Experimental Basicities of Superbasic Phosphonium Ylides and Phosphazenes

Article abstract of DOI:10.1021/acs.joc.6b00872

Experimental basicities of some of the strongest superbases ever measured (phosphonium ylides) are reported, and by employing these compounds, the experimental self-consistent basicity scale of superbases in THF, reaching a pK_α (estimate of pK_a) of 35 and spanning more than 30 pK_a units, has been compiled. Basicities of 47 compounds (around half of which are newly synthesized) are included. The solution basicity of the well-known t-Bu-N=P₄(dma)₉ phosphazene superbase is now rigorously linked to the scale. The compiled scale is a useful tool for further basicity studies in THF as well as in other solvents, in particular, in acetonitrile. A good correlation between basicities in THF and acetonitrile spanning 25 orders of magnitude gives access to experimentally supported very high (pK_a > 40) basicities in acetonitrile, which cannot be directly measured. Analysis of structure-basicity trends is presented.

Full text of DOI:10.1021/acs.joc.6b00872

Subscriber access provided by University of Sussex Library
Article
Experimental Basicities of Superbasic
Phosphonium Ylides and Phosphazenes
Jaan Saame, Toomas Rodima, Sofja Tshepelevitsh, Agnes Kütt,
Ivari Kaljurand, Tõiv Haljasorg, Ilmar A. Koppel, and Ivo Leito
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b00872 • Publication Date (Web): 08 Jul 2016
Downloaded from http://pubs.acs.org on July 10, 2016
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 free 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 accessible to all
readers and 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.
The Journal of Organic Chemistry 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 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Experimental Basicities of Superbasic Phosphonium Ylides and
Phosphazenes
Jaan Saame, Toomas Rodima, Sofja Tshepelevitsh, Agnes Kütt, Ivari Kaljurand, Tõiv
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
*
Haljasorg, Ilmar A. Koppel and Ivo Leito
Institute of Chemistry, University of Tartu, 14a Ravila Str, 50411 Tartu, Estonia.
Abstract
Experimental basicities of some of the strongest superbases ever measured (phosphonium
ylides) are reported and by employing these compounds the experimental selfꢀconsistent
α a
basicity scale of superbases in THF, reaching pK (estimate of pK ) 35 and spanning more
than 30 pK units, has been compiled. Basicities of altogether 47 compounds (around half of
a
which are newly synthesized) are included. The solution basicity of the wellꢀknown tꢀBuꢀ
4 9
N=P (dma) phosphazene superbase is now rigorously linked to the scale. The compiled scale
is a useful tool for further basicity studies in THF as well as in other solvents, in particular in
acetonitrile. A good correlation between basicities in THF and acetonitrile spanning 25 orders
of magnitude gives access to experimentally supported very high (pK over 40) basicities in
a
acetonitrile, which cannot be directly measured. Analysis of structureꢀbasicity trends is
presented.
*
Corresponding autor: eꢀmail: ivo.leito@ut.ee, phone: +372 51 84 176.
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 51
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
Introduction
Significant effort has been made in research of strongly basic compounds in recent decades
and base strength has always been one of the key considerations. Although the strongest
1
,2
known bases are mostly ionic compounds (alkali metal alkyls, amides and hydrides), nonꢀ
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
ionic bases are preferred in many applications as they usually offer milder reaction conditions,
3
–6
“cleaner” deprotonation and less sideꢀproducts. The most basic nonꢀionic compounds that
have been systematically experimentally studied are different substituted phosphazenes.
7,8
Initially synthesized and described by the Schwesinger group, they have been modified by
9–11
11–13
others,
studied in different solvents
and applied in numerous synthetic
3
–5,14–18
transformations.
3,5,14,15
The increasing interest in strong nonꢀionic bases – for use as catalysts
and auxiliary
3
reagents in synthesis – has motivated the creation of a variety of novel (potentially)
19
superbasic compound families, for example phosphatranes and their derivatives, BIG
20
9,21
10,22
bases, cyclopropenimine superbases,
bisphosphazene proton sponges
. Progress has
also been made in computational estimation of solutionꢀphase basicities and new exciting
23
families of bases have been proposed on the basis of computations: polycyclic proton
24
25
26
sponges, calixaꢀpyridines, etc.
Basicity is a core characteristic of any basic molecule, and practical application as well as
systematic development of new superbases require reliable basicity data of the existing
11,26,27
molecules and understanding of the factors governing basicity.
Other desired properties
of new compounds, besides high basicity, are high stability in nonꢀinert media (of both neutral
and protonated forms), low nucleophilicity, not too high molar mass, sufficient thermal
stability, nonꢀtoxicity and low cost.
2
ACS Paragon Plus Environment

Page 3 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
8
,12,13,20
Basicities of strong bases have been measured in different solvents,
but in contrast to
development of new bases, progress in basicity measurement has been less pronounced.
20
Although very strong experimental basicity values have been recently reported (BIG bases,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
8
9
imidazolidine ylidene amines, higher order cyclopropeneimine bases ), the most recent
1
1
rigorous extension of the selfꢀconsistent basicity scale in THF dates back to 2005 and
reaches pK value 28. pK values are estimates of free ion pK values that have been obtained
from measurements of ionꢀpair equilibria with correction for ion pairing, see below. The selfꢀ
consistent basicity scale in acetonitrile (MeCN) currently reaches pK value 32. Basicity
measurements of new superstrong bases are essentially comparisons with reference bases
and can be reliably carried out when basicity differences do not exceed 1.5 (preferably 1.0)
pK units and the role of selfꢀconsistent scales is providing reliable basicity values for such
α
α
a
a
13,20
a
reference bases. Thus, researchers are currently without a convenient tool for experimental
determination of basicities of the strongest bases.
The experimental studies of superbasicity in solution can be conveniently carried out in a
rather limited number of solvents. The solvent is required to be sufficiently differentiating
(
i.e. to have very low acidity), stable in the presence of highly basic compounds, available at
high degree of purity and have adequate dissolving ability and optical properties suiting the
chosen measurement method. The best known solvents that meet these requirements are
ethers (incl. THF), dimethyl sulfoxide, to some extent acetonitrile and some others.
Occasionally good correlations have been found between basicity values in different
8
,13
solvents.
This enables estimation of basicities in practically problematic cases by
correlation analysis from the experimental data in another solvent. A good example is
8,20
prediction of pK
a
values of strong bases in acetonitrile from data obtained in THF.
The
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 51
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
validity of such approach in the case of phosphazene bases has been recently indirectly
27
supported by computations. In view of the above, THF is an obvious choice for
superbasicity measurements. Its low acidity and high stability enable measuring very high
basicity values and a remarkable amount of basicity data is already available in THF.
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
The main aim of this work, was to significantly extend the existing selfꢀconsistent basicity
11,12,20,29
scale in THF, published in different papers,
towards stronger bases, reaching pK
α
around 35 and bringing its full span to 30 orders of magnitude. The resulting scale is
envisaged as a useful tool for future studies of very strong bases and will open up the
possibility of estimating (using correlation analysis) the pK
other solvents, importantly in acetonitrile, which is widely used for acidꢀbase studies but
where direct experimental pK measurements at this high basicity level are not possible.
a
values of the strongest bases in
a
It is preferable to conduct measurements of acidꢀbase equilibria in highly diluted solutions to
minimize the impact of various association reactions on the results. Of the many methods that
9,29,30
7,8,31,32
33
have been used for superbasicity studies (NMR,
, potentiometry,
conductometry,
1
2,13,20
UVꢀVis spectrophotometry
) the UVꢀVis spectrophotometric method has proven to yield
the most reliable results, as it is possible to work at low concentrations of compounds and
easily detect problems, such as side reactions and nonꢀreversibility of deprotonation.
In order to obtain reliable basicity values with the UVꢀVis spectrophotometric method,
compounds are needed that besides very high basicity also have suitable spectral properties –
i.e. the neutral and protonated forms of the compound have to have different spectra. The key
compound family for extending the scale, which satisfies both requirements, is a series of
newly synthesized phosphonium ylide superbases (Scheme 1). The higher basicities of
4
ACS Paragon Plus Environment

Page 5 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
phosphonium ylides compared to the corresponding phosphazenes with similar molar
2
7
34,35
weight give them practical importance in organic synthesis
and because of that their
basicities are also interesting by themselves. Only few members of this superbase family have
31,32,34
experimental basicity values available.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
The simple form of dissociation equation of a cationic acid HB (protonated base B)
(Equation 1) adequately represents the proton transfer process in polar media, but does not
hold in lowꢀpolarity solvents, such as THF, where ion aggregation takes place because of
36
insufficient screening of charged particles (relative permittivity 7.6 ).
+
K
a
+
HB + S
→

B + HS
(1)
(2)
←
+
a(B) ⋅ a(HS )
K =
a
+
a(HB )
As most ions in THF exist in the form of contact or solventꢀseparated ion pairs and direct
+
measurement of solvated proton activity a(HS ) in THF is very difficult, direct computation
of K according to Equation 2 is impractical. To overcome these difficulties, relative (instead
a
of absolute) basicity of two bases is measured and furthermore, equilibrium between ion pairs
(
Equation 3) is often studied in lowꢀpolarity solvents, instead of equilibrium between free
1
2
solvated ions.
+
−
+
−
HB A
2
d
HB A
1
K
K
1
/ K
d
+
1
ꢀ
+
2
ꢀ
+
2
ꢀ
+
1
ꢀ
→

→
→ HB + B + A
B +HB +A ←
HB A + B (3)
B +HB A
1
←
1
2
←
2
ꢀ
1 2 d
Protonated bases B and B form ion pairs with a counterꢀion A . K denotes the dissociation
constant of the respective ion pair. The value directly measured in titration experiments is the
difference of ionꢀpair basicities pKip of the bases, the soꢀcalled relative ionꢀpair basicity,
denoted as ꢀpKip and can be found from eq 4:
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 51
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
+
−
HB A
+
ꢀ
1
K ⋅ Kd
a(HB A ) ⋅ a(B )
2
1
+
ꢀ
+
ꢀ
(4)
ꢀ
pK = pK (HB A ) − pK (HB A ) = log
= log
ip
ip
2
ip
1
+
2
−
+
1
ꢀ
HB A
Kd
a(HB A ) ⋅ a(B )
2
3
7
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
Ion pair dissociation constants K
d
are estimated using the Fuoss equation and ꢀpK
α
(the
estimate of pK difference of bases B
a
1
2
and B ) is calculated according to Equation 5:
+
1
−
−
HB A
Kd
+
+
2
+
1
ꢀ
pK = pK (HB ) − pK (HB ) = ꢀpK − log
(5)
α
α
α
ip
HB A
2
Kd
6
ACS Paragon Plus Environment

Page 7 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 1. Bases studied in this work.
Phosphonium ylides
R3
1
2
R , R = H-, Me-, MeO-, Ph-,
-NO -C H -, 3-Cl-C H -
R1
H
N
4
R4
2
6
4
6
4
P
3
4
P
N
N
N
R2
R , R = Ph-, C H11-,1-Napht-,
R3
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
dma, pyrr, tmg
3-MeO-C H -, 4-MeO-C H -
2,6-(MeO) -C H -,
2 6 3
,5-(MeO) -C H -
,4,6-(MeO) -C H -,
,4,6-Me -C H -
6
4
6
4
1
2
3
4
R R C=P(R ) R
2
40
3
2
2
2 6 3
3
6
2
Ph-CH=P(-NMe-CH -CH -) N
2
2
3
3
6
2
P phosphazenes
P phosphazenes
2
1
R1
Et-
R2
imme 12
tmg 16
Ph(CH )CH- imme 20
R1
R2
N
N
R2
R2
P
R2
P
Et-
t-Bu-
Ph-
dma 42
dma 43
dma 50
Et-
R1
R2
R1
R2
N
P
N
N
3
R2
R2
R2
H-
t-Bu-
H-
imme 21
tmg 24
tmg 26
pyrr 55
2-Cl-C H - pyrr 54
6
4
t-Bu-
Ph- 56
1
2
1
2
R ꢀN=P (R )
R ꢀN=P (R )
1
3
2
5
Ph-
P phosphazenes
P phosphazenes
3
4
R2
R2
R1
R2
R3
R1
R2
pyrr
dma
N
N
1
2
4
R2
R2
t-Bu-
t-Bu-
t-Bu-
dma NEt2 25
dma N(i-Pr)2 32
R2
R2
t-Bu-
P
N
P
N
P
P
N
P
N
P
2
t-Bu-
R
Ph(CH )CH- pyrr
Ph(CH )CH- dma
Ph-
Ph- 39
3
R1
R3
R1
R2
N
N
N P
7
2
2
-Cl-C H - dma NEt2 48
,5-Cl-C H - pyrr NEt2 51
3
6
4
R2
R2
Ph(CH )CH- Ph- 22
3
6
3
R2
R2
R2
4
-MeO-C H - pyrr 27
6 4
Ph-
pyrr 30
R2
R2
1
2
3
1
2
R ꢀN=P (R ) R
R ꢀN=P (R )
3
6
4
9
P
N
N
N
N
N
N
N
N
N
N
N
N
45
dma
pyrr
tmg
imme
C H
6 11
P(-NMe-CH -CH -) N
2
2
3
a
See Table 1 for the substituent configurations of specific ylides.
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 51
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
Results
The pKip measurements were carried out with 47 bases using altogether 87 overlapping ∆pKip
measurements (see Table 1). This way the pKip range of 23 to 34 (pK range of 23 to 35) in
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
α
THF is now covered by rigorous multiply overlapping selfꢀconsistent basicity measurements.
Both ionꢀpair (pK ) values and values corrected for ion pairing (pK estimates, denoted as
ip
a
α
pK ) are given in Table 1. Together with the earlier data there is now continuous selfꢀ
α
consistent basicity scale available in THF solution ranging from pK 3 to 35.0, i.e. spanning
more than 30 orders of magnitude, see Table 1.
8
ACS Paragon Plus Environment

Page 9 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Selfꢀconsistent basicity scale in THF solution.
b
a
pK ip
THF)
pK
α
pK
a
No Compound
∆pK ip
(
(THF)
(MeCN)
c
1
2
3
4
5
6
7
8
9
t ꢀBuꢀN=P
t ꢀBuꢀN=P
4
(pyrr)
9
34.2
35.3
0
44.0
0.0
c
4
(dma)
9
1.4
0.8
33.0
33.9
0
42.7
0.0
0
.0 0.0
H
2
C=P(2,4,6ꢀ(MeO)
3
ꢀC
(pyrr)
)C=P(2,4,6ꢀ(MeO)
ꢀC
6
H
2
)
2
Ph
2.0
32.9
33.5
0.0
42.1
42.6
41.6
40.9
41.2
40.0
39.9
40.0
39.0
39.2
38.9
38.6
38.4
38.7
38.3
37.7
38.0
38.0
37.6
38.2
36.9
37.3
37.6
37.2
37.1
36.5
35.9
36.6
35.8
36.4
35.6
35.3
34.9
34.9
34.4
34.5
34.7
0.0
1
.1
Ph(CH
Ph(CH
3
3
)CHꢀN=P
4
9
1.2
32.9
33.9
0
0.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ꢀC
6
H
2
)
3
1.2
32.1
33.0
0
0.0
0.3
(CH
Ph(CH
MeOꢀCH=P(4ꢀMeOꢀC
(CH C=P(4ꢀMeOꢀC
)C=P(2,6ꢀ(MeO)
ꢀCH=P(C Ph
EtꢀN=P (imme)
(CH C=P(3,5ꢀ(MeO)
C=P(2,4,6ꢀ(MeO) ꢀC
CH ꢀCH=P(C 11)Ph
EtꢀN=P (tmg)
MeOꢀCH=P(C
(CH C=P(dma)
C=P(C Ph
20 Ph(CH )CHꢀN=P
(imme)
)CHꢀN=P
3
2
) C=P(2,4,6ꢀ(MeO)
3
6
H
2
)Ph
2
0.3
2.2
31.8
0.0
32.4
0.0
1.4
31.7
0.0
32.7
0
3
)CHꢀN=P (dma)
4
9
1.2
0
.8 1.1
6
H
4
)
3
30.9
31.7
0
0.0
)
3 2
6
H
)
4 3
0.1
30.8
31.5
0
0.0
0.0
2.0
10 Ph(CH
11 CH
3
2
ꢀC
6
H
3
)
3
30.8
31.6
0
0.0
0.4
3
6
H
11
)
2
30.3
30.8
0
0.0
0.8
1
1
1
1
1
1
1
2
3
4
5
6
7
8
1
3
1.0 1.1
30.2
30.9
0
0.0
0.5
)
3 2
2
ꢀC
6
H
)
3 3
0.5
29.9
30.7
0
0.0
0.0
0.4
0.0
H
2
3
6
H
2
)Ph
2
29.8
30.4
0
0.0
1.5
0.5
3
6
H
2
0.0
29.8
0.0
30.2
0
0.1
1
3
29.7
0.0
30.4
0
1.0
0.1 0.2
a
6
H
11)Ph
2
0.8
29.7
30.1
0.0
Colors are used according to
)
3 2
3
1.0
29.5
29.6
0
0.0
0.7
protonation center: carbon –
black, nitrogen – blue,
19
H
2
6
H
)
11 2
0.9
29.4
29.9
0
0.0
3
1
(imme)
3
29.0
29.7
0
0.0
2.0
1.0
2
2
2
2
2
2
2
2
1
2
3
4
5
6
7
8
HꢀN=P
Ph(CH
(CH C=PPh
1
3
28.8
29.4
0
0.0
0
.3 1.8
1.1
b
phosphorus – green. If not
3
4
(Ph)
9
28.8
30.0
0
0.0
)
3 2
3
28.7
28.9
0
0.0
1.1
a
stated otherwise, pK values
d
t ꢀBuꢀN=P
t ꢀBuꢀN=P
1
(tmg)
3
0.8
0.5
28.4
29.1
0
0.0
3
(dma)
6
NEt
2
28.3
0.0
29.4
0
in MeCN were estimated
from the data in THF using
1.2
HꢀN=P
4ꢀMeOꢀC
C=P(4ꢀMeOꢀC
ꢀCH=PPh
30 PhꢀN=P (pyrr)
ꢀCH=P(1ꢀNapht)Ph
t ꢀBuꢀN=P (dma) N(i- Pr)
PhꢀCH=P(pyrr) tmg
C=P(3,5ꢀ(MeO) ꢀC
C=P(3ꢀMeOꢀC
tmg
1
(tmg)
3
28.1
0.0
29.0
0
1.3
1.0
d
6
H
4
ꢀN=P
4
(pyrr)
9
27.8
28.9
0
0
0.8 0.6
H
2
6
H
4
)
3
27.8
28.5
0
0.0
0.2
1.3
linear regression. pK values
a
29 CH
3
3
0.6
27.5
27.9
0
0.0
0
.2
.2
4
9
0.2
27.4
28.5
0
0.0
of phosphazenes were
0
3
3
3
3
3
3
3
1
2
3
4
5
6
7
CH
3
2
27.3
27.8
0.0
0.8
3
6
2
3
1.0
0.6 0.2
27.1
28.2
0
calculated by Eq 9 and pK
a
0.0
1.7
0.9
2
0.9
27.1
27.7
0
0.0
0.6
values of phosphonium ylides
H
H
2
2
2
6
H
3
)
1.1
26.7
0.0
27.5
0
6
H
4
)
3
1.0
26.6
27.1
0
c
8
0
by Eq 10. Literature values.
0.4
PhꢀCH=P(dma)
C=PPh
CꢀCH=PPh
39 t ꢀBuꢀN=P (Ph)
40 PhꢀCH=P(ꢀNMeꢀCH
2
0.6
26.6
0.0
27.1
0.0
d
H
2
3
26.2
26.6
0
Anchor compounds of the
0.0
38 (CH
)
3 3
3
1.6
0.7
26.1
26.7
0.0
0.0
0.5
0.8
THF scale. Values in THF
3
7
1.6 1.0
0.7
25.6
26.7
0.0
0.0
1.5
e
2
ꢀCH
2
ꢀ)
3
N
25.4
26.1
0
33.18
11 e
0.0
from ref. . Experimental
1.6
4
4
4
4
4
4
4
1
2
3
4
5
6
7
H
2
C=P(2,4,6ꢀ(Me)
3
ꢀC
6
H
2
)
3
1.3
25.2
25.6
0
33.3
0.0
0.4
d
5
c
f
EtꢀN=P
t ꢀBuꢀN=P
PhꢀCH=P(pyrr)
P(ꢀNMeꢀCH ꢀCH
4ꢀNO ꢀC ꢀCH=P(2,4,6ꢀ(MeO)
PhꢀCH=P(dma)
2
(dma)
24.9
25.3
32.94
values from this work.
e
e
f
2
(dma)
5
24.4
25.0
0.0
33.13
32.46
0.0
0.5
.0 1.2
Experimental value from ref.
3
23.8
0.0
24.3
0.0
0
2
2
ꢀ)
3
N
0.4
23.8
0.0
24.1
0.0
32.90
38
. Prediction by Eq 8 gives
e
2
6
H
4
3
ꢀC
6
H
2
)
3
1.4
23.4
0.0
24.5
0.0
30.72
31.26
1.1
e
3
22.3
0
22.7
31.7.
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 51
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
The multiple overlapping measurements ensure the reliability of the obtained values and help
to estimate their selfꢀconsistency. The entire basicity range covered involves at least two
independent pathways of measurements and the relative basicity between any two bases can
be obtained by combining at least two independent sets of measurements. Reversibility of
protonation/deprotonation process was checked with all bases.
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
The absolute basicities were obtained by anchoring the scale to the pKip and pK
α
values of
three compounds for which basicity values are known: tꢀBuꢀN=P (tmg) , 4ꢀMeOꢀC H ꢀ
1
3
6 4
1
1
N=P
4
(pyrr)
9
and EtꢀN=P
2
(dma)
5
(their pKip and pK
α
values in Table 1 are from ref. ). The
12,13
pKip and pK
α
values for individual bases were found similarly to previous works.
The
essence of the approach is minimizing the sum of squares (SS) of differences between directly
measured relative ∆pKip (or ∆pK ) values and the assigned absolute pKip (or pK ) values by
α
α
allowing the absolute values to change, except those of the anchor compounds. SS is
calculated according to eq 6 (presented here for pKα):
nm
2
i
α
+
2
+
1
SS =
{
ꢀpK −
[
pK (HB ) − pK (HB )
]
}
(6)
∑
α
α
i=1
i
The sum is taken over all the measurements (altogether n
m
measurements) whereby ꢀpK
α
is
the result of a relative basicity measurement of bases B and B
1
2
(expressed as acidities of the
+
1
+
+
conjugate acids HB
and HB
2
). B
2
is the base with higher basicity. pK
α
(HB
1
) and
+
pK (HB ) are the absolute pK values for the two bases as found by the least squares
α
2
α
procedure. The precision and consistency of the results can be assessed using the consistency
13
standard deviation as defined by eq 7:
SS
s =
(7)
10
nm − nc
ACS Paragon Plus Environment

Page 11 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
The overall number of derived pKα values is denoted by nc. The consistency standard
deviation of the scale in Table 1 is 0.13, indicating good consistency for these highly
demanding measurements. In the highest basicity region of this scale it became gradually
more difficult to deprotonate the protonated bases with any of the common very strong
deprotonating agents, so that eventually only KH and metallic K worked. Indirect evidence
also points towards increasing influence of deprotonation of THF molecules in the high
basicity range. Given these increasing difficulties it is not expected that the selfꢀconsistent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
THF basicity scale can be extended much upwards from the present highest p
K
α
value 35.3.
Similarly it is not expected that p
in MeCN.
a
K values much higher than 33ꢀ34 can be directly measured
In addition to the compounds in Table 1 the basicities of four weaker bases were measured in
THF: 4ꢀNO ꢀC ꢀCH=P(dma) (53) (p ip = 17.3, p = 17.6), 4ꢀNO ꢀC ꢀCH=P(pyrr)
52) (p ip = 18.2, p = 18.7), ꢀBuꢀN=P (Ph) (56) (p ip = 19.6, p = 20.4) and 3ꢀClꢀC
CH=P(dma) (49) (p ip = 20.1, p = 20.4); and the basicities of some of the weaker bases
2
6
H
4
3
K
K
α
2
6
H
4
3
(
K
K
α
t
2
5
K
K
α
6 4
H ꢀ
3
K
K
α
were measured in MeCN. Details of these additional measurements are in the Supporting
Information.
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 51
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
Discussion
Basicities of compounds in THF and MeCN
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
3,20
Basicities in MeCN are known to correlate well with basicities in THF,
especially in the
groups of structurally similar compounds. pKa values of some phosphazenes in MeCN have
8
,20
been estimated from the experimental data in THF, but this has not yet been attempted with
20
phosphonium ylides. In this work the correlation from ref. , built upon a set of compounds
with sterically crowded nitrogen protonation sites (but otherwise belonging to different
families – amidines, guanidines, phosphazenes), was extended towards higher basicities by
8
including three alkylimino phosphazenes –
t
ꢀBuꢀN=P (pyrr) (1),
t
ꢀBuꢀN=P (dma) (2) and
4
9
4
9
EtꢀN=P (dma) (42) – and supplemented with six phosphonium ylides studied both in THF
2
6
and MeCN: PhꢀCH=P(ꢀNMeꢀCH
44), PhꢀCH=P(dma) (47), 4ꢀNO
CH=P(pyrr) (52) and 4ꢀNO
highest phosphazene homologues –
2
ꢀCH
ꢀC
ꢀCH=P(dma)
ꢀBuꢀN=P (dma) and
2
ꢀ)
3
N (40),
ꢀCH=P(2,4,6ꢀ(MeO)
(53). The acetonitrile p
ꢀBuꢀN=P (pyrr) – have previously
tꢀBuꢀN=P
2
(dma)
5
(43), PhꢀCH=P(pyrr)
3
(
3
2
6
H
4
3
ꢀC
6
H
2
)
3
(46), 4ꢀNO ꢀC
2
6
H
4
ꢀ
3
2
ꢀC
6
H
4
3
a
K values of the
t
t
4
9
4
9
8
been derived by extrapolation from data measured in THF, benzene or fluorobenzene, but
8
almost no experimental details are presented. Therefore, because (1) the extrapolation
procedure was very different from the correlation procedure used in this work, (2) three
8
different solvents were used and (3) the validity of the acetonitrile p
K
a
values of these two
27
bases has been recently indirectly supported by computations, we have decided to include
them in the overall correlation equation of this work. Figure 1 demonstrates that the new
additions fit very well into the correlation, even though phosphonium ylides have wider
scatter of data points than the other compounds. This good correlation confirms that the pK
a
12
ACS Paragon Plus Environment

Page 13 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
values of strong bases in MeCN may be estimated from THF data of the compounds from
different families.
Correlation equations have been built upon the composite set of all bases for which
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
experimental MeCN p
a
K values are available, namely MTBD, DBU, PhTMG, DMAN, 2ꢀ
NO ꢀ4ꢀCF ꢀC H ꢀN=P (pyrr) , 2,5ꢀCl ꢀC H ꢀN=P (pyrr) , 4ꢀCF ꢀC H ꢀN=P (pyrr) , Phꢀ
2
3
6
3
1
3
2
6
3
1
3
3
6
4
3
7
N=P (dma) , MeꢀN=P (dma) , PhꢀN=P (pyrr) , tꢀBuꢀN=P (pyrr) , 1, 2, 40, 42, 44, 46, 47, 52
1 3
3
7
1
3
1
3
and 53 (Equation 8), as well as subsets of 10 phosphazenes, namely 2ꢀNO
2
ꢀ4ꢀCF
(dma)
, 1, 2 and 42 (Equation 9) and 6 phosphonium
3
ꢀC
6
H
3
ꢀ
N=P
1
1
(pyrr)
3
, 2,5ꢀCl
2
ꢀC
H
6 3
ꢀN=P
1
(pyrr)3, 4ꢀCF
3
ꢀC
6
H
4
ꢀN=P
3
(pyrr)
7
, PhꢀN=P
3
7
, Meꢀ
N=P
(dma)
3
, PhꢀN=P
1
(pyrr)
3
,
t
ꢀBuꢀN=P
1
(pyrr)
3
ylides, namely 40, 44, 46, 47, 52 and 53 (Equation 10). Equations 8 (all bases) and 9
phosphazenes) are almost identical, while equation 10 is somewhat different and
(
considerably less accurate.
All bases:
p
n
p
n
p
n
K
α
a
(THF) = (0.92 ± 0.02) · pK (MeCN) – (5.15 ± 0.51) (8)
2
= 20,
R
=0.9936,
S
=0.6
Phosphazenes:
K
α
(THF) = (0.92 ± 0.02)·p
K
K
a
a
(MeCN) – (5.11 ± 0.52)
(MeCN) – (4.20 ± 3.25)
(9)
2
= 10,
R
=0.9973,
S
=0.5
Phosphonium ylides:
K
α
(THF) = (0.90 ± 0.11)·p
(10)
2
= 6,
R
=0.9441,
S
=0.9
The p
and the p
uncertainties of the predicted p
.5 and 1.5 pKa units, respectively. In the case of phosphazenes this corresponds to the overall
K
a
(MeCN) values of phosphazenes studied in this work were estimated using Equation
(MeCN) values of phosphonium ylides using Equation 10. The standard
values of phosphazenes and ylides are estimated equal to
9
K
a
K
a
0
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 51
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
standard deviation
S
of the corresponding regression. In the case of phosphorus ylides the
uncertainty has been estimated higher than
predictions.
S because of the extrapolation involved in
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
a 2 2 3
A pK (MeCN) estimate of the phosphane P(ꢀNMeꢀCH ꢀCH ꢀ) N found using Equation 8 is
38
31.7, while the corresponding experimental value is 32.90. The quality of the prediction is
good, considering that neither phosphanes nor any other bases with basicity center on
phosphorus atom are represented in the correlation.
Figure 1. Correlation between pK values of different bases in acetonitrile and pK
a
α
a
values in THF.
40
35
30
All compounds:
y = 0.92x - 5.15
25
20
15
10
5
15
20
25
30
pK
35
(MeCN)
40
45
50
a
a
Compound classes:  Phosphazenes, ꢀ amidines and guanidines, ∆ phosphonium ylides, ꢁ 1,8ꢀ
dimethylaminonaphthalene.
14
ACS Paragon Plus Environment

Page 15 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Effect of structure on basicity
The largest group of bases investigated in this work are phosphonium ylides. Their P=C bond
+
–
behaves rather as a zwitterionic P ꢀC bond and thus the ylidic carbon in many ways behaves
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
9,40
as a carbanionic center.
This feature is responsible for the main usage of ylides in today’s
organic chemistry for forming CꢀC bonds. The partly carbanionic nature of the ylidic carbon
27
causes the very high basicity of ylides, which is also typical to carbanions. Some of the
investigated ylides have structure and substitution pattern similar to phosphazenes with
known experimental basicities, so that the basicities can be directly compared. The ylide Phꢀ
2
CH=P(dma) tmg is by 8 orders of magnitude more basic than the phosphazene Phꢀ
11
N=P(dma) tmg (pKip = 18.1). The compound 4ꢀNO ꢀC H ꢀCH=P(pyrr) is by almost 5
2
2
6
4
3
11
orders of magnitude more basic than 4ꢀNO
2
ꢀC
6
H
4
1 3 α
ꢀN=P (pyrr) (pK = 13.3). Thus the
present results confirm that phosphonium ylides are by orders of magnitude more basic than
phosphazenes (phosphorus imines) with the same substitution pattern.
The set of bases used for investigation of structureꢀbasicity relation contains 16 ylides of the
1
2
3
4
1
2
3
4
2 3
general structure R R C=P(R ) R where R and R are either H or CH and R and R are
either phenyl rings with different number (from 0 to 3) of methoxy groups or cyclohexyl
groups (compounds 3, 5, 6, 9, 10, 11, 13, 14, 15, 18, 23, 28, 29, 33, 34, 36). It was found that
the basicity of this set of compounds, ranging for close to 7 orders of magnitude, can be well
described by the following equation:
pK = (26.7 ± 0.2) + (1.33± 0.12)n + (−5.0± 0.6)n + (1.40± 0.10)n
+
α
CH3
Ph
2,6ꢀMeO
(11)
15
+
(0.64± 0.09)n4ꢀMeO + (0.18± 0.06)n3,5ꢀMeO + (1.58± 0.17)nCꢀHexyl
2
(n
= 16,
R
= 0.98,
S
= 0.38)
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 51
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
1
2
where nCH3 is the number of CH groups among R and R ; nPh is the number of phenyl groups
3
1
2
among R and R ;
n
2,6ꢀMeO is the number of methoxy groups in 2 and/or 6 positions of phenyl
3
4
3
rings of R and R ; n4ꢀMeO is the number of methoxy groups in 4 position of phenyl rings of R
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
4
3
and R ;
n
3,5ꢀMeO is the number of methoxy groups in 3 and/or 5 positions of phenyl rings of R
4
3
4
and R ; nCꢀHexyl is the number of cyclohexyl rings among R and R . The intercept 26.7 ± 0.2
is formally expected to be equal to the pKα value of H C=P(Ph) and is in excellent agreement
2
3
with the experimental value 26.6 of this compound.
The good fit of Equation 11 with the experimental data shows that the influence of structural
features on basicity is to a very good approximation additive, and indirectly serves as a piece
of evidence for the good quality of the experimental data.
Examining the coefficient values in Equation 11 reveals that the most efficient way of
increasing basicity of this compound family is adding MeO substituents to the aromatic ring:
2
,4,6ꢀtrisubstitution of one ring leads to 3.2ꢀ3.5 pKa units of basicity increase. 3 and 5
substitution is significantly less efficient. Methylation of the ylidic carbon leads to more than
units of basicity increase, while introducing a phenyl substituent to the ylidic carbon
leads to dramatic – by more than 5 p units – basicity decrease. This extreme sensitivity of
the basicity center to phenyl substitution gives additional evidence that the P=C bond in ylides
1 pK
a
K
a
+
–
should rather be viewed as P ꢀC , i.e. zwitterionic. In the case of phenyl substitution the
negative charge on ylidic carbon will be efficiently delocalized in the aromatic ring and
protonation becomes significantly less favorable, as the aromatic ring stabilizes the neutral
ylide but does not offer any stabilization to its protonated form. The strong basicity redusing
effect of the 4ꢀNO
2
substituent is an additional piece of evidence.
16
ACS Paragon Plus Environment

Page 17 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Among the compounds measured in this work is (CH ) C=P(dma) (18), which has earlier
3
2
3
been reported to be by only an order of magnitude less basic than the P
4
phosphazene
28.7, which is more than six
phosphazene. This serves as a good
demonstration of the need for reliable basicity scales in studies of strong bases.
tꢀBuꢀ
34
N=P
4
(dma)
9
. The current results place this compound at p
K
α
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
orders of magnitude weaker than the
t
ꢀBuꢀN=P (dma)
4
9
1
2
3
4
2
Phosphonium ylides of the structure R R C=P(Ar ) Ar are the key family of compounds in
the present study. These compounds are useful for basicity studies, because they (in salt form)
are relatively easy to synthesize, have strong UV absorbance, and a very wide basicity range
is achievable with different substituents. Within the range of structural variations used for
compiling Equation 11 the basicity of the ylides can be easily predicted using the equation.
Already the ylides with just one phosphorus atom can have basicities comparable to P
phosphazenes, which enabled linking P phosphazenes to the scale. Equation 11 enables
estimating the p value of the strongest ylide that is possible to construct with the above
listed substituents: (CH ) C=P(2,4,6ꢀ(MeO) ꢀC6H ) . The estimate of pKα in THF is 39.6 and
4
4
K
α
3
2
3
2 3
p
Ka in MeCN (using eq 10) is 49. Both values are truly impressive for a P base.
1
The available data enable direct comparison of the effects pyrrolidino, dimethylamino and
phenyl substituents on basicity. The basicity increases in the row Ph(CH )CHꢀN=P (Ph)
Ph(CH )CHꢀN=P (dma) → Ph(CH )CHꢀN=P (pyrr) as follows: 30.0 → 32.7 → 33.9 (pKα
3
4
9
→
3
4
9
3
4
9
values). It can be seen that unsubstituted phenyl rings are weaker basicity enhancers than dma
or pyrr groups: replacing one Ph by dma would lead to roughly 0.9 p units of basicity
increase. Similar conclusion is reached by comparing ꢀBuꢀN=P (dma) and ꢀBuꢀN=P (Ph)
K
a
t
2
5
t
2
5
:
the former is by close to 5 orders of magnitude more basic.
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 51
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
At the same time, as demonstrated above, introducing methoxy groups into 2, 4 and 6
positions of a phenyl ring leads to basicity increase of more than 3 orders of magnitude. This
means that methoxylated phenyl rings connected to phosphorus are more efficient basicity
enhancers than dma or pyrr fragments. In the current scale this is demonstrated e.g. by
comparison of basicities of (CH ) C=P(2,4,6ꢀ(MeO) ꢀC H )Ph and (CH ) C=P(dma) : with
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
2
3
6
2
2
3 2
3
just one of the rings methoxylated, the former is by more than 2 orders of magnitude more
basic than the latter.
Conclusions
With the help of a series of substituted phosphonium ylides with widely tunable basicities
(more than 10 orders of magnitude) the span of the selfꢀconsistent basicity scale in THF is
α
now wider than 30 orders of magnitude and its highest basicity limit is at pK 35. The
2
excellent correlation (
R
= 0.996) between basicities in THF and acetonitrile covering a
basicity range of more than 25 orders of magnitude renders this scale a useful tool for further
basicity studies besides THF also in acetonitrile. In addition, since THF can be viewed as a
model lowꢀpolarity solvent and the basicity trends found in THF are expected to be well
applicable to other lowꢀpolarity solvents, as well as the gas phase, its potential usage is
expected to be even wider.
18
ACS Paragon Plus Environment

Page 19 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Experimental
Chemicals and consumables
Romil SpS tetrahydrofuran was used for basicity measurements after additional purification
with careful sodium/benzophenone treatment. Distillation of the 24 h stirred and 2 h refluxed
mixture yielded solvent with water content below 1 ppm (according to KarlꢀFisher titration
with Mettler Toledo DL32 Coulometer). Romil SpS acetonitrile was used after drying over
molecular sieves (3Å) for at least 24 h to achieve water content below 5 ppm. Potassium
hydride (KH, 30% suspension in mineral oil) was applied for liberation of bases in stock
solutions. The oil was washed out with THF and KH was dried under vacuum in the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
glovebox. In THF, methanesulfonic acid (>99%) solution was used as acidic titrant, and
tꢀBuꢀ
N=P (dma) (Initially 0.8M in hexane) solution as basic titrant. In MeCN the corresponding
4
9
solutions were made from trifluoromethanesulfonic acid (99+%) and EtꢀN=P
2 5
(dma) (98%) or
EtꢀN=P (pyrr) (see details of synthesis in SI).
2
5
Synthesis, Purification, and Characterization of Compounds
General
4
1
The Kirsanov reaction between PCl
5
and suitable amine/imine compounds was used to
synthesize phosphazene salts. Free phosphazene bases may be liberated from their salts by
1
1
means of
phosphazenes, if needed, were synthesized from free bases or by reaction from their HBF
HBPh salts. Trisꢀsubstituted phosphines are available commercially or were synthesized
t
ꢀBuOK in dimethoxyethane (DME) as described in ref. . Triflate salts of
4
or
4
according to literature procedures. The reaction between lithiated or Znꢀcatalyzed aryls or
Grignard reagents of halogen aryls with phosphorus halogenides was used. The following
3
6,37
42,43
phosphines were prepared: P(2,4,6ꢀ(MeO) ꢀC H ) ,
P(2,4,6ꢀ(MeO) ꢀC H ) Ph,
P(2,4,6ꢀ
3
6
2 3
3
6
2 2
4
2,43
44,45
46
42,43,47
(MeO) ꢀC H )Ph ,
P(Mes)3,
P(3ꢀMeOꢀC H ) , P(2,6ꢀ(MeO) ꢀC H ) ,
P(3,5ꢀ
3
6
2
2
6
4 3
2
6
3 3
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 51
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
4
7,48
49
49
50
50
51
(MeO)
2
ꢀC
6
H
3
)
3
,
P(dma)
3
,
P(pyrr)
3
,
P(C
6
H
11
)
2
Ph, P(C
6
H
11)Ph
2
,
2
P(Napht)Ph ,
52
52
P(dma)
2
Cl, P(pyrr)
2
Cl.
In the reaction between P(pyrr) Cl or P(dma) Cl and TMG and also between P(pyrr)Cl or
2
2
2
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
P(dma)Cl
2
and TMG the same products, i.e. P(pyrr)
2
tmg and P(dma)
2
tmg were obtained,
respectively. It seems to be due to the disproportionation of already formed compounds on the
5
3,54
reaction and/or distillation.
41,55
Phosphonium salts were synthesized according to eq 12:
R P + R´X → R3P+−R´ + X → R P ꢀR´ + CF SO
(
X = Cl, Br, I)
−
CF
3
SO
3
Na
+
−
(12)
The reactions were carried out sometimes as a neat mixture of reagents or in benzene, toluene,
DME, EtOH, ꢀPrOH or nitromethane solution under reflux. Benzene or alcohols as solvents
3
3
3
3
i
sometimes give strongly solvated phosphonium halogenides. To destroy these adducts
recrystallization from THF/MeOH may be recommended following crushing the crystal mass
with dichloromethane and concentrating the extracts in high vacuum. Dry solvents and argon
atmosphere were used for all syntheses except for metathesis reactions. To make
phosphonium salts more soluble in THF and to unify anions of compounds for acidꢀbase
measurements the halogenide anions were exchanged for triflate anion by metathesis reaction
with sodium triflate in water or in water/alcohol mixture.
8
11
11
The following compounds were synthesized as described in literature: 1, 16, 24 – 27,
1
1
8
56,57
58
8
59
58
30, 32, 37,
39, 43, 45, 56. Compounds 2 (0.8 M solution in hexane, hexane was
evaporated before use) and 42 (98%) have commercial origin and were used without
purification.
1
13
31
The structure and purity of synthesized compounds were checked by H, C and P NMR
spectrometric analyses on, a 200 MHz, 400 MHz or 700 MHz NMR spectrometer. Chemical
20
ACS Paragon Plus Environment

Page 21 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
shifts are given relative to TMS as internal standard for H NMR spectra, relative to the
60
13
31
solvent residual peak or to TMS for C NMR spectra, and relative to 85% H PO for P
3
4
NMR spectra. High resolution ESI mass spectra were obtained on a hybrid FTꢀICRꢀMS
system. Melting points are uncorrected. Sometimes in the course of melting of phosphonium
salts they slowly decompose. The synthesis of phosphonium salts and phosphazenes was not
optimized.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Typical examples of starting compounds synthesis
4
2
P(2,4,6ꢀ(MeO)
The mixture of 1,3,5ꢀtrimethoxybenzene, phenyldichlorophosphine and ZnCl
30:10:30 mmol (or 1,3,5ꢀtrimethoxybenzene, diphenylclorophosphine and ZnCl in relation
3 6 2 2 3 6 2 2
ꢀC H ) Ph and P(2,4,6ꢀ(MeO) ꢀC H )Ph
2
in ratio
2
1
0:10:30 mmol respectively) was heated at 65 °C for 6ꢀ8 hours and left overnight at room
temperature. Benzene (30 mL) and 12% NH OH solution (30 mL) were added to the sticky
̴
4
mass, an extra stirrer bar was added to the reaction flask, and the mixture was stirred until all
mass was dissolved. The organic layer was separated, washed with water and brine, and dried
over MgSO
4
. The benzene was evaporated in vacuum to get a crystalline mass, which may be
recrystallized from ethanol.
o
31
P(2,4,6ꢀ(MeO)
P(2,4,6ꢀ(MeO)
3
3
ꢀC
6
6
H
H
2
2
)
2
Ph: yield 40%, mp 152ꢀ153 C, P NMR (CDCl
3
, 81 MHz) δ ꢀ46.3.
o
ꢀC
)Ph
2
: yield 30%, mp 103ꢀ104 C, 31P NMR (CD
3
Cl, 81 MHz) δ ꢀ26.4.
1
43
The H NMR spectra were in accordance with literature data.
1,3ꢀDimethylꢀ2ꢀimidazolidinimine (immeH)
6
1
2
ꢀChloroꢀ1,3ꢀdimethylimidazolium chloride (65 g, 0.38 mol) was dissolved in 70 ml of
CH
2
Cl
2
and treated at 0 °C with threefold excess of NH
3
(19 g, 1.1 mol). The precipitate (1:1
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 51
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
mixture of immeHꢁHCl and NH Cl ) was filtered, washed with some CH Cl , dried and
4
2
2
dissolved in 180 ml of 26% NaOH. Ammonia was removed under reduced pressure on rotary
evaporator and the obtained precipitate was filtered off. The upper organic layer was
separated and distilled in vacuum at 46ꢀ48 °C/0.3 Torr to give 24 g (58%) of colorless liquid,
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
62
immeH. Synthesis of ImmeH has also been reported in literature.
1
13
H NMR (200 MHz, CDCl ) δ 4.04 (s, 1H), δ 3.16 (s, 4H), δ 2.73 (s, 6H). C NMR (50 MHz,
3
CDCl ) δ 163.2, 47.3, 33.0.
3
Synthesis of bases
2 3 6 2 2 3 3 3 3
H C=P(2,4,6ꢀ(MeO) ꢀC H ) PhꢁCF SO H (3ꢁCF SO H)
Bis(2,4,6ꢀtrimethoxyphenyl)phenylphosphine (0.30 g, 0.68 mmol) and 1.0 ml of CH I were
3
0
mixed and stirred with glass stick at 50 C for 15 min. The formed thick paste was diluted
with diethyl ether and filtered to give 0.39 g (0.66 mmol, 97%) of 3ꢁHI as yellowish fine
0
crystals, mp 192.8ꢀ193.7 C.
1
H NMR (200 MHz, CD CN) δ 7.68 – 7.40 (m, 5H), 6.24 (d,
J
=4.7 Hz, 4H), 3.84 (s, 6H),
=1.4 Hz),
=13.4 Hz), 126.7
PꢀC=1.9 Hz), 16.5 (d, PꢀC=64.4 Hz).
P , 457.17745; found,
3
13
3
.51 (s, 12H), 2.52 (d, JPꢀH=14.6 Hz, 3H). C NMR (50 MHz, CDCl ) δ 167.3 (d,
J
3
1
64.1 (d,
J
=1.6 Hz), 132.2 (d,
=7.0 Hz), 89.4, 87.3, 56.5 (d,
30 6
CN) δ 6.7. HRMS (ESI+): calc for C25H O
J=3.2 Hz), 130.8 (d, J=11.3 Hz), 128.7 (d, J
(d,
J
PꢀC=96.3 Hz), 92.0 (d,
J
J
J
3
1
+
P NMR (81MHz, CD
3
457.17745.
The triflate was prepared by mixing 3ꢁHI (0.22 g, 0.40 mmol) and sodium triflate (72 mg,
0.42 mmol) in warm solvent mixture (20 ml of water and 3.0 ml of ethanol). The solvent was
decanted and the glassy drops on the flask walls were treated with ethanol. The ethanolic
22
ACS Paragon Plus Environment

Page 23 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
solution was evaporated to dryness to give 0.13 g (0.21 mmol, 53%) of somewhat foamy solid
0
3
ꢁCF SO H, which melts at 60ꢀ70 C.
3
3
Ph(CH
3
)CHꢀN=P
4
(pyrr)
9
ꢁHBPh
4
(4ꢁHBPh
4
)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
3
Solution of Ph(CH
3
)CHꢀN=PCl
3
(2.2 g, 8.5 mmol, synthesized as described for compound
7ꢁHBF ) in 3.0 ml of THF was added to the solution of free phosphazene base Hꢀ
4
8
0
N=P (pyrr) (15.4 g, 60 mmol) in 12 ml of THF at ꢀ70 C. The mixture was allowed to warm
1
3
0
to room temperature, the solvent was removed and the residue was heated at 110 C for 18 h.
To the cooled mixture 30 ml of toluene and 2 ml of 1,1,2ꢀtrichloroethane was added, and the
mixture was set aside for a night. The mixture was filtered and the solution was evaporated to
dryness. The residue, raw 4ꢁHCl (11 g), was dissolved in 40 ml of 70% aqueous ethylamine
solution and 1.1 g of NaBF dissolved in 10 ml of water was added. Extra 8 ml of water was
4
added to the warm solution to generate a persistent turbidity, and then the mixture was set
4
aside for a day. The mixture was filtered to get 5.9 g (5.8 mmol, 69%) of raw 4ꢁHBF salt. It
4
was recrystallized from 1:3 mixture of Hex/EtOAc to get 2.6 g offꢀwhite crystal of 4ꢁHBF ,
0
mp 170ꢀ180 C (dec).
From the mother liquid 2.5 g (2.5 mmol) of 4ꢁHBF
4
salt was recovered. It was dissolved in 8
(0.9 g, 2.6 mmol) dissolved
ml of MeOH and transferred to 4ꢁHBPh salt by means of NaBPh
4
4
in 3 ml of MeOH. The white crystals were filtered off, recrystallized from 30 ml of 3:1
0
mixture of MeOH/MeCN to get 2.6 g (2.1 mmol, 84 %) of 4ꢁHBPh , mp 200ꢀ220 C (dec).
4
1
H NMR (200 MHz, CDCl
3
) δ 7.61 – 6.70 (m, 10H), 4.37 (dp,
J
PꢀH=13.0 Hz,
J
=6.1 Hz, 1H),
=6.6 Hz, 3H).
3.28 – 2.82 (m, 27H), 2.31 (t,
J
=6.3 Hz, 1H), 1.91 – 1.54 (m, 27H), 1.40 (d,
J
1
3
C NMR (50 MHz, CDCl
3
) δ 165.4 (d,
J
PꢀC=49.3 Hz), 163.5 (d,
J
PꢀC=49.3 Hz), 147.5 (d,
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 51
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
J
=7.5 Hz), 136.4 (dd,
Hz), 121.4 , 51.9 (d,
CDCl ) δ ꢀ0.23 (d,
J
=2.9 Hz,
J
=1.4 Hz), 128.3, 126.6, 125.9, 125.3 (dd,
J
=5.5 Hz,
=8.9 Hz). P NMR (81 MHz,
PꢀC=43.7 Hz), ꢀ19.8 (q, JPꢀC=46.9 Hz, JPꢀC=45.3 Hz).
J=2.7
31
J
=2.2 Hz), 46.6 (d, J=4.8 Hz), 26.3 (d, J
3
J
+
HRMS (ESI+): calc for C44
H
82
N
13
P
4
, 916.57611; found, 916.57607.
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 3 6 3 3 3 3 3 3
Ph(CH )C=P(2,4,6ꢀ(MeO) ꢀC H ) ꢁCF SO H (5ꢁCF SO H)
Tris(2,4,6ꢀtrimethoxyphenyl)phosphine (0.53 g, 1.0 mmol) was dissolved in 1.7 ml of hot
toluene and (1ꢀbromoethyl)benzene (0.19 g, 1.0 mmol) was added. The mixture was heated at
1
00 °C for 1 h and then cooled down. The formed crystals were filtered off, washed with
toluene and dried in vacuum. The 5ꢁHBr (0.60 g, 0.84 mmol, 84%) was obtained as colorless
crystals, mp 205.3ꢀ206.4 °C (dec).
1
H NMR (200 MHz, CD CN) δ 7.11 (s, 5H), 6.15 (d, J=4.4 Hz, 6H), 5.93 (dd, JPꢀH=20.4 Hz,
3
1
3
J
=7.2 Hz, 1H), 3.85 (s, 9H), 3.54 (s, 18H), 1.57 (dd,
50 MHz, CD CN) δ 166.4 (d, =2.0 Hz), 164.6, 139.4 (d,
128.6 (d, =3.6 Hz), 127.9 (d,
6.4, 39.5 (d, JPꢀC=48.1 Hz), 15.7.
J
PꢀH=21.5 Hz,
J
=7.2 Hz, 3H). C NMR
=5.1 Hz), 129.4 (d, =5.8 Hz),
PꢀC=99.0 Hz), 91.7 (d, =7.1 Hz), 56.6,
(
3
J
J
J
J
J
=4.2 Hz), 93.6 (d,
J
J
5
+
HRMS (ESI+): calc for C H O P , 637.25610; found, 637.25592.
35
42
9
The triflate salt was prepared by mixing 5·HBr (0.10 g, 0.14 mmol) and CF
.145 mmol) in hot water (3 ml). The crystals were washed with water and dried in vacuum to
give 80 mg (0.10 mmol, 72%) of 5ꢁCF SO H, mp 201.3ꢀ202.3 °C.
3 3
SO Na (25 mg,
0
3
3
1
H NMR (200 MHz, CD CN) δ 7.22 – 7.03 (m, 6H), 6.14 (m, 5H), 5.93 (dq, JPꢀH=20.5 Hz,
3
J
=6.9 Hz, 1H), 3.85 (d,
C NMR (50 MHz, CD
J=6.0 Hz, 9H), 3.54 (s, 18H), 1.57 (dd,
J
PꢀH=21.4 Hz,
J
=7.1 Hz, 3H).
=5.9 Hz),
1
3
3
CN) δ 163.4, 138.9 (d, =4.9 Hz), 136.0, 136.0, 129.4 (d, J
J
24
ACS Paragon Plus Environment

Page 25 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
128.7 (d,
J
=3.7 Hz), 128.2 (d,
J
=4.4 Hz), 105.0 (d,
J
=6.8 Hz), 101.6 (d, JPꢀC=91.5 Hz), 56.3,
3
1
39.4 (d, JPꢀC=44.8 Hz), 15.84. P NMR (81 MHz, CD CN) δ 19.9.
3
(
CH
ꢀPropyl(2,4,6ꢀtrimethoxyphenyl)diphenylphosphonium iodide (6ꢁHI) (0.10 g, 0.19 mmol)
was dissolved in 1.5 ml warm ethanol and CF SO Na solution (36 mg, 0.20 mmol) in 0.5 ml
of ethanol was added. For precipitation of the product 4.0 ml of water was added and the
3 2 3 6 2 2 3 3 3 3
) C=P(2,4,6ꢀ(MeO) ꢀC H )Ph ꢁCF SO H (6ꢁCF SO H)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
4
i
3
3
3 3
mixture was stored for a day. The collected colorless crystals of 6ꢁCF SO H were filtered off
and dried. Yield 80 mg (0.147 mmol, 77%), mp 155.0ꢀ155.6 °C.
1
H NMR (200 MHz, CD
H), 3.89 (s, 3H), 3.64 (s, 6H), 1.10 (dd,
CD CN) δ 26.7
3
CN) δ 7.90 – 7.52 (m, 10H), 6.32 (d,
J=4.4 Hz, 2H), 4.16 – 3.91 (m,
31
1
JPꢀH=19.2 Hz, J
=6.9 Hz, 6H). P NMR (81 MHz,
3
Ph(CH
Ph(CH
phenylꢀethylamine and 20.5 g (96 mmol) of PCl
3
)CHꢀN=P
4
(dma)
was synthesized by the Kirsanov method from 11 g (96 mmol) of
in 130 ml of CCl . The mixture was stirred
9 4 4
ꢁHBF (7ꢁHBF )
6
3
3
)CHꢀN=PCl
3
5
4
and heated until the evolution of HCl gas stopped. The solution was filtered, the solvent was
0
removed and the residual oil was distilled at 70ꢀ75 C/0.3 Torr. Yield 15 g (60 mmol, 62%).
8
HꢀN=P
1
(dma)
3
(10.5 g, 58.9 mmol) was dissolved in 12 ml of THF and the solution of
(2.2 g, 8.5 mmol) in 3 ml of THF was added at ꢀ70 °C with a syringe.
Ph(CH
3
)CHꢀN=PCl
3
The mixture was allowed to warm to room temperature, the solvent was removed and the
residue was heated at 110 °C for 24 h. To the cool mixture of 23 ml of THF and 2 ml of 1,1,2ꢀ
trichloroethane was added and the HꢀN=P
1 3
(dma) ꢁHCl was filtered off. The filtrate was
concentrated and the solvent was completely removed under high vacuum. The raw product
7ꢁHCl, brownish oil, was dissolved in 25 ml of 70% EtNH
2
aqueous solution and 1.0 g (9.1
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 51
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
mmol) of NaBF dissolved in 3 ml of water was added. The clear mixture was slightly
4
warmed and water was added up to persistent turbidity (about 15 ml). A soft waxy precipitate
formed. The solvent was decanted and the residue was once more treated with 70 % EtNH
solution (40 ml) and water (42 ml) and set aside for 2 days. Colorless crystals were filtered
off, washed with 30% EtNH aqueous solution and dried in vacuum. Yield 2.6 g (3.5 mmol,
2
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
41%) of 7ꢁHBF , mp 201ꢀ205 °C.
4
1
H NMR (200 MHz, CDCl
2.43 (m, 54H), 2.41 – 2.21 (m, 1H), 1.41 (d,
147.2 (d, =6.2 Hz), 128.2, 126.4, 125.8, 51.6 (d,
3
) δ 7.44 – 7.09 (m, 5H), 4.38 (dp,
=6.7 Hz, 3H). C NMR (50 MHz, CDCl
=1.7 Hz), 37.1 (d, =4.5 Hz), 26.6 (d,
J
=14.1 Hz,
J
=6.8 Hz, 1H), 2.90
13
–
J
3
) δ
J
J
J
31
J
=5.3 Hz). P NMR (81 MHz, CDCl ) δ 14.6 (d, JPꢀC=47.8 Hz), ꢀ19.6 (q, JPꢀC=49.6 Hz,
3
+
JPꢀC=47.8 Hz). HRMS (ESI+): calc for C H N P , 682.43526; found, 682.43526.
26
64 13 4
6 4 3 3 3 3 3
MeOꢀCH=P(4ꢀMeOꢀC H ) ꢁCF SO H (8ꢁCF SO H)
P(4ꢀMeOꢀC (0.25 g, 0.71 mmol) and excess of MeOCH Cl (0.57 g, 7.0 mmol) were
6
H
4
)
3
2
mixed in 1.0 ml of DME at 40 °C and stirred for 2 h. Colorless crystals were filtered off,
washed with DME and dried in vacuum to give 0.27 g, (0.62 mmol, 87%) of 8ꢁHCl, mp 194ꢀ
195 °C.
1
H NMR (400 MHz, CD
H), 3.91 (s, 9H), 3.53 (d,
Hz), 136.97 (d, =11.6 Hz), 116.88 (d,
Hz), 62.65 (d,
3
CN) δ 7.76 – 7.56 (m, 6H), 7.29 – 7.16 (m, 6H), 5.28 (d,
J
=4.7 Hz,
CN) δ 166.0 (d, =3.0
J=13.8 Hz), 108.6 (d, JPꢀC=94.5 Hz), 67.1(d, JPꢀC=71.1
13
2
J
=0.8 Hz, 3H). C NMR (101 MHz, CD
3
J
J
31
J
=13.3 Hz), 56.77. P NMR (162 MHz, CD CN) δ 15.5.
3
+
HRMS (ESI+): calc for C23
H
26
O
4
P , 397.15632; found, 397.15644.
26
ACS Paragon Plus Environment

Page 27 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
The triflate was prepared by mixing 8ꢁHCl (0.20 g, 0.46 mmol) and CF SO Na (80 mg, 0.46
3
3
mmol) in 5.0 ml of water. An oil separated which was washed with water and then dissolved
in ethanol. Ethanolic solution was concentrated to give colorless oil of 8ꢁCF SO H (0.16g,
.29 mmol, 63%,) which did not crystallize even at ꢀ15 °C.
3
3
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(CH
3 2 6 4 3 3 3 3 3
) C=P(4ꢀMeOꢀC H ) ꢁCF SO H (9ꢁCF SO H)
The mixture of P(4ꢀMeOꢀC H ) (0.52 g, 1.47 mmol), isopropyl iodide (0.36 g, 2.1 mmol)
6
4 3
and 6.0 ml of nitromethane was stirred at 100 °C for 2 h. The solvent was evaporated at
reduced pressure and the residue, a glassy mass, was triturated with glass stick with 3.0 ml of
benzene to wash off the starting phosphine. The solid was filtered off and dried under high
1
vacuum at 60 °C for 2 h. H NMR showed the compound to be 1:1 adduct of 9ꢁHI and
benzene (0.53 g, 0.88 mmol, 60%), and it does not have an exact melting point.
1
H NMR (200 MHz, CD
3
CN) δ 7.80 – 7.62 (m, 6H), 7.37 (s, C
6 6
H ), 7.21 (dd, J=9.0 Hz, J=2.6
3
1
Hz, 6H), 4.01 – 3.77 (m, 10H), 1.26 (dd,
CD CN) δ 28.6
JPꢀH=18.4 Hz, J
=7.0 Hz, 6H). P NMR (81MHz,
3
The triflate salt was prepared by mixing 9·HI (0.23 g, 0.38 mmol) and CF SO Na (76 mg,0.44
3
3
mmol) in 3.0 ml of warm isopropanol. The solvent was removed at 60 °C/0.05 Torr. The
residue, a sticky mass, was treated with CH Cl , filtered and solution was evaporated to
dryness to give 80 mg (0.147 mmol, 38%) of colorless glassy mass, 9ꢁCF SO H, mp 61ꢀ63 °C
2
2
3
3
(dec).
1
H NMR (200 MHz, CDCl ) δ 8.00 – 7.63 (m, 6H), 7.28 – 7.04 (m, 6H), 4.67 (qd,
J
=6.9 Hz,
=6.9 Hz, 6H). C NMR (50
=10.7 Hz), 116.3 (d, =13.2 Hz), 108.4 (d,
PꢀC=50.4 Hz), 16.6 (d, =2.1 Hz).
3
1
3
J
=3.7 Hz, 1H), 3.92 (d,
) δ 164.5 (d,
PꢀC=91.1 Hz), 56.1, 22.7 (d,
J
=7.1 Hz, 9H), 1.30 (dd,
JPꢀH=18.6 Hz, J
MHz, CDCl
3
J
=2.9 Hz), 135.8 (d,
J
J
J
J
J
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 51
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
+
HRMS (ESI+): calc for C H O P , 395.17706; found, 395.17700.
24
28
3
Ph(CH
3
)C=P(2,6ꢀ(MeO)
2 6 3 3 3 3 3 3
ꢀC H ) ꢁCF SO H (10ꢁCF SO H)
P(2,6ꢀ(MeO)
2
ꢀC (0.50 g, 1.13 mmol) was dissolved in 25 ml of warm toluene and (1ꢀ
6 3 3
H )
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
bromoethyl)benzene (0.22 g, 1.18 mmol) was added. The mixture was heated at 80 °C for 1 h,
cooled and the formed precipitate was filtered off, washed with toluene and dried in vacuum.
The 10ꢁHBr was obtained (0.52 g, 0.75 mmol, 73.3%) as a white powder, mp 217ꢀ218 °C
(dec).
1
H NMR (200 MHz, CD
3
CN) δ 7.50 (t,
=7.2 Hz, 1H), 3.55 (s, 18H), 1.62 (dd,
H). C NMR (50 MHz, CD CN) δ 165.4, 165.4, 163.6, 138.4 (d, =5.1 Hz), 128.4 (d,
=4.2 Hz), 117.2 , 92.6 (d, JPꢀC=99.0 Hz), 90.7 (d,
J
=8.4 Hz, 3H), 7.26 – 7.00 (m, 5H), 6.73 – 6.52 (m,
6H), 6.16 (dd,
J
PꢀH=20.8 Hz,
J
JPꢀH=21.8 Hz, J
=7.2 Hz,
1
3
3
J
J
J
=5.8
=7.1
3
Hz), 127.6 (d,
Hz), 55.5, 55.3, 38.5 (d,
HRMS (ESI+): calc for C32
J=3.6 Hz), 126.9 (d, J
31
J
3
PꢀC=48.1 Hz), 14.7. P NMR (81 MHz, CD CN) δ = 22.7.
+
H
36
O
6
P , 547.22440; found, 547.22437.
The triflate was prepared by reaction mixing the bromide salt 10ꢁHBr (0.33 g, 0.60 mmol)
and CF SO Na (0.11 g, 0.60 mmol) in warm 1:1 mixture of ꢀPrOH/H O to give 0.31 g (0.50
i
3
3
2
mmol, 83%) of 10ꢁCF SO H, mp 300.0ꢀ301.5 °C.
3
3
1
3
H NMR (200 MHz, CD CN) δ 7.63 – 7.25 (m, 3H), 7.28 – 6.83 (m, 5H), 6.77 – 6.33 (m,
1
3
6H), 6.32 – 5.96 (m, 1H), 3.53 (s, 18H), 1.63 (dd,
J
=21.8 Hz,
=4.9 Hz), 136.0 (d, =1.2 Hz), 129.4 (d,
=6.8 Hz), 101.6 (d, JPꢀC=91.5 Hz), 56.3, 39.4 (d,
J
=7.2 Hz, 3H). C NMR (50
MHz, CD
d, =3.7 Hz), 128.2 (d,
JPꢀC=44.8 Hz), 15.8.
3
CN) δ 163.4, 138.9 (d,
J
J
J
=5.9 Hz), 128.7
(
J
J
=4.4 Hz), 105.0 (d, J
3 6 11 2 3 3 3 3
CH ꢀCH=P(C H ) PhꢁCF SO H (11ꢁCF SO H)
28
ACS Paragon Plus Environment

Page 29 of 51
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
5
0
The salt 11ꢁHBr (0.10 g, 0.26 mmol) was converted into triflate by reaction with CF
50 mg, 0.29 mmol) in 1.0 ml of water at room temperature. The crystals were recrystallized
from 4:1 mixture of AcOEt/EtOH to give 60 mg (51%) of pure product (11ꢁCF SO H), mp
3 3
SO Na
(
3
3
1
61.4ꢀ162.5 °C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
H NMR (200 MHz, CDCl
3
) δ 7.92 – 7.61 (m, 5H), 2.91 – 2.57 (m, 2H), 2.24 – 0.96 (m,
) δ 32.9.
31
23H). P NMR (81 MHz, CDCl
3
EtꢀN=P (imme) ꢁHBF (12ꢁHBF )
1
3
4
4
In 25 ml of CH
2
2 5
Cl 2.05 g (9.86 mmol) of PCl was suspended at ꢀ40 °C and 6.7 g (59 mmol)
of immeH (vide supra) was slowly added. The mixture was stirred and allowed to warm up to
room temperature and additionally stirred for another 1 h. Then the mixture was cooled again
to ꢀ30 °C and a cold mixture of 0.68 g (15 mmol) ethylamine and 1.7 ml of triethylamine was
added with a syringe. The mixture was stirred until the ambient temperature was achieved and
then set aside for one day. The crystal mass that formed was filtered off and the filtrate was
concentrated in vacuum to yield 4.8 g of 12ꢁHCl. This salt was dissolved in 30 ml of water
and a solution of 1.13 g (10.2 mmol) NaBF
4 4
in 2 ml of water was added. The formed 12·HBF
was extracted with CH Cl . The CH Cl layer was separated and dried with MgSO . The
2
2
2
2
4
solvent was removed under reduced pressure to give 2.0 g of soft colorless crystals.
Recrystallization from 1:2 mixture of MeOH/AcOEt gave 1.90 g (4.0 mmol, 38%) of
1
4
2ꢁHBF , mp 142ꢀ144 °C.
1
H NMR (200 MHz, CDCl
3
) δ 1.14 (dt,
J
=0.90 Hz,
J
=7.16 Hz, 3H), 2.47 (br, 2H), 2.90 (s,
=7.5), 47.1, 36.2 (d, =1.5),
13
1
8H), 3.44 (s, 12H). C NMR (50 MHz, CDCl ) δ 156.7 (d,
J
J
3
31
3
33.7, 17.1 (d, J=8.5). P NMR (81 MHz, CDCl ) ꢀ11.06.
Anal. Calc for C17
8.26.
H
36
4
F BN10P: C, 40.97; H, 7.28; N, 28.11. Found: C, 40.99; H, 7.30; N,
2
29
ACS Paragon Plus Environment