Xanthene-4,5-diamine derivatives: A study of anion-binding catalysis

Article abstract of DOI:10.1016/j.tetlet.2015.07.058

This study describes the synthesis of a class of anion-binding catalysts based on a xanthene scaffold. Both unsymmetrical catalysts and C₂-symmetrical catalysts were generated, and were examined in the cyclization of 3- and 2-substituted furans onto N-acyliminium ions. Good conversion for each reaction was observed with a variety of anion-binding catalysts (42-76%).

Full text of DOI:10.1016/j.tetlet.2015.07.058

Tetrahedron Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Xanthene-4,5-diamine derivatives: a study of anion-binding catalysis
⇑
⇑
Alison E. Metz , Kailasham Ramalingam, Marisa C. Kozlowski
Department of Chemistry, Roy and Diana Vagelos Laboratories, University of Pennsylvania, Philadelphia, PA 19104, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
This study describes the synthesis of a class of anion-binding catalysts based on a xanthene scaffold. Both
unsymmetrical catalysts and C2-symmetrical catalysts were generated, and were examined in the
cyclization of 3- and 2-substituted furans onto N-acyliminium ions. Good conversion for each reaction
was observed with a variety of anion-binding catalysts (42–76%).
Received 8 May 2015
Revised 16 July 2015
Accepted 18 July 2015
Available online xxxx
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
Anion-binding catalyst
Asymmetric organocatalysis
Xanthene-4,5-diamine
Furan cyclization
N-acyliminium ions
Introduction
primary functional group found in organocatalysts used for coun-
teranion binding.
Ion-pairing chemistry has emerged as a useful form of catalysis
and relies on electrostatic interactions.1,2 There are two methods of
anion-pairing catalysis that have been utilized in the literature
(Fig. 1). The first is anion-binding catalysis, which is explored in
this paper. The second is chiral anion-directed catalysis and it
varies from the former by having the chiral center attached to
the anion rather than non-covalently bound to it. These catalyst
types are still in their early stages of development and relatively
few types have been explored.
We hypothesized that other types of known anion receptors
should have the ability to form good anion-binding catalysts.6
Acridones and similar anthracenyl scaffolds are well known in the
literature for their ability to bind to anions. Xanthene-derived
compounds have been used as hydrogen-bonding catalysts to pro-
mote the addition of 2-acetylcyclopentanone into a,b-unsaturated
nitroalkenes.7 This paper describes the investigation of
xanthene-derived compounds as anion-binding catalysts.
Jacobsen and Seidel have been the leading pioneers in the field
of organocatalysis involving anion-binding catalysis.3,4 The work in
this paper was inspired by Jacobsen’s Pictet–Spengler-type
reaction of hydroxylactams (Scheme 1). Here Jacobsen was able
to effect indole addition into N-acyliminium ions in good yield
(97%) and good enantiomeric excess (97%) using a thiourea catalyst
(TBME = tert-butyl methyl ether).4a A variety of experiments were
Results and discussion
4,5-Diaminoxanthene 3 was accessed in straightforward man-
ner requiring three steps with
a high overall yield (75%,
Scheme 2).8 Additionally, compound 3 can be mono-Boc protected
or coupled to one amino acid to yield compound 5. This reactivity
enables unsymmetrical catalysts to be made.
carried out to support this mechanism. Not only was
a
Using these conditions, various catalysts could be made expedi-
tiously, including unsymmetrical catalysts 6–8 and C2-symmetrical
catalysts 9 and 10 (Fig. 2). C2-Symmetrical catalysts are beneficial
because of the smaller number of possible diastereomeric transi-
tion states available.9 Although urea compounds are less acidic
and therefore weaker hydrogen-bond donors, they have been
found to give a greater amount of enantioinduction in certain sys-
tems.7 As such, both urea 7 and thiourea (6 and 8) were made. A
greater variety of amides are accessible using commercially avail-
able amino acids 4, but sulfonyl amides were hypothesized to be
better hydrogen-bond donors (again due to their increased acid-
ity).10 Thus, amides (6 and 7) and sulfonyl amide 8 were formed.
pronounced anion affect on enantioselectivity observed but also
rate enhancement was detected when tertiary alcohols are
utilized. Jacobsen extended this chemistry to an intermolecular
reaction4b and an intramolecular version with pyrroles.4c The
exceptional anion-binding properties of thioureas have been
known for quite some time.5 Because of this feature, they are the
⇑
Corresponding authors. Tel.: +1 215 898 3048; fax: +1 215 573 7165.
E-mail addresses: alimetz@sas.upenn.edu (A.E. Metz), marisa@sas.upenn.edu
(M.C. Kozlowski).
http://dx.doi.org/10.1016/j.tetlet.2015.07.058
0040-4039/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Metz, A. E.; et al. Tetrahedron Lett. (2015), http://dx.doi.org/10.1016/j.tetlet.2015.07.058

2
A. E. Metz et al. / Tetrahedron Letters xxx (2015) xxx–xxx
With structurally different compounds in hand, their catalytic
activity was evaluated using Jacobsen’s Pictet–Spengler-type reac-
tion (refer to Supporting information for further details). In this
study, thiourea 6a and urea 7 performed similarly. Literature sup-
ports that N-alkyl, N-aryl thioureas and ureas have a very similar
affinity for a chloride anion (Keq = 22 and 21 in DMSO, respectively)
indicating the activity of these catalysts is a function of their
affinity for a chloride anion rather than their pKa (13 and 19,
respectively).11 Bisthiourea 9 and thiourea 6a performed better
than a simple thiourea, 1-(3,5-bis(trifluoromethyl)phenyl)-3-phe
nylthiourea. These experiments suggest that four and three
hydrogen bonds, respectively, are superior for catalysis. Since high
selectivities have been reported for this reaction in the literature,
this substrate was not tested for enantiomeric excess.
Figure 1. Two methods of anion-pairing catalysis. (A) Anion-binding catalysis. (B)
Chiral anion-directed catalysis.
To determine the basis of the increased catalytic activity of the
xanthene-derived compounds relative to simple thioureas in the
Jacobsen’s Pictet–Spengler-type reaction, their binding constants
to chloride were measured. Binding studies on similar compounds
have been performed.6 Although xanthenyl diamide 10 was found
to have the lowest equilibrium constant of the compounds
measured for a chloride binding, it is still favorable in less polar
solvents such as pyridine. Figure 3 shows the change in chemical
shift of the NH protons for 10 (0.272 M in py-d5) in the 1H NMR
spectrum with increasing equivalents of chloride anion. Figure 4
displays the corresponding binding curve. Since only one peak is
observed, this system is undergoing fast exchange between 10
and complex 10ÁClNn-Bu4; an average of the bound NH protons
and unbound NH protons is observed rather than two distinct
peaks. The immediate change in chemical shift after addition of
chloride anion (t = 1 h) indicates equilibrium has been reached
before data collection (t = 1–24 h).
Scheme 1. Jacobsen’s Pictet–Spengler-type reaction of hydroxylactams and pro-
posed mechanism.
These data allowed calculation of Keq = 56 for binding of 10 to
chloride in pyridine (Table 1). This equilibrium constant corre-
sponds to a
D
G = À2.39 kcal/mol. The equilibrium constant of this
reaction was too low to be measured in DMSO. Next, analysis was
performed on 6b and 7 since they were predicted to be the next
strongest anion-binders. In line with other literature reports,11
these two catalysts have very similar binding constants, Keq = 127
versus 159 for 6b and 7 respectively (
D
G = À2.87 and À3.00 kcal/-
mol) in DMSO (Table 1). This similarity is supported by the
reactivity that we observed in Jacobsen’s Pictet–Spengler-type
reaction (refer to Supporting information for details). The steric
hindrance of the tert-leucine may decrease the affinity of
compound 6b toward chloride.
Finally, analysis was performed on 9. As predicted, this catalyst
has the highest affinity for a chloride anion of the compounds stud-
ied (Keq = 1517,
D
G = À4.34 kcal/mol) in DMSO. A summary of
these binding constants compared to other neutral, organic, anion
binders can be found in Table 1.12 The NH groups responsible for
chloride binding are highlighted in red.
Scheme 2. Summary of route to 4,5-diaminoxanthene followed by peptide
coupling and thiourea formation.
Given the good binding affinity of the xanthene-based catalysts
to chloride, we wanted to expand this Pictet–Spengler-type
reaction to more challenging substrates. We decided to investigate
the cyclization of 2 and 3-substituted furans (25 and 23, respec-
tively) onto N-acyliminium ions. To the best of our knowledge,
there are no reported enantioselective or organocatalytic methods
for these transformations.13
The cyclization of 3-substituted furans is slower (entry 4, 42%
conversion after 7 h at À42 °C) than the corresponding indole
(44% conversion after 1.5 h at À55 °C). This difference arises from
the lower nucleophilicity of the furan relative to indole.
Unsymmetrical sulfonamide thiourea
8 performs better than
amide urea 7 (entries 3 and 4, 60% vs 42% conversion) and this
result is proposed to be a consequence of sulfonamides being more
acidic than amides (pKa 16 vs 23 in DMSO, respectively).10
tert-Leucine-derived catalyst 6b showed small but significant
Figure 2. Anion-binding compounds used in this study. Ar = (CF3)2–C6H3.
Please cite this article in press as: Metz, A. E.; et al. Tetrahedron Lett. (2015), http://dx.doi.org/10.1016/j.tetlet.2015.07.058

A. E. Metz et al. / Tetrahedron Letters xxx (2015) xxx–xxx
3
Figure 3. Change in NH protons of 10, 0.0272 M, by 1H NMR with increasing equivalents of chloride anion, ClNn-Bu4, in py-d5.
Table 1
Summary of binding constants in this study and the literature
Figure 4. Binding curve for 10, 0.0272 M, with increasing equivalents of chloride
anion, ClNn-Bu4 in py-d5.
levels of enantioinduction (entry 5, 70% conversion and 17% ee).
Bisamide 10 (entry 6) did not promote the cyclization, in line with
its low affinity for a chloride anion.
We were uncertain if furan 25 would undergo cyclization given
the decreased stability of the positive charge in its non-aromatic
intermediate relative to furan 23. Cyclization does occur with the
addition of catalyst (Table 3), albeit with higher temperatures
(0 °C) and longer reaction times (17 h). Surprisingly, bisamide 10
(entry 4) performs well for this reaction (60% conversion) despite
its failure to promote the previous two reactions (0% conversion).
Compounds
Keq
Solvent
Refs.
10
7
6b
9
56
py-d5
DMSO-d6
This work
159
127
1517
840
1000
1930
>105
<10
10
21
22
96
28
6b
11
12
13
14
15
16
17
18
19
20
21
22
12d
12a
11a
This observation suggests
p-stacking with 25 promotes reactivity
of this substrate. Unfortunately, no significant enantioselectivity
was observed.
DMSO-d6/0.5% water
To gain a deeper understanding of the mechanism, DFT calcula-
tions were undertaken on a model system using a truncated
version of the anion-binding catalyst (Fig. 5). These calculations
support chloride as the deprotonating agent rather than water or
the sulfur moiety of the catalyst. They also support direct cycliza-
tion to form a six-membered ring (Fig. 6).14 This same trend was
observed in the absence of catalyst and for the corresponding
indole system of Scheme 1 (see SI for further details). Similar
reports are known in the literature.15 However, formation of a
spirocyclic intermediate has not been ruled out experimentally.
Additional screening was done around 3-substituted furans in
an attempt to optimize the best yielding catalyst in Table 2 (entry
5). Yields and enantiomeric excess were measured using several
solvents and varied temperatures (Table 4). Ethereal solvents
(entries 1 and 5) give good conversion and small but significant
amounts of enantiomeric excess, whereas more polar solvents16
such as CH2Cl2 tend to give good conversion and no enantioinduc-
tion (entry 3). Aromatic solvents decrease the conversion and
enantioselectivity, presumably due to decreased solubility of the
catalysts (entries 2 and 4). Additional experiments around ethereal
solvents (entries 6–9, CPME = cyclopentyl methyl ether) support
that polar solvents17 decrease enantioselectivity, possibly due to
6a
273
53
increased reaction rates (entry 6 vs 7, THF vs CPME, 14% vs 24%
ee). Also, it was determined that higher reaction concentrations
(0.1 M vs 0.01 M) increase the rate of reaction, subsequently low-
ering enantioselectivity (entry 9 vs 5, 6% vs 22% ee). In an attempt
to increase the enantioselectivity, the temperature was lowered to
À78 °C. However, this change inhibited reaction progression
(entries 11–13, <5% conversion). Decreased solubility of the cata-
lyst at these temperatures is most likely the cause for the low con-
versions (reaction mixtures appeared cloudy as opposed to clear).
Following the hypothesis that more polar linear ethereal solvents
may increase enantioselectivity (entry 1 vs 5, TBME vs ether, 16%
vs 22% ee), diglyme was tested.18 However, this solvent provided
no benefit (entry 10, <5% ee). An intermediate temperature was
also attempted (À60 °C). In CPME, little to no catalyzed reaction
Please cite this article in press as: Metz, A. E.; et al. Tetrahedron Lett. (2015), http://dx.doi.org/10.1016/j.tetlet.2015.07.058

4
A. E. Metz et al. / Tetrahedron Letters xxx (2015) xxx–xxx
Figure 5. Deprotonation with chloride, achiral catalyst 27 or water [B3LYP/6-31G(d)], in diethyl ether using a CPCM model for solvation.
Table 3
The cyclization of 2-substituted furans into N-acyliminium ions
Entry
Catalyst
Time (h)
Conversiona (%)
1
2
3
4
none
8
7
17
17
17
17
<5
60b
57b
60b
10
a
Determined by 1H NMR relative to an IS, butylated hydroxytoluene (BHT).
Significant enantioselectivity was not observed.
b
Table 4
Additional screening around catalyst 6b
Figure 6. Six- vs five-membered ring formation with achiral catalyst 27 [B3LYP/6-
31G(d)], in diethyl ether using a CPCM model for solvation.
Table 2
The cyclization of 3-substituted furans into N-acyliminium ions
Entry
Solvent
Temp (°C)
Time (h)
Conversiona (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
TBME
–42
–42
–42
–23
–42
–42
–42
–42
–42
–42
–78
–78
–78
5
5
5
5
5
7
7
7
7
5
15
15
15
71 (16)
65 (0)
75 (<5)
55 (<5)
73 (22)
67 (14)
42 (24)
<10 (N/A)
97 (6)
98 (<5)
0
toluene
CH2Cl2
CF3C6H5
ether
THF
CPME
ether (no cat)
ether (0.1 M)
diglyme
THF
ether
CPME
Entry
Catalyst
Time (h)
Conversiona (%)
1
2
3
4
5
6
None
9
8
7
6b
10
1,7
1
7
7
7
<5
76
60 (0)
42 (10)
70 (17)
0
<5
<5
7
a
Determined by 1H NMR relative to an IS, butylated hydroxytoluene (BHT).
a
Determined by 1H NMR relative to an IS, butylated hydroxytoluene (BHT).
Numbers in parentheses refer to enantiomeric excess (%).
Numbers in parentheses refer to enantiomeric excess (%).
precipitation. In ether, the reaction did progress, but no increase
in enantioselectivity was observed at À60 °C. From these results
the optimal reaction conditions remain CPME, at À42 °C for 7 h.
was observed (about 12% conversion) and there was not sufficient
product to determine accurately the enantiomeric excess. The
reactions in CPME were still cloudy indicating potential catalyst
Please cite this article in press as: Metz, A. E.; et al. Tetrahedron Lett. (2015), http://dx.doi.org/10.1016/j.tetlet.2015.07.058

A. E. Metz et al. / Tetrahedron Letters xxx (2015) xxx–xxx
5
K.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2013, 52, 534–561; (d) Mahlau, M.; List,
B. Angew. Chem., Int. Ed. 2013, 52, 518–533; For a recent review on (thio)urea
organocatalysts and anion recognition refer to: (e) Zhang, Z. G.; Schreiner, P. R.
Chem. Soc. Rev. 2009, 38, 1187–1198.
Future efforts are focused on catalyst modification or altering the
halogenating reagent (i.e., AcCl or TMSBr). Additional efforts are
aimed toward validating the computational model of the cycliza-
tion of furans onto N-acyliminium ions with the goal of designing
catalysts to achieve higher enantioinduction.
In summary, this paper discusses the effect of anion-binding
interactions on reaction catalysis. Anion-binding compounds based
on a xanthene scaffold are described. They are distinct from the
thiourea hydrogen-bonding catalysts currently in the literature,
because they are more effective in promoting Pictet–Spengler-type
reactions of hydroxylactams. These potent anion-binders have the
potential to pave the way for a new reaction development focused
on anion-binding catalysis.
2. (a) Moyano, A.; Rios, R. Chem. Rev. 2011, 111, 4703–4832; For a review on
recent advances in cooperative ion pairing in asymmetric organocatalysis refer
to: (b) Brière, J.; Oudeyer, S.; Dalla, V.; Levacher, V. Chem. Soc. Rev. 2012, 41,
1696–1707.
3. De, C. K.; Mittal, N.; Seidel, D. A. J. Am. Chem. Soc. 2011, 133, 16802–16805.
4. a) Raheem, I. T.; Thiara, P. S.; Peterson, E. A.; Jacobsen, E. N. J. Am. Chem. Soc.
2007, 129, 13404–13405; b) Peterson, E. A.; Jacobsen, E. N. Angew. Chem., Int.
Ed. 2009, 48, 6328–6331; c) Raheem, I. T.; Thiara, P. S.; Jacobsen, E. N. Org. Lett.
2008, 10, 1577–1580.
5. Schmidtchen, F. P.; Berger, M. Chem. Rev. 1997, 97, 1609–1646.
6. a) García-Garido, S. E.; Caltagirone, C.; Light, M. E.; Gale, P. A. Chem. Commun.
2007, 1450–1452; b) Bühlmann, P.; Nishizawa, S.; Xiao, K. P.; Umezawa, Y.
Tetrahedron 1997, 53, 1647–1654.
7. Nemoto, T.; Obuchi, K.; Tamura, S.; Fukuyama, T.; Hamada, Y. Tetrahedron Lett.
2011, 52, 987–991.
8. a) Hanawa, H.; Kii, S.; Maruoka, K. Adv. Synth. Catal. 2001, 343, 57–60; b)
Malaisé, G.; Barloy, L.; Osborn, J. A. Tetrahedron Lett. 2001, 42, 7417–7419.
9. Rasappan, R.; Laventine, D.; Reiser, O. Coord. Chem. Rev. 2008, 252, 702–714.
10. Bordwell, F. G.; Fried, H. E.; Hughes, D. L.; Lynch, T. Y.; Satish, A. V.; Whang, Y. E.
J. Org. Chem. 1990, 55, 3330–3336.
11. a) Andrews, N. J.; Haynes, C. J. E.; Light, M. E.; Moore, S. J.; Tong, C. C.; Davis, J.
T.; Harrell, W. A.; Gale, P. A. Chem. Sci. 2011, 2, 256–260; (b) Jakab, G.; Tancon,
C.; Zhang, Z.; Lippert, K. M.; Schreiner, P. R. Org. Lett. 2012, 14, 1724–1727.
12. a) Jagessar, R. C.; Shang, M.; Scheidt, W. R.; Burns, D. H. J. Am. Chem. Soc. 1998,
120, 11684–11692; b) Evans, N. H.; Beer, P. D. Angew. Chem., Int. Ed. 2014, 53,
11716–11754; c) Gale, P. A.; Busschaert, N.; Haynes, C. J. E.; Karagiannidis, L. E.;
Kirby, I. L. Chem. Soc. Rev. 2014, 43, 205–241; d) Chmielewski, M. J.; Jurczak, J.
Chem. Eur. J. 2005, 11, 6080–6094.
13. a) Tanis, S. P.; Deaton, M. V.; Dixon, L. A.; McMills, M. C.; Raggon, J. W.; Collins,
M. A. J. Org. Chem. 1998, 63, 6914–6928; b) Huang, H.; Ji, X.; Wu, W.; Jiang, H.
Chem. Commun. 2013, 3351–3353.
14. All calculations were optimized in diethyl ether (CPCM model) at the B3LYP/6-
31G(d) [and/or M06-2X/6-31G(p,d)] level, as implemented in Gaussian09:
Frisch, M. J. et al Gaussian 09, rev performed using GAUSSIAN09, revision A.02;
Gaussian: Wallington CT, 2009. Additional details can be found in the
Supporting information.
Acknowledgements
We are grateful to the NIH (GM-087605) and NSF
(CHE0911713) for financial support of this research. Partial
instrumentation support was provided by the NIH for MS
(1S10RR023444) and NMR (1S10RR022442). A.E.M. thanks the
ACS Division of Organic Chemistry and Organic Syntheses for a
graduate fellowship. K.R. and M.C.K. acknowledge XSEDE
(TGCHE120052) for computational resources. Dr. Osvaldo
Gutierrez is acknowledged for assistance in performing
calculations.
Supplementary data
Supplementary data (complete computational details, complete
Ref. 14 for Gaussian09, experimental procedures, full characteriza-
tion, including 1H NMR and 13C NMR spectra, for all new
compounds, HPLC chromatograms) associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/j.
tetlet.2015.07.058.
15. Maresh, J. J.; Giddings, L.-A.; Friedrich, A.; Loris, E. A.; Panjikar, S.; Trout, B. L.;
Stockigt, J.; Peters, B.; O’Connor, S. E. J. Am. Chem. Soc. 2008, 130, 710–723.
16. Ansyln, E. V.; Dougherty, D. A. Molecular Structure and Thermodynamics:
Introduction to Structure and Models of Bonding In Modern Physical Organic
Chemistry; University of Sciences Books: California, 2006; p 146.
17. Greener Solvent Alternatives: Supporting the Advancement of Chemistry
through Sounds Environmental, Social & Fiscal Responsibilities. 2010, Sigma-
Aldrich Co. https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/
Sigma/Brochure/greener_solvent_alternatives.pdf. (accessed March 31, 2015).
18. Marcheselli, L.; Pistoni, G.; Tagliazucchi, M.; Tassi, L.; Tosi, G. J. Chem. Eng. Data
1993, 38, 204–206.
References and notes
1. For recent reviews on chiral anion-mediated asymmetric ion pairing chemistry
refer to: (a) Lacour, J.; Moraleda, D. Chem. Commun. 2009, 7073–7089; (b)
Phipps, R. J.; Hamilton, G. L.; Toste, F. D. Nat. Chem. 2012, 4, 603–614; (c) Brak,
Please cite this article in press as: Metz, A. E.; et al. Tetrahedron Lett. (2015), http://dx.doi.org/10.1016/j.tetlet.2015.07.058