Stereochemistry of internucleotide bond formation by the H-phosphonate method. 5. the role of bronsted and H-bonding base catalysis in ribonucleoside H-phosphonate condensationchemical and stereochemical consequences

Article abstract of DOI:10.1080/15257770.2010.497014

Efficiency and stereoselectivity of condensations of ribonucleoside 3′-H-phosphonates with ethanol promoted by pivaloyl chloride were investigated as a function of tertiary amines used. Side reactions leading to an increased demand for the condensing agent were identified as derived from an attack of the pivalate anion at carbonyl centers of reactive pivaloyl derivatives. The conditions that secured quantitative yields of H-phosphonate diester condensations were assessed. Several tertiary amines promoted condensations with stereoselectivity higher than that observed for pyridine derivatives. A correlation between diastereoselectivity of the product formation and Brnsted and H-bonding basicities of the amine used was found.

Full text of DOI:10.1080/15257770.2010.497014

This article was downloaded by: [University of New Mexico]
On: 29 November 2014, At: 06:57
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Stereochemistry of Internucleotide
Bond Formation by the H-Phosphonate
Method. 5. The Role of Brønsted and H-
Bonding Base Catalysis in Ribonucleoside
H-Phosphonate Condensation—Chemical
and Stereochemical Consequences
Michal Sobkowski a , Jacek Stawinski a & Adam Kraszewski a
a Institute of Bioorganic Chemistry, Polish Academy of Sciences ,
Poznan, Poland
Published online: 19 Jul 2010.
To cite this article: Michal Sobkowski , Jacek Stawinski & Adam Kraszewski (2010) Stereochemistry
of Internucleotide Bond Formation by the H-Phosphonate Method. 5. The Role of Brønsted
and H-Bonding Base Catalysis in Ribonucleoside H-Phosphonate Condensation—Chemical and
Stereochemical Consequences, Nucleosides, Nucleotides and Nucleic Acids, 29:8, 628-645, DOI:
10.1080/15257770.2010.497014
To link to this article: http://dx.doi.org/10.1080/15257770.2010.497014
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Nucleosides, Nucleotides and Nucleic Acids, 29:628–645, 2010
Copyright ꢀC Taylor and Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770.2010.497014
STEREOCHEMISTRY OF INTERNUCLEOTIDE BOND FORMATION
BY THE H-PHOSPHONATE METHOD. 5. THE ROLE OF BRØNSTED
AND H-BONDING BASE CATALYSIS IN RIBONUCLEOSIDE
H-PHOSPHONATE CONDENSATION—CHEMICAL AND
STEREOCHEMICAL CONSEQUENCES
Michal Sobkowski, Jacek Stawinski, and Adam Kraszewski
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland
Efficiency and stereoselectivity of condensations of ribonucleoside 3ꢁ-H-phosphonates with
2
ethanol promoted by pivaloyl chloride were investigated as a function of tertiary amines used. Side
reactions leading to an increased demand for the condensing agent were identified as derived from
an attack of the pivalate anion at carbonyl centers of reactive pivaloyl derivatives. The conditions
that secured quantitative yields of H-phosphonate diester condensations were assessed. Several ter-
tiary amines promoted condensations with stereoselectivity higher than that observed for pyridine
derivatives. A correlation between diastereoselectivity of the product formation and Brønsted and
H-bonding basicities of the amine used was found.
Keywords H-phosphonates; stereochemistry; P-chirality; organocatalysis; reaction
mechanism
INTRODUCTION
While H-phosphonate chemistry was introduced to nucleotide field by Sir
Todd in 1957,[1] the modern approach to internucleotide bond formation by
H-phosphonate method dates back to 1985–1986 and papers by Stawinski[2]
and Froehler.[3] Due to these works, H-phosphonates emerged as particu-
larly efficient and versatile precursors for preparation of many biologically
active phosphorus compounds.[4,5] Despite this significant broadening of
the scope of applications of H-phosphonate esters, the initially devised con-
ditions for H-phosphonate diester formation, that is, condensations of nu-
cleoside H-phosphonate with an alcohol (or nucleoside) in pyridine (which
acted both as a base and as nucleophilic catalyst) promoted pivaloyl chloride
The financial support from Polish Ministry of Science and Higher Education is gratefully
acknowledged.
Address correspondence to Michal Sobkowski, Institute of Bioorganic Chemistry, Polish Academy
of Sciences, Noskowskiego 12/14 61-704, Poznan, Poland. E-mail: msob@ibch.poznan.pl
628

Internucleotide Bond Formation by the H-Phosphonate Method
629
(ca. 3 equiv.), remain practically unaltered in routine laboratory practice.[6]
Moreover, the underlying mechanism was recognized only superficial, since
the early findings of Stawinski, Froehler and Matteucci, and Efimov were
focused mainly on practical aspects of nucleoside 3ꢁ-H-phosphonate con-
densations, that is, increasing the yields and avoiding side reactions. Thus, it
was established that (i) acyl chlorides (preferentially pivaloyl chloride; PvCl)
were efficient activators for nucleoside H-phosphonate monoesters,[3,7,8] (ii)
H-phosphonic—carboxyl acid mixed anhydrides were putative reactive in-
termediates of the condensations,[9,10] and (iii) pyridine and quinoline sig-
nificantly enhanced the rate of H-phosphonate diester formation, acting
as nucleophilic catalysts,[11,12] Pyridine or pyridine-acetonitrile (Py-ACN; in
a ratio >1:1) were established as best media for nucleoside H-phosphonate
condensations, while the use of dichloromethane (DCM) or tetrahydrofuran
(THF) as co-solvents was found to be disadvantageous.[10]
The relatively voluminous knowledge on the chemistry of simple alkyl
H-phosphonates[13] is of limited use when comes to the chemistry of natural
product derivatives and mechanical aspects of nucleoside H-phosphonate
formation promoted by acyl chlorides. A complete chemoselectivity of a
nucleophilic attack at the phosphorus center of H-phosphonic-acyl mixed
anhydrides is a unique and still intriguing feature of H-phosphonate chem-
istry as many other acyl derivatives of phosphorus compounds, for example,
acyl C-phosphonates and acyl phosphates, show reverse chemoselectivity.
Only recently, the subject of pyridines-catalyzed reactions of nucleoside
3ꢁ-H-phosphonates was resumed in several mechanistic and computational
papers by Sigurdsson and Stro¨mberg,[14–17] while our group focused on stere-
ochemical aspects of nucleoside 3ꢁ-H-phosphonate condensations.[18–23] For
instance, it was found that by decreasing contents of pyridines in the reaction
mixture, stability of the reactive H-phosphonate intermediates involved was
increased by orders of magnitude, while the condensations with nucleophilic
species were still rapid and effective.[15,18,20,21,24] An advantage of low con-
centration of a base in the reaction mixture is suppression of the rate of side
reactions associated with an overactivation of H-phosphonate monoesters.
The amount of condensing agent could be reduced to almost quantitative
value (1.1 equiv.).
It was also found that condensation of ribonucleoside H-phosphonates 1
with alcohols and nucleosides owes its stereoselectivity to the dynamic kinetic
asymmetric transformation (DYKAT) that favors formation of DP(SP)[64]
diastereomer of H-phosphonate diesters 3 (Figure 1).[21]
According to this model, diastereomers of nucleoside H-phosphonic
—pivalic mixed anhydrides exist in a rapid equilibrium, with one diastere-
omer, namely the LP(SP), being significantly more reactive toward nucleo-
sides or alcohols than the other one. A variety of pyridine derivatives promote
efficient and highly stereoselective condensations of H-phosphonate 1, most

630
M. Sobkowski et al.
FIGURE 1 Stereoselective formation of DP(SP) H-phosphonate diesters 3 from ribonucleoside H-
phosphonate 1.
probably due to their ability to act as mild nucleophilic catalysts.[22] In con-
trast, the analogous reactions performed in the presence of more powerful
nucleophilic catalysts, for example, 4-dimethylaminopyridine (DMAP) or N -
methylimidazole (NMI), suffered from decreased stereoselectivity and low
yields that were attributed to side reactions at nucleobases and the phospho-
rus center.[11,22,29–31] Interestingly, in the corresponding condensations in
which tertiary aliphatic amines (triethylamine, TEA; diisopropylethylamine,
DIPEA) were used as bases, the reactions were rapid, and stereoselectivity
was usually higher than that observed for nucleophilic catalysts; however,
the yields were not quantitative.[22,23] This unusual observation prompted
us to study in more detail H-phosphonate condensations proceeding in the
presence of non-nucleophilic amines. In this article, we present mechanistic
investigations on the influence of general base and acid catalysis on chem-
istry and stereochemistry of ribonucleoside H-phosphonate condensations.
RESULTS AND DISCUSSION
Usually, tertiary aliphatic amines are not considered as suitable basic
components of the condensation reactions of H-phosphonates since they
are unlikely to act as nucleophilic catalysts, and their rather high basicity
may result in several side reactions, for example, transesterification of the
produced H-phosphonate diester,[32] disproportionation,[33] P-acylation,[11]
and bis-activation of H-phosphonates.[9,17] Nevertheless, we anticipated that
if these amines would be used in a small excess (for instance 3 molar equiv.),

Internucleotide Bond Formation by the H-Phosphonate Method
631
the above side reactions might be significantly reduced. Indeed, condensa-
tions of uridine 3ꢁ-H-phosphonate (1; c = 0.1 M) with ethanol (5–6 equiv.)
using pivaloyl chloride (1.6 equiv.) as a condensing agent, in the presence
of various amines (3 equiv.)[65] yielded the desired H-phosphonate diester
3a without detectable P-acylation or bis-activation (31P NMR).[66]
Moreover, despite the lack of nucleophilic catalysis, the condensations
appeared to be appreciably rapid in many cases. Both formation of the mixed
anhydride 2, and its subsequent esterification to produce H-phosphonate di-
ester 3a were accelerated not only due to nucleophilic catalysis, but probably
also due to base catalysis. For example, the reaction of the mixed anhydride 2
with 3ꢁ-O-(dimethoxytrityl)thymidine (TDMTr, 1 equiv.) in DCM in the pres-
ence of N,N -dimethyl aniline (DMA; pKa 5.1; 10 equiv.) required several
hours for completion, while for pyridine (pKa 5.1) it was complete after
ca. 10 minutes. The same reaction in the presence of non-nucleophilic N -
methylmorpholine, (NMM; pKa 7.4) was complete after ca. 2 minutes, and
for more basic amines the reaction was complete before the first 31P NMR
spectrum was recorded (ca. 1 minute).
Yield of Condensation
In contrast to many pyridine derivatives,[15,22] for most of the tertiary
amines investigated, including these of relatively low basicity, the con-
densations were not quantitative and usually ca. 5–30% of the starting
H-phosphonate monoester 1 remained unreacted (Table 1). From the
data collected, a general trend could be noticed, namely that the yield of
H-phosphonate diester 3a was lower for more basic amines. The distinct
exceptions were tertiary amines able to act as nucleophilic catalysts (hex-
amethylenetetramine; HMTA, 1,4-diazabicyclo[2.2.2] octane; DABCO), and
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) that is outstandingly basic in or-
ganic solvents.
The presence of significant amounts of unreacted starting material
(H-phosphonate 1) in the reaction mixtures suggested that pivaloyl chlo-
ride was partly consumed due to side reactions and its amount (1.6 equiv.)
was not sufficient for complete esterification of the substrate 1. Actually,
upon increasing the excess of the condensing agent the yield of diester in-
creased proportionally and reached quantitative value when >2.5 equiv. of
PvCl were used (Figure 2).
There are several possible reasons for the decreased yields of condensa-
tions in comparison to those observed in the presence of pyridine derivatives:
(i) reaction of PvCl with residual water, (ii) reaction of PvCl with alcohols
and nucleosides, (iii) reaction of PvCl with pivalic acid, and (iv) deacyla-
tion of the mixed anhydride 2 by pivalic acid. Other side reactions known
in H-phosphonate chemistry, for example, dealkylation of the produced H-
phosphonate diester 3, for example, by amines or chloride anions,[61,62] were

632
M. Sobkowski et al.
TABLE 1 The yields of the H-phosphonate diester 3a and diastereomeric excess (de) of its DP(SP)
diastereomer formed from H-phosphonate monoester 1 (0.1 M in DCM) and EtOH (6 equiv.) upon
activation with 1.6 equiv. of PvCl in the presence of various amines
Yield of
de (DP)d diesterd
pKa
pKa
pKa
pKa
c
Aminea
(H2O)a (DMSO) (ACN) (THF) pKHBb pKip
/%
/%
Pyridine derivatives[22]
1
2
3
4
2,6-Ditertbutyl-pyridine
Pyridine
2,4,6-Collidine
DMAP
5.0[34] 1.0[35]
47
62
68
53
70
100f
100f
85
5.2
7.5
9.7
3.2[36] 12.5[37] 8.3[38] 1.9[39]
15.0[37]
2.3[39]
2.8[39]
7.9[36] 17.7[37]
Tertiary amines
5
6
7
8
9
Tro¨ger’s basef
3.2g,[40]
5.1
5.2
1.5[41]
64
60
57
56
65
100e
95
66
95
100e
Dimethylaniline (DMA)
2.5[42] 11.4[37] 7.4[38] 0.5[41]
HMTAg
1.9[43]
0.3[41]
N,N -diethylaniline (DEA) 6.6
N ꢁ,N ꢁ,N 2,N 2-
6.7[44]
tetramethylphenyl-
enediamine
(TEMPhD)
N,N -diisopropylaniline
(DIPA)
N -methylmorpholine
(NMM)
10
11
7.4
7.4
53
72
100e
87
15.6[45] 8.5[46] 1.7[43]
12
13
14
15
N -ethylmorpholine
N,N’-dimethylpiperazine
DABCOg
7.7
72
73
53
72
84
79
42
80
8.3[47]
8.7
8.9[48] 18.3[49]
2.6[43] 0.8
2.3[43]
N ꢁ,N ꢁ,N 2,N 2-
9.1
tetramethylethylenediamine
ꢁ(TEMED)
16
N ,N ꢁ,N 2,N 2-
9.8
73
74
tetramethylpropylenediamine
(TEMPrD)
17
18
N-methylpiperidine
10.1
10.3[50]
72
71
76
74
N ꢁ,N ꢁ,N 2,N 2-
tetramethylbutylenediamine
(TEMBuD)
19
20
21
22
23
N -methylpyrrolidine
TEA
DIPEA
10.4
18.4[51]
69
75
70
73
70
79
71
71
13
72
11.0
11.4
11.6
12.0
9.0[52] 18.8[37] 13.7[38] 2.0[43] 2.1
10.0[46] 1.1[43]
DBU
24.3[37] 20.0[38] 3.8[53] −3.8
(−)− or (+/−)−
21.7[54]
Sparteinef
24
Proton Spongef
12.1[55] 7.5[48] 17.3[56]
0.2[41] 2.2
56
90
aAqueous pKa data, unless otherwise indicated, are taken from refs.[57,58]
bHydrogen bonding basicity, pKHB = logK(formation of HB complex), larger values correspond to greater
basicity.[59]
cIon pair basicity, pKip = -logKip, smaller values correspond to greater basicity.[60]
dDetermined via integration of the corresponding 31P NMR signals.
eNo signal of the starting monoester 1 could be detected in the spectrum.
fFor structure, see Chart 1.
gIn aqueous EtOH.
CHART 1

Internucleotide Bond Formation by the H-Phosphonate Method
633
FIGURE 2 Yields of condensations of uridine H-phosphonate 1 (0.05 mmol in 0.5 mL of DCM) with
EtOH (6 equiv.) promoted by PvCl (0.5–2.5 equiv.) in the presence of different amines (3 equiv.).
rejected as being very unlikely under the mild reaction conditions and short
reaction times used in this work.
In order to verify the possibility that water present in the reaction mix-
ture was responsible for lower yields of H-phosphonate condensations, the
reaction of H-phosphonate 1 with EtOH (6 equiv.) in the presence of var-
ious amines (3 equiv.) was repeated in ACN with controlled amounts of
water added (0.1–6.0 equiv.) to the reaction mixture, or in DCM saturated
with water (corresponding to ca. 1 equiv. of H2O) prior to pivaloyl chloride
addition (1.6 equiv.). For weakly basic amines (pyridine and DMA, pKa ∼5)
surprisingly high tolerance to water was observed, while for NMM and TEA
(pKa 7.4 and 11.0, respectively) even small amounts of water affected the
yield of condensation (Figure 3). Nevertheless, even relatively large amounts
of water did not deteriorate the yields dramatically, and the 1.6 equiv. of PvCl
used should to be sufficient to compensate for 0.3 equiv. of water in the reac-
tion mixture and preserve a quantitative yield also in the presence of NMM
and TEA. Thus, while spurious water might account for some variations in
the yields of the condensations investigated, it could not be considered as the
reason for a substantially increased demand for PvCl during condensations
of H-phosphonates in the presence of tertiary amines.
The second possible reason for incomplete condensations in the pres-
ence of tertiary amines could be consumption of pivaloyl chloride in acyla-
tion of the hydroxyl group and/or N3H function of uridine. While indeed,
the condensing agent was found to react rapidly for pivaloylation of a nu-
cleoside in the presence of strongly nucleophilic amines (e.g., DMAP) or

634
M. Sobkowski et al.
FIGURE 3 Yields of condensations of uridine H-phosphonate 1 (0.05 mmol in 0.5 mL of ACN or DCM)
with EtOH (6 equiv.) promoted by PvCl (1.6 equiv.) in the presence of water (0–6 equiv.) and different
amines (3 equiv.).
TEA-pyridine mixtures, these reactions were negligible when tertiary non-
nucleophilic amines were used alone, and could not account for the observed
low yields.[29,30]
Thus, the requirement of using relatively large excess of PvCl (>2.5
equiv.) to achieve a complete coupling in the presence of most tertiary
amines was tentatively assigned to an attack of the pivalic acid formed as a
side product during condensation at carbonyl center of one or both reactive
species present in the reaction mixture, that is, pivaloyl chloride and the
H-phosphonic-pivalic mixed anhydride 2, yielding pivalic anhydride (Pv2O)
that is unable to act as a condensing agent for nucleoside H-phosphonates[63]
(Figure 4). The observed general trend for an inverse relationship between
the pKa of an amine and the yield of diester 3a (Table 1) was in line with
this assumption and might be attributed to an increased ionization of the
pivalic acid, and the associated enhanced reactivity toward a carbonyl group.
Moreover, the differences in yields of the condensations performed in ACN
and DCM (Figure 3) were consistent with the assumption that the side reac-
tion involved a pivalate anion (Figure 4, paths c and d), since this reaction
might be expected to be slowed down in polar solvent, while the rate of
esterification of the mixed anhydride 2 (which is believed to be of SN2(P)
type and involves uncharged species), should be enhanced in polar acetoni-
trile. Indeed, kobs for the reaction of the mixed anhydride 2 (c = 0.1 M)
with TDMTr (1 equiv.)[15] in the presence of DMA (10 equiv.) was found
to be 0.014 M−1s−1 in DCM and 0.028 M−1s−1 in ACN. For pyridine and

Internucleotide Bond Formation by the H-Phosphonate Method
635
FIGURE 4 Putative reaction pathways induced by pivalate anions released during esterification (path b)
of the mixed anhydride 2. Path a, formation of 2 (presumably reversible but shifted quantitatively to the
right, 31P NMR). Path b, reaction with an alcohol yielding diester 3 and pivalic acid. Path c, deacylation
of 2 with pivalate. Path d, deactivation of the condensing agent with a simultaneous release of Cl− anions
that may shift the putative equilibrium in path a partly to the left. Path e, a hypothetic acylation of
H-phosphonate 1 by Pv2O (not observed[63]).
NMM, the reactions were too rapid for determination of the rate constants,
nevertheless, they were significantly faster in ACN than in DCM, as it could
be estimated when the reactions were performed in the presence of only
3 equiv. of the same bases.
The influence of the pivalate anion on the condensations of H-
phosphonate 1 was investigated in further experiments. Thus, when the
reactions of 1 with ethanol were performed in the presence of increasing
concentration of pivalic acid salts (Table 2), only pyridinium pivalate did not
affect the quantitative yield of H-phosphonate diester 3, irrespective of the
amount used. Pivalate of DMA caused a moderate reduction in the yield of
TABLE 2 The yield of H-phosphonate diester 3a and the diastereomeric excess of the DP(SP)
diastereomer of 3a formed during condensations promoted by PvCl (1.6 equiv.) performed in the
presence of pivalic salts of amines
Equivalents of pivalate
0
1.0
3.0
5.0
yield of 3a (de of 3a-DP)/%
Py•PvOH
100 (62)
95 (56)
70 (75)
100 (58)
92 (53)
58 (69)
100 (55)
88 (49)
17 (65)
100 (48)
81 (47)
8 (59)
DMA•PvOH
TEA•PvOH

636
M. Sobkowski et al.
TABLE 3 The fraction of H-phosphonate monoester 1 formed from the mixed anhydride 2 during a
stepwise addition of pivalic salts of amines
Equivalents of pivalate
0.4
0.8
1.2
1.6
2.0
3.0
Salt
pKa of amine
monoester 1 formed/%
Py•PvOH
5.2
5.1
11.0
0
0
34
0
0
77
0
0
100
14
0
100
39
3
100
58
5
100
DMA•PvOH
TEA•PvOH
3,[67] while in the presence of triethylammonium pivalate, the yield deterio-
ration was dramatic. In additional experiments it was found that the mixed
anhydride 2 was hardly deacylated by DMA•PvOH (Table 3) and that forma-
tion of the mixed anhydride 2 in the presence of DMA was rather sluggish
(∼5 minutes with 2.5 equiv. of PvCl). Thus, for weakly basic tertiary amines,
the lower yield of condensations could be attributed to the competitive re-
action of PvCl with pivalic acid (resulting Pv2O) instead of H-phosphonate
1 (Figure 4, path d). Since the formation of the mixed anhydride 2 in the
presence of TEA was very rapid and deacylation of 2 by TEA•PvOH was very
efficient,[68] it could not be excluded that in this case the last reaction was
a significant route leading to low yield of H-phosphonate diesters formation
(Figure 4, path c).
To resolve this ambiguity, the mixed anhydride 2 (prepared in situ)
was treated with EtOH in the presence of various amines (pyridine, TEA,
DIPEA, DMA, NMI, or DMAP). Under such conditions ethyl nucleoside H-
phosphonate diester 3a was formed quantitatively for pyridine and DMA,
and almost quantitatively (96–97%) for TEA and DIPEA. In contrast, in the
presence of strong nucleophilic catalysts, NMI and DMAP, the yield of 3a was
comparable to that of during routine condensations without preactivation
of H-phosphonate 1. (de 80% vs. 73% for NMI, and de 48% vs. 46% for
DMAP).[69] High yields in the reactions in the presence of non-nucleophilic
amines indicated that deacylation of the mixed anhydride 2 was too slow
to compete significantly with its esterification. However, when the mixed
anhydride 2 was reacted with 1 equiv. of TDMTr in the presence of 10 equiv.
of DMA and 3 equiv. of DMA•PvOH, the main product of the reaction was
monoester 1, indicating that the main pathway under such conditions was
deacylation of the mixed anhydride 2 by pivalate anions.
To conclude this part, during routine condensations of H-phosphonates
in the presence of tertiary amines, formation of pivalic anhydride was the
main side reaction responsible for increased demand for pivaloyl chloride in
comparison to condensations done in the presence of pyridine derivatives.
Most probably, the pivalic anhydride was formed mainly via the reaction of
pivaloyl chloride and the pivalate anion (path d in Figure 4). Deacylation of

Internucleotide Bond Formation by the H-Phosphonate Method
637
FIGURE 5 Diastereomeric excess (de) of the DP(SP) diastereomer of H-phosphonate diester 3a in the
presence of chosen tertiary amines (for which the pKHB values are known). The data are arranged
according to pKa values of the amines used. The reaction conditions: 0.05 mmol of 1 + EtOH (6 equiv.)
+ amine (3 equiv.) + PvCl (1.6 equiv.) in DCM (0.5 mL). The tertiary amines able to act as nucleophilic
catalysts are shown as empty bars.
the mixed anhydride 2 by the pivalate anion (path c) might be a significant
route only when the esterification pathway was retarded.
Stereoselectivity of Condensation
The stereoselectivity of condensations of H-phosphonate 1 with ethanol
in the presence of various tertiary amines differed in the range of de 53–75%
what corresponds to the ratio of diastereomers from 3:1 to 7:1. No clear-
cut correlation between the observed stereoselectivity and various parame-
ters of basicity could be found (Table 1). However, the stereoselectivity of
H-phosphonate condensations showed some positive correspondence with
both the pKa and pKHB[70] values of the amines used (Figure 5).
This might indicate that stronger bases can increase rates of the 2-DP ꢀ 2-
LP equilibrium responsible for the DYKAT efficiency (Figure 1), possibly due
to base–promoted ionization of the pivalic acid (the influence of pKa) and,
presumably independently, due to general acid catalysis by the conjugate
acids of the amines operating on the mixed anhydride 2 (the influence of
pKHB). A possible catalytic action of a conjugate acid of an amine is shown
in Figure 6.
We found that a possible contribution of both pKa and pKHB basicities
of the amines to the stereoselectivity of ribonucleoside 3ꢁ-H-phosphonates
condensations could be expressed in a form of a single parameter which we
tentatively called “DYKAT basicity” (pKDK). This is a weighted sum of pKa

638
M. Sobkowski et al.
FIGURE 6 A possible influence of general acid catalysis on reactivity of the mixed anhydride 2.
and pKHB basicities, given by Eqaution (1), where “n” is a proportionality
constant.
pKDK = n ∗ pKa + pKHB
(1)
The regression analysis of de versus pKDK for various investigated amines
gave the best linear fit (R2 = 0.916) for n = 0.18 (Figure 7).[71] The em-
pirically found relationship between de and pKDK has a form of a linear
FIGURE 7 A correlation between “DYKAT basicity” (pKDK) and the stereoselectivity of condensations
of uridine 3ꢁ-H-phosphonate 1 with EtOH in the presence of various amines. The data points for HMTA
and DABCO were excluded from calculation of the trend line.

Internucleotide Bond Formation by the H-Phosphonate Method
equation (2).
639
(2)
de = 0.0661 ∗ pKDK + 0.4875
The numerical value of pKDK does not have an obvious physical sense, but it
may be a measure of acid-base properties of an amine that are essential for
stereoselectivity of the reaction investigated.
In contrast to pKa and pKHB, we did not find any correlation of the yield
and stereoselectivity of condensations of ribonucleoside 3ꢁ-H-phosphonates
with the ion pair basicity[60] of the several amines investigated (pKip,
Table 1).
It should be noted that the stereoselectivity of condensations performed
in the presence of additional amounts of pivalic acid salts decreased irrespec-
tive of the amine used (Table 2). This was rather a surprising result since
pivalic acid was expected to increase stereoselectivity by accelerating isomer-
ization of the mixed anhydride 2 (Figure 1). To get more insight into the
influence of pivalic acid on reactivity of 2 we measured the rate constant of
its reaction with TDMTr in the presence of DMA•PvOH (3 equiv.) and DMA
(10 equiv.). It was found that kobs under such conditions was ca. half of that
without the pivalate added (kobs 0.0061 vs. 0.014 M−1s−1 respectively). These
results suggested that high concentrations of pivalic acid in the reaction
mixture decelerated both esterification and, presumably to higher extent,
P-epimerization of the mixed anhydride 2. Similar effect of deteriorated
stereoselectivity was observed when amine salts of other acids (tosylates and
hydrochlorides) were present in the reaction mixture. A possible reason for
this phenomenon could be decreased basicity of the reaction mixture due
to introduced salts and the following less efficient base catalysis. A practical
conclusion from those experiments is that in order to achieve the high-
est stereoselectivity of the condensation of H-phosphonates of type 1, one
should avoid increased concentration of salts in the reaction mixtures.
The stereoselectivity of condensation of H-phosphonate 1 with ethanol
in the presence of TEA remained at the same high level (de 76% 31P NMR)
when instead of 1.6 equiv. of PvCl, 2.5 equiv. (or more) were used, while
the yield of condensation became quantitative under these conditions (31P
NMR). This prompted us to check if tertiary amines could be used as basic
components for highly stereoselective and efficient internucleotide bond
formation. In initial experiments uridine H-phosphonate 1 (c = 0.1 M) was
reacted with TDMTr (3 equiv.) in the presence of TEA (3 equiv.) using PvCl
(3 equiv.) as a condensing agent. The expected DP (SP) diastereomer of
H-phosphonate diester 3b was formed with de >90% (i.e., the ratio of DP/LP
isomers was >95:5) and the starting monoester could not be detected in
31P NMR spectrum. Unfortunately, the yield of the desired H-phosphonate
diester was rather low (ca. 70%) due to formation of P-acylated compounds
(ca. 30%); also acylation of N3 position in pyrimidine moiety (up to 50%) was

640
M. Sobkowski et al.
observed. These side reactions could be reduced (ca 10% of P-acylation, no
N-acylation) upon dilution of the reactants. Thus, we expect that achieving
quantitative yield and very high stereoselectivity of condensation of ribonu-
cleoside H-phosphonates with nucleosides could be possible and fine-tuning
of this approach is a subject of further investigations in our laboratory.
CONCLUSIONS
Condensations of ribonucleoside H-phosphonates with alcohols can be
performed efficiently using tertiary, non-nucleophilic amines as base cata-
lysts. The reactions subject to general base catalysis and for amines having
pKa ≥ 7.4 (N -methylmorpholine) are significantly more rapid than those in
the presence of pyridine. In order to achieve quantitative yields of the con-
densations, >2.5 equiv. of the condensing agent (pivaloyl chloride) should
be used for most of tertiary amines (particularly for those of high pKa), and
non-hindered alcohols should be used in an excess of >3 equiv. to prevent
P-acylation. The necessity of using the condensing agent in excess arises from
its competitive reaction with pivalic acid released to the reaction mixture as a
side product of the condensations, and, in the cases when the esterification
rate was retarded, due to deacylation of nucleoside H-phosphonic-pivalic
mixed anhydride (the reactive intermediate) by pivalate anion.
Stereoselectivity of the condensations showed a positive trend with
Brønsted basicity (pKa) and H-bond basicity (pKHB) of the amines present
in the reaction mixture. This trend can be expressed in a form of an empir-
ical equation that correlates these parameters with diastereomeric excess of
DP (SP) diastereomer of ethyl uridine H-phosphonate diester 3a formed.
The highest stereoselectivity (de 75%) and quantitative yield of diester
3a (31P NMR) was achieved in the presence of triethylamine. In the prepa-
ration of dinucleoside H-phosphonate diester 3b the stereoselectivity was
exceptionally high, de >90%, however, ca. 10% of P-acylated products were
formed (the conditions are currently a subject of further optimization).
EXPERIMENTAL
The chemicals, instrumentation, and procedures were similar to those
reported earlier.[22] 31P NMR spectra were recorded at 121 MHz on a Varian
Unity BB VT spectrometer and the ratios of compounds were based upon
integration of the corresponding 31P NMR signals. The yield and diastere-
omeric excess for the data in Table 1 are an average of three experiments,
while for the data in Figures 5 and 7, of five repetitions.

Internucleotide Bond Formation by the H-Phosphonate Method
641
General Procedure for Condensation of H-Phosphonates
of Type 1 with Alcohols
Nucleoside H-phosphonate 1 (0.05 mmol) was rendered anhydrous by
repeated evaporation (3 × 5 mL) of the added toluene and dissolved in
0.5 mL of DCM. An amine (3 equiv.) and EtOH (6 equiv.) were added,
followed by PvCl (1.6 equiv.), and the reaction mixture was transferred to
an NMR tube and the 31P NMR spectra were recorded.
Procedure for the In Situ Preparation of the Mixed Anhydride 2
Nucleoside H-phosphonate 1 (0.1 mmol) was rendered anhydrous by
repeated evaporation (3 × 5 mL) of the added toluene and dissolved
in DCM or ACN (0.6 mL), then amine (0.2 equiv. of TEA or NMM, or
0.5 equiv. of DMA) and pivaloyl chloride (1.2 equiv) were added succes-
sively. Such a reaction mixture was analyzed by 31P NMR spectroscopy and
used for further transformations within 1 hour (no degradation products
formation was observed within that period of time).
Kinetic Measurements
To a solution of 3ꢁ-(dimethoxytrityl)thymidine[30] (0.05 mmol) in DCM
(1 mL), toluene (5 mL) was added and the solvents were evaporated to
dryness under reduced pressure; the procedure was repeated twice. The
residue was dissolved in DCM or ACN (0.2 mL) containing an appropriate
amine (3 or 10 equiv.) and the solution of the mixed anhydride 2 (pre-
pared as described above, 0.3 mL; 0.05 mmol) was added. The reaction mix-
ture was transferred to an NMR tube and 31P NMR spectra were recorded
immediately.
REFERENCES
1. Hall, R.H.; Todd, A.; Webb, R.F. Nucleotides. 41. Mixed Anhydrides as intermediates in the synthesis
of dinucleoside phosphates. J. Chem. Soc. 1957, 3291–3296.
2. Garegg, P.J.; Regberg, T.; Stawinski, J.; Stro¨mberg, R. Formation Of Internucleotidic Bonds Via
Phosphonate Intermediates. Chem. Scr. 1985, 25, 280–282.
3. Froehler, B.C.; Ng, P.G.; Matteucci, M.D. Synthesis Of DNA Via Deoxynucleoside H-Phosphonate
Intermediates. Nucleic Acids Res. 1986, 14, 5399–5407.
4. Stawinski, J.; Kraszewski, A. How To get the most out of two phosphorus chemistries. studies on
H-phosphonates. Acc. Chem. Res. 2002, 35, 952–960.
5. Kraszewski, A.; Stawinski, J. H-Phosphonates: Versatile Synthetic precursors to biologically active
phosphorus compounds. Pure Appl. Chem. 2007, 79, 2217–2227.
6. Stro¨mberg, R., Stawinski, J. In Current Protocols In Nucleic Acid Chemistry, Eds. S. L. Beaucage, D. E.
Bergstrom, G. D. Glick, R. A. Jones. John Wiley & Sons, New York, 2004.
7. Garegg, P.J.; Lindh, I.; Regberg, T.; Stawinski, J.; Stro¨mberg, R. Nucleoside H-phosphonates. III.
Chemical synthesis of oligonucleotides by the hydrogenophosphonate approach. Tetrahedron Lett.
1986, 27, 4051–4054.

642
M. Sobkowski et al.
8. Froehler, B.C.; Matteucci, M.D. Nucleoside H-phosphonates: Valuable intermediates in the synthesis
of deoxynucleotides. Tetrahedron Lett. 1986, 27, 469–472.
9. Garegg, P.J.; Regberg, T.; Stawinski, J.; Stro¨mberg, R. Nucleoside H-phosphonates. V. The mecha-
nism of hydrogenophosphonate diester formation using acyl chlorides as coupling agents in oligonu-
cleotide synthesis by the hydrogenophosphonate approach. Nucleosides Nucleotides, 1987, 6, 655–662.
10. Froehler, B.C.; Matteucci, M.D. The Use of nucleoside h-phosphonates in the synthesis of deoxy-
oligonucleotides. Nucleosides Nucleotides, 1987, 6, 287–291.
11. Dubey, I.Y.; Efimov, V.A.; Lyapina, T.V.; Fedoryak, D.M. Nucleophilic catalysts in H-phosphonate
internucleotide condensation reaction. Bioorg. Khim. 1992, 18, 911–919.
12. Regberg, T.; Stawinski, J.; Stro¨mberg, R. Nucleoside H-phosphonates. IX. Possible side-reactions
during hydrogen phosphonate diester formation. Nucleosides Nucleotides, 1988, 7, 23–35.
13. Troev K. Chemistry and Application of H-Phosphonates. Elsevier Science, 2006.
14. Sigurdsson, S.; Stro¨mberg, R. Evaluation of several economical computational methods for geometry
optimisation of phosphorus acid derivatives. Nucleosides Nucleotides Nucleic Acids, 2001, 20, 1381–
1384.
15. Sigurdsson, S.; Stro¨mberg, R. The H-phosphonate approach to oligonucleotide synthesis. an investi-
gation on the mechanism of the coupling step. J. Chem. Soc., Perkin Trans. 2, 2002, 1682–1688.
16. Sigurdsson, S.; Stro¨mberg, R. Comparison of some computational methods for geometry opti-
misation of phosphorus acid derivatives. Phosphorus, Sulfur Silicon Relat. Elem. 2002, 177, 2711–
2724.
17. Sigurdsson, S.; Stro¨mberg, R. Side reactions in the H-phosphonate approach to oligonucleotide syn-
thesis: a kinetic investigation on bisacylphosphite formation and 5’-O-acylation. Nucleosides Nucleotides
Nucleic Acids, 2003, 22, 1–12.
18. Almer, H.; Stawinski, J.; Stro¨mberg, R. Solid support synthesis of all-RP-oligo(ribonucleoside phos-
phorothioate)S. Nucleic Acids Res. 1996, 24, 3811–3820.
19. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. In Collection Symposium Series, Ed. M. Hocek.
Institute Of Organic Chemistry And Biochemistry, Academy Of Sciences Of The Czech Republic,
Prague, 2005, Pp. 183–187.
20. Sobkowski, M.; Jankowska, J.; Stawinski, J.; Kraszewski, A. Stereochemistry of internucleotide bond
formation by the H-phosphonate method. 1. synthesis and 31P NMR analysis of 16 diribonucleo-
side (3ꢁ-5ꢁ)-H-phosphonates and the corresponding phosphorothioates. Nucleosides Nucleotides Nucleic
Acids, 2005, 24, 1469–1484.
21. Sobkowski, M.; Kraszewski, A.; Stawinski, J. Stereochemistry of internucleotide bond formation by the
H-Phosphonate Method. 3. Investigations on a mechanism of stereoselectivity induction. Tetrahedron:
Asymmetry, 2007, 18, 2336–2348.
22. Sobkowski, M.; Stawinski, J.; Kraszewski, A. Stereochemistry of internucleotide bond formation by
the H-phosphonate method. 4. The role of nucleophilic catalysis in chemistry and stereochemistry
of ribonucleoside H-phosphonate condensation. New. J. Chem. 2009, 33, 164–170.
23. Sobkowski, M.; Stawinski, J.; Kraszewski, A. Stereochemistry of internucleotide bond formation by the
H-phosphonate method. 6. Optimization of the reaction conditions towards highest stereoselectivity.
Tetrahedron: Asymmetry, 2008, 19, 2508–2518.
24. Efimov, V.A.; Dubey, I.Y. Modification Of The H-phosphonate oligonucleotide synthesis on polymer
support. Bioorg. Khim. 1990, 16, 211–218.
25. Sobkowski, M.; Stawinski, J.; Kraszewski, A. A proposal for a new stereochemical notation for P-
chiral nucleotide analogues and related compounds. Nucleosides Nucleotides Nucleic Acids, 2005, 24,
1301–1307.
26. Sobkowski, M.; Stawinski, J.; Kraszewski, A. A proposal for a convenient notation for P-chiral nu-
cleotide analogues. Part 2. dinucleoside monophosphate analogues. Nucleosides Nucleotides Nucleic
Acids, 2006, 25, 1363–1375.
27. Sobkowski, M.; Stawinski, J.; Kraszewski, A. A proposal for a convenient notation for P-chiral nu-
cleotide analogues. Part 3. Compounds with one nucleoside residue and non-nucleosidic derivatives.
Nucleosides Nucleotides Nucleic Acids, 2006, 25, 1377–1389.
28. Sobkowski, M.; Stawinski, J.; Kraszewski, A. A proposal for a convenient notation for P-Chiral nu-
cleotide analogues. Part 4. A relationship between the DP/LP notation and stereochemistry of
reactions. Nucleosides Nucleotides Nucleic Acids, 2009, 28, 29–42.

Internucleotide Bond Formation by the H-Phosphonate Method
643
29. Sobkowski, M. Diverse chemoselectivity during acylation of nucleosides. In: Collection Symposium Series,
Ed. M. Hocek. Institute Of Organic Chemistry And Biochemistry, Academy Of Sciences Of The Czech
Republic, Prague, 2008, Pp. 277–281.
30. Sobkowski, M. A Convenient protection of 4-oxopyrimidine moieties in nucleosides By the pivaloyl
group. Collect. Czech. Chem. Commun. 2010, 75, 33–57.
31. Efimov, V.A.; Dubey, I.Y.; Chakhmakhcheva, O.G. NMR Study And improvement of H-phosphonate
oligonucleotide synthesis. Nucleosides Nucleotides, 1990, 9, 473–477.
32. Sobkowski, M.; Stawinski, J.; Sobkowska, A.; Kraszewski, A. Studies On reactions of nucleoside H-
phosphonates with bifunctional reagents. 2. Stability of nucleoside h-phosphonate diesters in the
presence of amino alcohols. J. Chem. Soc., Perkin Trans. 1, 1994, 1803–1808.
33. Kers, A.; Kers, I.; Stawinski, J.; Sobkowski, M.; Kraszewski, A. Studies on aryl H-phosphonates. 3. Mech-
anistic investigations related to the disproportionation of diphenyl H-phosphonate under anhydrous
basic conditions. Tetrahedron, 1996, 52, 9931–9944.
34. Hopkins, H.P.; Jahagirdar, D.V.; Moulik, P.S.; Aue, D.H.; Webb, H.M.; Davidson, W.R.; Pedley, M.D.
Basicities of the 2-, 4-, 2,4-di-, and 2,6-disubstituted tert-butyl pyridines in the gas phase and aqueous
phase: Steric effects in the solvation of tert-butyl-substituted pyridines and pyridinium cations. J. Am.
Chem. Soc. 1984, 106, 4341–4348.
35. Benoit, R.L.; Frechette, M.; Lefebvre, D. 2,6-Di-Tert-Butylpyridin—An unusually weak base in
dimethylsulfoxide. Can. J. Chem. 1988, 66, 1159–1162.
36. Augustin-Nowacka, D.; Makowski, M.; Chmurzynski, L. Potentiometric studies of cationic hetero-
conjugation equilibria in systems involving free and protonated pyridine derivatives in dimethyl
sulfoxide. J. Chem. Thermodynamics, 2002, 34, 391–400.
37. Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.; Leito, I.; Koppel, I.A. Extension of
the self-consistent spectrophotometric basicity scale in acetonitrile to a full span of 28 Pka units:
Unification of different basicity scales. J. Org. Chem. 2005, 70, 1019–1028.
38. Garrido, G.; Koort, E.; Rafols, C.; Bosch, E.; Rodima, T.; Leito, I.; Roses, M. Acid-base equilibria in
nonpolar media. absolute Pka scale of bases in tetrahydrofuran. J. Org. Chem. 2006, 71, 9062–9067.
39. Berthelot, M.; Laurence, C.; Safar, M.; Besseau, F. Hydrogen-bond basicity Pkhb scale of six-membered
aromatic N-heterocycles. J. Chem. Soc., Perkin Trans. 2, 1998, 283–290.
40. Wepster, B.M. Steric Effects On Mesomerism. 16. Comparative investigations involving ultra-violet
absorption spectra, molecular refractions, rates of reaction, and basic strengths of aromatic amines.
Recl. Trav. Chim. Pays-Bas, 1957, 76, 357–389.
41. Marquis, E.; Graton, J.; Berthelot, M.; Planchat, A.; Laurence, C. Hydrogen bonding of
arylamines—Competition with P and N sites. Can. J. Chem. 2004, 82, 1413–1422.
42. Benoit, R.L.; Mackinnon, M.J.; Bergeron, L. Basicity of N-substituted anilines and pyridine in
dimethylsulfoxide. Can. J. Chem. 1981, 59, 1501–1504.
43. Graton, J.; Besseau, F.; Berthelot, M.; Raczynska, E.; Laurence, C. The Pkhb scale for hydrogen bonds
in aliphatic tertiary amines. Can. J. Chem. 2002, 80, 1375–1385.
44. Krikstopaitis, K.; Kulys, J.; Pedersen, A.H.; Schneider, P. N-substituted P-phenylenediamines as per-
oxidase and laccase substrates. Acta Chem. Scand. 1998, 52, 469–474.
45. Beltrame, P.; Gelli, G.; Loi, A. Potentiometric study of the ionisation of some nitrogen bases in
acetonitrile. Gazz. Chim. Ital. 1980, 110, 491–494.
46. Nurminen, E.J.; Mattinen, J.K.; Lo¨nnberg, H. Alcoholysis of dialkyl tetrazolylphosphonites. J. Chem.
Soc., Perkin Trans. 2, 1999, 2551–2556.
47. Kallen, R.G.; Jencks, W.P. Equilibria for the reaction of amines with formaldehyde and protons in
aqueous solution. A re-examination of the formol titration. J. Biol. Chem. 1966, 241, 5864–5878.
48. Benoit, R.L.; Lefebvre, D.; Frechette, M. Basicity Of 1,8-Bis(dimethylamino)naphthalene and 1,4-
diazabicyclo[2.2.2]octane in water and dimethylsulfoxide. Can. J. Chem. 1987, 65, 996–1001.
49. Coetzee, J.F.; Padmanabhan, G.R. Properties of bases in acetonitrile as solvent. IV. Proton acceptor
power and homoconjugation of mono- and diamines. J. Am. Chem. Soc. 1965, 87, 5005–5010.
50. Leyden, D.E.; Mccall, J.M. Jr. Protolysis kinetics of diamines in sulfuric acid. J. Phys. Chem. 1971, 75,
2400–2402.
51. Leffek, K.T.; Pruszynski, P.; Thanapaalasingham, K. Basicity of substituted 2-phenyl-1,1,3,3-
tetramethylguanidines and other bases in acetonitrile solvent. Can. J. Chem. 1989, 67, 590–595.
52. Kolthoff, I.M.; Chantoon, M.K.; Bhowmik, S. Dissociation constants of uncharged and monovalent
cation acids in dimethyl sulfoxide. J. Am. Chem. Soc. 1968, 90, 23–28.

644
M. Sobkowski et al.
53. Cecchi, L.; De Sarlo, F.; Machetti, F. 1,4-Diazabicyclo[2.2.2]octane (DABCO) as an efficient reagent
for the synthesis of isoxazole derivatives from primary nitro compounds and dipolarophiles: The
role of the base. Eur. J. Org. Chem. 2006, 4852–4860.
54. Toom, L.; Kutt, A.; Kaljurand, I.; Leito, I.; Ottosson, H.; Grennberg, H.; Gogoll, A. Substituent effects
on the basicity of 3,7-diazabicyclo[3.3.1]nonanes. J. Org. Chem. 2006, 71, 7155–7164.
55. Alder, R.W. Strain effects on amine basicities. Chem. Rev. 1989, 89, 1215–1223.
56. Brzezinski, B.; Schroeder, G.; Grech, E.; Malarski, Z.; Sobczyk, L. Basicity, IR-spectra and protonation
of some proton sponges in acetonitrile. J. Mol. Struct. 1992, 274, 75–82.
57. Perrin DD. Dissociation Constants Of Organic Bases In Aqueous Solution. Butterworth, London, 1965.
58. CRC Handbook Of Chemistry And Physics, 87th Ed.. CRC Press LLC, Boca Raton, 2007.
59. Taft, R.W.; Gurka, D.; Joris, L.; Von Schleyer, P.; Rakshys, J.W. Studies Of hydrogen-bonded complex
formation with P-fluorophenol. V. Linear free energy relationships with oh reference acids. J. Am.
Chem. Soc. 1969, 91, 4801–4808.
60. Streitwieser, A.; Kim, Y.J. Ion pair basicity of some amines in THF: Implications for ion pair acidity
scales. J. Am. Chem. Soc. 2000, 122, 11783–11786.
61. Troev, K.; Tashev, E.; Borisov, G. On the interaction between dialkyl phosphites and amine hy-
drochlorides. Phosphorus Sulfur, 1981, 11, 363–367.
62. Bryant, D.E.; Kilner, C.; Kee, T.P. Facile one-pot mono-dealkylation of H-phosphonate esters in high
yield. Inorg. Chim. Acta, 2009, 362, 614–616.
63. Stro¨mberg, R.; Stawinski, J. Evaluation of some new condensing reagents for hydrogenphosphonate
diester formation. Nucleic Acids Res. Symp. Ser. 1987, 18, 185–188.
64. For the compounds presented in this paper the DP descriptor refers to a structure in which the P-H
bond is directed to the right in the Fischer projection, while for the LP one, to the left. The full
DP/LP notation is described in refs.[25–28]
65. The same conditions as used in previous studies.[22]
66. When the amount of an alcohol was reduced to 3 equiv., the above mentioned by-products appeared
as trace impurities (1–2%), while for stoichiometric amounts of an alcohol, these by-products ac-
counted for ca. 20% of the phosphorus-containing compounds present in the reaction mixtures,
most likely due to decreased rates of the condensations.
67. When the reaction was repeated in the presence of 10 instead of 3 equiv. of DMA, a significant
deacylation of the mixed anhydride 2 was observed, demonstrating the importance of the basicity of
the reaction mixture on the course of reaction.
68. Because analogous experiments revealed lack of deacylating activity of tosylates and hydrochlorides
of the same amines, the observed deacylation of 2 should be attributed to rather unique properties
(in discussed context) of the studied pivalates, while their deacylating power could be correlated
with pKa values of the conjugated amines. Thus, amines of higher pKa apparently might increase
nucleophilicity of pivalic acid via its more efficient deprotonation.[15]
69. The reasons for low yields of esterification of the mixed anhydride 2 in the presence of powerful
nucleophilic catalysts (NMI, DMAP) remain uncertain. One may speculate that the H-phosphonic
onium salts formed by those nucleophilic amines during the course of the reaction[11] may react
partly with the pivalate anion forming acyl onium salts (and H-phosphonate monoester 1), which
subsequently might rapidly form unreactive pivalic anhydride with the next pivalate molecule. In
contrast, all reactions during condensations of nucleoside H-phosphonates in the presence of mod-
erate nucleophilic catalyst (pyridine) were evidently fully chemoselective toward H-phosphonate
diester formation.
70. pKHB is a measure of a relative strength of the acceptor in hydrogen-bonded complex formed with a
reference acid (H-bonding basicity) and may indicate on propensity of a conjugate acid of an amine
to act as a general acid catalyst. pKa and pKHB may be unrelated.[59]

Internucleotide Bond Formation by the H-Phosphonate Method
645
71. HMTA and DABCO are strong nucleophilic catalysts and the outstandingly low stereoselectivity
found for these compounds were similar to those observed for N -methylimidazole and DMAP.[22]
It seems that these nucleophilic catalysts, efficient for reactions at carbon and PV centers, are poor
nucleofuges in H-phosphonate chemistry, and the relative stability of P-N+ intermediates makes room
for side reactions and slows down their epimerization, making the DYKAT process of asymmetric
induction less efficient.