Kinetic studies of the alkylation of metal cyanide complexes: M(CO)x(L)5-xCN (M = Mn, Re) and CpM′(CO)x(L)2-xCN (M′ = Fe, Ru)

Article abstract of DOI:10.1021/ic981014x

Reaction rates for the alkylation of Mn(CO)(dppm)₂CN (dppm = PPh₂CH₂PPh₂), Mn(CO)₂(tripod)CN [tripod = (PPh₂CH₂)₃CCH₃], Re(CO)₃(dppm)CN, (η⁶-C₆Me₆)Mn(CO)₂CN, CpFe(dppe)CN (dppe = PPh₂CH₂CH₂PPh₂), CpRu(dppe)CN, and CpRu(CO)(PPh₃)CN with methyl 4-nitrobenzenesulfonate (MeONs) to produce complexes of the type [L_nM-CNMe]^{+ -}ONs were investigated in 1,2-dichloroethane (DCE) at 30.0 °C. The reactions are first order in both the complex and the alkylating agent, which is consistent with a mechanism that involves nucleophilic attack of the cyanide nitrogen on the methyl of the MeONs. The second-order rate constants (k₂) correlate approximately with the stretching frequency of the cyanide ligand v(CN) in the complexes. There is a somewhat better correlation between v(CN) of the CNCH₃ ligands in the products and k₂. There is also a good correlation between the rate constant (k₂) and the calculated force constant (K_CO) for the v(CO) vibrational mode of the analogous [L_nM-CO]⁺ complex. These correlations appear to be useful for predicting nucleophilicities of metal cyanide complexes of different metals, in different oxidation states, and with ligands exhibiting a range of electronic and steric properties.

Full text of DOI:10.1021/ic981014x

1708
Inorg. Chem. 1999, 38, 1708-1712
Kinetic Studies of the Alkylation of Metal Cyanide Complexes: M(CO)x(L)5-xCN (M ) Mn,
Re) and CpM′(CO)x(L)2-xCN (M′ ) Fe, Ru)
Laurie A. Cardoza and Robert J. Angelici*
Department of Chemistry, Iowa State University, Ames, Iowa 50011
ReceiVed August 21, 1998
Reaction rates for the alkylation of Mn(CO)(dppm)2CN (dppm ) PPh2CH2PPh2), Mn(CO)2(tripod)CN [tripod )
(PPh2CH2)3CCH3], Re(CO)3(dppm)CN, (η6-C6Me6)Mn(CO)2CN, CpFe(dppe)CN (dppe ) PPh2CH2CH2PPh2),
CpRu(dppe)CN, and CpRu(CO)(PPh3)CN with methyl 4-nitrobenzenesulfonate (MeONs) to produce complexes
of the type [LnM-CNMe]+ -ONs were investigated in 1,2-dichloroethane (DCE) at 30.0 °C. The reactions are
first order in both the complex and the alkylating agent, which is consistent with a mechanism that involves
nucleophilic attack of the cyanide nitrogen on the methyl of the MeONs. The second-order rate constants (k2)
correlate approximately with the stretching frequency of the cyanide ligand ν(CN) in the complexes. There is a
somewhat better correlation between ν(CN) of the CNCH3 ligands in the products and k2. There is also a good
correlation between the rate constant (k2) and the calculated force constant (kCO) for the ν(CO) vibrational mode
of the analogous [LnM-CO]+ complex. These correlations appear to be useful for predicting nucleophilicities of
metal cyanide complexes of different metals, in different oxidation states, and with ligands exhibiting a range of
electronic and steric properties.
Introduction
shown in the structure of CpRu(PPh3)2CN-H+‚‚‚-OSO2CF3,
which was determined by an X-ray diffraction study.7
In this paper we describe kinetic studies of the nucleophi-
licities of cyanide ligands by determining the rates of reactions
(eq 2) of a series of d6, 18-electron transition metal complexes
with methyl 4-nitrobenzenesulfonate (MeONs) in 1,2-dichloro-
ethane at 30 °C,
Transition metal cyanide complexes have been the subject
of numerous studies since the discovery of Prussian blue in the
early 1700s.1-5 Of special interest is the ability of the nitrogen
atom of the cyanide ligand to behave as a nucleophile toward
various organic and inorganic electrophiles to form alkyl
isocyanide complexes3,4 and multinuclear cyanide-bridged metal
complexes.3-5 However, quantitative investigations of cyanide
ligand nucleophilicity have not been reported.
Related to the nucleophilicity of a cyanide ligand is its
basicity. Recently this group reported basicities of cyanide
ligands in CpM(L)2CN complexes, where M ) Fe or Ru, from
a study of heats of protonation (∆HCNH) of the complexes with
HOSO2CF3 in 1,2-dichloroethane (DCE) (eq 1).6 It was observed
where LnM-CN ) Mn(CO)(dppm)2CN (1CN), Mn(CO)2-
(tripod)CN (2CN), Re(CO)3(dppm)CN (3CN), (η6-C6Me6)Mn-
(CO)2CN (4CN), CpFe(dppe)CN (5CN), CpRu(dppe)CN (6CN),
and CpRu(CO)(PPh3)CN (7CN).
DCE
CpM(L)2CN + HOSO2CF3 25.0 °C8
CpM(L)2(CNH)+ -OSO2CF3; ∆HCNH (1)
The complexes were chosen so as to represent metals with
varying degrees of electron richness, which was adjusted using
different auxiliary ligands (L) in the LnM-CN complexes. A
major goal of the study was to determine whether the nucleo-
philicity of the cyanide ligand, as represented by the rate
constant (k2) for reaction 2, could be correlated with parameters
that reflect electron richness at the metal center.
that while the basicity (-∆HCNH) of the cyanide ligand increased
from CpRu(PPh3)2CN (20.5 kcal/mol) to CpRu(PMe3)2CN (22.4
kcal/mol) with increasing donor ability of the PR3 ligand, the
changes in -∆HCNH were relatively small. It was observed that
for the CpM(dppe)CN complexes, where dppe ) Ph2PCH2CH2-
PPh2, the Fe (20.9 kcal/mol) derivative was slightly more basic
than the Ru (19.5 kcal/mol) analogue.6 However, the basicity
determinations were somewhat complicated by hydrogen bond-
ing between the protonated product and the triflate anion as
Experimental Section
General Procedures. All preparative reactions, chromatography, and
manipulations were carried out under an inert atmosphere of nitrogen
or argon using standard Schlenk techniques. Solvents were purified
under nitrogen using standard methods.8 Hexanes and dichloromethane
(1) Sharpe, A. G. The Chemistry of the Cyano Complexes of Transition
Metals, 1st ed.; Interscience Pub.: New York, 1970.
(2) Dunbar K. R.; Heintz, R. A. Prog. Inorg. Chem. 1997, 45, 283.
(3) Fehlhammer, W. P.; Fritz M. Chem. ReV. 1993, 93, 1243.
(4) Shriver, D. F. Struct. Bonding 1966, 1, 32.
(5) Rigo, P.; Turco, A. Coord. Chem. ReV. 1974, 13, 133.
(6) Nataro C.; Chen, J.; Angelici, R. J. Inorg. Chem. 1998, 37, 1868.
(7) Sapunov, V. N.; Mereite, K.; Schmid, R.; Kirchner, K. J. Organomet.
Chem. 1997, 530, 105.
(8) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals, 2nd ed.; Pergamon: New York, 1980.
10.1021/ic981014x CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/19/1999

Kinetics of the Alkylation of Metal Cyanide Complexes
Inorganic Chemistry, Vol. 38, No. 8, 1999 1709
1
(DCM) were refluxed over CaH2 and then distilled. Tetrahydrofuran
(THF) and diethyl ether were distilled from sodium benzophenone.
Methanol was dried over magnesium turnings and iodine and distilled.
Hexanes and dichloromethane were stored under nitrogen over mo-
lecular sieves (4 Å). 1,2-Dichloroethane (DCE) was purified by washing
consecutively with concentrated sulfuric acid, distilled deionized water,
5% NaOH, and water again; after the solvent was predried over
anhydrous MgSO4, it was distilled from P4O10 and then stored in the
dark under nitrogen. Neutral Al2O3 (Brockmann activity I) was
deoxygenated at room temperature under high vacuum for 12 h,
deactivated with 5% (w/w) N2-saturated water, and stored under N2.
All solution infrared spectra were recorded on samples in CH2Cl2
or 1,2-dichloroethane in NaCl cells using a Nicolet 560 infrared
cm-1. H NMR (CDCl3): δ 1.49 (s, 3H, CCH3), 2.47 (m, 6H, CH2),
7.02-7.29, 7.49, 7.71 (m, 30H, Ph). 31P NMR (CDCl3): δ 21.89 (d,
JPP ) 88.0 Hz, 2P), 69.88 (t, JPP ) 90.0 Hz, 1P). MS (ESI): m/e 680
(M+).
Preparation of fac-Mn(CO)2(tripod)CN (2CN). A solution of
Mn(CO)2(tripod)Br (495 mg, 0.61 mmol) in CH2Cl2 (15 mL) was
treated with NaCN (119 mg, 2.42 mmol) dissolved in MeOH (32.0
mL). After the solution was refluxed for 2 h, the opaque orange solution
turned clear yellow-orange. This solution was cooled (-10 °C), the
solvent mixture was evaporated, and water (10 mL) was added to the
residue; then CH2Cl2 (3 × 5 mL) was added. The CH2Cl2 layer was
decanted, dried over anhydrous Na2SO4 and filtered through Celite.
The solvent was reduced to 5 mL under vacuum, and hexanes were
added to precipitate the product, which was recrystallized from CH2Cl2/
hexanes to give a light yellow solid (390 mg, 84%). IR (CH2Cl2): ν(CO)
1
spectrophotometer; H NMR spectra were obtained on compounds in
CDCl3 using a Bruker 200 AC MHz or a Varian VXR 300 MHz
spectrometer using TMS (δ ) 0.00 ppm) as the internal reference.
31P{1H} NMR spectra were obtained in CDCl3 on the Bruker AC 200
MHz instrument with 85% phosphoric acid (δ ) 0.00 ppm) as an
external reference. Elemental microanalyses were performed on a
Perkin-Elmer 2400 series II CHNS/O analyzer. Electron ionization mass
spectra (EIMS at 70 eV), chemical ionization (NH3) mass spectra
(CIMS), and electrospray ionization mass spectra (ESI in CH2Cl2) were
obtained on a Finnigan 4000 spectrometer.
1954 (vs), 1897 (s) cm-1. H NMR (CDCl3): δ 1.49 (s, 3H, CCH3),
1
2.24-2.33 (m, 6H, CH2), 7.09-7.25, 7.52, 7.70 (m, 30H, Ph). 31P NMR
(CDCl3): δ 31.0 (br, 2P), 49.0 (br, 1P). MS (ESI): m/e 761 (M+).
Anal. Calcd for C44H39MnNO2P3: C, 69.38; H, 5.16; N, 1.84. Found:
C, 68.78; H, 5.46; N, 1.79.
Preparation of fac-Re(CO)3(dppm)CN (3CN). To a solution of
Re(CO)3(dppm)OSO2CF3 (1.95 g, 2.42 mmol) in acetone (20.0 mL)
was added a 2-fold molar excess of KCN or NaCN. After the solution
was stirred for 3 h at room temperature, the solvent was removed under
vacuum. The resulting white solid was washed with water (10 mL)
and extracted with CH2Cl2 (3 × 5 mL). The CH2Cl2 layer was filtered
and dried over Na2SO4. A white solid (3CN) was obtained from the
CH2Cl2 solution after recrystallization from CH2Cl2/hexanes (1.31 g,
The alkylating agent methyl 4-nitrobenzenesulfonate (MeONs) was
purchased from Aldrich or Sigma. The metal complexes, Mn2(CO)10
and Re2(CO)10, as well as the phosphines (Ph2PCH2)3CCH3 (tripod),
Ph2PCH2CH2PPh2 (dppe), and Ph2PCH2PPh2 (dppm) were purchased
from Strem Chemical Co. NaCN and KCN were purchased from Fisher
Chemical Co. The complexes Mn(CO)5Br,9 (C6Me6)Mn(CO)2CN
(4CN),10 Mn(CO)(dppm)2Br,11 CpFe(dppe)CN (5CN),12 CpRu(dppe)-
CN (6CN),13 CpRu(CO)(PPh3)CN (7CN),6 and Re(CO)3(dppm)(OSO2-
CF3)14 were prepared by literature methods. IR spectra in the ν(CN)
80%). IR (CH2Cl2): ν(CO) 2032 (vs), 1952 (s, sh) cm-1. H NMR
1
(CDCl3): δ 4.60-4.80 (m, 1H, CHH), 5.32-5.61 (m, 1H, CHH), 7.31-
7.48, 7.69 (m, 30H, Ph). 31P NMR (CDCl3): δ -39.5 (s).
Reactions of 1CN-7CN with MeONs. The compounds 1CN-7CN
were reacted with MeONs for infrared characterization of the [LnM-
CNMe]ONs products by dissolving approximately 10 mg of LnM-
CN in 2.0 mL of DCE in a 16 mm Pyrex test tube under nitrogen. To
the solution was added 10 equiv of MeONs. The color of the solution
lightened upon alkylation within the reaction time (45 min to 7 days).
The yields of the alkylated products were quantitative as determined
by the complete disappearance of the reactant IR spectrum and the
appearance of the spectrum for only the LnM-CNMe+ product.
+
region of the LnM-CN and LnM-CNCH3 complexes are given in
the Discussion section. The ν(CN) bands of the CN- ligands are of
weak to medium intensity, while those of the CNCH3 ligands are of
medium to strong intensity.
Preparation of trans-Mn(CO)(dppm)2CN (1CN). To a 50 mL
Schlenk flask containing Mn(CO)(dppm)2Br (415 mg, 0.69 mmol) and
CH2Cl2 (10.0 mL) was added a 3-fold molar excess of KCN (135 mg,
2.07 mmol) in methanol (20.0 mL) under nitrogen. The solution was
stirred for 3 h, at which time the solution had changed from dark orange
to orange-yellow. The solvent was removed under vacuum. The
remaining residue was treated with 10 mL of water, and the Mn(CO)-
(dppm)2CN was extracted from the mixture with CH2Cl2 (3 × 5 mL).
The CH2Cl2 solution was dried over anhydrous Na2SO4, the solvent
was reduced to ∼5 mL under vacuum, and hexanes were added to
precipitate the orange product (264 mg, 68%). Spectra of 1CN are
similar to those reported previously for this compound.15 IR (CH2Cl2):
ν(CO) 1861 (s) cm-1. 1H NMR (CDCl3): δ 4.84 (m, 4H, CH2), 7.16-
7.31, 7.41-7.46 (m, 40H, Ph). 31P NMR (CDCl3): 41.5 (s). MS (EI):
878 (M+).
Preparation of fac-Mn(CO)2(tripod)Br (2Br). To a 50 mL Schlenk
flask containing Mn(CO)5Br (704 mg, 2.56 mmol) and tripod (1.76 g,
2.82 mmol) was added toluene (20 mL) under nitrogen. While the
solution was refluxed with stirring, it turned from light orange to dark
orange-red. After 1 h, the solution was allowed to cool to room
temperature. The volume of toluene was reduced under vacuum to 10
mL, and hexanes were added; a dark orange solid precipitated. The
solution was filtered, and the solid was washed with hexanes (4 × 5
mL), dried under vacuum, and recrystallized from CH2Cl2/hexanes to
give 1.86 g (89%) of 2Br.16 IR (CH2Cl2): ν(CO) 1951 (s), 1893 (m)
Isolation of the alkylated products was accomplished by reaction of
the metal cyanide complexes (100 mg) with equimolar MeONs in
refluxing 1,2-dichloroethane (30 mL). These reactions were monitored
by IR spectroscopy until only the product peaks were visible. The
solvent was reduced to 5 mL, and the product was precipitated by the
addition of 50 mL of hexanes. The product was recrystallized from
MeOH/diethyl ether. Spectroscopic data for the alkylated complexes
[LnM-CNMe]ONs are given below.
[Mn(CO)(dppm)2(CNCH3)]ONs (1CNCH3+). IR (DCE): ν(CO)
1901 (s) cm-1. H NMR (CDCl3): δ 2.70 (s, 3H, CNCH3), 4.88 (m,
1
2H, CHH), 4.76 (m, 4H, CHH), 7.01, 7.11, 7.23-7.74 (m, 40H, Ph),
8.15 (s, 4H, C6H4NO2). 31P NMR (CDCl3): 38.8 (s).
[Mn(CO)2(tripod)(CNCH3)]ONs (2CNCH3+). IR (DCE): ν(CO)
1975 (vs), 1924 (s) cm-1. H NMR (CDCl3): δ 1.80 (s, 3H, CCH3),
1
2.83 (s, 6H, CH2), 2.49 (s, 3H, CNCH3), 7.26 (s, 30H, Ph), 8.15 (s,
4H, C6H4NO2). 31P NMR (CDCl3): δ 34.5 (s, br).
[Re(CO)3(dppm)(CNCH3)]ONs (3CNCH3+). IR (DCE): ν(CO)
2049 (vs), 1977 (s) cm-1. H NMR (CDCl3): δ 5.24-5.45 (m, 2H,
1
CHH), 6.18-6.32 (m, 2H, CHH), 2.78 (s, 3H, CNCH3), 7.28, 7.33-
7.87 (m, 20H, Ph; 4H, C6H4NO2). 31P NMR (CDCl3): δ -43.7 (s).
[(C6Me6)Mn(CO)2(CNCH3)]ONs (4CNCH3+). IR (DCE): ν(CO)
2007 (vs), 1962 (s) cm-1. H NMR (CDCl3): δ 2.40 (s, 18H, CCH3),
1
(9) Quick, M. H.; Angelici, R. J. Inorg. Synth. 1979, 19, 160.
(10) Pike, R. D.; Carpenter, G. B. Organometallics 1993, 12, 1416.
(11) Reimann, R. H.; Singleton, E. J. Organomet. Chem. 1972, 38, 113.
(12) Treichel, P. M.; Molzahn, D. C. Synth. React. Inorg. Met.-Org. Chem.
1979, 9, 21.
(13) Bair, G. J.; Davies, S. G. J. Organomet. Chem. 1984, 262, 215.
(14) Dahlenburg, L.; Hillmann, G.; Ernst, M.; Moll, M.; Knoch, F. J.
Organomet. Chem. 1996, 525, 115.
3.73 (s, 3H, CNCH3), 8.18 (s, 4H, C6H4NO2).
[CpFe(dppe)(CNCH3)]ONs (5CNCH3+). 1H NMR (CDCl3):
2.55, 2.62 (m, 4H, CH2), 2.88 (s, 3H, CNCH3), 4.54 (s, 5H, C5H5),
7.27, 7.43-7.69 (m, 20H, Ph; 4H, -C6H4NO2). 31P NMR (CDCl3): δ
97.1 (s).
δ
(15) Carriedo, G. A.; Crespo, M. C.; Riera, V.; Sanchez, M. G.; Valin, M.
L.; Moreiras, D.; Solans, X. J. Organomet. Chem. 1986, 302, 47.
(16) Ellermann, J.; Lindner, H. A.; Moll, M. Chem. Ber. 1979, 112, 3441.

1710 Inorganic Chemistry, Vol. 38, No. 8, 1999
Cardoza and Angelici
[CpRu(dppe)(CNCH3)]ONs (6CNCH3+). 1H NMR (CDCl3): δ
2.61-2.71 (m, 4H, CH2), 2.71 (s, 3H, CNCH3), 4.98 (s, 5H, C5H5),
7.09-7.72 (m, 20H, Ph; 4H, C6H4NO2). 31P NMR (CDCl3): δ 77.7
(s).
Preparation of 1CN, 2CN, and 3CN. Two common methods
for the preparation of metal cyanide complexes from metal
halides LnM-X involve either (i) refluxing LnM-X in methanol
with an alkali metal cyanide6,12,13 or (ii) stirring LnM-X with
AgCN in CH2Cl2.15 The preparation of fac-Mn(CO)2(tripod)-
CN (2CN) gave the highest yield following the first method by
refluxing a MeOH/CH2Cl2 solution of Mn(CO)2(tripod)Br with
an excess of NaCN.
[CpRu(CO)(PPh3)(CNCH3)]ONs (7CNCH3+). IR (DCE): ν(CO)
1
2006 (s) cm-1. H NMR (CDCl3): δ 3.34 (s, 3H, CNCH3), 5.38 (s,
5H, C5H5), 7.28-7.73 (m, 15H, Ph; 4H, C6H4NO2). 31P NMR
(CDCl3): δ 49.3 (s).
Attempted Alkylations of Nitriles. To compare the reactivities of
the cyano complexes with those of organic nitriles, acetonitrile and
benzonitrile (4-5 µL) were treated with a 10-fold excess of methyl
4-nitrobenzene sulfonate in a solution of DCE (3.0 mL) at 30.0 °C in
a 16 mm Pyrex test tube under nitrogen. Periodic monitoring of the
reaction solutions by infrared spectroscopy showed that no reaction
occurred for either reaction over a 7 day period. Benzonitrile was
previously reported to be ethylated with the strong electrophilic reagents
[Et3O]BF4 and diethoxycarbonium hexachloroantimonate.17-19
Kinetic Studies of the Reactions of LnM-CN with MeONs. In a
typical experiment, 10-15 mg of LnM-CN was dissolved in DCE in
a 16 mm Pyrex test tube under nitrogen. This solution was placed in
a constant-temperature bath at 30.0 ( 0.2 °C for 20-30 min. The Cryo
Therm variable-temperature IR cell (International Crystal Laboratories,
No. 2405-4763) was placed in a constant-temperature jacket also at
30.0 ( 0.2 °C for 30 min before it was used. To the solution of LnM-
CN was added 0.5-1.5 mL of the desired concentration of MeONs
also at 30.0 ( 0.2 °C. This reaction solution was then injected into the
IR cell. The formation of the methylated product was monitored to
completion by measuring the increase in the ν(CNMe) peak intensity
of the methylated product. Rate constants were calculated from the
15-50 spectra taken during the first two half-lives of the reactions. In
the experiments, where at least a 10-fold excess of alkylating agent
was utilized, the absorbances (P) of the ν(CNMe) product band were
fitted to eq 320 utilizing Kaleidagraph (Synergy Software) to obtain a
pseudo-first-order rate constant kobs (s-1). The bimolecular rate constants
The synthesis of complex 1CN from trans-Mn(CO)(dppm)2Br
(1Br) and silver cyanide has been described previously by Riera
et al.15 However, 1Br is also oxidized by AgCN, producing a
black solution with only a small quantity of the desired product
1CN. To avoid this oxidation, KCN or NaCN was reacted with
1Br to give 1CN in 68% yield.
Because there was no reaction between fac-Re(CO)3(dppm)-
Br and excess AgCN even after 48 h, the triflate complex
Re(CO)3(dppm)(OSO2CF3) was reacted with KCN or NaCN to
produce 3CN (eq 6) in 80% yield. Two ν(CO) bands were
AgOSO2CF3
Re(CO)3(dppm)Br
8
DCM
3Br
NaCN
Re(CO) (dppm)CN
Re(CO)3(dppm)(OSO2CF3) acetone8
(6)
3
3CN
observed in the IR spectrum of 3CN; they are at lower
wavenumber than those of Re(CO)3(dppm)(OSO2CF3). One
phosphorus signal was observed in the 31P NMR spectrum, and
1H NMR signals for Ph and both CHH protons were similar in
appearance to those of the facial complexes Re(CO)3(dppm)-
(OSO2CF3)14 and Re(CO)3(dppm)Br.14 Complexes (C6Me6)Mn-
(CO)2CN (4CN),10 CpFe(dppe)CN (5CN),12 CpRu(dppe)CN
(6CN),13 and CpRu(CO)(PPh3)CN (7CN)6 were prepared by
previously reported methods.
P∞ - Pt ) (P∞ - P0) exp(-kobst)
(3)
k2 (M-1 s-1) were calculated using the expression k2 ) kobs/[MeONs]av,
where [MeONs]av is the average of the [MeONs] at the beginning and
the end of the reaction.
Rate data obtained when [MeONs]0 was less than 10 times as large
as [LnM-CN]0 were fitted to linear plots of eq 4,20 ln([B]/[A]) vs t,
where the rate constant was determined by first converting the
absorbance for LnM-CNCH3 ([Mn(CO)(dppm)2CNMe]ONs, ꢀ ) 708
cm-1 M-1; [Mn(CO)2(tripod)CNMe]ONs, ꢀ ) 1033 cm-1 M-1) to
concentration. The second-order rate constant k2 (M-1 s-1) was obtained
from the slope of the best fit straight line.
Characterization of Products in Reaction 2. The methylated
products were prepared by reaction of the LnM-CN complexes
1CN-7CN with excess (10-fold) MeONs in DCE. The large
increase in ν(CN) and small increase in ν(CO) indicate that the
alkylation occurs at the cyanide ligand. For example, ν(CN) in
[(C6Me6)Mn(CO)2(CNMe)]ONs increases by 103 cm-1 from
that in (C6Me6)Mn(CO)2CN, but ν(CO) increases by only 27
and 33 cm-1. The coordination geometries of the alkylated
complexes were established to be the same as those of their
precursor cyanide complexes, LnM-CN, by the similar appear-
ance of their IR and 31P and 1H NMR spectra. Also trans-[Mn-
(CO)(dppm)2(CNMe)]ONs (1CNCH3+), [(C6Me6)Mn(CO)2-
(CNMe)]ONs (4CNCH3+), and CpFe(dppe)(CNMe)]ONs
(5CNCH3+) have spectra very similar to those of the following
reported compounds that contain the same cationic complexes:
trans-[Mn(CO)(dppm)2(CNMe)]PF6,15 [CpFe(dppe)(CNMe)]-
BF4,13 and (C6Me6)Mn(CO)2(CNMe)]OTf.10 In the 31P NMR
[MeONs]0 - [LnM-CNCH3]t
[LnM-CN]0 - [LnM-CNCH3]t
[MeONs]0
ln
) ln
+ k2∆0t (4)
[LnM-CN]0
where [B] ) [MeONs]0 - [LnM-CNCH3]t, [A] ) [LnM-CN]0
[LnM-CNCH3]t and ∆0 ) [MeONs]0 - [LnM-CN]0.
-
Results
+
spectra of 6CNCH3 and 7CNCH3+, the 31P signal is shifted
Preparation of Mn(CO)(tripod)Br (2Br). This complex was
previously prepared by Ellermann et al.16 by an indirect method.
A more efficient route involves refluxing a solution of Mn(CO)5Br
with 1 equiv of tripod in toluene (eq 5).
1
upfield from that of the precursor cyanide complexes. The H
NMR spectra of 6CNCH3+ and 7CNCH3+ show a Cp resonance
that is downfield with respect to the nonalkylated starting
complexes 6CN and 7CN, as is also observed in the methylation
of CpFe(dppe)(CN) (5CN). Overall shifts in the IR and 31P and
1H NMR bands are similar to those reported for the methylation
of CpFe(dppe)(CN) with [Me3O]BF4.13 Upon methylation, the
31P NMR spectrum of Mn(CO)2(tripod)CN (2CNCH3+) shows
an upfield shift in the signal of the phosphorus trans to the
cyanide ligand and a slight downfield shift in the signal for the
toluene
Mn(CO)5Br + (Ph2PCH2)3CCH3
8
reflux
Mn(CO)2[(Ph2PCH2)3CCH3]Br
(5)
2Br
(17) Broch, R. F. J. Org. Chem. 1969, 34, 629.
(18) Meerwein, H.; Laasch, P.; Mersch, R.; Spille, J. Chem. Ber. 1965,
89, 209.
(19) Kabuss, S. Angew. Chem., Int. Ed. Engl. 1966, 5, 675.
(20) Espenson, J. Chemical Kinetics and Reaction Mechanisms, 1st ed;
McGraw-Hill: New York, 1981.

Kinetics of the Alkylation of Metal Cyanide Complexes
Inorganic Chemistry, Vol. 38, No. 8, 1999 1711
Figure 1. Plots of kobs vs [MeONs]0 for reactions (eq 2) of (a) CpFe-
(dppe)CN (1), (b) (C6Me6)Mn(CO)2CN (b), and (c) Re(CO)3(dppm)-
CN (9) with MeONs at 30.0 °C.
Table 1. Second-Order Rate Constants (k2) for Reaction 2 and kCO
for LnM-CN Complexes
metal complex
105k2,a M-1 s-1
kCO, N m-1
Figure 2. Correlation between log k2 rate constants for reaction 2 and
(a) ν(CN) for the LnM-CN complexes (2), (b) ν(CN)for the LnM-
CNMe+ products ([), and (c) calculated force constants (kCO) for LnM-
CO+ corresponding to the LnM-CN complexes 1CN-7CN (b).
Re(CO)3(dppm)CN (3CN)
(C6Me6)Mn(CO)2CN (4CN)
CpRu(CO)(PPh3)CN (7CN)
CpRu(dppe)CN (6CN)
CpFe(dppe)CN (5CN)
Mn(CO)2(tripod)CN (2CN)
Mn(CO)(dppm)2CN (1CN)
0.54(3)
0.97(1)
1.09(4)
16.1(4)
16.7(4)
20(2)
1724
1650
1691
1617
1615
1634
1551
chosen as the electrophile for detailed study because it gave
the most convenient rates for the reactions with the cyanide
complexes.
152(6)
a Average of the values in Table S1 of Supporting Information;
numbers in parentheses are average deviations.
Alkylation (eq 2) of the CN- ligand in complexes 1CN-
7CN with MeONs follows a second-order rate law which is
consistent with a mechanism that involves nucleophilic attack
of the nitrogen atom of the cyanide ligand on the carbon of the
MeONs methyl group. This attack results in the displacement
of the nosylate anion (-ONs) and the formation of the product
[LnM-CNMe]+ -ONs. As shown in Table 1, rate constants k2
for the reaction in eq 2 increase with the nucleophilic complex
in the order (105k2 in parentheses) 3CN (0.54) < 4CN (0.97)
< 7CN (1.09) < 6CN (16.1) ∼ 5CN (16.7) < 2CN (20) <
1CN (152). In an attempt to explain the 280-fold increase in
rate from the slowest 3CN to the fastest 1CN reaction, we first
considered the possibility that ν(CN) values of the complexes
would serve as a guide. It was observed previously that there is
a correlation between the basicity (-∆HCNH, eq 1) and ν(CN)
for a series of (η5-Cp)Ru(PR3)2CN complexes;6 the lower ν(CN),
the higher the basicity. For the present series of complexes, the
ν(CN) values (measured in CH2Cl2 except for 5CN and 6CN
in DCE) decrease in the order 3CN (2119 cm-1) > 7CN (2104)
> 4CN (2093) > 2CN (2081) > 1CN (2079) > 6CN (2077)
> 5CN (2065). The correlation (R ) 0.615) of ν(CN) with log
k2 is shown in Figure 2. A somewhat better correlation (R )
0.934) is observed (Figure 2) between ν(CN) values of the CN-
CH3 ligands in the products and the k2 rate constants. These
ν(CN) values (measured in DCE solvent) decrease in the order
two phosphorus atoms cis to the cyanide ligand. The resulting
spectrum contains only one broad peak. The 31P NMR spectrum
of 3CNCH3+ shows only one signal similar in position to those
of the related facial complexes fac-Re(CO)3(dppm)Br and fac-
Re(CO)3(dppm)OTf.14
Kinetic Studies. Second-order rate constants (k2) for the
methylation (eq 2) of LnM-CN, 1CN-7CN, were determined
from rate studies conducted under pseudo-first-order (>10-fold
excess of MeONs) or second-order (<10-fold excess) conditions
in 1,2-dichloroethane (DCE) at 30.0 °C. The rate constants k2
(Table S1) were obtained by one of two methods: (i) indirectly
from the relationship k2 ) kobs/[MeONs]av or (ii) directly from
a linear plot of the data for a second-order reaction using eq 4.
Plots (Figure 1) of kobs vs [MeONs]0 gave straight lines with
y-intercepts near 0, within experimental error. Rate constants
k2 for the 5-13 runs for each complex are within at least 17%
of the average values listed in Table 1. All reactions were shown
to follow the second-order rate law: rate ) k2 [LnM-CN]-
[MeONs]. No ν(CN) or ν(CO) bands other than those of the
reactant and product were detected during the course of the
reactions.
Discussion
3CNCH3 (2221 cm-1) > 7CNCH3 (2211) > 4CNCH3
+
+
+
+
(2195) > 2CNCH3+ (2167) > 6CNCH3+ (2165) > 5CNCH3
(2150) > 1CNCH3 (2149).
Prior to performing the kinetic studies, we compared qualita-
tive rates of reactions of Mn(CO)2(tripod)CN (2CN) with 10-
fold excesses of several electrophiles in CH2Cl2 solvent at room
temperature. After 3 h of reaction with MeI, approximately 25%
of 2CN was converted to 2CNCH3+. Under the same conditions,
approximately 60% of 2CN was converted to 2CNCH3 after
7 h of reaction with methyl tosylate (MeOTs). With MeONs,
50% of 2CN reacted in 1.1 h. With CF3SO3Me (MeOTf), 100%
of 2CN was converted to 2CNCH3 in less than 3 min. Thus,
the reaction rates increase with the alkylating agent in the order
MeI e MeOTs < MeONs , MeOTf. A similar order of
reactivity has been observed in reactions with organic nucleo-
philes.21 On the basis of these and other studies, MeONs was
+
Seeking a more general parameter that would account for
the rates of reactions (eq 2) involving these complexes contain-
ing four different metals (Mn, Re, Fe, Ru), two oxidation states,
and several different ligands (Cp, C6Me6, tripod, dppe, dppm,
PPh3, and CO), we turned to a parameter (kCO) that this research
group used previously22 to correlate nucleophilic attack on
coordinated ethylene and benzene ligands. The kCO in its present
+
+
(21) March, J. AdVanced Organic Chemistry; Reactions, Mechanisms, and
Structure, 4th ed.; John Wiley and Sons: New York, 1992.
(22) Bush, R. C.; Angelici, R. J. J. Am. Chem. Soc. 1986, 108, 2735.

1712 Inorganic Chemistry, Vol. 38, No. 8, 1999
Cardoza and Angelici
usage is simply the C-O stretching force constant for a CO
ligand that occupies the site of the CN- ligand in a LnM-CN
complex. Thus, the kCO value for [CpRu(dppe)(CO)]+ is the
parameter for CpRu(dppe)CN and is a measure of the electron
richness of the CpRu(dppe)+ fragment. Complexes with the
lowest kCO values have the highest electron richness at the metal
center. A high electron richness at the metal center should make
the CN- ligand more electron rich and therefore more nucleo-
philic. In principle, kCO values for the LnM-CO+ complexes
can be calculated from experimental ν(CO) values for the
complexes or they can be more easily estimated by an empirical
procedure (eq 7) developed by Timney.23 He noted that kCO
values depend on the following parameters: kd (associated with
the metal and its number of d electrons), ꢀLθ (contributions from
each of the ligands L in the complex),24 and 197 N m-1 for the
effect of the charge on ionic complexes. Using Timney’s eq 7
the line is a least-squares fit of all of the compounds except
4CN. Only the complex (C6Me6)Mn(CO)2CN (4CN) deviates
significantly from the otherwise very good correlation (R )
0.986). The reason for this deviation is not obvious, but it is
perhaps related to the method of estimating kCO. Because
Timney parameters for the η6-C6Me6 ligand are not available,23
the previously reported25 kCO value, calculated from experi-
mental ν(CO) data for [(η6-C6Me6)Mn(CO)3]+, was used. Thus,
with the exception of 4CN, there is a clear correlation between
a high rate of reaction and a low kCO. This correlation, together
with those of ν(CN) values for the CN- and CNCH3 ligands,
supports the proposed SN2 nucleophilic mechanism. The cor-
relation also suggests that it is the electronic, rather than the
steric, properties of the complexes that control the rate of
nucleophilic attack.
In conclusion, results described in this paper show that the
nucleophilicities of transition metal cyanide complexes LnM-
CN depend substantially upon the metals and ligands (L) in
the complexes and that these nucleophilicities are generally
correlated with ν(CN) for the CN- and CNCH3 ligands and
with the kCO parameter. Low values of all of these parameters
indicate complexes with more highly nucleophilic CN- ligands.
kCO ) k +
ꢀ
θ + (197 N m-1)
L
(7)
∑
d
L
and parameters,23 we calculated kCO values for the LnM-CO+
complexes corresponding to each of the cyanide complexes in
Table 1, except for 4CN. In Figure 2 is shown a plot of the ln
k2 rate constant (eq 2) vs kCO values for the complexes, where
Acknowledgment. We appreciate the support of the National
Science Foundation through Grant No. CHE-9414242.
Supporting Information Available: Rate constants for the reactions
(eq 2) of LnM-CN with MeONs in DCE at 30.0 ( 0.2 °C. This material
is available free of charge via the Internet at http://pubs.acs.org.
(23) Timney J. A. Inorg. Chem. 1979, 18, 2502.
θ
(24) The ꢀL parameter for the phosphorus donor groups in dppm, dppe,
θ
and tripod were approximated by the ꢀL value for PPh2Me, which
IC981014X
θ
was estimated from ꢀLθ values for PPh3 and PMe3 as follows: (2ꢀPPh3
+ ꢀPMe3θ)/3.
(25) Angelici, R. J.; Blacik, L. J. Inorg. Chem. 1972, 11, 1754.