Reaction of Boranephosphonate Diesters with Pyridines or Tertiary Amines in the Presence of Iodine: Synthetic and Mechanistic Studies

Article abstract of DOI:10.1021/acs.joc.9b03506

Boranephosphonate diesters react with heteroaromatic and certain tertiary amines in the presence of an oxidant (I₂) to afford the boron-modified phosphodiester analogues containing a P-B-N structural motif. Our multinuclear ³¹P and ¹¹B NMR spectroscopy studies lend support for a two-step mechanism involving generation of a λ³-boranephosphonate intermediate that immediately coordinates an amine in the solvent cage, leading to B-pyridinium or B-ammonium boranephosphonate betaine derivatives. We found that the type of the solvent used (e.g., dichloromethane vs acetonitrile) significantly affected the course of the reaction, resulting in either formation of boron-modified derivatives or loss of the boron group with a subsequent oxidation of the phosphorus atom. In aprotic, electron-donating, polar solvents., e.g., acetonitrile (ACN) and tetrahydrofuran (THF), a λ³-boranephosphonate intermediate can also coordinate solvent molecules forming P-B-ACN or P-B-THF complexes that may influence the type of the products formed.

Full text of DOI:10.1021/acs.joc.9b03506

Subscriber access provided by Murdoch University Library
Article
Reaction of boranephosphonate diesters with pyridines or tertiary
amines in the presence of iodine. Synthetic and mechanistic studies.
Justyna Go##biewska, and Jacek Stawinski
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b03506 • Publication Date (Web): 19 Feb 2020
Downloaded from pubs.acs.org on February 19, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 36
The Journal of Organic Chemistry
1
1
2
3
4
5
6
7
8
9
Reaction of boranephosphonate diesters with pyridines or tertiary amines
in the presence of iodine. Synthetic and mechanistic studies.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Justyna Gołębiewska and Jacek Stawinski*
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14,
6
1-704 Poznań, Poland
Corresponding author:
Jacek Stawinski: tel. (+) 48 61 852 85 02 ext. 181, fax (+) 48 61 852 05 32, e-mail
jstaw@ibch.poznan.pl
Keywords: boranephosphonates; pyridiniumboranephosphonates; oxidation with
iodine; mechanism; nucleotide analogues
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 36
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract:
Boranephosphonate diesters react with heteroaromatic and certain
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
tertiary amines in the presence of an oxidant (I ) to afford the boron-modified
2
phosphodiester analogues containing a P-B-N structural motif. Our multinuclear 31P
11
and B NMR spectroscopy studies lend support for a two steps mechanism involving
3
generation of a  -boranephosphonate intermediate that immediately coordinates an
amine in the solvent cage, leading to B-pyridinium or B-ammonium
boranephosphonate betaine derivatives. We found that type of the solvent used (e.g.,
dichloromethane vs acetonitrile) significantly affected course of the reaction resulting
either in the formation of boron-modified derivatives or lost of the boron group with a
subsequent oxidation of the phosphorus atom. In aprotic, electron-donating, polar
3
solvents. e.g., acetonitrile (ACN), tetrahydrofuran (THF), a  -boranephosphonate
intermediate can also coordinate solvent molecules forming P-B-ACN or P-B-THF
complexes that may influence the type of the products formed.
ACS Paragon Plus Environment

Page 3 of 36
The Journal of Organic Chemistry
3
1
2
3
4
5
6
7
8
9
INTRODUCTION
During the last decades boranephosphonate analogues emerged as
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
important bioisosters of nucleic acids and their components. New properties
imparted by the presence of the BH group, located in a nonbridging position
3
of the phosphate function, e.g., chirality at the phosphorus center, resistance
2
to nucleases, high lipophilicity, low toxicity, etc. constituted basis for
potential applications of these analogues as antisense/antigene agents, siRNA
gene modulators,1a,2a,3 alternative substrates for polymerase chain reaction
4
5
(
PCR), or new structural motifs for nucleic acids-based diagnostics.
Apart from these biology-related applications, there has been an
increasing interest in exploring boranephosphonates/boranephosphinates as
new, potentially important synthetic equivalents for P(III) phosphorus
synthons with the hidden P-H functionality (e.g. H-phosphonate or H-
phosphinate esters). Temporary conversion of the P-H function into the P-BH3
5
4
one, freezes these compounds in a   phosphonate form that, in contrast to
that of the parent P-H compounds, cannot isomerize to a nucleophilic 33
tervalent form with a lone electron pair on the phosphorus. Thus,
boranephosphonates are generally more resistant to various chemical
transformations than H-phosphonate derivatives, but the P-H function and its
reactivity can be retrieved at the end of synthesis and subject, if so desired, to
additional synthetic tasks. The feasibility and usefulness of such an approach
has been successfully demonstrated in the synthesis of biologically important
phosphorus compounds.6
7
Recently, Caruthers et al. exploring further chemistry of
boranephosphonates proposed the P-B bond as a new chemical marker for a
postsynthetic modification of oligonucleotides, and showed that the BH3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 36
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
group under mild, oxidative conditions (e.g., in the presence of iodine) can be
stereospecifically replaced by various nucleophiles, e.g., amines, alcohols,
thiols, etc, forming the corresponding P(V) derivatives. When such oxidative
transformation is carried out in the presence of pyridine (or other
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
heteroaromatic, or certain 3° amines), boranephosphonate diester derivatives
bearing a pyridine or a 3° amine moiety attached to the borane group, are
8
formed. The authors showed that a pyridiniumboranephosphonate
modification, when present in oligonucleotides, facilitated cellular uptake of
8
such molecules, and modulated reducing properties of borane-modified
nucleic acids during formation of DNA-templated metal nanostructures.8-9
With a focus on chemistry of boranephosphonates as important
nucleotide analogs and potentially valuable synthetic intermediates, we
recently investigated mechanistic aspects of a formal replacement of the BH3
1
0
group by amines, promoted by iodine. In contradistinction to the literature
7
report we found that this transformation occurred with total inversion of the
configuration at the phosphorus center and involved the intermediacy of the
1
0
corresponding H-phosphonate diesters. These findings are of importance in
the context of a possible use of boranephosphonate analogues in postsynthetic
modifications of nucleic acids.
In this paper we investigated the reaction of boranephosphonate
8
diesters with heteroaromatic and 3° amines, promoted by iodine. This
7
reaction, in contrast to that involving 1° or 2° amines, preserves the P-B bond
in boranephosphonates, and leads to the formation of a new structural motif,
8
P-B-N, where N is part of a heteroaromatic or 3° amine. Taking into account
the potential importance of B-modified boranephosphonate nucleotide
analogues in nucleic acid chemistry, medicinal chemistry, or metal
nanotechnology, we undertook these studies to gain better understanding of
the underlying chemistry, evaluate substrate scope of this reaction in term of
ACS Paragon Plus Environment

Page 5 of 36
The Journal of Organic Chemistry
5
1
2
3
4
5
6
7
8
9
the kind of the amine used, and to clarify some mechanistic aspects that can
be of synthetic relevance. We used in our studies multinuclear NMR
spectroscopy to detect putative intermediates and define a mechanistic
framework of the whole transformation. As a model system for our studies
we chose a reaction of diethyl boranephosphonate diester 1 with pyridine (or
its derivatives), or certain 3° amines promoted by iodine under various
experimental conditions.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
For the reaction of boranephosphonate diesters with 1° and 2° amines
in the presence of iodine, and that with heteroaromatic and certain 3° amines,
two distinct mechanisms have been advocated. For that involving 1° and 2°
amines, a nucleophilic substitution of the activated borane group by iodide to
form the corresponding iodophosphate with inversion of the configuration at
7
phosphorus, was proposed as a key reaction step. For the second reaction,
that makes use of heteroaromatic or 3° amines as substrates and preserves the
borane group in the molecule, a different mechanism, involving a four center
addition of an N-iodoammonium species to the borane moiety with a
simultaneous oxidation of the boron-bound hydride, was suggested.8
Studies from our laboratory showed that the mechanism suggested
for the reaction of boranephosphonates with 1° and 2° amines was incorrect.10
We proposed that the most likely reaction pathway for this process involved
3
the intermediacy of an electron deficient  -boranephosphonate species that
coordinated the amine present in the reaction mixture to form aminoborane
intermediate, followed by its dissociation to the corresponding H-
phosphonate, and finally after oxidation, a phosphoramidate diester (Scheme
1).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 36
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Reaction of boranephosphonate diesters with amines, promoted by
iodine.
As for the reaction of boranephosphonate diesters with
heteroaromatic and 3° amines promoted by iodine, our working hypothesis
was that the initial, key mechanistic steps were the same as those proposed
for 1° and 2° amines, i.e., involved formation of an electron deficient 3-
boranephosphonate species, followed by coordination of the amine used
(Scheme 1). The fate of the reaction is probably determined by stability of the
3
formed intermediates, i.e., complexes of  -boranephosphonate with amines
(ABP vs PyBP, Scheme 1). In case of instability of an
aminoboranephosphonate complex (e.g ABP), the reaction proceeds further to
the corresponding P(V) derivative, but if the complex is stable enough (e.g.,
pyridiniumborane complex PyBP), it may remain the final product of the
reaction. This hypothesis was based on the expected reactivity of
boranephosphonate diesters towards oxidants, the known affinity of electron-
31
deficient boron species towards amines, and some preliminary P NMR
spectroscopy experiments on this reaction.
Although the conversion of boranephosphonate diesters into the
corresponding pyridiniumborane- or ammoniumboranephosphonate
derivatives formally looks like a simple two steps reaction, a careful scrutiny
ACS Paragon Plus Environment

Page 7 of 36
The Journal of Organic Chemistry
7
1
2
3
4
5
6
7
8
9
of the reaction conditions revealed that it might be a complex process with
various competing reaction pathways and multiple equilibria systems
8
involved. In agreement with these, Caruthers et al. reported on the formation
on variable amounts of phosphates in this reaction (the lost of the borane
group and oxidation at the phosphorus centre), and a complex dependency
on the amount of pyridine and the oxidant used. For our part, we observed
also irreproducibility of certain experiments that can point to sensitivity of the
reaction to factors that have not been fully appreciated.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To get a deeper insight into the mechanistic features of the
investigated reaction and to pinpoint synthetically important factors, we
carried out series of experiments to identify the intermediates involved,
clarify the role of the solvents used, and to elucidate a possible involvement of
amine-oxidant complexes in the oxidation step.
Reaction of boranephosphonate 1 with pyridine promoted by iodine in
different solvents.
Preliminary experiments showed that simple diethyl
boranephosphonate 1 reacted with pyridine in the presence of iodine
analogously to the derivatives containing more complex esters residues (e.g.,
nucleoside moieties), and for this reason we selected it as a model
boranephosphonate diester for our studies. We expected that type of the
solvent used for this reaction can be important on at least two counts. Firstly,
iodine is known to form brown- or violet-colored solutions depending upon
the solvent. Brown solutions, which contain iodine coordinated to the solvent
molecules appear to be more reactive than those with presumably free iodine
1
1
molecules, the violet ones. Secondly, assuming an electron-deficient borane
intermediate to be involved in the investigated reaction (Scheme 1), the
coordination ability or electron-donating properties of the solvent may be of
particular importance for stabilization of such species. Thus, taking into
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 36
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
account solubility requirements for the substrates and the theoretical premises
mentioned above, we chose dichloromethane (DCM) and acetonitrile (ACN)
as two differing solvents for the studied reaction.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Reaction of diethyl boranephosphonate 1 with pyridine and iodine in dichloromethane
vs acetonitrile.
A standard protocol for the investigated reaction consisted of adding
+
iodine (3 equiv.) to a ca. 0.1 M solution of boranephosphonate 1 (TEAH salt)
in a selected solvent containing a heterocyclic or 3° amine (10 equiv.). First,
we run the reaction in DCM with pyridine as a heterocyclic amine. Inspection
31
of the P NMR spectrum after 1 h since the addition of iodine showed a
complete disappearance of the starting material 1 ( = 94.2 ppm) and the
P
presence of two signals at  = 59.8 ppm (m, ca. 80%) and  = 45.0 ppm (m, ca.
P
P
2
0%). The compound resonating at ca. 60 ppm was isolated from the reaction
1
13
11
mixture and identified ( H, C, B NMR, and HRMS) as the expected product,
diethyl pyridiniumboranephosphonate 2 (Scheme 2, PyBP). For identification
1
2
of the side product at  = 45.0 ppm, see later in the text.
P
O
H
EtO P B N
DCM
ACN
H
EtO
1
2
. 10 equiv. Py
. 3 equiv. I2
O
H
2 (PyBP)
EtO P B H
6
0 min.
H
O
O
EtO
EtO P O P OEt
EtO OEt
1 (BP)
TEAH+ salt
(for clarity, this cation part
3
has been omitted on all the
following Schemes)
Scheme 2. Different courses of the reaction 1 + pyridine + iodine in
3
1
dichloromethane (DCM) vs acetonitrile (ACN). For the relevant P
and B NMR spectra, see Fig. S1 and Fig. S2 in the supporting
information.
11
The analogous reaction of BP 1 with pyridine and iodine in ACN took
completely different course and afforded tetraethyl pyrophosphate 3 ( = -
P
ACS Paragon Plus Environment

Page 9 of 36
The Journal of Organic Chemistry
9
1
2
3
4
5
6
7
8
9
1
3.0 ppm) as the sole phosphorus containing product (Scheme 2). This
indicated deboronation occurring during the course of the reaction but its
mechanistic pathway was far from clear. To get some insight into a possible
3
1
mechanism, we followed progress of the reaction in acetonitrile by P NMR
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
1
spectroscopy. In the first P NMR spectrum recorded (after ca. 5 min.), a
compound resonating at ca.  = 60 ppm (m, ca. 80%) appeared in the reaction
P
mixture and during 30-60 min it gradually faded away to be converted into
the final product of this reaction, pyrophosphate 3. The shape of the signal
and its chemical shift values ( = 59.8 ppm and  = -13.4 ppm) strongly
P
B
suggested formation of PyBP 2 but, in contradistinction to the reaction in
DCM, this species was unstable under the reaction conditions and ultimately
(
within 60 min) produced pyrophosphate 3 (Scheme 3).
O
H
O
O
O
H
ACN
30-60 min
EtO P B H
EtO P B N
EtO P O P OEt
EtO OEt
H
1
2
. 10 equiv. Py
H
EtO
EtO
. 3 equiv. I2
3
1
2
Scheme 3. A transient formation of PyBP 2 in the reaction of 1 with pyridine
3
1
11
and iodine in acetonitrile. For the relevant P and B NMR spectra,
see Fig. S1 in the supporting information.
Since the isolated PyBP 2 was found to be completely stable in an
acetonitrile solution, to prove its intermediacy in the investigated reaction, we
reacted BP 1 with pyridine (10 equiv.) and iodine (3 equiv.) in ACN in the
presence of the added putative intermediate, PyBP 2 (1 equiv.). Somewhat
surprisingly, in this experiment BP 1 was converted into PyBP 2 that did not
undergo further transformation (few hours) to the expected under these
conditions final product 3. However, when we doubled the amount of iodine
and pyridine in the above reaction, both substrates (BP 1 and PyBP 2) were
converted into tetraethyl pyrophosphate 3 (Fig. 1). Thus, in this context, two
questions should be addressed: (i) why sometimes the produced PyBP 2 was
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 36
1
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
stable even when the investigated reaction was run in acetonitrile, and related
to it, (ii) what is a possible mechanism of transformation of PyBP 2 into
pyrophosphate 3 during the course of the reaction in acetonitrile?
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Fig. 1. The reaction of boranephosphonate 1 with pyridine (20 equiv.) in
acetonitrile mediated by iodine (6 equiv.), in the presence of the added
postulated intermediate, pyridiniumboranephosphonate 2 (1 equiv.).
1
1
31
(a) B NMR spectrum of 1+2 before the addition of iodine; (b) P NMR
spectrum of the same reaction mixture as in (a); (c) P NMR spectrum
31
of the reaction mixture as in (b) but after a few minutes after adding
iodine.
The transformation of PyBP 2 into pyrophosphate 3 during the course
3
1
of the reaction in ACN (Scheme 3) occurred without any detectable by the P
NMR spectroscopy intermediates, and thus no mechanistic clue existed to this
ACS Paragon Plus Environment

Page 11 of 36
The Journal of Organic Chemistry
1
1
1
2
3
4
5
6
7
8
9
process. However, on a chemical ground one can assume as plausible an
initial nucleophilic attack of iodide (present in the reaction mixture) on carbon
1
3
C2 of the pyridinium cation moiety in 2 to form aminoboranephosphonate
intermediate 4 (Scheme 4). Such a species, or its protonated form, was
expected to be unstable and should dissociate to generate first phosphonate
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
anion 5 (or the corresponding H-phosphonate), and after its oxidation by
iodine in the presence of spurious water, pyrophosphate 3. This reaction
pathway was partly supported by the findings that under strictly anhydrous
conditions, also pyridinium adduct of metaphosphate 6 was formed.14
O
O
EtO P O P OEt
EtO OEt
I
H
I
H
I2
H2O
O
O
O
P
H
H
H
3
I
EtO P B N
EtO P B N
EtO
+ B N
H
ACN
H
H
O
EtO
EtO
EtO
I2
EtO P O
N
2
4
5
6
Scheme 4. A putative mechanism for the conversion of PyBP 2 during the
reaction of BP 1 with pyridine and iodine in acetonitrile.
As for the variable persistence of PyBP 2 during the reaction in
acetonitrile, we carried out a series of experiments in which we reacted BP 1
with pyridine (10 equiv.) by varying the amount of iodine (0.5, 1, 1.5, 2, and 3
equiv.). We observed that with less iodine (0.5 - 1.5 equiv.), the produced
PyBP 2 was stable under the reaction conditions (24 h), while larger excess of
iodine (2-3 equiv.) favoured decomposition of 2 into tetrasubstituted
pyrophosphate 3 (ca. 1h). These can be rationalized in light of the mechanism
1
5
in Scheme 4, by assuming a possible formation of polyiodides with excess of
iodine that could be more effective nucleophiles than iodides.
Another important aspect of the reaction in acetonitrile, in contrast to
that run in DCM, was absence of the side-product resonating at ca.  = 45
P
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 36
1
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ppm. This could be of synthetic importance and possibly might provide an
additional mechanistic insight into the investigated reaction. Since the most
likely scenario for the formation of side-products in this reaction was a
subsequent oxidation of the initially formed PyBP 2 by iodine, first we
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
doubled the amount of iodine and pyridine (6 and 20 equiv., respectively) in
the reaction in DCM. Unfortunately, this did not result in an increase in side-
product formation. However, repeated addition of new portions of iodine and
pyridine to the reaction mixture (three times 3 and 10 equivalents,
respectively) each time generated more by-product resonating at ca.  = 45
P
1
6
ppm (increase of ca. 20% upon each addition). A similar phenomenon was
also observed when isolated PyBP 2 was subjected to the incremental addition
of iodine and pyridine in DCM. We speculated that a plausible explanation
why the reaction demanded a repeated addition of several portions of fresh
iodine and pyridine could be that oxidation of PyBP 2 apparently required
more reactive oxidizing species than that involved in oxidation of BP 1. Such
species could be a short lived and transiently generated upon mixing of
iodine with pyridine (more on iodine-pyridine complexes, see later in the
text). At this point we opted for two possible structures for this side-product,
namely, iodopyridiniumboranephosphonate 7 (compound with a betaine
structure) and dipyridiniumboranephosphonate 8 (a cationic species), shown
in Scheme 5.
O
I
EtO P B N
H
I
EtO
I2
7
O
O
EtO P B
EtO
H
N
pyridine
EtO P B N
DCM
H
H
EtO
py
O
H
2
N
EtO P B
EtO
3-pyridiniumborane
I-
N
intermediate
8
ACS Paragon Plus Environment

Page 13 of 36
The Journal of Organic Chemistry
1
3
1
2
3
4
5
6
7
8
9
Scheme 5. A possible outcome of oxidation of PyBP 2 with iodine-pyridine
reagent system.
Propitiously enough, we succeeded in isolation from the reaction
mixture of the side-product resonating at  = 45.0 ppm and found that it
P
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
contained two pyridine moieties ( H NMR). This, and other spectral analysis
13
31
11
data ( C, P, B NMR, and HRMS) allowed us to determine the structure of
this side-product as derivative of dipyridiniumboranephosphonate 8.
Since formation of the side-product 8 was not observed when the
investigated reaction was run in acetonitrile, it seemed that further oxidation
of PyBP 2 by iodine did not occurred in this solvent. This paralleled our
previous observation that also oxidation of BP 1 in acetonitrile was slower
than that in dichloromethane and usually required more iodine.10
Preactivation of boranephosphonate 1 with iodine in dichloromethane vs acetonitrile.
The widely differing reactivity of BP 1 in dichloromethane compared
to acetonitrile (Scheme 2) prompted us to investigate what were the initial
products of activation of boranephosphonate diesters with iodine in these
solvents. To this end we reacted BP 1 with 3 equiv. of iodine separately in
3
1
11
DCM and in ACN, and followed progress of these reactions by the P and B
NMR spectroscopy. We found that in DCM, the major reaction product
(
>70%) resonated at ca.  = 69 ppm and  = -39 ppm, and in ACN at  = 75
P
B
P
ppm and  = -25 ppm (also the major product, >70%). Upon mixing these two
B
3
1
reaction mixtures, only one signal in the P NMR (at ca. 75 ppm) and one in
1
1
the B NMR (at ca. -25 ppm) spectra, were observed. This might have
indicated that the same product was formed in both reactions, and the
observed differences in chemical shifts values were due to differences in the
solvents’ polarity. However, additional experiments testified to something to
the contrary. When to the reaction mixture in ACN, dichloromethane was
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 36
1
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
added (1:1, v/v), this did not affect the chemical shift values of the signals in
3
1
11
P (ca. 75 ppm) and B NMR (-25 ppm) spectra. However, addition of
acetonitrile to the reaction mixture in DCM (1:1, v/v), resulted in an
immediate change in the positions of the NMR signals from  = 69 and  = -
P
B
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
9 ppm to those observed in the reaction mixture in ACN ( = 75 and  = -25
P
B
ppm). By adding only limited amounts of acetonitrile to the reaction mixture
3
1
in DCM, two resonances in the P NMR (at ca. 67 ppm and 75 ppm) and in
11B NMR spectra (at ca. -39 and -25 ppm) could be observed. These
experiments strongly suggested that (i) in the preactivation reaction of BP 1
with iodine in DCM and in the reaction in ACN, different products were
formed, and (ii) the product formed in DCM was apparently converted to that
formed in acetonitrile, upon addition of the later solvent.
Although none of these preactivation products could be isolated, the
MS spectra of the crude reaction mixtures and the observed reactivity of the
intermediates formed, pointed to iodoboranephosphonate 9 (the preactivation
in DCM,  = 69 and  = -39 ppm) and to a new type of complex of
P
B
3
acetonitrile- -boranephosphonate 10, (the preactivation in ACN,  = 75 ppm
P
and  = -25 ppm), as the most likely structures (Scheme 6). Mechanistic
B
aspects of how these intermediates could be formed will be discussed later in
this work.
O
H
O
H
1
0 equiv. Py
ACN
0 equiv. Py
DCM
ACN
EtO P B I
EtO P B N
O
H
H
H
EtO
3
equiv. I2
EtO
2
EtO P B H
9
H
EtO
O
H
O
O
1
1
EtO P B N C
EtO P O P OEt
EtO OEt
H
EtO
10
3
Scheme 6. Products of the preactivation of BP 1 with iodine in DCM and in
ACN, and their subsequent reactions with the added pyridine.
ACS Paragon Plus Environment

Page 15 of 36
The Journal of Organic Chemistry
1
5
1
2
3
4
5
6
7
8
9
As a next step we studied reactivity of the generated intermediates 9
and 10 towards pyridine. We found that addition of pyridine (10 equiv.) to
the reaction mixture containing the putative iodoborane intermediate 9 in
DCM afforded immediately the expected product,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pyridiniumboranephosphonate 2, presumably via a nucleophilic substitution
at the boron center (Scheme 6). This might have supported the view that 9
could be an intermediate involved in the reaction of boranephosphonate
diesters with pyridine promoted by iodine. However, the analogous reactions
with the added 2,6-lutidine or triethylamine did not produce the anticipated
1
7
products, i.e., the corresponding 2,6-lutidiniumborane- or
triethylammoniumboranephosphonate, although these were formed in the
reactions without the preactivation step (compounds 12b and 12l, vide infra).
On this basis, and assuming that the same mechanism operated for all the
investigated amines, we could dismiss iodoboranephosphonate 9 as a
common reaction intermediate, and considered instead a path via a four-
center transition state addition, or a mechanism involving the intermediacy of
3
an electron-deficient  -boranephosphonate species (vide infra and Scheme 1)
for the formation of products 2 and 12a-l.
The reactivity of intermediate 10 generated in acetonitrile turned out
to be somewhat unexpected since the addition of pyridine (10 equiv.) resulted
in an immediate formation of pyrophosphate 3 (Scheme 6). This reaction, in
contradistinction to that in Scheme 3, was fast, proceeded without
pyridiniumboranephosphonate 2 as an intermediate, and thus could be an
alternative reaction pathway leading to pyrophosphate 3. A possible
explanation for the observed reactivity of 10 could be that a betaine-type of
structure of this complex should make the B-N bond stronger and effectively
prevent a nucleophilic substitution of the acetonitrile moiety by pyridine at
the boron center (path a in Scheme 7). On the other hand, the presence of a
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 36
1
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
positive charge on the nitrile nitrogen in 10, might invite a nucleophilic attack
of pyridine on the nitrile carbon (path b in Scheme 7) possibly generating an
aminoborane intermediate 10’. This step, which can be considered analogous
1
8
19
to that in the Pinner or the Ritter reaction, should lead to labilization of the
P-B bond and dissociation of the aminoborane intermediate 10’ into
phosphonate anion 5 and the borane part, and then in the presence of iodine,
to subsequent P-oxidation reactions, as it was discussed above for the
transformation shown in Scheme 4.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
H
N
B
EtO P B N
a
H
2
O
P
a
b
EtO
H
H
EtO
N C
O
P
H
O
O
EtO
b
EtO
B N C
H
EtO P O P OEt
N
10
N
EtO
EtO OEt
10’
3
Scheme 7. A possible mechanism for the reaction of acetonitrile-3-
boranephosphonate complex 10 with pyridine in the presence of
iodine.
In line with the mechanism suggested in Scheme 7, the added 2,6-
lutidine (10 equiv.) or triethylamine (10 equiv.) reacted similarly as pyridine
2
0
with acetonitrile-borane complex 10, to afford pyrophosphate 3 (see also
later in the text).
Reactivity of boranephosphonate 1 in other solvents.
High sensitivity of boranephosphonate diesters to the nature of the
solvent used for the reaction, prompted us to investigate the activation
process of BP 1 and its reaction with amines mediated by iodine in different
organic solvents, namely, monoglyme (1,2-dimethoxyetane, DME), dioxane,
tetrahydrofuran (THF), N,N-dimethylformamide (DMF), and toluene. The
reactions of 1 with iodine alone (3 equiv.; the preactivation process) in
ACS Paragon Plus Environment

Page 17 of 36
The Journal of Organic Chemistry
1
7
1
2
3
4
5
6
7
8
9
monoglyme, dioxane, and toluene occurred similarly, but less efficiently than
that in DCM, producing iodoboranephosphonate 9 (Scheme 6 and 8) as a
minor product (e.g., in DME, 50% of unreacted 1, 25% of 9, and 25% of diethyl
phosphoroiodidate). In DMF the reaction of BP 1 with iodine afforded
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
1
immediately only deboronated products without any detectable by the P
NMR spectroscopy intermediates.
We observed that the produced iodoboranephosphonate 9 was much less
stable in monoglyme or dioxane than in DCM (>24 h) and decomposed
(deboronation and oxidation) to phosphate derivatives within 1 h.
O
H
10 equiv. Py
DME, dioxane
or toluene
EtO P B
I
O
H
O
H
H
EtO
3 equiv. I2
EtO P B H
9
THF
EtO P B N
H
H
EtO
EtO
THF
O
H
2
1
EtO P B O
10 equiv. Py
H
EtO
11
Scheme 8. Preactivation of BP 1 with iodine in monoglyme (DME), dioxane,
toluene, and THF.
The preactivation of BP 1 with iodine in THF did not afford
3
1
iodoboranephosphonate 9, but a new compound resonating in the P NMR
1
1
spectrum at 78.2 ppm, and in the B NMR, at -7.3 ppm. This later chemical
shift value indicated presence of a boron-oxygen bond, and thus we
3
tentatively identified this compound as a THF- -boranephosphonate
complex 11 (Scheme 8), analogously to that containing the acetonitrile instead
of the THF moiety (compound 10 in Scheme 6). This assignment was
substantiated by the experiment in which iodoboranephosphonate 9,
generated in DCM ( = 69.4 ppm and  = -38.7 ppm), was converted into the
P
B
3
putative THF- -boranephosphonate complex 11 ( = 78.2 and  = -7.3 ppm)
P
B
upon the addition of THF (20 equiv.). In light of these it seems that THF, in
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 36
1
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
contradistinction to monoglyme or dioxane, is a strongly participating solvent,
resembling in this respect ACN. One should note, however, that the bonding
3
pattern in THF- -boranephosphonate 11 is quite different from that in the
acetonitrile complex 10. This probably made complex 11 susceptible to a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
1
nucleophilic substitution by pyridine at the boron center to form
pyridiniumboranephosphonate 2 (Scheme 8), in contradistinction to the
acetonitrile complex 10, where this reaction pathway was disfavored (Scheme
6
).
In all the investigated solvents the reaction of BP 1 with pyridine (10
equiv.), and iodine (3 equiv.) without the preactivation proceeded similarly as
in DCM and afforded the expected product Py-BP 2 together with small
amounts of dipyridiniumboranephosphonate 8 (10-20%) as a side-product.
The substrate scope of the investigated reaction
To explore the substrate scope of the investigated reaction we
8
appended the series of N-nucleophiles used by Caruthers et al. by additional
heteroaromatic and 3° amines a-l with diverse structural features (aromatic
amines, pyridine derivatives, bicyclic amidines, aliphatic acyclic, cyclic and
bicyclic amines, etc; Scheme 9).
ACS Paragon Plus Environment

Page 19 of 36
The Journal of Organic Chemistry
1
9
1
2
3
4
5
6
7
8
9
O
H
DCM
EtO P B N
H
EtO
1
2a-l
O
H
yields: 57-95% (31P NMR, Table S1)
(12b and 12k were isolated as exemplary
compounds - see the Experimental Section)
1
. 10 equiv. amine
EtO P B H
H
2
. 3 equiv. I2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EtO
1
(BP)
O
O
ACN
EtO P O P OEt +
12a-l
EtO
OEt
N
3
amines:
N
N
N
N
N
N
N
N
N
OH
a
b
c
d
e
N
N
N
h
O
i
f
g
N
N
N
N
j
k
l
Scheme 9. Substrate scope of the reaction of diethyl boranephosphonate 1
with tertiary amines.
To have comparison with the reactivity of pyridine, we reacted
amines a-l both with BP 1 in DCM, ACN, and THF in the presence of iodine,
and also separately with the products of activation of BP 1 with iodine, i.e.,
iodoboranephosphonate 9 (reactions in DCM), acetonitrile-3-
boranephosphonate 10 (reactions in ACN) and tetrahydrofuran-3-
boranephosphonate 11 (reactions in THF). The analysis of data from Table S1
(
see the supplementary information) revealed some general trends regarding
the reactivity of the tested amines. In dichloromethane, amines a-l reacted
promptly with BP 1, but less cleanly than pyridine, to afford the
3
1
corresponding B-modified boranephosphonate derivatives 12 ( P NMR
experiments,  = 58 - 65 ppm, broad signals) and some deboronated products
P
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 36
2
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
ca. 5-20%). In all instances, presence of additional boron-containing products
3
1
(up to 20%), resonating at ca. 40 ppm in the P NMR spectra, could be
2
2
detected. A similar reactivity was observed for the reactions in ACN and
THF (participating solvents), and the yields of the P-B-N products of type 12
were usually noticeable higher in THF than those in ACN or DCM. For
certain amines, however, e.g., b, d in ACN, and a, b in THF, only the
deboronated products were formed. For the reaction of amines a-l (10 equiv.)
with the generated iodoboranephosphonate intermediate 9 (reactions in
DCM), formation of 12 with variable proportions of tetraethyl pyrophosphate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
(20-60%) was observed. Certain sterically hindered amines (e.g., b, e, i, j, k),
apparently could not substitute the iodide at the boron center in 9, and
produced only the deboronated products (phosphate derivatives). This may
have a mechanistic bearing concerning a possible intermediate involved in the
investigated reaction.
3
In the instance of acetonitrile- -boranephosphonate complex 10, no
formation of the P-B-N derivatives of type 12 was observed for all the tested
amines, analogously to the reaction with pyridine (Scheme 7). In sharp
3
contrast to the reactivity of 10, tetrahydrofuran- - boranephosphonate
complex 11 reacted smoothly with amines c-l giving the P-B-N derivatives 12
usually in higher yields than in the reaction of BP 1 in DCM. This was most
apparent for amidine bases d and e. Sterically hindered amines of low basicity,
e.g., N,N-dimethylaniline (a) and 2,6-lutidine (b), were not able to substitute
the THF moiety in complex 11, but instead they probably facilitated further
oxidation of the borane group eventually producing deboronated products
(phosphate derivatives).
The broad substrate scope of this reaction is consistent with studies
8
by Caruthers et al. but with a notable exception of 2,6-lutidine and
triethylamine, for which no formation of complexes of type 12 was reported,
ACS Paragon Plus Environment

Page 21 of 36
The Journal of Organic Chemistry
2
1
1
2
3
4
5
6
7
8
9
due to the claimed extensive steric hindrance. Contrary to these we found that
in DCM compounds 12b (2,6-lutidine derivative) and 12k (triethylamine
derivative) could be formed and their structures were confirmed by
1
13
31
11
spectroscopic methods ( H, C, P, B NMR, and HRMS) after silica gel
chromatographic isolation. This highlighted the importance of the solvent
used and might pointed to the intermediacy of an electron-deficient borane
species in this reaction (see Scheme 1). In agreement with these, in the
reaction of boranephosphonate diesters with sterically hindered amines,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
participating solvents (e.g., dioxane in the Caruthers’ studies , and acetonitrile
or THF in the present work) may efficiently prevent capturing of the borane
intermediate by the amines, in contrast to non-participating solvents (e.g.,
3
dichloromethane), where the transient  -borane intermediate can be
intercepted even by sterically hindered amines to form B-modified
boranephosphonate dieters of type 12.
Mechanistic aspects of the reaction of boranephosphonate diesters with
amines promoted by iodine.
The issue of the oxidizing species - iodine-amine complexes.
Although iodine is implicated as an oxidant in the investigated
reaction, a chemical nature of the oxidizing entity involved is far from clear.
The problem can be particularly difficult to resolve for oxidative
transformations involving amines7-8 since these are known to form with
iodine various type of complexes, connected via complex equilibria systems,23
that ultimately undergo time- and concentration-dependent decomposition
processes.23d,24 The interaction of amines with halogens has been the subject of
numerous investigations since almost a century11b,23c,25 and in recent years
these studies have received new stimuli due to the revival of the concept of a
2
6
N-X-N halogen bonding as a new molecular motif in rational drug design
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 36
2
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
7
and in material sciences. For a pyridine/iodine mixture, a typical equilibria
system that is established in organic solvents is shown in Scheme 10.23c,28
I2
Py
N + I
I
N
I
I
N
I + I3
N
I
N
+ I3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PyI2
PyI+
PyIPy+
Scheme 10. A putative (simplified) equilibria system for iodine and pyridine
in organic solvents.
An increase in iodine concentration is shifting the equilibrium in
+
Scheme 10 towards N-iodopyridinium cationic species PyI , while an increase
in base concentration (pyridine) favours formation of a charge transfer
-
1
complex PyI (K ca. 100-200 M ) and another cationic species, N-
2
a
iododipyridinium PyIPy+.28a,29 In nonpolar solvents, the charge transfer
complexes of type PyI are relatively stable but in polar systems, they tend to
2
+
+
dissociate to form ionic species PyI and PyIPy that undergo slow
decomposition.28a According to the Schuster’s and Roberts’ H- and C NMR
1
13
3
0
studies, in nonpolar solvents iodine and pyridine are in fast equilibrium
+
with PyI , while the equilibrium between PyI and PyIPy is slow on the NMR
2
2
1
time scale. In line with these, our H NMR experiments on 2:1 mixture of
pyridine and iodine in DCM revealed only one set of signals from the
pyridine moiety, indicating a rapid equilibrium of iodine and pyridine with
the charge transfer complex PyI . For the ACN solutions, an additional set of
2
+
signals, attributed to the formation of cationic species PyIPy (ca. 15%), was
1
observed in the H NMR spectra (data not shown). We could also confirm that
solutions of iodine with pyridine were not completely stable (even in DCM)
as judging from the inferior results of the formation of
pyridiniumboranephosphonates 2, when pyridine with and iodine were
31
mixed 15-60 min prior to the reaction with BP 1. Due to complexity of the
system and its sensitivity to various factors (e.g., kind of the solvent used,
concentration and the ratio of the reactants, instability in time), no attempt
ACS Paragon Plus Environment

Page 23 of 36
The Journal of Organic Chemistry
2
3
1
2
3
4
5
6
7
8
9
was made to identify, which of the species, elemental iodine, charge transfer
+
+
complex PyI , PyI , or PyIPy was the reactive entity accepting a hydride ion
2
from the borane group during synthesis of 2. However, selected experiments
shed some light on this problem. Since that there was no discernible
difference in the rate of oxidation of boranephosphonate 1 in DCM with
iodine alone (the preactivation process), and with iodine in the presence of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pyridine (the reaction to produce 2, vide supra), we assumed for the purpose of
our studies that reactivity of elemental iodine was probably similar to that of
the other complexes formed by iodine with pyridine (Scheme 10). For this
reason in our mechanistic discussion we found it justified to invoke the
participation of elemental iodine or its complexes with pyridine as possible
oxidants.
Reaction pathways for oxidative transformation of boranephosphonate 1 into
pyridiniumboranephosphonate derivative 2.
All the experiments discussed above cumulatively pointed to three
possible reaction pathways for the conversion of BP 1 into
pyridiniumboranephosphonates of type 2 that are summarized in an
abbreviated form in Scheme 11.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 36
2
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 11. Some possible reaction pathways for the formation of
pyridiniumboranephosphonate 2.
At the present stage of this study path a in Scheme 11 seems most
likely because all the synthetic results discussed above can be simply
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
accounted for by assuming the intermediacy of  -boranephosphonate A with
a vacant p-orbital on the boron atom. In line with this, in non-participating
solvents such as DCM, the formation compounds of type 12 with the P-B-N
functionality bearing even highly sterically hindered amines, was observed.
While the generated in the preactivation process in DCM
iodoboranephosphonate 9 also reacted smoothly with pyridine and other 3°
amines (Table 1S), it failed to provide the corresponding P-B-N derivatives in
the reaction with certain sterically hindered amines, e.g., 2,6-lutidine (b), DBU
(
e), N-methylmorpholine (i), trimetylamine (j), and triethylamine (k) (Table
S). By this token, path b, that assumed iodoboranephosphonate 9 as an
alternative common intermediate in this reaction, must be rejected. Thus, in
1
1
0
the light of our previous studies path a may constitute a uniform mechanism
for the iodine-mediated reactions of boranephosphonate diesters with amines,
regardless of the type of final products formed, i.e., the corresponding
1
0
phosphoramidates (for 1° and 2° amines) or B-modified boranephosphonate
derivatives (for 3° and heteroaromatic amines, this study).
Another possible way for the formation of
pyridiniumboranephosphonate 2 can be path c that involves a 4-center
transition state C (Scheme 11) in which oxidation of the borane hydride and
coordination of the pyridine moiety occurs simultaneously. Such a
8
mechanism was previously proposed by Caruthers et al to explain a different
course of the mediated by iodine reaction of boranephosphonate diesters with
pyridine vs 1° or 2° amines. This seems reasonable in light of the oxidizing
species present in the reaction mixture, but on the geometrical ground, a
ACS Paragon Plus Environment

Page 25 of 36
The Journal of Organic Chemistry
2
5
1
2
3
4
5
6
7
8
9
transition state of type C would call for a square pyramidal arrangement of
the substituents around the boron center, and require a frontside attack of the
amine moiety. While probably possible for pyridine, this scenario can be
somewhat problematic for sterically hindered bases. Thus, we suggest that if
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
PyI would be involved as a reactive oxidizing species, the B-H bond breaking
(
oxidation of the borane hydride) might run significantly ahead of the B-N
bond formation, and the transition state could adopt all the characteristic of
3
-boranephosphonate A intermediate with the pyridine molecule nearby the
boron center, available for coordination. We refer to such a mechanism as
3
occurring via  -boranephosphonate A intermediate in a solvent cage. A
similar scenario can be envisaged for iodine or its charge-transfer complex
with pyridine (PyI ) acting as an oxidizing species (path a), where the incipient
2
3
-boranephosphonate A intermediate could be intercepted by pyridine
present in the solvent cage. Assuming the above mechanism, it can be
expected that in participating solvents (ACN and THF, see Table 1S)
competitive capture of intermediate A by the solvent molecules may occur.
3
This would lead to the formation of acetonitrile- -boranephosphonate
3
complex 10 (reactions in ACN) or tetrahydrofuran- -boranephosphonate
complex 11 (reactions in THF). Since compound 10 cannot be converted into
P-B-N products (see Scheme 6 and Table 1S), lower yields of the desired
products 12 would be expected for the reaction in acetonitrile. This is indeed
apparent from the data in Table 1S. In addition, for certain amines, e.g., 2,6-
lutidine (b) and DBN (d), only deboronated products were formed in ACN,
probably due to complete capture of A by the solvent molecules. This
interpretation is congruous with the data for the reactions in THF (Table 1S).
In this instance the interception of the putative intermediate A by the solvent
3
molecules to form tetrahydrofuran- -boranephosphonate complex 11, should
not have a detrimental effect on the efficiency of the P-B-N compounds
formations, because unlike the complex with acetonitrile 10, complex 11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 36
2
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
would return the desired products 12 upon the reaction with amines. Indeed,
for the most part, the yields of products 12 in the reactions in THF were
higher than in acetonitrile and comparable with those in DCM. The exception
here were N,N-dimethylaniline (a) and 2,6-lutidine (b) that, as it was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
mentioned before, could not substitute the tetrahydrofuran moiety in 11, but
instead, promoted its degradation to deboronated products. Considering this
reactivity and the assumption that during the course of the reaction of BP 1
with amines, intermediate A can be captured by the solvent molecules to form
1
1, the formation of P-B-N products for amines a and b was not observed in
THF.
Finally, a possible mechanism for the formation of the preactivation
products from BP 1 with iodine in DCM (compound 9), ACN (compound 10),
and THF (compound 11), deserved some comments. Iodoboranephosphonate
9
can be formed, in principle, via (i) the initial generation of intermediate A,
followed by coordination of iodide present in the reaction mixture, (ii) in a
one step reaction via a 4-center transition state B, analogous to C, or (iii)
3
coordination of the iodide anion by the insipient  -boranephosphonate A in a
solvent cage, as it was suggested above for the reactions with amines. Since
reactivity of iodine alone, and iodine in the presence of amines, towards
boranephosphonate diesters seems to be comparable, at this stage of our
investigations we tend to favour the last option. As to the formation of
3
3
acetonitrile- -boranephosphonate 10 and tetrahydrofuran- -
boranephosphonate 11, direct coordination of the solvent molecules
3
(acetonitrile and tetrahydrofuran, respectively) by  -boranephosphonate A
most likely prevailed due significantly higher concentration of the solvent
molecules compared to iodide, but some participation of 9 cannot be
completely excluded.
ACS Paragon Plus Environment

Page 27 of 36
The Journal of Organic Chemistry
2
7
1
2
3
4
5
6
7
8
9
To closing this part, it should be remembered that reactions of
boranephosphonate diesters with amines mediated by iodine are also very
sensitive to the ratio of the reagents used, as it was shown above for the
model amine, pyridine. This can have a quantitative and qualitative impact
on the outcome of these reactions.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSIONS
We found that the reaction of diethyl boranephosphonate 1 with
pyridine promoted by iodine in dichloromethane afforded the corresponding
pyridiniumboranephosphonate 2 and ca. 20% of
dipyridiniumboranephosphonate derivative 8. In an analogous reaction in
acetonitrile, a complete deboronation occurred, with tetraethyl
pyrophosphate 3 as a final product. It seems that the formation of different
products in dichloromethane vs acetonitrile was most likely due to instability
of the formed pyridiumboranephosphonate derivative under the reaction
conditions or an alternative reaction pathway (formation of acetonitrile
adduct 10) in the later solvent. A similar course of the reaction was observed
for other pyridine derivatives and 3° amines investigated. Preactivation of BP
1 with iodine in dichloromethane vs acetonitrile without amine also produced
distinct intermediates, namely, the corresponding iodoboranephosphonate 9
3
(in DCM) and the hitherto unknown complex of  -boranephosphonate with
acetonitrile 10 (in ACN). These, upon addition of pyridine afforded the
desired diethyl pyridiniumboranephosphonate 2, and tetraethyl
pyrophosphate 3, respectively. We also found that THF in this reaction acted
as a strongly participating solvent forming BP-THF complex 11 that, in
contradistinction to the analogous complex with acetonitrile 10, reacted with
the added pyridine yielding PyBP 2.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 36
2
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
1
11
On the basis of P and B NMR spectroscopy we formulated a
general mechanism for the reaction of boranephosphonate diesters with 1°, 2°,
°, and heteroaromatic amines mediated by iodine. It features, as key
3
mechanistic steps, generation of an electron-deficient intermediate 3-
boranephosphonate A (Scheme 11), followed by its coordination by the amine
present to form compounds with a P-B-N structural motif. Given the overall
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
complexity of this reaction, two distinct mechanisms can also be envisaged, one via 4-
center transition state of type C, for less sterically hindered amines (e.g., 1° and 2°
3
amines, pyridine), and the other, via a  -boranephosphonate intermediate A, for
sterically hindered amines (e.g., 2,6-lutidine).
EXPERIMENTAL SECTION
All reagents were of analytical grade, obtained from commercial suppliers and used
without further purification. Diethyl boranephosphonate 1 was prepared as previously
1
0
described. Anhydrous solvents used for the reactions were stored over molecular
sieves 4 Å. TLC analyses were carried out on Merck silica gel 60 F 254 precoated
plates using CH Cl –MeOH 9:1 (v/v) solvent system. All evaporations were carried
2
2
out on rotatory evaporators under reduced pressure at 40 °C.
The NMR spectra were recorded using Bruker Avance II 400 MHz instrument. The
1
13
chemical shifts are reported in ppm, relative to solvent peaks ( H, C NMR), 2%
3
1
11
H PO solution in D O ( P NMR), and BF diethyl etherate ( B NMR). Assignment
3
4
2
3
of the NMR signals was accomplished on the basis of the expected chemical shift
values, splitting pattern of the signals, and 2D correlation experiments. High-
resolution mass spectra (HRMS) were recorded on Thermo Fisher Scientific Q-
Exactive Orbitrap mass spectrometer. Identification of intermediates observed in the
31
11
P and B NMR spectra of the reaction mixtures was done on the basis of the
expected chemical shifts, coupling constants, shapes of the multiples, and spiking of
the reaction samples with compounds of known structures.
1
Purity of the isolated compounds was >97% ( H NMR spectroscopy) unless it was
stated otherwise.
3
1
11
General protocol for P and B NMR experiments
To a solution of diethyl boranephosphonate 1 (0.5 mmol) and pyridine (5 mmol, 10
equiv.) in CH
2
Cl (3 mL), iodine (1.5 mmol, 3 equiv.) was added, and the reaction
2
mixture was kept at room temperature with a continuous stirring. Progress of the
3
1
11
reaction was checked by P and B NMR spectroscopy by taking samples in time
ACS Paragon Plus Environment

Page 29 of 36
The Journal of Organic Chemistry
2
9
1
2
3
4
5
6
7
8
9
intervals and recording the spectra. Analogous reactions were run also in other
solvents (e.g. acetonitrile, THF) and with other heteroaromatic and 3° amines (for
details, see in the text). To detect possible intermediates involved, 1 in a suitable
solvent was reacted with variable amounts of iodine in the absence of an amine (see,
the preactivation experiments). Reactivity of the intermediates formed was studied by
adding to the reaction mixture suitable reactants as indicated in the text.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To prove chemical identity of the formed compounds, selected reactions were run on
a preparative scale, and the products were isolated by silica gel chromatography. The
isolated compounds were characterized by spectroscopic methods and, is so desired,
subjected to additional reactions as specified in the text.
Preactivation of diethyl boranephosphonate 1 with iodine in dichloromethane.
Diethyl boranephosphonate 1 (TEAH+ salt, 127 mg, 0.5 mmol) was dissolved in
CH
2
Cl
Cl
2
(10 mL) and evaporated repeatedly (3 x) the added solvent (3 x 10 mL
). The residue was dissolve in CH Cl (3 mL) and iodine (380 mg, 1.5 mmol, 3
CH
2
2
2
2
3
1
11
equiv.) was added. Progress of the reaction was checked by P and B NMR
spectroscopy by taking samples in time intervals and recording the spectra. After ca
10 min upon mixing the reagents, the major signal (>70%) at ca 69 ppm in the P
NMR or at ca -39 ppm in B NMR spectrum, appeared. This was tentatively
identified as iodoboranephosphonate 9 on the basis of its reactivity towards pyridine
and acetonitrile, and by MS spectra of the crude reaction mixture. For more details,
see also Scheme 6 and the discussion in the text.
3
1
1
1
Preactivation of diethyl boranephosphonate 1 with iodine in acetonitrile.
Diethyl boranephosphonate 1 (TEAH+ salt, 127 mg, 0.5 mmol) was dissolved in
acetonitrile(10 mL) and evaporated repeatedly (3 x) the added solvent (3 x 10 mL
acetonitrile). The residue was dissolve in acetonitrile (3 mL) and iodine (380 mg, 1.5
3
1
11
mmol, 3 equiv.) was added. Progress of the reaction was checked by P and B
NMR spectroscopy by taking samples in time intervals and recording the spectra.
After ca 15 min upon mixing the reagents, the major signal (>75%) at ca 75 ppm in
31
11
the P NMR or at ca -25 ppm in B NMR spectrum appeared. This was tentatively
identified as iodoboranephosphonate 9 on the basis of its reactivity towards pyridine
and by MS spectra of the crude reaction mixture. For more details, see also Scheme 6
and the discussion in the text.
Synthesis of diethyl pyridiniumboranephosphonate 2
+
Diethyl boranephosphonate 1 (TEAH salt, 254 mg, 1mmol) was placed in a 100 mL
pear-shaped flask and evaporated twice the added anhydrous dichloromethane (2 × 20
mL). The residue was dissolved in the same solvent (5 mL), and then pyridine (10
mmol) and elemental iodine (3 mmol) were added. The mixture was stirred and
3
1
progress of the reaction was monitored by P NMR spectroscopy. After 10 min
excess of iodine was decomposed by the addition of an aqueous ethanethiol (25%, 0.2
mL), the solvents evaporated under reduced pressure, and the product was purified by
ACS Paragon Plus Environment