PROJECT SUMMARY REPORT

submitted to

THE U.S. DEPARTMENT OF ENERGY

for

TRANSITION METAL ACTIVATION AND FUNCTIONALIZATION OF CARBON-HYDROGEN BONDS

William D. Jones, Principal Investigator University of Rochester

Department of Chemistry Rochester, New York 14627

Phone: 585-275-5493

Grant No. DE-FG02-86ER13569

Total Project Period: December 1, 2001 - November 30, 2004 Total Award Amount (3 years): $ 395,000

Unexpended Balance from Previous Year: $0

2

Executive Summary of FY 2001-4 Research in the Chemical Sciences Transition Metal Activation and Functionalization of Carbon-Hydrogen Bonds Grant FG02-86ER13569 William D. Jones, Department of Chemistry, University of Rochester, Rochester, NY 14627 Total Grant Period: 12/1/01-11/30/04, $395,000 for three years

Overview of Research Goals and Accomplishments: This project has as its overall goal improvement in the intelligent use of our energy

resources, specifically pertroleum derived products, and is aimed at the development of new routes for the manipulation of C-H and C-C bonds. During the 3 year project period, our research has focussed on the following specific goals: (1) discovery of new carbon-carbon bond cleavage reactions, (2) fundamental studies of C-H bond cleavage reactions of trispyrazolyl-boraterhodium complexes, (3) catalytic C-H and C-C bond functionalization, and (4) carbon-fluorine bond activation. This year we have made progress in several of these areas, as described in the following report.

The specific accomplishments of the current grant period include:(1) we have successfully measured and qunatitatively modeled the processes available to metal alkyl hydride complexes in a trispyrazolylborate-rhodium complex; (2) we have for the first time determined the isotope effects on the specific fundamental steps involved in alkane oxidative addition and reductive elimination; (3) we have preliminary results that measure for the first time the selectivity for a metal fragment binding to methyl vs methylene groups in a linear hydrocarbon; (4) we have cleaved C-C bonds in biphenylene, aryl-nitriles, and aryl-acetylenes, expanding tremendously the breadth of C-C cleavage; (5) we have established 3 different mechanisms for C-F bond cleavage of fluorocarbons using Cp*2ZrH2; (6) we have synthesized and studied several hemi-labile P-N complexes of platinum and nickel, and demonstrated that the labile nitrogen give enhanced reactivity not observed with the well-known P-P chelate complexes; (7) we have discovered new C-H and C-C functionalizations that allow introduction of reactive boronate and olefin functional groups; (8) we have cleaved C-C bonds in allyl nitriles, leading to isomerization of the C-C skeleton. This reaction is critical to the DuPont synthesis of Nylon from butadiene; (9) we have measured C-H activation selectivities in chloroalkanes, showing a preference for activation of the terminal methyl groups and the facile β-chloride elimination reaction.

A wide variety of chemistry has been examined, resulting in publication of several manuscripts. The work has been communicated at both national and international meetings. DOE funds have been used for the partial support of 4 graduate students (Andrew Vetter, Brad Kraft, Steve Oster, Karlyn Skugrud) and 1 postdoc (Nicole Brunkan) during the current grant period, as well as several undergraduates (Jason Holt, Mike Evans, John Curley, Brian Warsop, Susan Golisz).

The continued success of this work will lead to the development of new techniques and processes for the manipulation of petroleum-based hydrocarbons. These new processes will be based upon the new methods for making and breaking strong bonds in organic molecules of the type studied here. The work has the potential to have a significant impact in science and in technologies of interest to DOE.

A detailed report follows, followed by a listing of the 20 DOE supported publications for the current grant cycle and recent special recognitions received by the PI.

DOE Report, 2001-2004 3 William D. Jones

Detailed Progress Report for the Project Period Dec. 1, 2001- Nov. 30, 2004.

This report summarizes research that has been performed since during the current 3-year

grant, as well as work that will be completed and published by the end of the grant period.

1. Tris-pyrazolylborate Rhodium C-H Activation Studies.

Our rhodium-trispyrazolylborate studies on hydrocarbon activation make use of the

reactive 16-electron fragment [HB(3,5-dimethylpyrazolyl)3]Rh(CNCH2CMe3), abbreviated

herein as Tp'RhL. In the prior 3-year project period, we established that the Tp'RhL fragment

coordinates an alkane to give a σ-complex. A series of labelling studies allowed the

determination of the relative rates of the processes available to the alkane σ-complex,

specifically: (1) C-H activation (called oxidative cleavage), (2) migration down the alkane chain,

or (3) simple dissociation. These experiments involved modelling of the scrambling of

deuterium in complexes such as Tp'Rh(L)(CH2CH2CH3)D before loss of propane-d1. Figure 1

shows the relative rates of these processes for methyl, ethyl, n-propyl, and n-butyl derivatives.

For methane, C-H activation is strongly preferred over dissociation, whereas for ethane, the rates

of these two processes are closer. End-to-end migration in ethane is intermediate. For propane,

terminal C-H activation is favored over dissociation to a lessor extent than methane, but

comparable to ethane. Migration from the end to the middle of propane is slightly slower than

C-H activation. For the secondary propane complex, migration to the end and dissociation occur

at about the same rate. Interestingly, migration down a butane chain (secondary to secondary) is

the fastest process, accounting for the observed kinetic preference for terminal C-H activation.

0 5 10 15

methane

ethane

propane

butane

migration from 2°-2°migration from 2°-1°2° dissociation

migration from 1°1° dissociationactivation

relative rate (kd1=1)

fixed ratio

Figure 1. Relative rates of σ-alkane processes.

4

These conclusions can be expressed in a schematic fashion as shown in Scheme 1. Our

studies do not permit the determination of either the absolute or relative energies of the primary

and secondary alkane complexes, so we cannot establish the relative rates of processes between

these two intermediates. Relative rates are shown assuming the energies of the two alkane

complexes are equal.

Scheme 1: Relative rates of processes available in primary (1°) and secondary (2°) alkane

complexes.

*Assuming 1° alkane σ-complex is at the same energy as the 2° alkane σ-complex

(6x)

(3x)

(1x)

MH

M C C2H5H

HCH3

M +

M CH

HH

koxidative cleavage

kmigration

kdissociation

MH

primary

primary

butane σ-complex

secondarykoxidative cleavage

kmigration(back)(3x)*

(11x)*

(1.6x)*secondarykdissociation

CH

HH

M

CH

HCH3

Mkmigration(2°-2°)

(0.01x)*

With the relative rates of all of these processes now known for any linear alkane, we have

completed studies to determine which C-H bond of an alkane first binds to the Tp'RhL fragment.

The execution of this experiment is not completely obvious, as reaction of the fragment with any

linear hydrocarbon only gives a single product, the n-alkyl hydride (eq 1). One cannot determine

how the alkane initially bound if a single product is observed.

kprimary

ksecondary

MH

M C CH3H

HCH3

M +

MH

M CH

HH

(1)

In the current project period, we established that (1) the Tp'RhL fragment coordinates to a

linear alkane to give a σ-complex and that the coordination is favored at the methylene group

over the methyl group by a ratio of 1.5:1. (2) a methyl group in pentane coordinates 1.9 times

5

faster than the more hindered methyl group of isobutane. (3) the oxidative cleavage of a methyl

C-H bond (primary C-H) occurs 65K12 times faster than the C-H bond in a methylene group

(secondary C-H), where K12 represents the equilibrium constant between primary and secondary

alkane complexes. These conclusions were made building upon our earlier studies of the relative

rates of oxidative cleavage, migration, and dissociation indicated in Scheme 1. The experimental

evidence for these conclusions is given below.

The first new result is determination of the rate at which a reactive metal complex with a

vacant site coordinates a methyl group C-H vs a methylene C-H. This was determined by

looking at the product distribution in a competition experiment between pentane and decane. In

this experiment, the metal intermediate would see an equal number of methyl groups in each

substrate but differing numbers of methylenes (see Scheme 2). Note that in either case, only the

terminal activation product would be seen since coordination to an internal methylene would

lead to migration of the metal along the chain to the end, where oxidative cleavage would occur.

If

Scheme 2: Pathways for C-H coordination/activation in pentane-decane competitition.

kd1kd2kd2 kd2

kd2

Tp'LRh

H3kOC

H

km12

km21

RhLTp'

C HH

H

RhLTp'

C HH

H3C

RhLTp'

C HH RhLTp'

C HH

RhLTp'

C HHkm22

km22

km22

km22

km22

km22

km22

km22

km22

km22

km22

km22Tp'LRh

CHH

Tp'LRh

CHH

Tp'LRh

CHH Tp'LRh

CHH

CH3 CHH

H

Tp'LRh

km12

3kOCH

RhLTp'

H

km22

km22

kCH3kCH2kCH2kCH2 kCH2kCH2 kCH2

kCH2kCH2kCH3 kd2kd2

kd2 kd2kd1

A

B C D E FG H

I J K L M N O P Q RS T

1 2 3 4 5 6

78 9 10 11

1213 14 15 16 17 18 19 20 21 22

2324

2526

27 28 29 3031

32

km21

L = CNCH2CMe3

kCH2kCH3

km21

km12

km22

km22

RhTp'L

H

Tp'LRh

C HH

HRhLTp'

C HH

H3CRhLTp'

C HH

pentane/decane

3kOCH

kd1 kd2kd2

RhLTp'

C HH

H3CRhLTp'

C HH

H

Tp'L Rh

H

km22

km22 km12

km21

3kOCH

kCH2 kd2 kd1kCH3kCH2

Tp'LRh(PhN=L)hν

-PhN=L[Tp'LRh]

kCH3 @ 7, 11, 23, 32

kCH2 @ 8, 9, 10, 24, 25, 26, 27, 28, 29, 30, 31

kRCH

kRCH

the reactive intermediate were to bind only to methyl groups C-H bonds, then a 1:1 product ratio

would be expected. If the reactive intermediate were to bind only to methylene C-H bonds, then

6

a product ratio favoring decane activation would be expected (corresponding to the increased

number of methylene groups).

Scheme 2 shows all possible intermediates and pathways. It is rather complicated (!), but

since the earlier studies established many of these relative rates, only the rates of primary

coordination (kCH3) and secondary coordination (kCH2) need to be determined. In fact, we only

want to know the ratio of these rates so that the ratio can be adjusted in a simulation to match the

experimentally observed ratio of products. In the experiment, Tp*Rh(CNR)(carbodiimide) was

irradiated in a 1:1 mixture of pentane/decane. The alkyl hydride products were quenched with

CCl4 to give the stable chloro derivatives in a 1.1:1 ratio (Scheme 3). Consequently, the results

are consistent with a slight preference for methylene group coordination, and the simulation

indicates an actual preference of kCH2/kCH3 = 1.5:1.

Scheme 3. Competition between pentane and decane.

[Tp'LRh]-PhN=L

hνTp'LRh(PhN=L)

L = CNCH2CMe3

Tp'LRhH

Tp'LRhH

+

-20 °C

CCl4

+

Tp'LRhCl

Tp'LRhCl

1

:

1.1

Steric hinderence can interfere with the ability to coordinate to a methyl group C-H bond.

This can be seen in a competition between pentane and isobutane. In this experiment, two

methyl groups in a linear alkane compete with three methyl groups in a branched alkane. The

competitive reactions involved are shown in Scheme 4, with the two different rate constants for

methyl group binding shown as kCH3 and kCH3'.

7

Scheme 4. Pathways for C-H coordination/activation in pentane-isobutane competitition.

3kCH3'

3kOC1°

Tp'LRhC

H

H

Tp'LRhC

H

HCH3

Tp'LRhC

H

HH

Tp'LRhH

3kOC1°

km22 km22

km12km212kCH3

2kCH2

kCH2

Tp'LRh

H

Tp'LRhC

H

HH

[Tp'LRh]

-PhN=Lhν

Tp'LRh(PhN=L)kd1

kd2

kd2

kd2

kCH3/kCH3' = 1.9

In the experiment, Tp*Rh(CNR)(carbodiimide) was irradiated in a 1:1 mixture of

pentane/isobutane. The alkyl hydride products were quenched with CCl4 to give the stable

chloro derivatives in a 1.35:1 ratio (Scheme 5). Once again, simulation of the experiment while

varying only the ratio kCH3/kCH3' allows determination of the ratio as 1.9:1. Note that in this

experiment, we assume that the rate of oxidative cleavage of the C-H bond in the two methyl

complexes is the same. It is possible to interpret this experiment in terms of an equal rate of

binding to the two types of methyl groups, but with a 1.9:1 ratio between the rate of the two

oxidative cleavage rates.

Scheme 5. Competition between pentane and isobutane.

[Tp'LRh]-PhN=L

hνTp'LRh(PhN=L)

L = CNCH2CMe3

Tp'LRhH

Tp'LRhH

+

-20 °C

CCl4+

Tp'LRhCl

Tp'LRhCl

1.35

:

1

In the pentane/decane competition experiment, we learned about the relative rates of

binding of unhindered CH3 vs CH2. It would be reasonable to expect this rate ratio to apply in

other competitions involving the binding of similar types of bonds. For example, the methylene

8

C-H bonds in cyclohexane would be expected to bind to the [Tp*RhL] fragment with the same

ease as those in pentane, since cyclohexane is an unstrained, 'natural' conformation similar to

that in a linear alkane. Interestingly, cyclohexane does undergo activation of its C-H bonds by

the [Tp*RhL] fragment because secondary C-H bonds are the only ones available. In a

competition between pentane and cyclohexane, the metal fragment should bind to both alkanes

in a predictable fashion since kCH3/kCH2 is known. From the ratio of products obtained, one

should be able to determine the relative rates of oxidative cleavage of a secondary C-H bond

(kOC2°) vs a primary C-H bond (kOC1°), provided the alkane σ-complexes are at the same energy.

These pathways and the corresponding rate constants are shown in Scheme 6.

Scheme 6. Pathways for C-H coordination/activation in pentane-cyclohexane competitition.

kRC2

2kOC2°

1312kd2 kCH2

Tp'L Rh

H

RhLTp'

HH

kRCH

kRCH

[Tp'LRh]-PhN=L

hνTp'LRh(PhN=L)

kCH2 kCH3kd1

kd2kCH2

3kOCH

km21

km12km22

km22

Tp'L Rh

H

RhLTp'

C HH

H

RhLTp'

C HH

H3C

kd2kd2

kd1

3kOCH

pentane/cyclohexane

RhLTp'

C HH

RhLTp'

C HH

H3C

Tp'LRh

C HH

H RhTp'L

Hkm22

km22

km12

km21

kCH3kCH2

L = CNCH2CMe3

1110987

654321

HG FEDCB

A

kCH2 @ 8, 9, 10, 12

kCH3 @ 7, 11, 23, 32

kOC1°/kOC2° = 65 (with 1° and 2° alkane complexes at same E)

or otherwise:kOC

1°/kOC2° = 65 (km12/km21)= 65 (K12)

I J

In the competition experiment, Tp*Rh(L)(carbodiimide) was irradiated in a 1:1 mixture of

pentane/cyclohexane and quenched with CCl4. A 6.5:1 ratio of the n-pentyl and cyclohexyl

chloride products was seen (Scheme 7). From the simulation, this corresponds to a 65:1 ratio for

kOC1°/kOC2°, assuming the alkane σ-complexes are at the same energy. If they are at different

energies, then a 'correction' must be applied corresponding to Keq between the two σ-complexes.

9

Scheme 7. Competition between pentane and cyclohexane.

[Tp'LRh]-PhN=L

hνTp'LRh(PhN=L)

L = CNCH2CMe3

Tp'LRhH

Tp'LRhH

+

-20 °C

+

Tp'LRhCl

6.5

:

1 Tp'LRhCl

CCl4

Finally, we have also looked at activation of cyclopentane vs pentane in a competition

experiment. Here, due to the strain in the cyclopentane ring, it is not reasonable to assume that

the rate of methylene coordination is the same as in a linear alkane. Consequently, the scheme

for this experiment shows different rates for the two types of methylene coordination (Scheme

8), but the same rate of oxidative cleavage in the methylene-alkane complex.

Scheme 8. Pathways for C-H coordination/activation in pentane-cyclopentane competitition.

5kCH2'

3kOC1°

Tp'LRhH

2kOC2°

km22 km22

km12km212kCH3

2kCH2

kCH2

Tp'LRh

H

kd1

kd2

kd2

kd2

kCH2 /kCH2 = 1.7'

Tp'LRhC

H

H

Tp'LRhC

H

HCH3

Tp'LRhC

H

HH

Tp'LRhC

H

H

[Tp'LRh]

-PhN=Lhν

Tp'LRh(PhN=L)

In the competition experiment, Tp*RhL(carbodiimide) was irradiated in a 1:1 mixture of

pentane and cyclopentane, and CCl4 added to quench the products. A 4.5:1 ratio of n-pentyl to

cyclopentyl products were seen (Scheme 9), showing that cyclopentane is slightly more reactive

than cyclohexane, as anticipated. From the kinetic simulation of the competition, a ratio for

kCH2'/kCH2 of 1.7 was obtained.

10

Scheme 9. Competition between pentane and cyclopentane.

[Tp'LRh]-PhN=L

hνTp'LRh(PhN=L)

L = CNCH2CMe3

Tp'LRhH

+

-20 °C

+Tp'LRhCl

4.5 : 1Tp'LRhH

Tp'LRhCl

CCl4

As a check that this value represents the relative coordination abilities of cyclopentane vs

cyclohexane, a competition experiment was run by irradiation of Tp*RhL(carbodiimide) in a 1:1

mixture of cyclopentane/cyclohexane. A 1.9:1 ratio of cycloalkyl chloride products was

observed. The excellent agreement between the competition ratio from these two independent

experiments argues that the assumptions made in the simulations are reasonable.

One of the more interesting side-lights from these studies comes from the independent

determination of isotope effects for both the 'oxidative cleavage' and the 'reductive coupling'

steps of the C–H activation reaction indicated in equation 2. These isotope effects, both kinetic

isotope effects on a fundamental reaction step, were found to be normal isotope effects. The

overall effect on alkane reductive elimination, however, is to generate an inverse kinetic isotope.

The initial equilibrium isotope effect between the alkyl hydride complex and the alkane sigma-

complex is inverse, not because either of the individual rates are inverse, but because the ratio of

these isotope effects is inverse. As this was the first known system where these effects had been

completely sorted out, we published a more didactic article in Accounts of Chemical Research to

shed light on the analysis of this controversial subject.

slow+Lfast Ln+1M

LnMCH2R

D

[LnM] + R-CH2D

LnMCH2R

DLnM

CHDR

H kRCH

kOCH

kOCD

kRCD

(2)

2. C-C Bond Cleavage Studies

Our DOE supported work showed that several types of C-C bonds can be cleaved. We

have discovered 3 distinct classes of C-C bonds that can be cleaved: (1) strained rings such as

biphenylene undergo sp2-sp2 C-C cleavage with a number of metal complexes to give a variety

of products. (2) diphenylacetylene undergoes sp-sp2 cleavage photochemically when attached to

PtL2 complexes. (3) arylnitriles undergo sp2-sp C-CN cleavage when reacted with NiL2

11

fragments, and η2-nitrile adducts can be observed as reaction intermediates. These results are

described in more detail below.

Earlier DOE supported work showed that the complexes Pt(PEt3)3, Pd(PEt3)3, and [Ni

(dippe)H]2 cleave the C-C bond of biphenylene to give (PEt3)2Pt(2,2'-biphenyl), (PEt3)2Pd(2,2'-

biphenyl), and (dippe)Ni(2,2'-biphenyl), respectively. These complexes underwent catalytic

chemistry to give functionalized products, such as tetraphenylene, biphenylene, fluorenone, or

phenanthrenes (Scheme 10).

Scheme 10:

ML

L

+ ML2

L2 = (PEt3)2, dippeM = Ni, Pd, Pt

H2

CO

O

RC CRR R

In the case of Ni(dippe), it was found that catalysis required the introduction of small

amounts of oxygen, just enough to oxidize the phosphine to phosphine oxide. The remaining

'naked' metal was efficient at catalyzing the insertion of alkynes into biphenylene to give

phenanthrenes. The requirement of a labile chelate led us to investigate the use of the hemi-

labile ligand Pri2PCH2CH2NMe2 in these same metal systems during the current grant period.

The platinum complexes proved to be the easiest to synthesize and study, since the adducts

are fairly stable. The strategy was to use a source of Pt(0) in the presence of the P-N chelate and

an alkyne, to isolate the complex, and then to look at reactions of the complex. With

diphenylacetylene, the preparation of the adduct was straightforward and allowed comparison of

the P-P chelate with the P-N chelate (eq 3).

Pt

PR2

ER2´

+ PtP

E

R2

R2´- CODCH2Cl2

E = N, R = iPr, R´= Me: 1E = P, R = R´= iPr: 2 (3)

12

While the P-P chelated complex showed no reactivity with added diphenylacetylene even

after heating, the P-N chelated complex reacted at room temperature to give a

metallacyclopentadiene complex. Further studies of the system showed that the nitrogen of the

chelate is quite labile, and that once it dissociates the phosphine ligands can then redistribute

between metals to generate two observable intermediates (Scheme 11). Ultimately, however,

thermodynamics takes over and one winds up with quantitative production of the metallacycle.

Heating this sample results in the slow catalytic formation of hexaphenylbenzene.

Scheme 11:

+ 2 NMe2

Pt

(Pri)2 P C

C

Ph

Ph

NMe2

Pt

(Pri)2 P

Ph

Ph

Ph

Ph

Pt

Me2NPt

PPri2

PhPh

PhPh+ PtMe2N

R2P

Pt

PR2

Me2N

2 2

With trimethylsilylphenylacetylene, the preparation of the initial complex is similar but

now the alkyne complex undergoes insertion into the C-Si bond at room temperature (eq 4).

Once again, no such reaction is seen for the complex with a bis-phosphine chelate. Just as in the

case of diphenylacetylene, ligand dissociation and redistribution is observed at intermediate

times to give two observable intermediates (Scheme 12). Once again, thermodynamics takes

over and only a single product, the C-Si insertion complex, is observed at longer reaction times.

NMe2

Pt

(Pri)2 P

SiMe3

C6D6, 5 days

22 °C

NMe2

Pt

(Pri)2 P SiMe3

(4)

13

Scheme 12:

+

2 2

SiMe3

+ 2 NMe2

Pt

(Pri)2 P C

C

SiMe3

Ph

PtMe2N

R2P

Pt

PR2

Me2N

SiMe3

NMe2

Pt

(Pri)2 P SiMe3

NMe2

Pt

(Pri)2 P SiMe3

Me2 N

Pt P(Pri)2

SiMe3

Pt

Me3Si

While the dippe P-P chelate complex is unreactive thermally, it does react photochemically

to give a C-Si insertion product (eq 5). Remarkably, however, this complex reverts to the Pt(0)

alkyne complex thermally! This observation leads to the important conclusion that oxidative

addition is favored thermodynamically by the presence of the P-N ligand, whereas reductive

elimination is thermodynamically favored by the P-P ligand. This conclusion implies that the

study of a chelating, sterically hindered, bis-(dialkylamino)ethane ligand should give a metal

fragment that will strongly favor C-C cleavage. This hypothesis remains to be tested.

C6D6

PtP

P

SiMe3Pt

P

P

SiMe3Cy2

Cy2 Cy2

Cy2hν

∆

3 4 (5)

In addition to the above work with platinum, we have also prepared the analogous nickel

complex (eq 6). These complexes have proven to be efficient catalysts for the selective coupling

of biphenylene and alkynes to give phenanthrenes (eq 7).

Ni + +

P(iPr)2

NMe2

NiP

N- 2 COD(1)

(iPr)2

Me2

1 (6)

PhPhPhPh

+

T = 70°C

NiP

NPh

Ph(iPr)2

Me2

(7)

14

We have also discovered that the nickel complex [Ni(dippe)H]2 reacts with benzonitrile to

give first an η2-nitrile complex, which then undergoes C-C cleavage of the carbon-CN bond (eq

8). Furthermore, the reaction does not go to completion but forms and equilibrium mixture of

the η2-nitrile and C-CN oxidative addition product. We know of no such example of reversible

C-C cleavage in the literature. Other examples of aryl C-CN cleavage are under investigation.

N

C

Ph

Ni

Pri2 P

PPri2

Ni

Pri2 P

PPri2

HNi

Pri2 P

PPri2

H

+ 2 Ph-C N

2-H2

KeqNi

Pri2 P

PPri2

Ph

CN

(8)

We have also looked at the effects of electron withdrawing and electron donating groups

on the aryl cyanide. The results show that both the rate and equilibrium for C-CN cleavage is

affected, as shown in the plots in Figure 2. For the equilibration, Keq is found to have a ρ value

of +6.1. This large and positive value for ρ indicates that negative charge is being stabilized on

the ipso carbon of the substituted aryl group, consistent with the organometallic nature of this

bond as possessing substantial Niδ+ —Cδ– character. From the rate of approach to equilibrium,

the rate of the forward reaction k1 can be determined. Similarly, a plot of ln k1 vs. σ, also shown

in Figure 2b, gives a ρ value of +1.3. While there is somewhat more scatter in this plot, the

positive slope correlation is unmistakable and is consistent with the localization of charge

density on the ipso carbon in the transition state for C-CN bond cleavage, although not so much

as the Ni(II) product.

-5

-3

-1

1

3

5

-1 -0.5 0 0.5 1

σp

ln( K

eq/ K

H)

(a)

-1

-0.5

0

0.5

1

1.5

-0.75 -0.25 0.25 0.75σp

ln( k

1/ kH)

(b)

Figure 2. (a) Hammett plot for the equilibrium constants (Keq) from equation 8 vs. σP at 54 °C. (b) Hammett plot for the forward rate constants (k1) vs. σP at 54 °C.

15

We have also discovered an important new type of C-C bond oxidative addition, cleavage

of sp-sp2 C-C bonds in aryl acetylenes. This is a new class of C-C bond cleavage, and offers

many exciting possibilities. We have found that irradiation of either P-P or P-N chelate

complexes of Pt-(diphenylacetylene) leads to the clean and quantititive formation of the

oxidative addition product (eq 9). The reaction works for all complexes we have examined to

date. Furthermore, the oxidative cleavage reaction is reversible, in that heating the Pt(II)

complexes results in their reversion to the Pt(0)-alkyne complexes. Consequently, we now have

methodology to break and make C-C———C bonds, and further development of this reaction will be

the subject of future studies.

ER2'

R2P

Pthν

C6D6 ER2'

R2P

Pt

E = N, R = iPr, R'= MeE = P, R = R'= iPrE = P, R = R'= Cy

(9)

Recently, we have initiated investigations of a new class of C-C cleavage, that of allyl-

nitriles. At a metal center this cleavage reaction generates both a strong metal-cyanide bond and

a π-allyl ligand, and hence has been found to be both facile and reversible. Indeed, at nickel(0),

this reaction forms the basis of DuPont's synthesis of adiponitrile for the production of nylon via

addition of HCN to butadiene, to the tune of over 400 thousand metric tons per year!

We have discovered that the reactive hydride [(dippe)NiH]2, which serves as a room

temperature source of [Ni(dippe)], reacts with allylcyanide to give initially a π-olefin complex

(dippe = bis-(diisopropylphosphino)ethane). This species can be seen at low temperature by

NMR spectroscopy, and upon warming to RT competitive C-H and C-CN cleavage takes place.

C-H activation gives a π-allyl hydride complex that is not observed, because the hydride is

transferred back to the opposite end of the allyl group to give a very stable crotononitrile

complex (both cis and trans are formed). C-CN activation, however, leads to a metastable π-

allyl cyanide complex that can be isolated and was structurally characterized. C-CN cleavage is

reversible, so that ultimately, all nickel winds up as the crotononitrile complexes (Scheme 13).

16

Scheme 13. C-C and C-H Bond Activation in allylcyanide.

k2, k3

[Ni(dippe)H2]2

CN+

PPri2

Ni

Pri2 P

CN

PPri2

Ni

Pri2 P

H

CN

PPri2

Ni

Pri2 P CN

PPri2

Ni

Pri2 P

NC

k1, C-C activation

C-H activation

C-C formation, k-1 ∆H‡ (kcal/mol) ∆S‡ (e.u.) ∆G‡ at 40 °C (kcal/mol)k1 (C-C cleavage) 18.0(0.3) -14.8(0.8) 22.71k-1 (C-C formation) 27.5(0.5) 11.7(1.6) 23.75k2 (C-H cleavage) 13.4(0.2) -29.9(0.6) 22.95k3 (C-H cleavage) 13.2(0.2) -31.2(0.7) 23.14

By monitoring the distribution of species over time, we have been able to extract the rate

constants for all of these species by kinetic simulation. In addition, by measuring the

distribution of species as a function of temperature, we can obtain activation parameters for the

various steps. The results are quite interesting, in that we find that while C-H activation and C-C

activation have small temperature dependences, C-C cleavage has a large temperature

dependence. The result is that by raising the temperature, one can selectively drive the reaction

in the direction of the less-favorable π-allyl cyanide complex. This is good news, since the

DuPont catalysis requires that the C-C cleavage dominate the C-H cleavage. The activation

parameters support the mechanism for C-H anc C-C cleavage shown in Scheme 14.

Scheme 14:

Ni

CN r.d.s.CN

Ni

H

crotononitrilesfast

‡

Ni

CN H

H

r.d.s.Ni

CNNi

CN

Ni

NC

‡

N

CNi

NiN

C

HH

Ni

CN

We have also investigated the effect of the Lewis acid BPh3 upon the isomerization

reaction. Addition of BPh3 to a cold solution of (dippe)Ni(η2-allylCN) leads to the immediate

and quantitative formation of the BPh3 adduct of the π-allyl cyanide adduct. Addition of a

second equivalent of the Lewis acid leads to the removal of the cyanide ligand as the

17

[Ph3B-CN-BPh3]- anion, leaving behind the [(dippe)Ni(π-allyl)]+ cation (Scheme 15). With only

one equivalent of BPh3, there is evidence of linkage isomerism of the initial Ni-C-N-BPh3 adduct

to the Ni-N-C-BPh3 adduct.

Scheme 15. Effect of Lewis acid on the C-C and C-H Bond Activation in allylcyanide.

PR2

Ni

R2P

HNi

H

PR2

R2P CN

BPh3

slow

PiPr2

Ni

iPr2 P

RT PiPr2

Ni

iPr2 P

BPh3NC

PiPr2

Ni

iPr2 P

BPh3CN

slow

[Ph3B-NC] [Ph3B-CN]

3. C-H and C-C Bond Functionalization Studies

We have also initiated investigations of the above systems for their ability to serve in

further functionalization reactions of hydrocarbons. For example, we have found that C-C bonds

can not only be cleaved, but functionalized using 'standard' organic reagents. A nice example of

the application of this chemistry appears in our work with palladium catalyzed reactions of

biphenylene. Using a Pd(0) precursor, we can activate the C-C bond of biphenylene and then

protonate one of the Pd-C bonds using a weak acid of the appropriate pH. Next, one can perform

Heck or Suzuki-type couplings using olefins or boronic acids to give functionalized products (eq

10, 11). Furthermore, other acidic C-H bonds can be added across the activated C-C bond using

pH control. For example, α-keto C-H bonds or α-nitrile C-H bonds can add across biphenylene

to give functionalized biaryls (eq 12, 13).

+R

(1.3 equiv)

Pd(PPh3)4 (5 mol%)

p-cresol (1 equiv)

C6D6, 120 °C, 6 dR

+ R

(10)

18

C6D6, 120 °C

p-cresol (1 equiv)

Pd(PPh3)4 (5 mol%)

(1 equiv)

+

R

B(OH)2

R

(11)

C6D6, 120 °C

p-cresol (10 mol%)

Pd(PPh3)4 (5 mol%)+ R

OR

O (12)

C6D6, 120 °C, 1 d

p-cresol (10 mol%)

Pd(PPh3)4 (5 mol%)

(1 equiv)

+

Me

CN

CN

Me (13)

We have conducted an extensive investigation of C-H activation in chloroalkanes using the

reactive precursor Tp*Rh(CNR)(carbodiimide). Remarkably, the C-Cl bond does not undergo

oxidative addition. Rather, we find a strong selectivity for exclusive terminal methyl group C-H

bond activation. Thus, 1-chloro alkane gives the 5-chloropentyl hydride as the only product. 3-

chloropentane gives the 3-chloropentyl hydride product. In this case, however, two

diastereomers are formed in a 1:1 ratio, since the metal is chiral and the 3-chloro substituent

renders the ligand chiral. If a chlorine is present in a β-position, then β-chloride elimination

occurs to give an olefin and the metal chloride. Therefore, 2-chloropropane gives only propene

and the hydrido chloride Tp*Rh(CNR)HCl. 2-chloropentane gives a mixture of the C-H

activation product 4-chloropentyl hydride, pentene, and the hydrido chloride. These reactions

are summarized in Scheme 16.

We have also begun to investigate aliphatic nitrile complexes. It appears that there is once

again a selectivity for terminal methyl groups. With acetonitrile, the adduct Tp*LRh(CH2CN)H

is formed, and is found to be stable for days at 60 °C! This is the most stable alkyl hydride in

this series yet, and we may be able to do new functionalization chemistry with this derivative.

19

Scheme 16. Reactions of Chloroalkanes with [Tp*Rh(CNR)].

Tp'LRh

+ PhNCNR

hν

L = CNCH2CMe3

-20 °C

Cl

ClTp'LRh

HCl

Tp'LRh

Cl

H

Cl

+

Tp'LRhClH

+ Tp'Rh(L)HCl

+ Tp'Rh(L)HCl

ClTp'LRhN

C

R

NPh

4. C-F Bond Cleavage Studies

We have now begun studies with the soluble, more reactive Cp*2ZrH2 and found that this

molecule cleaves a wide variety of aromatic and aliphatic C-F bonds. Systematic studies have

shown that primary, secondary, and tertiary C-F bonds can all be cleaved with progressively

greater difficulty (Scheme 17). In addition, di-fluorosubstituted carbons can be made to react

with even more forcing conditions. Trifluoromethyl groups scarcely react at all even under

extreme conditions.

Scheme 17:

F

F

F

H

H

H

++

C6D12, H2 2 d, 25oC

C6D12, H2

C6D12, H2

4 d, 120oC

1 d, 150oC

primary C-F:

secondary C-F:

tertiary C-F:

ZrH

H

Me5

Me5

ZrF

H

Me5

Me5

Most remarkable, however, even trifluoromethyl C-F bonds can be easily cleaved if they

are adjacent to a double bond. 3,3,3-trifluoropropene is completely defluorinated in 5 min at

room temperature to give the zirconium-n-propyl hydride complex (Scheme 18).

Perfluoropropene undergoes a similar reaction to give the same product. Details of the

20

Scheme 18:

F2C CF3

F

d12

Cp*2ZrHF +

CF3

F2C CH3

H2

CH3CH2CH3 + Cp*2ZrH2

22°C

+ Cp*2ZrHF

Cp*2ZrH2

22°C

d12

excess

+ Cp*2ZrHF

H2

ZrH

H

Me5

Me5

ZrH

CH2CH2CH3

Me5

Me5

ZrH

CH2CH2CH3

Me5

Me5

mechanism are under further study. Defluorination reactions are also seen with

nonafluorohexene, perfluorocyclobutene, perfluorocyclopentene, perfluorobenzene,

trifluorotoluene, and related substrates. Chlorofluorocarbons (CFCs) react rapidly to give first

fluorocarbons (HFCs), which then are converted to hydrocarbons (HCs) in accord with the above

established reactivities (Scheme 19). Mechanistic investigations into the aliphatic fluorocarbons

has revealed evidence for a radical chain mechanism. Further mechanistic work with the

fluoroolefins is underway suggesting an insertion/β-fluoride elimination pathway.

Scheme 19. x Cp*2ZrH2

CFCl2H

CF2Cl2

CF2ClH

CH3F

CF2H2

CF2H2

CH4

CH4

CH4

25oC, 5 min.

+

25oC, 5 min.

25oC, 5 min.

RT1 day

120oC

120oCx = 3, 4

slow

slow

We have now continued our investigation of the C-F cleavage in perfluoroolefins. These

appear to be a special class of substrate, in that the mechanism of C-F cleavage may be different

than that seen in our earlier studies with Cp*2ZrH2. Reaction with perfluoropropene gives first

the selective formation of E-CHF=CFCF3. Further reaction with zirconium hydride leads to

complete defluorination with no further intermediates being seen (Scheme 20).

21

Scheme 20. Reaction of Cp*2ZrH2 with perfluoropropene.

+CF3

F

H

F

6 Cp*2ZrH2

- 5 Cp*2ZrHFCF3

F

F

F+

ZrF

H

ZrH

H-40 C

ZrH

The mechanism for the reaction could involve hydridic attack on the olefin with H/F

metathesis, or an insertion/elimination mechanism. The olefin could approach centrally,

between the two hydrides, or laterally, with the two hydrides remaining cis to each other. In

order to investigate these possibilities, we have initiated a collaboration with a theory group in

Montpellier. Odile Eisenstein and Eric Clot have provided high level calculations investigating

these systems, and the work is providing guidance for the mechanism of reaction. Initial studies

were done on the insertion reaction of ethylene with Cp2ZrH2. We then progressed to

trifluoropropene, which has been shown experimentally to undergo an insertion/β-fluoride

elimination pathway. The calculations done thus far are able to reproduce the selective internal

insertion product, and further work is underway to look at the β-fluoride elimination step

(Scheme 21).

Scheme 21. Calculations on the reaction of Cp*2ZrH2 with trifluoropropene.

externalcoordination

internalcoordination

internalcoordination

0.0

-22.8

-36.8

C2

D2

C1

Cp

Zr

CpH

H

CF3Cp

Zr

CpH

H

CF3

Cp

Zr

CpH

H

CF3

-22.6

Cp2ZrH2 + C2H3CF3

E2

-31.6

Cp

Zr

CpH

H

CF3D1E1

-32.1

Cp

Zr

CpH

CF3Cp

Zr

CpH

CF3

-32.0

Further calculations have been recently completed with Cp2ZrH2 reacting with

perfluoropropene. The results show that interaction of the olefin internally with Cp2ZrH2 leads

to olefin insertion followed by an external β-fluoride elimination. Other pathways were

investigated but all were found to lie at higher energies (Scheme 22).

22

Scheme 22. Calculations on the reaction of Cp*2ZrH2 with perfluoropropene.

���

���

����

�

����

����

�

�����

�����

��

����

�����

��

23

Publications appearing during the current grant cycle acknowledging DOE support, December 1, 2001 - November 30, 2004:

Manuscripts in print:

1. “Perspectives: Synthetic Chemistry. The Key to Successful Organic Synthesis is…,” William D. Jones, Science, 2002, 295, 289-290.

2. “Formation of Tetrafluorobenzyne by β-Fluoride Elimination In Zirconium-Perfluorophenyl Complexes,” Bradley M. Kraft, Rene J. Lachicotte, and William D. Jones, Organometallics, 2002, 21, 727-731.

3. “Cleavage of the Carbon-Carbon Bond in Biphenylene using Transition Metals,” Christophe Perthuisot, Brian L. Edelbach, Deanna L. Zubris, Nira Simhai, Carl N. Iverson, Christian Müller, Tetsuya Satoh, and William D. Jones, J. Mol. Catal. A, Chemical, 2002, 189, 157-168.

4. “Chelating P,N versus P,P Ligands: Differing Reactivity of Donor Stabilized Pt-(η2-PhC———CPh) Complexes Towards Diphenylacetylene,” Christian Müller, Rene J. Lachicotte, and William D. Jones, Organometallics, 2002, 21, 1118-1123.

5. “Thermal and Photolytical Silicon-Carbon Bond Activation in Donor Stabilized Pt(0)-Alkyne Complexes,” Christian Müller, Rene J. Lachicotte, and William D. Jones, Organometallics, 2002, 21, 1190-1196.

6. “Catalytic C-C Bond Activation in Biphenylene and Cyclotrimerization of Alkynes: Reactivity of P,N versus P,P Substituted Nickel Complexes,” Christian Müller, Rene J. Lachicotte, and William D. Jones, Organometallics, 2002, 21, 1975-1981.

7. “Mechanism of Vinylic and Allylic Carbon-Fluorine Bond Activation of Non-Perfluorinated Olefins using Cp*2ZrH2,” Bradley M. Kraft, Rene J. Lachicotte, and William D. Jones, J. Am. Chem. Soc., 2002, 124, 8681-8689.

8. “Cleavage of Carbon-Carbon Bonds in Aromatic Nitriles using Nickel(0),” Juventino J. Garcia, Nicole M. Brunkan, and William D. Jones, J. Am. Chem. Soc., 2002, 124, 9547-9555.

9. “Carbon-Fluorine Bond Activation of Perfluorinated Arenes with Cp*2ZrH2,” Bradley M. Kraft and William D. Jones, J. Organomet. Chem., 2002, 658, 132-140.

10. “η2- Coordination and C–H Activation of Electron-poor Arenes,” Carl N. Iverson, Rene J. Lachicotte, Christian Müller and William D. Jones, Organometallics, 2002, 21, 5320-5333.

11. “Isotope Effects in C-H Bond Activation Reactions by Transition Metals,” William D. Jones, Acc. Chem. Res. 2003, 36, 140-146.

12. “Preparation, Structure, and Dynamics of a Nickel π-Allyl Cyanide Complex,” Nicole M. Brunkan and William D. Jones, J. Organomet. Chem. 2003, 683, 77-82.

13. “Activation of C-F Bonds using Cp*2ZrH2: A Diversity of Mechanisms,” William D. Jones, J. Chem. Soc., Dalton Trans. 2003, 3991-3995.

14. “Synthesis, Characterization, and Reactivity of a Rhenium Complex with a Corannulene Based Ligand,” Robert M. Chin, Benjamin Baird, Michael Jarosh, Shane Rassman, Brian Barry, and William D. Jones, Organometallics 2003, 22, 4829-4832.

24

15. “Structural Properties and Inversion Mechanisms of [Rh(dippe)(µ−SR)]2 Complexes,” Steven S. Oster and William D. Jones, Inorg. Chim. Acta, 2004, 357, 1836-1846.

16. “Carbon-hydrogen bond activation of chloroalkanes by a rhodium trispyrazolylborate complex,” Andy J. Vetter and William D. Jones, Polyhedron, 2004, 23, 413-417.

17. “Kinetics, Thermodynamics and Effect of BPh3 on Competitive C-C and C-H Bond Activation Reactions in the Interconversion of Allyl Cyanide by [Ni(dippe)],” Nicole M. Brunkan, Donna M. Brestensky, and William D. Jones, J. Am. Chem. Soc., 2004, 126, 3627-3636.

18. “Defluorination of Perfluoropropene using Cp*2ZrH2 and Cp*2ZrHF: A Mechanistic Investigation from a Joint Experimental-Theoretical Perspective,” Eric Clot, Claire Megret, Bradley M. Kraft, Odile Eisenstein, and William D. Jones, J. Am. Chem. Soc., 2004, 126, 5647-5653.

Manuscripts Submitted or in press:

19. “Alkane Complexes as Intermediates in C-H Bond Activation Reactions,” William D. Jones, Andrew J. Vetter, Douglas D. Wick, Todd O. Northcutt, ACS Symposium Series, in press.

20. “Cleavage of Carbon-Carbon bonds in Alkyl Nitriles Using Nickel(0),”.Juventino J. García*, Alma Arévalo, Nicole M. Brunkan, and William D. Jones, Organometallics, submitted.

Manuscripts in Preparation:

21. “Alkane Coordination Selectivity in Hydrocarbon Activation by [Tp'Rh(CNneopentyl)]: The Role of Alkane Complexes,” Andrew J. Vetter, and William D. Jones, J. Am. Chem. Soc. 2004, to be submitted.

25

Recent Special Recognitions Received by the PI:

ACS Award in Organometallic Chemistry, 2003

Associate Editor, J. Am. Chem. Soc., January 2003-

Chair of Organometallic Subdivision, Inorganic Division of the American Chemical Society, 2001.

Charles F. Houghton Professor of Chemistry, 2000-present

July 2000- 2003: Chairman, Department of Chemistry

Organometallic Gordon Conference, Chairman, Newport, RI, 2000.

Opening speaker at Inorganic Reaction Mechanisms Gordon Conference, Ventura, 2003.

Opening speaker at Organometallic Gordon Conference, Newport, 2003.

Speaker at Inorganic Gordon Conference, Newport, 2003.

Speaker at Isotope Effects Gordon Conference, Ventura, 2004.

Additional Comments: Our renewal budget for 2001-2004 was less than our budget for the prior grant period. On

the basis of our success and productivity, as delineated by the DOE request for the specific information included in this progress report, it is becoming increasingly difficult to accomplish such a large body of work with so little funding. It is imperative that the grant be increased so that 2.5 people can work on the projects. Our current stipend for a good student is $22K/year, and the current budget is supporting only about 80% of these costs. We usually obtain several undergraduate workers for 'free' (i.e., paid by the University), but supply and instrument usage costs still need to be paid.

We apologize for the length of this report.