1,2-Dibro-
General Procedures. All preparative reactions, chromatography,
and manipulations were carried out under an atmosphere of nitrogen
or argon using standard Schlenk techniques,10 unless otherwise
(11) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals, 2nd ed.; Pergamon: New York, 1980.
(12) Tate, C. P.; Augl, J. M. ; Kipple, W. R. Inorg. Chem. 1962, 1, 433.
(13) Ross, B. L.; Grasselli, J. L.; Richey, W. M.; Kaesz, H. D. Inorg. Chem.
1963, 2, 1023.
(14) Smith, R. J.; Salek, S. N.; Went, M. J.; Blower, P. J.; Barnard, N. J.
J. Chem. Soc., Dalton Trans. 1994, 3165.
(8) Sowa, J. R., Jr.; Zanotti, V.; Angelici, R. J. Inorg. Chem. 1991, 30,
4108.
(9) Zanotti, V.; Rutar, V.; Angelici, R. J. J. Organomet. Chem. 1991,
414, 177.
(10) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-SensitiVe
Compounds, 2nd ed.; Wiley: New York, 1986.

434 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
mobutane (1.12 g, 5.19 mmol) freshly distilled under nitrogen was
added dropwise to a suspension of (NMe4)2[W(CO)3(S(C2H4S)2)] (2.82
g, 4.96 mmol), which was prepared exactly as described for the Mo
analog,15 in acetonitrile (20 mL). The resulting brown suspension was
stirred for 20 h at room temperature under N2, and then the solvent
was removed under vacuum. The residue was washed with hexanes
and then dissolved in CH2Cl2 (ca. 30-35 mL). The solution was
filtered through a neutral alumina column (2.5 × 3 cm) to remove the
Me4NBr and insoluble decomposition products that formed during the
reaction. The filtrate was then concentrated and purified on a neutral
alumina column (1.5 × 10 cm) eluting with a 4:1 (v/v) mixture of
CH2Cl2 and hexanes. The desired product eluted last as a pale yellow
band, as determined by checking the fractions by IR spectroscopy.
Recrystallization from CH2Cl2- ethyl ether gave (SSSc)W(CO)3 (1W)
(0.732 g, 31%) as a yellow powder. 1H NMR (CD2Cl2) for the mixture
Cl2): δ 93.13 (s). Anal. Calcd for C15H17O3PS2Mo: C, 41.29; H,
3.93. Found: C, 41.13; H, 3.97. Complex 2Mo was recently prepared
by a very similar route.17a
Preparation of (SSS)M(CO)3, M ) W (3W) and Mo (3Mo).
These syntheses are similar to that used for (2,5,8-trithianonane)Mo-
(CO)3.17b
S(C2H4SEt)2 (SSS). To a solution of S(C2H4SH)2 (1.18 g, 7.66
mmol) in 30 mL of dry THF cooled to -78 °C was added dropwise a
solution of n-BuLi (2.5 M in hexanes, 6.2 mL, 15.5 mmol). The
reaction mixture was allowed to stir for 15 min, and then EtBr (1.69 g,
15.5 mmol) dissolved in 5 mL of THF was added. After 15 more
minutes of stirring, the mixture was warmed to room temperature and
concentrated to ca. 10 mL on a rotary evaporator. Water (ca. 30 mL)
was added, and the product was extracted with ether. After drying of
the ether extract over anhydrous Na2SO4, the solvent was removed under
vacuum to yield the product as a very pale yellow oil (1.42 g, 88%),
which was used without further purification. 1H NMR (CDCl3): δ
1.28 (t, 3JHH ) 6 Hz, 6H, CH3), 2.58 (q, 3JHH ) 6 Hz, 4 H, CH2-Me),
2.72-2.79 (m, 8H, S-CH2-CH2-S).
3
3
of isomers: δ [1.12 (t, JHH ) 6 Hz), 1.19 (t, JHH ) 7.5 Hz), 3H,
CH3(C-Et)], 1.26-3.59 (m, 13H, CH2-S). Isomer ratio ∼ 1:2. Anal.
Calcd for C11H16O3S3W: C, 27.74; H, 3.39. Found: C, 27.53; H, 3.38.
The Mo analog, 1Mo, was obtained in 37% yield (0.731 g) from
1,2-dibromobutane (1.10 g, 5.09 mmol) and (NMe4)2[Mo(CO)3-
(S(C2H4S)2)]15 (2.45 g, 5.09 mmol) in 15 mL of CH3CN following the
same procedure used for 1W. 1H NMR (CD2Cl2) for the mixture of
(SSS)W(CO)3 (3W). A solution of W(CO)3(CH3CN)3 (1.27 g, 3.25
mmol) in CH3CN (25 mL) was transferred by cannula under N2 to a
flask containing S(C2H4SEt)2 (0.684 g, 3.25 mmol) in an inert
atmosphere. The reaction mixture was allowed to stir at room
temperature for 5 h. The solvent was then removed under vacuum,
and the residue was washed with hexanes and then dissolved in a
minimum amount of CH2Cl2. The CH2Cl2 solution was chromato-
graphed on a neutral alumina column (1.5 × 8 cm) using CH2Cl2 as
the eluent. The yellow band was collected, and the resulting compound
was recrystallized from CH2Cl2-ether yielding 3W (0.793 g, 51% based
on W(CO)6) as a yellow microcrystalline mass. 1H NMR (CD2Cl2):
3
3
isomers: δ [1.11 (t, JHH ) 6 Hz), 1.17 (t, JHH ) 7.5 Hz), 3H, CH3-
(C-Et)], 1.27-3.48 (m, 13H, CH2-S). Isomer ratio ∼ 1:2. Anal.
Calcd for C11H16O3S3Mo: C, 34.02; H, 4.15. Found: C, 33.82; H,
4.20.
Preparation of (SPSc)M(CO)3, M ) W (2W) and Mo (2Mo). The
synthesis of these compounds uses a molybdenum template reaction
previously applied to the preparation of (1,4,7-trithiacyclononane)Mo-
(CO)3.15
(NMe4)2[PhP(C2H4S)2]. A solution of PhP(C2H4SH)216 (1.08 g, 4.69
mmol) in 25% methanolic Me4NOH (3.42 g, 3.95 mL, 9.38 mmol)
was prepared under N2, stirred at room temperature for 5 min, and
evaporated to dryness under vacuum. The yellowish-green residue of
(NMe4)2[PhP(C2H4S)2] was used for subsequent reactions without
further purification.
(NMe4)2[W(CO)3(PhP(C2H4S)2)]. A solution of W(CO)3(CH3CN)3
(1.83 g, 4.68 mmol) in acetonitrile (30 mL) was transferred under N2
by cannula to a flask containing (NMe4)2[PhP(C2H4S)2] (1.76 g, 4.67
mmol) in an inert atmosphere. The reaction mixture was allowed to
stir for 8-10 h at room temperature. The resulting suspension
containing a yellow precipitate was used as such in subsequent reactions.
(SPSc)W(CO)3 (2W). Freshly distilled 1,2-dibromoethane (0.883
g, 4.70 mmol) that was saturated with N2 was added dropwise to a
suspension of (NMe4)2[W(CO)3(PhP(C2H4S)2)] (3.01 g, 4.67 mmol) in
acetonitrile (30 mL). After the reaction mixture was allowed to stir at
room temperature for 2 h, the solvent was removed under vacuum and
the brown residue was washed with hexanes and dissolved in CH2Cl2
(ca. 30 mL). The suspension was filtered through a neutral alumina
column (2.5 × 3 cm) to remove Me4NBr and insoluble decomposition
products that formed during the reaction. The filtrate was concentrated
and then purified on a neutral alumina column (1.5 × 10 cm) by eluting
with a 4:1 (v/v) mixture of CH2Cl2 and hexanes. The desired product
elutes last, as determined by IR spectroscopy, as a pale yellow band.
Recrystallization from CH2Cl2-ethyl ether gave (SPSc)W(CO)3 (2W)
(0.958 g, 39%) as a yellow powder. 1H NMR (CD2Cl2): δ 1.80-1.91
(m, 12H, CH2-S and CH2-P), 7.50-7.98 (m, 5H, Ph). 31P{H} NMR
(CD2Cl2): δ 92.32 (s). Anal. Calcd for C15H17O3PS2W: C, 34.37; H,
3.27. Found: C, 34.44; H, 3.30.
3
δ 1.37 (t, JHH ) 7.5 Hz, 6 H, CH3), 2.32-2.76 (m, 8H, S-CH2-
3
CH2-S), 2.83 (q, JHH ) 7.5 Hz, 4H, CH2-Me). Anal. Calcd for
C11H18O3S3W: C, 27.62; H, 3.79. Found: C, 27.80; H, 3.80.
The Mo analog, 3Mo, was obtained in 55% yield (0.751 g) from
Mo(CO)3(CH3CN)3 (1.06 g, 3.50 mmol) in 25 mL of CH3CN and
S(C2H4SEt)2 (0.736 g, 3.50 mmol) according to the above procedure.
3
1H NMR (CD2Cl2): δ 1.40 (t, JHH ) 7.5 Hz, 6 H, CH3), 2.34-2.80
(m, 12H, CH2-S). Anal. Calcd for C11H18O3S3Mo: C, 33.84; H, 4.65.
Found: C, 34.05; H, 4.65.
Preparation of (SPS)M(CO)3, M ) W (4W), and Mo (4Mo).
PhP(C2H4SEt)2 (SPS) was prepared in 81% yield (1.91 g), as described
16
for the synthesis of S(C2H4SEt)2, from PhP(C2H4SH)2 (1.89 g, 8.22
mmol), n-BuLi (6.56 mL, 16.4 mmol, 2.5 M in hexanes), and EtBr
3
(1.79 g, 16.4 mmol). 1H NMR (CDCl3): δ 1.20 (t, JHH ) 6 Hz, 6H,
CH3), 2.45-2.61 (m, 4H, CH2-S), 1.98-2.07 (m, 4H, P-CH2), 7.35-
7.58 (m, 5H, Ph). MS (CI): m/e 287 (MH+).
(SPS)W(CO)3 (4W) was obtained in 56% yield (1.09 g) , as
described for the synthesis of (SSS)W(CO)3, from W(CO)3(CH3CN)3
(1.38 g, 3.53 mmol) in acetonitrile (25 mL) and PhP(C2H4SEt)2 (1.01
g, 3.53 mmol). 1H NMR (CDCl3): δ 1.37 (t, 3JHH ) 6 Hz, 6H, CH3),
1.86-2.90 (m, 12H, CH2-S and CH2-P), 7.51-7.99 (m, 5H, Ph).
31P{H} NMR (CD2Cl2): δ 80.46 (s). Anal. Calcd for C17H23O3-
PS2W: C, 36.84; H, 4.18. Found: C, 37.06; H, 4.10.
The molybdenum analog, 4Mo, was prepared in 59% yield (1.03
g), from Mo(CO)3(CH3CN)3 (1.13 g, 3.73 mmol) in 25 mL of CH3CN
and PhP(C2H4SEt)2 (1.07 g, 3.73 mmol). 1H NMR (CDCl3): δ 1.38
(t, 3JHH ) 6 Hz, 6H, CH3), 1.89-2.95 (m, 12H, CH2-S and CH2-P),
7.50-8.00 (m, 5H, Ph). 31P{H} NMR (CD2Cl2): δ 81.68 (s). Anal.
Calcd for C17H23O3PS2Mo: C, 43.78; H, 4.97. Found: C, 43.80; H,
5.08.
The Mo analog, 2Mo, was obtained in 43% yield (0.849 g) from
1,2-dibromoethane (0.849 g, 4.52 mmol) and (NMe4)2[Mo(CO)3(PhP-
(C2H4S)2)] (2.51 g, 4.52 mmol), which was prepared from Mo(CO)3-
(CH3CN)3 (1.37 g, 4.52 mmol) in 25 mL of CH3CN and (NMe4)2[PhP-
(C2H4S)2] (1.70 g, 4.52 mmol), according to the procedure used for
the synthesis of 2W. 1H NMR (CD2Cl2): δ 1.82-2.93 (m, 12H,
CH2-S and CH2-P), 7.50-7.98 (m, 5H, Ph). 31P{H} NMR (CD2-
Preparation of (SNSc)M(CO)3, M ) W (5W) and Mo (5Mo).
The following sequence of reactions for the synthesis of EtN(C2H4-
SH)2 was adapted from a known procedure for the preparation of TsN-
(C2H4SH)2.18
(17) (a) Blower, P. J.; Jeffery, J. C.; Miller, J. R.; Salik, S. N.; Schmal-
johann, D.; Smith, R. J.; Went, M. J. Inorg. Chem. 1997, 36, 1578.
(b) Ashby, M. T.; Enemark, J. H.; Lichtenberger, D. L.; Ortega, R. B.
Inorg. Chem. 1986, 25, 3154.
(15) Sellmann, D.; Zapf, L. J. Organomet. Chem. 1985, 289, 57.
(16) Blower, P. J.; Dilworth, J. R.; Leigh, G. J.; Neaves, B. D.; Normanton,
F. B.; Hutchinson, J.; Zubieta, J. A. J. Chem. Soc., Dalton Trans.
1985, 2647.
(18) McAuley, A.; Subramanian, S. Inorg. Chem. 1990, 29, 2830.

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 435
N-Ethylbis(2-chloroethyl)ammonium chloride. To a stirred solu-
tion of SOCl2 (14.7 g, 123 mmol) in dry chloroform (10 mL) under N2
was added dropwise a solution of EtN(C2H4OH)2 (7.10 g, 53.3 mmol)
in 10 mL of dry chloroform. The reaction mixture was refluxed for
20 min and stirred at room temperature for 1 h. Ethanol (20 mL) was
added, and the mixture was refluxed for 2-3 min. Concentrating the
solution under vacuum and adding ethyl ether precipitated the product
(9.14 g, 83%) as a white solid which was filtered out, washed with
ether, and air dried.
Formation of the Thiouronium Salt. N-Ethylbis(2-chloroethyl)-
ammonium chloride (6.02 g, 29.1 mmol) and thiourea (4.57 g, 60.0
mmol) were refluxed in ethanol (100 mL) under nitrogen for 8-10 h.
The fine white precipitate that formed (eq 6) during reflux was filtered
equipped with a condenser was added the thiouronium salt (3.02 g,
8.42 mmol) slowly in small amounts. A gentle stream of N2 was
bubbled through the solution at all times. The reaction mixture was
then refluxed for 1 h under a N2 atmosphere. After the sample was
cooled to room temperature, EtBr (1.83 g, 16.8 mmol) was added
dropwise, and the reaction mixture was allowed to stir for 30 min at
room temperature. The solution was then concentrated to ca. 15 mL
on a rotary evaporator, and 40 mL of water was added. The product
was extracted from the aqueous solution with ether; the ether extract
was dried over anhydrous Na2SO4. Removal of the solvent under
vacuum gave EtN(C2H4SEt)2 as a pale yellowish oil (1.17 g, 63% based
on the thiouronium salt) that was used without further purification for
the syntheses of the metal complexes.
(SNS)W(CO)3 (6W) was prepared in 74% yield, based on W(CO)6
(1.52 g), from W(CO)3(CH3CN)3 (1.64 g, 4.19 mmol) in 30 mL of
acetonitrile and EtN(C2H4SEt)2 (0.928 g, 4.19 mmol) as described for
the preparation of 3W. 1H NMR (CD2Cl2): δ 1.27 [t, 3JHH ) 7.5 Hz,
3
3 H, CH3(N-Et)], 1.36 [t, JHH ) 7.5 Hz, 6H, CH3(S-Et)], 2.69-
3
3.03 (m, 12H, S-CH2 and N-CH2), 3.48 [q, JHH ) 7.5 Hz, 2H,
CH2(N-Et)]. Anal. Calcd for C13H23NO3S2W: C, 31.91; H, 4.74; N,
2.86. Found: C, 32.17; H, 4.83; N, 2.82.
off, washed with ether and air dried. Concentration of the filtrate and
addition of ether yielded more product to give a total yield of 9.53 g
(91%). The thiouronium salt was used without further purification.
Hydrolysis of the Thiouronium Salt. A 10% (w/w) aqueous KOH
solution (50 mL) was saturated with N2 for at least 30 min in a 3-necked
flask equipped with a condenser. The thiouronium salt (3.02 g, 8.42
mmol) was slowly added in small amounts with stirring while a gentle
stream of N2 was bubbled through the solution. The reaction mixture
was then refluxed for 30-40 min under N2. After cooling, the solution
was filtered under N2 and the filtrate was diluted with 50 mL of de-
aerated H2O. Concentrated HCl was then added to the solution until
pH ∼ 8 was reached. After the product was extracted with CH2Cl2,
the resulting solution was dried over anhydrous MgSO4, and the CH2-
Cl2 solvent was removed under vacuum leaving the EtN(C2H4SH)2
product as a colorless oil (0.989 g, 71%). 1 H NMR (CDCl3): δ 1.03
The molybdenum analog, 6Mo, was obtained in 78% yield (1.36 g)
from Mo(CO)3(CH3CN)3 (1.32 g, 4.36 mmol) in 25 mL of CH3CN
and EtN(C2H4SEt)2 (0.965 g, 4.36 mmol) following the same procedure
described above for the synthesis of 6W. 1H NMR (CD2Cl2): δ 1.27
[t, 3JHH ) 7.5 Hz, 3 H, CH3(N-Et)], 1.39 [t, 3JHH ) 7.5 Hz, 6H, CH3-
3
(S-Et)], 2.53-2.91 (m, 12H, S-CH2 and N-CH2), 3.38 [q, JHH
)
7.5 Hz, 2H, CH2(N-Et)]. Anal. Calcd for C13H23NO3S2Mo: C, 38.90;
H, 5.78; N, 3.49. Found: C, 39.08; H, 5.85; N, 3.51.
Preparation of (NSNc)M(CO)3, M ) W (7W) and Mo (7Mo).
4,7-Diethyl-1-thio-4,7-diazacyclononane (NSNc). To an ice-cooled
mixture of 1-thia-4,7-diazacyclononane dihydrobromide19 (2.07 g, 6.72
mmol), methanol (25 mL), and acetaldehyde (1.28 g, 29.1 mmol) was
slowly added NaBH3CN (0.599 g, 9.51 mmol) with continuous stirring.
When the addition was complete, the reaction mixture was allowed to
warm to room temperature and was stirred for 72 h. The solution was
then acidified with concentrated HCl to pH ) 2, the methanol was
removed under vacuum, and 16 mL of H2O was added to the residue.
The solution was brought to pH > 10 with KOH and then saturated
with NaCl. After the product was extracted with ether, the ether
solution was dried over anhydrous Na2SO4 and evaporated to give the
product (NSNc) as an oil in 69% yield (0.935 g). 1H NMR (CDCl3):
3
3
(t, JHH ) 6 Hz, 3 H, CH3), 1.74 (t, JHH ) 6 Hz, 2H, S-H), 2.51-
2.69 (m, 10H, N-CH2 and S-CH2). MS (EI): m/e 165 (M+).
(NMe4)2[EtN(C2H4S)2] was prepared, as described for (NMe4)2[PhP-
(C2H4S)2], from EtN(C2H4SH)2 (0.511 g, 3.09 mmol) and Me4NOH
(2.25 g, 2.60 mL of 25% (w/w) solution in methanol, 6.18 mmol) and
used as such in subsequent reactions.
3
(NMe4)2[W(CO)3(EtN(C2H4S)2)] was prepared, as described for
(NMe4)2[W(CO)3(PhP(C2H4S)2)], from W(CO)3(CH3CN)3 (1.21 g, 3.09
mmol) in 25 mL of acetonitrile and (NMe4)2[EtN(C2H4S)2] (0.963 g,
3.09 mmol); the resulting suspension was used in subsequent reactions.
(SNSc)W(CO)3 (5W) was prepared in 13% yield (0.196 g), as
described for the preparation of 1W, from 1,2-dibromobutane (0.669
g, 3.10 mmol) and (NMe4)2[W(CO)3(EtN(C2H4S)2)] (1.79 g, 3.09 mmol)
in 25 mL of acetonitrile. 1H NMR (CD2Cl2) for the mixture of
δ 1.03 (t, JHH ) 6 Hz, 6H, CH3), 2.54-2.62 (m, 8H), 2.95 (s, 8H).
MS (EI): m/e 202 (M+).
(NSNc)W(CO)3 (7W) was prepared in 66% yield (0.808 g) from
W(CO)3(CH3CN)3 (1.02 g, 2.60 mmol) in 25 mL of acetonitrile and
the ligand (NSNc) (0.527 g, 2.60 mmol) as described for the preparation
of 3W but with a longer reaction time, 10 h of stirring at room
temperature. 1H NMR (CD2Cl2): δ 1.29 [t, 3JHH ) 7.5 Hz, 6 H, CH3-
(N-Et)], 2.63-2.85 (m, 8H, S-CH2-CH2-N), 3.07-3.14 (m, 4H,
N-CH2-CH2-N), 3.43-3.55 (m, 4H, CH2(N-Et)]. Anal. Calcd for
C13H22N2O3SW: C, 33.20; H, 4.72; N, 5.96. Found: C, 33.36; H, 4.77;
N, 5.90.
3
3
isomers: δ [1.12 (t, JHH ) 7.5Hz), 1.19 (t, JHH ) 7.5Hz), 3H for
both isomers, CH3(C-Et)],)], [1.28 (t, 3JHH ) 7.5 Hz), 1.29 (t, 3JHH
)
7.5 Hz), 3H for both isomers, CH3(N-Et)], 1.60-3.54 (m, 15H, CH2N
and CH2S). Isomer ratio ∼ 1:8. Anal. Calcd for C13H21NO3S2W: C,
32.04; H, 4.34; N, 2.87. Found: C, 31.31; H, 4.32; N, 2.88.
The molybdenum analog, 5Mo, was obtained in 15% yield (0.194
g) from 1,2-dibromoethane (0.702 g, 3.25 mmol) and (NMe4)2[Mo-
(CO)3(EtN(C2H4S)2)] (1.59 g, 3.24 mmol), which was prepared from
Mo(CO)3(CH3CN)3 (0.982 g, 3.24 mmol) in 25 mL of CH3CN and
(NMe4)2[EtN(C2H4S)2] (1.01 g, 3.24 mmol), according to the procedure
used for 5W. 1H NMR (CD2Cl2) for the mixture of isomers: δ [1.11
(t, 3JHH ) 7.5 Hz), 1.17 (t, 3JHH ) 7.5 Hz), 3H for both isomers, CH3-
(C-Et)], [1.29 (t, 3JHH ) 7.5 Hz), 1.28 (t, 3JHH ) 7.5 Hz), 3H for both
isomers, CH3(N-Et)], 1.60-3.34 (m, 15H, CH2N and CH2S). Isomer
ratio ∼ 1:1. Anal. Calcd for C13H21NO3S2Mo: C, 39.10; H, 5.30; N,
3.51. Found: C, 39.30; H, 5.41; N, 3.62.
The molybdenum analog, 7Mo, was obtained in 69% yield (0.760
g) from Mo(CO)3(CH3CN)3 (0.873 g, 2.88 mmol) in 25 mL of CH3CN
and the ligand (NSNc) (0.583 g, 2.88 mmol) according to the procedure
3
described above for 7W. 1H NMR (CD2Cl2): δ 1.29 [t, JHH ) 7.5
Hz, 6 H, CH3(N-Et)], 2.51-2.81 (m, 8H, S-CH2-CH2-N), 2.98-
3.10 (m, 4H, N-CH2-CH2-N), 3.27-3.45 (m, 4H, CH2(N-Et)]. Anal.
Calcd for C13H22N2O3SMo: C, 40.84; H, 5.80; N, 7.33. Found: C,
41.09; H, 5.90; N, 7.20.
Preparation of (NNNc)M(CO)3, M ) W (8W)20 and Mo (8Mo),21
was performed as described in the literature. 1H NMR (CDCl3) for
8W: δ 2.78-2.87 (m, 6H, N-CH2), 2.97-3.06 (m, 6H, N-CH2), 3.16
(19) (a) Hart, S. M.; Boeyens, J. C. A.; Michael, J. P.; Hancock, R. D. J.
Chem. Soc., Dalton Trans. 1983, 1601. (b) Hoffmann, P.; Steinhoff,
A.; Mattes, R. Z. Naturforsch. 1987, 42b, 867.
(20) Backes-Dahmann, G.; Wieghardt, K. Inorg. Chem. 1985, 24, 4044.
(21) Backes-Dahmann, G.; Herrmann, W.; Wieghardt, K. Inorg. Chem.
1985, 24, 485.
Preparation of (SNS)M(CO)3, M ) W (6W) and Mo (6Mo). The
ligand for these complexes was prepared starting with the thiouronium
salt that was used in the synthesis of compounds 5W and 5Mo.
EtN(C2H4SEt)2 (SNS). To a 10% (w/w) ethanolic KOH solution
(50 mL) saturated with N2 for at least 30 min in a 3-necked flask

436 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
Table 1. IR Data for (L3)W(CO)3 and (L3)W(CO)3(H)+ Complexes in CH2Cl2 Solution
ν(CO), cm-1
complex
avg ν(CO), cm-1
(SSSc)W(CO)3, 1W
1929 (s)
2023 (s)
1930 (s)
2030 (s)
1923 (s)
2031 (s)
1925 (s)
2032 (s)
1914 (s)
2025 (s)
1911 (s)
2022 (s)
1905 (s)
2015 (s)
1895 (s)
2010 (s)
1922 (s)
1951 (w)
2032 (m)
1930 (s)
2025 (s)
1822 (s, br)
1950 (s)
1834 (s)
1952 (s)
1815 (s, br)
1959 (s)
1829 (s)
1961 (s)
1792 (s, br)
1947 (s)
1789 (s, br)
1942 (s)
1858
(SSSc)W(CO)3H+, 1WH+
(SPSc)W(CO)3, 2W
1930 (sh)
1819 (s)
1940 (sh)
1861
1851
1853
1833
1830
1818
1804
(SPSc)W(CO)3H+, 2WH+
(SSS)W(CO)3, 3W
(SSS)W(CO)3H+, 3WH+
(SPS)W(CO)3, 4W
1942 (sh)
1804 (s)
1948 (sh)
(SPS)W(CO)3H+, 4WH+
(SNSc)W(CO)3, 5W
(SNSc)W(CO)3H+, 5WH+
(SNS)W(CO)3, 6W
1929 (sh)
1920 (sh)
1908 (sh)
(SNS)W(CO)3H+, 6WH+
(NSNc)W(CO)3, 7W
1775 (s, br)
1932 (s)
1758 (s, br)
1921 (s)
(NSNc)W(CO)3H+, 7WH+
(NNNc)W(CO)3, 8W
(NNNc)W(CO)3H+, 8WH+
fac-(PNP)W(CO)3, 9W-fac
mer-(PNP)W(CO)3, 9W-mer
(PNP)W(CO)3H+, 9WH+
1902 (sh)
1797 (s)
1798 (m)
1919 (vs)
1823 (s)
1839 (vs)
1950 (m)
1835 (s, br)
1954 (s)
a
[MeC(CH2PPh2)3]W(CO)3
1867
[MeC(CH2PPh2)3]W(CO)3(H)+ a
a Reference 9.
1941 (sh)
(s, 9H, N-CH3). 1H NMR (CD2Cl2) for 8Mo: δ 2.68-2.79 (m, 6H,
N-CH2), 2.91-3.02 (m, 6H, N-CH2), 3.04 (s, 9H, N-CH3).
Preparation of fac-(PNP)M(CO)3, M ) W (9W-fac) and Mo
(9Mo-fac). Compound 9W was prepared in 79% yield (4.19 g) from
W(CO)3(CH3CN)3 (2.87 g, 7.34 mmol) in 45 mL of CH3CN and MeN-
(C2H4PPh2)222 (PNP) (3.35 g, 7.35 mmol) using a procedure similar to
that for the synthesis of 3W. 1H NMR (CDCl3): δ 2.53-3.10 (m,
8H, P-CH2-CH2-N), 3.06 (s, 3H, CH3-N), 6.84-7.84 (m, 20H, Ph).
31P{H} NMR (CDCl3): δ 33.07 (s).
) 92-88% for the Mo system) were determined using Beer plots for
pure 9W-fac and 9Mo-fac and absorbances of the ν(CO) peaks in DCE
at 1922 cm-1 (for 9W-fac) and 1929 cm-1 (for 9Mo-fac) of the
mixtures.
Protonation Reactions. Compounds 7W, 7Mo, 8W, 8Mo, 9W-
fac, 9Mo-fac, 9W-mer, and 9Mo-mer were protonated for spectroscopic
characterization by dissolving approximatively 5 mg of the complex
in 0.7 mL of either CD2Cl2 or CDCl3 (for NMR) or in CH2Cl2 (for IR)
in an NMR tube under nitrogen. To the solution was added 1 equiv of
CF3SO3H using a gastight microliter syringe. The solutions im-
mediately changed color from yellow to pale yellow with the exception
of 9W-mer and 9Mo-mer, which were orange and turned pale yellow.
The facial isomers (9W-fac and 9Mo-fac) and the meridional isomers
(9W-mer and 9Mo-mer) gave the same protonated complex (9WH+
and 9MoH+) upon addition of 1 equiv of CF3SO3H. Yields of the
Compound 9Mo-fac was prepared by following the procedure for
the synthesis of 9W-fac, in 76% yield (4.13 g) from Mo(CO)3(CH3-
22
CN)3 (2.58 g, 8.51 mmol) in 30 mL of CH3CN and MeN(C2H4PPh2)2
(PNP) (3.88 g, 8.51 mmol). 1H NMR (CD2Cl2): δ 2.48-2.82 (m, 8H,
P-CH2-CH2-N), 2.86 (s, 3H, CH3-N), 6.89-7.82 (m, 20H, Ph).
31P{H} NMR (CD2Cl2): δ 42.40 (s). Anal. Calcd for C32H31NO3P2-
Mo: C, 60.48; H, 4.92; N, 2.20. Found: C, 60.25; H, 5.15; N, 2.13.
Preparation of mer-(PNP)M(CO)3, M ) W (9W-mer) and Mo
1
protonated products were determined to be quantitative by IR and H
NMR or 31P{H} NMR spectroscopy. No precipitates formed, and no
unidentified bands were present in the 1H NMR spectra which indicates
that no other products were formed. Attempts to isolate any of the
(L3)M(CO)3(H)+ complexes were unsuccessful due to their air-
sensitivity. To our knowledge, the only previously reported (L3)M-
(CO)3(H)+ complexes in addition to those in eqs 3 and 4 are (1,4,7-
triazacyclononane)M(CO)3(H)+ (M ) W, Mo), which were characterized
by their elemental analyses and infrared spectra.23 1H NMR and 31P-
{H} NMR data for the protonated complexes at room temperature
except where indicated otherwise are given below. IR data are
presented in Table 1 for the W complexes and in Table 2 for the Mo
complexes.
-
(9Mo-mer). When the protonated complexes 9WH+CF3SO3 and
9MoH+CF3SO3- are deprotonated with 1 equiv of 1,3-diphenylguani-
dine base (DPG) in CH2Cl2 or DCE solvent, the reactions are rapid
and quantitative, but they yield in both cases the meridional isomers,
which are orange compounds as compared with the yellow facial
isomers. Attempts to isolate the pure mer isomers from these solutions
by chromatography on alumina eluting with CH2Cl2 yielded only
mixtures of facial and meridional isomers; the amounts of the facial
isomers increased the longer time the solution was on the column.
Chromatography on silica gel afforded almost exclusively the facial
isomer. In order to obtain mixtures as rich as possible in the meridional
isomers, the solutions of 9WH+CF3SO3 (4.55 g, 5.21 mmol) and
-
[(NSNc)W(CO)3(H)](CF3SO3) (7WH+CF3SO3-). 1H NMR
9MoH+CF3SO3 (4.20 g, 5.35 mmol) in 20 mL of CH2Cl2 were
-
3
(CD2Cl2): δ -4.45 (s, 1H,W-H), 1.36 [t, JHH ) 7.5 Hz, 6H, CH3
deprotonated with 1 equiv of DPG (1.10 g, 5.21 mmol, for
9WH+CF3SO3- and 1.13 g, 5.35 mmol, for 9MoH+CF3SO3-) and then
chromatographed under N2 on neutral alumina columns (1.5 × 5 cm)
that had been previously rinsed with a dilute solution of NMe3 in CH2-
Cl2. Solutions of the complexes eluting from the columns were
concentrated under vacuum; addition of ether gave orange precipitates
that were filtered out and dried under vacuum. Yield ) 81% (3.05g)
for the 9W-mer + 9W-fac mixture, and yield ) 85% (2.89 g) for the
9Mo-mer + 9Mo-fac mixture. Spectroscopic data for 9W-mer are as
follows: 1H NMR (CDCl3): δ 2.72 (s, 3H, CH3-N), 2.63-3.40 (m,
8H, P-CH2-CH2-N), 7.27-7.79 (m, 20H, Ph). 31P{H} (CDCl3): δ
41.46 (s). Spectroscopic data for 9Mo-mer are as follows: 1H NMR
(CD2Cl2): δ 2.48 (s, 3H, CH3-N), 2,68-3.31 (m, 8H, P-CH2-CH2-
N), 7.14-7.79 (m, 20H, Ph). 31P{H} (CD2Cl2): δ 59.51 (s). The
compositions (mole percent) of these mixtures (9W-fac ) 1-3%, 9W-
mer ) 99-97% for the W system and 9Mo-fac ) 8-12%, 9Mo-mer
(N-Et)], 2.82-2.92 (m, 2H), 3.16-3.25 (m, 6H), 3.38-3.52 (m, 4H),
3.58-3.81 (m, 4H).
[(NSNc)Mo(CO)3(H)](CF3SO3) (7MoH+CF3SO3-). 1H NMR
3
(CD2Cl2): δ -5.12 (s, 1H, Mo-H), 1.37 [t, JHH ) 7.5 Hz, 6H, CH3
(N-Et)], 2.51-2.81 (m, 8H), 2.98-3.10 (m, 4H), 3.27-3.45 (m, 4H).
[(NNNc)W(CO)3(H)](CF3SO3) (8WH+CF3SO3-). 1H NMR
(CDCl3): δ -4.10 (s, 1H, W-H), 3.06-3.16 (m, 6H, CH2-N), 3.21-
3.31 (m, 6H, CH2-N), 3.39 (s, 9H, CH3N).
[(NNNc)Mo(CO)3(H)](CF3SO3) (8MoH+CF3SO3-). 1H NMR
(CD2Cl2): δ -4.87 (s, 1H, Mo-H), 2.99-3.09 (m, 6H, CH2-N), 3.17-
3.27 (m, 15H, CH2-N and CH3-N).
(22) Sacconi, L.; Morassi, R. J. Chem. Soc. A 1968, 2997.
(23) Chaudhury, P.; Wieghardt, K.; Tsai, Y.-H.; Kru¨ger, C. Inorg. Chem.
1984, 23, 427.

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 437
avg ν(CO), cm-1
Table 2. IR Data for (L3)Mo(CO)3 and (L3)Mo(CO)3(H)+ Complexes in CH2Cl2 Solution
ν(CO), cm-1
complex
(SSSc)Mo(CO)3, 1Mo
1936 (s)
2040 (s)
1937 (s, br)
2040 (s)
1930 (s)
2037 (s)
1932 (s)
2038 (s)
1924 (s)
2030 (s)
1920 (s)
2028 (s)
1914 (s)
2022 (s)
1905 (s)
2018 (s)
1929 (s)
1958 (w)
2039 (m)
1827 (s, br)
1963 (s)
1839 (s)
1989 (s)
1820 (s, br)
1967 (s)
1836 (s)
1970 (s)
1800 (s, br)
1954 (s)
1796 (s, br)
1952 (s)
1784 (s, br)
1941 (s)
1765 (s, br)
1930 (s)
1828 (s)
1849 (vs)
1939 (vs)
1863
(SSSc)Mo(CO)3H+, 1MoH+
(SPSc)Mo(CO)3, 2Mo
1936 (sh)
1824 (s)
1952 (sh)
1867
1857
1859
1841
1837
1827
1812
(SPSc)Mo(CO)3H+, 2MoH+
(SSS)Mo(CO)3, 3Mo
(SSS)Mo(CO)3H+, 3MoH+
(SPS)Mo(CO)3, 4Mo
1951 (sh)
1810 (s)
1955 (sh)
(SPS)Mo(CO)3H+, 4MoH+
(SNSc)Mo(CO)3, 5Mo
(SNSc)Mo(CO)3H+, 5MoH+
(SNS)Mo(CO)3, 6Mo
1938 (sh)
1936 (sh)
1921 (sh)
(SNS)Mo(CO)3H+, 6MoH+
(NSNc)Mo(CO)3, 7Mo
(NSNc)Mo(CO)3H+, 7MoH+
(NNNc)Mo(CO)3, 8Mo
(NNNc)Mo(CO)3H+, 8MoH+
fac-(PNP)Mo(CO)3, 9Mo-fac
mer-(PNP)Mo(CO)3, 9Mo-mer
(PNP)Mo(CO)3H+, 9MoH+
1911 (sh)
1804 (s)
1803 (m)
[(PNP)W(CO)3(H)](CF3SO3) (9WH+CF3SO3-).
1H NMR
(C-Et)], [1.35 (t, 3JHH ) 7.5 Hz), 1.39 (t, 3JHH ) 7.5 Hz), 3H for both
isomers, CH3(N-Et)], 1.86-3.82 (m, 15H, CH2-S and CH2-N).
Isomer ratio ∼ 1:8.
(CDCl3): δ -4.71 (dd, 2JHP1 ) 20 Hz, 2JHP2 ) 47 Hz, 1H, W-H), 3.04-
4.11 (m, 8H, P-CH2-CH2-N), 7.45-7.76 (m, 20H, Ph). 31P{H}
2
2
[(SNSc)Mo(CO)3(H)](CF3SO3) (5MoH+CF3SO3-). 1H NMR
(CD2Cl2): δ [-5.16 (s), -5.09 (s), 1H for both isomers, Mo-H], [1.14
(CDCl3): δ 41.81 (d, JP1P2 ) 76 Hz), 31.40 (d, JP2P1 ) 76 Hz).
[(PNP)Mo(CO)3(H)](CF3SO3) (9MoH+CF3SO3-). 1H NMR (CD2-
2
2
3
3
Cl2) (-80 °C) : δ -5.3 (dd, JHP1 ) 20 Hz, JHP2 ) 40 Hz, 1H, Mo-
H), 2.82-3.79 (m, 8H, P-CH2-CH2-N), 7.45-7.83 (m, 20H, Ph).
(t, JHH ) 7.5 Hz), 1.17(t, JHH ) 7.5 Hz), 3H for both isomers, CH3-
(C-Et)], [1.34 (t, 3JHH ) 7.5 Hz), 1.35 (t, 3JHH ) 7.5 Hz), 3H for both
isomers, CH3(N-Et)], 1.58-3.58 (m, 15H, CH2-S and CH2-N).
Isomer ratio ∼ 1:1.
31P{H} (CD2Cl2): δ 58.73 (d, JP1P2 ) 80 Hz), 46.68 (d, JP2P1 ) 80
2
2
Hz).
[(SNS)W(CO)3(H)](CF3SO3) (6WH+CF3SO3-).
1H NMR
Compounds 1W-6W and 1Mo-6Mo were protonated similarly,
but as indicated by the spectroscopic data, the protonations were not
complete with 1 equiv of CF3SO3H. However, 3 equiv of CF3SO3H
gave complete protonation in all cases. 1H NMR and 31P{H} NMR
data of 1WH+-6WH+ and 1MoH+-6MoH+ are listed below; IR data
are presented in Table 1 for the W complexes and in Table 2 for the
Mo complexes.
3
(CD2Cl2): δ -4.47 (s, 1H, W-H), 1.37 (t, JHH ) 7.5 Hz, 3H, CH3-
(N-Et)), 1.48 (t, 3JHH ) 7.5 Hz, 6H, CH3(S-Et)), 3.01-3.41 (m, 12H,
3
S-CH2 and N-CH2), 3.76 (q, JHH ) 7.5 Hz, 2H, CH2(N-Et).
[(SNS)Mo(CO)3(H)](CF3SO3) (6MoH+CF3SO3-). 1H NMR
3
(CD2Cl2): δ -4.89 (s, 1H, Mo-H), 1.37 (t, JHH ) 6 Hz, 3H, CH3-
(N-Et)), 1.49 (t, 2JHH ) 7.5 Hz, 6H, CH3(S-Et)), 2.93-3.35 (m, 12H,
[(SSSc)W(CO)3(H)](CF3SO3) (1WH+CF3SO3-). 1H NMR
(CD2Cl2): δ [-4.16 (s), -4.15 (s), 1H for both isomers, W-H], [1.20
3
S-CH2 and N-CH2), 3.66 (q, JHH ) 6 Hz, 2H, CH2(N-Et).
Calorimetric Studies. Heats of protonation (∆HHM) for complexes
7W, 7Mo, 8W, 8Mo, 9W-fac, and 9Mo-fac, which were protonated
by 1 equiv of acid, were determined under an argon atmosphere by
titration with CF3SO3H in DCE solution at 25.0 °C, using a Tronac
model 458 isoperibol calorimeter as originally described24 and then
modified.25 A typical calorimetric run consisted of three sections:26
initial heat capacity calibration, titration, and final heat capacity
calibration. Each section was preceded by a baseline acquisition period.
During the titration, about 1.2 mL of an approximately 0.1 M CF3-
SO3H solution (standardized to a precision of (0.0002 M) in DCE
was added at a rate of 0.3962 mL/min to 50 mL of a 2.6 mM solution
of the complex (5-10% excess) in DCE at 25.0 °C.
Two separate standardized acid solutions were used for the deter-
mination of the ∆HHM of each complex. The reported values are the
average of at least four titrations and as many as five. The reaction
enthalpies were corrected for the heat of dilution (∆Hdil)) of the acid in
DCE (-0.2 kcal/mol).25 The reported error in ∆HHM is the average
deviation from the mean of all determinations. Titrations of 1,3-
diphenylguanidine (GFS Chemicals) with CF3SO3H in DCE (-36.9 (
0.3 kcal/mol; lit.24 -37.2 ( 0.4 kcal/mol) were used to monitor the
performance of the calorimeter before each set of determinations.
Heats of protonation for complexes 1W-6W were determined by a
modified titration procedure which involved adding excess acid (3
equiv) to the complex solution during the 3-min titration period to
ensure complete protonation. A typical calorimetric run consisted of
the same three sections: initial heat capacity calibration, titration, and
3
3
(t, JHH ) 7.5 Hz), 1.23 (t, JHH ) 7.5 Hz), 3H for both isomers, CH3
(C-Et)], 1.78-4.11 (m, 13H, CH2-S). Isomer ratio ∼ 1:2.
[(SSSc)Mo(CO)3(H)](CF3SO3) (1MoH+CF3SO3-). 1H NMR
(CD2Cl2): δ [-4.33(s), -4.32 (s), 1H for both isomers, Mo-H], [1.18
(t, 3JHH ) 7.5 Hz), 1.21 (t, 3JHH ) 7.5 Hz), 3H for both isomers, CH3-
(C-Et)], 1.79-3.93 (m, 13H, CH2-S). Isomer ratio ∼ 1:2.
[(SPSc)W(CO)3(H)](CF3SO3) (2WH+CF3SO3-). 1H NMR
(CD2Cl2): δ -4.67 (d, 2JHP ) 20 Hz, 1H, W-H), 2.25-3.30 (m, 12H,
CH2S and CH2P), 7.61-7.81 (m, 5H, Ph). 31P{H} (CD2Cl2): δ 88.61
(s).
[(SPSc)Mo(CO)3(H)](CF3SO3) (2MoH+CF3SO3-). 1H NMR
(CD2Cl2): δ -4.67 (d, 2JHP ) 20 Hz, 1H, W-H), 2.30-3.26 (m, 12H,
CH2S and CH2P), 7.59-7.79 (m, 5H, Ph). 31P{H} (CD2Cl2): δ 95.26
(s).
[(SSS)W(CO)3(H)](CF3SO3) (3WH+CF3SO3-). 1H NMR (CD2-
Cl2): δ -4.75 (s, 1H, W-H), 1.50 (t, 3JHH ) 7.5 Hz, 6H, CH3), 2.89-
3.30 (m, 12H, S-CH2).
[(SSS)Mo(CO)3(H)](CF3SO3) (3MoH+CF3SO3-). 1H NMR
(CD2Cl2): δ -4.77 (s, 1H, Mo-H), 1.51 (t, 3JHH ) 7.5 Hz, 6H, CH3),
2.95-3.23 (m, 12H, S-CH2).
[(SPS)W(CO)3(H)](CF3SO3) (4WH+CF3SO3-). 1H NMR (CD2-
Cl2): δ -5.07 (d, 2JHP ) 22 Hz, 1H, W-H), 1.49 (t, 3JHH ) 8 Hz, 6H,
CH3), 2.17-3.22 (m, 12H, CH2S and CH2P), 7.63-7.91 (m, 5H, Ph).
31P{H} (CD2Cl2): δ 78.88 (s).
[(SPS)Mo(CO)3(H)](CF3SO3) (4MoH+CF3SO3-). 1H NMR
2
3
(CD2Cl2): δ -5.14 (d, JHP ) 26 Hz, 1H, W-H), 1.50 (t, JHH ) 8
Hz, 6H, CH3), 2.19-3.20 (m, 12H, CH2S and CH2P), 7.64-7.89 (m,
5H, Ph). 31P{H} (CD2Cl2): δ 85.13 (s).
(24) Bush, R. C.; Angelici, R J. Inorg. Chem. 1988, 27, 681.
(25) Sowa, J. R., Jr.; Angelici, R. J. J. Am. Chem. Soc. 1991, 113, 2537.
(26) Eatough, D. J.; Christensen, J. J.; Izatt, R. M. Experiments in
Thermometric and Titration Calorimetry; Brigham Young Univer-
sity: Provo, UT, 1974.
[(SNSc)W(CO)3(H)](CF3SO3) (5WH+CF3SO3-). 1H NMR
(CD2Cl2): δ [-4.66 (s), -4.56 (s), 1H for both isomers, W-H], [1.15
(t, 3JHH ) 7.5 Hz), 1.19 (t, 3JHH ) 7.5 Hz), 3H for both isomers, CH3-

438 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
final heat capacity calibration, each section being preceded by a baseline
acquisition period. During the titration, about 1.2 mL of an ap-
proximately 0.3 M CF3SO3H solution (standardized to a precision of
(0.001 M) in DCE was added at a rate of 0.3962 mL/min to 50 mL
of an approximately 2.4 mM solution of the complex in DCE at 25.0
°C. Because the voltage/time data for the titration period were, as
expected, not linear for these complexes, only the initial and the final
titration points were used for calculation of the total reaction heat. The
reaction enthalpies (∆Hexp), obtained as averages of at least four
titrations with two separately standardized acid solutions, were corrected
for the heat of dilution of the acid and the heat of hydrogen bonding
(see below) between the triflate anion of the protonated complex and
the excess triflic acid (eq 9). The magnitude of the correction for
dilution and hydrogen bonding was determined by titrating 9Mo-fac
with 3 equiv of acid using the same conditions as for complexes 1W-
6W and then by subtracting the previously determined ∆HHM(9Mo-
fac) from the average value [∆Hexp(9Mo-fac)] of four titrations with
two different excess acid solutions (eqs 7 and 8). The error in
hydrogen bonding between the triflate anion of the protonated complex
and the excess acid. The heat of protonation (∆Hexp) of the three-
component mixture was determined using the value at the point where
the change of slope occurred, i.e., when the complexes and DPG were
completely protonated. This heat of reaction was then corrected for
acid dilution (∆Hdil ) -0.2 kcal/mol) to give ∆Htotal (eq 10).
∆Htotal ) ∆Hexp - ∆Hdil ) ∆Hexp - (-0.2 kcal/mol) (10)
∆Htotal ) (% molfac)∆Hfac + (% molmer)∆Hmer
+
(% molDPG)∆HDPG (11)
∆Hmer ) {∆Htotal - [(% molfac)∆Hfac
+
(% molDPG)∆HDPG]}/% molmer (12)
Subtraction of the contributions of the DPG (∆H ) -36.9 kcal/mol)
and the 9W-fac (∆HHM ) -22.1 kcal/mol) or 9Mo-fac (∆HHM ) -19.2
kcal/mol) from the overall heat of reaction yielded ∆HH9W-mer and
∆HHbond+dil ) ∆Hexp(9Mo-fac) - ∆HHM(9Mo-fac)
∆HHbond+dil ) -32.4(2) - [-19.2(3)] ) -13.2(5) kcal/mol
∆HHM(1W-6W) ) ∆Hexp(1W-6W) - ∆HHbond+dil
(7)
(8)
(9)
∆HH9Mo-mer, respectively (eq 12). The experimental values (∆Hexp
)
varied from run to run due to the differences in the compositions of
the mixtures but were in the range 24.9-25.7 kcal/mol for the W
complexes and 24.5-25.6 kcal/mol for the Mo compounds. The
reported values are the average of at least four titration runs with two
separately standardized acid solutions. The reported error is the average
deviation from the mean of all determinations.
∆HHbond+dil is the sum of the errors in ∆Hexp(9Mo-fac) and ∆HHM(9Mo-
fac). The values of ∆Hexp (kcal/mol) for complexes 1W-6W are given
in brackets: 1W [-23.8(3)], 2W [-24.7(3)], 3W [-25.7(3)], 4W
[-26.5(1)], 5W [-27.4(3)], 6W [-28.6(4)]. The reported error in
∆HHM for complexes 1W-6W is the sum of the average deviation
from the mean of all the determinations of ∆Hexp and the calculated
Results
Syntheses of the (L3)M(CO)3 Complexes. Both the mo-
lybdenum and tungsten complexes were prepared by following
the same procedures. All the complexes with noncyclic ligands
were synthesized by the reaction of the ligand L3 and
M(CO)3(CH3CN)3 (M ) Mo, W) in acetonitrile at room
temperature (eq 13).
error in ∆HHbond+dil
.
As noted above, the titration of 9Mo-fac which requires only 1 equiv
of CF3SO3H for complete protonation continues to liberate heat when
one and two more equivalents of CF3SO3H are added. We attribute
this additional heat to the heat of dilution of the acid and to the heat of
-
hydrogen bonding between the CF3SO3 anion generated when 9Mo-
+ L CH CN8 (L3)M(CO)3
(13)
fac was protonated and the free CF3SO3H acid (∆HHbond+dil ) -13.2
kcal/mol). There is evidence in the literature that [CF3SO3H‚‚‚O3SCF3]-
hydrogen bonding occurs in nonpolar solvents. Bullock and co-
3
M(CO)3(CH3CN)3
3
rt
M ) W, Mo
1
workers27 observed two CF3SO3H resonances in the H NMR spectra
of mixtures of [Cp*Os(CO)2(H)2]+CF3SO3 and of [Cp*Os(CO)2(η2-
-
In the series of complexes with cyclic ligands, compounds
8W20 and 8Mo21 were prepared as described in the literature
by refluxing W(CO)6 or Mo(CO)6 with the ligand in decalin
solvent. The syntheses of complexes 7W and 7Mo were based
on that used for (1-thia-4,7-diazacyclononane)Mo(CO)319b which
proved to be extremely insoluble in either CH2Cl2 or DCE and
therefore not suitable for calorimetric titrations. In order to
increase the solubility, complexes of alkylated 1-thia-4,7-
diazacyclononane were prepared. Methylation of the diamine
with a mixture of formic acid and formaldehyde, as described
for the preparation of tetramethylcyclam from cyclam,28 led to
the formation of the still insoluble (4,7-dimethyl-thia-4,7-
diazacyclononane)Mo(CO)3. However, compounds 7W and
7Mo, obtained by ethylation of 1-thia-4,7-diazacyclononane with
acetaldehyde in the presence of NaBH3CN followed by reaction
of the ligand with M(CO)3(CH3CN)3 (M ) Mo, W), proved to
have satisfactory solubilities for the calorimetric studies.
H2)]+ CF3SO3 in the presence of excess acid in CD2Cl2 solvent and
-
-
assigned them to CF3SO3H hydrogen bonded to itself and to CF3SO3
as [CF3SO3H‚‚‚O3SCF3]-. A CD2Cl2 solution of equimolar CF3SO3H
and [(Ph3P)2N]+CF3SO3 also shows 1NMR evidence for the
-
[CF3SO3H‚‚‚O3SCF3]- species.27
Because 9W-mer and 9Mo-mer could not be completely separated
from their fac-isomers, their heats of protonation were determined by
calorimetric titration of three-component mixtures with known com-
positions of 9W-mer + 9W-fac + DPG and 9Mo-mer + 9Mo-fac +
DPG, respectively. During the 3-min titration period, about 1.2 mL
of an approximately 0.1 M CF3SO3H solution (standardized to a
precision of (0.0002 M) in DCE was added at a rate of 0.3962 mL/
min to 50 mL of a 1.8 mM solution of the mixture in DCE. A small
excess of acid (20-25%) was used in order to ensure complete
protonation of all the species. The compositions (mole percent) of the
mixtures varied from run to run but were approximately as follows:
9W-mer, ) 85-92%; 9W-fac, ) 3-1%; DPG, 12-7% for the W
system; 9Mo-mer, 70-80%; 9Mo-fac, 15-8%; DPG, 15-12% for the
Mo system. The exact composition of each mixture was determined
by IR intensities (ν(CO) for fac) and weight (mer + fac and DPG).
The extinction coefficients in DCE for 9W-fac and 9Mo-fac were
428 M-1 mm-1 (at 1922 cm-1) and 480 M-1 mm-1 (at 1929 cm-1),
respectively. During the titration, there is a clear change in slope in
the voltage/time data after complete protonation of the complexes and
DPG. The heat that is given off beyond that point is attributed to
The other members of the series of complexes with cyclic
ligands were prepared by a method that was initially used for
the synthesis of (1,4,7-trithiacyclononane)Mo(CO)3.15 The
ligand precursor, a dithiolate, is attached as a dianion to the
metal-tricarbonyl fragment, and the macrocycle is then closed
with a 1,2-dibromoalkane, using the metal-tricarbonyl as a
(27) Bullock, R. M.; Song, J.-S.; Szalda, D. J. Organometallics 1996, 15,
(28) (a) Barefield, E. K.; Wagner, F. Inorg. Chem. 1973, 12, 2435. (b)
Buxtorf, R.; Kaden, T. A. HelV. Chim. Acta 1974, 57, 1035.
2504.

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 439
template (eq 14). The synthesis of (SPSc)M(CO)3 (2W, 2Mo)
Scheme 1
was based on this approach.29 Complexes (1,4,7-trithiacy-
clononane)Mo(CO)315 and its readily synthesized W analog were
not sufficiently soluble in DCE to be studied calorimetrically.
In order to increase the solubility, complexes 1W and 1Mo were
prepared, each as a mixture of isomers, by the method described
in the literature for the preparation of carbon-substituted 1,4,7-
trithiacyclononanes14 that uses 1,2-dibromobutane instead of 1,2-
dibromoethane to close the ligand cycle. The two isomers differ
by the orientation of the ethyl group, which points toward or
away from the metal. Their ratio is approximately 1:2 for both
1
the W and Mo complexes, as indicated by H NMR data, but
complexes have an octahedral geometry with the carbonyl
ligands occupying mutually cis positions. The fac geometry is
supported by the IR spectra (Tables 1 and 2) which show a
strong, sharp ν(CO) A1 band at 1895-1937 cm-1 and a broad
E band at 1758-1839 cm-1. The latter band becomes extremely
broad or even splits into two bands when the symmetry of the
complex is lowered, especially in the case of tridentate ligands
with mixed donor atoms. These spectra are very similar to those
of the closely related complexes (1,4,7-trithiacyclononane)Mo-
(CO)3,30 (1-thia-4,7-diazacyclononane)Mo(CO)3,19b and (2,5,8-
trithianonane)Mo(CO)3,17 whose structures have been estab-
lished as being octahedral with facially coordinated L3 ligands
by X-ray diffraction studies. The only complexes of the series
that exist both as the fac- and mer- isomers are those with the
PNP ligand. Compounds 9W-mer and 9Mo-mer are thermo-
dynamically unstable and isomerize to the fac isomers when
refluxed in DCE or when traces of acid are present in their
solutions. The isomerization of related complexes of the type
fac-[(η2-dppm)(PR3) M(CO)3] and fac-[(η2-dppm)(η1-dppm-
)M(CO)3] (M ) W, Mo; dppm ) PPh2CH2PPh2, PR3 ) PEt2-
Ph, PEt3) to their mer isomers have been reported31 to be
catalyzed by acid; a cationic seven-coordinate hydride complex
is proposed as an intermediate. The mer isomers 9W-mer and
9Mo-mer could be isolated only as mixtures that contained small
amounts of the fac-isomers. The mer geometry of 9W-mer and
9Mo-mer was assigned on the basis of their weak (∼1955
the predominant isomer was not determined. The ν(CO) values
were the same for the two isomers, which suggests that the
orientation of the Et group does not significantly affect the
electron-richness of the metal.
Although the preparation of 7-aza-1,4-dithiacyclononane18 has
been reported, it was not chosen as a ligand in these studies
because of the difficulty in detosylating 1-tosyl-1-aza-4,7-
dithiacyclononane and the expected low solubility in CH2Cl2
or DCE of its molybdenum and tungsten tricarbonyl complexes.
The related compounds (7-ethyl-7-aza-1,4-dithiacyclononane)M-
(CO)3 (M ) Mo,W) were prepared by following the above-
described template synthesis (eq 14) using 1,2-dibromoethane
for the ring closure, but their solubilities were still too low. In
order to achieve a higher solubility, the cycle was closed with
1,2-dibromobutane, yielding compounds 5W and 5Mo, again
as a mixture of isomers. The preparation of the ligand precursor
EtN(C2H4SH)2 was adapted from the synthesis of TsN(C2H4-
SH)2.18 The ratio of the two isomers, as indicated by NMR
1
data, is approximately 1:8 for 5W and 1:1 for 5Mo, but H
NMR signals were not assigned to specific isomers. Infrared
data for the mixture of isomers show only one set of ν(CO)
values.
The introduction of ethyl groups on the C-bridge in the cyclic
ligands 1W, 1Mo, 5W, and 5Mo or replacement of the methyl
group with an ethyl group on the N in 5W, 5Mo, 7W, 7Mo,
8W, and 8Mo in order to increase the solubility of the metal
complexes in DCE is expected to have a minor effect on the
donor ability of the ligand. This is supported by the observation
that such substitutions do not measurably affect ν(CO) values
for the complexes. It has also been shown that complexes of
iron, cobalt, nickel, copper, and silver with 1,4,7-trithiacy-
clononane or 2-methyl-1,4,7-trithiacyclononane ligands have
stoichiometries, electronic spectra, and electrochemistries that
are not significantly affected by the presence of the methyl group
in the ring.14
Characterization of the (L3)M(CO)3 Complexes. The
identity and purity of the 1W-8W, and 1Mo-8Mo complexes
were established by elemental analysis, IR, 1H NMR, and 31P-
{H} NMR spectroscopy. The complexes are pale yellow to
yellow solids that are air-stable in the solid state for several
days, but upon longer exposure to air they show signs of
decomposition as indicated by a greenish coloration. All
cm-1), very strong (∼1844 cm-1), and medium (∼1800 cm-1
)
pattern of ν(CO) bands. The same pattern was reported for other
related mer-compounds, mer-[(η2-dppm)(η1-dppm)W(CO)3]32
[1958 (w), 1856 (s), 1833 (m) cm-1] and mer-[(η2-dppm)-
(PEt3)W(CO)3]31 [1954 (w), 1852 (vs), 1832 (m) cm-1].
Characterization of the Protonated Complexes (L3)M-
(CO)3(H)+. The hydride resonances observed in the 1H NMR
spectra of the complexes 1WH+-9WH+ and 1MoH+-9MoH+
indicate that the protonation occurs quantitatively at the metal
center with 1 equiv of CF3SO3H for compounds 7W, 7Mo, 8W,
8Mo, 9W-fac, 9Mo-fac, 9W-mer, and 9Mo-mer and with 2-3
equiv for compounds 1W-6W and 1Mo-6Mo. Protonation
of the mer compounds, 9W-mer and 9Mo-mer, gives the same
protonated forms, 9WH+ and 9MoH+, that are obtained by
protonation of the fac-isomers, 9W-fac and 9Mo-fac (Scheme
1).
(30) Ashby, M. T.; Lichtenberger, D. L. Inorg. Chem. 1985, 24, 636.
(31) Vila, J. M.; Shaw, B. L. J. Chem. Soc., Chem. Commun. 1987, 1778.
(32) Blagg, A.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1987, 221.
(29) Smith, R. J.; Powell, A. K.; Barnard, N; Dilworth, J. R.; Blower, P.
J. J. Chem. Soc., Chem. Commun. 1993, 54.

440 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
Scheme 2
Chart 2
ature in the hydride region consists of a quartet at -4.62 ppm
due to equal coupling (JPH ) 21 Hz) to the three P atoms. The
equivalence of the three P atoms suggests that the hydride is
fluxional. The phosphorus-containing compounds 2W and 2Mo
both have hydride signals at -4.67 with JPH ) 20 Hz. The
The ν(CO) pattern of compounds 9WH+ and 9MoH+,
medium (∼2035 cm-1), medium (∼1950 cm-1), and very strong
(∼1920 cm-1), is the same as that observed for other protonated
complexes of the type (L)3W(CO)3(H)+, where L ) PR3, such
1
similarities of the IR and H NMR spectra of [CH3C(CH2-
as (PEt3)3W(CO)3(H)+ [2022 (m), 1943 (m), 1908 (vs) cm-1 9
]
PPh2)3]W(CO)3(H)+ and the protonated complexes with cyclic
ligands suggest that they have the same structure (eq 15), a
and [PhP(C2H4PPh2)2]W(CO)3(H)+ [2038 (m), 1976 (m), 1918
(vs) cm-1].9 These phosphine-containing complexes were
proposed to have the pentagonal bipyramidal structure shown
1
in eq 3. The variable-temperature H NMR of [PhP(C2H4-
PPh2)2]W(CO)3(H)+ was investigated extensively.9 At low
temperature (-35 °C) the hydride resonance (-3.78 ppm) was
a doublet of doublets which was attributed to R- and â-isomers
(Scheme 2), where JPaHR ) JPcHâ ) 46 Hz and JPcHR ) JPaHâ
)
18 Hz. At higher temperatures (37 °C), hydride migration is
fast and the hydride resonance appears as a triplet as Pa and Pc
become equivalent on the NMR time scale.
Coupling of the hydride ligand with the axial phosphorus Pb
is not observed as Pb is perpendicular to the equatorial plane
that contains the hydride and JPbH ∼ 0.9 The 1H NMR spectrum
of 9WH+ is very similar to that of [PhP(C2H4PPh2)2]W(CO)3-
(H)+, but the hydride migration is slow at room temperature
and the hydride resonance (-4.71 ppm) appears as a doublet
pentagonal bipyramid with mutually cis CO ligands. Complexes
1W, 1Mo, 5W, and 5Mo each exist as mixtures of two isomers
resulting from the ethyl group being on the same or opposite
side of the ring from the metal. Upon protonation, each isomer
yields a separate protonated complex, as indicated by the two
1
hydride peaks in their H NMR spectra.
While the structures of the (L3)M(CO)3(H)+ complexes are
proposed to be based on a trigonal bipyramidal geometry, it is
possible that they have a 4:3 piano stool structure as was
established for the related cation [(L3)Mo(CO)3Br]+ (L3 ) 1,4,7-
triazacyclononane).23 The exact disposition of the ligand donor
atoms in these complexes is not known. Assuming a pentagonal
bipyramidal structure as in eq 15, it is possible to assign a
specific location to the X, Y, and Z atoms for complexes with
ligands that contain a P donor atom. This occurs only for
(SPSc)W(CO)3(H)+ (2WH+) and (SPSc)Mo(CO)3(H)+ (2MoH+),
for which the coupling constant between the P donor and hydride
ligand is JPH ) 20 Hz. This is very similar to that (18 Hz) for
the P-donor cis to the hydride in [PhP(C2H4PPh2)2]W(CO)3-
(H)+ (Scheme 2),9 which suggests that the P atom in 2WH+
and 2MoH+ occupies the Y position (eq 15) cis to the hydride
in the pentagonal plane.
of doublets with JPaHR ) JPcHâ ) 47 Hz and JPcHR ) JPaHâ
)
1
20 Hz. In the case of 9MoH+, the H NMR spectrum in the
hydride region at room temperature was poorly resolved.
However, at low temperature (-80 °C), the hydride resonance
(-5.3 ppm) appears as a doublet of doublets with JPaHR ) JPcHâ
) 40 Hz and JPcHR ) JPaHâ ) 20 Hz. The similarities in their
1
infrared and H NMR spectra lead to the conclusion that the
protonated complexes 9WH+ and 9MoH+ and [PhP(C2H4-
PPh2)2]W(CO)3(H)+ share the same structure, a pentagonal
bipyramid with the ligand having a nearly coplanar disposition
of the donor atoms.
In contrast to the structure for 9WH+ and 9MoH+, the
protonated complexes 1WH+-8WH+ and 1MoH+-8MoH+
have a different disposition of donor atoms in the trigonal
bipyramidal structure. The pattern of ν(CO) bands in their IR
spectra (Tables 1 and 2), a strong band at 2010-2040 cm-1, a
strong band at 1921-1989 cm-1, and a shoulder at 1902-1955
cm-1, is characteristic of three mutually cis CO ligands. The
positions of the ν(CO) bands in the protonated complexes are
100-150 cm-1 higher than those of the (L3)M(CO)3 compounds,
and this shift is consistent with an increase in the positive charge
on the metal in these formally divalent seven-coordinate species.
Complexes 1W, 2W, 5W, 7W, 8W, 1Mo, 2Mo, 5Mo, 7Mo,
and 8Mo with cyclic ligands have protonated forms that can
be compared with [CH3C(CH2PPh2)3]W(CO)3(H)+.9 All of
these complexes have rigid tridentate ligands that force the donor
atoms to be mutually cis. Compound [CH3C(CH2PPh2)3]W-
(CO)3(H)+ was proposed9 to have the pentagonal structure
shown in eq 4. The IR spectra of [CH3C(CH2PPh2)3]W(CO)3-
(H)+ [2025 (s), 1954 (s), 1941 (sh) cm-1]9 and of the protonated
complexes with cyclic ligands [∼2025 (s), ∼1940 (s), ∼1930
The protonated complexes 3WH+, 3MoH+, 4WH+, 4MoH+,
6WH+ and 6MoH+, which have noncyclic ligands, have the
same ν(CO) pattern as the protonated complexes with cyclic
ligands. Thus, they are assigned the same pentagonal bipyra-
midal structure (eq 15) with mutually cis CO ligands. There
are three possible structures of this type (Chart 2) that might
be considered for 3WH+, 3MoH+, 4WH+, 4MoH+, 6WH+,
and 6MoH+. In structure a (Chart 2), the hydride ligand is cis
to the central donor atom X; in structure b, these ligands are
approximately trans. In structure c, the bond is perpendicular
to the plane that contains the hydride ligand. Complexes 4WH+
and 4MoH+, that have X ) PPh, are the only compounds for
which the structure can be assigned on the basis of the coupling
constants JPH ) 22 Hz (4WH+) and JPH ) 26 Hz (4MoH+).
As discussed previously (Scheme 2), [PhP(C2H4PPh2)2]W(CO)3-
(H)+ has a structure in which the hydride lies in the pentagonal
plane cis to one P atom (JPH ) 18 Hz) and approximatively
trans to the other (JPH ) 46 Hz); JPH coupling to the phosphine
that is perpendicular to the pentagonal plane that contains the
hydride in [PhP(C2H4PPh2)2]W(CO)3(H)+ is unobservably small
1
(sh) cm-1] have the same pattern. The H NMR (CD2Cl2)
spectrum9 of [CH3C(CH2PPh2)3]W(CO)3(H)+ at room temper-

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 441
Table 3. Heats of Protonation (∆HHM) for the (L3)W(CO)3
Complexes
Chart 3
compd
-∆HHM, kcal/mol
(SSSc)W(CO)3, 1W
(SPSc)W(CO)3, 2W
(SSS)W(CO)3, 3W
(SPS)W(CO)3, 4W
(SNSc)W(CO)3, 5W
(SNS)W(CO)3, 6W
(NSNc)W(CO)3, 7W
(NNNc)W(CO)3, 8W
fac-(PNP)W(CO)3, 9W-fac
mer-(PNP)W(CO)3, 9W-mer
fac-[PhP(C2H4PPh2)2]W(CO)3
[MeC(CH2PPh2)3]W(CO)3
10.6(8)a
11.5(8)a
12.5(8)a
13.3(6)a
14.2(8)a
15.4(9)a
17.4(1)b
20.9(3)b
22.1(1)b
24.1(2)b
16.7(1)c
10.5(1)c
(<2 Hz). On the basis of these coupling constants and the 22
and 26 Hz JPH coupling constants for compounds 4WH+ and
4MoH+, we assign structure a (Chart 2), in which the hydride
and P donor groups are cis to each other, to complexes 4WH+
and 4MoH+. It is not possible to assign a specific structure a,
b, or c to 3WH+, 3MoH+, 6WH+, or 6MoH+.
a Protonation with excess (3 equiv) of CF3SO3H in DCE solvent at
25.0 °C. Numbers in parentheses are average deviations of at least 4
determinations, as described in the text. b Protonation with 1 equiv of
CF3SO3H in DCE solvent at 25.0 °C. Numbers in parentheses are
average deviations from the mean of at least four titrations. c Reference
8.
To summarize the structures of the (L3)M(CO)3(H)+ com-
plexes, the protonated forms 1WH+-8WH+ and 1MoH+-
8MoH+ are assigned a pentagonal bipyramidal structure with
mutually cis ligand donor atoms and mutually cis CO ligands
(structure A, Chart 3) and complexes 9WH+ and 9MoH+
a
Table 4. Heats of Protonation (∆HHM) for the (L3)Mo(CO)3
Complexes
pentagonal bipyramidal structure with the ligand donor atoms
approximately coplanar with the metal and the CO ligands also
approximately coplanar with the metal (structure B, Chart 3).
compd
-∆HHM,a kcal/mol
(NSNc)Mo(CO)3, 7Mo
(NNNc)Mo(CO)3, 8Mo
fac-(PNP)Mo(CO)3, 9Mo-fac
mer-(PNP)Mo(CO)3, 9Mo-mer
14.6(3)
18.3(3)
19.2(3)
24.0(3)
Complexes 1H+-8H+ and 1MoH+-8MoH+ are deproto-
nated rapidly (< 5 s) by 1,3-diphenylguanidine base (DPG) in
CH2Cl2 or DCE solvent. This deprotonation gives back
complexes 1W-8W and 1Mo-8Mo as their fac isomers, which
are recovered by chromatography on an alumina column (1.5
× 8 cm) by elution with CH2Cl2 and recrystallization of the
chromatographed product from CH2Cl2/ether. In the cases of
9WH+ and 9MoH+, deprotonation with DPG yields isomers
that are different than the starting 9W-fac and 9Mo-fac (Scheme
1). Addition of 1 equiv of DPG to a solution of 9WH+ or
9MoH+ in DCE or CH2Cl2 generates immediately the deep
orange color of 9W-mer and 9Mo-mer, as opposed to the yellow
color of 9W-fac or 9Mo-fac solutions. Chromatography on
alumina eluting with CH2Cl2 yields a mixture of the two isomers,
but when silica gel is used, the final mixture contains almost
exclusively the fac isomer. The mer complexes proved to be
stable at room temperature in the solid state; however, in
solution, heat (refluxing in DCE) or even traces of acid trigger
the mer-to-fac isomerization. However, in the presence of small
amounts (1-2%) of basic DPG, which presumably scavenges
any acidic species, the mer isomers are stable in solution at
room temperature for several hours.
a Protonation with 1 equiv of CF3SO3H in DCE solvent at 25.0 °C.
Numbers in parentheses are average deviations from the mean of at
least four titrations.
Discussion
Thioether and Tertiary Amine Ligand Effects on Metal
Basicity (-∆HHM). The heats of protonation of the series of
(L3)W(CO)3 complexes with tridentate cyclic ligands containing
S and N donor atoms (Table 3) increase with the ligand in the
following order (-∆HHM, kcal/mol, in parentheses): SSSc
(10.6) < SNSc (14.2) < NSNc (17.4) < NNNc (20.9). In an
effort to understand this trend, one might consider the relative
electron-donating abilities of the S and NR (R ) Me or Et)
groups. As discussed in the Introduction, the basicities (-∆HHM
)
of metal complexes increase linearly with the basicities (as
measured by -∆HHP or pKa) of their phosphine ligands. Thus
one might expect the pKa’s of the conjugate acids of thioethers
(Et2S) and amines (Et2N-R) to be a guide to the donor abilities
of the S and NR groups in the cyclic ligands and to the -∆HHM
values of their complexes. The pKa of the conjugate acid of
Et3N is 10.8033 while that of Et2S34 is only -6.8. Therefore,
we expect the basicities (-∆HHM) of the complexes to increase
as S donors are replaced by NR, as observed. Replacement of
each S by a NR group (R ) Me or Et) increases the basicity by
3.4 ( 0.2 kcal/mol; this value is the same within experimental
error whether it is for the first S atom, the second, or the third.
The same trend is observed in the molybdenum series (Table
4), although only two ∆HHM values are available for comparison.
Thus, the increase in basicity (-∆HHM) for replacement of a S
by NR in (NSNc)Mo(CO)3 (7Mo) (14.6 kcal/mol) to give
(NNNc)Mo(CO)3 (8Mo) (18.3 kcal/mol) is 3.7 kcal/mol, which
is the same within experimental error as for the W series (3.4
kcal/mol). That replacement of ligand donor groups affects the
Calorimetry Studies. Heats of protonation (∆HHM) deter-
mined by calorimetric titration of the complexes, 1W-8W, 9W-
fac, 9W-mer, and 7Mo, 8Mo, 9Mo-fac, and 9Mo-mer, with
CF3SO3H in DCE solvent at 25.0 °C are presented in Tables 3
and 4. For the compounds that protonated completely with 1
equiv of acid (7W, 8W, 9W-fac, 7Mo, 8Mo, and 9Mo-fac)
the plots of temperature Vs amount of acid added were linear.
There was no decomposition of either the neutral or protonated
complexes during the titration experiment, as evidenced by
normal pre-and post-titration baseline slopes. For the complexes
that required excess acid (3 equiv) for complete protonation,
plots of temperature vs amount of acid added were not linear,
as expected for equilibrium protonations, but the normal pre-
and post-titration baseline slopes indicated that there was no
decomposition during the experiment. However, this was not
the case with the molybdenum complexes 1Mo-6Mo, and we
were unable to obtain reproducible ∆HHM values for these
compounds.
(33) Perrin, D. D. Dissociation Constants of Organic Bases in Aqueous
Solution; Butterworths: London, 1972.
(34) Arnett, E. M.; Mitchell, E. J.; Murty, T. S. S. R. J. Am. Chem. Soc.
1974, 96, 3875.

442 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
Thus, for reasons that are not known at this time, the replace-
ment of PPh by NEt increases the basicity by 5.4 kcal/mol when
the protonated products have structure B (Scheme 1 and Chart
3) but only 2.7 or 2.1 kcal/mol when the products have structure
A (Chart 3).
Phosphine and Thioether Ligand Effects on Metal Basicity
(-∆HHM). Replacement of a S donor atom by a PPh in
(SSSc)W(CO)3 (1W) (10.6 kcal/mol) to give (SPSc)W(CO)3
(2W) (11.5 kcal/mol) increases the basicity of the metal only
very slightly (by 0.9 kcal/mol). A comparison of -∆HHM for
1W with that of the related fac-[MeC(CH2PPh2)3]W(CO)3 (10.5
kcal/mol),6,8 which has a ligand that also requires the three donor
atoms to be fac-coordinated in the reactant and protonated
product, shows that they have the same -∆HHM values within
experimental error. Thus, in this latter comparison, the thioether
and PPh2 donor groups have about the same effect on the
basicity of the metal assuming the different structures of the
SSSc and MeC(CH2PPh2)3 ligands do not affect the -∆HHM
values significantly.
Figure 1. Correlation (eq 16) of the average value of the ν(CO) bands
of (L3)W(CO)3 complexes (Table 1) with the average value of the ν-
(CO) bands of (L3)Mo(CO)3 complexes (Table 2).
In the noncyclic ligand series, replacement of the central S
(structure a, Chart 2) in (SSS)W(CO)3 (3W) (12.5 kcal/mol)
by a PPh donor group to give (SPS)W(CO)3 (4W) (13.3 kcal/
mol) results again in a very small increase of the basicity of
the metal center (0.8 kcal/mol), which leads to the conclusion
that the donor ability of PPh is slightly larger than that of S for
complexes that give protonated complexes of type A structure
(Chart 3). The relative donor abilities of these groups toward
H+ can be evaluated by comparing the pKa values of the
protonated forms of Et2S (-6.8)34 and Et2PPh (6.25).35 On this
basis, S is expected to be a much weaker donor than PPh. In
order to understand why the basicity of 3W is nearly as large
as that of 4W, one might consider that, besides the σ-donor
electrons on S, there is also a lone electron pair that can raise
the energy of the metal dπ orbitals by repulsive interactions,
which would make the S atom a stronger overall donor than
expected from σ-donation alone. Effects of lone electron pairs
on properties of sulfur ligands have been noted in other
systems.36
A comparison of -∆HHM values for noncyclic ligand
complexes (SNS)W(CO)3 (6W) (15.4 kcal/mol) and fac-
(PNP)W(CO)3 (9W-fac) (22.1 kcal/mol) would indicate at first
glance that replacement of two SEt donor groups with two PPh2
donor groups enhances the basicity of the metal center by 6.7
kcal/mol. However, the two complexes give protonated forms
that have different structures; in 6WH+ the donor atoms occupy
mutually cis positions (structure A, Chart 3), while in 9WH+
the donor atoms are approximately coplanar with the metal
(structure B, Chart 3). Considering that the thioether and PPh2
donor groups make essentially the same contribution to the
basicity of the metal (see two paragraphs above), the 6.7 kcal/
mol difference can be attributed primarily to steric repulsion
between the PPh2 groups which is relieved by rearrangement
to structure B in the 9WH+ product. The 6.7 kcal/mol
difference between ∆HHM values for 9W-fac and 6W is similar
to that (6.2 kcal/mol) between fac-[MeC(CH2PPh2)3]W(CO)3
(∆HHM ) -10.5 kcal/mol)6,8 and fac-[PhP(C2H4PPh2)2]W(CO)3
(∆HHM ) -16.7 kcal/mol),6,8 whose protonated forms have
structures (A and B, Chart 3) that differ in the same way as
9WH+ and 6WH+ (eqs 3 and 4). The more exothermic value
for fac-[PhP(C2H4PPh2)2]W(CO)3 is also presumably due to
relief of steric repulsion between the PPh2 groups upon
protonation to give a product with structure B (eq 3).
electron richness of both tungsten and molybdenum complexes
to the same extent is also supported by a plot of the average of
the three ν(CO) frequencies (Table 1) for the (L3)W(CO)3
complexes vs the average ν(CO) frequencies (Table 2) for the
analogous Mo complexes (Figure 1). The correlation (r )
0.9996) is expressed in eq 16, and the slope of approximately
1 indicates that the ligands have nearly the same effect on both
W and Mo complexes.
avg ν(CO)W ) -139.8 + 1.072[avg ν(CO)Mo
]
(16)
For complexes with noncyclic tridentate ligands containing
S and N donor atoms, replacement of the central S by NEt
increases the basicity by 2.9 kcal/mol from (SSS)W(CO)3 (3W)
(12.5 kcal/mol) to (SNS)W(CO)3 (6W) (15.4 kcal/mol). Al-
though this replacement seems to have a somewhat smaller
effect on the basicity of the metal, the overall trend is the same
as in the case of the cyclic ligand complexes.
Phosphine and Amine Ligand Effects on Metal Basicity
(-∆HHM). As expected from the pKa values of the conjugate
acids of Et3N (pKa ) 10.8)33 and Et2PhP (pKa ) 6.25)35 and
from their enthalpies of protonation (-∆HHN ) 39.3 kcal/mol7b
and -∆HHP ) 27.8 kcal/mol,6 respectively), the NEt group is
a better donor than the PPh group. This is reflected in the higher
basicity of (SNSc)W(CO)3 (5W) (14.2 kcal/mol) as compared
with that of (SPSc)W(CO)3 (2W) (11.5 kcal/mol). Thus in the
cyclic ligand complexes, replacement of PPh by NEt results in
a 2.7 kcal/mol increase in the basicity of the metal.
For complexes with noncyclic ligands, the basicity of the
metal in (SPS)W(CO)3 (4W) (13.3 kcal/mol) increases by 2.1
kcal/mol when the central PPh group is substituted by NEt to
give (SNS)W(CO)3 (6W) (15.4 kcal/mol). Thus the complexes
with noncyclic ligands follow the same trend as those with cyclic
ligands when their protonated complexes have the same structure
(structure A, Chart 3).
Complex fac-(PNP)W(CO)3 (9W-fac) (22.1 kcal/mol), which
also has a noncyclic ligand, is 5.4 kcal/mol more basic than the
previously studied fac-[PhP(C2H4PPh2)2]W(CO)3 (16.7 kcal/
mol).6,8 However, both protonated forms 9WH+ and [PhP(C2H4-
PPh2)]W(CO)3(H)+ have structure B (Schemes 1 and Chart 3).
(35) Streuli, C. A. Anal. Chem. 1960, 32, 985.
(36) Caulton, K. G. New J. Chem. 1994, 18, 25.

Protonation of (L3)M(CO)3 Complexes
Inorganic Chemistry, Vol. 37, No. 3, 1998 443
Enthalpy for the Isomerization of mer-(PNP)M(CO)3 to
fac-(PNP)M(CO)3. The (PNP)M(CO)3 (M ) W, Mo) com-
plexes were the only ones that could be prepared as both the
fac and mer isomers. They both protonate to give the same
product, 9WH+ or 9MoH+ (Scheme 2). Using ∆HHM values
for both isomers allows one to calculate the heat of reaction
(∆Hmer-fac) for the isomerization of the mer complex to the
thermodynamically more stable fac isomer using the thermo-
dynamic cycle given in eqs 17-19 and represented in Scheme
N donor groups in the coordination sphere. This π-back-
bonding to the CO groups may be particularly important in these
cationic complexes. On the other hand, the steric bulkiness of
the terminal PPh2 groups in 9WH+, 9MoH+, and [PhP(C2H4-
PPh2)2]W(CO)3(H)+ may force these complexes out of structure
A to give B. Thus, all of the (L3)M(CO)3(H)+ complexes with
relatively small SEt groups adopt structure A, while complexes
with bulky PPh2 groups in the terminal positions favor the less
crowded structure B.
Effect of the Metal on the Basicity of (L3)M(CO)3
Complexes. Enthalpies of protonation for W-Mo pairs of
complexes with the same ligands may be compared as follows:
(NNNc)W(CO)3 (8W) (20.9 kcal/mol) and (NNNc)Mo(CO)3
(8Mo) (18.3 kcal/mol), (NSNc)W(CO)3 (7W) (17.4 kcal/mol)
and (NSNc)Mo(CO)3 (7Mo) (14.6 kcal/mol), fac-(PNP)W(CO)3
(9W-fac) (22.1 kcal/mol) and fac-(PNP)Mo(CO)3 (9Mo-fac)
(19.2 kcal/mol). These results demonstrate that the W com-
plexes are more basic than their Mo analogs by an average of
2.8 kcal/mol (2.6 kcal/mol for 8W and 8Mo, 2.8 kcal/mol for
7W and 7Mo and 2.9 kcal/mol for 9W-fac and 9Mo-fac).
Previously it was found7b that the -∆HHM values for the
protonation of the cis-M(CO)2(Ph2PCH2PPh2)2 complexes are
1.8 kcal/mol higher for the W complex (31.5 kcal/mol) than
the Mo analog (29.7 kcal/mol). The same trend was observed
in pKa values of the conjugate acids of [CpM(CO)3]- complexes
determined in CH3CN solvent: Mo (13.9) < W (16.1).37 The
trend that complexes of third-row metals are more basic than
their second-row analogs1,38 is also noted in other transition
metal groups. In previous ∆HHM studies,7a it was shown that
Os complexes of the type CpM(PR3)2X were more basic than
their Ru analogs by an average of 7.4 kcal/mol.
∆Hmer-fac ) ∆Hmer - ∆Hfac
(17)
W system: ∆Hmer-fac(W) ) -24.1(2) - [-22.1(1)] )
-2.0(3) kcal/mol (18)
Mo system: ∆Hmer-fac(Mo) ) -24.0(3) -
[-19.2(3)] ) -4.8(6) kcal/mol (19)
1. In order to understand the less exothermic mer-to-fac
rearrangement for 9W-mer as compared with that for 9Mo-
mer, one can consider that tungsten behaves as a more electron
rich metal center than Mo as suggested by pKa,37 enthalpy of
protonation,7b and redox potential38 studies of their complexes.
This greater electron richness allows W to π-back-bond to a
larger degree to the CO groups that are trans to each other in
the mer isomer, thereby stabilizing this isomer more in the
tungsten complex than the molybdenum. The greater stabiliza-
tion of 9W-mer as compared to 9Mo-mer results in a less
exothermic ∆Hmer-fac for the tungsten complex rearrangement.
Cyclic and Noncyclic Tridentate Ligand Effects on Metal
Basicity (-∆HHM). A comparison of ∆HHM values for
complexes with cyclic and noncyclic ligands that have the same
donor groups, (SNSc)W(CO)3 (5W) (14.2 kcal/mol) and (SN-
S)W(CO)3 (6W) (15.4 kcal/mol), (SSSc)W(CO)3 (1W) (10.6
kcal/mol) and (SSS)W(CO)3 (3W) (12.5 kcal/mol), and (SP-
Sc)W(CO)3 (2W) (11.5 kcal/mol) and (SPS)W(CO)3 (4W) (13.3
kcal/mol), shows that, in each pair, the complex with the
noncyclic ligand is the more basic (by 1.2 kcal/mol for the pair
5W and 6W, by 1.9 kcal/mol for the pair 1W and 3W, and by
1.8 kcal/mol for the pair 2W and 4W). The greater exother-
micity of the noncyclic ligand complex protonations may be
due to the greater flexibility of the noncyclic ligands which
allows the protonated products (structure A, Chart 3) to adopt
more stable structures than can be achieved with the less flexible
cyclic ligands. That this is likely has been shown for (1,4,7-
trithiacyclononane)Mo(CO)3 and the analogous complex with
the noncyclic ligand (2,5,8-trithianonane)Mo(CO)3.17 Crystal-
lographically-determined structures of these complexes suggest
that the flexibility of the 2,5,8-trithianonane ligand allows for
better overlap between the orbitals of the sulfur donor atoms
and the orbitals of the metal than is possible for the 1,4,7-
trithiacyclononane ligand. This greater overlap puts more
electron density on the metal in the 2,5,8-trithianonane complex
whose ν(CO) bands therefore occur at lower frequencies than
those of (1,4,7-trithiacyclononane)Mo(CO)3.
Conclusions
Comparisons of basicities of tungsten complexes (1W, 2W,
and 5W) with cyclic ligands establish that the trend in donor
ability relative to S is as follows: S (0 kcal/mol) e PPh (0.9
kcal/mol) , NR (R ) Me, Et) (3.6 kcal/mol). While ∆HHM
values serve as a measure of the relative donor abilities of the
S, PPh, and NR groups, Lever4d has described EL as another
parameter for ligand donor ability. It is an electrochemical
parameter defined as 1/6 of the Ru(III)/Ru(II) potential for RuL6
complexes in CH3CN, assuming that all ligand contributions
are additive. This definition means that the most strongly
4d
donating ligands have the lowest EL values. The EL values
for ligands similar to the S, P, and N donor groups in complexes
1W, 2W, and 5W are as follows: Et2S (0.35 V) > Me2PPh
(0.34 V) . NH3 (0.07 V). Although there are no data for Et2-
PPh and Et3N, it is to be expected that the EL values for Et2-
PPh and Me2PPh are very similar, and EL for Et3N is likely to
be nearly the same as that (0.07 V) for NH3 since EL values for
saturated amines all fall in a very narrow range between 0 and
0.1 V. Thus, the trend in donor ability as measured by EL
parallels the one found from our calorimetry studies. Both
∆HHM and EL parameters suggest that the donor abilities of the
R2PPh group is only slightly higher than that of R2S. If one
uses pKa values of the conjugate acids as a measure of ligand
It is interesting to note that of all the (L3)M(CO)3(H)+
complexes discussed in this paper only 9WH+, 9MoH+, and
[PhP(C2H4PPh2)2]W(CO)3(H)+ have structure B (Chart 3), while
all of the others have structure A. In order to understand this
result, opposing steric and electronic factors must be considered.
Stabilization of the π-bonding CO groups would be greater in
structure A in which all of the CO groups are trans to S, P, or
donor ability, one finds the same trend, Et2S (pKa ) -6.8)34
<
Et2PPh (pKa ) 6.2535) < Et3N (pKa ) 10.8033), but the very
large difference in pKa values between Et2S and Et2PPh indicates
that thioethers are much weaker donors than phosphines. That
thioethers behave as donors that are nearly as strong as
phosphines in their metal complexes suggests that the lone
electron pairs on sulfur, which are not present in phosphines or
amines, are able to increase the electron density on the metal
by π-donation17 or repulsion with filled metal d orbitals.36
(37) Jordan, R. F.; Norton, J. R. J. Am. Chem. Soc. 1982, 104, 1255.
(38) Tilset, M.; Parker, V. D. J. Am. Chem. Soc. 1989, 111, 6711.

444 Inorganic Chemistry, Vol. 37, No. 3, 1998
Siclovan and Angelici
In the series of metal complexes (3W, 4W, and 6W) with
noncyclic ligands, where the structures of the protonated
products are the same (structure A, Chart 3) as those of the
complexes with cyclic ligands, a similar trend in donor ability
relative to S is observed: S (0 kcal/mol) e PPh (0.8 kcal/mol)
, NR (R ) Me, Et) (2.9 kcal/mol). With the goal of correlating
this trend with another measure of electron density on the metal,
we plot (Figure 2) -∆HHM Vs average ν(CO) values of
complexes 1W-8W and fac-[MeC(CH2PPh2)3]W(CO)3 that
have the same reactant and product structures. This correlation
(r ) 0.973) is expressed in eq 20 and shows that as the basicity
-∆HHM ) 300.1 - 0.1553[ν(CO)] (kcal/mol) (20)
(-∆HHM) of a complex increases, its average ν(CO) decreases.
Equation 20 may be useful for estimating basicities of other
(L3)W(CO)3 complexes with S, P, and N donor ligands.
In contrast to this correlation with ν(CO), an attempt to plot
-∆HHM for complexes 1W-8W Vs the 1H NMR chemical shift
of the hydride signal in the 1WH+-8WH+ complexes resulted
in a very poor correlation (r ) 0.415).
Figure 2. Correlation (eq 20) of (L3)W(CO)3 basicity (-∆HHM) (Table
3) with average ν(CO) values of the (L3)W(CO)3 complexes (Table
1). PPP ) MeC(CH2PPh2)3.
Several other factors besides the donor groups affect the
basicities of the (L3)M(CO)3 complexes. Noncyclic ligands
make the metal more basic, by 1.6 ( 0.3 kcal/mol, than cyclic
ligands with the same donor atoms. The tungsten complexes
are 2.8 ( 0.1 kcal/mol more basic than their molybdenum
analogs. And the isomeric form of the (PNP)M(CO)3 (M )
W, Mo) complexes influences the -∆HHM of their protonation.
For both the W and Mo complexes the mer isomer is more basic
than the fac. These protonations also show that the fac isomer
is more stable than the mer by 2.0 kcal/mol for (PNP)W(CO)3
and by 4.8 kcal/mol for (PNP)Mo(CO)3.
Acknowledgment. We are grateful to the National Science
Foundation (Grant CHE-9414242) for support of this work
IC970670E