Electrostatically enhanced phosphoric acids: A tool in Br?nsted acid catalysis

Article abstract of DOI:10.1021/acs.orglett.6b02750

A novel type of phosphoric acid catalyst with enhanced reactivity is reported. These compounds possess one or two positively charged centers which electrostatically activate them. This is illustrated in several bond-forming transformations including Friedel-Crafts and Diels-Alder reactions as well as a ring-opening polymerization. Rate accelerations corresponding to orders of magnitude are observed.

Full text of DOI:10.1021/acs.orglett.6b02750

Letter
pubs.acs.org/OrgLett
Electrostatically Enhanced Phosphoric Acids: A Tool in Brønsted Acid
Catalysis
Jie Ma and Steven R. Kass*
Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, United States
S
* Supporting Information
ABSTRACT: A novel type of phosphoric acid catalyst with
enhanced reactivity is reported. These compounds possess one
or two positively charged centers which electrostatically
activate them. This is illustrated in several bond-forming
transformations including Friedel−Crafts and Diels−Alder
reactions as well as a ring-opening polymerization. Rate
accelerations corresponding to orders of magnitude are
observed.
rønsted acids are among the most commonly employed
catalysts and are used to promote a wide variety of
B
chemical transformations.1,2 They provide an excellent
alternative to transition-metal-containing catalysts in that they
tend to be tolerant of water and air and generally are
environmentally friendlier.2,3 Substrate activation commonly
takes place by proton transfer or hydrogen bonding, both of
which lower the energy of the lowest unoccupied molecular
orbital (LUMO) and facilitate nucleophilic attack.4,5 Different
types of Brønsted acid catalysts with a wide range of acidities
have been developed including thioureas,6 α,α,α,α-tetraaryl-1,3-
dioxolane-4,5-dimethanols (TADDOLs),7 1,1′-bi-2-naphthol
(BINOL) derivatives,8 and phosphoric acids,9 and their utility
has been amply demonstrated. Of these compounds,
phosphoric acids are inherently the strongest acids, and their
structural flexibility enables them to be introduced into rigid
organic frameworks. The presence of a phosphoryl oxygen also
can serve as a hydrogen bond accepting site, enabling dual or
bifunctional activation modes that increase their catalytic
potential.1,9a,10
Figure 1. Charged phosphoric acids reported in this study.
positively charged substituents should act as powerful electron-
withdrawing groups in low polarity media, but not in polar
solvents, and the long-chain octyl groups were selected to
enhance the solubility of these salts in solvents such as
dichloromethane, chloroform, and toluene. A noncoordinating
counteranion is required to avoid deactivating hydrogen bonds
with the phosphoryl hydroxyl group, and tetrakis(3,5-bis-
(trifluoromethyl)phenyl)borate (BArF −) was chosen for this
4
purpose since it also has good solubility and reactivity
characteristics.19−21
Both catalysts were synthesized as illustrated in Scheme 1
starting with commercially available 3-hydroxypyridine. Its N-
alkylation with 1-iodooctane afforded the pyridinium iodide salt
(3) in a reasonable yield (66%) as previously described.16 Upon
reaction of 3 with phenyl dichlorophosphate or phosphorus
oxychloride followed by an aqueous hydrolysis, the halide salts
corresponding to 1 and 2 were produced (4 and 5). Ion
Diphenyl phosphate [(PhO)2P(O)OH, DPP] is widely used
and can catalyze a variety of transformations such as Friedel−
Crafts and Diels−Alder reactions11,12 and ring-opening
polymerizations.13 Its reactivity is limited due to its modest
acidity but can be enhanced by the presence of strong electron-
withdrawing groups on the two aromatic rings.14 More acidic
derivatives are of interest given Rueping et al.’s recent report
demonstrating the relationship between catalyst acidity and
activity.15 That is, more acidic catalysts led to faster rates of a
Nazarov cyclization test reaction.
The incorporation of a positively charged center into phenols
and thioureas recently was found to enhance their acidities and
catalytic reactivities.16,17 We therefore decided to investigate
charge-containing phosphoric acid derivatives of DPP (1 and 2,
Figure 1) and report the results herein.
exchange with NaBArF in CH2Cl2 occurred readily since the
4
sodium halide salts are not soluble and precipitate out of
solution. A sulfuric acid wash was subsequently carried out to
ensure that the target compounds were produced in their
protonated form (HCl is not suitable; presumably, it leads to
loss of the octyl group) and this was verified via titration.
To evaluate the catalytic reactivity of 1 and 2, the room-
temperature Friedel−Crafts alkylation of indole and N-
methylindole with β-nitrostyrene were examined at various
catalyst loadings in dichloromethane and chloroform (eq 1).
Two electrostatically enhanced phosphoric acids 1 and 2
were envisioned in which one or both of the phenyl groups in
DPP are replaced by N-alkylated pyridinium ions.18 The
Received: September 12, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02750
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
and 48 weeks for half of the starting material to be converted to
product depending upon the reaction conditions. Incorporation
of an electron-withdrawing nitro group in DPP at the 3-
position of both phenyl rings [(3-O2NC6H4O)2P(O)OH,
DPP*] has a modest accelerating effect corresponding to a
factor of 4.5. Monocharged phosphoric acid derivative 1,
however, speeds up the reactions of indole and N-methylindole
relative to DPP by factors of 12 and 210 in CD2Cl2 (entries 4
and 14) and 7.2 and 78 in CDCl3 (entries 10 and 20). These
differences can be enhanced by 1 order of magnitude if a
second charged center is incorporated into the catalyst. That is,
k2/kDPP = 120 and 2100 for indole and N-methylindole in
CD2Cl2 (entries 5 and 15) and 38 and 870, respectively, in
CDCl3 (entries 11 and 21). As a result, the half-lives for these
reactions are reduced to as little as 3.5 h.
Scheme 1. Synthetic Route for Catalysts 1 and 2
Bifunctional activation brought about by phosorphoric acid
derivatives has been invoked previously1,9a,10,22 and would
involve a N−H···OP hydrogen bond when indole is used as
the reactant (Scheme S1). With N-methylindole, this
interaction is absent, and consequently, one might expect this
latter transformation to proceed more slowly. This is the case
for the DPP-catalyzed reactions where kindole/kN‑methylindole
13−15 in CD2Cl2 and CDCl3. When 1 and 2 were used,
N‑methylindole/kindole = 1.4−1.5 except for when the former
=
1
k
These transformations were monitored by H NMR, and the
catalyst is used in CDCl3, then kindole/kN‑methylindole = 1.4. These
results are consistent with bifunctional activation being
involved in the rate-determining step when DPP is used, but
it is unclear if this is important with the charge-containing
phosphoric acids. These species undoubtedly are more acidic
than DPP, and thus, their N−H···OP hydrogen bonds
should be weaker. However, N-methylindole is more
nucleophilic than indole, and this could counterbalance the
loss of the hydrogen bond when 1 and 2 are used.
kinetic data displayed second-order behavior as expected. Rate
constants, and background corrected half-lives and relative rates
under the different conditions that were employed are
summarized in Table 1. In the absence of a catalyst, neither
indole or N-methylindole reacts with β-nitrostyrene in CD2Cl2
or CDCl3 for all practical purposes. That is, the reaction half-
lives were estimated to be 9 months or more (see Tables S1
and S2 for the kinetic data); a control experiment in which
CH2Cl2 was washed with H2SO4 as done in the preparation of
1 and 2 had no impact on the background rate of the reaction
with indole. These transformations are accelerated by 10 mol %
of DPP, but it is an ineffective catalyst in that it takes between 3
A small solvent effect (≤∼2) is observed for the Friedel−
Crafts reactions when CD2Cl2 and CDCl3 are used. If DPP is
the catalyst, the background corrected rate constants are ∼2
a
Table 1. Kinetic Data of Friedel−Crafts Reactions
b
b
entry
solvent
R
cat.
mol %
k (M−1 h−1
)
t1/2 (h)
krel
1
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CDCl3
CDCl3
CDCl3
CDCl3
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CDCl3
CDCl3
CDCl3
CDCl3
H
0.000544
0.0185
0.0813
0.215
19400
587
2
H
DPP
10
10
10
10
5.0
2.0
1.0
4.5
12
3
H
DPP*
131
4
H
1
2
2
2
49.2
5.12
11.7
28.8
26600
530
5
H
2.06
120
c
6
H
0.904
100
100
c
7
H
0.367
8
H
0.000396
0.0203
0.143
9
H
DPP
5.0
5.0
5.0
1.0
7.2
38
10
11
12
13
14
15
16
17
18
19
20
21
H
1
2
74.0
14.1
32600
7350
35.6
3.45
7.59
13.7
6760
8110
104
H
0.750
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
0.000324
0.00176
0.297
DPP
10
10
1.0
210
1
2
2
2
10
3.06
2100
c
5.0
2.5
1.39
1900
2200
c
0.771
0.00156
0.00286
0.103
DPP
5.0
5.0
5.0
1.0
78
1
2
1.13
9.35
870
a
b
c
Both reactant concentrations were 94.8 mM. Background corrected half-lives and relative rates. Linearly corrected for the amount of catalyst.
B
DOI: 10.1021/acs.orglett.6b02750
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
times larger in the latter solvent upon making a 2-fold linear
correction for the catalyst loading (i.e., 5 vs 10 mol %). In
contrast, this transformation takes place 1.2−1.5 times faster in
the former medium when 1 and 2 are employed, except for
than DPP and t1/2 = 1.5 min without any loss in the
diastereoselectivity relative to 1 (entry 5). Moreover, a reduced
catalyst loading of 0.1 mol % still leads to a very efficient
transformation (i.e., t1/2 = 17 min) without any falloff in the
selectivity.
when 1 is used in the reaction with indole and then kCDCl
/
3
Similar results were obtained for the Diels−Alder reaction
between 1,3-cyclopentadiene and acrylonitrile, but all three
catalysts (DPP, 1 and 2) are less effective. Diphenyl phosphate
only enhances the reaction rate of the noncatalyzed process by
7% (entries 7 and 8), and this difference is well within the
experimental uncertainties of the two measurements; we
estimate the errors in the rate constants to be 10%. There
is also no change in the observed endo/exo ratio of 65:35, and
thus, 1 mol % of DPP has little, if any, effect on this reaction.
An equivalent amount of 1 does speed up the transformation
and improves the diastereoselectivity, but the reaction is still
slow with an observed nonbackground corrected half-life of 84
h and an endo/exo ratio of 70:30 (entry 9). This kinetic
acceleration, however, does correspond to a factor of 11
compared to DPP. A further improvement is found with the
doubly charged catalyst 2 in terms of the rate (k2/kDPP = 50)
and the product selectivity (endo/exo = 75:25, entry 10). The
apparent half-life for this reaction is 33 h, but this is for a 1 mol
% loading, and when 5 mol % is employed (entry 11) this
decreases to 6.6 h.
kCD Cl = 1.3. Presumably the more polar solvent tends to lead
2
2
to faster reactions with the charged catalysts, as observed with
charged thioureas,17 because it stabilizes the polar transition
state. Lack of solubility under the reaction conditions prevented
toluene from being explored, but 2 was found to be unreactive
in THF-d8 since it is a hydrogen bond acceptor and a more
polar solvent. Chlorobenzene is also a poorer solvent for this
reaction than CD2Cl2 and CDCl3 even though it is less polar by
some measures.23,24
Catalyst loading was examined with 2 in CD2Cl2 (entries 5−
7 and 15−17), and smaller amounts of it led to slower
transformations with both indole and N-methylindole. Linear
relationships between the mol % of 2 and the background-
corrected reaction rate constants are observed in both cases
(Figure S1). This indicates that these Friedel−Crafts reactions
are first order in catalyst, which contrasts with the second-order
behavior reported for a charged thiourea.17
To examine the catalytic abilities of 1 and 2 further, Diels−
Alder reactions between 1,3-cyclopentadiene and methyl vinyl
ketone (MVK) or acrylonitrile (AN) were carried out (eq 2).
A ring-opening polymerization of δ-valerolactone with benzyl
alcohol as the initiator in four different solvents was examined
(eq 3) since this is an important process in which DPP is
1
These reactions were monitored by H NMR to obtain the
reactant and product concentrations at different times (Table
S3). Both transformations displayed second-order behavior,
and the resulting data are summarized in Table 2. For the
commonly used.13,26 The conversion of the monomer in these
1
living polymerizations was followed by H NMR (Table S4),
a
and the data can be fit using a first-order kinetic model.13,27,28
No conversion was observed over the course of one month in
the absence of a catalyst, whereas the ring-opening polymer-
ization is facile when 0.5 mol % of DPP or 2 is used. Rate
constants for the propagation stage are provided along with
reaction half-lifes and relative rates in Table 3. In all four
Table 2. Kinetic Data of Diels−Alder Reactions
k
endo/
exo
b
b
entry
R
cat.
mol % (M−1 h−1
)
t1/2 (h)
krel
1
Ac
0.660
15.2
61.3
81:19
81:19
82:18
88:12
88:12
87:13
65:35
65:35
70:30
75:25
76:24
2
Ac
DPP
1.0
1.0
1.0
1.0
0.10
0.823
3.89
1.0
19
3
Ac
DPP*
3.26
4
Ac
1
2
2
163
3.7 min
1.5 min
17 min
143
1000
2500
a
5
Ac
412
Table 3. Kinetic Data of Polymerizations
6
Ac
35.9
entry
solvent
cat.
k (h−1
)
t1/2 (h)
krel
7
CN
CN
CN
CN
CN
0.0698
0.0744
0.119
0.302
1.58
1
2
3
4
5
6
7
8
C6D5CD3
C6D5CD3
C6D6
DPP
2
0.444
0.833
0.325
0.574
0.0380
0.0759
0.0460
0.108
1.6
0.8
2.1
1.2
18
1.0
1.9
1.0
1.8
1.0
2.0
1.0
2.3
8
DPP
1.0
1.0
1.0
5.0
2170
203
1.0
11
50
9
1
2
2
DPP
2
10
11
43
C6D6
6.6
CD2Cl2
CD2Cl2
CDCl3
CDCl3
DPP
2
a
b
Both reactant concentrations were 100 mM. Background corrected
9.1
15
half-lives and relative rates.
DPP
2
6.4
reaction with MVK, 1.0 mol % of DPP was found to be a very
poor catalyst as the rate constant is only 25% larger than for the
noncatalyzed transformation (entries 1 and 2). DPP* is almost
20 times more reactive (entry 3), but the endo/exo selectivity
for these three processes is the same. Our monocharged
phosphoric acid, however, is 1000 times more active than DPP
(entry 4) and reduces the reaction half-life to under 5 min. It
also increases the endo/exo ratio from 81:19 to 88:12 as
commonly observed with Lewis acids.25 The doubly charged
derivative is even better in that it is 2500 times more reactive
a
Experiments performed with 0.5 M δ-valerolactone, 1.0 mol % of
PhCH2OH, and 0.5 mol % of the catalyst.
solvents, our doubly charged catalyst outperforms DPP by a
factor of about two. The actual reaction rates vary by a factor of
12 regardless of the catalysts with the faster processes occurring
in benzene-d6 and toluene-d8 and the slower ones taking place
in CD2Cl2 and CDCl3. When smaller concentrations of the
monomer are used, the polymerization takes longer but krel
C
DOI: 10.1021/acs.orglett.6b02750
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(8) (a) Matsui, K.; Takizawa, S.; Sasai, H. J. Am. Chem. Soc. 2005,
127, 3680−3681. (b) Wieting, J. M.; Fisher, T. J.; Schafer, A. G.; Visco,
M. D.; Gallucci, J. C.; Mattson, A. E. Eur. J. Org. Chem. 2015, 2015,
525−533. (c) Zhang, Y. D.; Li, N.; Qu, B.; Ma, S. L.; Lee, H.;
Gonnella, N. C.; Gao, J.; Li, W. J.; Tan, Z. L.; Reeves, J. T.; Wang, J.;
Lorenz, J. C.; Li, G. S.; Reeves, D. C.; Premasiri, A.; Grinberg, N.;
Haddad, N.; Lu, B. Z.; Song, J. H. J.; Senanayake, C. H. Org. Lett.
2013, 15, 1710−1713.
increases. This was also observed for the Friedel−Crafts and
Diels−Alder reactions presumably because the reactants are
more polar than the solvent and enhance the polarity of the
medium.
The effects of the two catalysts on the number-average
molecular weights (Mn,NMR) of the polymers at different
1
conversions were also investigated by H NMR. In all four
solvents, Mn,NMR was found to be similar at equivalent
conversions of the monomer regardless of the catalyst (Figure
S2).
(9) (a) Rueping, M.; Kuenkel, A.; Atodiresei, I. Chem. Soc. Rev. 2011,
40, 4539−4549. (b) Honjo, T.; Phipps, R. J.; Rauniyar, V.; Toste, F. D.
Angew. Chem., Int. Ed. 2012, 51, 9684−9688. (c) Held, F. E.; Grau, D.;
Tsogoeva, S. B. Molecules 2015, 20, 16103−16126.
(10) Akiyama, T.; Itoh, J.; Fuchibe, K. Adv. Synth. Catal. 2006, 348,
999−1010.
In summary, we successfully designed and synthesized two
electrostatically enhanced phosphoric acid derivatives (1 and
2). To the best of our knowledge, these compounds are the first
examples of phosphoric acids with positively charged centers
incorporated to enhance their acidities and catalytic abilities in
nonpolar media. This was evaluated by investigating a series of
acid-catalyzed processes and comparing the catalytic efficiencies
of 1 and 2 with diphenyl phosphate (DPP) and di(3-
nitrophenyl) phosphate (DPP*), their noncharged analogues.
Impressively, both charged phosphoric acid derivatives out-
performed DPP and DPP* in all of the reactions that were
examined, in some cases by more than 3 orders of magnitude.
Extension of this strategy to other Brønsted acid catalysts such
as chiral phosphoric acid derivatives is a promising research
avenue and currently is the focus of ongoing efforts in our
group.
(11) Wang, B. L.; Zhang, J. X.; Li, N. K.; Liu, G. G.; Shen, Q.; Wang,
X. W. Tetrahedron Lett. 2011, 52, 4671−4674.
(12) Takagi, R.; Kondo, A.; Ohkata, K. J. Mol. Catal. A: Chem. 2007,
278, 120−126.
(13) Makiguchi, K.; Ogasawara, Y.; Kikuchi, S.; Satoh, T.; Kakuchi, T.
Macromolecules 2013, 46, 1772−1782.
(14) Li, Y. H.; Lu, L. Q.; Das, S.; Pisiewicz, S.; Junge, K.; Beller, M. J.
Am. Chem. Soc. 2012, 134, 18325−18329.
(15) Kaupmees, K.; Tolstoluzhsky, N.; Raja, S.; Rueping, M.; Leito, I.
Angew. Chem., Int. Ed. 2013, 52, 11569−11572.
(16) Samet, M.; Buhle, J.; Zhou, Y. W.; Kass, S. R. J. Am. Chem. Soc.
2015, 137, 4678−4680.
(17) Fan, Y.; Kass, S. R. Org. Lett. 2016, 18, 188−191.
(18) Only small reactivity differences are expected for 4-N-
alkylpyridinium ion DPP derivatives since the negative charge in
their conjugate bases is not delocalized onto the aromatic rings.
(19) Nishida, H.; Takada, N.; Yoshimura, M.; Sonoda, T.; Kobayashi,
H. Bull. Chem. Soc. Jpn. 1984, 57, 2600−2604.
ASSOCIATED CONTENT
* Supporting Information
■
S
(20) Rosokha, S. V.; Stern, C. L.; Ritzert, J. T. CrystEngComm 2013,
15, 10638−10647.
(21) Scherer, A.; Mukherjee, T.; Hampel, F.; Gladysz, J. A.
Organometallics 2014, 33, 6709−6722.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b02750.
(22) (a) Simon, L.; Goodman, J. M. J. Am. Chem. Soc. 2008, 130,
8741−8747. (b) Marcelli, T.; Hammar, P.; Himo, F. Chem. - Eur. J.
2008, 14, 8562−8571. (c) Yamanaka, M.; Hirata, T. J. Org. Chem.
2009, 74, 3266−3271.
Experimental procedures, NMR spectra, and reaction
data (PDF)
AUTHOR INFORMATION
Corresponding Author
(23) Reichardt, C., Welton, T. Solvents and Solvent Effects in Organic
Chemistry, 4th ed.; Wiley-VCH: Weinheim, 2010.
■
(24) Burdick & Jackson Solvent Guide, 3rd ed.; Burdick & Jackson
Laboratories: Muskegon, MI, 1990.
(25) Nie, J.; Kobayashi, H.; Sonoda, T. Catal. Today 1997, 36, 81−
84.
(26) Makiguchi, K.; Satoh, T.; Kakuchi, T. Macromolecules 2011, 44,
*E-mail: kass@umn.edu.
Notes
The authors declare no competing financial interest.
1999−2005.
ACKNOWLEDGMENTS
■
(27) Kan, S. L.; Jin, Y.; He, X. J.; Chen, J.; Wu, H.; Ouyang, P. K.;
Guo, K.; Li, Z. J. Polym. Chem. 2013, 4, 5432−5439.
(28) Chen, J.; Kan, S. L.; Xia, H. D.; Zhou, F.; Chen, X.; Jiang, X. Q.;
Guo, K.; Li, Z. J. Polymer 2013, 54, 4177−4182.
We thank Ms. Deborah K. Schneiderman (University of
Minnesota) for helpful discussions. Generous support from the
National Science Foundation (CHE-1361766) and the
Petroleum Research Fund (556313-ND4) as administered by
the American Chemical Society is also gratefully acknowledged.
REFERENCES
■
(1) Akiyama, T. Chem. Rev. 2007, 107, 5744−5758.
(2) Yu, J.; Shi, F.; Gong, L. Z. Acc. Chem. Res. 2011, 44, 1156−1171.
(3) Rueping, M.; Koenigs, R. M.; Atodiresei, I. Chem. - Eur. J. 2010,
16, 9350−9365.
(4) Parmar, D.; Sugiono, E.; Raja, S.; Rueping, M. Chem. Rev. 2014,
114, 9047−9135.
(5) Liao, H. H.; Chatupheeraphat, A.; Hsiao, C. C.; Atodiresei, I.;
Rueping, M. Angew. Chem., Int. Ed. 2015, 54, 15540−15544.
(6) Takemoto, Y. Org. Biomol. Chem. 2005, 3, 4299−4306.
(7) (a) Seebach, D.; Beck, A. K.; Heckel, A. Angew. Chem., Int. Ed.
2001, 40, 92−138. (b) Huang, Y.; Unni, A. K.; Thadani, A. N.; Rawal,
V. H. Nature 2003, 424, 146. (c) Rueping, M.; Sugiono, E.; Moreth, S.
A. Adv. Synth. Catal. 2007, 349, 759−764.
D
DOI: 10.1021/acs.orglett.6b02750
Org. Lett. XXXX, XXX, XXX−XXX