Activation of carboxylic acids in asymmetric organocatalysis

Article abstract of DOI:10.1002/anie.201400169

Organocatalysis, catalysis using small organic molecules, has recently evolved into a general approach for asymmetric synthesis, complementing both metal catalysis and biocatalysis.1 Its success relies to a large extent upon the introduction of novel and generic activation modes.2 Remarkably though, while carboxylic acids have been used as catalyst directing groups in supramolecular transition-metal catalysis,3 a general and well-defined activation mode for this useful and abundant substance class is still lacking. Herein we propose the heterodimeric association of carboxylic acids with chiral phosphoric acid catalysts as a new activation principle for organocatalysis. This self-assembly increases both the acidity of the phosphoric acid catalyst and the reactivity of the carboxylic acid. To illustrate this principle, we apply our concept in a general and highly enantioselective catalytic aziridine-opening reaction with carboxylic acids as nucleophiles. Activation by dimerization: There is still no general activation mode for carboxylic acids in organocatalysis. The formation of heterodimers between chiral phosphoric acid diesters and carboxylic acids can be used to activate and direct reactivity of the latter in asymmetric reactions. This novel principle has been applied to the ring-opening desymmetrization and kinetic resolution of aziridines leading to valuable amino alcohols.

Full text of DOI:10.1002/anie.201400169

Angewandte
Chemie
DOI: 10.1002/anie.201400169
Organocatalysis
Activation of Carboxylic Acids in Asymmetric Organocatalysis**
Mattia Riccardo Monaco, Belꢀn Poladura, Miriam Diaz de Los Bernardos, Markus Leutzsch,
Richard Goddard, and Benjamin List*
Abstract: Organocatalysis, catalysis using small organic
molecules, has recently evolved into a general approach for
asymmetric synthesis, complementing both metal catalysis and
biocatalysis.[1] Its success relies to a large extent upon the
introduction of novel and generic activation modes.[2] Remark-
ably though, while carboxylic acids have been used as catalyst
directing groups in supramolecular transition-metal catalysis,[3]
a general and well-defined activation mode for this useful and
abundant substance class is still lacking. Herein we propose the
heterodimeric association of carboxylic acids with chiral
phosphoric acid catalysts as a new activation principle for
organocatalysis. This self-assembly increases both the acidity
of the phosphoric acid catalyst and the reactivity of the
carboxylic acid. To illustrate this principle, we apply our
concept in a general and highly enantioselective catalytic
aziridine-opening reaction with carboxylic acids as nucleo-
philes.
exist. We were intrigued about the nature of the electronic
structure of such a hypothetical heterodimer. On the one
hand, protonation of the carboxylic acid is expected to lower
its LUMO, generating a more reactive electrophile. On the
other hand, the well-established basicity of the oxygen atom
=
within the P O bond of the phosphoric acid catalyst should
lead to a partial deprotonation of the carboxylic acid, which
would increase the energy of the HOMO and provide a more
nucleophilic species. Potentially, both activation modes could
even occur simultaneously (Scheme 1).
C
arboxylic acids, including all amino acids, tartaric acid,
lactic acid, mandelic acid, and pyruvic acid, to name just a few,
are ubiquitous molecules in life. They feature both a hydro-
gen-bond donor and a hydrogen-bond acceptor, which can
stabilize each other by self-association. The dimerization of
carboxylic acids was first observed at the beginning of the
20th century and is now well appreciated to be a general
phenomenon.[4] In the middle of the last century, a similar
interaction was observed for phosphoric acid diesters and the
strength of the dimerization was found to be even larger for
Scheme 1. Activation of carboxylic acids by heterodimerization with
phosphoric acids.
The concept proposed here originated from the observa-
tion that in contrast to simple and small phosphoric acid
diesters, which readily form dimers,[8] their sterically much
more demanding binol-derived congeners are generally
monomeric in solution. Presumably, their dimerization is
sterically hindered. Dimerization also appears to be difficult
in the solid state.[9] For example, in the crystal structure of the
popular phosphoric acid catalyst TRIP, a water molecule is
required to bridge the dimer. On this basis, we hypothesized
that the heterodimerization with a small carboxylic acid,
which could enter the catalytic pocket in the absence of
repulsive forces, might be a highly favorable process.
Indeed, initial NMR studies supported our hypothesis (for
details, see the Supporting Information). A stoichiometric
amount of benzoic acid significantly shifted the signals of
TRIP in the 1H NMR spectrum. In addition, the signal of the
phosphoric acid in the 31P NMR spectrum was shifted
considerably downfield. To gain further evidence of the
suggested heterodimerization, diffusion-ordered spectrosco-
py (DOSY) measurements were performed.[10] The transla-
tional diffusion coefficient of the two molecules significantly
decreased upon self-assembly. As expected, the effect on the
smaller carboxylic acid is larger due to the higher variation of
its hydrodynamic volume (Figure 1a).
=
these species on account of the dipolar nature of the P O
bond and their higher acidity.[5,6] In the meantime, chiral
binaphthol-derived phosphoric acid diesters have been used
as remarkably general asymmetric Brønsted acid catalysts
that activate not only imines but also certain carbonyl
compounds and related substrates.[7] Interestingly, carboxylic
acids have not previously been utilized as substrates in
asymmetric phosphoric acid catalysis. We speculated that in
mixtures of phosphoric acids and carboxylic acids, in addition
to the two expected homodimers, an equilibrium favoring the
corresponding and previously undescribed heterodimer may
[*] M. R. Monaco, B. Poladura, Dr. M. Diaz de Los Bernardos,
M. Leutzsch, Dr. R. Goddard, Prof. Dr. B. List
Max-Planck-Institut fꢀr Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Mꢀlheim an der Ruhr (Germany)
E-mail: list@kofo.mpg.de
[**] Generous support by the Max-Planck-Society and the European
Research Council (Advanced Grant “High Performance Lewis Acid
Organocatalysis, HIPOCAT”) is gratefully acknowledged. We thank
the members of our mass spectrometry department for their
excellent service.
We were also able to determine the binding isotherms of
the association of TRIP with benzoic acid and with acetic acid
by following the shift of the phosphorus signal in a 31P NMR
titration experiment (Figure 1b). The constants Ka were
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201400169.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
served as the actual catalyst (heteroconjugation
phenomenon).[14–16] Moreover, as the association
has not previously been recognized, its real poten-
tial, especially for the activation of carboxylic acids,
has remained untapped. We were particularly
intrigued by the upfield shift of the proton signals
in the NMR spectrum of the carboxylic acids. This
appears to indicate the somewhat counterintuitive
possibility that heterodimerization with a phosphoric
acid raises the energy of the HOMO. In other words,
the (stronger) phosphoric acid would formally
behave like a base within the dimer. Thus, upon
self-assembly a synergistic activation may take
place: the nucleophilicity of the carboxylic acid
and the acidity of the catalyst are both increased.
This observation suggested to us the possibility of
designing a catalytic system in which the heterodi-
meric species acts as a source of nucleophilic
carboxylic acids (Scheme 2).
Specifically, we decided to investigate whether
the system could facilitate an asymmetric ring
opening of aziridines to afford protected amino
alcohols. Molecules containing a vicinal amino
alcohol moiety are interesting synthetic targets,
due to their abundance in naturally occurring
molecules and pharmacologically active compounds,
and their application as chiral ligands, auxiliaries,
and catalysts in asymmetric transformations.[17,18]
Stereoselective transformations of aziridines have
Figure 1. Physical characterization: a) 2D-DOSY analysis; b) Binding isotherm
determination; c) Crystal structure of TRIP·AcOH. TRIP-HD=heterodimer of TRIP
with a carboxylic acid.
determined by a nonlinear regression approach in which the
heterodimer was treated as a 1:1 host–guest system.[11] The
measured Ka values (TRIP·BzOH Ka = (3981 Æ 98)mÀ1
,
TRIP·AcOH Ka = (1948 Æ 26)mÀ1) are consistent with this
kind of interaction and lie between the binding constants for
the homodimerization of carboxylic acids (10–100mÀ1) and
those of phosphoric acids (usually above 105 mÀ1). The
stronger association with benzoic acid may be explained on
the basis of the dual contribution of the aromatic ring, which
increases the availability of the lone pairs on the carbonyl
moiety and the acidity of the hydroxy group. A final
confirmation of heterodimer formation was obtained by
cocrystallizing acetic acid and TRIP, which gave a crystal
suitable for X-ray diffraction. The resulting structure (Fig-
ure 1c) shows that this interaction is not only favored in
solution, but also in the solid state.
Scheme 2. A catalytic cycle that utilizes TRIP·AcOH as a chiral and
nucleophilic source of AcOH.
Interestingly, carboxylic acids have previously been used
as additives in chiral phosphoric acid catalyzed asymmetric
reactions. In 2006, the Akiyama group described the benefi-
cial effect of a stoichiometric amount of acetic acid in a TRIP-
catalyzed aza-Diels–Alder reaction without further investi-
gating its role.[12] Later Rueping and co-workers described
a Mannich reaction, in which acetic acid was crucial for the
reactivity and suggested that its role would be to promote the
enolization of the nucleophile, while the chiral phosphoric
acid was suggested to activate the electrophile.[13]
been widely investigated in the last few years but an
asymmetric catalytic conversion that leads to amino alcohols,
that is, a hydrolytic kinetic resolution or desymmetrization,
has not yet been developed.[19,20] We therefore anticipated
that the designed reaction would be an ideal testing ground
for our new carboxylic acid activation.
Initially, we focused our attention on the development of
a desymmetrization strategy for meso-aziridines with benzoic
acid as the nucleophile and the designed system was indeed
found to be effective in the transformation. We selected an
electron-poor benzamide moiety with two nitro groups in the
Although other explanations cannot be excluded, we
propose here that in both cases the more acidic heterodimer
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
ortho and para positions as the protecting group since the
simple one-step deprotection of both ester and amide
moieties of the products would afford the “naked” 1,2-
amino alcohols. Optimization of the reaction conditions (see
the Supporting Information) revealed that the transformation
proceeds smoothly when the association between the phos-
phoric acid catalyst and the carboxylic acid is promoted by an
excess of the carboxylic acid. Degradation of the catalyst by
means of direct attack of the phosphate on the aziridine was
observed when equilibrium was not reached.[21] With seven
equivalents of benzoic acid, ring opening of the six-mem-
bered-ring-fused aziridine 1a to give the expected protected
amino alcohol 2a occurred with quantitative yield and
essentially perfect enantioselectivity (98% yield, e.r. >
99.5:0.5, Scheme 3).
Compound 2e, bearing only one nitro group on the benza-
mide moiety, required lower reaction temperature (À108C)
and more dilute conditions.
Acyclic substrates were also tolerated although with
a slight loss in stereocontrol. Protected amino alcohols 2 f,g
could be obtained in excellent yield and good enantiocontrol
when the reaction was simply performed under more dilute
conditions. Notably, product 2h was also delivered smoothly
from the challenging cis-stilbene-derived aziridine 1h in good
yield and excellent stereoselectivity. The reaction also toler-
ates other carboxylic acids. For example, acetic acid gives
similarly high yield and enantioselectivity (products 2i and
2j). The longer reaction time required and higher nucleophile
concentration may suggest that the affinity between TRIP
and the carboxylic acid plays a role in the transformation.
Chiral terminal aziridines are considered linchpins for the
synthesis of important building blocks such as 1,2-amino
alcohols, 1,2-diamines, and 1,2-amino thiols.[22] Therefore,
having shown the potential of TRIP/carboxylic acid hetero-
dimers in the desymmetrization of meso-aziridines, we
decided to also apply our activation concept to the kinetic
resolution of racemic terminal aziridines. It is noteworthy that
in this transformation both the product and the unreacted
starting material are potentially useful.
Several different cyclic and acyclic meso-aziridines 1a–h
were successfully employed in the transformation. The
corresponding cyclic products 2a–e were generally obtained
with good to excellent yields and enantioselectivities. Impor-
tantly, the methodology seems to be only slightly influenced
by the ring strain and the size of the starting materials.
We found that our catalyst system was indeed also
effective in this reaction. Lowering the temperature to
À308C and employing terminal aziridines with only one
nitro group in the benzamide protecting group, we could
obtain the desired products and the unreacted starting
materials with excellent selectivity factors (S = 37–51,
Table 1). The preliminary scope of this transformation is
shown in Table 1. Not only the linear-chain aziridines 3a–c
proved to be suitable for the resolution, but even the
branched compound 3d reacted smoothly.
As expected for ring-opening reactions occurring in acidic
media, the nucleophilic attack exclusively occurs on the
internal carbon of the aziridine substrate. The reason for the
perfect regioselectivity may be due to a partial positive charge
on the aziridine fragment in the transition state (vide infra)
and no regioisomeric product could be detected in any case.
On the basis of the results obtained during our inves-
tigation, a catalytic cycle can be proposed (Scheme 4).
Accordingly, the initial equilibrium between the free phos-
phoric acid and the heterodimer is shifted to the right due the
high concentration of the carboxylic acid. This saturation of
TRIP suppresses its degradation through reaction with the
aziridine. Thus, the carboxylic acid plays a dual role in the
reaction by both performing the ring opening as well as
protecting the catalyst from degradation. The heterodimer
then engages in the nucleophilic attack, which possibly occurs
through a concerted transition state, in which the aziridine is
subject to additional Brønsted acid activation. While the
heterodimer has in principle two different possible nucleo-
philic sites, the carboxylate and the phosphate, the reactivity
of the phosphate moiety is sterically hindered. Based on the
experimental results we speculate that the reaction might
occur by means of an asynchronous SN2 pathway. The
presence of a carbocationic species in the catalytic cycle is
unlikely because of the exclusivity of the trans product in the
Scheme 3. Reactions were performed on a 0.2 mmol scale. Substrate
1a was reacted on a 1 mmol scale. Substrate 1e was reacted at
À108C. The loading of AcOH was increased to 10 equivalents in the
reactions to products 2i,j. Enantiomeric ratios were determined by
HPLC on a chiral stationary phase.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Kinetic resolution of terminal aziridines.[a]
d+ charge at the reacting carbon center (for mechanistic
investigations see the Supporting Information).
As suggested above, the success of organocatalysis relies
on the possibility to design novel transformations based on
established activation modes. Our activation of carboxylic
acids is not an exception and it is potentially broadly
applicable. A related catalytic system has been developed
concurrently in our laboratories for the first asymmetric
organocatalytic carboxylysis of epoxides.[23]
In conclusion a new concept for the activation of
carboxylic acids in asymmetric catalysis has been designed
and realized, which involves their heterodimerization with
chiral sterically hindered phosphoric acids. This interaction
may open new perspectives by not only providing a new
highly acidic motif for asymmetric Brønsted acid catalysis, but
also by providing a generic activation mode for carboxylic
acids in organocatalysis. The system was shown to be effective
in the first highly enantioselective ring opening of aziridines
to afford protected amino alcohols. Further investigations to
define the scope of this catalyst system are currently in
progress in our laboratories.
Entry Aziridine
Product
Yield [%] e.r.[b]
S[c]
48
42
92.5:7.5
1
48
98:2
49
46
94:6
2
3
50
51
95.5:4.5
51
44
92.5:7.5
98.5:1.5
Experimental Section
Catalytic asymmetric synthesis of 2a: In a dried screw-cap vial
a solution of benzoic acid (855 mg, 7 equiv) and TRIP (60 mg,
8 mol%) was prepared in chloroform (6 mL). The mixture was stirred
at the reaction temperature for 20 min and then a solution of the
aziridine 1a (291 mg, 1 mmol) in chloroform (2 mL) was added at
once. When the reaction was determined to be complete (thin-layer
chromatography), the reaction mixture was diluted with hexanes/
MTBE (1:1) and the products were directly purified by column
chromatography. For complete experimental details and character-
ization of compounds, see the Supporting Information.
28
58
96:4
4
37
71.5:28.5
[a] Reactions were performed on 0.1 mmol scale in 0.016m solution.
[b] Determined by HPLC on a chiral stationary phase. [c] Measured by
Kagan’s equation.
Received: January 7, 2014
Revised: March 6, 2014
Published online: && &&, &&&&
Keywords: amino alcohols · aziridines · carboxylic acids ·
.
organocatalysis · self-assembly
[1] B. List, J. W. Yang, Science 2006, 313, 1584 – 1585.
[2] D. W. C. MacMillan, Nature 2008, 455, 304 – 308.
ˇ
[3] a) T. Smejkal, B. Breit, Angew. Chem. 2008, 120, 317 – 321;
Angew. Chem. Int. Ed. 2008, 47, 311 – 315; b) J. Meeuwissen,
J. N. H. Reek, Nat. Chem. 2010, 2, 615 – 621.
[4] N. V. Sidgwick, Inorg. Chem. Ann. Rep. 1933, 30, 114 – 115.
[5] D. F. Peppard, J. R. Ferraro, G. W. Mason, J. Inorg. Nucl. Chem.
1957, 4, 371 – 372.
[6] J. DeFord, F. Chu, E. V. Anslyn, Tetrahedron Lett. 1996, 37,
1925 – 1928.
[7] a) D. Kampen, C. M. Reisinger, B. List, Top. Curr. Chem. 2010,
291, 395 – 456; b) R. J. Phipps, G. L. Hamilton, F. D. Toste, Nat.
Chem. 2012, 4, 603 – 614; c) M. Terada, Synthesis 2010, 1929 –
1982; d) T. Akiyama, J. Itoh, K. Fuchibe, Adv. Synth. Catal. 2006,
348, 999 – 1010.
Scheme 4. Proposed catalytic cycle for the TRIP·RCO2H-mediated ring
opening of aziridines.
[8] D. F. Peppard, J. R. Ferraro, J. W. Mason, Inorg. Nucl. Chem.
1958, 7, 231 – 244.
[9] a) M. Klussmann, L. Ratjen, S. Hoffmann, V. Wakchaure, R.
Goddard, B. List, Synthesis 2010, 2189 – 2192; b) M. Rueping, E.
desymmetrization, while the complete regioselectivity of the
kinetic resolution indicates the presence of a strong localized
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Sugiono, C. Azap, T. Theissmann, M. Bolte, Org. Lett. 2005, 7,
3781 – 3783.
[10] K. F. Morris, C. S. Johnson, Jr., J. Am. Chem. Soc. 1992, 114,
3139 – 3141.
[11] L. Fielding, Tetrahedron 2000, 56, 6151 – 6170.
[12] T. Akiyama, Y. Tamura, J. Itoh, H. Morita, K. E. Fuchibe, Synlett
2006, 141 – 143.
[13] a) M. Rueping, C. Azap, Angew. Chem. 2006, 118, 7996 – 7999;
Angew. Chem. Int. Ed. 2006, 45, 7832 – 7835; b) M. Rueping, E.
Sugiono, F. R. Schoepke, Synlett 2007, 1441 – 1445.
[14] A. Kꢀtt, I. Leito, I. Kaljurand, L. Soovꢁli, V. M. Vlasov, L. M.
Yagupolskii, I. A. Koppel, J. Org. Chem. 2006, 71, 2829 – 2838.
[15] H. Yamamoto, K. Futatsugi, Angew. Chem. 2005, 117, 1958 –
1977; Angew. Chem. Int. Ed. 2005, 44, 1924 – 1942.
[16] a) G. M. Barrow, J. Am. Chem. Soc. 1956, 78, 5802 – 5806; b) F.
Chu, L. S. Flatt, E. V. Anslyn, J. Am. Chem. Soc. 1994, 116,
4194 – 4204.
[19] a) S. E. Larson, J. C. Baso, G. Li, J. C. Antilla, Org. Lett. 2009, 11,
5186 – 5189; b) B. Wu, J. R. Parquette, T. V. RajanBabu, Science
2009, 326, 1662; c) B. Wu, J. C. Gallucci, J. R. Parquette, T. V.
RajanBabu, Angew. Chem. 2009, 121, 1146 – 1149; Angew.
Chem. Int. Ed. 2009, 48, 1126 – 1129; d) K. Ohmatsu, Y.
Hamajima, T. Ooi, J. Am. Chem. Soc. 2012, 134, 8794 – 8797;
e) T. Mita, E. N. Jacobsen, Synlett 2009, 1680 – 1684; f) J.
Cockrell, C. Wilhelmsen, H. Rubin, A. Martin, J. B. Morgan,
Angew. Chem. 2012, 124, 9980 – 9983; Angew. Chem. Int. Ed.
2012, 51, 9842 – 9845.
[20] J. L. Jat, M. P. Paudyal, H. Gao, Q.-L. Xu, M. Yousufuddin, D.
Devarajan, D. H. Ess, L. Kꢀrti, J. R. Falck, Science 2014, 343, 61 –
65.
[21] N. D. Shapiro, V. Rauniyar, G. L. Hamilton, J. Wu, F. D. Toste,
Nature 2011, 470, 245 – 250.
[22] W. McCoull, F. A. Davis, Synthesis 2000, 1347 – 1365.
[23] M. R. Monaco, S. Prevost, B. List, Angew. Chem. 2014, DOI:
10.1002/ange.201400170; Angew. Chem. Int. Ed. 2014, DOI:
10.1002/anie.201400170.
[17] S. C. Bergmeier, Tetrahedron 2000, 56, 2561 – 2576.
[18] J. A. Birrell, E. N. Jacobsen, Org. Lett. 2013, 15, 2895 – 2897.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Organocatalysis
M. R. Monaco, B. Poladura,
M. Diaz de Los Bernardos, M. Leutzsch,
R. Goddard, B. List*
&&&& — &&&&
Activation of Carboxylic Acids in
Asymmetric Organocatalysis
Activation by dimerization: There is still
no general activation mode for carboxylic
acids in organocatalysis. The formation of
heterodimers between chiral phosphoric
acid diesters and carboxylic acids can be
used to activate and direct reactivity of
the latter in asymmetric reactions. This
novel principle has been applied to the
ring-opening desymmetrization and
kinetic resolution of aziridines leading to
valuable amino alcohols.
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!