A comprehensive investigation of the chemistry and basicity of a parent amidoruthenium complex

Published on Web 04/06/2002

A Comprehensive Investigation of the Chemistry and Basicity

of a Parent Amidoruthenium Complex

J. Robin Fulton, Stepan Sklenak, Marco W. Bouwkamp, and Robert G. Bergman*

Contribution from the Department of Chemistry and the Center for New Directions in Synthesis

(CNDOS), UniVersity of California, Berkeley, California 94720

Received August 2, 2001

Abstract: trans-(DMPE)₂Ru(H)(NH₂) (1) dehydrogenates cyclohexadiene and 9,10-dihydroanthracene to

yield benzene (or anthracene), (DMPE)₂Ru(H)₂, and ammonia. Addition of fluorene to 1 results in the

formation of the ion pair [trans-(DMPE)₂Ru(H)(NH₃)⁺][A^-] (A^-) fluorenide, 4a). Complex 1 also reacts

with weak acids A-H (A-H ) phenylacetylene, 1,2-propadiene, phenylacetonitrile, 4-(R,R,R-trifluoromethyl)-

phenylacetonitrile, cyclobutanone, phenol, p-cresol, aniline) to form ammonia and trans-(DMPE)₂Ru(H)(A)

(7, 8, 9a, 9b, 10, 11b, 11c, 12, respectively). In the cases where A-H ) phenylacetylene, cyclobutanone,

aniline, phenol, and p-cresol, the reaction was observed to proceed via ion pairs analogous to 4a. Compound

1 is reactive toward even weaker acids such as toluene, propylene, ammonia, cycloheptatriene, and

dihydrogen, but in these cases deuterium labeling studies revealed that only H/D exchange between A-H

and the ND₂group is observed, rather than detectable formation of ion pairs or displacement products.

Addition of triphenylmethane to 1 results in the formation of an equilibrium mixture of 1, triphenylmethane,

and the ruthenium/triphenylmethide ion pair 4h. If the energetics of ion-pair association are ignored, this

result indicates that the basicity of 1 is similar to that of triphenylmethide. All these observations support

the conclusion that the NH₂group in amido complex 1 is exceptionally basic and as a result prefers to

abstract a proton rather than a hydrogen atom from a reactive C-H bond. The energetics and mechanism

of these proton-transfer and -exchange reactions are analyzed with the help of DFT calculations.

Introduction

have shown that it readily oxidizes cyclohexadiene (CHD) to

benzene and the corresponding Fe²⁺-MeOH complex. This

Late transition metal complexes containing nondative metal-

oxygen or -nitrogen single bonds (for brevity, defined here as

“M-X” bonds) serve many different roles in biological systems

and have been implicated as intermediates in several industrially

important catalytic reactions.^1-3However, the nature of the

M-O (alkoxo, hydroxo) and M-N (amido) bonds is not

completely understood. Some of these complexes exhibit the

ability to abstract a hydrogen atom from a weak C-H bond,

suggesting that the metal-heteroatom bond can participate in

homolytic reactions.^4-6For example, Stack and co-workers

prepared [Fe^III(PY5)(OMe)][OTf]₂(PY5 ) 2,6-bis((2-pyridyl)-

methoxymethane)pyridine)⁴as a model for lipoxygenase^7-13and

reaction is proposed to proceed by initial H-atom abstraction

by the Fe-OMe group to give an intermediate cyclohexadienyl

radical.

However, direct observation of this type of odd electron

reactivity in simple M-X complexes appears to be rare and

may depend on the possibility of a favorable change in the

oxidation state of the metal. Most late metal amido and alkoxo

M-X complexes exhibit reactivity consistent with a strongly

polarized M-X bond, sometimes to the extent that the reactivity

of the heteroatom is analogous to that of an alkali metal salt.^1,14

Despite this, there is little evidence for direct ionization of M-X

bonds, and in some systems there is strong evidence that the

heteroatom is tightly bound to the metal center.¹⁵Recent

quantitative equilibrium studies by our group,^16,17in combination

with earlier studies by Bryndza and Bercaw,¹⁸have concluded

that the heteroatom ligands that are better able to stabilize a

negative charge form significantly stronger M-X bonds than

* To whom correspondence should be addressed. E-mail: bergman@

cchem.berkeley.edu.

(1) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163.

(2) Roundhill, D. M. Chem. ReV. 1992, 92, 1.

(3) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. ReV. 1996, 96, 2239.

(4) Jonas, R. T.; Stack, T. D. P. J. Am. Chem. Soc. 1997, 119, 8566.

(5) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441.

(6) Lockwood, M. A.; Wang, K.; Mayer, J. M. J. Am. Chem. Soc. 1999, 121,

11894.

(7) Samuelsson, B.; Dahle´n, S.-E.; Lindgren, J. A.; Rouzer, C. A.; Serhan, C.

N. Science 1987, 237, 1171.

(8) Glickman, M. H.; Wiseman, J. S.; Klinman, J. P. J. Am. Chem. Soc. 1994,

116, 793.

(13) Clapp, C. H.; Senchak, S. E.; Stover, T. J.; Potter, T. C.; Findeis, P. M.;

Novak, M. J. J. Am. Chem. Soc. 2001, 123, 747.

(14) Caulton, K. G. New J. Chem. 1994, 18, 25.

(9) Nelson, M. J.; Cowling, R. A.; Seitz, S. P. Biochemistry 1994, 33, 4966.

(10) Nelson, M. J.; Seitz, S. P.; Cowling, R. A. Biochemistry 1990, 29, 6897.

(11) Moiseyev, N.; Rucker, J.; Glickman, M. H. J. Am. Chem. Soc. 1997, 119,

3854.

(12) Ogo, S.; Wada, S.; Watanabe, Y.; Iwase, M.; Wada, A.; Garata, M.;

Jitsukawa, K.; Masuda, H.; Einaga, H. Angew. Chem., Int. Ed. 1998, 37,

2102.

(15) Simpson, R. D.; Bergman, R. G. Organometallics 1993, 12, 781.

(16) Holland, P. L.; Andersen, R. A.; Bergman, R. G.; Huang, J.; Nolan, S. P.

J. Am. Chem. Soc. 1997, 119, 12800.

(17) Holland, P. L.; Andersen, R. A.; Bergman, R. G. Comments Inorg. Chem.

1999, 21, 115.

(18) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.; Tam, W.; Bercaw, J. E. J.

Am. Chem. Soc. 1987, 109, 1444.

9

4722

J. AM. CHEM. SOC. 2002, 124, 4722-4737

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

those which do not have this ability. This provides further

evidence for the significant polar nature of these complexes.

We recently reported the synthesis of trans-(DMPE)₂Ru(H)-

(NH₂) (1) (DMPE ) 1,2-dimethylphosphinoethane).¹⁹Complex

1 is the first isolable monomeric structurally characterized late-

metal parent amido complex. The handful of known monomeric

parent amido complexes exhibit nucleophilic behavior;^20-22

however, the instability of these complexes, and/or the impure

reaction mixtures in which they are generated, has prevented a

thorough study of their properties. Most known monomeric late

metal amido complexes are anilido based and are therefore

stabilized by electronic delocalization into the π-system of the

aryl ring.^17,23-27Our monomeric parent amido complex 1 lacks

this added stabilization, providing us with an opportunity to

directly study the unperturbed metal-nitrogen bond. This paper

describes studies that provide evidence that is consistent with

2e^-chemistry in this system and suggest that the highly polar

nature of the Ru-N bond dictates the reactivity of this molecule.

Results

Figure 1. ORTEP diagram of 4a.

Reaction of trans-(DMPE)₂Ru(H)(NH₂) with Cyclohexa-

diene and 9,10-Dihydroanthracene. Treatment of the ruthe-

nium amido complex 1 with 1,4-cyclohexadiene (1,4-CHD) in

THF-d₈(eq 1) produces benzene in greater than 95% yield as

Scheme 1

1

observed by H NMR spectroscopy. The coproduct was not a

Ru-NH₃complex, but instead was a mixture of the known

(DMPE)₂Ru(H)₂(3) and free ammonia, both in >95% yield.²⁸

Thus, in analogy to Stack’s system, 1 oxidizes 1,4-CHD to

benzene. However, in contrast to Stack’s system, the formal

oxidation state of the metal does not change.⁴

atom or a proton (vide infra). To explore this question, we

extended the above investigation to a series of organic com-

pounds having C-H bonds characterized by a wide range of

bond dissociation energies and acidities.

Deprotonation of Fluorene by Amido Complex 1 and

Hydroxo Complex 2: Formation of Ion Pairs. Ammonia-

bound complex 4a was formed in 90% yield upon treatment of

1

A similar, albeit slower, reaction was observed when hydroxo

complex 2 was used as the oxidant. Water was presumably

formed instead of ammonia, although no resonances corre-

sponding to water could be found. In a similar fashion, both

amido complex 1 and hydroxo complex 2 oxidize 9,10-

dihydroanthracene (DHA) to anthracene (eq 2) in greater than

95% yield. However, compounds 1 and 2 also catalyze (more

rapidly) the interconversion of 1,4- and 1,3-CHD (t_1/2< 5 min)

prior to dehydrogenation (eq 3), a process not reported in the

Stack system.

amido complex 1 with fluorene in THF (Scheme 1). The H

NMR spectrum of this material shows coordinated ammonia

proton signals at δ -0.6 and the bridgehead proton on the

fluorenide anion at δ 5.94.

Consistent with the proposed ionic structure, this complex is

sparingly soluble in aromatic solvents and insoluble in aliphatic

solvents. Crystals suitable for an X-ray diffraction study were

obtained by slow diffusion of pentane into a THF solution of

4a in THF at -30 °C. An ORTEP diagram is shown in Figure

1. Crystal and data collection parameters are given in Table 1,

and selected intramolecular distances and bond angles are given

in Table 2. The crystal structure confirms the presence of a

coordinated ammonia ligand. The Ru-N bond distance is 2.252-

(7) Å, longer than the Ru-N bond distance of the parent

complex (2.191(6) Å). The distance between the fluorenide plane

and the nitrogen atom of the coordinated ammonia ligand is

3.00 Å.

Among several possible mechanisms, these processes could

in principle occur by initial abstraction of either a hydrogen

(24) (a) Bergman, R. G. Polyhedron 1995, 14, 3227. (b) Fulton, J. R.; Holland,

A. W.; Fox, D. J.; Bergman, R. G. Acc. Chem. Res. 2002, 35, 44. (c) Some

of the results described here were reported earlier in preliminary form:

Fulton, J. R.; Bouwkamp, M. W.; Bergman, R. G. J. Am. Chem. Soc. 2000,

122, 8799.

(19) Kaplan, A. W.; Ritter, J. C. M.; Bergman, R. G. J. Am. Chem. Soc. 1998,

120, 6828.

(20) Dewey, M. A.; Knight, D. A.; Gladysz, J. A. Chem. Ber. 1992, 125, 815.

(21) Joslin, F. L.; Johnson, M. P.; Mague, J. T.; Roundhill, D. M. Organome-

tallics 1991, 10, 2781.

(25) Martin, G. C.; Palenik, G. J.; Boncella, J. M. Inorg. Chem. 1990, 29, 2027.

(26) Cowan, R. L.; Trogler, W. C. Organometallics 1987, 6, 2451.

(27) Li, J. J.; Li, W.; James, A. J.; Holbert, T.; Sharp, T.; Sharp, P. R. Inorg.

Chem. 1999, 38, 1563.

(28) Dihydride complex 3 exists as an equilibrium mixture of cis and trans

isomers in solution, see: Field, L. D.; Hambley, T. W.; Yau, B. C. K.

Inorg. Chem. 1994, 33, 2010.

(22) Tahmassebi, S. K.; McNeil, W. S.; Mayer, J. M. Organometallics 1997,

16, 5342.

(23) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. J. Am. Chem. Soc. 1988,

110, 6738.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4723

A R T I C L E S

Fulton et al.

Table 1. Crystallographic Data for 4a, 9b, 10, and 4e

4a

9b

10

4e

empirical formula

fw

crystal system

a (Å)

b (Å)

c (Å)

R (deg)

â (deg)

γ (deg)

V (Å³)

space group

Z

RuP₄NC₂₉H₅₃

656.71

monoclinic

17.5506(3)

9.2847(2)

35.0153(7)

RuP₄F₃NC₂₁H₃₈

586.5

orthorhombic

18.00(2)

12.5999(14)

11.6542(13)

RuP₄OC₁₆H₃₈

471.44

RuP₄O_1.41NC_21.71H₄₃

565.63

monoclinic

22.1659(1)

11.6666(1)

32.4664(2)

triclinic

9.3817(2)

15.8315(3)

16.5172(2)

117.042(1)

90.520(1)

89.994(1)

2184.92(7)

Ph1 (No. 2)

4

1.433

984.00

50.63

52.2

0.029; 0.035; 0.041

1.59

91.7570(10)

106.977(1)

5703.13(17)

2643.1(4)

5803.98(8)

P(2)₁/a (No. 14)

Pnma (No. 62)

P(2)₁/a (No. 14)

8

4

8

D

_calcg/cm³

1.530

1.474

1.295

F_ooo

2768.00

1208.00

2364.32

µ (Mo KR) cm^-1

2θ_max

residuals: R; R_w; R_all

GOF indicator

7.99

52.5°

8.66

52.1°

7.75

49.4°

0.043; 0.045; 0.134

0.98

0.049; 0.052; 0.157

1.47

0.028; 0.028; 0.075

0.88

Table 2. Selected Intermolecular Distances (Å) and Angles (deg)

for 4a, 4e, 9b, and 10

Scheme 2

4a

Ru-P (av)

Ru-N1

2.230

2.252(7)

3.00

P-Ru-N1 (av)

Ru-N1-C1

93.3

N1‚‚‚fluorenide

plane

4e

Ru-P (av)

Ru-N (av)

O‚‚‚N (av)

O-C_aryl(av)

Ru-N1

N1-C1

Ru-P (av)

Ru-C13

C13-C14

C13-C16

C14-C15

C15-C16

C16-O1

2.297

2.251)

2.932

1.305

2.13(2)

1.18(2)

2.303

2.338(3)

1.546(5)

1.451(5)

1.556(5)

1.527(5)

1.218(4)

93.1

9b

10

171(2)

P-Ru-C13 (av)

Ru-C13-C14

Ru-C13-C16

C14-C13-C16

C13-C14-C15

C14-C16-C16

C13-C16-C16

O1-C16-C13

O1-C16-C15

94.9

119.0(2)

112.6(2)

89.9(3)

88.3(2)

86.8(3)

92.9(3)

136.7(3)

130.3(3)

Reaction of Amido Complex 1 with Weak Acids: Dis-

placement of Ammonia. Treatment of amido complex 1 with

phenylacetylene results in the loss of ammonia and formation

of the phenyl-acetylide complex 7 in greater than 95% yield

(Scheme 2). This is analogous to the earlier-reported reaction

that gives compound 7 when hydroxo complex 2 is treated with

phenylacetylene.³⁰Addition of 1,2-propadiene to amido complex

1 results in loss of ammonia and a net 1,3-hydrogen shift to

yield ruthenium methylacetylide complex 8 in 73% yield after

2 days at room temperature in toluene (Scheme 2). Complex 8

can be synthesized independently by treating 1 with propyne.

Intriguingly, the rates of reaction of 1 with propyne and 1,2-

propadiene are not significantly different.

Complex 1 also functions as a prototropic isomerization

catalyst in this system: addition of an excess of either 1,2-

propadiene or propyne to the amido complex 1 results in the

interconversion of the two organic species to an equilibrium

ratio of 1:4 (1,2-propadiene:propyne, eq 5). No prototropic

isomerization is observed when 1,2-propadiene is added to a

solution of methylacetylide complex 8.

In the UV-vis spectrum, the λ_maxof 4a in a dilute THF

solution is 368 nm, similar to that of the fluorenide-tetrabuty-

lammonium ion pair.²⁹Displacement of ammonia from 4a to

generate a covalent ruthenium-fluorenide complex 5 was not

observed, even on prolonged heating (Scheme 1).

Treatment of the hydroxo complex 2 with fluorene results in

the formation of an equilibrium mixture of 2, fluorene, and the

aquo-fluorenide complex 6a (eq 4). The aromatic resonances

in the ¹H NMR spectrum in THF-d₈at 25 °C are broad,

indicating the presence of a relatively fast exchange process

occurring on the NMR time scale. Two hydride resonances

corresponding to the hydroxo complex 2 (δ -21.4) and the

aquo-fluorenide salt complex 6a (δ -20.7) are observed. Upon

cooling the reaction mixture, both hydride resonances shift to

higher field, and the hydride resonance corresponding to hydroxo

complex 2 decreases in intensity such that at -80 °C, the only

hydride resonance present in the ¹H NMR spectrum corresponds

to 6a (δ -23.4). In addition to this resonance, four distinct aryl

resonances and a new singlet at δ 5.97 are present. These data

are consistent with a shift in the equilibrium to favor 6a at lower

temperatures. Repeated attempts to isolate this complex failed.

The addition of benzonitrile and acetonitrile to amido complex

1 led to a complex mixture of products. However, reaction with

phenylacetonitrile in benzene-d₆resulted in ammonia formation

and clean conversion to keteniminate complex 9a in 79% yield

(29) Smid, J. In Spectrophotometric Studies of Ion-Pair Equilibria; Szwarc, M.,

Ed.; Wiley-Interscience: New York, 1972; Vol. 1, pp 85-151.

(30) Kaplan, A. W.; Bergman, R. G. Organometallics 1998, 17, 5072.

9

4724 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

Figure 3. ORTEP diagram of 10, first enantiomer in the unit cell.

Scheme 3

Figure 2. ORTEP diagram of 9b. The disorder in the structure is not shown.

after 8 h at 25 °C (Scheme 2). Similarly, addition of 4-(R,R,R-

trifluoromethyl)phenylacetonitrile to amido complex 1 resulted

in the formation of a purple solution of keteniminate complex

9b in 74% yield after 1 h at 25 °C (Scheme 2). X-ray quality

crystals were grown by slowly cooling a concentrated toluene

solution of 9b to -30 °C. An X-ray diffraction study confirmed

the connectivity shown in Scheme 2; the ORTEP diagram is

shown in Figure 2. Crystal and data collection parameters are

given in Table 1, and selected intramolecular distances and bond

angles are given in Table 2. The molecule lies on a crystal-

lographic mirror plane within the unit cell, and thus only half

of the molecule is unique. This requires disorder in the

trifluoromethyl group and the two DMPE ligands, limiting the

accuracy of the structure determination. However, the analysis

clearly shows the connectivity of 9b, and the bond distances

are consistent with other known keteniminate complexes.^31,32

Addition of cyclobutanone to amido complex 1 also resulted

in ammonia displacement and the generation of a carbon-bound

enolate complex (10) in 59% yield after 8 h at 25 °C (Scheme

2). The ¹H NMR spectrum of 10 reveals five inequivalent broad

proton resonances integrating to one proton each. The ¹³C{¹H}

NMR spectrum contains signals due to two methylene carbons

at δ 41.8 and 23.5, a quaternary carbon at δ 212.7, and a

quintuplet corresponding to a methine resonance at δ 40.47 (J_C-P

) 18.5 Hz). The IR spectrum reveals a strong band at 1636

cm^-1, indicative of a carbonyl moiety, along with a band at

1832 cm^-1corresponding to the Ru-H stretch. The carbonyl

frequency is substantially red-shifted as compared to that in

cyclobutanone, presumably due to interaction of the CdO π

system with the R-metal-carbon bond.^33-35These data imply

that the enolate moiety is bound through the â-carbon of the

cyclobutanone functionality.

The proposed connectivity was confirmed by an X-ray

diffraction study on a single crystal grown by slowly cooling a

concentrated diethyl ether solution to -30 °C. Two independent

molecules exist in the asymmetric unit; an ORTEP diagram of

one of the enantiomers is shown in Figure 3. Crystal and data

collection parameters are given in Table 1, and selected

intramolecular distances and bond angles are given in Table 2.

The C-O bond distance of 1.218(4) Å is longer than that of a

typical carbon-oxygen double bond, and the C1-C2 bond

length of 1.451(5) Å is shorter than that of the other carbon-

carbon bonds in the cyclobutanone moiety, indicating significant

electronic delocalization within the enolate system.

Mechanism of Ammonia Displacement: Formation of a

Metastable Ammonia-Bound Ion Pair Intermediate. Mecha-

nistic insight was gained by monitoring the reactions of amido

complex 1 with cyclobutanone and phenylacetylene. An inter-

mediate that we believe is the (possibly hydrogen-bonded; see

next section) ammonia-bound complex 4b (Scheme 3) is

observed upon monitoring the reaction between amido complex

1

1 and cyclobutanone by H and ¹³C{¹H} NMR spectroscopy.

The ¹H NMR spectrum taken after 10 min at 25 °C in THF-d₈

reveals a singlet corresponding to the enolate proton H_aat δ

3.36 and singlet resonances due to protons H_band H_cat δ 1.71

and 2.13, respectively. The resonances corresponding to the

bound NH₃hydrogens are found at δ 2.38. The NH₃resonance

is observed downfield relative to the NH₃protons of the

fluorenide ion pair 4a, indicating a decrease in electron density

around these protons perhaps due to hydrogen bonding to the

enolate anion. The ¹³C{¹H} NMR spectrum revealed the

(31) Bruce, M. I.; Wallis, R. C.; Skelton, B. W.; White, A. H. J. Chem. Soc.,

Dalton Trans. 1981, 2205.

(32) Tellers, D. M.; Ritter, J. C. M.; Bergman, R. G. Inorg. Chem. 1999, 38,

4810.

(33) Albe´niz, A. C.; Catalina, N. M.; Espinet, P.; Redo´n, R. Organometallics

1999, 18, 5571.

(34) Hartwig, J. F.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1990,

112, 5670.

(35) Hartwig, J. F.; Bergman, R. G.; Andersen, R. A. Organometallics 1991,

10, 3326-3344.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4725

A R T I C L E S

Fulton et al.

Table 3. Selected IR Data^a(in cm^-1) for Ion Pairs 4a, 4b, 4c, 4d,

4e, 4f, and 4g

4a

4b

4c

4d

4e

4f

4g

NH₃:^b3331 3353

3272 3296

3361

3351 3357 not observed 3235

3256

1920

3312 3295

3223 3282

1901 1878

3182

1902

Ru-H 1916 1872

1869

other

C-O: 1597 CtC: 1994

O-H: 2653, 2518

^aAll data taken in Nujol mulls. The O-H stretch for MeOH in Nujol

appears at 3370 cm^-1(br). ^bSymmetric and asymmetric.

Table 4. Selected ¹H NMR (in ppm, THF-d₈Solvent) Data for Ion

Pairs 4a, 4b, 4c, 4d, 4e, 4f, and 4g

4a

4b

4c

4d

4e

4f

4g

NH₃-0.6

2.38

2.50

2.53^b

2.55^c

3.31^b

2.92^d

Ru-H -20.34 -19.4^a-19.19^a-19.48^b-19.45^c-19 (br) -19.34^d

Figure 4. ORTEP diagram of 4e. The ruthenium hydrides and the nitrogen

hydrogens were located. The disordered solvent molecule is not shown.

^a222 K. ^b193 K. ^c203 K. ^d213 K.

Scheme 4

1 with a hexanes solution of the substrate followed by removal

of the volatile materials under reduced pressure.

X-ray quality crystals of 4e were grown by slowly cooling a

concentrated THF solution of cresolate ion pair 4e layered with

pentane to -30 °C for 1 month, and an X-ray diffraction study

was carried out. An ORTEP diagram is shown in Figure 4.

Crystal and data collection parameters are given in Table 1,

and selected intramolecular distances and bond angles are given

in Table 2. The hydride and the N-H protons were located in

the difference Fourier map. A disordered solvent molecule

cocrystallized within the asymmetric unit cell. The ion pair 4e

exists as a hydrogen-bonded dimer in the solid state with two

ruthenium ammonia cations bridged by two cresolate anions.

The average distance between nitrogen and oxygen is 2.932 Å,

indicative of hydrogen bonding between these two atoms. Two

of the ammonia hydrogens participate in the hydrogen bonding,

while the other is orthogonal to the hydrogen bonding plane.³⁶

The ruthenium-nitrogen bond length of 2.251 Å is very similar

to that found in the crystal structure of the fluorenide ion pair

4a.

In the aryl alcohol cases, ammonia displacement was

observed. However, in sharp contrast to the reactions with

carbon acids, an equilibrium was established over 1 h between

the ion pairs 4d and 4e and their corresponding ruthenium-

phenolate or -cresolate complexes 11b and 11c and ammonia

(Scheme 4).^30,37Cooling the reaction mixture to -50 °C results

in a change in the ratio of ruthenium complexes favoring the

ion pairs 4d or 4e. Warming the reaction mixture to 25 °C results

in the reestablishment of the original distribution. Addition of

excess ammonia to the reaction mixtures at 25 °C shifts the

equilibrium in favor the ion pairs 4d or 4e. Similarly, addition

of ammonia to pure ruthenium-phenolate complex 11b or

ruthenium-cresolate complex 11c results in the formation of ion

pair 4d. These observations are all consistent with the presence

of a rapidly equilibrating mixture of 1, 4d, and 11b or 1, 4e,

and 11c.

presence of two methylene carbons and one methine carbon.

The symmetric and asymmetric stretching vibrations for the

N-H bond were observed at 3353 and 3296 cm^-1in the IR

spectrum, along with a Ru-H stretch at 1872 cm^-1and a C-O

stretching vibration at 1597 cm^-1. Addition of a hexanes solution

of cyclobutanone to a hexanes solution of amido complex 1

resulted in the precipitation of complex 4b. This white solid

can be kept indefinitely at room temperature. Upon redissolution

of the solid in THF or benzene, the ion pair 4b can be observed,

but it disappears concurrently with formation of the carbon-

bound enolate complex 10 and ammonia.

Similarly, addition of phenylacetylene to 1 in THF-d₈resulted

in immediate formation of an ammonia-bound ruthenium cation/

phenylacetylide anion pair 4c (Scheme 3, see Tables 3 and 4

1

for relevant IR and H NMR data). Addition of a hexanes

solution of phenylacetylene to a hexanes solution of amido

complex 1 resulted in the precipitation of complex 4c. Like 4b,

this tan solid can be kept indefinitely at low temperatures.

However, upon redissolution in THF-d₈or C₆D₆, the ion pair

4c is observed by NMR spectroscopy to disappear concurrently

with formation of the phenylacetylide complex.

Reaction of 1 with Alcohols and Amines: Evidence for

Hydrogen Bonding in the Ion Pair. Addition of phenol,

p-cresol, methanol, and aniline to amido complex 1 resulted in

the formation of ion pairs 4d, 4e, 4f, and 4g, respectively

(Scheme 4). The relevant IR and NMR data are presented in

Tables 3 and 4. In all cases, white powders of the ion pairs are

isolated cleanly by mixing a hexanes solution of amido complex

In the case of methanol, the expected ruthenium-methoxide

complex was not observed after ion pair formation. Instead,

(36) For an example of a platinum-bound ammonia complex which is hydrogen

bonded to an aryloxide anion, see: Sigel, R. K. O.; Freisinger, E.; Metzger,

S.; Lippert, B. J. Am. Chem. Soc. 1998, 120, 12000.

(37) Burn, M. J.; Fickes, M. G.; Hollander, F. J.; Bergman, R. G. Organome-

tallics 1995, 14, 137.

9

4726 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

Table 5. pK_a’s and BDEs of Compounds Reactive toward Amido

Complex 1

complex 4f disappears concurrent with formation of dihydride

3, formaldehyde, and ammonia over 1.5 d at 25 °C. However,

in analogy to the reactions observed between 1 and the

aforementioned carbon acids, ion pair 4g forms the known

anilido complex 12 and ammonia after 1 d at 25 °C in THF or

benzene.³⁷Confirming the exoergicity of this reaction, addition

of ammonia to a THF-d₈solution of ruthenium-anilido complex

12 did not result in the formation of anilido ion pair 4g or amido

complex 1.

compound

pK

a

BDE (kcal mol^-1

)

phenol

18.0 (DMSO)^a

18^c

89.95^b

∼84

88.7^d,e

84.7^d

80^g

4-(R,R,R-trifluoromethy)-

phenylacetonitrile

p-cresol

phenylacetonitrile

fluorene

cyclobutanone

phenylacetylene

methanol

water

triphenylmethane

dihydrogen

cycloheptatriene

ammonia

toluene

18.9 (DMSO)

22.3^c

22.9 (THF)^f

25.1 (DMSO)^h

28.7 (DMSO)^h

29.0 (DMSO)^j

31.2 (DMSO)^c

31.5 (THF)^f

35^j

>125^a,j

102 (O-H)^j

118.1^k

84^g

H/D Exchange Reactions with Very Weak Acids. Amido

complex 1 also reacts with exceedingly weak acids. One of the

most striking reactions of 1 occurs upon its dissolution in

104^j

1

2

toluene-d₈. Monitoring this reaction by H and H{¹H} NMR

spectroscopy revealed complete deuterium incorporation into

the N-bound positions of the amido complex 1 followed by

deuterium incorporation into the methylene and methyl protons

of the DMPE ligand itself, concurrent with hydrogen incorpora-

tion into the benzyl position of toluene (eq 6).

38.8^j

73ⁱ

41^h

107.4ⁱ

88.0^l

41^j

propene

benzene

43^j

86.3^l

43^j

110.9^l

^aBordwell, F. G.; McCallum, R. J.; Olmstead, W. N. J. Org. Chem.

1984, 49, 1424. ^bBordwell, F. G.; Cheng, J.-P. J. Am. Chem. Soc. 1991,

113, 1736. ^cBordwell, F. G.; Bauxch, M. J. J. Am. Chem. Soc. 1986, 108,

1979. ^dReference 65b. ^eZhang, X.-M.; Bordwell, F. G. J. Am. Chem. Soc.

1994, 116, 968. ^fStreitwieser, A.; Wang, D. Z.; Stratakis, M.; Facchettis,

A.; Gareyev, R.; Abbotto, A.; Krom, J. A.; Kilway, K. V. Can. J. Chem.

1998, 76, 765. ^gBordwell, F. G.; Cheng, J.-P.; Ji, G.-Z.; Satish, A. V.;

Zhang, X. J. Am. Chem. Soc. 1991, 113, 9790. ^hReference 54. ⁱReference

52. ^jReference 55. ^kBerkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys.

Chem. 1994, 98, 2744. ^lEllison, G. B.; Davico, G. E.; Bierbaum, V. M.;

Depuy, C. H. Int. J. Mass Spectrom. Ion Processes 1996, 156, 109.

A similar, albeit slower, exchange is observed with hydroxo

complex 2. Hydrogen incorporation into the aryl ring of toluene

is not observed, and deuterium is not incorporated into the

ruthenium hydride position. Surprisingly, in contrast to the

behavior of 1 with the stronger acids discussed above, no

displacement of ammonia to form the benzyl complex 13 is

observed, even at elevated temperatures (eq 7). In control

experiments, treatment of a solution of trans-(DMPE)₂Ru(H)-

(Cl) with benzylmagnesium chloride generated the benzyl

complex 13 in 35% yield after 3 h at 25 °C. Dissolution of 13

into liquid ammonia did not result in the conversion of the

benzyl complex 13 to the amido complex 1.

bicycloheptatrienyl (from cycloheptatriene), 1,5-hexadiene (from

propene), hexaphenylethane (from triphenylmethane), or bi-

fluorenyl (from fluorene) are observed in any of these reac-

tions.^38,39

In analogy to the toluene reaction and in contrast to the

behavior observed with stronger acids, no displacement of

ammonia is observed in any of the exchange reactions between

1 and very weak acids listed above. This extends to NH₂as a

potential entering group; despite rapid H/D exchange between

1 and ND₃, addition of ¹⁵NH₃to the amido complex does not

result in incorporation of labeled nitrogen into the amido

complex 1, even after extended reaction times (eq 9). Labeled

nitrogen is observed in the presumably more thermodynamically

stable cis isomer, which is formed irreversibly after 1 month at

room temperature.

H/D Exchange Reactions Catalyzed by Amidoruthenium

Complex 1. Consistent with the observation that amido complex

1 undergoes H/D exchange with toluene, complex 1 can also

be used to catalyze deuterium exchange between toluene-d₈and

carbon acids of similar or stronger acidities but widely varying

bond dissociation energies (Table 5 and eq 8).

Equilibrium Measurements between the Amido and Hy-

droxo Complexes and Weak Acids of Similar Conjugate

Base Strength. To obtain an estimate of the intrinsic basicity

of 2, we examined the reaction of the hydroxo complex with

several different carbon acids of moderate acidity in THF-d₈.

In the case of fluorene, NMR analysis showed an equilibrium

mixture of 2, fluorene, and the aquo-fluorenide complex 6a (eq

4). The λ_maxof 368 nm compares well with other known contact

ion pair (CIP) of fluorenide salts (vide infra). Assuming that

the product is a CIP, the data at several different concentrations

were analyzed according to the equilibrium expression in eq

Compounds that do undergo H/D exchange include triph-

enylmethane, cycloheptatriene, ammonia, dihydrogen, and pro-

pene; those that do not (even at elevated temperatures) include

benzene, THF, ethylene, and N-N-dimethyl-p-toluidine. Expo-

sure of 1 for extended periods to an atmosphere of dihydrogen

or ethylene results in the loss of ammonia and formation of

either the dihydride complex 3 or the ethylene complex

(DMPE)₂Ru(C₂H₄), respectively. No detectable amounts of

radical coupling products such as bibenzyl (from toluene),

(38) Vincow, G.; Dauben, H. J.; Hunter, F. R.; Volland, W. V. J. Am. Chem.

Soc. 1969, 91, 2823.

(39) Arends, I. W. C. E.; Mulder, P.; Clark, K. B.; Wayner, D. D. M. J. Phys.

Chem. 1995, 99, 8182.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4727

A R T I C L E S

Fulton et al.

Scheme 5

Scheme 6

Figure 5. Van’t Hoff plot of the equilibrium between amido 1, triphenyl-

methane, and the salt complex 4h. This gives ∆H° ) -7.2 ( 0.4 kcal/mol

and ∆S° ) -26 ( 2 eu.

10, leading to a good fit and an equilibrium constant (K_measured(6a)

)

of 6.4 ( 0.5 M^-1at 25 °C for this process. The fact that an

essentially invariant K_measuredvalue was obtained at several

different concentrations supports the hypothesis that the products

exist as a simple CIP in THF solution at 25 °C.

The details are provided in the Supporting Information, and the

results are briefly covered at the end of the Discussion section

below.

Discussion

Oxidation of 1,4-Dienes. The addition of 1,4-CHD to the

amido complex 1 leads to the formation of benzene, ammonia,

and the dihydride complex 3 (eq 1). This result may be relevant

to the action of lipoxygenase, a non-heme iron-containing

enzyme, which converts 1,4-fatty acid dienes to alkylhydrop-

eroxides.⁷The rate-determining step in this process has been

proposed to be the homolytic cleavage of the weak C-H bond

in the organic substrate by an Fe-OH intermediate.^8-11,46,47

Stack’s [Fe^III(PY5)(OMe)][OTf]₂complex also dehydrogenates

1,4-CHD.⁴This reaction was proposed to serve as a model for

lipoxygenase, taking place by initial abstraction of a hydrogen

atom. Mayer and co-workers have shown the feasibility of M-O

mediated radical reactions by demonstrating that [(phen)₂Mn-

(µ-O)₂Mn(phen)₂][PF₆]₃abstracts hydrogen atoms from weak

C-H bonds such as those found in DHA.^5,6,48

The more strongly basic amido complex 1 completely

deprotonates fluorene. Addition of the weaker carbon acid

triphenylmethane to amido complex 1 in THF-d₈resulted in

the formation of a pale-pink solution, suggesting that some

triphenylmethide had been generated. However, it must have

1

been a very small amount, since the H NMR spectrum of the

reaction mixture at 25 °C showed resonances due only to

unreacted starting material. Upon cooling the reaction mixture

to 253 K, the color intensified, and three resonances in the aryl

region of the ¹H NMR spectrum appeared in a 2:2:1 ratio. We

have assigned these to the aryl resonances of Ph₃C^-. Further

cooling resulted in a change in the intensities of the new

resonances for Ph₃C^-relative to the resonances corresponding

to Ph₃CH. These observations are consistent with the existence

of a temperature-dependent equilibrium in which amido complex

1 deprotonates triphenylmethane to generate the Ph₃C^-/ammonia-

bound ruthenium ion pair 4h (eq 11). Integration at five different

temperatures permitted evaluation of the equilibrium constants

using the expression in eq 12. A van’t Hoff plot allowed us to

calculate ∆H° ) -7.2 ( 0.4 kcal mol^-1and ∆S° ) -26 ( 2

eu (Figure 5). This leads to an extrapolated equilibrium constant

of 4.0 ( 0.4 × 10^-1M^-1at 25 °C.

The above transformations involve a thermodynamically

accessible reduction of the metal species upon transfer of the

hydrogen atom to the alkoxo ligand.^4,5,11The oxidation state

change for our amido and hydroxo complexes (1 and 2) is

presumably less favorable than the analogous reactions in these

(42) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.

A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,

R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,

K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Mennucci, B.;

Cossi, M.; Adamo, C.; Jaramillo, J.; Cammi, R.; Pomelli, C.; Ochterski,

J.; Petersson, G. A.; Ayala, P. Y.; Morokuma, K.; Malick, D. K.; Rabuck,

A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;Baboul, A.

G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;

Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith,

T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.;

Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;

Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian

99, Development Version, revision B; Gaussian, Inc.: Pittsburgh, PA, 2000.

(43) The geometries of all species were fully optimized. Vibration frequencies

were also computed for all species to determine whether these species are

minima, transition states, or higher order saddle points on the potential

energy surface (PES) and for evaluating zero-point energies (ZPEs). The

enthalpies of the species at 0 K (DH) were calculated as the sum of the

absolute energies and the ZPE. The absolute energies, ZPEs, and Cartesian

coordinates of all species are listed in the Supporting Information.

(44) See Supporting Information for experimental write-up.

Theoretical Calculations. To gain a better understanding of

the thermodynamics involved in proton transfer and ammonia

displacement, several calculations were performed using the

hybrid density functional theory (DFT) method with the B3LYP

functional^40,41employing the Gaussian 99 series of programs.^42-44

(45) To simplify our calculations, the methyl groups on the DMPE ancillary

ligands were replaced with hydrogen atoms (compound labels 1′, 2′, etc.).

(46) Jonsson, T.; Glickman, M. H.; Sun, S.; Klinman, J. P. J. Am. Chem. Soc.

1996, 118, 10319.

(47) Glickman, M. H.; Klinman, J. P. Biochemistry 1996, 35, 12882.

(48) Wang, K.; Mayer, J. M. J. Am. Chem. Soc. 1997, 119, 1470.

(40) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.

(41) Lee, C.; Tang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.

9

4728 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

Scheme 7

Scheme 8

first row systems. Still, the existence of a more difficult

oxidation state change does not rigorously rule out a one-electron

pathway in the oxidation of CHD or DHA by 1 or 2. One can

envision a mechanism by which the nitrogen abstracts a

hydrogen from the weak C-H bond of CHD (bond dissociation

energy (BDE) ) 73 kcal/mol)⁴⁹to generate ammonia, a

cyclohexadienyl radical species, and a 17 electron Ru(I) complex

(Scheme 5, which we will refer to as the “radical pathway”).

The ruthenium radical can then abstract a second hydrogen atom

from the cyclohexadienyl radical to generate benzene and the

ruthenium dihydride complex 3.

An alternative mechanism postulates deprotonation of the

weakly acidic C-H bond of CHD to generate a transient

ammonia-bound cationic ruthenium center and the cyclohexa-

dienide anion (Scheme 6, the “ionic pathway”). This complex

could lose ammonia and abstract a hydride directly from

cyclohexadienide to afford the observed products or it could

collapse to a cyclohexadienylruthenium complex, followed by

â-hydride elimination to generate the dihydride complex 3 and

benzene. To gain further evidence for either hypothesis, we

extended this study to include the reactions between amido

complex 1 and a variety of different organic substrates pos-

sessing a wide range of bond strengths and pK_a’s.

The observation of intermediates in the addition of cyclobu-

tanone, phenylacetylene, aniline, methanol, and aryl alcohols

to amido complex 1 provided evidence that displacement

reactions proceed via ammonia-bound ruthenium ion pairs 4

(Scheme 7). The chemical shift of the resonance corresponding

1

to the N-H protons in H NMR spectra of the intermediates

appeared downfield of the N-H resonance in the fluorenide

ion pair 4a. This is indicative of a decrease in electron density

around the N-H protons and provides evidence that the above

reactants form ion pairs that are held together (and presumably

stabilized) by hydrogen bonding. All of the conjugate bases of

the acids listed above are capable of forming strong hydrogen

bonds to hydrogen bond donors.⁵⁵

Displacement of Ammonia Is Preceded by Formation of

an Ammonia-Bound Cationic Ruthenium Complex. Reaction

of Brønsted acids possessing pK_a’s < 35 with amido complex

1 led to exothermic deprotonation and, in many cases, displace-

ment of ammonia. For example, exothermic deprotonation is

observed in the reaction between 1 and fluorene, generating the

fluorenide complex 4a (Scheme 1). The ionic nature of this

complex is clearly demonstrated by its insolubility in aliphatic

solvents and by its crystal structure (Figure 1). The salt 4a shows

a nitrogen-aromatic ring distance of 3.00 Å, indicating a

potentially stabilizing cation-π interaction between the am-

monium nitrogen and the fluorenide anion.⁵⁰No displacement

of ammonia to generate the neutral ruthenium-fluorenide

complex (5) was ever observed (Scheme 1).

Displacement reactions occurred with substrates possessing

fewer steric constraints than fluorene and encompassing a wide

range of C-H bond strengths. For example, reaction of 1 with

arylacetonitriles having pK_avalues similar to or lower than that

of fluorene results in the formation of keteniminate complexes

9a or 9b and ammonia (Scheme 2). Similarly, addition of

cyclobutanone to 1 leads to the formation of a carbon-bound

(in preference to an oxygen-bound) enolate complex, a binding

motif common for this ruthenium system (Scheme 2).^30,33,34,51

Amido complex 1 also reacts with the terminal C-H bond of

phenylacetylene to generate the phenylacetylide complex 7

(Scheme 2), demonstrating that 1 reacts with very strong C-H

bonds (BDE > 125 kcal/mol)^52,53as long as they are relatively

acidic (pK_a) 28.7).⁵⁴Similarly, amido complex 1 reacts with

the more acidic O-H bond (pK_a) 29.0,⁵⁴BDE ) 102 kcal/

mol)⁵³as opposed to the weaker C-H bond (BDE ) 93 kcal/

mol)⁵³of methanol (Scheme 4).

Addition of 1,2-propadiene to amido complex 1 results in

the formation of the methylacetylide complex 8 exclusively,

even though the amido complex catalyzes the interconversion

of 1,2-propadiene and propyne (Scheme 8). Strong bases,

including sodium amide in liquid ammonia, affect prototropic

acetylene-allene rearrangements via carbanion intermediates.^56,57

We believe the amido complex promotes the isomerization via

a similar mechanism. Deprotonation of 1,2-propadiene or

propyne generates the propynyl anion/ruthenium-ammonia

complex 4j. Reprotonation can occur at either end of the anion;

repetition of this sequence ultimately generates a thermody-

namically controlled mixture of C₃species. Apparently, the more

exoergic reaction in this system involves loss of ammonia and

generation of a methylacetylide complex (8) via deprotonation

of the terminal hydrogen of propyne which initially generates

the ion pair 4k. If 3-propynyl or allenylruthenium complexes

are formed during this reaction, they must either be generated

reversibly or rearrange to the methylacetylide complex 8.

Addition of methanol to amido complex 1 results in immedi-

ate formation of the methoxide ion pair 4f (Scheme 9). In

contrast to our observations with other nucleophilic counterions,

no covalent ruthenium methoxide complex 11a is observed after

ammonia displacement. Instead, formaldehyde and dihydride

complex 3 are generated. These are the expected products of

â-hydride transfer in the ion pair, or â-hydride elimination from

the neutral ruthenium-methoxide complex 11a (Scheme 9). This

(49) Burkey, T. J.; Mafewski, M.; Griller, D. J. Am. Chem. Soc. 1986, 108,

2218.

(50) Ma, J. C.; Dougherty, D. A. Chem. ReV. 1997, 97, 1303.

(51) Slough, G. A.; Bergman, R. G.; Heathcock, C. H. J. Am. Chem. Soc. 1989,

111, 938.

(52) McMillen, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982, 33, 493.

(53) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic

Chemistry, 3rd ed.; Harper and Row: New York, 1987.

(54) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.

(55) Chabinyc, M. L.; Brauman, J. I. J. Am. Chem. Soc. 2000, 122, 8739.

(56) Cram, D. J. Stereochemistry of Substitution of Carbon Acids and Orga-

nometallic Compounds; Academic Press: New York, 1965; Vol. 4.

(57) Landor, P. D. In Synthesis of Allenes; Landor, S. R., Ed.; Academic Press:

London, 1982; Vol. 1, pp 44-61.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4729

A R T I C L E S

Fulton et al.

Scheme 9

Scheme 10

reactions discussed above clearly demonstrate the basic nature

of the metal-bound NH₂ligand in amido complex 1. The

reaction of 1 with exceptionally weak acids, such as ammonia,

toluene, and the ruthenium-bound DMPE ligands, is even more

dramatic. In these cases, the use of deuterium-labeled reaction

partners reveals that the ruthenium amide is capable of effecting

H/D exchange at a reasonable rate at room temperature. In

contrast to the behavior of 1 with stronger acids, no displacement

of ammonia is observed in these cases.

overall reaction is a common pathway of decomposition for

many late-metal alkoxide complexes.^1,58-60

Reaction between Amido Complex 1 and Aryl Alcohols:

Reversible Displacement of Ammonia. Addition of phenol to

amido complex 1 results in the formation of an equilibrium

mixture of the phenolate ion pair 4d and the ruthenium-

phenolate complex 11b and ammonia (Scheme 4). The ion pair

exists as a hydrogen bonded dimer in the solid state, similar to

our previously reported ruthenium hydroxo/aquo complex

[trans-(DMPE)₂Ru(H)(OH)]₂[H₂O]₂.⁶¹Most aryloxide hydrogen

bonded complexes are monomeric in the solid state.^{15,16,59,62-64}

However, Canty’s palladium-aquo/pentafluoro phenolate com-

plex was observed to crystallize as a dimer.⁶²Ion pair 4d may

form dimers or more complicated aggregates in solution; we

have not yet been able to obtain definitive evidence for its

solution structure by NMR.

The equilibrium established between the aryloxide ion pairs

4d and 4e and their corresponding ruthenium-aryloxide com-

plexes 11b and 11c and ammonia in THF-d₈is intriguing

(Scheme 4). Apparently, the ion pairs and neutral complexes

have comparable free energies in THF at 25 °C. Although

equilibrium constant measurements in other cases were suc-

cessful, attempts to quantify the equilibrium constant between

11 and 4 led to inconclusive results. Apparent K_eq’s measured

in dilute solutions were larger than those obtained in more

concentrated solutions, indicating the possible formation of

larger aggregates in the more concentrated solutions.

The reason for the low barrier to ammonia displacement in

the aryloxide ion pairs 4d and 4e relative to those in the anilido

(4g) and methoxide (4f) ion pairs is unclear. Similar systems,

such as phenylacetylide ion pair 4c, methoxide ion pair 4f, or

anilido ion pair 4g irreversibly lose ammonia over the course

of at least 1 day, and the enolate ion pair 4b loses ammonia

over the course of 8 h. However, in the aryloxide examples,

loss of ammonia is fairly rapid and reversible. This indicates a

low-energy pathway for ammonia displacement, potentially a

function of the relative stability of the anion. Qualitatively, as

long as they are not too sterically hindered (e.g., as in fluorenide

ion pair 4a), the more stable anions displace ammonia faster

than do the less stable anions. Possible reasons for this are

discussed below.

As with the displacement reactions discussed above, the pK_a,

and not the bond dissociation energy, of the substrate appears

to be the physical property that controls the rate of H/D

exchange. For example, no H/D exchange was observed with

the benzyl methyl hydrogens of N-N-dimethyl-p-toluidine, even

though they have bond dissociation energies similar to that of

toluene. We believe this lack of reactivity is due to the lower

acidity of the benzyl protons of N,N-dimethyl-p-toluidine.⁶⁵

Similarly, amido complex 1 undergoes H/D exchange with

ammonia, despite the exceptionally high NH₃bond dissociation

energy (107.4 kcal/mol). Additional evidence against a radical

pathway was provided by the treatment of amido complex 1

with a variety of poorly acidic substrates that possess particularly

weak C-H bonds, such as cycloheptatriene (BDE ) 73 kcal/

mol). Although these complexes were observed to undergo H/D

exchange, no products from radical dimerization, such as the

formation of bicycloheptatrienyl from cycloheptatriene, were

ever observed.^38,39These results provide further support for a

two-electron, rather than a one-electron, pathway in these

reactions, and by extension in the oxidation of the weakly acidic

hydrocarbons cyclohexadiene and 9,10-dihydroanthracene.

The most obvious mechanism for H/D exchange between

amido complex 1 and the benzyl protons of toluene is the

reversible formation of ion pair 4i (Scheme 10). However, as

noted above, displacement of ammonia by a transiently gener-

ated benzyl anion to generate benzylruthenium complex 13 is

not observed (eq 7). Analogously, ¹⁵N from free ¹⁵NH₃is not

incorporated into amido complex 1 (eq 9). Because rapid H/D

exchange between the amido protons 1 and the benzyl protons

of toluene occurs at 25 °C, either ion pair 4i must be

energetically accessible at that temperature or the proton transfer

must proceed via a concerted transition state similar to that

shown in Figure 6.

Interestingly, acids which lead to strongly basic anions result

in ammonia displacement more slowly than those which lead

to weaker bases (vide supra). We propose to address this

surprising result using the picture of the overall displacement

mechanism illustrated in Scheme 11. After the initial reversible

proton transfer step, it seems likely that loss of ammonia is

preceded by the (presumably also reversible) rupture of the

RuNH₃⁺‚‚‚A^-hydrogen bond in ion pair 4, leading to solvent-

separated ion pair 14. Rapid collapse of the resulting ion pair

15 (a dissociative mechanism for this process is illustrated, but

H/D Exchange Reactions: Evidence for an Extremely

Basic Amido Ligand. The deprotonation and displacement

(58) Blum, O.; Milstein, D. J. Organomet. Chem. 2000, 594, 479.

(59) Kuznetsov, V. F.; Yap, G. P. A.; Bensimon, C.; Alper, H. Inorg. Chim.

Acta 1998, 280, 172.

(60) Newman, L. J.; Bergman, R. G. J. Am. Chem. Soc. 1985, 107, 5314.

(61) Burn, M. J.; Fickes, M. G.; Hartwig, J. F.; Hollander, F. J.; Bergman, R.

G. J. Am. Chem. Soc. 1993, 115, 5875.

(62) Canty, A. J.; Jin, H.; Skelton, B. W.; White, A. H. J. Organomet. Chem.

1995, 503, C16.

(65) The BDEs of C-H bonds in substituted toluenes are not substantially

affected by para-substituents, see: (a) Zhang, X.-M.; Bruno, J. W.;

Enyinnaya, E. J. Org. Chem. 1998, 63, 4671. (b) Wayner, D. D. M.; Sim,

B. A. J. Org. Chem. 1991, 56, 4853.

(63) Haarman, H. F.; Kaagman, J.-W. F.; Smeets, W. J. J.; Spek, A. L.; Vrieze,

K. Inorg. Chim. Acta 1998, 270, 34.

(64) Kim, Y.-J.; Choi, J.-C.; Osakada, K. J. Organomet. Chem. 1996, 491, 97.

9

4730 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

ion pair does not dissociate easily. Our results contrast with

the observation of Schaad and Landis in the similar system,

trans-(DMPE)₂Ru(H)(NHCOCF₃).⁶⁶Addition of ¹⁵NH₂COCF₃

to this complex does result in ligand exchange, providing support

for a facile dissociation of the ligand from the metal center.

The only difference between their system and ours is the relative

stability of the free anions: [NHCOCF₃]^-is a much weaker

-

Figure 6. Proposed transition state for H/D exchange between amido

complex 1 and toluene.

base than NH₂

.

pK_aDetermination and the Nature of the Ion Pair. The

equilibrium constant measured (K_measured) for the conversion of

the hydroxo complex 2 and fluorene to the fluorenide salt (eqs

4, 17) 6a is controlled by the relative pK_a’s of the ruthenium

aquo complex 6a and that of fluorene. Similarly, the equilibrium

constant measured for the reaction of amido complex 1 with

triphenylmethane to give triphenylmethide salt 4h (eqs 11, 16)

is controlled by the pK_aof the ammonia-bound ruthenium

complex 4 and the pK_aof triphenylmethane. Spectroscopic and

chemical investigations (see below) showed that the product of

the proton transfer is the contact ion pair (CIP) 6a rather than

a solvent-separated ion pair (SSIP) or free ions. Therefore, the

equilibrium is most appropriately written as the three component

equation shown in eqs 13 and 14.

Scheme 11

the displacement could be associative) would then lead to the

covalent product.

In the case of acids such as phenol, cyclobutanone, etc., the

initial reaction with 1 is either energetically downhill or close

to thermoneutral, leading to substantial concentrations of the

initial ion pair that results from proton transfer. The free energy

barrier that must be crossed to give the final displacement

product is then only that associated with further reaction of the

ion pair. In the case of very weak acids, the positive ∆G°

associated with proton transfer must be added to that of ∆G^q

for displacement reaction itself to give the overall free energy

of the displacement reaction from starting material to product.

It seems likely that this combined ∆G^qis too high to lead to

displacement, and so only exchange is observed. There may

also be a higher energetic cost for the second (ion-pair

dissociating) step in Scheme 11 if the anion is extremely basic.

Finally, in the case of some very weak acids, proton transfer

may be so costly energetically that the mechanism shifts over

to the concerted H/D exchange process illustrated in Figure 6,

where the ion pair is never formed as a true intermediate.

The above disussion ignores the question of whether, in cases

where displacement of ammonia by A^-from the acid HA is

not observed, the conversion of the amide to the Ru-A product

is exoergic. Our theoretical calculations (vide infra) suggest that

in the case of the reaction with toluene, the conversion of the

amide to the benzylruthenium complex may be thermodynami-

cally unfavored. We have tried to explore this question

experimentally by treating the potential displacement product

(DMPE)₂Ru(H)(CH₂Ph), which can be synthesized indepen-

dently, with ammonia, but this gives no reaction up to

temperatures where slow decomposition of the benzyl complex

to intractable products sets in. Therefore, there is clearly a kinetic

barrier to this transformation, but we do not know if this barrier

is high enough to account for the lack of displacement in the

“forward” direction. This ambiguity does not exist in the

attempted displacement of the ruthenium amide with ¹⁵NH₃,

which must be essentially thermoneutral - here it is clear that

the lack of displacement must be due to a kinetic rather than a

thermodynamic unfavorability.

Evidence for the assignment of 4 as CIPs was acquired from

UV-vis measurements of the reaction solutions and from

stereochemical experiments. For example, the λ_maxof the UV-

vis spectrum of fluorenide anion is highly dependent upon its

environment, the nature of the cation, the temperature of the

solution, and the solvent.²⁹Free fluorenide, defined as the fully

solvated anion, has λ_max) 375 nm in THF. In fluorenide CIPs,

the adsorption band exhibits a hypsochromic shift whose

magnitude depends on the size of the cation. For example,

lithium fluorenide CIP has λ_max) 348 nm, cesium fluorenide

CIP has λ_max) 364 nm, and tetrabutylammonium fluorenide

CIP has λ_max) 368 nm. At room temperature, lithium fluorenide

exists primarily as a mixture of CIPs and SSIPs, but upon

cooling to -50 °C, the equilibrium favors the SSIP. Large, soft

cations such as Cs⁺are not readily solvated by THF, and their

fluorenide salt exists primarily as a CIP in solution. Because

[(DMPE)₂Ru(H)(NH₃)]⁺is a large, soft cation, its fluorenide

salt should also be a CIP. This assumption is supported by its

λ_maxat 368 nm, very close to that of the tetrabutylammonium

salt.

On the basis of our evidence for a CIP, the room-temperature

equilibrium constants measured between the hydroxo complex,

fluorene, and the fluorenide salt 6a and extrapolated for the

amido complex 1, triphenylmethane, and the triphenylmethide

salt 4h were calculated using the CIP model (eqs 13 and 16). If

the extra stabilization due to ion pairing could be ignored, we

would conclude that the basicity of the hydroxo complex 2 is

similar to that of fluorenide and the basicity of amido complex

The lack of ¹⁵N incorporation upon addition of free ¹⁵NH₃to

the amido complex 1 also suggests that the RuNH₃⁺‚‚‚NH₂

-

(66) Schaad, D. R.; Landis, C. R. J. Am. Chem. Soc. 1990, 112, 1628.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4731

A R T I C L E S

Fulton et al.

1 is slightly less than that of triphenylmethide. However, this

assumption ignores the significant entropic contribution from

ion pair formation, and a more accurate estimation of the pK_a’s

of these complexes must take into account the dissociation

constants (K_dissoc) of salts 4h and 6a (eqs 15 and 16).

of the corresponding cesium salts of triphenylmethide and

fluorenide, respectively, or 10^-7to 10^-8

.

Solving eq 19 for each system using these (admittedly very

approximate) dissociation constants yields an estimated pK_aof

the ammonia complex 4 of 23 to 24 and an estimated pK_aof

the aquo complex 6 of 16 to 17 (in THF). Even though the

basicity of amido complex 1 is, at first, reminiscent of the alkali

salts of NH₂^-, the pK_aof the cationic ammonia complex is

significantly lower than that of free ammonia (41). Ammonia

+

complex 4 is much less acidic than a protonated amine (NH₄

:

10.5, DMSO, PhNH₃⁺: 3.6, DMSO),⁶⁸indicative of a more

polarized Ru-N bond as compared to a C-N (or H-N in the

case of NH₄⁺) bond. The pK_aof aquo complex 2 is similar to

that of Mayer’s protonated [(phen)₂Mn(µ-O)₂Mn(phen)₂][PF₆]₃

(pK_a) 14.6 ( 0.5, CH₃CN) and Pecoraro’s protonated [Mn-

(salpn)(µ-O)]₂complex (salpn ) N,N′-bis(salicylidene)-1,3-

propanediamine, pK_a) 13.4, CH₃CN).^48,69These are larger than

the pK_aof Stack’s Fe(MeOH) (pK_a) 9.7, methanol).⁴It is

unfortunate that so many of these measurements were made in

different solvents. A rigorous comparison of the basicity of these

complexes will have to wait until pK_astudies are carried out

under more comparable conditions.

Because the pK_aof an acid is defined as the negative log of

the equilibrium constant for the dissociation of a proton from

the acid, the pK_{a(4 or 6)}of the ammonia complex 4 and the aquo

complex 6 should be similarly defined (eqs 17 and 18).

Theoretical Calculations. As mentioned at the end of the

Results section, we have carried out several DFT calculations

to help us better understand our results. The details are provided

in the Supporting Information, but the most important results

are summarized here.

The K_a4(or the acid dissociation constant, eq 19) of ammonia

complex 4 can be calculated from the ratio of the triphenyl-

methane acid dissociation constant (K_acid) to the room-temper-

ature equilibrium constant measured for the conversion of 1

and triphenylmethane to the triphenylmethide complex 4h

(K_measured1), and the ion pair dissociation constant of complex

4h (K_dissoc4, eqs 15 and 16). Similarly, the K_a6of aquo complex

6 can be calculated from the ratio of the K_acidof fluorene to the

room-temperature equilibrium constant measured for the con-

version of 2 and fluorene to the fluorenide complex 6a

(K_measured2), and the dissociation constant of complex 6a (K_dissoc6).

We investigated the enthalpies of the reactions between amido

complex 1′⁴⁵or hydroxo complex 2′⁴⁵and six different substrates

(R-H): phenol (Scheme 4), phenylacetonitrile (Scheme 2),

cyclobutanone (Scheme 2), phenylacetylene (Scheme 2), triph-

enylmethane (eq 11), and toluene (eqs 7 and 10). Four different

reactions of complexes 1′ or 2′ with R-H were considered:

the gas-phase formation of ion pairs (GP-IP), gas-phase

formation of infinitely separated ions (ISI) from starting

complexes, formation of ISI in THF (ISI-THF), and gas-phase

formation of products. In the interest of brevity, the results will

only be briefly summarized.

The computational results mirrored information available from

experiment. For example, the enthalpy required for the formation

of gas-phase ion pairs (GP-IPs) correlated with the gas-phase

acidities of the substrate in question (Tables 6 and 7). That is,

the reaction between the metal complexes 1′⁴⁵or 2′⁴⁵and the

stronger gas-phase acids is calculated to be less endothermic

than the reactions between the metal complexes and the weaker

gas-phase acids. This trend was also followed in the formation

of infinitely separated ion pairs (ISIs) and THF-solvated ISIs.

The enthalpies of reaction for the formation of ISIs are highly

endothermic in the gas phase due to the energetic cost of charge

Streitwieser and co-workers have measured the dissociation

constants for a variety of different lithium and cesium salts,

including those of fluorenide and triphenylmethide, in the

relatively low dielectric constant solvent THF.⁶⁷These dis-

sociation constants are highly dependent upon the cation and,

to a lesser degree, the anion. For example, the dissociation

constant (K_D) for lithium fluorenide in THF at 25.0 °C is 6.90

× 10^-6M, and the analogous cesium salt has a K_Dof 1.49 ×

10^-8M, 2 orders of magnitude smaller. The K_Dfor the

triphenylmethide salts shows similar trends of 2.34 × 10^-5and

2.03 × 10^-7M for the lithium and cesium salts, respectively.

The difference between the dissociation constants of the cesium

and lithium salts is due to the increased ability of THF to solvate

the hard lithium ions relative to the soft cesium ions. As stated

earlier, our ruthenium cations 4 and 6 are predicted to be fairly

soft and will not be extensively solvated by THF. Therefore,

we estimate that the dissociation constant K_Dfor the salt

complexes 4h and 6a is on the same order of magnitude as that

separation. As expected, significant energy (∼90 kcal mol^-1

)

was gained upon formation of the GP-IPs from the ISIs.

However, only the GP-IPs for the formation of the ion pairs

containing enolate 4b′, phenylacetylide 4c′, and phenolate 4d′

(68) The authors would like to advise the reader to view these numbers as a

qualitative comparison rather than a quantitative comparison. DMSO forms

stronger hydrogen bonds with free protons than does THF, and thus the

absolute acidity scales in these two solvent systems vary. However, the

relative ordering should be the same, see: (a) Kolthoff, I. M.; Chantooni,

M. K.; Bhowmik, S. J. Am. Chem. Soc. 1968, 90, 23. (b) Streitwieser, A.;

Wang, D. Z.; Stratakis, M.; Facchetti, A.; Gareyev, R.; Abbotto, A.; Krom,

J. A.; Kilway, K. V. Can. J. Chem. 1998, 76, 765.

(67) Kaufman, M. J.; Gronert, S.; Streitwieser, A. J. Am. Chem. Soc. 1988,

110, 2829.

(69) Baldwin, M. J.; Pecoraro, V. L. J. Am. Chem. Soc. 1996, 118, 11325.

9

4732 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

Table 6. Calculated Reaction Enthalpies (in kcal mol^-1) for the

Reaction of Amido Complex 1′ with Substrates Listed (See Text)

(GPA ) Gas-Phase Acidity of Substrate (lit.); GP ) Gas Phase;

IP ) Ion Pair; ISI ) Infinitely Separated Ions; Prod. ) Products)

reactions involving 2′ than those involving 1′ (17 kcal mol^-1

in THF). In addition, the gas-phase enthalpy difference in the

displacement/product formation reactions between the reactions

involving amido complex 1′ and hydroxo complex 2′ is -9.7

kcal mol^-1. This is the enthalpy for the reaction of amido

complex 1′ with water to generate hydroxo complex 2′ and

ammonia (eq 20), which experimentally is known to lie far on

the side of 2′, in agreement with the calculations.

GP−

THF−

GP−

THF−

substrate

phenol

phenylacetonitrile 350.7^b

GPA

IP

ISI

prod.

351.6^a-22.3 67.5

-8.1 -17.4

-4.2 -9.2

9.8

13.8

7.5

65.3

88.1

92

70.9

103.8 30.2

cyclobutanone

phenylacetylene

triphenylmethane

toluene

367.1^c-6.1

-7.9

-21.1

-8.9

370.6^d1.3

358.7^e

380.6^f

3.5

^aDeTuri, V. F.; Ervin, K. M. Int. J. Mass Spectrom. 1998, 175, 123.

^bFujio, M.; McIver, R. T.; Taft, R. W. J. Am. Chem. Soc. 1981, 103, 4017.

^cBrickhouse, M. D.; Squires, R. R. J. Am. Chem. Soc. 1988, 110, 2706.

^dBartmess, J. E.; Scott, J. A.; McIver, R. T. J. Am. Chem. Soc. 1979, 101,

6047. ^eTaft, R. W.; Bordwell, F. G. Acc. Chem. Res. 1988, 21, 463.

^fGunion, R. F.; Gilles, M. K.; Polak, M. L.; Lineberger, W. C. Int. J. Mass

Spectrom. Ion Processes 1992, 117, 601.

Because the N-H bond of ammonia is 11 kcal mol^-1weaker

than the O-H bond of water, the driving force for this reaction

must be the formation of a more stable Ru-O bond as compared

to the Ru-N bond. This indicates that the Ru-O bond is

Table 7. Calculated Reaction Enthalpies (in kcal mol^-1) for the

Reaction of Hydroxo Complex 2′ with Substrates Listed (See Text)

(GPA ) Gas-Phase Acidity of Substrate (lit.); GP ) Gas Phase;

IP ) Ion Pair; ISI ) Infinitely Separated Ions; Prod. ) Products)

stronger than the Ru-N bond by approximately 20 kcal mol^-1

.

This provides the first direct comparison of bond strengths for

a late-metal parent-amido complex and its corresponding

hydroxo complex. Other studies have been performed that

compare the relative bond energies of substituted late metal

amido complexes and hydroxo complexes. In all of these studies,

the M-O bond is considerably stronger (by a magnitude similar

to that predicted by our calculations) than the M-N bond.^1,16

GP−

THF−

GP−

THF−

substrate

phenol

phenylacetonitrile

cyclobutanone

phenylacetylene

triphenylmethane

toluene

GPA

IP

ISI

prod.

351.6^a

350.7^b

367.1^c

370.6^d

358.7^e

380.6^f

83.8

81.6

104.4

108.3

87.2

9.2

13.1

27

31.2

24.8

47.5

-7.7

0.5

1.8

1.2

-11.3

120.1

13.3

Conclusions

^aDeTuri, V. F.; Ervin, K. M. Int. J. Mass Spectrom. 1998, 175, 123.

^bFujio, M.; McIver, R. T.; Taft, R. W. J. Am. Chem. Soc. 1981, 103, 4017.

^cBrickhouse, M. D.; Squires, R. R. J. Am. Chem. Soc. 1988, 110, 2706.

^dBartmess, J. E.; Scott, J. A.; McIver, R. T. J. Am. Chem. Soc. 1979, 101,

6047. ^eTaft, R. W.; Bordwell, F. G. Acc. Chem. Res. 1988, 21, 463.

^fGunion, R. F.; Gilles, M. K.; Polak, M. L.; Lineberger, W. C. Int. J. Mass

Spectrom. Ion Processes 1992, 117, 601.

Our study of the properties of the new monomeric late metal

parent amido complex trans-(DMPE)₂Ru(H)(NH₂) (1) originally

focused on its potential to abstract a hydrogen atom from a weak

C-H bond. This putative one-electron process initially appeared

plausible because, in analogy to Stack’s Fe^IIIsystem, amido

complex 1 oxidizes cyclohexadiene (CHD) to benzene.⁴How-

ever, the observed reactivity between 1 and an array of C-H

bonds suggests that the oxidation of CHD occurs via a two-

electron rather than a one-electron process. Deprotonation of

sterically encumbered weak acids (such as fluorene and triph-

enylmethane) by 1 generates stable ion pairs containing a

cationic ammonia-bound ruthenium complex. Smaller weak

acids (such as phenylacetylene, cyclobutanone, alcohols, and

amines) are also deprotonated by 1 to generate similar ion pairs;

however, the transiently generated anion in these intermediates

eventually displaces ammonia, yielding covalent ruthenium

complexes.

In addition to the reactions observed with weak acids, amido

complex 1 also promotes H/D exchange between the protons

of its NH₂group and a variety of very weak acids (such as

ammonia-d₃, toluene-d₈, and propene-d₃). We suggest that these

H/D exchange reactions proceed either via the reversible

formation of ion pair 4 or through a concerted transition state

similar to that shown in Figure 6. However, in these cases, no

displacement of the ammonia ligand is observed, perhaps

because of strong hydrogen bonding in the ion pair that is

formed. These reactions provide additional evidence against the

intervention of odd-electron processes in this system since the

H/D exchange rate correlates more strongly with the pK_aof the

carbon acid than with the bond dissociation energy of the C-H

bond in question.

were calculated. We were unable to localize the other ammonia-

based ion pairs and all of the aquo-based ion pairs; in these

cases, all attempts resulted in the proton essentially “jumping”

back onto the anion. In solution, the reactions are predicted to

be much nearer to thermoneutral.

Interestingly, the enthalpies of displacement-product forma-

tion from starting complexes involving amido complex 1′ were

calculated to be exothermic except for the reaction between 1′

and toluene to form benzyl complex 13′ and ammonia. In

contrast, with the ruthenium hydroxo complex 2′, only the

displacement reactions with phenol or phenylacetylene were

calculated to be exothermic.

The calculated reaction enthalpies for the formation of

displacement products from the starting complexes do not

correlate with either the pK_aor the BDE of the organic substrates

in question. This is in agreement with this group’s earlier

experimental results in a nickel system.¹⁶Thus, the strength of

the Ru-X bond, as compared to the R-H bond, is determined

by a combination of both electrostatic and covalent interactions,

the balance of which is dependent upon the substrate (R-H) in

question.¹⁷

In all cases and in agreement with our experimental results,

the calculated enthalpies for the reactions involving amido

complex 1′ were significantly more negative than the reactions

involving hydroxo complex 2′, confirming that 1′ is more basic

than 2′. For example, the enthalpy of ISI formation from starting

complexes in the gas phase is 16 kcal mol^-1higher for the

The relative basicity of our amido complex 1 was at first

reminiscent of that of the very strong base, sodium amide.

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4733

A R T I C L E S

Fulton et al.

dissolved in 1 mL of THF. After layering with 1 mL of pentane, the

However, the contrast between amido complex 1 and sodium

amide is marked by their differing reactivity toward benzene:

NaNH₂undergoes H/D exchange with benzene, while 1 does

not.⁵⁶We estimate the pK_aof the ammonia complex 4 to be

approximately 23 to 24; thus, 1 is a much stronger base than

are ordinary alkylamines. There is much more negative charge

on nitrogen than is observed in compounds with a C-N bond,

but less than that observed in alkali metal salts. At present, we

do not know the cause of the high basicity of 1. Two possible

explanations are the highly polar nature of the Ru-N σ-bond

and dπ-pπ repulsion between the filled d-orbitals on ruthenium

and the nitrogen lone pair orbital.^14,17Further work will be

required to distinguish these two effects.

solution was cooled to -30 °C to produce orange crystals of 4a (37.5

1

mg, 6.4 × 10^-5mol, 72%). H NMR (d₈-THF): δ 7.82 (d, 2H, J )

7.6, ArH), 7.26 (d, 2H, J ) 8.4, ArH), 6.77 (t, 2H, J ) 7.2, ArH), 6.40

(t, 2H, J ) 7.0 Hz, ArH), 5.94 (s, 1H, ArH), 1.3 (br m, 8H, DMPE),

1.2 (s, 12H, DMPE), 0.9 (s, 12H, DMPE), -0.6 (br, 3H, NH₃), -20.34

(qn, 1H, J ) 21.2, RuH). ¹³C{¹H} NMR (d₈-THF): δ 138.2 (quat),

123.4 (CH), 119.8 (CH), 119.5 (CH), 117.0 (CH), 108.8 (quat), 83.6

(CH), 31.3 (m, DMPE), 22.7 (m, DMPE), 14.2 (m, DMPE). ³¹P{¹H}

NMR (d₈-THF): δ 46.41. IR (Nujol): 3331, 3272, 3044, 3027, 1916,

1571, 1467, 1440, 1423, 1321, 1284, 1233, 1221, 1107, 982, 937, 925,

888, 840, 794, 743, 731, 711, 647 cm^-1. Anal. Calcd for RuP₄-

NC₂₅H₄₅: C, 51.36; H, 7.76; N, 2.40. Found: C, 51.53; H, 7.40; N,

2.32.

[trans-(DMPE)₂Ru(H)(NH₃)][OC₆H₅] (4d). To solution consisting

of trans-(DMPE)₂Ru(H)(NH₂) (125 mg, 2.99 × 10^-4mol) and 15 mL

of hexanes was added a solution consisting of phenol (27.8 mg, 2.95

× 10^-4mol) and 5 mL of hexanes. A powdery white precipitate formed.

The volatile materials were removed under reduced pressure, and the

white solid was stored at -30 °C. H NMR (THF-d₈, 193 K): δ 6.67

(t, 2H, J ) 7.3, ArH), 6.02 (d, 2H, J ) 7.5, ArH), 5.84 (t, 1H, J ) 6.5,

ArH), 2.53 (s, 3H, NH₃), 1.67 (s, 4H, DMPE), 1.59 (s, 12H, DMPE),

The calculations on our model systems amido complex 1′

and hydroxo complex 2′ correlated well with our experimental

results. The enthalpy of ion formation (either as ion pairs or

infinitely separated ions) from 1′ or 2′ and the weakest acids

was substantially larger than the enthalpy of ion formation from

1′ or 2′ and the stronger acids. Accordingly, the enthalpy of

ISI formation was much greater than the enthalpy of IP

formation. The calculations predict that the formation of benzyl

complex 13′ and ammonia from the amido complex 1′ and

toluene is endothermic in the gas phase. Whether this is actually

the case, or whether 13 is not formed because of the high barrier

of ion pair dissociation, is arguable, and our experimental results

do not provide enough evidence to exclude either hypothesis.

1

1.46 (s, 4H, DMPE), 1.28 (s, 12H, DMPE), -19.48 (qn, 1H, J_HP

)

20.8, Ru-H). ¹³C{¹H} NMR (THF-d₈, 203 K): δ 172.6 (quat), 129.5

(CH), 121.0 (CH), 107.8 (CH), 30.8 (m, DMPE), 23.2 (m, DMPE),

14.5 (m, DMPE). ³¹P{¹H} NMR (THF-d₈): δ 47.7. IR (Nujol): 3351,

3312, 3223, 2658, 2551, 2480, 2397, 2302, 2235, 2147, 1901, 1798,

1636, 1578, 1546, 1537, 1483, 1426, 1329, 1285, 1240, 1217, 1154,

1141, 1124, 1085, 1059, 1012, 979, 929, 890, 923, 938, 820, 795 cm^-1

.

Experimental Section

Anal. Calcd for RuP₄NOC₁₈H₄₁: C, 42.19; H, 8.06; N, 2.73. Found:

C, 42.47; H, 8.15; N, 3.02.

General Comments. All manipulations were carried out under inert

[trans-(DMPE)₂Ru(H)(NH₃)][OC₆H₄Me] (4e). A vial was charged

with trans-(DMPE)₂Ru(H)(NH₂) (217 mg, 5.19 × 10^-3mol) and 0.5

mL of THF. To it was added a solution consisting of p-cresol (55.3

mg, 5.11 × 10^-4mol) and 0.5 mL of THF. The resulting solution was

layered with 2 mL of pentane and placed into a -30 °C freezer for 1

month to give pale-yellow crystals of 4e in 59% yield (159 mg, 3.01

× 10^-3mol). ¹H NMR (THF-d₈, 203 K): δ 6.50 (d, 2H, J ) 7.5, ArH),

5.93 (d, 2H, J ) 7.5, ArH), 2.55 (s, 3H, NH₃), 1.68 (s, 4H, DMPE),

1.43 (s, 12H, DMPE), 1.37 (s, 4H, DMPE), 1.28 (s, 12H, DMPE),

-19.45 (s, 1H, Ru-H). ¹³C{¹H} NMR (THF-d₈, 223 K): δ 170.5 (quat),

130.1 (CH), 120.3 (CH), 114.8 (quat), 31.1 (m, DMPE), 23.4 (m,

DMPE), 21.3 (CH₃), 14.5 (DMPE). ¹³P{¹H} NMR (THF-d₈, 298 K):

δ 47.7. IR (Nujol): 3357, 3295, 3282, 3184, 3042, 2854, 2812, 2724,

2640, 2545, 2853, 2317, 2263, 2149, 2070, 1982, 1878, 1818, 1734,

1650, 1597, 1536, 1494, 1425, 1322, 1302, 1289, 1231, 1158, 1121,

1095, 1085, 990, 930, 913, 887, 838, 825, 797 cm^-1. Anal. Calcd for

RuP₄NOC₁₉H₄₃: C, 43.34; H, 8.23; N, 2.66. Found: C, 43.70; H, 8.33;

N, 2.58.

atmosphere or by using standard Schlenk or vacuum line techniques

1

unless noted otherwise. The H and ¹³C NMR spectra were obtained

on Bruker 300 and 400 MHz Fourier transform spectrometers with

commercial Bruker AMX series interfaces or a Bruker 500 MHz Fourier

transform spectrometer with a commercial Bruker DRX series interface.

The latter instrument was used in the equilibrium experiments, and

the thermocouple was calibrated with either neat methanol or neat

ethylene glycol prior to each experiment.

trans-(DMPE)₂Ru(H)(NH₂) (1), trans-(DMPE)₂Ru(H)(OH) (2), and

(DMPE)₂Ru(H)₂(3) were prepared by literature methods.^19,30,61,70All

other reagents were purchased from commercial vendors, checked for

purity, and used without further purification unless otherwise noted.

Fluorene and triphenylmethane were purified via sublimation. Liquids

were degassed using three freeze-pump-thaw cycles and dried over

4 Å activated molecular sieves or passed through an activated alumina

column. Solids were stored in an inert atmosphere box. Toluene,

benzene, pentane, and hexanes (UV grade, alkene free) were passed

through a column of activated alumina (A1, 12 × 32, Purifry) under

nitrogen pressure and sparged with nitrogen prior to use.⁷¹Diethyl ether

and tetrahydrofuran were distilled from purple sodium/benzophenone

ketyl under nitrogen. Benzene-d₆and toluene-d₈were degassed by three

freeze-pump-thaw cycles, placed over 4 Å activated molecular sieves,

and filtered through glass-fiber filter paper prior to use. THF-d₈was

vacuum transferred from sodium/benzophenone ketyl and stored under

inert atmosphere.

Generation, Spectroscopic Characterization, and Isolation in

Crude Form of Ion Pair Salts 4b, 4c, 4f, and 4g: General

Procedure. A vial was charged with trans-(DMPE)₂Ru(H)(NH₂) (107

mg, 2.54 × 10^-4mol) and 8 mL of hexanes. To it was added a solution

consisting of phenylacetylene (25.9 mg, 2.54 × 10^-4mol) and 1 mL

of hexanes. A white powder precipitated from solution, and the volatile

material was removed in vacuo. Pure material could not be obtained

because some final covalent product formed in each case during this

procedure, as ascertained by NMR spectroscopy.

[trans-(DMPE)₂Ru(H)(NH₃)][C₄H₅O] (4b). ¹H NMR (THF-d₈, 222

K): δ 3.38 (s, 1H, CHC(O)), 2.38 (s, 3H, NH₃), 2.13 (s, 2H, C(O)-

CH₂), 1.97 (br m, 4H, DMPE), 1.71 (s, 2H, CH₂CHC(O)CH₂), 1.54 (s,

12H, DMPE), 1.46 (br m, 4, DMPE), 1.33 (s, 12H, DMPE), -19.4

(qn, 1, J ) 25, RuH). ¹³C{¹H} NMR (THF-d₈, 222 K): 166.48 (CO),

82.40 (CH), 39.35 (CH₂), 31.16 (qn, J_C-P) 51, DMPE), 23.45 (m,

DMPE), 19.50 (CH₂), 14.96 (m, DMPE). ³¹P{¹H} NMR (d₈-THF, 25

[trans-(DMPE)₂Ru(H)(NH₃)][C₁₃H₉] (4a). A vial was charged with

trans-(DMPE)₂Ru(H)(NH₂) (37.5 mg, 9.0 × 10^-5mol) and 10 mL of

pentane. To this was added a solution consisting of fluorene (14.9 mg,

9.0 × 10^-5mol) and 2 mL of pentane. A yellow-orange precipitate

formed immediately. The volatile material was removed in vacuo, and

the product was washed with 5 mL of pentane. The residue was

(70) Field, L. D.; Hambley, T. W.; Yau, B. C. K. Inorg. Chem. 1994, 33, 2010.

(71) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem. Educ.

2001, 78, 64.

9

4734 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

°C): δ 40.38. IR (Nujol): 3353, 3296, 1872, 1643, 1581, 1458, 1425,

1378, 1284, 1159, 1121, 1078, 981, 912, 889, 841, 794, 729, 707, 644.

[trans-(DMPE)₂Ru(H)(NH₃)][CCPh] (4c). ¹H NMR (THF-d₈, 222

K): δ 7.07 (d, 1H, J ) 7.5, ArH), 7.06 (t, 1H, J ) 7.3 ArH), 7.04 (d,

1H, J ) 7.5 ArH), 7.03 (t, 1H, J ) 7.5 ArH), 6.94 (t, 1H, J ) 8.0

ArH), 2.50 (s, 3H, NH₃), 2.36 (br s, 4H, DMPE), 1.68 (s, 12H, DMPE),

1.48 (br s, 4H, DMPE), 1.36 (s, 12H, DMPE), -19.19 (qn, 1H, J )

20.8, RuH). ¹³C{¹H}NMR (THF-d₈, 222 K): δ 133.20 (Ru-CC),

131.42 (CH), 128.27 (CH), 123.71 (CH), 109.36 (ipso-C), 31.67 (m,

DMPE), 23.30 (DMPE), 16.78 (DMPE). ³¹P{¹H}NMR (THF-d₈, 25

°C): δ 48.5. IR (Nujol): 3361, 3256, 3068, 2347, 2236, 2140, 1994,

1920, 1885, 1636, 1561, 1589, 1479, 1424, 1299, 1284, 1266, 1234,

12H), 1.51 (m, 4H), 1.29 (m, 4H), 1.24 (s, 12H), -11.95 (qn, 1H, J )

20.0). ¹³P{¹H} NMR (C₆H₆): δ 45.5 (s).

trans-(DMPE)₂Ru(H)(CCMe) (8) (Method A). Allene (7.99 × 10^-4

mol) was condensed onto a frozen solution consisting of trans-

(DMPE)₂Ru(H)(NH₂) (333 mg, 7.97 × 10^-4mol) and 30 mL of toluene.

The solution was warmed to room temperature and allowed to stir for

3 d, after which time the volatile materials were removed in vacuo.

The residue was dissolved in a minimal amount of diethyl ether (2

mL). The resulting solution was filtered through glass fiber and cooled

1

to -30 °C for 3 d to give yellow needles of 8 (257 mg, 73%). H

NMR (C₆D₆): δ 2.16 (s, 3H, CCH₃), 1.56 (m, 4H, DMPE), 1.54 (s,

12H, DMPE), 1.32 (m, 4H, DMPE), 1.27 (s, 12H, DMPE), -12.46

(qn, 1H, J ) 21, RuH). ¹³C NMR (C₆D₆): δ 107.67 (qn, J ) 45.0,

RuCC), 97.00 (RuCC), 32.38 (qn, J ) 55.4, DMPE), 24.61 (qn, J )

28.3, DMPE), 17.92 (qn, J ) 25.8, DMPE), 7.84 (CCH₃). ³¹P NMR

(C₆D₆): δ 46.32 (d, J ) 10). IR (Nujol): 2715, 2395, 2348, 2079,

1731, 1420, 1291, 1275, 1232, 1119, 1073, 991, 935, 888, 839, 794

cm^-1. Anal. Calcd for RuP₄C₁₅H₃₆: C, 40.82; H, 8.22. Found: C, 41.15;

H, 8.26.

1183, 1168, 1121, 1083, 1065, 1025, 932, 914, 889, 842, 798 cm^-1

.

1

[trans-(DMPE)₂Ru(H)(NH₃)][MeO] (4f). H NMR (THF-d₈, 193

K): 3.73 (s, 3H, CH₃O), 3.31 (s, 3H, NH₃), 2.11 (s, 4H, DMPE), 1.63

(s, 12H, DMPE), 1.45 (s, 4H, DMPE), 1.33 (s, 12H, DMPE). ¹³C{¹H}

NMR (THF-d₈, 223 K): δ 57.5 (CH₃O), 31.6 (m, DMPE), 32.6 (m,

DMPE), 14.1 (m, DMPE). ³¹P{¹H} NMR (THF-d₈): δ 45.2. IR

(Nujol): 2653, 2518, 1869, 1662, 1423, 1320, 1291, 1234, 1085, 934,

912, 888, 841, 798 cm^-1

.

trans-(DMPE)₂Ru(H)(CCMe) (8) (Method B). Propyne (7.79 ×

10^-4mol) was condensed onto a frozen solution consisting of trans-

(DMPE)₂Ru(H)(NH₂) (82.5 mg, 1.97 × 10^-4mol) and 10 mL of

toluene. The solution was warmed to room temperature and allowed

to stir for 3 d, after which time the volatile materials were removed

under reduced pressure. The tan residue was dissolved in 1 mL of

toluene, and the resulting solution was cooled to -30 °C for 1 d to

give yellow crystals of 7 in 64% yield (55.9 mg, 1.27 × 10^-4mol).

The spectroscopic data for this material were identical to those obtained

from Method A.

[trans-(DMPE)₂Ru(H)(NH₃)][NHC₆H₅] (4g). ¹H NMR (C₆D₆, 213

K): δ 6.42 (t, 2H, J ) 7.5, ArH), 5.81 (d, 2H, J ) 7.5, ArH), 5.43 (t,

1H, J ) 6.8, ArH), 2.92 (s, 3H, NH₃), 1.84 (m, 4H, DMPE), 1.40 (m,

4H, DMPE), 1.35 (s, 12H, DMPE), 1.29 (s, 12H, DMPE), -19.34 (qn,

1H, J ) 21.1). ¹³C{¹H} NMR (THF-d₈, 223 K): δ 165.3 (quat), 129.4

(CH), 115.7 (CH), 103.8 (CH), 31.2 (m, DMPE), 25.2 (m, DMPE),

15.5 (m, DMPE). ³¹P{¹H} NMR (THF-d₈): δ 44.8. IR (Nujol): 3235,

3182, 1902, 1596, 1487, 1442, 1422, 1300, 1165, 1091, 1079, 990,

934, 919, 889, 837, 813 cm^-1

.

[trans-(DMPE)₂Ru(H)(NH₃)][Ph₃C] (4h). A vial was charged with

trans-(DMPE)₂Ru(H)(NH₂) (21.5 mg, 5.15 × 10^-5mol), triphenyl-

methane (12.5 mg, 5.15 × 10^-5mol), and 0.5 mL of THF-d₈. Complex

4h is observed in solution in equilibrium with 1 and triphenylmethane

(see text); therefore, we were unable to cleanly isolate 4h for full

characterization and could only obtain low-temperature NMR evidence

trans-(DMPE)₂Ru(H)(NCC(H)Ph) (9a). A vial was charged with

trans-(DMPE)₂Ru(H)(NH₂) (68.5 mg, 1.64 × 10^-4mol) and 5 mL of

toluene. To it was added a solution consisting of phenylacetonitrile

(19.7 mg, 1.68 × 10^-4mol) and 5 mL of toluene. The solution turned

yellow after 2 d at 25 °C. The volatile materials were removed under

reduced pressure. The yellow residue was dissolved into 4 mL of THF,

and the resulting solution was layered with 8 mL of pentane and cooled

to -30 °C for 2 d. Yellow crystals were isolated in 79% yield (69.2

1

for its existence. H NMR (d₈-THF, 202 K): δ 7.35 (d, 6H, J ) 8.0,

ArH), 6.50 (t, 6H, J ) 7.5, ArH), 5.94 (t, 3H, J ) 7, ArH), 1.65 (br m,

4H, DMPE), 1.54 (br m, 4H, DMPE), 1.36 (s, 12H, DMPE), 1.29 (s,

12H, DMPE), 0.08 (br s, 3H, NH₃), -19.0 (qn, 1H, J ) 20H, RuH).

¹³C{¹H} NMR (d₈-THF, 202 K): δ 150.0 (quat), 128.4 (CH), 124.2

(CH), 113.3 (CH), 91.4 (quat), 31.3 (m, DMPE), 22.9 (m, DMPE),

14.0 (m, DMPE). ³¹P{¹H} NMR (d₈-THF, 202 K): δ 45.1.

[trans-(DMPE)₂Ru(H)(OH₂)][C₁₃H₉] (6a). A vial was charged with

trans-(DMPE)₂Ru(H)(OH) (20.1 mg, 4.8 × 10^-5mol), fluorene (8.0

mg, 4.8 × 10^-5mol), and 0.5 mL of THF-d₈. Complex 6a is observed

in solution in equilibrium with 2 and triphenylmethane (see text);

therefore, we were unable to cleanly isolate 6a for full characterization

and could only obtain low-temperature NMR evidence for its existence.

¹H NMR (d₈-THF, 203 K): δ 7.87 (d, 2H, J ) 7.5, ArH), 7.31 (d, 2H,

J ) 7.2, ArH), 6.82 (t, 2H, J ) 7.8, ArH), 6.36 (t, 2H, J ) 6.9, ArH),

5.97 (s, 1H, ArH), 1.62 (br m, 4H, DMPE), 1.46 (br m, H, DMPE),

1.30 (br, 24H, DMPE), -23.4 (br, 1H, RuH). ¹³C{¹H} NMR (d₈-THF,

203 K): δ 150.1 (quat), 145.1 (quat), 128.4 (CH), 127.2 (CH), 124.3

(CH), 113.3 (CH), 91.4 (CH), 31.5 (DMPE), 32.2 (m, DMPE), 14.0

(m, DMPE). ³¹P{¹H} NMR (d₈-THF, 203 K): δ 45.3.

trans-(DMPE)₂Ru(H)(CCPh) (7). A vial was charged with 80.8

mg (1.93 × 10^-4mol) of trans-(DMPE)₂Ru(H)(NH₂) and 2 mL of THF.

To it was added a solution consisting of phenylacetylene (19.5 mg,

1.91 × 10^-4mol) and 1 mL of THF. The volatile materials were

removed under reduced pressure after 8 h at 25 °C. The remaining

residue was dissolved in 1 mL of toluene and cooled to -30 °C to

give peach crystals in 71% yield (68.2 mg, 1.34 × 10^-4mol). ¹H NMR

(C₆D₆): δ 7.47 (d, 2H, J ) 7), 7.16 (t, 2H, J ) 7.8), 6.95 (t, 1H, J )

7.5), 1.50 (s, 12H), 1.49 (m, 4H), 1.29 (m, 4H), 1.24 (s, 12H), -11.95

(qn, 1H, J ) 21.5). ³¹P{¹H} NMR: δ 45.6 (s). lit.^{30 1}H NMR (C₆D₆):

δ 7.47 (d, 2H, J ) 6.9), 7.18 (s, 2H), 6.95 (t, 2H, J ) 7.4), 1.50 (s,

1

mg, 1.30 × 10^-4mol). H NMR (d₈-THF): δ 6.56 (t, 2H, J ) 7.5,

ArH), 6.34 (d, 2H, J ) 7.5, ArH), 5.90 (t, 1H, J ) 7.5, ArH), 2.60 (s,

1H, CHPh), 1.71 (br m, 4H, DMPE), 1.62 (br m, 4H, DMPE), 1.45 (s,

12H, DMPE), 1.43 (s, 12H, DMPE), -18.9 (qn, 1H, J ) 21, RuH).

¹³C{¹H} NMR (C₆D₆, room temperature): δ 150.45 (quat C), 144.65

(quat C), 129.15 (CH), 118.81 (CH), 113.22 (CH), 36.77 (NCCHPh),

31.42 (m, DMPE), 22.24 (DMPE), 15.78 (DMPE). ³¹P{¹H} NMR (d₈-

THF): 48.15. IR (Nujol): 2120, 1922, 1580, 1547, 1485, 1461, 1421,

1374, 1323, 1289, 1235, 1198, 1172, 1150, 1125, 1093, 1074, 983,

928, 888, 840, 793, 731, 703, 692, 643 cm^-1. Anal. Calcd for RuP₄-

NC₂₀H₃₉: C, 46.33; H, 7.58; N, 2.70. Found: C, 46.53; H, 7.90; N,

2.61.

trans-(DMPE)₂Ru(H)(NCC(H)C₆H₄CF₃) (9b). A vial was charged

with trans-(DMPE)₂Ru(H)(NH₂) (76.6 mg, 1.83 × 10^-4mol) and 5

mL of toluene. To it was added a solution consisting of 4-(R,R,R-

trifluoromethyl)phenylacetonitrile) (34.2 mg, 1.85 × 10^-4mol) and 5

mL of toluene. After 1 d, the volatile materials were removed in vacuo.

The purple residue was dissolved in 10 mL of THF and cooled to -30

°C for 2 d. Light purple/green crystals were isolated in 74% yield (81.8

mg, 1.35 × 10^-4mol). ¹H NMR (d₈-THF, 203 K): δ 6.80 (d, 1H, J )

8.7, ArH), 6.72 (d, 1H, J ) 9.2, ArH), 6.42 (d, 1H, J ) 8.7, ArH),

6.15 (d, 1H, J ) 8.7, ArH), 3.01 (s, 1H, NCCHAr), 1.69 (br m, 8H,

DMPE), 1.44 (s, 24H, DMPE), -18.67 (qn, 1H, J ) 21.1, RuH). ¹³C-

{¹H} NMR (d₈-THF, room temperature): δ 152.5 (quat), 136.6 (quat),

127.9 (CH), 127.1 (CH), 124.2 (CH), 123.8 (CH), 38.47 (NCCHAryl),

29.86 (m), 20.28 (m), 13.86. ³¹P{¹H} NMR (d₈-THF, room tempera-

ture): δ 47.67. IR (Nujol): 2134, 1943, 1594, 1531, 1512, 1461, 1420,

1393, 1379, 1308, 1234, 1214, 1177, 1143, 1100, 1067, 1049, 936,

912, 890, 841, 815, 796, 733, 705, 646 cm^-1. Anal. Calcd for RuP₄-

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4735

A R T I C L E S

Fulton et al.

NC₂₁F₃H₃₈: C, 43.01; H, 6.53; N, 2.39. Found: C, 43.35; H, 6.89; N,

2.69.

(4.10 mL, 4.10 × 10^-4mol) via a syringe. After 3 h at 25 °C, the

volatile materials were removed under reduced pressure. Pentane (2

mL) was added and evaporated to remove residual THF, after which

time the crude product was extracted with THF (10 mL), and the

resulting solution was filtered through glass fiber. Recrystallization from

pentane (5 mL) resulted in yellow crystals of 14 (69.9 mg, 1.42 ×

trans-(DMPE)₂Ru(H)(C₄H₅O) (10). A vial was charged with 79.3

mg (1.90 × 10^-4mol) of trans-(DMPE)₂Ru(H)(NH₂) and 5 mL of

toluene. To it was added a solution consisting of cyclobutanone (13.8

mg, 1.97 × 10^-4mol) and 5 mL of toluene. After 1 d, the volatile

materials were removed in vacuo, and the residue was dissolved in 3

mL of ether. The solution was cooled to -30 °C for 3 d to give orange

crystals in 59% yield (52.8 mg, 1.12 × 10^-4mol). ¹H NMR (THF-d₈,

room temperature): δ 2.83 (br m, 1H), 2.76 (br m, 1H), 2.29 (br m,

1H), 1.87 (br m, 1H), 1.79 (br m, 1H), 1.46 (s, 24H, DMPE), 1.35 (s,

8H, DMPE). ¹³C{¹H} NMR (THF-d₈, room temperature): δ 212.67

(CO), 41.84 (CH₂C(O)), 40.47 (qn, J_CP) 18.5, RuCH), 31.73 (br m,

DMPE), 26.75 (br m, DMPE), 23.13 (RuCHCH₂C(O)), 15.77 (br m,

DMPE). ¹³P{¹H} NMR (THF-d₈, room temperature): δ 39.77 (m), 38.73

(m). IR (Nujol): 1832, 1636, 1523, 1458, 1421, 1377, 1292, 1276,

1

10^-4mol, 35%). H NMR (THF-d₈): δ 6.69 (d, 2H, J ) 7.6, ArH),

6.64 (t, 2H, J ) 6.8, ArH), 6.35 (t, 1H, J ) 6.8, ArH), 1.6 (br, 2H,

RuCH₂Ph), 1.5 (br m, 4H, DMPE), 1.4 (br m, 4H, DMPE), 1.4 (m,

12H, DMPE), 1.2 (m, 12H, DMPE), -15.40 (qn, 1H, J ) 22.5, RuH).

¹³C{¹H} NMR (C₆D₆): δ 165.54 (quat), 128.77 (CH), 127.88 (CH),

125.90 (CH), 31.67 (DMPE), 27.03 (DMPE), 15.42 (DMPE), 15.76

(qn, J_CP) 20, Ru-CH₂Ph). ³¹P{¹H} NMR (THF-d₈): δ 48.86. IR

(Nujol): 1786, 1586, 1565, 1469, 1416, 1378, 1291, 1274, 1231, 1210,

1169, 1119, 1073, 1042, 1023, 1010, 973, 925, 885, 835, 790, 751,

719, 694, 644, 633 cm^-1. Anal. Calcd for C₁₉H₄₀P₄Ru: C, 46.24; H,

8.17. Found: C, 45.90; H, 8.27.

1250, 1234, 1146, 1120, 1081, 1010, 936, 888, 839, 796, 721 cm^-1

.

Anal. Calcd for RuP₄OC₁₆H₃₈: C, 40.76; H, 8.12. Found: C, 41.00;

H, 8.42.

Oxidation of 1,4-Cyclohexadiene or Anthracene with Amido

Complex 1 or Hydroxo Complex 2: General Procedure. In a typical

experiment, 0.5 mL of a THF-d₈solution was prepared containing trans-

(DMPE)₂Ru(H)(NH₂) (10 mg, 2.4 × 10^-5mol), 9,10-dihydroanthracene

(4.4 mg, 2.4 × 10^-5mol), and 1,3,5-trimethoxybenzene (internal

standard, 3.4 mg, 2.0 × 10^-5mol). The reactions were monitored by

¹H and ³¹P{¹H} NMR spectroscopy. The identities of (DMPE)₂Ru(H)₂

and anthracene were ascertained by comparison of their NMR spectra

with those of authentic samples. Anthracene was also identified by

comparison of its GC retention time with that of an authentic sample.

H/D Exchange Experiments: Typical Procedure. An NMR tube

was charged with trans-(DMPE)₂Ru(H)(NH₂) (12 mg, 2.9 × 10^-5mol),

triphenylmethane (6.5 mg, 2.7 × 10^-5mol), 1,3,5-trimethoxybenzene

(internal standard, 4.2 mg, 2.5 × 10^-5mol), and 0.5 mL of toluene-d₈.

After 4 h, a one-pulse ¹H NMR spectrum was acquired, and the proton

resonances of triphenylmethane and toluene-d₈were integrated against

the proton resonances of the internal standard 1,3,5-trimethoxybenzene.

The volatile materials were removed in vacuo, and the residue was

trans-(DMPE)₂Ru(H)(OC₆H₅) (11b). A vial was charged with trans-

(DMPE)₂Ru(H)(NH₂) (52.6 mg, 1.26 × 10^-4) and 10 mL of THF. To

it was added a solution consisting of phenol (12.7 mg, 1.17 × 10^-4

mol) and 5 mL of THF. After allowing the resulting solution to stand

for 1 h at 25 °C, the volatile materials were slowly removed (4 h)

under reduced pressure. The white residue was dissolved in a minimal

amount of toluene (0.5 mL), and 2 mL of pentane was layered onto

this solution. The resulting solution was placed at -30 °C for 2 d to

1

produce white crystals of 11b (49%, 30.5 mg, 6.17 × 10^-5mol). H

NMR (C₆D₆): δ 7.34 (t, 2H, J ) 7.4), 6.61 (t, 1H, J ) 6.8), 6.42 (d,

2H, J ) 6.3), 1.64 (m, 4H), 1.30 (s, 12H), 1.24 (m, 4H), 1.09 (s, 12H),

-23.30 (qn, 1H, J ) 22.0). ³¹P{¹H} NMR (C₆D₆): δ 45.4. lit.^{30 1}H

NMR (C₆D₆): δ 7.35 (t, 2H, J ) 7.8), 6.62 (t, 1H, J ) 6.8), 6.43 (d,

2H, J ) 7.6), 1.65 (m, 4H), 1.30 (s, 12H), 1.23 (m, 4H), 1.09 (s, 12H),

-23.3 (qn, 1H, J ) 21.8). ³¹P{¹H} NMR (C₆D₆): δ 45.4.

trans-(DMPE)₂Ru(H)(OC₆H₄Me) (11c). trans-(DMPE)₂Ru(H)-

(NH₂) (57.3 mg, 1.37 × 10^-4mol) was dissolved in 10 mL of THF.

To this was added a solution consisting p-cresol (12.7 mg, 1.35 × 10^-4

mol) and 5 mL of THF. The resulting solution was allowed to stand

for 1 h at 25 °C, after which the volatile materials were slowly removed

(4 h) under reduced pressure. The white residue was dissolved in a

minimal amount of toluene (0.5 mL), and the resulting solution was

layered with 2 mL of pentane and cooled to -30 °C for 2 d to give

white crystals of 11c (57%, 39.2 mg, 7.70 × 10^-5mol). ¹H NMR

(C₆D₆): δ 7.11 (d, 2H, J ) 8.0), 6.36 (d, 2H, J ) 7.9), 2.41 (s, 3H),

1.66 (m, 4H), 1.32 (s, 12H), 1.25 (m, 4H), 1.10 (s, 12H), -23.19 (qn,

1H, J ) 20.0). ³¹P{¹H} NMR (C₆D₆): δ 45.6. lit.^{37 1}H NMR (C₆D₆):

δ 7.11 (d, 2H, J ) 8.1), 6.38 (d, 2H, J ) 7.7), 2.40 (s, 3H), 1.66 (m,

4H), 1.32 (s, 12H), 1.26 (m, 4H), 1.00 (s, 12H), -23.20 (qn, 1H, J )

21.8). ³¹P{¹H} NMR (C₆D₆): δ 45.5.

trans-(DMPE)₂Ru(H)(NHC₆H₅) (12). trans-(DMPE)₂Ru(H)(NH₂)

(134 mg, 3.21 × 10^-4mol) was dissolved in 3 mL of THF. To it was

added a solution consisting of aniline (29.8 mg, 3.20 × 10^-4mol) and

1 mL of THF. The solution was allowed to stand for 1.5 d, after which

the volatile materials were removed in vacuo. The yellow residue was

dissolved in a minimal amount of toluene (1 mL). After layering with

2 mL of pentane, the solution was cooled to -30 °C for 2 d to give

pale-yellow crystals of 11 in 57% yield (89.9 mg, 1.82 × 10^-4mol).

¹H NMR (C₆D₆): δ 7.20 (t, 2H, J ) 6.9), 6.35 (t, 1H, J ) 6.8), 6.22

(d, 2H, J ) 7.1), 1.43 (m, 4H), 1.27 (s, 12H), 1.19 (m, 4H), 1.11 (s,

12H), -19.13 (qn, 1H, J ) 23.0). ³¹P{¹H} NMR (C₆D₆): δ 44.6. lit.³⁷

¹H NMR (C₆D₆): δ 7.22 (t, 2H, J ) 7.4), 6.38 (m, 1H), 6.24 (d, 2H,

J ) 7.8), 1.42 (m, 4H), 1.27 (s, 12H), 1.16 (m, 4H), 1.10 (s, 12H),

-19.13 (qn, 1H, J ) 22.9). ³¹P{¹H} NMR (C₆D₆): δ 44.5.

2

dissolved in 0.5 mL of benzene (protiated). H NMR (C₆H₆): δ 5.41

(s, Ph₃C-H), 1.52 (br s, DMPE), 1.34 (br s, DMPE), 1.29 (s, DMPE).

Equilibrium Studies of the Reaction between trans-(DMPE)₂Ru-

(H)(OH) and Fluorene: General Procedure. In a typical experiment,

a stock solution was made by charging a 2.00 mL volumetric flask

with trans-(DMPE)₂Ru(H)(OH) (39.0 mg, 9.30 × 10^-5mol, 4.65 ×

10^-2M), fluorene (11.6 mg, 6.90 × 10^-5mol, 3.45 × 10^-2M), 1,3,5-

trimethoxybenzene (internal standard, 10.4 mg, 6.18 × 10^-5mol, 3.09

× 10^-2M), and freshly distilled THF-d₈until the volume totaled 2.00

mL. The solution was stored in a Teflon-sealed vial at -30 °C. A 0.5

mL aliquot was transferred to an NMR tube with a Teflon-seal and

inserted into an NMR probe that had been precalibrated with a neat

ethylene glycol sample to 25.0 °C. After 15 min, a 16-pulse spectrum

using a 60 s delay to allow for complete proton relaxation was acquired.

Equilibrium Studies of the Reaction between trans-(DMPE)₂Ru-

(H)(NH₂) and Triphenylmethane: General Procedure. In a typical

experiment, a stock solution was made by charging a 5.00 mL

volumetric flask with trans-(DMPE)₂Ru(H)(NH₂) (32.3 mg, 7.73 ×

10^-5mol, 1.55 × 10^-2M), 1,3,5-trimethoxybenzene (internal standard,

15.0 mg, 8.92 × 10^-5mol, 1.78 × 10^-2M), and freshly distilled THF-

d₈until the volume totaled 5.00 mL (stock solution 1). The solution

was stored in a Teflon-sealed vial at -30 °C. A 1.00 mL volumetric

flask was charged with triphenylmethane (13.1 mg, 5.36 × 10^-5mol,

5.36 × 10^-2M) and enough stock solution to bring the total volume to

1.00 mL. A 0.5 mL aliquot was transferred to an NMR tube with a

Teflon-seal. The tube was inserted into an NMR probe, precooled to

250.2 K. After 15 min, a one-pulse spectrum was acquired. Spectra

were acquired at four more temperatures after allowing for a 15 min

equilibration to the desired temperature. ¹H NMR (d₈-THF, 202 K): δ

7.35 (d, 6H, J ) 8.0, ArH), 6.50 (t, 6H, J ) 7.5, ArH), 5.94 (t, 3H, J

) 7, ArH), 1.65 (br m, 4H, DMPE), 1.54 (br m, 4H, DMPE), 1.36 (s,

trans-(DMPE)₂Ru(H)(CH₂Ph) (13). A vial was charged with trans-

(DMPE)₂Ru(H)(Cl) (177.7 mg, 4.06 × 10^-4mol) and 10 mL of THF.

To it was added a 1.0 M hexanes solution of benzylmagnesium chloride

9

4736 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Chemistry of a Parent Amidoruthenium Complex

A R T I C L E S

12H, DMPE), 1.29 (s, 12H, DMPE), 0.08 (br s, 3H, NH₃), -19.0 (qn,

1H, J ) 20H, RuH). ¹³C{¹H} NMR (d₈-THF, 202 K): δ 150.0 (quat),

128.4 (CH), 124.2 (CH), 113.3 (CH), 91.4 (quat), 31.3 (m, DMPE),

22.9 (m, DMPE), 14.0 (m, DMPE). ³¹P {¹H} NMR (d₈-THF, 202 K):

δ 45.1.

fixed B_iso(B_iso) 3.32). Their positions were refined, and their isotropic

thermal parameters were held fixed.

For 9b. Golden platelike crystals were obtained by the slow cooling

of a THF solution of 9b to -30 °C. On the basis of the systematic

absences of 0k1: k + 1 * 2n and hk0: h * 2n, packing considerations,

a statistical analysis of intensity distribution, and the successful solution

and refinement of the structure, the space group was determined to be

Pnma (No. 62). Attempts to refine the carbon atoms anisotropically

led to unreasonable values for the components of the anisotropic

displacement parameters and a low data-to-parameter ratio. The hydride

(H1) was located in the difference electron density map. The hydrogen

atom coordinates were refined isotropically.

For 10. Colorless blocklike crystals were grown by slowly cooling

a concentrated diethyl ether solution of 10 to -30 °C. On the basis of

a statistical analysis of intensity distribution, and the successful solution

and refinement of the structure, the space group was determined to be

P1 (No. 2). The hydride ligands and the hydrogen atoms of the R

carbons C13 and C29 were located in the difference electron density

map and their positions refined with fixed thermal parameters.

X-ray Structure Determinations of 4a, 4e, 9b, and 10: General

Procedure. Crystals were mounted on a quartz fiber using Paratone N

hydrocarbon oil and were cooled by a nitrogen-flow low-temperature

apparatus. Crystal quality was evaluated via measurements of intensities

and inspection of peak scans. All measurements were made on a

Siemens SMART (Siemens Industrial Automation, Inc.) diffractometer

with a CCD area detector using graphite monochromated Mo KR

radiation and an exposure time of 10 s/frame. The raw data were

integrated by the program SAINT (SAX Area-Detector Integration

Program) and were corrected for Lorentz and polarization effects. Data

were analyzed for agreement and possible adsorption using XPREP.⁷²

Empirical adsorption corrections were based on comparisons of

redundant and equivalent reflections as applied using XPREP or

SADABS.⁷³The structures were solved by direct methods and expanded

using Fourier techniques. Except as noted, all non-hydrogen atoms were

refined anisotropically, and hydrogen atoms were included as fixed

contributions but not refined. The function minimized in the full-matrix

least-squares refinement was ∑w(|F_o| - |F_c|)². The weighting scheme

was based on counting statistics and included a factor to downweight

the intense reflections.

Acknowledgment. The authors are grateful for financial

support from the National Science Foundation (grant no. CHE-

0094349). M.W.B. acknowledges the Dow Chemical Co. for

support. We would like to thank Dr. F. J. Hollander, Dr. Dana

Caulder, and Dr. Allen Oliver for determining the solid-state

structures of 4a, 9, 10, and 4e. We are very grateful to Prof.

Kenneth Wiberg and the Yale chemistry department for allowing

S.S. to perform calculations on their computers (NIH); the work

in the Wiberg laboratory was supported by NIH grant GM-

53708. We would also like to thank Prof. Andrew Streitwieser

for helpful discussions on acidity questions.

For 4a. Orange platelike crystals were obtained by slowly cooling

a solution of 4a in THF and pentane to -30 °C. The systematic absences

of h01: h * 2n and 0k0 * 2n uniquely determined the space group to

be P2₁/a (No. 14). The hydrides (H71 and H72) were located in the

difference electron density map. Their positions were refined, but their

isotropic thermal parameters were held fixed.

For 4e. Colorless crystals were obtained by the slow diffusion of

pentane into a concentrated THF solution of 4e at -30 °C. The

systematic absences of h01: h + 1 * 2n and 0k0: k * 2n uniquely

determined the space group to be P2₁/n (No. 14). The hydride and

ammonia hydrogens on both ruthenium complexes were located from

a difference Fourier map and included with refined coordinates and a

Supporting Information Available: Tables giving details of

the data collection, refinement, atomic coordinates, thermal

parameters, bond lengths, and bond angles for complexes 4a,

9, 10, and 4e. In addition, the experimental used in the

calculations along with tables of computational data including

the absolute energies, ZPEs, and Cartesian coordinates for all

calculated species (PDF). This material is available free of

charge via the Internet at http://pubs.acs.org.

(72) XPREP, version 5.03; Part of the SHELXTL Crystal Structure; Determi-

nation Siemens Industrial Automation, Inc.: Madison, WI, 1995.

(73) SADABS: Siemens Area Detector ABSorption correction program; George

Sheldrick, 1996. Advance copy, private communication.

JA011876O

9

J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4737

Article Doi

A comprehensive investigation of the chemistry and basicity of a parent amidoruthenium complex

DOI: 10.1021/ja011876o

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja011876o

Full text of DOI:10.1021/ja011876o

Products guided by the article

R&D Labs maybe for 433301-63-4

Relevant to this article

Hot Product