Finding the proton in a key intermediate of anti-markovnikov alkyne hydration by a bifunctional catalyst

Article abstract of DOI:10.1021/ja803106z

The secondary structure of a bifunctional catalyst positions a crucial reactive proton in the final intermediate of anti-Markovnikov alkyne hydration to give an aldehyde. NMR coupling and isotopic labeling studies elucidate the location of this proton and its involvement in hydrogen bonding. Copyright

Full text of DOI:10.1021/ja803106z

Published on Web 07/24/2008
Finding the Proton in a Key Intermediate of anti-Markovnikov Alkyne
Hydration by a Bifunctional Catalyst
Douglas B. Grotjahn,*,† Elijah J. Kragulj,† Constantinos D. Zeinalipour-Yazdi,‡
Valent´ın Miranda-Soto,† Daniel A. Lev,† and Andrew L. Cooksy†
Department of Chemistry and Biochemistry, 5500 Campanile DriVe, San Diego State UniVersity,
San Diego, California 92182-1030, and Heterogeneous Catalysis Laboratory, Department of Chemistry,
UniVersity of Cyprus, Nicosia, CY 1678 Cyprus
Received April 27, 2008; E-mail: grotjahn@chemistry.sdsu.edu
Scheme 1. Formation of Key Intermediate 6
Many of Nature’s metalloenzyme active sites feature a metal
ion or ions and a variety of acidic or basic groups, which act in
concert, allowing enzymes to achieve turnover frequencies rarely
rivaled by synthetic catalysts.1 A major contributor to enzymatic
rate accelerations is secondary structure, with a network of hydrogen
bond or proton donors and acceptors that lower the energy of
intermediates or transition states. We are interested in the increased
activity and selectivity of small-molecule bifunctional catalysts2
and their secondary structure. Catalysts for anti-Markovnikov
hydration of terminal alkynes to give aldehydes3a–e or for alkene
isomerization3f are made 1000-10 000 times faster using bifunc-
tional phosphine ligands. However, the same features that increase
the efficiency of bifunctional catalysts also complicate studies of
their mechanism. We are developing methods to study these
complex systems.4 Here, we have unambiguously determined the
location and hydrogen bonding of a reactive proton in a bifunctional
catalyst structure. Our methods should be useful in other studies
of bifunctional catalysis.
Complex 3 and its imidazolyl analog 3-Im (Scheme 1) were
synthesized5 from 1 and 1-Im3b,6 by a route similar to that reported
recently for an analog.4 In the pyridyl case, on removal of solvent
and water, nearly complete formation of 4 is seen, because the
complex is in equilibrium with water and 3. With the ligands shown,
under anhydrous conditions, 4 or 4-Im could also be obtained
directly as the major product. When a terminal alkyne is added to
pyridyl species 3 or 4 at -40 °C, vinylidene 5 (R ) H) forms
within 3 h. An intermediate alkyne π-complex4 is not detected with
this phosphine ligand set, even under what would be optimal
conditions,7 with acetylene, the smallest alkyne.
When pyridyl derivative 3 is allowed to react with acetylene
and water (4 equiv) at 0 °C for 4 h, up to 91% of an acyl complex
(A) is observed. Possible tautomeric structures 6, 7, and 8 (R )
H) are collectively referred to as A. Acyl complexes have been
identified in other alkyne hydrations,8 but these were not syntheti-
cally useful catalytic systems. The acyl complex (R ) H) is
somewhat stable near 0 °C but at higher temperatures cleanly
releases CH3CHO, forming 3 or 4, depending on the amount of
water. The fact that A accounts for almost all of the catalyst added
means that turnover under the conditions described is determined
by the rate at which A releases product. This makes the identity of
A and its structure and reactivity of special interest.
The 1H NMR data for A revealed a broadened singlet near δ 19
ppm,9a,b suggesting a 1H nucleus involved in hydrogen bonding as
shown in tautomers 6-8. When the alkyne H13C13CH was used,
this resonance remained unchanged, but intense 13C NMR signals
1
near9a,b δ 293 (broad) and 50 ppm (d, JCC ) 22.3 Hz9b) were
now seen. The chemical shift of the downfield carbon fits better a
hydroxycarbene complex (7) than an acyl.10 However, we note that
the few literature comparisons available are between neutral acyl
† San Diego State University.
‡ University of Cyprus.
9
10860 J. AM. CHEM. SOC. 2008, 130, 10860–10861
10.1021/ja803106z CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
(like 9) and cationic hydroxy- or alkoxycarbene complexes
(resembling 7, but without the intramolecular hydrogen bond). To
probe the effect of protonation, to A-(13C)2 was added Et3N (3
equiv), forming 9-(13C)2 with 13C shifts9a of 259.4 and 53.0 ppm
coupling and isotopic perturbation are expected to have applicability
in other catalytic reactions.
Acknowledgment. We thank the NSF for continuing support
under CHE 0415783 and 0719575, Cambridge Isotope Laboratories,
Inc. for labeled building blocks, and Dr. LeRoy Lafferty for assisting
with NMR experiments.
1
(two d, JCC ) 18.2 Hz). For 9, υCO ) 1567 cm-1. Notably, 9
could be isolated, unlike A which releases aldehyde above 0 °C.
The NH-N unit in 8 resembles that proposed for a hydride
complex with ligands like 1 but without the t-butyl groups.11 Our
DFT calculations suggest that because of the t-Bu groups, 8 is
energetically unfavorable, with either 6 or 7 being favored
depending on the methods used.12 Experimental data arguing against
8 include the large changes in 13C NMR and IR data for the acyl
moiety accompanying deprotonation of A-(13C)2.
Supporting Information Available: Details of compound prepara-
tion and characterization and calculations. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Lippard, S. J.; Berg, J. M. Principles of Bioinorganic Chemistry, University
Science Books: Mill Valley, CA, 1994.
In order to determine which tautomer of A formed, ligand 1-15N
was synthesized and incorporated into the various intermediates
shown in Scheme 1. Of particular interest, the 15N chemical shifts
for 1-15N, 2-, 3-, and 5-(15N)2 are -55.7, -59.1, -80.7, and -59.8
ppm, respectively.9c,d These data show that ligand coordination at
P or overall charge of a complex do not significantly affect the
15N shift (1, 2, 5) whereas hydrogen bonding does (3 vs 2 and 5).
(2) Reviews:(a) van den Beuken, E. K.; Feringa, B. L. Tetrahedron 1998, 54,
12985. (b) Rowlands, G. J. Tetrahedron 2001, 57, 1865. (c) Clapham, S. E.;
Hadzovic, A.; Morris, R. H. Coord. Chem. ReV. 2004, 248, 2201. (d)
Grotjahn, D. B. Chem.-Eur. J. 2005, 11, 7146. (e) Ikariya, T.; Murata,
K.; Noyori, R. Org. Biomol. Chem. 2006, 4, 393. (f) Natale, D; Mareque-
Rivas, J. C. Chem. Commun. 2008, 425. (g) Das, S.; Brudvig, G. W.;
Crabtree, R. H. Chem. Commun. 2008, 413.
(3) (a) Grotjahn, D. B.; Incarvito, C. D.; Rheingold, A. L. Angew. Chem., Int.
Ed. 2001, 40, 3884. (b) Grotjahn, D. B.; Lev, D. A. J. Am. Chem. Soc.
2004, 126, 12232. (c) Labonne, A.; Kribber, T.; Hintermann, L. Org. Lett.
2006, 8, 5853. (d) Hintermann, L.; Labonne, A. Synthesis 2007, 1121. (e)
Kribber, T.; Labonne, A.; Hintermann, L. Synthesis 2007, 2809. (f) Grotjahn,
D. B.; Larsen, C. R.; Gustafson, J. L.; Nair, R.; Sharma, A. J. Am. Chem.
Soc. 2007, 129, 9592.
(4) Grotjahn, D. B.; Miranda-Soto, V.; Kragulj, E. J.; Lev, D. A.; Erdogan,
G.; Zeng, X.; Cooksy, A. L. J. Am. Chem. Soc. 2008, 130, 20.
(5) See Supporting Information for full experimental and computational details.
(6) Baur, J.; Jacobsen, H.; Burger, P.; Artus, G.; Berke, H.; Dahlenburg, L.
Eur. J. Inorg. Chem. 2000, 1411.
(7) Lomprey, J. R.; Selegue, J. P. J. Am. Chem. Soc. 1992, 114, 5518.
(8) (a) Bianchini, C.; Casares, J. A.; Peruzzini, M.; Romerosa, A.; Zanobini,
F. J. Am. Chem. Soc. 1996, 118, 4585. (b) Ogo, S.; Uehara, K.; Abura, T.;
Watanabe, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2004, 126, 16520.
(9) Solvents used: (a) THF-d8; (b) acetone-d6; (c) CDCl3; (d) CD2Cl2. Of
particular note, the data for complex 6 in solvents of different polarity were
quite similar, consistent with an intramolecular hydrogen bond.
(10) (a) Bruce, M. I.; Swincer, A. G. Aust. J. Chem. 1980, 33, 1471. (b)
Esteruelas, M. A.; Gomez, A. V.; Lahoz, F. J.; Lopez, A. M.; Onate, E.;
Oro, L. A. Organometallics 1996, 15, 3423.
1
Significantly, the H NMR peak for coordinated water protons in
3-(15N)2 appears as a slightly broadened singlet,13a consistent with
location of the water protons on O and not on N.
In sharp contrast, the 15N NMR data for A-(15N)2(13C)2 at -100
°C9b show two signals at very different chemical shifts (-63.6 and
-146.8 ppm). The large upfield shift of the latter resonance shows
protonation at one nitrogen (6 or 8), as does the 1H NMR spectrum
at -100 °C9a showing the downfield NH resonance as a doublet of
large magnitude (1JHN ) 56.8 Hz).13b However, the value of 1JNH
is only 62% of that in the model salt 10-15N made by protonating
1-15N (1JHN ) 92.3 Hz,9d consistent with related data14), which
underscores the presence of hydrogen bonding in 6.
The foregoing data conclusively place a hydrogen-bonded proton
on one nitrogen. To show that the proton indeed interacts with the
acyl oxygen, A-(13C)2 was formed as usual, using H2O in THF-d8.
Then, D2O was added to create a mixture of A and its N-deuterated
isotopomer. A remarkably large15 perturbation of the 13C chemical
shift for the acyl carbon of 1.6 ppm was observed, unequivocally
demonstrating an O-HN interaction as in 6.15
(11) Jalon, F. A.; Manzano, B. R.; Caballero, A.; Carrion, M. C.; Santos, L.;
Espino, G.; Moreno, M. J. Am. Chem. Soc. 2005, 127, 15364.
(12) The predicted stability of 6 equaled that of 7 only with explicit quantum
treatment of the phenyl groups, and with a triple-ꢀ basis set for atoms in
and adjacent to the H-bond. In 3-Im, where the heavy atom geometry is
determined from x-ray diffraction,3a the computed geometry is in good
agreement with experiment. At the optimized 6 geometry (RNH ) 1.09 Å,
1
RHO ) 1.48 Å), the predicted JHN value is -68.6 Hz. The lower
experimental magnitude (56.8 Hz) is consistent with a larger vibrationally
averaged N-H separation, as zero-point motion carries the proton towards
the O atom.
Having identified 6 under near-stoichiometric conditions of high
catalyst loading and low water content, we looked for its presence
during catalyzed alkyne hydration using 3 or 11 on Ph(CH2)3C13CH.
Indeed, hydration of this alkyne in d6-acetone (0.25 M) using 5
(13) (a) The 1H and 15N NMR data5 for the imidazole analog 3-Im-(15N)2 made
with 1-Im-15N are similar, showing no observable coupling between 15N
and water protons; moreover, X-ray data for the triflate salt of 3-Im3a show
the H2O protons on O not N. (b) Interestingly, there is evidence of a
fluxional process by which the NH proton may be transferred between the
two nitrogens, perhaps via 7. As the sample of 6-(15N)2 is warmed above
-80 °C, the 15N peaks broaden to the point of invisibility, whereas the 1H
resonance becomes more complicated between -80 and -40 °C, and a
triplet (J ) 25.5 Hz)9a at higher temperatures. The changes in 1H spectra
were modeled as a dynamic AA′X-A2X system (X ) H, A ) 15N), line-
shape analysis giving Ea ) 8 kcal mol-1 for the fluxional process.13c
1
equiv of water and 4 mol % of catalyst revealed 13C, 31P, and H
NMR peaks for 6-13C [R ) (CH2)3Ph], along with peaks for the
vinylidene intermediate 5-13C preceding it.
Here for the first time in a practical catalytic system for anti-
Markovnikov alkyne hydration, we have identified an acyl inter-
mediate (6). More significantly, we have conclusively shown the
special role of bifunctional ligands: the proton needed for catalyst
turnover is located on one pyridine base, and this pyridinium moiety
donates a hydrogen bond to the acyl oxygen. Ongoing work seeks
to clarify role of the bifunctional ligands in formation of 6 and in
its release of the aldehyde product, where intramolecular proton
transfer is likely to be crucial. Elucidation of the roles of proton
transfer or hydrogen bonding with NMR techniques such as 15N
Similarly, the 31P signals for 6 (δ 60.7 and 51.9 ppm at -100 °C8b
)
coalesced near -55 °C. (c) The structure of 3-Im (X ) OTf)3a shows
slightly different O-H-N bonds. Low-temperature spectra of 3 and 3-Im
show two 31P and 15N NMR signals, the latter very close (with in 1 ppm
of each other),5 suggesting slightly different hydrogen bonding environ-
ments. Line-shape analyses give Ea ) 7 and 10 kcal mol-1 for 3 and 3-Im,
respectively, consistent with a rocking motion of the two ligands.
(14) Andreeva, D. V.; Ip, B.; Gurinov, A. A.; Tolstoy, P. M.; Denisov, G. S.;
Shenderovich, I. G.; Limbach, H.-H. J. Phys. Chem. A 2006, 110, 10872.
(15) Vidossich, P.; Piana, S.; Miani, A.; Carloni, P. J. Am. Chem. Soc. 2006,
128, 7215.
JA803106Z
9
J. AM. CHEM. SOC. VOL. 130, NO. 33, 2008 10861