Insight into binding of alkanes to transition metals from NMR spectroscopy of isomeric pentane and isotopically labeled alkane complexes

Article abstract of DOI:10.1021/ja044208m

Alkane complexes of the type Cp′Re(CO)₂(alkane) (Cp′ = cyclopentadienyl or (isopropyl)cyclopentadienyl; alkane = isotopomers of n-pentane and cyclopentane) have been characterized using NMR spectroscopy following photolysis of Cp′Re(CO)₃ in the appropriate alkane at 163-193 K. In the case of n-pentane, three different complexes are observed corresponding to binding of the three different types of carbon in this alkane. ROESY NMR experiments indicate that these isomeric complexes are slowly interconverting intramolecularly at 173 K. The order of the energetically preferred site of coordination is methylene (C2) ≈ central methylene (C3) > methyl (C1) but with a spread of <0.2 kcal mol^-1. Isotopic perturbation of resonance (IPR) experiments, conducted on several isotopomers of (i-PrCp)Re(CO)₂(1-pentane), showed a large shielding of the ¹H NMR chemical shift of the proton in a bound CHD₂ moiety (δ -3.62) and CH₂D (δ -2.64) compared with that of a bound CH₃ moiety (δ -1.99). Likewise, the value of ¹J_CH for the coordinated methyl group of isotopomers of (i-PrCp)Re(CO)₂(1-pentane) reduces in the order CH₃ > CH₂D > CHD₂. This suggests that the alkane coordinates in an η²-C,H fashion with a rapid exchange of bound hydrogen or deuterium within a methyl or methylene group, and that binding of a hydrogen atom is preferred over a deuterium by an amount of 0.23 ± 0.03 kcal mol^-1. Copyright

Full text of DOI:10.1021/ja044208m

Published on Web 03/04/2005
Insight into Binding of Alkanes to Transition Metals from NMR Spectroscopy
of Isomeric Pentane and Isotopically Labeled Alkane Complexes
Douglas J. Lawes, Spili Geftakis, and Graham E. Ball*
NMR Facility and School of Chemistry, UniVersity of New South Wales, Sydney NSW 2052, Australia
Received September 23, 2004; E-mail: g.ball@unsw.edu.au
It is now well established that, in alkane solvent, photodisso-
ciation of a ligand from many transition metal complexes leads to
the generation of a complex containing a weakly bound alkane
ligand.1,2 In this work, we have directly observed and clearly
differentiated three complexes of pentane binding via the three
different types of hydrogen present in the pentane molecule using
NMR spectroscopy. Binding site preferences have been determined.
This work builds on extensive IR studies by George and co-
workers3,4 and our previous observations.5,6
Potential modes of interaction of CH3 groups with a metal include
η2-C,H (I - asymmetric interaction mostly with just one C-H
bond), η1-H (II), η2-H,H (III - symmetric equal interaction with
two hydrogens), and η3-H,H,H (IV) as shown below:
There are two X-ray studies of simple alkanes interacting
significantly with transition or lanthanide metals in the solid state.
In a recent study, several alkanes were found to interact with a
uranium(III)-hexadentate macrocyclic ligand complex and an
1
Figure 1. The 500 MHz H NMR spectra of products obtained from the
photolysis of (i-PrCp)Re(CO)3 (ca. 3 mM) at 163 K in various pentane
isotopomers. With 5% pentane-d12 present in A and D. See text for details.
η2-C,H binding mode was proposed, although the hydrogens were
not directly located.7 In the second example, n-heptane interacts
with an iron(II)-double A-frame porphyrin complex. Positional
disorder complicated direct crystallographic analysis, and an
asymmetric mode of binding between modes I and III was
calculated.8 Notably, neither of these types of complexes retain their
alkane ligands long enough in solution for them to have been
characterized using routine NMR experiments, and the interaction
of the alkane with co-ligands may be responsible for significantly
stabilizing these complexes in the solid state.
Since the observation of Cr(CO)5(C6H12) following laser flash
photolysis of Cr(CO)6 in cyclohexane,9 numerous other alkane
complexes have been observed in solution, primarily using “fast”
techniques, such as time-resolved infrared (TRIR) spectroscopy.
Often, they have been detected as intermediates in the extremely
important C-H activation process prior to formation of the alkyl
hydride product.10,11 Classic examples in this category include
CpRh(CO)(C6H12)12 and Tp*Rh(CO)(alkane)-type species (Tp* )
tris(3,5-dimethylpyrazolyl)borate).13
Indirect evidence has been found that alkyl hydride complexes,
such as {(1,4,7-triazacyclononane)Rh(hexyl)H[P(OMe)3]}+ and
[Tp*Rh(CNR)(i-Pr)H] (R ) neopentyl), may convert to transient
alkane complexes in which the Rh center may bind to each of the
methylene and methyl sites of the corresponding propane or longer
alkane and switch between these sites.14-16 Significantly, these
complexes are typical in their preference for C-H activation
occurring at primary carbons; most complexes that activate linear
alkanes selectively activate the C-H bonds of CH3 groups in
preference to CH2 groups.10 Oxidative cleavage to form alkyl
hydrides does not occur readily in CpRe(CO)2(alkane) complexes.
Figure 1 shows the region of the 1H NMR spectrum that includes
the resonances of protons in alkane complexes, formed after
photolysis of (i-PrCp)Re(CO)3 (i-PrCp ) η5-(isopropyl)cyclopen-
tadienyl), that are directly bound to the metal center. The photolysis
was conducted on precooled samples in the probe of a 500 MHz
NMR spectrometer, as described previously.5 The spectra were
obtained using an excitation sculpting scheme to suppress the
resonances of the free alkane.17 The scheme effectively excites the
region containing the resonances shown in Figure 1 and those of
the Cp rings (not shown) relatively uniformly and avoids excitation
of the window between δ 0.6 and 1.6 containing the free solvent.
Spectrum A and resolution-enhanced expansions (arrowed) show
the result of photolysis in natural abundance n-pentane. The visible
resonances are assigned to the three possible pentane complexes,
wherein the metal is bound to C1, C2, or C3, namely, (i-PrCp)-
Re(CO)2(1-pentane) (1) at δ -1.99 (t, 3JH1H2 ) 6.4 Hz), (i-PrCp)-
Re(CO)2(2-pentane) (2) at δ -3.10 (m, J ≈ 6 Hz), and
3
(i-PrCp)Re(CO)2(3-pentane) (3) at δ -3.42 (quintet, JH3H2 ) 6.0
Hz), respectively. Integration of the resonances due to 1, 2, and 3
showed relative intensities of 6:6.07((0.21):2.90((0.11), compared
with a statistical ratio of 6:4:2.18 This slight favoring for binding
of CH2 sites over CH3 in this mixture corresponds to a thermody-
namic preference of 0.13 ( 0.02 kcal mol-1. That this is an
equilibrium mixture is confirmed by performing ROESY NMR
experiments17 that show the isomers 1, 2, and 3 are slowly
interconverting intramolecularly (∼1-10 s-1) at 173 K.19 The
favoring of binding to CH2 sites is in marked contrast to the
observation that C-H activation occurs preferentially at the terminal
carbon of linear alkanes in most systems that activate C-H bonds10
and supports the idea that activation at the methyl carbon occurs
9
4134
J. AM. CHEM. SOC. 2005, 127, 4134-4135
10.1021/ja044208m CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
1
due to other factors rather than a preference for the intermediate
alkane complex. It may also confirm why activation of ethane has
been found to be more rapid than for higher alkanes15,20 since ethane
contains only CH3 groups. Binding to methylenes of higher alkanes
requires a rearrangement process to occur to give the methyl-bound
isomer prior to C-H activation, reducing the rate.
methyl groups, with JCH values being 116.5 ( 0.5, 113.2 ( 1.0,
and 108.5 ( 2.5 Hz for 1-13C, 1-13C,d1, and 1-13C,d2, respectively.
Using the value of K ) 2.02, a model in which the bound C-H
1
1
has JCH ) 85 Hz and the unbound C-H bonds have JCH ) 132
Hz fits the data.
The IPR data indicate an asymmetric interaction. Combined with
the shielded 13C shift data, this suggests an η2-C,H interaction (type
I). This binding model is also supported by theoretical calculation.
Ab initio calculations on the ethane derivative, CpRe(CO)2(CH3-
CH3), show that one C-H bond is coordinated and elongated to
1.17 Å.24,25 Calculations have also predicted 2-propane complexes
to be more stable than 1-propane complexes in the case of W(CO)5-
(propane).26
The resonances in spectrum A show the same multiplicity and
3
similar JHH values to those in free pentane, indicative of alkane
In conclusion, the CpRe(CO)2 fragment binds to all three types
of C-H bonds in pentane with a slight preference for those in CH2
units. Binding to rhenium occurs primarily through one C-H bond
at any one instant, and this bond exchanges rapidly with the other
bonds in the CH2 or CH3 unit that is attached to the Re center.
Acknowledgment. We thank Dr. G. Edwards, Professor B.
Hibbert, A/Professor R. Read, and Dr. N. Roberts for helpful
discussions, and the Australian Research Council for funding.
Supporting Information Available: General Experimental. Prepa-
ration of labeled alkanes. NMR experimental. NMR spectra: 1H NMR
of CpRe(CO)2(pentane) and CpRe(CO)2(pentane-2,2,4,4-d4) at 163-
193 K; 1H NMR of CpRe(CO)2(cyclopentane-cis-1,2-d2); HSQC of
1-13C-d0-2; exchange (ROESY) spectrum of a mixture of 1, 2, and 3 at
complexes. That the bound CH2 resonances of 2 and 3 are shifted
to significantly lower frequency than the bound CH3 of 1 suggests
that just one hydrogen in each of these groups interacts with the
metal at one instant, and that these are time-averaged shifts. Hence,
each of the CH2 hydrogens interacts directly for 50% of the time
and each of the CH3 hydrogens for 33% of the time. Spectra of
complexes containing the Cp ligand17 are almost identical to those
containing the more soluble, hence, preferable, i-PrCp ligand. No
evidence of an agostic interaction involving the i-Pr group is seen.
Spectrum B shows the same experiment conducted using
pentane-2,2,4,4-d4. The resonance at δ -3.10 due to 2 is now absent
due to deuteration, confirming its assignment. The remaining res-
onances due to i-PrCpRe(CO)2(1-pentane-2,2,4,4-d4) and i-PrCpRe-
(CO)2(3-pentane-2,2,4,4-d4) are now broad singlets, confirming that
the protons on C2 were the source of the couplings for these
resonances in spectrum A. Spectrum C is acquired on the same
sample as B, and only 13C satellites were selected in this exper-
173 K. Calculation of K(C-H
for accurate extraction of integrals in H NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
for isotopomers of 1. Method
)
/C-D
bound bound
1
References
1
iment.17 Large JCH values, diagnostic for alkane complexes (1JCH
(1) Hall, C.; Perutz, R. N. Chem. ReV. 1996, 96, 3125.
(2) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.
(3) Sun, X.-Z.; Grills, D. C.; Nikiforov, S. M.; Poliakoff, M.; George, M. W.
J. Am. Chem. Soc. 1997, 119, 7521.
) 109 ( 3 Hz for 3), are seen. Spectrum D shows the result of
employing pentane-1-13C as solvent. Fifty percent of the bound methyl
(4) Childs, G. I.; Grills, D. C.; Sun, X. Z.; George, M. W. Pure Appl. Chem.
2001, 73, 443.
1
groups are 13C labeled and show the JCH splitting. The observed
1JCH ) 116.5 ( 0.5 Hz is only slightly reduced from that in free
(5) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120, 9953.
(6) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1999, 121, 6336.
(7) Castro-Rodriguez, I.; Nakai, H.; Gantzel, P.; Zakharov, L. N.; Rheingold,
A. L.; Meyer, K. J. Am. Chem. Soc. 2003, 125, 15734.
(8) Evans, D. R.; Drovetskaya, T.; Bau, R.; Reed, C. A.; Boyd, P. D. W. J.
Am. Chem. Soc. 1997, 119, 3633.
(9) Kelly, J. M.; Hermann, H.; Koerner Von Gustore, E. J. Chem. Soc., Chem.
Commun. 1973, 105.
(10) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507.
(11) Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 2437.
(12) Bengali, A. A.; Arndtsen, B. A.; Burger, P. M.; Schultz, R. H.; Weiller,
B. H.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. Pure Appl. Chem.
1995, 67, 281.
(13) Bromberg, S. E.; Yang, H.; Asplund, M. C.; Lian, T.; McNamara, B. K.;
Kotz, K. T.; Yeston, J. S.; Wilkens, M.; Frei, H.; Bergman, R. G.; Harris,
C. B. Science 1997, 278, 260.
(14) Flood, T. C.; Janak, K. E.; Iimura, M.; Zhen, H. J. Am. Chem. Soc. 2000,
122, 6783.
(15) Northcutt, T. O.; Wick, D. D.; Vetter, A. J.; Jones, W. D. J. Am. Chem.
Soc. 2001, 123, 7257.
pentane (124.7 Hz). Again, this coupling is an averaged value, con-
1
sistent with one of the three C-H bonds having a reduced JCH
,
averaged with two typical JCH values. The 13C NMR shift of the
bound carbon of 1 (δ -42.4) is highly shielded, suggesting a
significant metal-carbon interaction. Spectra E and F (same sample;
only 13C isotopomers are selected in F17) show the result of using
a mixture of 13C and 2H doubly labeled isotopomers, namely, pen-
tane-1-13C, pentane-1-13C,1-d1, and pentane-1-13C,1,1-d2. This
sample contains three types of 13C-labeled bound methyl groups,
13CH3, 13CH2D, and 13CHD2, allowing evaluation of deuterium
1
isotope effects on H and 13C chemical shifts and J-couplings. A
1
large isotopic perturbation of resonance (IPR)21-23 is observed for
the 1H shifts in these three isotopomers of 1, with δ -1.99, -2.64,
and -3.62 being observed for zero (1-13C), one (1-13C,d1), or two
(1-13C,d2) deuteriums incorporated, respectively. A correspondingly
large IPR is observed in CpRe(CO)2(cyclopentane-cis-1,2-d2) with
δ -4.09 for bound CHD groups versus δ -2.32 for bound CH2
groups.17 These effects are consistent with a rapid exchange of the
hydrogen within a methyl or methylene unit that is interacting with
the metal center and a shifting of equilibria when deuterium is incor-
porated in favor of isotopomers in which a C-H is bound to the
metal center rather than C-D. We calculate17,23 δ(bound) -8.22,
(16) Jones, W. D. Acc. Chem. Res. 2003, 36, 140.
(17) Details are given in the Supporting Information.
(18) Integration values were corrected for minor site-dependent excitation
effects; see Supporting Information. Errors are 95% confidence limits.
(19) Accurate quantitative, mechanistic studies are currently underway.
(20) McNamara, B. K.; Yeston, J. S.; Bergman, R. G.; Moore, C. B. J. Am.
Chem. Soc. 1999, 121, 6437.
(21) Saunders, M.; Jaffe, M. H.; Vogel, P. J. Am. Chem. Soc. 1971, 93, 2558.
(22) Saunders, M.; Telkowski, L.; Kates, M. R. J. Am. Chem. Soc. 1977, 99, 8070.
(23) Calvert, R. B.; Shapley, J. R. J. Am. Chem. Soc. 1978, 100, 7726.
(24) Yang, H.; Asplund, M. C.; Kotz, K. T.; Wilkens, M. J.; Frei, H.; Harris,
C. B. J. Am. Chem. Soc. 1998, 120, 10154.
(25) DFT calculations on the cyclopentane and pentane analogues also support
this view, and calculations on the methane analogue show that the barrier
to exchange of coordinated hydrogen on the same carbon is predicted to
be less than 2 kcal mol-1. Dance, I. G.; Harris, H.; Ball, G. E. Manuscript
in preparation. See also: Bergman, R. G.; Cundari, T. R.; Gillespie, A.
M.; Gunnoe, T. B.; Harman, W. D.; Klinckman, T. R.; Temple, M. D.;
White, D. P. Organometallics 2003, 22, 2331.
δ(unbound) 1.12, and an equilibrium constant, K(C-H
)
bound/C-Dbound
)
2.02 ( 0.06, for one of each type of bond in 2 and 3 and estimate
K ) 2.6 for a CHD group in cyclopentane complexes. This leads
to a preference of C-H over C-D bonds by 0.23 ( 0.03 kcal
mol-1 in 1-pentane complexes. Preferential binding of the C-H
bonds leads to a parallel reduction in 1JCH in the deuterated bound
(26) Zaric, S.; Hall, M. B. J. Phys. Chem. A 1997, 101, 4646.
JA044208M
9
J. AM. CHEM. SOC. VOL. 127, NO. 12, 2005 4135