Significance of reagent addition sequence in the amidation of carboxylic acids mediated by PPh3 and I2

Article abstract of DOI:10.1039/c5ra03184b

The outcome of the amidation reaction mediated by PPh₃-I₂ was found to be highly dependent on the addition sequence of the reagents. When triethylamine was subjected to a mixture containing PPh₃, I₂, and a carboxylic acid, acid anhydride was generated almost instantly, before treatment with an amine, presumably via an attack of carboxylate ion onto the acyl function of an acyloxyphosphonium salt. Nevertheless, when a PPh₃-I₂ mixture was treated with an amine, then a carboxylic acid, prior to adding the base, amide was rapidly formed in high yield with high chemoselectivity, most likely through an intermediate O,N-pentacoordinate phosphorane species as confirmed by ESI-MS technique.

Full text of DOI:10.1039/c5ra03184b

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
Significance of reagent addition sequence in the
amidation of carboxylic acids mediated by PPh3
and I2†
Cite this: RSC Adv., 2015, 5, 25789
Received 19th February 2015
Accepted 4th March 2015
Sirilak Wangngae, Chuthamat Duangkamol, Mookda Pattarawarapan
*
and Wong Phakhodee
DOI: 10.1039/c5ra03184b
www.rsc.org/advances
The outcome of the amidation reaction mediated by PPh3–I2 was competitive formation of acid anhydride has oen been
found to be highly dependent on the addition sequence of the encountered in several cases. Although its presence has been
reagents. When triethylamine was subjected to a mixture containing well documented,5a,5b,9d to the best of our knowledge, no further
PPh3, I2, and a carboxylic acid, acid anhydride was generated almost attempt has been made to circumvent this undesired side
instantly, before treatment with an amine, presumably via an attack of reaction. Recently, during the synthesis of N-acylbenzotriazoles
carboxylate ion onto the acyl function of an acyloxyphosphonium salt. from carboxylic acids mediated by PPh3–I2.10 we also observed
Nevertheless, when a PPh3–I2 mixture was treated with an amine, then formation of benzoic anhydride upon mixing benzoic acid with
a carboxylic acid, prior to adding the base, amide was rapidly formed in PPh3 and I2 in the presence of triethylamine base. Aer a
high yield with high chemoselectivity, most likely through an inter- thorough investigation, we have discovered that when adding
mediate O,N-pentacoordinate phosphorane species as confirmed by tertiary amine base in the last step, not only anhydride forma-
ESI-MS technique.
tion could be prevented, but the desired products also rapidly
formed in high yields. This observation prompts us to extend
our investigation toward amide bond formation using various
amines with a range of nucleophilicity. Although the amidation
mediated by PPh3–I2 has previously been reported,5b the reac-
tion gave relatively low yields of the amides and, in some cases,
required long reaction times with large excess of reagents
possibly due to an involvement of anhydride side reaction.
Thus, in this study, the utility of PPh3–I2 combination was re-
explored. Indeed, the outcome of reaction was found to be
highly dependent on sequences of reagent addition. Herein, we
wish to report a new improved protocol for amidation of
carboxylic acids mediated by PPh3–I2 which allows rapid access
to a series of amides in good to excellent yields with high
chemoselectivity.
In our preliminary studies, the amidation mediated by
PPh3–I2 was investigated in the reaction between benzoic acid and
benzyl amine as model substrates. The outcome of the reactions
using different sequences of reagent addition was carefully
examined within 10 min aer the last reagent was added.
As shown in Scheme 1 (method A), when benzoic acid was
added to the mixture containing PPh3–I2 in the presence of
Et3N, benzoic anhydride was exclusively obtained almost
instantly within 5 min. Aer 10 min stirring in the presence of
benzylamine, no detectable amount of the amide product was
observed possibly due to the relatively low reactivity of the
formed anhydride. When benzylamine was added to the PPh3–I2
mixture, followed by addition of Et3N then benzoic acid
Organic reactions mediated by phosphines play important roles
in a number of functional group transformations including the
classic Appel, Garegg–Samuelsson, Wittig, Mitsunobu, and
Staudinger reactions.1 Mechanistically, these reactions are
driven by the strong oxophilicity of tertiary phosphines which,
as a result, activate alcohols, carbonyl compounds, carboxylic
acids and their derivatives toward nucleophilic substitution.
Among the available phosphines, triphenylphosphine (PPh3)
is highly attractive since it is inexpensive, readily available, and
easy to handle.1b,2 Especially, in the preparation of carboxylic
acid derivatives such as acid chlorides, amides, esters and
thioesters, PPh3 and its polymer bound analogs have been used
in combination with various additives such as NCS,3 NBS,4 I2,5
I2/Zn(OTf)2 (Tf ¼ triuoromethylsulfonyl),6 hypervalent iodi-
ne(III) reagent,7 BrCCl3,8 and diethyl azodicarboxylate (Mitsu-
nobu reaction)9 for an in situ activation of a carboxylic acid prior
to acyl displacement with an appropriate nucleophile.
A highly reactive acyloxyphosphonium species has been
proposed as the key intermediate in these reactions which acted
as an acylating agent for the acyl transfer process. Nevertheless,
Department of Chemistry and Center of Excellence for Innovation in Chemistry,
Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand. E-mail:
wongp2577@gmail.com; Fax: +66 53892277; Tel: +66 53943341
† Electronic supplementary information (ESI) available: Experimental procedure
and spectroscopic data. See DOI: 10.1039/c5ra03184b
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25789–25793 | 25789

View Article Online
RSC Advances
Communication
Table 1 Synthesis of amides promoted by Ph3P–I2 systema,b
Entry NHR1R2 1
R3CO2H 2
Time (min) % Yield of 3 (ref.)
1
2
10 (90)
10
99 (77) (ref. 5b)
Scheme 1 Outcomes of the reactions between benzylamine and
benzoic acid following different sequences of reagents addition.
99 (ref. 12)
3
4
5
6
30 (720)
30 (60)
30
93 (80)
(method B), no trace of the corresponding amide could be
detected either. Nevertheless, we were able to isolate a major
component from this reaction which was identied as (benzy-
lamino)triphenylphosphonium iodide11 based on spectroscopic
analysis. To our delight, it was found that when the amine base
was added in the last step (methods C and D), N-benzylbenza-
mide was obtained quantitatively within 10 min.
89 (63) (ref. 5b)
82 (ref. 13)
86 (ref. 14)
30
Since method C required longer time than method D for acid
activation (see ESI† for details), the scope and limitation of the
developed protocol were further evaluated following the proce-
dure described for method D except that the reaction time was
varied depending on the reactivity of the substrates. Different
types of aliphatic and aromatic amines were treated with
various carboxylic acids including aromatic acids, aliphatic
acid, and heterocyclic acid. According to Table 1, treatment of
amines 1 with carboxylic acids 2 generally provided amides 3 in
good to excellent yields within short reaction times. Quantita-
tive yields of amides were obtained in the reactions between
primary amines and benzoic acid (entries 1 and 2). The reaction
with sterically hindered tert-butylamine also gave high product
yield (entry 3). Secondary aliphatic amines as well as primary
aromatic amines gave the corresponding amides in good yields
(entries 4–7). However, the reaction with electron decient 4-
nitroaniline was sluggish providing the product in low yield
(entry 8). No signicant difference based on the electronic effect
of the substituents on benzoic acid was observed in the reac-
tions with cyclohexylamine (entries 9–11) although the presence
of strong withdrawing nitro group on benzoic acid resulted in a
slightly lower yield (entry 11).
7
15 (360)
60
90 (53) (ref. 5b)
27
8
9
15
95 (ref. 15)
96 (ref. 16)
10
15
11
30
88 (ref. 17)
12
13
15
60
87
75 (ref. 18)
14
15
60
60
98 (ref. 19)
87 (ref. 20)
Cinnamic acid exclusively gave 1,2 addition product without
Michael addition product or double bond isomerization (entry
12). In case of aliphatic acid, 5-phenylvaleric acid provided the
expected amide in moderate yield under relatively long reaction
time (entry 13). Lower yield from aliphatic acid in comparison
to those with aromatic acids implied an involvement of p–p
stacking interactions between aromatic acids and the phos-
phonium intermediates. For heterocyclic containing substrates,
both nicotinic acid and 4-aminopyridine were viable reactants
which underwent amidation without any interference (entries
14 and 15). In addition, the reaction between steric 2-methyl-
aniline with 3,5-dinitrobenzoic acid and diethylamine with
2-napthanoic acid provided the corresponding amides in high
yields (entries 16 and 17). It is interesting to note that, with our
16
60 (180)
60 (720)
94 (52) (ref. 5b)
99 (60) (ref. 5b)
17
a
The reactions were carried out by adding amine (0.49 mmol) into a
mixture containing I2 (0.49 mmol) and PPh3 (0.49 mmol) at 0 ꢀC
under N2. Aer stirring at 0 ꢀC until amine was no longer observed by
TLC (ꢁ5–30 min), carboxylic acid (0.41 mmol) was added, followed by
adding Et3N (0.82 mmol). The mixture was further stirred at room
b
temperature for
a
specied time. The reaction times and the
product yields of the reported method using PPh3–I2 as the acid
activator5b were given in parenthesis.
25790 | RSC Adv., 2015, 5, 25789–25793
This journal is © The Royal Society of Chemistry 2015

View Article Online
Communication
RSC Advances
developed strategy, both the reaction times and the product sequence of reagent addition in the methods A–D using CDCl3
yields were signicantly improved in comparison to those as a solvent. 31P NMR spectra were recorded aer each addition.
derived from the previously reported method using the PPh3/I2 Based on the results obtained (see Fig. S1–S4 in ESI† for details),
system5b as shown in parenthesis.
the general mechanisms for these reactions were proposed as
Another important factor to be considered when developing shown in Scheme 2. The 31P chemical shis of the plausible
a new method in organic synthesis is chemoselectivity. We thus intermediates were listed accordingly.
turn our attention to investigate functional groups compati-
In all reactions, addition of I2 to a solution of PPh3 resulted
bility of the protocol using carboxylic acids containing reactive in a slight shi of the free phosphine signal with an appearance
functionalities such as hydroxyl and amino groups as well as N- of two new resonance peaks at around 45 and ꢂ17 ppm which
protected a-amino acids with tert-butyloxycarbonyl (Boc), ben- were assigned to a slow equilibrium between triphenylphos-
zyloxycarbonyl (Cbz), and 9-uorenylmethyloxycarbonyl (Fmoc) phonium iodide (Ia) and pentacoordinate phosphorus diiodide
groups. The standard reaction condition was applied with all (Ib), respectively.5a–c Under method A, addition of benzoic acid
substrates without further optimization. As shown in Fig. 1, to I caused a slight shi of the signal at ꢂ16.93 to ꢂ14.83 ppm
primary aliphatic or aromatic amines reacted smoothly with which could be contributed to the formation of an intermediate
4-hydroxybenzoic acid or 4-hydroxy-3-methoxybenzoic acid to II. In the presence of Et3N, acid anhydride formed spontane-
provide 3a–d21–24 in good yields. Combination between the ously with concomitant released of Ph3PO (29.50 ppm).5c This
heterocyclic 2-picolylamine with nicotinic acid also afforded data suggested the intermediacy of an acyloxyphosphonium ion
amide 3e25 in moderate yield. In the cases of benzoic acid III though this species could not be observed directly due to its
derivatives pending with amino group, due to their limit solu- high reactivity. Aer addition of benzylamine, amide bond
bility in dichloromethane, the amide products 3f26 and 3g27 were formation did not proceed within the 10 min observation time.
obtained in moderate yields. Interestingly, no side products
from acylation of the arylamino groups were detected indicating amine led to a rapid formation of solid precipitate. 31P{H} NMR
the reactions were highly chemoselective.
spectrum of the solution showed an appearance of a 31P signal
In method B, treatment of the intermediates I with benzyl-
Benzylamine reacted with N-Boc-protected glycine to afford at 39.20 ppm (major) as well as two minor peaks at ꢂ18.70 and
the amide 3h28 in quantitative yield. However, with the highly 35.50 ppm (broad) which could be attributed to the presence of
steric hindered N-Cbz-protected phenylalanine, low yield of the aminophosphonium iodide,11 and possibly
a pentavalent
amide 3i29 was obtained. In the reaction with Fmoc-protected phosphorus species IV as well as an iminiumphosphorane V,
glycine, partially cleavage of the Fmoc group was observed by respectively.5a,30 Upon addition of Et3N, the solids which could
TLC indicating that the applied reaction condition was incom- be ammonium salts including IV and V became dissolve. As a
patible with the Fmoc group.
result, the equilibrium was shied toward the formation of
Since there are several reports using polymer supported PPh3 aminophosphonium iodide. This species was inactive toward
(PS-PPh3) to facilitate product purication,5b,5c,9e our reaction the applied acid and thus gave no amide product under the
condition was further examined using PS-PPh3 instead of the subjected reaction condition.
free phosphine in the synthesis of the amides 3a–i. As shown in
In method C, addition of benzoic acid to I led to a slight shi
Fig. 1, it was found that, in most cases, the respective amides of the 31P signals. The shi of the peak at ꢂ16.82 to ꢂ13.85 ppm
could be prepared in comparative yields using 1.5 equiv. each of could be due to the formation of specie II. Subsequent treat-
PS-PPh3 and I2.
ment with benzylamine resulted in solid precipitate, while the
To explain why the sequence of reagent addition is so crucial, 31P NMR spectrum revealed three new resonance peaks at dp
31P{H} NMR studies of the reaction between benzylamine and 39.21 ppm (minor), 34.55 (minor, broad) and ꢂ16.27 ppm
benzoic acid were carried out in an NMR tube following the (major). At this stage, acid–base reaction between the acid and
amine is most likely to be predominant which shied the
equilibrium toward species Ia (ꢂ16.27 ppm). Amino-
phosphonium iodide (39.21 ppm) and the intermediate IV
could be generated from the remained amine as minor
components. Upon treatment with Et3N, all the formed solids
became soluble. A release of Ph3PO (29.48 ppm) with concom-
itant formation of the amide product (based on TLC) indicated
that the reaction might proceed through an acylox-
yphosphonium ion III or possibly via an O,N-pentacoordinate
phosphorane VI.4b,9f The presence of an amine when III is being
generated seems to be crucial in preventing the anhydride
formation.
In method D where the addition of the carboxylic acid and
the amine was reversed, treatment of I with benzylamine again
resulted in solid precipitate. 31P-NMR spectrum of the mixture
revealed the presence of aminophosphonium iodide
Fig. 1 Synthesis of amides from carboxylic acids containing various
functionalities using PPh3 and PS-PPh3 (the yields obtained with PS-
PPh3 were given in parenthesis).
(39.16 ppm, major) as well as the ammoniumphosphorane IV
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25789–25793 | 25791

View Article Online
RSC Advances
Communication
Scheme 2 Proposed mechanism for the reactions mediated by Ph3P–I2 system following the methods A, B, C, and D. 31P NMR chemical shifts of
important intermediates were given for R1 ¼ Ph, R2 ¼ H, and R3 ¼ Ph.
(ꢂ18.74 ppm, minor) and iminiumphosphorane V (35.56 ppm, systems which will greatly enhance the reactions in terms of the
minor). Addition of a carboxylic acid did not lead to any reaction rate, product yields, and selectivity.
signicant change in these 31P signals. However, in the pres-
ence of Et3N where all the solids dissolved, apart from the
remained aminophosphonium iodide (38.90 ppm), Ph3PO
Acknowledgements
(29.50 ppm) was generated spontaneously upon amide bond
Financial support from the Thailand Research Fund through
forming.
the Royal Golden Jubilee Ph.D. Program (Grant no. PHD/0206/
When the ESI-MS technique was used to monitor the
2556) to S. W. and the Center of Excellence for Innovation in
reaction at the early stage of Et3N addition, the mass spec-
Chemistry (PERCH-CIC). are gratefully acknowledged. We also
gratefully acknowledge Chulabhorn Research Institute (CRI),
Thailand for support with ESI-MS and Dr Poonsakdi Ploypra-
dith (CRI) for valuable discussion.
trum revealed a major peak at m/z 368.1576 which was
assigned to the iminophosphorane V and (benzylamino)tri-
phenylphosphonium salt (see Fig. S5 in ESI†). Interestingly,
we were able to observe a peak at m/z 490.1915 which
conrmed the presence of the proposed pentacoordinated
phosphorous intermediate VI. This data strongly suggested
that this reaction proceeded mainly via an attack of benzoate
Notes and references
anion on to the phosphorus center of V. The generated species
1 (a) D. H. Valentine Jr and J. H. Hillhouse, Synthesis, 2003,
VI then concomitantly activates both the carboxylic acid and
317–334; (b) D. W. Allen, Organophosphorus Chem., 2012,
the amine moieties to participate in an intramolecular acyl
41, 1–55; (c) S. Xu and Z. He, RSC Adv., 2013, 3, 16885–
16904; (d) A. D. Kosal, E. E. Wilson and B. L. Ashfeld,
Angew. Chem., Int. Ed., 2012, 51, 12036–12040; (e) M. Koehn
and R. Breinbauer, Angew. Chem., Int. Ed., 2004, 43, 3106–
3116; (f) E. Saxon, J. I. Armstrong and C. R. Bertozzi, Org.
Lett., 2000, 2, 2141–2143; (g) B. L. Nilsson, L. L. Kiessling
and R. T. Raines, Org. Lett., 2000, 2, 1939–1941; (h)
A. D. Kosal, E. E. Wilson and B. L. Ashfeld, Chem.–Eur. J.,
2012, 18, 14444–14453.
2 L. F. Pedrosa, Synlett, 2008, 1581–1582.
3 P. Froeyen, Synth. Commun., 1995, 25, 959–968.
4 (a) P. Froyen, Phosphorus, Sulfur Silicon Relat. Elem., 1994, 91,
145–151; (b) P. Froyen, Tetrahedron Lett., 1997, 38, 5359–
5362; (c) P. Gopinath, R. S. Vidyarini and
S. Chandrasekaran, J. Org. Chem., 2009, 74, 6291–6294; (d)
transfer process to provide an amide with a release of Ph3PO.
It is important to point out that although the commonly
proposed acyloxyphosphonium intermediate III could not be
detected by both 31P NMR and ESI-MS techniques, its
formation still could not be completely ruled out since the
reaction was under equilibrium.
In summary, a new improved protocol for amidation of
carboxylic acids mediated by PPh3–I2 system was described
through changing the sequence of the reagents addition. The
method is proven to be mild, highly effective with functional
group tolerant and high chemoselectivity. Since acylox-
yphosphonium species have been proposed to be involved in
various nucleophilic acyl substitution reactions which should
not only be limited to the PPh3–I2 activation, this technique
could be potentially applied to other phosphine mediated
P. Gopinath, C. Debasree, R. S. Vidyarini and
S. Chandrasekaran, Tetrahedron, 2010, 66, 7001–7011.
25792 | RSC Adv., 2015, 5, 25789–25793
This journal is © The Royal Society of Chemistry 2015

View Article Online
Communication
RSC Advances
5 (a) S. P. Morcillo, L. Alvarez de Cienfuegos, A. J. Mota, 18 I.-H. Kim, F. R. Heirtzler, C. Morisseau, K. Nishi, H.-J. Tsai
J. Justicia and R. Robles, J. Org. Chem., 2011, 76, 2277– and B. D. Hammock, J. Med. Chem., 2005, 48, 3621–3629.
2281; (b) A. Kumar, H. K. Akula and M. K. Lakshman, Eur. 19 X. Cui, Y. Zhang, F. Shi and Y. Deng, Chem.–Eur. J., 2011, 17,
J. Org. Chem., 2010, 2709–2715; (c) P. J. Garegg, T. Regberg, 1021–1028.
J. Stawinski and R. Stroemberg, J. Chem. Soc., Perkin Trans. 20 K. Sakakibara, N. Yoneshima and T. Osawa, Preparation of
2, 1987, 271–274; (d) R. Caputo, E. Cassano,
L. Longobardo, D. Mastroianni and G. Palumbo, Synthesis,
1995, 141–143.
6 N. Mamidi and D. Manna, J. Org. Chem., 2013, 78, 2386–2396.
7 J. Tian, W.-C. Gao, D.-M. Zhou and C. Zhang, Org. Lett., 2012,
14, 3020–3023.
N-(4-pyridyl)benzamide derivatives as cardiotonics, Chem.
Abstr., 1988, 108, 131583.
21 V. Bavetsias, A. Faisal, S. Crumpler, N. Brown,
M. Kosmopoulou, A. Joshi, B. Atrash, Y. Perez-Fuertes,
J. A. Schmitt, K. J. Boxall, R. Burke, C. Sun, S. Avery,
K. Bush, A. Henley, F. I. Raynaud, P. Workman, R. Bayliss,
S. Linardopoulos and J. Blagg, J. Med. Chem., 2013, 56,
9122–9135.
8 Y. Kawagoe, K. Moriyama and H. Togo, Tetrahedron, 2013,
69, 3971–3977.
9 (a) J. McNulty, A. Capretta, V. Laritchev, J. Dyck and 22 H. C. Hansen, F. S. Chiacchia, R. Patel, N. C. W. Wong,
J. Robertson Al, J. Org. Chem., 2003, 68, 1597–1600; (b)
D. L. Hughes and R. A. Reamer, J. Org. Chem., 1996, 61,
V. Khlebnikov, R. Jankowska, K. Patel and M. M. Reddy,
Eur. J. Med. Chem., 2010, 45, 2018–2023.
2967–2971; (c) C. Ahn, R. Correia and P. DeShong, J. Org. 23 I. A. Pearl and D. L. Beyer, J. Am. Chem. Soc., 1953, 75, 2627–
Chem., 2002, 67, 1751–1753; (d) C. Ahn and P. DeShong, J. 2630.
Org. Chem., 2002, 67, 1754–1759; (e) G. L. Thomas, 24 O. Vajragupta, P. Boonchoong and Y. Wongkrajang, Bioorg.
C. Boehner, M. Ladlow and D. R. Spring, Tetrahedron, Med. Chem., 2000, 8, 2617–2628.
2005, 61, 12153–12159; (f) P. J. Garegg, T. Regberg, 25 L. A. Tetyueva and P. A. Petyunin, Zh. Obshch. Khim., 1958,
J. Stawinski and R. Stroemberg, J. Chem. Soc., Perkin Trans.
2, 1987, 271–274.
28, 739–742.
26 X. Wu and L. Hu, J. Org. Chem., 2007, 72, 765–774.
10 C. Duangkamol, S. Wangngae, M. Pattarawarapan and 27 S.-K. Anandan and R. D. Gless, Bioorg. Med. Chem. Lett.,
W. Phakhodee, Eur. J. Org. Chem., 2014, 7109–7112. 2010, 20, 2740–2744.
11 G. Albertin, S. Antoniutti and J. Castro, Z. Anorg. Allg. Chem., 28 K. C. Nadimpally, K. Thalluri, N. B. Palakurthy, A. Saha and
2014, 640, 136–139. B. Mandal, Tetrahedron Lett., 2011, 52, 2579–2582.
12 J.-C. Hsieh and C.-H. Cheng, Chem. Commun., 2005, 4554– 29 B. Kolesinska and Z. J. Kaminski, Org. Lett., 2009, 11, 765–
4556. 768.
13 T. Ohshima, T. Iwasaki, Y. Maegawa, A. Yoshiyama and 30 (a) J. Garcia, F. Urpi and J. Vilarrasa, Tetrahedron Lett., 1984,
K. Mashima, J. Am. Chem. Soc., 2008, 130, 2944–2945.
14 K.-J. Han, B. S. Tae and M. Kim, Org. Prep. Proced. Int., 2005,
37, 198–203.
15 G. Pelletier, W. S. Bechara and A. B. Charette, J. Am. Chem.
Soc., 2010, 132, 12817–12819.
16 N. Auzeil, M. Largeron and M.-B. Fleury, J. Chem. Soc., Perkin
Trans. 2, 1999, 1703–1709.
17 F. Abraham, S. Ganzleben, D. Han, P. Smith and
H.-W. Schmidt, Macromol. Chem. Phys., 2010, 211, 171–181.
25, 4841–4844; (b) R. Bielsa, R. Navarro, E. P. Urriolabeitia
and A. Lledos, Inorg. Chem., 2007, 46, 10133–10142; (c)
K. E. Elson, I. D. Jenkins and W. A. Loughlin, Aust. J.
Chem., 2004, 57, 371–376; (d) S. S. Kim, K. H. Lee and
B. Liu, Bull. Korean Chem. Soc., 1998, 19, 879–880; (e)
G. F. de la Fuente and J. E. Huheey, Phosphorus, Sulfur
Silicon Relat. Elem., 1993, 78, 23–36; (f) K.-W. Lee and
L. A. Singer, J. Org. Chem., 1974, 39, 3780–3781.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25789–25793 | 25793