Direct observation of molecular preorganization for chirality transfer on a catalyst surface

Article abstract of DOI:10.1126/science.1208710

The chemisorption of specific optically active compounds on metal surfaces can create catalytically active chirality transfer sites. However, the mechanism through which these sites bias the stereoselectivity of reactions (typically hydrogenations) is gen

Full text of DOI:10.1126/science.1208710

Direct Observation of Molecular Preorganization for Chirality Transfer
on a Catalyst Surface
Vincent Demers-Carpentier et al.
Science 334, 776 (2011);
DOI: 10.1126/science.1208710
This copy is for your personal, non-commercial use only.
If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of June 4, 2012 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/334/6057/776.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2011/11/10/334.6057.776.DC1.html
This article cites 38 articles, 2 of which can be accessed free:
http://www.sciencemag.org/content/334/6057/776.full.html#ref-list-1
This article appears in the following subject collections:
Chemistry
http://www.sciencemag.org/cgi/collection/chemistry
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2011 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

REPORTS
12. Z. Chen, J. G. Bohnet, S. R. Sankar, J. Dai,
J. K. Thompson, Phys. Rev. Lett. 106, 133601 (2011).
13. J. J. Bollinger, W. M. Itano, D. J. Wineland, D. J. Heinzen,
Phys. Rev. A 54, R4649 (1996).
14. M. J. Holland, K. Burnett, Phys. Rev. Lett. 71, 1355 (1993).
15. A. Kuzmich, L. Mandel, Quantum Semiclass. Opt. 10,
493 (1998).
16. Z. Y. Ou, J.-K. Rhee, L. J. Wang, Phys. Rev. Lett. 83, 959 (1999).
17. P. Walther et al., Nature 429, 158 (2004).
18. M. W. Mitchell, J. S. Lundeen, A. M. Steinberg, Nature
429, 161 (2004).
19. I. Afek, O. Ambar, Y. Silberberg, Science 328, 879 (2010).
20. V. Meyer et al., Phys. Rev. Lett. 86, 5870 (2001).
21. D. Leibfried et al., Nature 438, 639 (2005).
22. L.-M. Duan, A. Sørensen, J. I. Cirac, P. Zoller,
Phys. Rev. Lett. 85, 3991 (2000).
32. P. Hyllus, L. Pezzé, A. Smerzi, Phys. Rev. Lett. 105,
120501 (2010).
33. P. Hyllus, O. Gühne, A. Smerzi, Phys. Rev. A 82, 012337
(2010).
perspectives for highly sensitive measurements
with a new generation of atom interferometers.
Bücker et al. (34) recently reported reduced
atom number fluctuations in twin-atom beams,
and the group of M. Oberthaler has independent-
ly used spin dynamics in Bose-Einstein conden-
sates to generate atomic two-mode entanglement
detected by a homodyning technique.
34. R. Bücker et al., Nat. Phys. 7, 608 (2011).
Acknowledgments: We thank E. Rasel for stimulating
discussions and P. Zoller for helpful comments. We
acknowledge support from the Centre for Quantum
Engineering and Space-Time Research (QUEST), the
European Science Foundation (EuroQUASAR), and the
Danish National Research Foundation Center for
Quantum Optics. P.H. acknowledges financial support
of the European Research Council Starting Grant
GEDENTQOPT. This work was supported in part by
Provincia Autonoma di Trento within the activities of the
BEC center. L.P. acknowledges support by the Laboratoire
Charles Fabry de L’Institut d’ Optique where part of this
work was completed.
References and Notes
1. N. F. Ramsey, Phys. Rev. 78, 695 (1950).
2. C. Bordé, Phys. Lett. A 140, 10 (1989).
3. L. Essen, J. V. L. Parry, Nature 176, 280 (1955).
4. R. Wynands, S. Weyers, Metrologia 42, S64 (2005).
5. G. Santarelli et al., Phys. Rev. Lett. 82, 4619 (1999).
6. L. Pezzé, A. Smerzi, Phys. Rev. Lett. 102, 100401
(2009).
23. H. Pu, P. Meystre, Phys. Rev. Lett. 85, 3987 (2000).
24. T. Kim, O. Pfister, M. J. Holland, J. Noh, J. L. Hall,
Phys. Rev. A 57, 4004 (1998).
7. J. Appel et al., Proc. Natl. Acad. Sci. U.S.A. 106,
10960 (2009).
8. J. Estève, C. Gross, A. Weller, S. Giovanazzi, M. K. Oberthaler,
Nature 455, 1216 (2008).
9. C. Gross, T. Zibold, E. Nicklas, J. Estève, M. K. Oberthaler,
Nature 464, 1165 (2010).
10. I. D. Leroux, M. H. Schleier-Smith, V. Vuletić,
Phys. Rev. Lett. 104, 073602 (2010).
11. M. F. Riedel et al., Nature 464, 1170 (2010).
25. T. Nagata, R. Okamoto, J. L. O’brien, K. Sasaki,
Supporting Online Material
S. Takeuchi, Science 316, 726 (2007).
www.sciencemag.org/cgi/content/full/science.1208798/DC1
26. C. Klempt et al., Phys. Rev. Lett. 104, 195303 (2010).
27. S. M. Barnett, D. T. Pegg, Phys. Rev. A 42, 6713 (1990).
28. M. Scherer et al., Phys. Rev. Lett. 105, 135302 (2010).
29. Supporting material is available on Science Online.
30. C. Klempt et al., Phys. Rev. Lett. 103, 195302 (2009).
31. R. A. Campos, B. E. A. Saleh, M. C. Teich, Phys. Rev. A 40,
1371 (1989).
SOM Text
Figs. S1 to S5
References (35–40)
23 May 2011; accepted 23 September 2011
Published online 13 October 2011;
10.1126/science.1208798
the modifier, and Pt(111) as the metal surface.
Preorganization in this chemisorption system can
be described in terms of molecular recognition
between a fully chiral molecule and a molecule
that is chiral only by virtue of its confinement
on the surface (4, 9–15). The experiment resolves
interactions at the level of functional groups at
individual modifier sites and thereby provides
insight into the structural preorganization that
biases the stereochemical outcome of the reac-
tion, in essence the mechanism of chirality trans-
fer. The scanning tunneling microscopy (STM)
Direct Observation of Molecular
Preorganization for Chirality
Transfer on a Catalyst Surface
Vincent Demers-Carpentier,1 Guillaume Goubert,1 Federico Masini,1 Raphael Lafleur-Lambert,1
Yi Dong,1 Stéphane Lavoie,1 Gautier Mahieu,1 John Boukouvalas,1 Haili Gao,2
Anton M. H. Rasmussen,2 Lara Ferrighi,2 Yunxiang Pan,2 Bjørk Hammer,2* Peter H. McBreen1*
The chemisorption of specific optically active compounds on metal surfaces can create catalytically
active chirality transfer sites. However, the mechanism through which these sites bias the stereoselectivity measurements were carried out at room temper-
of reactions (typically hydrogenations) is generally assumed to be so complex that continued progress in ature in parallel with room temperature catalytic
the area is uncertain. We show that the investigation of heterogeneous asymmetric induction with
single-site resolution sufficient to distinguish stereochemical conformations at the submolecular
level is finally accessible. A combination of scanning tunneling microscopy and density functional
theory calculations reveals the stereodirecting forces governing preorganization into precise
chiral modifier-substrate bimolecular surface complexes. The study shows that the chiral modifier
induces prochiral switching on the surface and that different prochiral ratios prevail at different
submolecular binding sites on the modifier at the reaction temperature.
measurements on the asymmetric hydrogena-
tion of TFAP to 2,2,2-trifluorophenylethanol over
(R)-NEA modified Pt on an alumina support.
The interactions of the chiral modifier and
the prochiral substrate with the metal surface
were first studied separately. Two modifier mo-
tifs, present in a 7:3 ratio, are observed in the STM
images (Fig. 1, A and B). The two motifs may
hirality transfer and amplification on of sought-after advantages, including ease of be distinguished by the position of the bright
surfaces is fundamentally important to separation of the catalyst from the product. The protrusion: It extends from the central region of
progress in the synthesis of enantiopure best-explored examples in terms of potential ap- the image in the majority motif (A) and from
C
compounds for pharmaceutical and agrochem- plications involve the hydrogenation of activated the left-hand side in the minority motif (B). The
ical applications (1–4) and to discussions of the ketones to chiral alcohols and hydroxy esters on two lowest energy structures predicted by den-
origin of biological homochirality (5). Chirally cinchona-modified Pt (6, 7). A number of syn- sity functional theory (DFT) are the (R)-NEA-1
modified heterogeneous catalysts, in which the thetic modifiers that share some key structural and (R)-NEA-2 conformers (Fig. 1, E and F).
reaction is stereocontrolled at surface sites formed and functional characteristics of cinchonidine, Their calculated adsorption energies are ~2 eV
by adsorbing a chiral molecule, offer a number such as 1-(1-naphthyl)ethylamine (NEA) and its with a pronounced 0.13 eV preference for the
condensation derivatives (8), can also be used. (R)-NEA-1 conformation (figs. S10 and S11), sug-
However, a lack of mechanistic understanding gesting that their relative surface coverages are
has long been seen as an impediment to the ra- determined in part by adsorption dynamics. Sur-
tional development of chirally modified hetero- face vibrational spectroscopy measurements (figs.
1Centre de recherche sur les propriétés des interfaces et la
catalyse (CERPIC) and Département de chimie, Université Laval,
Québec, QC G1V 0A6, Canada. 2Interdisciplinary Nanoscience
Center (iNANO) and Department of Physics and Astronomy,
Aarhus University, DK 8000 Aarhus, Denmark.
geneous catalysts (6, 7).
S1 and S2) confirm that the NEA conformers are
We directly characterized the surface com- chemisorbed in the geometry found in the DFT
plexation of a prochiral reagent with a chiral calculations. In particular, the C-CH3 bond is
modifier by using 2,2,2-trifluoroacetophenone, nearly perpendicular to the surface, as also de-
TFAP, as the prochiral substrate, (R)-(+)-NEA as scribed for the Pd(111) surface (16).
*To whom correspondence should be addressed. E-mail:
hammer@phys.au.dk (B.H.); peter.mcbreen@chm.ulaval.
ca (P.H.M.)
776
11 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org

REPORTS
Fig. 1. STM images and DFT-
calculated structures of the single
enantiomer modifier, (R)-NEA, and
the prochiral substrate, TFAP, chem-
isorbed separately on Pt(111) at
room temperature. The modifier,
(R)-(+)-1-(1-naphthyl)ethylamine,
is imaged in two distinct forms in
a 7-to-3 ratio. These two confor-
mations [(A) and (B)], (R)-NEA-
1 and (R)-NEA-2, differ in whether
the NH2 group points to the edge
or the center of the two-ring
aromatic framework [(E) and
(F)]. The STM image (C) and DFT-
calculated structure (G) of 1-(1-
anthryl)ethylamine are used to
identify the bright protrusion as
the ethylamine group. Chemi-
sorption of the substrate, 2,2,2-
trifluoroacetophenone, leads to
the formation of homochiral dimers
(17). A pro-S dimer and the cor-
responding calculated structure
are shown in (D) and (H), respec-
tively. STM images were recorded
at room temperature, 1.2-V sample
bias, and 0.3-nA tunnel current.
A
B
C
D
1 nm
1 nm
1 nm
CH3
CH3
CH3
H
H
NH2
H
NH2
H
NH2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
1 nm
H
H
7 : 3
E
F
G
H
Fig. 2. STM image of the chemisorption system
formed by adding TFAP to (R)-NEA–modified Pt(111).
(A) Large scan (11.5 nm by 5.5 nm) image, taken
with the sample held at room temperature, shows
that TFAP forms both dimers and modifier-substrate
complexes. The distribution of geometries in the dia-
stereomeric complexes, as determined from a visual
inspection, is plotted in (C) and (D) in terms of the
angular position (q) and orientation (f) of TFAP with
respect to (R)-NEA, as defined in (B). A color code is
used to emphasize the existence of five distinct abun-
dant families, three (M1 to M3) involving (R)-NEA-1
and two (M4 and M5) involving (R)-NEA-2. STM im-
ages were recorded at 1.2-V bias and 0.3-nA current.
A
B
(°)
(°)
C
D
150
100
50
150
100
50
M
3
M
5
0
0
M
1
M
4
-50
-100
-150
-50
-100
-150
M
2
(R)-NEA-1
(R)-NEA-2
-100
-50
0
50
100
150
-100
-50
0
50
100
150
(°)
(°)
www.sciencemag.org SCIENCE VOL 334 11 NOVEMBER 2011
777

REPORTS
Upon chemisorption, the substrate forms di- would yield (S)-2,2,2-trifluorophenylethanol fol- substrate complexes in terms of TFAP’s angular
mers, and each TFAP is imaged (Fig. 1D) as a lowing attack by adsorbed atomic hydrogen at position (q) and orientation (f) with respect to
bowling-pin motif in which the small brighter the enantioface in direct contact with the metal the modifier (as defined in Fig. 2B) shows tight
protrusion is caused by the trifluoroacetyl sub- surface.
groupings of the experimental structures into five
stituent. DFT calculations predict a binding en- When both the modifier and the substrate distinct geometries (M1 to M5; Fig. 2, C and D).
ergy of ~1 eV per TFAP monomer and suggest are present on the surface, STM images taken The corresponding STM motifs are shown in Fig.
that dimers are stabilized (17) by two simulta- at room temperature show a dynamic system 3. By using DFT calculations (Fig. 3 and figs.
neous aryl-CH…OC interactions in a homochiral where, in addition to forming dimers, TFAP S17 and S18), we explored the best possible (R)-
conformation (Fig. 1H). Because the DFT cal- binds to (R)-NEA. Video measurements (movie NEA+TFAP complexes on the Pt surface. Three
culations establish the directionality of the car- S1) reveal that the dimers disassemble and reas- of the four most stable (R)-NEA-1+TFAP com-
bonyl groups within each dimer, the prochirality semble on the time scale of the experiment and plexes and the two most stable (R)-NEA-2+TFAP
of dimerized TFAP can be determined by a vi- that individual TFAP molecules form short-lived complexes found by DFT are included in Fig. 3.
sual inspection of the STM images. Pro-S dimers complexes (seconds to a few minutes) with the These calculated structures match with the five
are shown in Fig. 1, D and H, and additional ex- modifier. The ratio of (R)-NEA-1 to (R)-NEA-2 most abundant families observed by STM both
amples of pro-R and pro-S dimers are presented is conserved upon adding the substrate. At the in terms of the location of the TFAP around
in fig. S6. In keeping with the objective of chi- coverage used in this study, about 60% of the the ethylamine group and the direction in which
rality transfer to the hydrogenation product, the NEAs are interacting with a TFAP molecule at any it points (see fig. S19 for simulated STM im-
pro-S configuration is defined as the one that given time. A visual inspection of these modifier- ages). We note that one additional rather stable
M1
M2
M3
M4
M5
34%
18%
13%
10%
6%
-0.21 eV
-0.28 eV
-0.22 eV
-0.31 eV
-0.28 eV
Fig. 3. STM images and DFT-calculated structures of the five most frequently ob-
served modifier-substrate diastereomeric complexes, M1 to M5. The indicated
percentages are relative to the total population of 900 complexes. The DFT-
calculated structures establish the prochirality of the substrate in families M1 to
M5. The numbers below the calculated structures are the potential energies relative
to the separated substrate and modifier (figs. S16 and S19). STM images were
recorded at a 1.2-V bias and 0.3-nA tunnel current, with the sample at room
temperature.
γ
β
α
Pro-S
Pro-R
Pro-S
Pro-R
Pro-S
Pro-R
Mirror Counterpart
Mirror Counterpart
Mirror Counterpart
M1
M3
M2
Fig. 4. DFT-calculated structures for mirror-image complexes formed by bind- reflection operation relative to the dashed line gives the mirror counterpart
ing TFAP to (R)-NEA-1 on Pt(111). For each binding site (a, b, and g), two stable [b-(R)]. The same symmetry operation applied to all of the Mi groups [plus
complexes of inverse prochirality are found. Dashed lines indicate the mirror a possible g-(R) species for (R)-NEA-2] reveals the 12 families used for the sta-
planes perpendicular to the surface that bring pro-R TFAP into pro-S TFAP and tistical analysis presented in Fig. 5. Red arcs indicate stereodirecting repulsive
vice versa. For example, by taking the M3 family (left-hand side of the middle interactions for the low population families. According to the energy calcula-
panel) where a pro-S TFAP binds at the top of the ethylamine group [b-(S)], a tions, the mirror counterparts are about 0.1 to 0.3 eV less stable (fig. S20).
778
11 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org

REPORTS
(R)-NEA-1+TFAP complex is identified with
A statistical analysis of over 900 complexes NH…OC interaction and a simple set of repul-
DFT (complex X1 in fig. S17) but does not cor- (Fig. 5) shows that we are observing a strong sive forces in the low population families a-(S),
respond to any of the abundant families ob- expression of regiospecific prochiral differen- b-(R), and g-(S). Simulation of the coadsorption
served in the STM images, suggesting that it tiation. For example, the most frequently ob- of (R)-NEA-1 and trifluoromethyl 2-pyridyl ke-
undergoes facile conversion into one of the ob- served complex is formed by pro-R TFAP at the tone to form b-(R) and b-(S) complexes evens
served structures. The excellent overall corre- a position of (R)-NEA-1 corresponding to fam- out the energy difference between the two
spondence between experiment and theory allows ily M1 (Figs. 2 and 3). It accounts for 34% of prochiralities (fig. S21), revealing that a repul-
us to discuss the imaged complexes in terms all of the complexes, whereas its pro-S coun- sive interaction between the amine group of the
of the calculated structures. In particular, it re- terpart accounts for 3%. This 11-to-1 pro-R– modifier and the phenyl moiety of TFAP is the
veals that the five most abundant families all to–pro-S ratio is evidence of strong prochiral reason for the low populations of b-(R) and g-(S)
form via NH2…OC interactions. Hence, the pro- steering at the a site. Further, 1-to-6 and 6-to-1 families. Similarly, the a-(S) family is disfav-
chirality of TFAP substrate in each of the com- pro-R–to–pro-S ratios are measured at the b and ored relative to a-(R) because of steric interaction
plexes can be determined by visual inspection g sites, respectively. The overall stereoselectiv- between the phenyl group of the substrate and the
of the STM images.
ity of (R)-NEA-1 is good, with a ratio of 3 to naphthyl group of the modifier. Phenyl-naphthyl
Labeling the five most abundant families 1 (obtained from adding the TFAP populations repulsion also causes the absence of a g-(R)/(R)-
in terms of the combination of the prochirality at all three binding sites for this conformer), NEA-2 family. Our observations thus identify re-
of TFAP (pro-R or pro-S) and the position where whereas for (R)-NEA-2 a less-pronounced 3- pulsive interactions with the phenyl group as the
it binds to the ethylamine function (a, b, and g to-2 ratio is observed.
main stereodirecting force at the origin of the
for the left-hand, top, and right-hand sites, re- In order to test the actual enantioselectivity of enantioselectivity for the NEA-TFAP pair. This
spectively) gives a-(R) for M1 and M4, b-(S) for (R)-NEA–modified Pt for TFAP hydrogenation, subtle steric effect occurs in the presence of the
M3 and M5, and g-(R) for M2. It then becomes we carried out catalytic measurements using binding NH2…OC interaction and the yet stronger
natural to look for less abundant families as Pt/Al2O3 at room temperature. Indeed, enantio- chemisorption interaction of TFAP.
complexes of inverse TFAP prochirality at the selection is observed with an enantiomeric ex-
Importantly for a chirality transfer process
given sites, that is, to look for a-(S), b-(R), and cess of ~34% (corresponding to an enantiomeric (18, 19), the relative abundances of TFAP surface
g-(S) families (Fig. 4). It turns out that ~93% of ratio of roughly 2:1) in favor of the (R)-product. enantiomers averaged over ensembles of coexist-
all complexes in a sample of 900 can be accounted This moderate expression of asymmetry is in ing TFAP dimers and TFAP/(R)-NEA complexes
for by including the less abundant mirror-image line with our STM measurements on (R)-NEA– reveal that the modifier induces prochiral switch-
families. The remaining nonassigned complexes modified Pt(111), also at room temperature, ing of the adsorbed substrate. There is an overall
consist mainly of TFAP interacting with the naph- identifying predominantly pro-R preorganization excess of pro-R TFAP that increases with the cov-
thyl (q < 0 data points, Fig. 2C) instead of the and a smaller but significant fraction of pro-S erage of the modifier. Symmetry breaking is ob-
ethylamine group and complexes resulting from preorganization.
served to occur irrespective of the sequence in
high local coverage. Termolecular complexes of The classification of STM images, calculated which the substrate and the modifier are added
one (R)-NEA and two TFAPs are observed where structures, and energies (Figs. 3 and 5) gives de- to the surface. When the surface is first exposed
both the a and g sites are simultaneously occu- tailed information on the diastereomeric complexes to TFAP and then to (R)-NEA to a surface cov-
pied (fig. S5), implying that the ethylamine group inducing chirality transfer from the modifier to erage of roughly 1/3 monolayer of (R)-NEA and
is in the same conformation for both binding the substrate. The DFT calculations (Fig. 4) re- 1/3 monolayer of TFAP (as shown in Fig. 2), a
events, in agreement with the DFT calculations veal that the observed prochiral excesses can be pro-R excess up to 15 T 2% is observed. For ex-
(Fig. 3, M1 and M2).
explained by the balance between an attractive periments performed by first adsorbing (R)-NEA
β
β
α
CH 3
CH3
CH3
CH3
CH3
CH3
γ
α
γ
H
NH2
H
NH2
H
H
NH2
H
NH2
H
NH2
NH2
H
H
H
H
H
H
H
H
H
H
Total
66%
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
O
CF3
M1
M2
M4
Pro-R
34%
2%
18%
10%
2%
0%
1%
F C
O
3
27%
M3
M5
Pro-S
3%
13%
3%
1%
6%
Fig. 5. Classification of 900 modifier-substrate complexes formed by (R)-NEA and population it represents. The remaining 7% mostly consist of complexes where
TFAP on Pt(111) at room temperature into distinct stereochemical and regiochem- TFAP binds to the naphthyl instead of the ethylamine group and irregularly shaped
ical arrangements. Each entry shows a 2 nm–by–2 nm STM image of an archetypal complexes likely caused by high local coverages. STM images were recorded
complex of a particular family, along with the percentage of the total complex at a 1.2-V bias and 0.3-nA current with the sample at room temperature.
www.sciencemag.org SCIENCE VOL 334 11 NOVEMBER 2011
779

REPORTS
13. L. Ning, S. Haq, G. R. Darling, R. Raval, Angew. Chem.
Int. Ed. 40, 7613 (2007).
14. M. Lingenfelder et al., Angew. Chem. Int. Ed. 46,
4492 (2007).
15. S. Blankenburg, W. G. Schmidt, Phys. Rev. Lett. 99,
196107 (2007).
16. L. Burkholder, M. Garvey, M. Weinert, W. T. Tysoe,
J. Phys. Chem. C 115, 8790 (2011).
17. V. Demers-Carpentier et al., J. Phys. Chem. C 115,
1355 (2011).
18. R. R. Knowles, E. N. Jacobsen, Proc. Natl. Acad. Sci. U.S.A.
107, 20678 (2010).
19. A. Zehnacker, M. A. Suhm, Angew. Chem. Int. Ed. 47,
6970 (2008).
and then dosing TFAP on the modified surface for the NEA-TFAP pair a higher enantiomeric
with similar coverages, pro-R excesses of up excess could be achieved by molecular design
to 16 T 3% are observed. This behavior can be to limit the chiral pocket to one binding site,
explained simply by low barrier rotation of the rather than the multiple sites observed. Efforts
trifluoroacetyl group followed by preferential could be directed toward blocking the counter-
capture of pro-R TFAP to form diastereomer- productive b site and/or favoring chemisorption
ic complexes at free binding sites around the of NEA-1 over the less-selective NEA-2 con-
ethylamine group of the modifier. The observed former. In future work, STM/DFT-guided asym-
symmetry breaking is a direct manifestation of metric catalyst design could be applied to other
the action of (R)-NEA as a chiral modifier of the modifier substrate pairs and a broader range of
Pt(111) surface.
reactions.
The previous lack of direct information on
modifier-substrate stereodirecting interactions
has been the major roadblock in the comprehen-
sion and optimization of chirally modified het-
erogeneous asymmetric catalysts. As this study
demonstrates, efficient stereodirection occurs
through subtle energy differences in the presence
of stronger ones, including chemisorption bonding.
It is extraordinarily difficult to identify the origin
of these differences, and their relative impor-
tance, in the absence of direct measurements.
The combination of STM and DFT used here
to describe chirality transfer preorganization
in terms of regiospecific prochiral ratios is a
key step forward. It can serve as a basis for
profound studies of the kinetic relevance of in-
dividual submolecular sites as well as for studies
of the stereodynamics of heterogeneous asymmetric
induction. Our model-free analysis suggests that
Acknowledgments: This work was supported by a Natural
Sciences and Engineering Research Council of Canada
(NSERC) discovery grant, a Fonds Québécois de la
Recherche sur la Nature et les Technologies (FQRNT)
team grant, Canadian Foundation for Innovation (CFI)
grants, and the FQRNT Centre in Green Chemistry
and Catalysis (CCVC). This work was in part supported
by the Lundbeck Foundation, the Danish Research
Councils, and the Danish Center for Scientific
Computing. V.D.-C. acknowledges an NSERC graduate
student scholarship.
References and Notes
1. G. Kyriakou, S. K. Beaumont, R. M. Lambert, Langmuir
27, 9687 (2011).
2. M. Forster, M. S. Dyer, M. Persson, R. Raval, Top. Catal.
54, 13 (2011).
3. A. J. Gellman, ACS Nano 4, 5 (2010).
4. A. G. Trant, C. J. Baddeley, J. Phys. Chem. C 115, 1025
(2011).
5. K. H. Ernst, Orig. Life Evol. Biosph. 40, 41 (2010).
6. H.-U. Blaser, M. Studer, Acc. Chem. Res. 40, 1348
(2007).
7. M. Heitbaum, F. Glorius, I. Escher, Angew. Chem. Int. Ed.
45, 4732 (2006).
8. E. Orglmeister, T. Mallat, A. Baiker, Adv. Synth. Catal.
347, 78 (2005).
9. J. M. Bonello, F. J. Williams, R. M. Lambert, J. Am.
Chem. Soc. 125, 2723 (2003).
10. Q. Chen, N. V. Richardson, Nat. Mater. 2, 324
(2003).
11. S. Weigelt et al., Nat. Mater. 5, 112 (2006).
12. A. Kühnle, T. R. Linderoth, B. Hammer, F. Besenbacher,
Nature 415, 891 (2002).
Supporting Online Material
www.sciencemag.org/cgi/content/full/334/6057/776/DC1
Materials and Methods
SOM Text
Figs. S1 to S19
Tables S1 to S4
References (20–38)
Movie S1
20 May 2011; accepted 19 September 2011
10.1126/science.1208710
surfaces, as well as the kinetics of N2 reduction
(3). These studies have shown that iron is predom-
inantly in the zero oxidation state in the catalyst.
Importantly, the rate-limiting step of the catalytic
process is N2 chemisorption and N-N bond cleav-
age to give surface-bound nitrides (N3–), which
react with H2 to form the N-H bonds in NH3
(Fig. 1). However, many details of the bond-
breaking and bond-forming steps are difficult to
discern by using surface science techniques, for
example, how many iron atoms are involved in
N2 Reduction and Hydrogenation
to Ammonia by a Molecular
Iron-Potassium Complex
Meghan M. Rodriguez,1 Eckhard Bill,2 William W. Brennessel,1 Patrick L. Holland1*
The most common catalyst in the Haber-Bosch process for the hydrogenation of dinitrogen (N2)
to ammonia (NH3) is an iron surface promoted with potassium cations (K+), but soluble iron
complexes have neither reduced the N-N bond of N2 to nitride (N3–) nor produced large amounts the N-N dissociation step and whether the K+
of NH3 from N2. We report a molecular iron complex that reacts with N2 and a potassium reductant
to give a complex with two nitrides, which are bound to iron and potassium cations. The product has sociation step (4). Similar uncertainties surround
a Fe3N2 core, implying that three iron atoms cooperate to break the N-N triple bond through a
postulated mechanisms for N2 reduction by nitro-
six-electron reduction. The nitride complex reacts with acid and with H2 to give substantial yields of genase enzymes, which have iron (in the +2 and
promoter interacts with N2 during the N-N dis-
N2-derived ammonia. These reactions, although not yet catalytic, give structural and spectroscopic
insight into N2 cleavage and N-H bond-forming reactions of iron.
+3 oxidation states) at their active sites (5, 6).
Reactions of iron coordination compounds
with N2 are of interest because the products can
he Haber-Bosch process is a large-scale process (iron, ruthenium, osmium, uranium, and shed light on Fe-N2 interactions in atomic detail.
method for catalytic reduction of dinitro- cobalt-molybdenum), but iron has received the Although the behavior of iron atoms in a co-
gen (N2) with dihydrogen (H2) to give most industrial and theoretical attention because ordination compound (where they have a posi-
T
ammonia (NH3). This process is vital for pro- of its hardiness and low cost (2–4). The most tive formal charge) is inevitably different than the
duction of synthetic fertilizer, which is needed to common iron catalyst is promoted by K+ addi- zero valent iron atoms on a metallic surface, co-
produce food for the world’s expanding popula- tives and was developed by Mittasch in 1910 (2). ordination compounds are more amenable to de-
tion (1). Several metals catalyze the Haber-Bosch In the century since then, chemists have studied tailed structural characterization. Unfortunately,
the nature of active iron and iron/potassium iron coordination complexes are poor at reduc-
1Department of Chemistry, University of Rochester, Rochester,
NY 14627, USA. 2Max-Planck-Institut für Bioanorganische
Chemie, Stiftstrasse 34-36, D-45470 Mülheim an der Ruhr,
Germany.
Fig. 1. Simplified scheme
of the ammonia formation
pathway in the Haber-Bosch
process.
*To whom correspondence should be addressed. E-mail:
holland@chem.rochester.edu
780
11 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org