Electrophilic activation of aminocarboxylic acid by phosphate ester promotes Friedel-Crafts acylation by overcoming charge-charge repulsion

Article abstract of DOI:10.1039/c7ob02158e

Friedel-Crafts acylation of aromatic compounds with aminocarboxylic acids proceeds efficiently in the presence of a tailored phosphate ester and a strong Br?nsted acid, despite the strong charge-charge repulsion associated with acylium ion formation. Here, we investigate the mechanism of this electrophilic aromatic acylation reaction, focusing on how the aminocarboxylic acid is activated by the phosphate ester and how the charge-charge repulsion is overcome. In the first step of the reaction, an acyl phosphate is generated from the aminocarboxylic acid through the intervention of the phosphate ester, which possesses three methyl salicylate ester linkages. The o-methyl salicylates enhance the reactivity of the phosphate ester via a protonation-induced conformational change, thereby overwhelming the charge-charge repulsion associated with the acylium ion formation. Weakening of the resonance interaction in the C(O)-O(P) bond by the lone-pair electrons of the ether oxygen atom of the carboxylic acid functionality contributes to the rapid formation of the acylium ion. Thus, our results show that the formation of aromatic ketones from various carboxylic acids proceeds because the strong leaving ability of the acyl phosphate overwhelms the charge-charge repulsion associated with the formation of the acylium ion. This information will be helpful to improve the design of tailored phosphate reagents.

Full text of DOI:10.1039/c7ob02158e

View Article Online
View Journal
Organic &
Biomolecular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. Sumita, Y. Otani
and T. Ohwada, Org. Biomol. Chem., 2017, DOI: 10.1039/C7OB02158E.
This is an Accepted Manuscript, which has been through the
Volume 14 Number
1 7 January 2016 Pages 1–372
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Page 1 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
Electrophilic activation of aminocarboxylic acid by phosphate ester
promotes Friedel-Crafts acylation by overcoming charge-charge
repulsion
Akinari Sumita,a Yuko Otani, a Tomohiko Ohwada *a
a) Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7ꢀ3ꢀ1
Hongo, Bunkyoꢀku, Tokyo 113ꢀ0033, Japan
Eꢀmail: ohwada@mol.f.uꢀtoko.ac.jp
Abstract
FriedelꢀCrafts acylation of aromatic compounds with aminocarboxylic acids proceeds
efficiently in the presence of a tailored phosphate ester and a strong Brønsted acid,
despite the strong chargeꢀcharge repulsion associated with acylium ion formation.
Here, we investigate the mechanism of this electrophilic aromatic acylation reaction,
focusing on how the aminocarboxylic acid is activated by the phosphate ester and
how the chargeꢀcharge repulsion is overcome. In the first step of the reaction, an acyl
phosphate is generated from the aminocarboxylic acid through the intervention of the
phosphate ester, which possesses three methyl salicylate ester linkages. The oꢀmethyl
salicylates enhance the reactivity of the phosphate ester via a protonationꢀinduced
conformational change, thereby overwhelming the chargeꢀcharge repulsion associated
with the acylium ion formation. Weakening of the resonance interaction in the C(O)ꢀ
O(P) bond by the loneꢀpair electrons of the ether oxygen atom of the carboxylic acid
functionality contributes to the rapid formation of the acylium ion. Thus, our results
show that the formation of aromatic ketones from various carboxylic acids proceeds
1

Organic & Biomolecular Chemistry
Page 2 of 32
View Article Online
DOI: 10.1039/C7OB02158E
because the strong leaving ability of the acyl phosphate overwhelms the chargeꢀ
charge repulsion associated with the formation of the acylium ion. This information
will be helpful to improve the design of tailored phosphate reagents.
(200 words)
2

Page 3 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
Introduction
The FriedelꢀCrafts reaction is the classical method for synthesizing aromatic ketones,
which are important structures in medicinal chemistry and materials science.[1] There
have been many improvements in FriedelꢀCrafts methodologies over the years, but
the generation of acylium ions from amino acids to produce aromatic ketones remains
challenging.[2] The reason for this is the difficulty of selective activation of the
carboxylic acid in the vicinity of the positively charged ammonium or equivalent ion
in the substrate, even if strong Lewis acids or transition metal catalysts are employed.
In other words, it is difficult to produce the acylium ion (2) or its equivalent from an
+
amino acid (1) because of chargeꢀcharge repulsion between NH3 and CO+ (Figure 1);
in addition, the nucleophilic amino group may react with the activated carboxylic
functional group.
Figure 1. Generation of an acylium ion from an anthranic acid compound
Consequently, phosphorus reagents such as phosphorus pentoxide or Eaton reagent
have been employed to promote the acylation by dehydrating the carboxylic
functional group.[3] Usually, the cleavage of PꢀO bonds in phosphate esters is highly
energyꢀdemanding, but PꢀO bond cleavage can proceed smoothly when a carboxylic
or methyl ester functional group is present at the ortho position. For example, Bender
reported that salicylic acid (4) is a good leaving group from the phosphate ester (3),
because intramolecular hydrogen bonding from the carboxylic acid to phenolic
3

Organic & Biomolecular Chemistry
Page 4 of 32
View Article Online
DOI: 10.1039/C7OB02158E
oxygen activates the PꢀO bond (Figure 2 [a])[4]. Brown reported that methyl salicylate
is also a good leaving group. In this case, a cationic species (the trication of
ytterbium) is trapped between the carbonyl oxygen and phenolic oxygen atoms of 6,
thereby activating the PꢀO bond (Figure 2 [b]).[5] Kirby, on the other hand, reported
that salicylic acid could interact with the phosphorus atom (which serves as a
neighboring group participant) to kick out another salicylic acid from the phosphorus
atom.[6] In this case, the remaining salicylic acid (a nonꢀleaving spectator group) plays
a crucial role in the hydrolysis reaction (Figure 2 [c]). [7]
Figure 2. Representative P-O bond cleavage reactions employing salicylic acid
derivatives as the leaving group.
4

Page 5 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
This promising methodology facilitates PꢀO bond cleavage, and thus is likely to
favour smooth generation of a phosphorus cation (such as ROꢀ+P(=O)ꢀOR’) upon PꢀO
bond cleavage. The cation is then trapped by a nucleophilic species. In biological
systems, even carboxylic acid can react with a phosphorus cation species to generate
acyl phosphate (soꢀcalled coupled reactions),[8] which in turn shows reactivity toward
various nucleophilic functional groups.[9] We recently reported that a carefully tailored
organophosphate ester (13, Figure 3), which possesses three methyl salicylate ester
linkages to phosphorus, is a good promoter of acylation of carboxylic acids in acidic
solution.[10] Aminocarboxylic acids readily produced the corresponding acylium ion
in the presence of the tailored phosphate ester and a strong Brønsted acid, resulting in
efficient acylation reaction with aromatic compounds. Acyl phosphate is an
intermediate in this reaction, which could be used to introduce the acyl functional
group into aromatic compounds. Thus, the reaction can be used to transform various
carboxylic acids into aromatic ketones in the presence of aromatic compounds.
The aim of the present study is to examine how aminocarboxylic acid is activated by
the phosphate ester (13), which can be regarded as a structured analogue of
oligomeric phosphorus pentoxide, to generate acyl phosphate intermediates and how
the chargeꢀcharge repulsion is overcome in the acylation of aromatic compounds
(Figure 1). An understanding of the reaction mechanism should provide clues to
improve the design of tailored phosphate reagents, as well as giving insight into the
reaction mechanisms of phosphorus pentoxide.
Results and Discussion
Requirement for an ortho-substituted ester functional group
When a regioꢀisomeric phenol (4ꢀhydroxymethyl benzoate) (19) was linked to the
5

Organic & Biomolecular Chemistry
Page 6 of 32
View Article Online
DOI: 10.1039/C7OB02158E
phosphorus atom instead of methyl salicylate (phosphate ester (16) shown in Figure
3), the desired aromatic ketone (18) was not obtained (Figure 3). Interestingly,
however, the phosphate ester containing one methyl salicylate and two regioꢀisomeric
phenols (phosphate ester (15)) gave a low yield (10%) of the desired aromatic ketone
(18). Contrary to our expectation, no methyl salicylate (8) was obtained from the
reaction mixture, and instead the regioꢀisomeric phenol (19) was obtained.[11] This
result led us to consider that 8 acted as a nucleophilic catalyst, rather than as a proton
donor to phenolic oxygen to form an intramolecular hydrogen bond.[12]
The large difference in product yields observed between the phosphate esters 14 and
15 was consistent with the idea of an intramolecular nucleophilic catalytic role of the
orthoꢀsubstituted methyl ester, resulting in a change of the valency of the central
phosphine atom: i.e., one of the methyl salicylates interacts with the cationic
phosphorus atom as a Lewis base. In this situation, the other methyl salicylates hardly
interact with the cationic phosphorus atom, and thus can serve as good leaving groups
due to the intramolecular hydrogen bonding. [13] These two distinct functions of the
methyl salicylates enable the phosphorus cation to be generated efficiently; thus, at
least two methyl phosphates are required for efficient reaction.
6

Page 7 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
Ph-H
O
P
O
17:
O
O
3.0 mmol
COOH
O
MeOOC
+
COOMe
Ph
triflic acid
20°C, 20 min
12: 1.0 mmol
18
MeOOC
13~16: 1.0 mmol
O
P
O
COOMe
MeOOC
O
O
o
o
o,o,o : 92%
o
13
MeOOC
O
P
O
COOMe
MeOOC
O
O
p,o,o : 76%
o
o
14
p
COOMe
O
P
O
MeOOC
OH
MeOOC
O
O
o
p,p,o : 10%
p
o
COOMe
p
phenol 8
15
COOMe
OH
O
P
O
p
O
O
p,p,p : < 1%
MeOOC
p
p
phenol 19
COOMe
p
MeOOC
16
COOMe
0
100
Yield of ketone 18 (%)
Figure 3. Isolation yields of benzophenone (18) from the acylation reactions of benzene
(17) in the presence of various phosphates (13~16).
Acidity-dependence of reactions and conformations of phosphate ester in acidic
solution
The phosphate ester bearing three methyl salicylates (13) required a strongly acidic
condition to promote the acylation reaction of amino acid (1) with benzene (17)
(Figure 4). When trifluoroacetic acid (TFA, Hammett acidity function (H0), ꢀH0 = 2.7)
was employed, the desired aromatic ketone (20) was not obtained, and the phosphate
ester (13) was recovered in almost quantitative yield. On the other hand, when the
acidity was increased (triflic acid–trifluoroacetic acid system; –H0 = 11.8),[14] the
7

Organic & Biomolecular Chemistry
Page 8 of 32
View Article Online
DOI: 10.1039/C7OB02158E
phosphate ester (13) was completely consumed, and the desired aromatic ketone (20)
was obtained. However, the yield was rather low (10%) and a side product (21) was
also obtained (38% yield). When only triflic acid (ꢀH0 = 14.1) was employed, the
yield of the desired aromatic ketone (20) was dramatically increased to 88%. These
results may mean that high acidity is needed to block the intermolecular nucleophilic
attack of methyl salicylate on the resultant acylation electrophile, giving the side
product 21. Thus, a strongly acidic condition is needed both to activate the phosphate
ester and to enable the acylation reaction to proceed efficiently.
Figure 4. Phosphate-assisted acylation reaction of anthranilic acid (1) at various levels of
acidicity.
The 1H NMR spectrum of the phosphate ester (13) indicated that it was diꢀprotonated
in triflic acid (SI IIIꢀb, Figure S1). While no peak was detected at around 15 ppm in
TFA (SI IIIꢀb, Figure S4), a single peak was detected at around 15 ppm in triflic acid.
8

Page 9 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
The proton attached to the nonꢀbridged oxygen atom of the phosphate ester (H+ꢀ
O=PR3) could not be detected, as noted by Olah and coꢀworkers, so the downfieldꢀ
shifted peak was assumed to be due to the proton intercalated between the phenolic
oxygen and carbonyl oxygen atoms of methyl ester, i.e. an intramolecular hydrogen
bond is formed.[15] Thus, diꢀprotonation was considered to cause a conformational
change of the phosphate ester (13). The 31P NMR signal of the phosphate ester was
dramatically shifted to higher field in triflic acid (31P NMR = ꢀ78 ppm) compared with
the peak in chloroform (31P NMR = ꢀ19 ppm) or in TFA (31P NMR = ꢀ19 ppm). It has
been reported that monoꢀprotonation of the oxygen atom of simple phosphotriesters
(such as O=Pꢀ(OEt)3) did not induce a significant shift of the 31P NMR signal.[15]
Therefore, our observation that the 31P NMR signal of 13 appears at the same
chemical shift value in chloroform and in TFA is reasonable. Furthermore, in contrast
to the behavior of 13, the phosphate esters 14 (31P NMR = ꢀ21 ppm, in triflic acid) and
31
15 (31P NMR = ꢀ24 ppm, in triflic acid) did not show a marked shift of the P NMR
signal in triflic acid. These results indicate that diꢀprotonated 13 takes a different
conformation from diꢀprotonated 14 or 15, presumably due to the interaction between
orthoꢀoxygen of the methyl ester and the phosphorus atom.[16]
DFT calculations[18] were carried out to understand the structure of the diꢀ
protonated phosphate ester (13) in triflic acid solution (Figure 5). The diꢀprotonated
structure with one proton attached to phosphate nonꢀbridging oxygen, and the other
proton intercalated between phenolic oxygen and carbonyl oxygen of methyl ester
was optimized.
In the absence of interaction between the carbonyl oxygen of the methyl ester and the
cationic phosphorus atom (conformation A, 22), the 31P NMR signal was calculated to
9

Organic & Biomolecular Chemistry
Page 10 of 32
View Article Online
DOI: 10.1039/C7OB02158E
appear at ꢀ19 ppm, which is almost the same value as in the neutral state of the
phosphate ester (Figure 5). On the other hand, single (conformation B, 23) or dual
(conformation C, 24) interactions between the carbonyl oxygen atom(s) of the methyl
ester(s) and the cationic phosphorus atom caused a highꢀfield shift of the 31P NMR
signal.
Figure 5. Calculated 31P NMR signals of various dicationic conformers.
(GIAO-CPCM/PBE0/6-311++(d,p)//CPCM/M06-2X/6-31G(d), solvent: triflic acid)[17]
The experimental 31P NMR signal (31P NMR = ꢀ78 ppm (experimental value), 13 in
triflic acid) indicates the presence of a single interaction between the carbonyl oxygen
atom and the cationic phosphorus atom (conformation B, 23, 31P NMR = ꢀ72 ppm
(calculated value)). In addition, the trigonal bipyramidal conformation (conformation
B, 23) is consistent with the experimentally observed 13C NMR signals (Figure S14,
S15).
Reaction pathways of the phosphate ester
We found that the phosphate ester (13) dissociates into the cationic phosphate ester
preferentially via a unimolecular reaction pathway, i.e., not involving the carboxylic
acid, and then the resultant phosphorus cation smoothly reacts with the carboxylic
acid.[19]
10

Page 11 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
To investigate this reaction, we carried out a 31P NMRꢀbased kinetic study. Figure 6
illustrates the progress of the reaction of the phosphate ester (13) in triflic acid, with
or without benzonic acid (12). When the carboxylic acid was present in the reaction
mixture, the peak of the starting material (23) (at around ꢀ76 ppm) disappeared and
the peak at around ꢀ5 ppm increased as time passed (Figure 6 [a]). The new peak was
attributed to acyl phosphate generated from diꢀprotonated phosphate ester (23) and
benzonic acid (12). In the absence of benzonic acid (12), a peak appeared at around ꢀ
20 ppm.
85% H3PO4 aq.
20 min
0 min
ppm
0
−5
−10
−15 −20
−25 −30 −35
−40
−45 −50 −55
−60 −65
−70
−75
[b] time course spectra without benzonic acid (12)
85% H3PO4 aq.
22 min
0 min
0
−5
−10
−15
−20
−25
−30
−35
−40
−45
−50
−55
−60
−65
−70
−75 ppm
H
MeO
O
H
O
P
MeO
H
O
O
O
O
O
O
O
O
P
O
P
O
MeO
OMe
O
O
O
O
O
MeO
O
O
OMe
H
MeO
25
27
23
Figure 6. Changes of the 31P NMR signals during the reaction of the phosphate [a] with
benzonic acid (12) (one equivalent with respect to the phosphate ester (13)) and [b] in
the absence of benzonic acid (12). Spectra were measured every 2 minutes.
11

Organic & Biomolecular Chemistry
Page 12 of 32
View Article Online
DOI: 10.1039/C7OB02158E
This peak at around ꢀ20 ppm was considered to be due to a phosphoryl cationic
species (25)[20], based on the following experiments. When the reaction mixture of 13
and triflic acid was quenched with methanol under basic conditions, the phosphate
ester (26) and one equivalent of methyl salicylate (8) were obtained (Figure 7). The
phosphate ester (26) was derived from 13 by replacement of one methyl salicylate
with one methanol moiety. The yield of 26 was fairly low (46%), possibly due to
instability of the phosphoryl cation species (25). However, formation of one
equivalent of methyl salicylate supports the view that the cationic phosphate
intermediate (25) was generated. Moreover, DFT calculation gave a chemical shift
value of the 31P NMR signal of the cationic phosphate ester (25) of ꢀ19 ppm, which is
consistent with the experimental value.
O
P
O
O
P
O
COOMe
COOMe
MeOOC
MeOOC
1) Triflic acid,
20°C, 20 min
O
O
O
OMe
HO
+
2) Na2CO3, MeOH,
-78°C, 1 h
MeOOC
MeOOC
8: 1.02 mmol
(46 % yield)
13: 1.01 mmol
26: 0.46 mmol
H
O
Triflic acid,
20°C, 20 min
MeOH,
-78°C, 1 h
O
OMe
P
O
O
MeO
O
25
25
31P NMR: -19 ppm (85% H3PO4aq. = 0 ppm)
31P NMR(calcd): ꢀ19 ppm (H3PO4 = 0 ppm)
Figure 7. Trapping of the cationic intermediate (25) by work-up with methanol under
basic conditions. Calculation level: GIAO/CPCM/PBE0/6-311++G(d,p)//CPCM/M06-
2X/6-31G(d) (solvent: triflic acid)
12

Page 13 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
The 31P NMR spectra indicate that the phosphate ester (13) spontaneously dissociated
into the cationic phosphorus species (25) and methyl salicylate (8) through cleavage
of the PꢀO bond (Figure 6 [b]). Interestingly, the rate of decrease of 13 became larger
as the concentration of the carboxylic acid (12) was increased (Table S1).
Figure 8 shows the changes in the concentration of the phosphate ester (13)
observed during the reaction with equimolar benzonic acid (12), together with the
simulated concentration changes of 13 calculated according to Equation 1 with the
experimentally obtained rate constants in the absence of benzonic acid (12) (firstꢀ
order rate constant k1 with respect to [13]) and in the presence of ten equivalents of
benzonic acid (12) (secondꢀorder rate constant k2 with respect to [13] and [12]) .[21]
ꢀd[phosphate 13]/dt = k1[phosphate 13] + k2[phosphate 13][carboxylic acid 12] Eq. 1.
These decay curves are very similar, suggesting that the phosphate ester (13) could
follow a combination of unimolecular (k1) and bimolecular (k2) reaction pathways.
13

Organic & Biomolecular Chemistry
Page 14 of 32
View Article Online
DOI: 10.1039/C7OB02158E
0.11$
0.10$
0.09$
0.08$
0.07$
0.06$
0.05$
0.04$
COOMe
COOH
O
O
+
P
O
O
trflic acid
12 °C
COOMe
MeOOC
12
13
0.1 mmol/mL
0.1 mmol/mL
Reaction of equimolar
12 and 13
Simulation of
decrease of 13 based
on Equation 1
0$
5$
10$
15$
20$
25$
Time%[min]%
Figure 8. Changes in the concentration of the phosphate ester (13) experimentally
observed (red) for the reaction of equimolar 13 and 12, and simulated (blue) according
to Equation 1 with the rate constants (k1 ( = 0.25 x 10-3 [s-1] ) and k2 ( = 2.6 x 10-3 [L mol-
•
1
•
s-1] )) experimentally obtained under first-order and second-order reaction conditions.
Figure 9 shows the calculated reaction pathways for the phosphate ester (13) with
benzoic acid (12) from protonated phosphate ester (23) to the generation of acyl
phosphate (30). The stable conformation of the phosphate ester (13) in acidic solution
was the trigonal bipyramidal structure (23) depicted in Figure 5.
In the unimolecular reaction pathway (Figure 9 [a]), the rateꢀdetermining step was
cleavage of the PꢀO bond to dissociate one methyl salicylate from the phosphorus
atom (23 to 29, or 24 to 28). Dissociation could occur via a trigonal bipyramidal
conformation (mechanism 1) (magenta line) or via an octahedral conformation
(mechanism 2) (blue line). The enthalpy of activation of the reaction pathway via the
trigonal bipyramidal conformation was 16.7 kcal/mol, and the corresponding enthalpy
of activation via the octahedral conformation was 15.7 kcal/mol. These values are
14

Page 15 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
consistent with the experimental results (kinetic parameters of the unimolecular
G‡ H‡ = 18 kcal/mol, S‡ = ꢀ12
reaction pathway:
cal/mol·K[22] [23]
ꢀ
= 22 kcal/mol,
ꢀ
ꢀ
298K
)
The octahedral conformation (24) is rather sterically crowded, so the
activation energy for departure of methyl salicylate is lower due to release of the
steric repulsion.[24]
On the other hand, in the bimolecular reaction pathway (Figure 9 [b]), the
phosphate ester (13) could only react via the trigonal bipyramidal conformation
(mechanism 3) (red line).[25] The rateꢀdetermining step of the bimolecular reaction
was different between the trigonal bipyramidal pathway and the octahedral
pathway.[26] In the trigonal bipyramidal pathway (red line), the rateꢀdetermining step
was cleavage of the PꢀO bond with loss of one methyl salicylate from the phosphorus
atom (including rotation of methyl salicylate; 32 to 34). A similar endothermic
transformation (like 32 to 34)[27] has previously been proposed to cause steric
acceleration of the subsequent leaving process of the orthoꢀsalicylate through the TS
37 leading to the acyl phosphate 34.
15

Organic & Biomolecular Chemistry
Page 16 of 32
View Article Online
DOI: 10.1039/C7OB02158E
Figure 9. Energy diagrams of reaction pathways of the phosphate ester. [a]:
unimolecular reaction pathways. [b]: bimolecular reaction pathways.
CPCM/M06-2x/6-311G(d,p)//CPCM/M06-2X/6-31G(d) (solvent: triflic
acid).
On the other hand, in the octahedral pathway (green line) (Figure 9(b)), the rateꢀ
determining step was approach of the carboxylic acid to the phosphorus atom to form
a new PꢀO bond (24 to 36). Because the octahedral conformation was rather sterically
crowded, this step required both a high enthalpy of activation and a high Gibbs free
16

Page 17 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
energy. In the trigonal bipyramidal pathway, the enthalpy of activation and Gibbs free
energy of the rateꢀdetermining step were lower. From the viewpoint of the enthalpy of
activation of the reaction, the trigonal bipyramidal pathway gave a better match with
the experimental values (kinetic parameters of the bimolecular reaction pathway:
ꢀ
G‡298K = 21 kcal/mol, H‡ = 12 kcal/mol, S‡ = ꢀ30 cal/mol·K). [23]
ꢀ ꢀ
In all three reaction pathways (mechanisms 1-3), the departure of methyl salicylate
from the phosphorus atom is rateꢀdetermining.
Reactivity of acyl phosphate
Both the unimolecular and bimolecular reaction pathways generated the acyl
phosphate species (27) (shown in Figure 6), which is expected to show high reactivity
toward aromatic compounds. Previously, we demonstrated that methyl salicylate (8)
was a good leaving group from the carbonyl carbon, enabling smooth generation of
the acylium ion.[28] However, methyl salicylate could not leave the carbonyl carbon of
the anthranilic ester (21) (Figure 10 [b]). On the other hand, the acyl phosphate (38),
generated from the phosphateꢀassisted reaction, could smoothly react with an
aromatic compound (17) to form an aromatic ketone (20). Thus, 38 has a greater
+
leaving ability than Oꢀprotonated methyl salicylate (21+H ). Importantly, no ester
species (21) was obtained after the phosphateꢀassisted acylation reaction (Figure 10
[c]), indicating that 38 reacted directly with 17.
17

Organic & Biomolecular Chemistry
Page 18 of 32
View Article Online
DOI: 10.1039/C7OB02158E
Figure 10. Reactions of anthranilic acid derivatives.
To examine the role of the acyl phosphoryl functional group, acyl phosphates (39
and 41) were synthesized.[29] Purified acyl phosphate (39) bearing unsubstituted
benzonic acid and phenol on the phosphorus atom showed similar reactivity to 13,
and gave the desired benzophenone (18) in 81 % yield (Figure 11). Importantly,
diphenyl phosphate (40) was obtained as a side product, indicating selective cleavage
of the CꢀO bond, rather than the PꢀO bond. The similar yields of 18 and 40 indicate
that the acyl phosphoryl functional group itself possesses high leaving ability. Similar
reaction behavior was observed when anthranilic acid was linked to the phosphorus
atom instead of benzonic acid (41). Acyl phosphate (41) was obtained from the
reaction of diphenyl phosphorochloridate with anthranilic acid (Scheme S5), but was
relatively unstable and difficult to purify, so we used the crude product, which gave
moderate yields of diphenyl phosphate (40) (47%) and aromatic ketone (20) (40%)
(Figure 11). Generation of these products in comparable yields indicates a strong
18

Page 19 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
ability to generate the acylium ion from acyl phosphate (41), overcoming the chargeꢀ
charge repulsion.
Figure 11. Acylation reactions starting from the acyl phosphates (39 and 41).
Enhanced leaving ability of acyl phosphate
Figure 12 illustrates the reaction pathways elucidated by DFT calculation. Carboxylic
acid (42), ester (44), and acyl phosphate (46) were compared from the viewpoint of
enthalpy of activation. In the case of the carboxylic acid (42), attempts to optimize a
transitionꢀstate structure starting from N,Oꢀdiꢀprotonated anthranilic acid failed: as the
distance between the carboxyl carbon atom and the oxygen atom of the protonated
hydroxyl group of the carboxylic acid became longer, the energy monotonously
increased. The situation remained the same even when a counterꢀion was located near
the protonated amino functional group. Thus, the intramolecularly hydrogenꢀbonded
state (42, Figure 12) was employed as the starting state for the generation of the
acylium ion (43). In the case of ester (44), the reaction pathway started from the
conformation involving
19

Organic & Biomolecular Chemistry
Page 20 of 32
View Article Online
DOI: 10.1039/C7OB02158E
Figure 12. Energy diagrams of the reactions from anthranilic acid derivatives to the
acylium ion. CPCM/M06-2X/6-311G(d,p)//CPCM/M06-2X/6-31G(d) (solvent: triflic
acid).
an intramolecular hydrogen bond (44, Figure 12), but in this case too, cationic charge
on the carbonyl carbon was rather energetically unfavored, and we could not obtain an
optimized transition structure. However, when a counterꢀion was placed near the
protonated amino functional group, a transition state was obtained.
In the case of acyl phosphate, the starting state was the diꢀprotonated species
(46): one proton was attached to the amino group and the other was intercalated
between the phosphate nonꢀbridging oxygen and the carbonyl oxygen of the methyl
ester (Figure 12). This structure is supported by experimental results[10]: benzonic acid
reacted with tailored phosphate ester to generate acyl phosphate with one proton
intercalated between the phosphate nonꢀbridging oxygen and the carbonyl oxygen of
20

Page 21 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
the methyl ester in triflic acid solution. From the acyl phosphate, a transition state for
cleavage of the PꢀO bond was successfully optimized.
As expected from the experimental results (Figure 10), the enthalpy of activation of
the reaction from acyl phosphate (46) was the lowest (ꢀ
H‡ = 15.8 kcal/mol) among
those of carboxylic acid (42), ester (44), and acyl phosphate (46). Although the
enthalpy of activation of the ester species (44) (25.0 kcal/mol) is also lower than that
of the carboxylic acid (42) (33.4 kcal/mol), the calculated energy profile indicated that
the reaction of the ester species (44) would not proceed smoothly under conditions of
a short reaction time and at room temperature.
A possible explanation for the enhanced leaving ability of acyl phosphate (46)
would be a resonance effect in the phosphoryl leaving group. Figure 13 compares the
bond lengths (Figure 13 [a]), NPA charges (Figure 13 [b]), and results of NBO
analysis[30] (Figure 13 [c]) of the optimized structures of a simplified carboxylic acid
(48) and acyl phosphates (49, 50). The results support an enhanced leaving ability of
the acyl phosphate. A comparison of bond lengths shows that the C2ꢀO3 bond length
of the acyl phosphate (49) (1.38 ꢀ ) is longer than that of the carboxylic acid (48)
(1.34 ꢀ ). The NPA charges of the acyl group reflected the differences of C2ꢀO3 bond
lengths (Figure 13 [b], highlighted in yellow). The NPA charge of the acyl group of 49
(+0.35) was more cationic than that of the acyl group of 48 (+0.26). When hydrogen
chloride was placed at the phosphoryl group of 50, the NPA charge was even more
cationic (+0.40), which would imply an even lower activation energy for generation
of the acylium ion.
Elongation of the C2ꢀO3 bond of the acyl phosphate was due to the weakened
resonance between the lone pair of the O3 atom and the carbonyl group, as evidenced
21

Organic & Biomolecular Chemistry
Page 22 of 32
View Article Online
DOI: 10.1039/C7OB02158E
Figure 13. Comparison of [a]: bond lengths, [b] NPA charges, and [c] NBO analysis
results of the carboxylic acid (48), acyl phosphate (49), and partially protonated acyl
phosphate (50). CPCM/MP2/6-311++G(d,p)//CPCM/M06-2X/6-31G(d) (solvent : triflic
acid).
by the NBO analysis (53.32 kcal/mol in the case of the acyl phosphate (49), and 69.84
kcal/mol in the case of the carboxylic acid (48)) (Figure 13 [c]). The weak interaction
in 49 was caused by the contribution of another resonance interaction between the
lone pair of the O3 atom and the phosphoryl group (2.89 kcal/mol).[31] Also, when a
proton (H+) was coordinated to the oxygen atom of the phosphoryl group (50), this
induced greater resonance interaction of the O3 lone pair with the phosphoryl group
22

Page 23 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
(4.58 kcal/mol), and less toward the carbonyl group (44.21 kcal/mol).
Thus, in 50, the resonance effect of the ether atom (O3) and the phosphoryl group
weakened the resonance effect of the ether atom (O3) and the carbonyl group, which
would favor cleavage of the C2ꢀO3 bond to generate the acylium ion.
Conclusion
In this paper, we examine the reaction mechanism of the phosphateꢀassisted Friedelꢀ
Crafts acylation reaction. A combination of a tailored phosphate ester and triflic acid
enables the acylation reaction of amino acids with aromatic compounds to proceed
efficiently, affording aromatic ketones in high yields. We show that the tailored
phosphate ester works both via a unimolecular reaction pathway to generate an
unstable phosphorus cation, and via a bimolecular reaction pathway with carboxylic
acids to generate acyl phosphates directly. In the unimolecular pathway, one of the
methyl salicylates, having a proton intercalated between phenolic oxygen and
carbonyl oxygen, leaves the phosphorus atom to generate the phosphorus cation
species. This cation shows high reactivity toward carboxylic acid, resulting in rapid,
though indirect, formation of acyl phosphate. In the bimolecular reaction pathway, the
carbonyl oxygen appears to enhance both the nucleophilicity of the carboxylic acid
and the leaving ability of the methyl salicylate. Both reaction pathways generate acyl
phosphate, which has a high leaving ability, favoring generation of the acylium cation,
possibly due to the resonance effect of the phosphoryl functional group, which
overwhelms the chargeꢀcharge repulsion associated with acylium ion formation.
23

Organic & Biomolecular Chemistry
Page 24 of 32
View Article Online
DOI: 10.1039/C7OB02158E
Acknowledgement
This research was partially supported by a GrantꢀinꢀAid for Research Fellow for
Young Scientists (JSPS) to A.S. (16J08260) and by the University of Tokyo. The
computations were performed at the Research Center for Computational Science,
Okazaki (Japan).
Experimental Section
General Procedures
Melting points were determined with a Yanaco micro melting point apparatus without
1
31
correction. Hꢀ (400 MHz), 13Cꢀ (100 MHz), and Pꢀ (162 MHz) NMR spectra were
recorded on a Bruker Avance400. Chemical shifts were calibrated with
tetramethylsilane as an internal standard or with the solvent peak, and are shown in
ppm (δ) values. Coupling constants are shown in hertz (Hz). The following
abbreviations are used: s = singlet, d = doublet, t = triplet, q = quartet, dd = double
doublet, dt = double triplet, dq = double quartet, h = hextet, m = multiplet, and brs =
broad singlet. Electrospray ionization timeꢀofꢀflight mass spectra (ESIꢀTOF MS) were
recorded on a Bruker micrOTOFꢀ05 as highꢀresolution mass spectra (HRMS). All
reagents were commercially available, and were used without further purification,
unless otherwise noted. Flash column chromatography was carried out on silica gel
(silica gel (40~63µm)). The combustion analyses were carried out in the
microanalytical laboratory of this department.
Organophosphate-catalyzed reaction of 2-aminobenzoic acid (1) with benzene
(17)
in
triflic
acid
solution
(-H0
=
14.1)
to
give
(2-
aminophenyl)(phenyl)methanone (20) (Figure 3)
To a mixture of benzene (17) (239.4 mg, 3.06 mmol), 2ꢀaminobenzoic acid (1) (137.7
24

Page 25 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
mg, 1.00 mmol) and trimethylsalicyl phosphate (13) (504.7 mg, 1.01 mmol), triflic
acid (2.0 mL) was added at 20°C. The reaction mixture was stirred at 20°C for 20
minutes, and then poured into iceꢀwater. 2 M aqueous NaOH (50 mL) was added to it,
and the resulting mixture was extracted with CH2Cl2 (40 mL x 3). The organic phase
was washed with brine (40 mL), dried over Na2SO4, and evaporated under reduced
pressure to give a residue, which was columnꢀchromatographed on silicaꢀgel (eluent:
EtOAc : nꢀhexane = 1 : 4) to (2ꢀaminophenyl)(phenyl)methanone (20) (173.0 mg,
0.88 mmol, 88%) as an yellow solid.
1
Mp. 105.9ꢀ107.1°C (yellow plates, recrystallized from CH2Cl2/nꢀhexane). HꢀNMR
(CDCl3, 400 MHz) δ (ppm): 7.621 (2H, d, J = 6.8 Hz), 7.496 (1H, dddd, J = 7.3, 7.3,
1.6, 1.6 Hz), 7.444ꢀ7.405 (3H, m), 7.278ꢀ7.231 (1H, m), 6.702 (1H, dd, J = 8.4, 0.8
Hz), 6.573 (1H, ddd, J = 7.2, 7.2, 0.8 Hz), 6.097 (2H, brs). 13CꢀNMR (CDCl3, 100
MHz) δ (ppm): 199.64, 151.47, 140.64, 135.14, 134.79, 131.59, 129.66, 128.63,
118.70, 117.55, 116.06. HRMS (ESIꢀTOF, [M+Na]+): Calcd. for C13H11NNaO+:
220.0733. Found: 220.0733. Anal. Calcd. for C13H11NO: C, 79.17; H, 5.62; N, 7.10.
Found: C, 79.22; H, 5.89; N, 7.08.
Work-up procedure with methanol under basic conditions to trap the reactive
cation (Figure 7).
To a trimethylsalicyl phosphate (13) (503.8 mg, 1.01 mmol), triflic acid (2.0 mL) was
added at 20°C. The mixture was stirred at 20°C for 20 minutes, and poured into a
solution of methanol and Na2CO3 at ꢀ78°C. The whole mixture was stirred for 1 hour
at ꢀ78°C, then filtered, and evaporated under reduced pressure to give a residue,
which was columnꢀchromatographed on silicaꢀgel (eluent: EtOAc : nꢀhexane = 1 : 1)
to afford bis(phenol oꢀmethylester)ꢀmono(methanol)ꢀphosphate ester (26) (176.1 mg,
25

Organic & Biomolecular Chemistry
Page 26 of 32
View Article Online
DOI: 10.1039/C7OB02158E
0.46 mmol, 46%) as a colorless oil, and methyl salicylate (8) (155.1 mg, 1.02 mmol,
102%) as a colorless oil.
1HꢀNMR (CDCl3, 400 MHz) δ (ppm): 7.992 (2H, d, J = 8.0 Hz), 7.507ꢀ7.424 (4H, m),
13
7.245 (2H, dd, J = 7.6, 7.6 Hz), 4.045 (3H, d, J = 12.0 Hz), 3.831 (6H, s). CꢀNMR
(CDCl3, 100 MHz) δ (ppm): 165.80, 149.94 (JCP = 7 Hz), 134.14, 132.39, 125.85,
123.62 (JCP = 7 Hz), 122.24 (JCP = 2 Hz), 56.64 (JCP = 6 Hz), 52.73. 31PꢀNMR
(CDCl3, 162 MHz) δ (ppm): ꢀ11.97. HRMS (ESIꢀTOF, [M+Na]+): Calcd. for
C17H17NNaO8P+: 403.0553. Found: 403.0550. Anal. Calcd. for C17H17O8P: C, 53.69;
H, 4.51. Found: C, 53.60; H, 4.63.
Superacid reaction of acyl phosphate (39)
To a solution of benzene (17) and benzoic (diphenyl phosphoric) anhydride (39)
(354.8 mg, 1.00 mmol), triflic acid (2.0 mL) was added at 20°C. The reaction mixture
was stirred at 20°C for 20 minutes, and then poured into iceꢀwater. 2 M aqueous
solution of NaOH (50 mL) was added, and the resulting mixture was extracted with
CHCl3 (50 mL x 3). The aqueous phase was poured into 2 M aqueous HCl (100 mL),
then extracted with CH2Cl2 (50 mL x 3)), washed with brine (40 mL), dried over
Na2SO4, evaporated and vacuumꢀdried to afford diphenyl phosphate (40) (205.3 mg,
0.82 mmol, 82%) as a brown oil. The organic phase was washed with brine (40 mL),
dried over Na2SO4, and the solvent was evaporated under reduced pressure to give a
residue, which was columnꢀchromatographed on silicaꢀgel (eluent: EtOAc : nꢀhexane
= 1 : 6) to afford benzophenone (18) (147.1 mg, 0.81 mmol, 81%) as a colorless oil.
Kinetic Study
Weighed substrates (phosphate ester 13 and benzoic acid 12) were placed in a dry
26

Page 27 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
flask filled with dry argon gas, and distilled triflic acid was added at 0°C to
completely dissolve the substrates. For the kinetic study of the bimolecular reaction,
phosphate ester 13 in triflic acid solution and benzoic acid 12 in triflic acid solution
were mixed at ꢀ40°C.
An aliquot of the solution was transferred into a dry NMR tube filled with argon and
shielded from light. The NMR tube, containing a solution of the phosphate ester 13 or
a mixture of the phosphate ester 13 and benzoic acid 12 was cooled in the NMR
machine to a specified temperature. The temperature of the sample in the NMR probe
was calibrated by using methanol according to Geet’s method.[32] Magnetic field
locking was not used, but the shimming was maximized with CDCl3 before
measurements. The peak of an 85% aqueous solution of H3PO4 sealed in a tube in the
solution was used as an internal standard of integration.
The integration value of the phosphorus atom of the phosphate ester, as compared
with that of 85% aqueous solution of H3PO4, was obtained and used to calculate the
concentration of the phosphate ester. The logarithms of the concentration of the
starting substrate (log[C]) were plotted against time (t, min) to give pseudoꢀfirstꢀorder
kinetics, and the rate constants were calculated.
Computational Study
We carried out computational studies by using the Gaussian 09 suites of programs.[20]
The geometries of the starting structure, transition state, and product at each step were
fully optimized at the CPCM/M06ꢀ2X/6ꢀ31G(d) level. Harmonic vibrational
frequencies were calculated to characterize the optimized structures. Bulk solvation
effects (selfꢀconsistent reaction field, SCRF) were simulated by the CPCM method[33]
in trifluoromethanesulfonic acid (eps = 77.4, rsolv = 2.5985274, density = 1.696,
epsinf = 1.882384 (the value of acetic acid was employed)), water, or CHCl3 as a
27

Organic & Biomolecular Chemistry
Page 28 of 32
View Article Online
DOI: 10.1039/C7OB02158E
solvent. Singleꢀpoint energies were calculated at the CPCM/M06ꢀ2X/6ꢀ311G(d,p)
level (solvent = triflic acid) on the basis of the optimized structures. The zeroꢀpoint
vibrational energy corrections were done without scaling. 31P chemical shifts were
elucidated at various calculation levels using GIAO methods.[34] Isotropic values were
employed as the calculated NMR peaks. The peak values of 31P NMR were calibrated
with H3PO4 as a standard compound.
References
[1] G. Sartori, R. Maggi, Advances in FriedelꢀCrafts Acylation Reactions: Catalytic and
Green Processes; CRC Press: France, 2017.
[2] (a) O. Itoh, T. Honnami, A. Amano, K. Murata, Y. Koichi, T. Sugita, J. Org. Chem., 1992,
57, 7334: (b) T. F. Buckley III, H. Rapoport, J. Am. Chem. Soc., 1981, 103, 6157: (c) I.
Ivanov, S. Nikolova, S. StatkovaꢀAbeghe, Synth. Comm., 2006, 36, 1405.
[3] (a) P. E. Eaton, G. R. Carlson, J. T. Lee, J. Org. Chem., 1973, 38, 4071: (b) D. Zewge, C.ꢀ
Y. Chen, C. Deer, P. G. Domer, D. L. Hughes, J. Org. Chem., 2007, 72, 4276.
[4] M. L. Bender, J. M. Lawlor, J. Am. Chem. Soc., 1963, 85, 3010.
[5] D. R. Edwards, A. A. Neverov, R. S. Brown, J. Am. Chem. Soc., 2009, 131, 368.
[6] K. W. Y. Abell, A. J. Kirby, J. Chem. Soc., Perkin Trans. 2, 1983, 1171.
[7] (a) A. J. Kirby, F. Noem, Acc. Chem. Res., 2015, 48, 1806: (b) A. J. Kirby, J. R. Mora, F.
Noem, Biochim. Biophys. Acta., 2013, 454..
[8] (a) W. W. Krajewski, T. A. Jones, S. L. Mowbray, Proc. Natl. Acad. Sci. USA, 2005, 102,
10499; (b) H. S. Gill, G. M. U. Pfluegl, D. Eisenberg, Biochemistry, 2002, 41, 9863; (c) C.
Moreira, M. J. Ramos, P. A. Fernandes, Chem. Eur. J., 2016, 22, 9218; (d) S.ꢀH. Liaw, D.
Eisenberg, Biochemistry, 1994, 33, 675.
[9] (a) R. Kluger, Synlett, 2000, 1708. (b) M. Pal, S. L. Bearne, Org. Biomol. Chem., 2014,
12, 9760.
[10] A. Sumita, Y. Otani, T. Ohwada, Chem. Comm., 2017, 53, 1482.
28

Page 29 of 32
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C7OB02158E
[11] The reaction scheme is shown in detail in Supporting Information (from Scheme SI 1 to
Scheme SI 4).
[12] Other examples of intramolecular nucleophilic reaction are as follows: (a) P. M. Cullis,
M. J. P. Jarger, J. Chem. Soc., Perkin Trans. 2, 2002, 1538: (b) R. Kluger, J. F. Chow, J. J.
Croke, J. Am. Chem. Soc., 1084, 106, 4017: (c) J. Rahil, R. F. Pratt, J. Chem. Soc., Perkin
Trans. 2, 1991, 947.
[13] A similar mechanism was reported: E. S. Orth, T. A. S. Brandão, B. S. Souza, J. R.
Pliego, B. G. Vaz, M. N. Eberlin, A. J. Kirby, F. Noem, J. Am. Chem. Soc., 2010, 132, 8513.
[14] S, Saito, S. Saito, T. Ohwada, K. Shudo, Chem. Pharm. Bull., 1991, 39(10), 2718.
[15] G. A. Olah, C. W. McFarland, J. Org. Chem., 1971, 36, 1374.
[16] R. R. Holmes, Chem. Rev., 1996, 96, 927.
[17] Among the various calculation levels, GIAO/CPCM/PBE0/6ꢀ311++G(d,p) (solvent:
triflic acid) gave results that coincided relatively well with the experimental results. See
Supporting information, section Vꢀa.
[18] Gaussian 09, Revision C.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L.
Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S.
Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox,
Gaussian, Inc., Wallingford CT, 2010.
[19] The reaction rates of phosphate ester (13) are shown in Supporting Information (Section
IVꢀb, Table SI 1).
29