Acid-base equilibria in nonpolar media. 2. Self-consistent basicity scale in THF solution ranging from 2-methoxypyridine to EtP1(pyrr) phosphazene

Article abstract of DOI:10.1021/jo016185p

Relative ion-pair basicities ΔpK_ip of 25 substituted aryl and alkyl iminophosphoranes (phosphazenes) and 20 other N-bases (various pyridines, amines, amidines) have been measured in THF medium using the UV-Vis and/or ¹³C NMR methods. The ΔpK_ip values were corrected for ion pairing using the Fuoss equation to obtain relative ionic basicities ΔpK_α. Based on the measurements, a basicity scale ranging from 2-methoxypyridine to EtP₁(pyrr) and having a total span over 18 pK units has been created. The scale has been anchored to the pK_α value of triethylamine (pK_α = 12.5). The results are compared to pK_a values in various other solvents and in the gas phase. The pK_α values give better correlations than the pK_ip values, thus indirectly validating the procedure of correction for ion pairing. The predictability of the basicity together with suitable spectral properties in the UV range make the phenylphosphazenes convenient neutral indicators in the high basicity range where the choice of neutral indicators is very limited.

Full text of DOI:10.1021/jo016185p

J . Org. Chem. 2002, 67, 1873-1881
1873
Acid -Ba se Equ ilibr ia in Non p ola r Med ia . 2.1 Self-Con sisten t
Ba sicity Sca le in THF Solu tion Ra n gin g fr om 2-Meth oxyp yr id in e
to EtP 1(p yr r ) P h osp h a zen e
Toomas Rodima, Ivari Kaljurand, Aino Pihl, Vahur Ma¨emets, Ivo Leito,* and Ilmar A. Koppel
Institute of Chemical Physics, Department of Chemistry, University of Tartu, J akobi 2,
51014 Tartu, Estonia
leito@ut.ee
Received October 9, 2001
Relative ion-pair basicities ∆pKip of 25 substituted aryl and alkyl iminophosphoranes (phosphazenes)
and 20 other N-bases (various pyridines, amines, amidines) have been measured in THF medium
using the UV-Vis and/or 13C NMR methods. The ∆pKip values were corrected for ion pairing using
the Fuoss equation to obtain relative ionic basicities ∆pKR. Based on the measurements, a basicity
scale ranging from 2-methoxypyridine to EtP1(pyrr) and having a total span over 18 pK units has
been created. The scale has been anchored to the pKR value of triethylamine (pKR ) 12.5). The
results are compared to pKa values in various other solvents and in the gas phase. The pKR values
give better correlations than the pKip values, thus indirectly validating the procedure of correction
for ion pairing. The predictability of the basicity together with suitable spectral properties in the
UV range make the phenylphosphazenes convenient neutral indicators in the high basicity range
where the choice of neutral indicators is very limited.
In tr od u ction
pKip values for a number of other organic compounds
were established and the pKR (here and henceforth the
term “pKa of the base” is understood as the acidity of
conjugate acid of the base, and its definition is given in
eq 1) value 12.5 of triethylamine was suggested as the
secondary standard for anchoring acidity and basicity
scales in THF.10
Numerous acidity studies, mostly focused on CH-acids,
have been carried out in THF.2,3,4 Alkali metal amides
or carbanions with alkali metal counterions have mostly
been used as deprotonating agents. On the basis of these
measurements, ion-pair acidity scales have been devel-
oped relative to 9-phenylfluorene or fluorene.2,3,4 Those
scales have been anchored to the pKa values of these
reference compounds in aqueous sulfolane and DMSO,
respectively. This approach has also been used in con-
structing acidity scales in various other nonpolar medias
cyclohexylamine,5 benzene,6,7 dimethoxyethane,8 diethyl
ether,9 etc.
Contrary to acidity measurements, studies on basicity
in THF are scarce. Recently, the Morris group compiled
an acidity scale in THF based on NMR measurements
including numerous metal hydrides, phosphines, etc.10
The observed pKip values were corrected for ion-pairing
using the Fuoss equation.11 Besides the phosphines the
In nonpolar media the measured equilibrium constants
do not generally reflect the free ion acidity but rather
refer to ion pairs. An attempt to suppress the interactions
between cations and anions of the CH-acids in nonpolar
media was made by the Konovalov group.12,2 They have
used lithium [2.1.1]cryptate as the counterion for the
anions of CH-acids. In this cryptate the [2.1.1]cryptand
acts as a layer of solvent molecules separating the ions
and so eliminating the specific interactions. In a previous
work by some of us phosphazene t-BuP4(dma) (here and
henceforth “dma” denotes dimethylamino (N(CH3)2) and
“pyrr” denotes 1-pyrrolidinyl (N(CH2 CH2)2) radical) was
used as the deprotonating agent to create an acidity scale
in n-heptane.1 Protonated t-BuP4(dma) is a large cation
with a delocalized charge and has very weak interaction
with anions.
(1) Part
1 of this series: Leito, I.; Rodima, T.; Koppel, I. A.;
Schwesinger, R.; Vlasov, V. M. J . Org. Chem. 1997, 62, 8479-8483.
(2) Antipin, I. S.; Gareyev, R. F.; Vedernikov, A. N.; Konovalov, A.
I. J . Phys. Org. Chem. 1994, 7, 181-191 and references therein.
(3) Streitwieser, A., J r.; Ciula, J . C.; Krom, J . A.; Thiele, G. J . Org.
Chem. 1991, 56, 1074-1076.
(4) Kaufman, M. J .; Gronert S.; Streitwieser A. J . Am. Chem. Soc.
1988, 110, 2829-2835 and references therein.
(5) Streitwieser, A., J r.; J uaristi, E.; Nebenzahl, L. L. In Compre-
hensive Carbanion Chemistry. Part A. Structure and Reactivity; Buncel,
E.; Durst, T., Eds.; Elsevier: Amsterdam, 1980; pp 323-381.
(6) McEwen, W. K. J . Am. Chem. Soc. 1936, 58, 1124-1129.
(7) Antipin, I. S.; Vedernikov, A. N.; Konovalov, A. I. Zh. Org. Khim.
1986, 22, 446-447.
(8) Petrov, E. S.; Terekhova, M. I.; Lebedeva, T. I.; Basmanova, B.
M.; Shatenshtein, A. I. Zh. Obshch. Khim. 1978, 48, 616-623 and
references therein.
We have found earlier that aryl-substituted P1 phos-
phazenes ArNdP(R)3, where R ) dma or pyrr, are
suitable indicators together with amines to compile a
basicity scale in acetonitrile (AN).13,14
In this paper we report a basicity scale in THF medium
that incorporates aryl and alkyl P1 phosphazenes
(11) Fuoss, R. M. J . Am. Chem. Soc. 1958, 80, 5059-5061.
(12) Antipin, I. S.; Vedernikov, A. N.; Konovalov, A. I. Zh. Org. Khim.
1985, 21, 1355-1356.
(9) Conant, J . B.; Wheland, G. W. J . Am. Chem. Soc. 1932, 54, 1212-
1221.
(10) Abdur-Rashid, K.; Fong, T. P.; Greaves, B.; Gusev, D. G.;
Hinman, J . G.; Landau, S. E.; Lough, A. J .; Morris, R. H. J . Am. Chem.
Soc. 2000, 122, 9155-9171.
(13) Rodima, T.; Ma¨emets, V.; Koppel, I. J . Chem. Soc., Perkin Trans.
1 2000, 2637-2644.
(14) Kaljurand, I.; Rodima, T.; Leito, I.; Koppel, I. A.; Schwesinger,
R. J . Org. Chem. 2000, 65, 6202-6208.
10.1021/jo016185p CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/22/2002

1874 J . Org. Chem., Vol. 67, No. 6, 2002
Rodima et al.
+
-
A
HB
Sch em e 1
Kd
{
1
K
B2 + HB+1 A-
} B2 + HB+1 + A-
{\}
+
-
HB
A
1/Kd
{
2
HB+2 + B1 + A-
} HB+2 A- + B1 (7)
The Kd are the dissociation constants of the respective
ion pairs. The directly measured quantity is the relative
ion-pair basicity, ∆pKip, of bases B1 and B2. It is expressed
as follows:
∆pKip ) pKip(HB+2 A-) - pKip(HB+1 A-) )
HB1+A-
KKd
a(HB+2 A-)a(B1)
a(HB+1 A-)a(B2)
log
) log
(8)
HB2+A-
Kd
R′NdP(R)3 (see Scheme 1), some aryl P2 phosphazenes
R′NdP(R)2NdP(R)3, various substituted pyridines, and
several other bases.
For a better comparison with the THF data and as a
further extension of the earlier established basicity scale
in AN, some additional pKa values for several bases were
also measured in the latter solvent.
In a polar solvent S (water, AN, etc.) at low or moderate
concentrations the basicity of base B is defined using eq
1 and is expressed as dissociation constant Ka (eq 2) of
the respective conjugate acid HB+ of the base B or, more
commonly, its negative logarithm pKa
If the Kd values can be measured or estimated then
the pKR (an estimate of the pKa) can be found as follows:
∆pKR ) pKR(HB2+) - pKR(HB1+) )
HB1+A-
Kd
∆pKip - log
(9)
HB2+A-
Kd
Exp er im en ta l Section
Ch em ica ls. Most of the compounds were the same as used
earlier.13,14 The following compounds were of commercial origin
and were used without further purification: TMG (Aldrich,
99%), 1,8-bisdimethylaminonaphthalene (DMAN) (Aldrich,
>99%), tert-butylimino-tris(dimethylamino)phosphorane (t-
BuP1(dma)) base (Fluka, >98%), tert-butylimino-tris(pyrroli-
dino)phosphorane (t-BuP1(pyrr)) base (Fluka, g98%), and all
substituted anilines (Aldrich) used for synthesis of new phos-
phazenes. Aniline (Reakhim) was purified by refluxing with
acetone and following recrystallization of HCl salt.16 N,N-
Dimethylaniline (Reakhim) was purified by refluxing with
acetic anhydride and following recrystallization of HCl salt.16
o-Toluidine (Reakhim) was distilled twice, the second time
from CaH2. Pyrrolidine (Merck or Fluka, >98%) was distilled
from NaOH and was kept over NaOH. 2-Methoxypyridine
(Aldrich, 98%) was distilled fractionally from MgSO4 under
reduced pressure.
Ka
\
HB+ + S {
} B + HS+
(1)
(2)
a(B)a(HS+)
a(HB+)
Ka )
In media of low polarity (D e 15-20)15 there is an
extensive ion-pairing and formation of aggregates be-
tween ions and neutral molecules (homo- and heterocon-
jugation) that has to be considered. The extent of ion-
pairing depends on the solvent, the size of the ions, and
the charge distribution in ions. The general trend is that
small ions tend to form solvent-separated ion-pairs (SSIP)
(eq 3) while large ions with delocalized charge tend to
form contact ion-pairs (CIP) (eq 4).
Solutions of trifluoromethanesulfonic acid (TfOH) (Aldrich,
99+%) or methanesulfonic acid (MeSO3H) (Fluka, >99%) were
used as acidic titrants. Phosphazene bases EtP1(pyrr)17 or EtP2-
(dma) (Fluka, >98%) were used as basic titrants.
Solven ts. THF was used as purchased (Romil, >99.9%,
Super Purity Solvent, water content <0.005%) or purified
(REAKHIM, pure) as follows:16 THF was kept on KOH pellets
and then boiled 2-2.5 h on CaH2 in the flow of dried Ar. After
that, THF was distilled through the 60 cm long column filled
with steel rings, and the fraction with bp 66.2-66.5 °C was
collected. To this distillate was added LiAlH4, and then it was
boiled with same column for 1 h under Ar and then distilled
fractionally, fraction with bp 66.2-66.3 °C was collected. For
syntheses was used THF stored over KOH and distilled from
LiAlH4. CCl4 was distilled from P2O5, benzene from LiAlH4 (all
Reakhim).
We observed that ∼0.2 M TfOH polymerizes THF within a
few hours, and in dilute solutions the active concentration of
this acid was lower than analytical concentration. Also we
observed that with some batches of THF the absorbance below
280 nm decreased upon addition of TfOH.
HB+ + A- a HB+s‚A-
HB+ + A- a [HBA]s
(3)
(4)
s
In media of low polarity the homo- and heteroconju-
gation processes (eqs 5 and 6, respectively) occur to a
much lower extent as compared to ion-pairing. At a low
concentration these conjugation processes between ions
and neutrals can be neglected because most of the ions
present in the solution are ion-paired.
HB+ + B a BHB+
(5)
(6)
+
HB+ + B1 a BHB1
AN (Romil, >99.9%, Super Purity Solvent (far UV), water
content <0.005%) was the same used in previous work14 and
was used without further purification.
To exclude the necessity for measuring the hydrogen
ion activity (see eq 2), the equilibrium between two (ion-
paired) bases B1 and B2 was studied:
(16) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon: Sydney, 1989.
(17) Schwesinger, R.; Willaredt, J .; Schlemper, H.; Keller, M.;
Schmitt, D.; Fritz, H. Chem. Ber. 1994, 127, 2435-2454 and references
therein.
(15) Reichardt, C., Solvents and Solvent Effects in Organic Chem-
istry, 2nd ed.; VCH: Weinheim, 1988.

Acid-Base Equilibria in Nonpolar Media
J . Org. Chem., Vol. 67, No. 6, 2002 1875
Gen er a l Meth od for th e Syn th esis of th e P 1 P h osp h a -
zen es. New phenyl-substituted P1 phosphazenes (s-PhP1(pyrr))
were synthesized by the Kirsanov reaction18 according to the
scheme:
122.4 (br m), 125.0 (d, J C-P ) 9.2), 125.7 (q, J C-F ) 270.2),
128.2 (m), 144.9 (d, J C-P ) 26.3), 150.0 (d, J C-P ) 7.8).
2,6-Cl2-4-NO2-C6H 2P 1(p yr r ). According to the general
method, 4.2 g (yield 63,2%, mp 95.3-96.9 °C, recrystallized
from 70% EtNH2 aqueous solution or from hexane) of yellow
crystals was obtained. Anal. Calcd for C18H26Cl2N5O2P: C,
48.77%; H, 5.91%; N, 15.8%. Found: C, 48.37%; H, 5.87%; N,
PyrrH
s-PhNH2 + PCl5
8 s-PhNdPCl3 Bz or THF8 s-PhP1(pyrr)
CCl4
1
(10)
15.80%. H NMR (200 MHz, CDCl3) δ 1.82 (m, 12H), 3.21 (dt,
12H, J H-H ) 6.6, J P-H ) 3.4), 8.13 (d, J P-H ) 1.2). 13C NMR
(50 MHz, CDCl3) δ 26.4 (d, J C-P ) 8.6), 47.0 (d, J C-P ) 5.0),
124.0, 128.2 (d, J C-P ) 9.0), 136.1, 151.4 (d, J C-P ) 11.7).
2,6-(NO2)2-C6H3P 1(p yr r ). According to the general method,
3.6 g (yield 57%, mp 132.5-132.8 °C, recrystallized from 3:1
mixture MeOH-CHCl3) of orange crystals was obtained. Anal.
Calcd for C18H27N6O4P: C, 51.17%; H, 6.44%; N, 19.89%.
An amount of 20 mmol of the corresponding aniline was
dissolved in 20 mL of CCl4, and 20 mmol of PCl5 was added
(in some cases it was necessary to cool the mixture in the
beginning of the reaction). The mixture was warmed and
refluxed until the evolution of HCl finished. Insoluble in CCl4,
trichlorophosphazenes were filtered off and washed succes-
sively with CCl4, benzene, and diethyl ether. When the
trichlorophosphazene was soluble in CCl4, the solvent was
removed on a rotavapor, and the residue was dried in vacuo.
An amount of 15 mmol of obtained phenylimino trichloro-
phosphazene was dissolved in 40 mL of dried THF (in the case
of 2-NO2-4-CF3 and 5-Cl-2-NO2 substituents, 50 mL of dry
benzene was used instead), and 90 mmol of pyrrolidine
(solution in 15 mL of THF) was added by means of a dropping
funnel. The mixture was heated to 50 °C and stirred ca. 1 h.
Then the mixture was cooled to +5 °C, the HCl salt of
pyrrolidine was filtered off, and the solvent was removed at
reduced pressure (60 °C/5 Torr). The brown residue was
washed with 70% aqueous solution of EtNH2. The crystals of
the product were collected, washed with 40% aqueous EtNH2
solution, dried in vacuo, and recrystallized (2-NO2-4-CF3-
C6H3P1(pyrr) was isolated as HBF4 salt).
1
Found: C, 50.77%; H, 6.36%; N, 19.89%. H NMR (200 MHz,
CD3Cl) δ 1.81 (m, 12H), 3.09 (dt, 12H, J H-H ) 6.6, J P-H ) 3.0),
6.59 (dt, 1H, J H-H ) 8.0, J P-H ) 1.3), 7.56 (dd, 2H, J H-H ) 8.0,
J P-H ) 1.2). 13C NMR (50 MHz, CD3Cl) δ 26.4 (d, J C-P ) 8.7),
46.6 (d, J C-P ) 5.1), 113.4, 126.6, 137.6 (d, J C-P ) 12.8), 148.1
(d, J C-P ) 7.7).
2,5-Cl2-C6H3P 1(p yr r ).13 1H NMR (200 MHz, THF) δ 1.8
(overlapped by solvent, 12H), 3.05 (dt, 12H, J H-H ) 6.7,
J P-H ) 4.3), 6.37 (ddd, 1H, J H-H ) 8.4, 2.5, J P-H ) 0.5), 6.68
(dd, 1H, J H-H ) 2.5, J P-H ) 1.2), 7.05 (dd, 1H, J H-H ) 8.4,
J P-H ) 2.5). 13C NMR (50 MHz, THF) δ 27.0 (d, J C-P ) 7.5),
47.6 (d, J C-P ) 3.7), 115.9, 122.3 (d, J C-P ) 8.3), 127.3 (d,
J C-P ) 26.4), 130.1 (d, J C-P ) 1.5), 132.2, 150.5 (d, J C-P ) 5.8).
Syn t h esis of HBP h 4 Sa lt s of t h e P 2 P h osp h a zen es.
2-Cl-C6H4P 2(p yr r )‚HBP h 4. In three-necked flask, equipped
with magnetic stirrer, powder adding system, and rubber gas
balloon with Ar, of 15 mL (140 mmol) of 2-chloroaniline was
added 0.28 g (12 mmol) of NaH. The mixture was heated slowly
to 60 °C. After the evolution of H2 was ceased, the mixture
was stirred and cooled to room temperature, and 2.6 g (4.7
mmol) Cl-P(pyrr)2dN-P+(pyrr)3‚BF4- was added gradually by
means of powder adding system.19 The mixture was heated
again to 60 °C, held at that for 6 h, and then left for 2 days.
The excess of chloroaniline was removed under reduced
pressure (110 °C/6 mmHg). To the residue dissolved in CH2-
Cl2 was added some water, and the CH2Cl2 extract of product
HBF4 salt was washed with acidified water to remove the rest
of chloroaniline. The CH2Cl2 was removed in vacuo, the
residue, 0.8 g of dark oil, was dissolved in 10 mL MeOH,
treated with charcoal, and product was precipitated as HBPh4
salt by adding 0.43 g of NaBPh4 solution in 1.5 mL of MeOH.
It was filtered and recrystallized from a 1:1 mixture of EtOH
and AN. Yield 0.2 g (18%, mp 155.0-156.1 °C) of nearly
colorless crystals. Anal. Calcd for C50H65BClN7P2: C, 68.84%;
H, 7.51%; N, 11.24%. Found: C, 68.82%; H, 7.71%; N, 11.20%.
1H NMR (200 MHz, THF) δ 1.8 (overlapped by solvent, 20H),
2.95 (dt, 12H, J H-H ) 6.6, J P-H ) 3.6), 3.22 (m, 8H), 6.34
(br d, 1H, J P-H ) 11.4), 6.69 (t, 4H, J H-H ) 7.2), 6.84 (t, 8H,
J H-H(av) ) 7.4), 7.0 (m, 2H), 7.19 (d, 1H, J H-H ) 4.5), 7.26 (m,
8H), 7.39 (d, 1H, J H-H ) 4.5). 13C NMR (50 MHz, THF) δ 26.9
(d, J C-P ) 9.0), 27.1 (d, J C-P ) 9.4), 47.5 (d, J C-P ) 5.2), 47.9
(d, J C-P ) 5.7), 121.7, 121.8 (d, J C-P ) 2.9), 124.8, 125.6 (m,
J C-B ) 2.9), 128.5, 130.5, 130.6 (d, J C-P ) 5.0) 137.2 (m,
J C-B ) 1.5), 137.5 (d, J C-P ) 2.7), 165.3 (m, J C-B ) 49.9).
P h P 2(d m a )‚HBP h 4. The same procedure was used as for
2-Cl-C6H4P2(pyrr)‚HBPh4. To 15 mL of aniline (distilled from
Zn-dust) was added 15.2 mmol of NaH, and the mixture was
heated for a short time. At room temperature 7.6 mmol (5.2
4-CF 3-C6H4P 1(p yr r ). According to the general method, after
recrystallization from 70% EtNH2 aq soln, 1.9 g of colorless
crystals was obtained, yield 32%, mp 102.4-103.3 °C. Anal.
Calcd for C19H28F3N4P: C, 56.99%; H, 7.07%; N 13.99%.
1
Found: C, 57.18%; H, 7.12%; N, 14.01%. H NMR (200 MHz,
CD3Cl) δ 1.83 (m, 12H), 3.19 (dt, 12H, J H-H ) 6.6, J P-H ) 3.4),
6.78 (d, 2H, J H-H ) 8.5), 7.27 (d, 2H, J H-H ) 8.5). 13C NMR
(50 MHz, CD3Cl) δ 26.4 (d, J C-P ) 8.0), 46.9 (d, J C-P ) 4.0),
117.3 (q, J C-F ) 32.3), 122.2 (d, J C-P ) 18.2), 125.72 (d,
J C-P ) 2.7), 125.74 (q, J C-F ) 270.3), 155.7.
5-Cl-2-NO2-C6H3P 1(p yr r ). According to the general method,
after recrystallization from 70% EtNH2 aq soln, 4.5 g of
yellowish crystals was obtained, yield 73%, mp 82.7-83.2 °C.
Anal. Calcd for C18H27ClN5O2P: C, 52.49%; H, 6.61%; N,
17.0%. Found: C, 52.47%; H, 6.59%; N, 16.80%. 1H NMR (200
MHz, THF) δ 1.8 (overlapped by solvent, 12H), 3.19 (dt, 12H,
J H-H ) 6.7, J P-H ) 4.2), 6.42 (dd, 1H, J H-H ) 8.6, 2.2), 6.77
(dd, 1H, J H-H ) 2.2, J P-H ) 1.0), 7.42 (dd, 1H, J H-H ) 8.6,
J P-H ) 2.4). 13C NMR (50 MHz, THF) δ 27.0 (d, J C-P ) 7.7),
47.6 (d, J C-P ) 3.7), 114.34, 124.2 (d, J C-P ) 9.4), 126.0, 137.4,
144.2 (d, J C-P ) 24.3), 147.9 (d, J C-P ) 7.4).
2-NO2-4-CF 3-C6H3P 1(p yr r ). The raw product (4.9 g) syn-
thesized according to the general method was dissolved in 40
mL of 5% aq HCl, and the solution of 1.3 g of NaBF4 in water
was added. The precipitate was filtered, dried, and recrystal-
lized from ethyl acetate. A 3.3 g (6.2 mmol) amount of light
yellow crystals of the HBF4 salt of the phosphazene (mp 186-
187 °C) was dissolved in the mixture of 8 mL MeOH and 12
mL AN. A 6.2 mmol amount of MeOK as 25% solution in
MeOH was added. The solvent was removed, and the residue
was refluxed with hexane for 0.5 h. The mixture was filtered,
and the hexane was removed under the reduced pressure. The
brownish solid residue was recrystallized from 70% EtNH2 aq
soln to give 2.0 g (yield 72.5%, mp 70.5-71.2 °C) of light yellow
crystals of the product. Anal. Calcd for C19H21F3N5O2P: C,
51.23%; H, 6.11%; N, 15.72%. Found: C, 51.42%; H, 6.12%;
N, 15.66%. 1H NMR (200 MHz, THF) δ 1.8 (overlapped by
solvent, 12H), 3.20 (dt, 12H, J H-H ) 6.7, J P-H ) 4.2), 6.90 (d,
1H, J H-H ) 8.8), 7.29 (dd, 1H, J H-H ) 8.8, 2.5, J P-H ) 0.6),
7.72 (br m, 1H, J H-H ) 2.5). 13C NMR (50 MHz, THF) δ 27.1
(d, J C-P ) 7.7), 47.6 (d, J C-P ) 4.0), 115.5 (q, J C-F ) 33.8),
g) of ClP(dma)2dN-P+(dma)3‚BPh4 (made from BF4 salt19
)
-
-
was added. The excess of aniline was washed out with acidified
water and the residue extracted with warm 70% EtNH2 aq
soln. By diluting this extract with water, a brownish precipi-
tate of desired (raw) product as a HBPh4 salt was collected.
The recrystallization from 70% EtNH2 aq soln and finally from
(19) Schwesinger, R.; Schlemper, H.; Hasenfratz, C.; Willaredt, J .;
Dambacher, T.; Breuer, T.; Ottaway, C.; Fletschinger, M.; Boele, J .;
Fritz, H.; Putzas, D.; Rotter, H. W.; Bordwell, F. G.; Satish, A. V.; J i
G.-Z.; Peters, E.-M.; Peters, K.; von Schnering, H. G.; Walz, L. Liebigs
Ann. 1996, 1055-1081 and references therein.
(18) Zhmurova, I. N.; Kirsanov, A. U. J . Gen. Chem. USSR (Engl.
Transl.) 1960, 30, 3018.

1876 J . Org. Chem., Vol. 67, No. 6, 2002
Rodima et al.
ethyl acetate gave 1.0 g of crystals with gentle violet tint (yield
δ 37.3 (d, J C-P ) 4.4), 38.1 (d, J C-P ) 2.6), 114.3, 123.2 (d,
J C-P ) 20.1), 128.4 (d, J C-P ) 1.7), 154.9.
18.8%, mp 135.2-136.3 °C). Anal. Calcd for C40H56BN7P2: C,
67.89%; H, 7.99%; N, 13.85%. Found: C, 67.30%; H, 8.07%;
P h P 2(p yr r ): colorless needles. 1H NMR (200 MHz, THF)
δ 1.8 (overlapped by solvent, 20H), 3.12 (dt, 20H, J H-H ) 6.7,
J P-H ) 4.3), 6.58 (m, 2H, J H-H ) 8.3), 6.78 (m, 2H). 13C NMR
(50 MHz, THF) δ 27.1 (d, J C-P ) 8.1), 27.3 (d, J C-P ) 8.4),
47.4 (d, J C-P ) 4.7), 47.9 (d, J C-P ) 3.8), 113.9, 122.9 (d,
J C-P ) 20.8), 128.3 (d, J C-P ) 1.9), 155.3.
1
N, 13.87%. H NMR (200 MHz, THF) δ 2.41 (d, 18H, J P-H
)
10.2), 2.63 (d, 12H, J P-H ) 10.8), 6.70 (t, 4H, J H-H ) 7.2), 6.85
(t, 8H, J H-H(av) ) 7.4), 6.9-7.0 (m, 3H), 7.2 (m, 2H), 7.28 (m,
8H). 13C NMR (50 MHz, THF) δ 36.9 (d, J C-P ) 5.2), 119.9 (d,
J C-P ) 7.1), 121.7, 123.2, 125.6 (m, J C-B ) 2.8), 130.0, 137.2,
140.8, 165.2 (m, J C-B ) 49.5).
Syn th esis of HBP h 4 Salts of P1 Phosphazenes and Amines.
A methanolic solution of bases (10 mmol in 10 mL MeOH) was
acidified with 15% HCl aq soln, and a slight excess of NaBPh4
solution in small quantity of MeOH was added. Precipitate of
the salt was filtered, washed several times with MeOH,
recrystallized (except TBD) from a 4:1 mixture of MeOH and
CHCl3, and dried in vacuo.
P h P 2(p yr r )‚HBP h 4. The same procedure was used as for
2-Cl-C6H4P2(pyrr)‚HBPh4, except that NaH was not used. To
5 mL of aniline 5 mmol (2.4 g) of ClP(pyrr)2dN-P+(pyrr)3‚
BF4- was added,19 and the mixture was stirred and heated at
150 °C for 20 h. Then aniline was removed under the reduced
pressure. The residue was extracted with 10 mL of CH2Cl2,
and the extract was washed twice with water and acidified
water, respectively. The extract was concentrated, and a dark
sticky residue (1.9 g) was dissolved in 8 mL of MeOH. The
solution was filtered, and a solution of 1.22 g (3.5 mmol) of
NaBPh4 in 3.5 mL MeOH was added. Collected violet crystals
of product were recrystallized from 70% EtNH2 aq soln as
described in ref 19 or from 4:1 mixture of MeOH/CHCl3. A 1.8
g (yield 44%, mp 179.2-180.2 °C) amount of crystals with beige
tint was obtained. Anal. Calcd for C50H66BN7P2: C, 71.67%;
H, 7.94%; N, 11.70%. Found: C, 71.64%; H, 7.90%; N, 11.64%.
1H NMR (200 MHz, THF) δ 1.8 (overlapped by solvent, 20H),
1
Et3N‚HBP h 4: mp 184-187 °C (dec); H NMR (200 MHz,
THF) δ 0.88 (t, 3H, J H-H ) 7.4) 2.53 (q, 2H, J H-H ) 7.4), 6.75
(t, 4H, J H-H ) 7.2), 6.89 (t, 8H, J H-H(av) ) 7.4), 7.31 (m, 8H).
13C NMR (50 MHz, THF) δ 9.3, 47.8, 122.0, 125.9 (m, J C-P
)
2.7), 137.1, 165.1 (m, J C-P ) 49.5).
1
2-Cl-C6H4P 1(d m a )‚HBP h 4: mp 162.0-163.3 °C. H NMR
(200 MHz, THF) δ 2.47 (d, 18H, J P-H ) 10.2), 6.72 (t, 4H,
J H-H ) 7.2), 6.86 (t, 8H, J H-H(av) ) 7.6), 7.10 (m, 1H), 7.2 (m,
2H) 7.28 (m, 8H), 7.47 (m, 1H). 13C NMR (50 MHz, THF) δ
37.3 (d, J C-P ) 4.4), 121.9, 125.7 (m, J C-B ) 2.8), 128.7 (d,
J C-P ) 2.0), 129.2 (d, J C-P ) 5.7), 131.3, 132.1 (d, J C-P ) 6.6),
134.3, 137.2, 165.1 (m, J C-B ) 49.5).
2.99 (dt, 12H, J H-H ) 6.6, J P-H ) 3.6), 6.68 (t, 4H, J H-H
)
2,5-Cl2-C6H3P 1(p yr r )‚HBP h 4: mp 153.4-153.6 °C. 1H
NMR (200 MHz, THF) δ 1.8 (overlapped by solvent, 12H), 3.06
(dt, 12H, J H-H ) 6.7, J P-H ) 3.7), 6.71 (t, 4H, J H-H ) 7.2),
7.2), 6.83 (t, 8H, J H-H(av) ) 7.5), 6.9-7.0 (m, 3H), 7.19 (d, 2H,
J H-H ) 7.6), 7.25 (m, 8H). 13C NMR (50 MHz, THF) δ 27.0 (d,
J C-P ) 8.6), 27.1 (d, J C-P ) 9.2), 47.5 (d, J C-P ) 5.2), 47.7 (d,
J C-P ) 5.8) 119.8 (d, J C-P ) 7.2), 121.6, 123.0, 125.6 (m,
J C-B ) 2.8), 129.9, 137.2 (br m), 141.2, 165.3 (m, J C-B ) 49.5).
6.86 (t, 8H, J H-H(av) ) 7.4), 7.13 (dd, 1H, J H-H ) 2.4, J P-H
)
1.1), 7.22 (ddd, 1H, J H-H ) 8.6, 2.4, J P-H ) 0.9), 7.26 (m, 8H),
7.44 (dd, 1H, J H-H ) 8.6, J P-H ) 1.2). 13C NMR (50 MHz, THF)
δ 26.8 (d, J C-P ) 8.5), 48.4 (d, J C-P ) 4.7), 121.8, 125.6 (m,
J C-B ) 2.9), 125.7 (overlapped by anions peak), 128.0, 128.7
(d, J C-P ) 8.0), 132.4, 134.1 (d, J C-P ) 1.7), 136.2, 137.1 (m,
J C-B ) 1.4), 165.2 (m, J C-B ) 49.5).
4-MeO-C6H4P 2(p yr r )‚HBP h 4. 5 mmol (0.61 g) of anisidine,
5 mmol (2.75 g) of Cl-P(pyrr)2dN-P+(pyrr)3‚BF4- and 5 mmol
(0.5 g) of Et3N were mixed in a 20 mL solution of THF/AN
(1:2) at room temperature. The mixture was refluxed for 20
h. The solvent was removed under reduced pressure, and the
residue, 2.2 g of viscous oil, was dissolved in 10 mL of AN. A
1.7 g amount of NaBPh4 in 5 mL of MeOH was added. An oily
substance separated. The solution was decanted from the oily
substance, and the substance was twice extracted with 40 mL
of hot MeOH. The crystals collected from the methanolic
solutions were recrystallized from 1:4:9 mixture of water-AN-
MeOH. Yield 0.7 g (16%, mp 165.0-166.4 °C) of colorless
crystals. 1H NMR (200 MHz, THF) δ 1.8 (overlapped by
solvent, 20H), 2.98 (dt, 12H, J H-H ) 6.5, J P-H ) 3.6), 3.7
(overlapped by solvent, 3H), 6.54 (d, 4H, J H-H ) 12.0), 6.68 (t,
4H, J H-H ) 7.2), 6.8 (br m, 12H,), 7.25 (m, 8H). 13C NMR (50
MHz, THF) δ 27.0 (d, J C-P ) 8.6), 27.1 (d, J C-P ) 9.2), 47.5 (d,
J C-P ) 5.0), 47.7 (d, J C-P ) 5.7), 55.7, 115.2, 121.7, 122.0 (d,
J C-P ) 6.7), 125.6 (m, J C-B ) 2.8), 133.6, 137.2, 156.8, 165.3
(m, J C-B ) 49.5).
Liber a tion of P 2 P h osp h a zen e Ba ses fr om Th eir HB-
P h 4 Sa lts w ith KOMe. The corresponding HBPh4 salt was
dissolved in a possibly small amount of dried MeOH, and a
calculated (with light excess) amount of 25% KOMe solution
in MeOH was added. The precipitated KBPh4 was filtered off
in a glovebox, and MeOH was removed under reduced pres-
sure. The residue was extracted with hexane, the extract was
filtered, and hexane was removed in vacuo. The free bases
were used for spectrometric measurements.
t-Bu P 1(d m a )‚HBP h 4: mp 266-267 °C (dec). 1H NMR (200
MHz, THF) δ 1.22 (d, 9H, J P-H ) 1.0), 2.50 (d, 18H, J P-H
)
10.0), 4.5 (br d, 1H, J P-H ) 9.7), 6.71 (t, 4H, J H-H ) 7.2), 6.86
(t, 8H, J H-H(av) ) 7.4), 7.28 (m, 8H). 13C NMR (50 MHz, THF)
δ 31.5 (d, J C-P ) 4.6), 37.8 (d, J C-P ) 4.7), 53.2, 121.8, 125.7
(m, J C-B ) 2.8), 137.2, 165.2 (m, J C-B ) 49.5).
TBD‚H BP h 4 (1,5,7-t r ia za b icyclo[4.4.0]d ec-5-en ylt et -
r a p h en ylbor a te): mp 240-241 °C (dec); 1H NMR (200 MHz,
DMSO-d6) δ 1.81 (q, 4H, J H-H(av) ) 5.9), 3.13 (t, 4H, J H-H
)
5.8), 3.18 (t, 4H, J H-H ) 6.0). 13C NMR (50 MHz, DMSO-d6) δ
20.2, 37.5, 46.2, 121.5, 125.3 (m, J C-B ) 2.7), 135.5 (m, J C-B
1.2), 150.6, 163.3 (m, J C-B ) 49.5).
)
t-Bu P 1(p yr r )‚HBP h 4: mp 233-235 °C. 1H NMR (200 MHz,
THF) δ 1.24 (d, 9H, J P-H ) 0.8), δ 1.8 (overlapped by solvent,
12H), 3.05 (dt, 12H, J H-H ) 6.7, J P-H ) 3.7), 6.71 (t, 4H,
J H-H ) 7.2), 6.85 (t, 8H, J H-H(av) ) 7.4), 7.26 (m, 8H).13C NMR
(50 MHz, THF) δ 26.8 (d, J C-P ) 8.2), 31.7 (d, J C-P ) 4.5),
48.4 (d, J C-P ) 4.7), 53.2 (d, J C-P ) 0.8) 121.8, 125.7 (m,
J C-B ) 2.8), 137.2, 165.2 (m, J C-B ) 49.5).
H2NP 1(p yr r )‚HBP h 4: mp 183.0-184.3 °C. 1H NMR (200
MHz, THF) δ 1.8 (overlapped by solvent, 12H), 3.05 (dt, 12H,
J H-H ) 6.6, J P-H ) 3.8), 3.35 (d, 2H, J P-H ) 10.4), 5.50 (d, 1H,
J P-H ) 35.6), 6.73 (t, 4H, J H-H ) 7.2), 6.87 (t, 8H, J H-H ) 7.2),
7.28 (m, 8H). 13C NMR (50 MHz, THF) δ 26.9 (d, J C-P ) 8.1),
47.9 (d, J C-P ) 4.0), 121.9, 125.8 (m, J C-B ) 2.8), 137.1, 165.2
(m, J C-B ) 49.7).
2-Cl-C6H4P 2(p yr r ): colorless crystals. 1H NMR (200 MHz,
THF) δ 1.8 (overlapped by solvent, 20H), 3.12 (dt, 12H,
1
EtP 1(p yr r )‚HBP h 4: H NMR (200 MHz, THF) δ 1.08 (dt,
J H-H ) 6.6, J P-H ) 4.2), 3.23 (m, 8H), 6.17 (ddd, 1H, J H-H
)
3H, J H-H ) 7.4, J P-H ) 1.2), δ 1.8 (overlapped by solvent, 12H),
7.7, 6.8, 1.9), 6.71 (ddd, 1H, J H-H ) 8.1, 6.8, 1.7), 6.79 (ddd,
1H, J H-H ) 8.1, 1.9, J P-H ) 1.2), 7.01 (ddd, 1H, J H-H ) 7.7,
2.82 (dq, 2H, J H-H ) 7.4, J P-H ) 9.0), 3.03 (dt, 12H, J H-H
)
6.7, J P-H ) 3.8), 6.71 (t, 4H, J H-H ) 7.2), 6.86 (t, 8H,
J H-H(av) ) 7.4), 7.26 (m, 8H). 13C NMR (50 MHz, THF) δ 17.3
(d, J C-P ) 7.1), 26.9 (d, J C-P ) 8.0), 36.8, 48.0 (d, J C-P ) 4.6),
121.8, 125.7 (m, J C-B ) 2.8), 137.2, 165.2 (m, J C-B ) 49.5).
UV-Vis Sp ectr op h otom etr ic Deter m in a tion of p Ka in
AN a n d p Kip in THF . The spectrophotometric titration
method in THF and in AN media in the glovebox was similar
to that used in the previous work.14 Perkin-Elmer Lambda 2S
or Lambda 40 spectrophotometer equipped with quartz fiber
1.7, J P-H ) 2.6). 13C NMR (50 MHz, THF) δ 27.1 (d, J C-P
8.2), 27.4 (d, J C-P ) 8.4), 47.4 (d, J C-P ) 4.6), 47.9 (d, J C-P
)
)
4.0), 113.8, 122.1 (d, J C-P ) 11.9), 126.5, 127.9 (d, J C-P ) 30.7),
129.4 (d, J C-P ) 2.7), 151.3 (d, J C-P ) 5.1).
1
P h P 2(d m a ): colorless crystals. H NMR (200 MHz, THF)
δ 2.55 (d, 18H, J P-H ) 10.1), 2.65 (d, 12H, J P-H ) 10.0), 6.27
(ddt, 1H, J H-H ) 7.1, 1.5, J P-H ) 1.2), 6.60 (br d, 2H,, J H-H
)
7.9), 6.82 (br t, 2H,, J H-H(av) ) 7.5). 13C NMR (50 MHz, THF)

Acid-Base Equilibria in Nonpolar Media
J . Org. Chem., Vol. 67, No. 6, 2002 1877
Ta ble 1. Ion ic Ra d ii Used for Cor r ection for Ion P a ir in g
ion
ion-pair radius, Åa
ion
ion-pair radius, Åa
RNH+dPR′3; RNH+dPR′′3
RNH+dPR′2NdPR′3
RNH+dPR′′2NdPR′′3
TBDH+
4
pyrrolidineH+
2
3.2
2
2.2
2.2
2.5
4.4
4.8
5.4
3
2.5
2.2
2.7
DMANH+
X-anilineH+; X-pyridineH+
N,N-Me2-anilineH+
2,6-X2-pyridineH+
DBUH+; DABH+b
TMGH+; TEAH+
PhTMGH+
CF3SO3-; CH3SO3
-
-
BPh4
a
Ionic radii from literature (ref 10) were used when available. In cases when no literature data were available, the radii were estimated
b
by PM3 calculations; R ) alkyl or aryl; R′ ) dma; R′′ ) pyrr; X ) H or substituent. DAB ) 1,4-diaminobutane.
optic system, and an external sample compartment positioned
into the glovebox was used. Two different gloveboxes were
used. Earlier measurements were carried out in a Mecaplex
glovebox in the atmosphere of nitrogen, continuously purified
from water vapor, volatile acidic and basic impurities with
molecular sieves, powdered P2O5, and KOH pellets. Later
measurements were carried out in MBraun glovebox in the
atmosphere of argon that was constantly circulated through
a purification system containing molecular sieves and acti-
vated copper for removal of water vapor and oxygen, respec-
tively. The residual concentrations of water and oxygen in the
atmosphere of the glovebox during the measurements were
constantly monitored and were generally below 1 ppm.
All solutions were prepared in glovebox and were made fresh
daily. The concentrations of acidic titrants were from 1 to 3
mM, and basic titrants were from 0.5 to 5.5 mM. Higher
concentrations (up to 0.36 and 0.025 M for methanesulfonic
acid and EtP1(pyrr), respectively) were made for studying
relative ion-pair basicity dependence from concentrations in
THF. The concentrations of studied bases were generally in
mM range; higher concentrations (up to 0.23 M) were used
for studying relative ion-pair basicity dependence from con-
centrations.
of a regression line. From the values of [B1] and [B2], the
calculation of ∆pKa of the bases is straightforward. In many
cases (for example, when the bases have absorption maxima
in different wavelength ranges) it was possible to use various
simpler calculation procedures (see refs 1 and 20). The mixture
of bases as well as both bases separately was titrated with an
optically transparent acid and/or base, and the data for ∆pKa
calculations was obtained from UV-Vis spectra. From each
titration experiment, the ∆pKa was determined as the mean
of 5-20 values.
In THF the general principle of the ∆pK calculations is the
same. As UV-Vis spectrophotometry does not make any
difference between free ions, solvent separated ion-pairs, and
loosely bonded contact ion-pairs (these last two are the main
forms of monocharged ions at concentration below 0.01 M in
THF),10 we get ∆pKips instead of ∆pKas. The ∆pKip values given
in Table 2 are a mean of 5-23 measurements. The correction
for ion pairing was calculated using the Fuoss equation as
described in ref 10, and a ∆pKR is then obtained. The ionic
radii that were used are given in Table 1.
In some cases (“invisible bases”, e.g., aliphatic amines
(pyrrolidine and triethylamine) and t-BuP1(dma), also DBU
and TMG versus “visible” aromatic bases) the calculations have
been carried out on a molar basis. The solution containing a
mixture of known amounts (in moles) of “invisible” and visible
base was titrated with titrant of known concentration. From
the added titrant mass and its concentration, the amount (in
moles) of titrant in the cell was found. Combining the spectra
of solutions containing both bases fully deprotonated, fully
protonated, and the mixture of protonated and deprotonated
forms, we calculated the indicator ratio of the visible base, and
knowing the amounts of the visible base and titrant added,
we calculated the indicator ratio for “invisible” base. The ∆pKip
calculation is then straightforward. The agreement between
relative ion-pair basicities obtained with different calculation
methods was satisfactory.
The phosphazene bases are more stable and convenient to
-
handle if they are used as salts. Several counterions (BPh4
,
PF6-, ClO4-) were used in the salts. From these, only salts
with tetraphenylborate anion were soluble enough in THF.
Various disturbances were frequently observed with the
tetraphenylborate anion both in UV-Vis (strange behavior of
the spectra upon titration and the nonreversibility of the
spectra upon back-titration, change of spectrum in time) and
NMR measurements (change of spectrum in time, appearance
of alien peaks). This could be due to the somewhat unstable
nature of this anion. Therefore, we avoided this anion as
counterion in UV-Vis spectrophotometric measurements and
used free bases instead.
NMR p Kip Deter m in a tion . The standard 1D 1H and
proton-decoupled 13C NMR spectra were recorded on a Bruker
AC-200 NMR spectrometer at 200 and 50 MHz, correspond-
ingly. Solutions (∼ 0.1 M) were prepared and sealed off in 5
mm NMR tubes. Chemical shifts were determined relative to
TMS as an internal standard.
Ca lcu la tion Meth od s for UV-Vis Sp ectr op h otom etr ic
Mea su r em en ts. The ∆pKa calculation methods in AN are
similar to those of the previous works.1,20 The essence of the
general calculation method is the following. When two partially
protonated bases B1 and B2 are in the same solution, then the
following equation holds for absorbance A at wavelength λ (1
cm path length):
The NMR spectra of phosphazenes and the method used for
determination of their ∆pKip in THF were analogous to the
corresponding NMR spectra and ∆pKa calculations applied for
phosphazenes in the AN.13 There is no large difference of the
13C and 1H chemical shifts for phosphazenes in THF as
compared to the corresponding shifts in AN. Usually, the
differences were below 1 ppm for 13C and 0.1 ppm for 1H. The
∆pKips of phosphazenes were determined using approximately
equimolar mixture of phosphazene and indicator in THF. As
it was in the case of AN solutions, there is the fast (in NMR
time scale) exchange between phosphazene and indicator base
and acid forms leading to the coalescence of NMR lines in the
13C and 1H spectra. Correspondingly, the chemical shifts of
these forms were determined separately from the single
component THF solutions of these species. The ∆pKip values
were calculated as it was done previously with ∆pKa values.13
The indicator ratios for eq 8 were calculated (eqs 13 and 14)
from chemical shifts of the individual species (both neutral
and protonated forms of the bases) and their averaged values
in the mixtures containing both forms.
Aλ ) [HB1+]ꢀHλ B + + [B ]ꢀλ + [HB2+]ꢀHλ B + + [B ]ꢀλ (11)
1
B1
2
B2
1
2
The molar absorptivities ꢀ can be found separately from the
spectra of the free bases and fully protonated bases. If we use
concentrations, that are normalized to 1 then we may write:
[HB1+] ) 1 - [B1] and [HB2+] ) 1 - [B2]. After a mathematical
transformation of eq 11 we get
Aλ - ꢀλHB + - ꢀλ
(ꢀλB - ꢀλHB
)
)
+
+ ) [B1] ‚
+ [B2] (12)
HB2
1
1
1
2
(ꢀλB - ꢀλHB
)
(ꢀλB - ꢀλHB
+
+
2
2
2
If the spectra are recorded over a range of wavelengths then
[B1] and [B2] can be found from eq 12 as the slope and intercept
(20) Leito, I.; Kaljurand, I.; Koppel, I. A.; Yagupolskii, L. M.; Vlasov,
V. M. J . Org. Chem. 1998, 63, 7868-7874.

1878 J . Org. Chem., Vol. 67, No. 6, 2002
Ta ble 2. Con tin u ou s Self-con sisten t Ba sicity Sca le of Neu tr a l Ba ses in THF Solu tion a
Rodima et al.
a
The numbers on the arrows are the direct experimental ∆pKip values (uncorrected for ion pairing) obtained from UV-Vis
spectrophotometric measurements if not indicated otherwise. NMR measurements. c Absolute pKip(THF) and pKR(THF) estimated values
b
for conjugate acids of the respective bases.
a(HB+2 A-)
a(B2)
Resu lts
δB - δ
2
)
)
(13)
(14)
δ - δHB
+
-
A
Altogether 96 individual acid-base equilibrium mea-
surements in THF involving 45 bases were carried out
using the UV-Vis spectrophotometric or 13C NMR method.
These measurements give a continuous basicity scale in
THF as presented in Table 2, and some 13C NMR results
are given in Table 3.
Multiple overlapping measurements make the results
more reliable and help to estimate their self-consistency.
The entire basicity range covered involves at least two
independent pathways of measurements and the relative
basicity of any two bases can be obtained by combining
2
δ - δB
a(B1)
a(HB+1 A-)
1
δ
- - δ
+
A
HB1
In the case of alkylphosphazenes, the differences of the 13C
chemical shifts for alkyl substituent between phosphazene
base and acid forms are markedly lower as compared to the
corresponding differences of the arylic carbons of arylphos-
phazenes13 and were in the range of 3-6 ppm for the ones
measured.

Acid-Base Equilibria in Nonpolar Media
J . Org. Chem., Vol. 67, No. 6, 2002 1879
a
Ta ble 3. 13C NMR ∆p Kip Resu lts Not In clu d ed in
Ta ble 4. Ba sicity Da ta of th e Ba ses in AN, H2O, a n d th e
Ga s P h a se
Ta ble 2
base
reference base
PhP2(dma)
∆pKip
pKa values
GB
pKa values
base
in ANa
(kcal/mol)b
in H2Oc
t-BuP1(dma)
DBU
0.39
0.77
0.00
t-BuP1(dma)
EtP1(pyrr)
28.89d
28.35d
26.28
25.96e
26.88d
25.24
23.71
24.16
22.95
22.17
21.05
23.3f
2-Cl-C6H4P2(pyrr)
4-Me2N-C6H4P1(pyrr)
PhP1(pyrr)
t-BuP1(pyrr)
0.48
PhP2(dma)
TBD
TMG
DAB
DMAP
-1.11
-0.53
0.38
-0.45
0.87
0.08
-0.94
0.31
244.3
239.6
238.4
2-Cl-C6H4P1(pyrr)
2-Cl-C6H4P1(dma)
2,5-Cl2-C6H3P1(pyrr)
2-Cl-C6H4P1(dma)
2,5-Cl2-C6H3P1(pyrr)
4-Cl-2-NO2-C6H3P1(pyrr)
2,5-Cl2-C6H3P1(pyrr)
Et3N
t-BuP1(dma)
2-Cl-C6H4P2(pyrr)
4-Me2N-C6H4P1(pyrr)
DBU
4-MeO-C6H4P1(pyrr)
PhP1(pyrr)
Et3N
DMAN
4-Br-C6H4P1(pyrr)
TMG
0.25
13.6
PhP1(dma)
21.07
19.93
20.42
20.63
19.34
19.97
18.87
18.63
18.32
18.36
17.75
18.42
17.48
17.07
16.33
14.78
14.68
14.25
13.91
14.04
13.92
11.66
13.11
12.33i
10.42
a
∆pKip ) pKip(conjugate acid of the reference base) - pKip(conjugate
acid of the base).
4-CF3-C6H4P1(pyrr)
1-NaphtP1(pyrr)
PhTMG
236.9
218.8
12.18g
11.1
at least two independent sets of measurements. Reversi-
bility of protonation/deprotonation process of all bases
was checked. All equilibria presented in Table 2 were
reached in minutes and were stable. Both ion-pair (pKip)
values and values corrected for ion pairing (pKR) are given
in Table 2. Although somewhat arbitrary, the correction
for ion-pairing is useful because it makes our data
comparable to the data by the Morris group.10
The absolute pKR values have been obtained by an-
choring the scale to the pKR value of triethylamine in
THF (pKR ) 12.5),10 a secondary standard proposed by
the Morris’ group. This is not a perfect choice, but is the
most suitable anchoring point available for our data. See
Discussion for additional comments.
It is not easy to find a suitable anchoring point for the
ion-pair pKip values. The amount of available absolute
pKip values of bases in THF is scarce. The data on acids
is abundant but not directly comparable to pKip data of
bases (see ref 21 for further discussion). Therefore the
pKip values have been anchored to the pKR value of PhP1-
(pyrr). This anchorage is arbitrary, but this way the core
of the scalesthe P1 phosphazenesshave all practically
the same values in both scales, which facilitates the
discussion.
pyrrolidine
2-Cl-C6H4P1(pyrr)
2-Cl-C6H4P1(dma)
Et3N
2,5-Cl2-C6H3P1(pyrr)
2,6-Cl2-C6H3P1(pyrr)
DMAP
227
10.7
232.1
238.0
9.53
DMAN
12.1h
4-Cl-2-NO2-C6H3P1(pyrr)
5-Cl-2-NO2-C6H3P1(pyrr)
2-NO2-4-CF3-C6H3P1(pyrr)
2,4,6-Me3-pyridine
2,4-(NO2)2-C6H3P1(pyrr)
2,6-Cl2-4-NO2-C6H2P1(pyrr)
2,6-(NO2)2-C6H3P1(pyrr)
4-MeO-pyridine
2,6-Me2-pyridine
4-MeO-aniline
2-Me-pyridine
pyridine
aniline
2-Me-aniline
N,N-Me2-aniline
4-Br-aniline
2-MeO-pyridine
222.2
222.5
207.6
219.2
214.7
203.3
6.5
6.70
5.3
5.94
5.25
4.6
4.4
5.1
3.9
3.1
11.23
9.25
217.3
a
Slightly revised pKa values of conjugate acids of corresponding
bases obtained in previous work (ref 14) or in this work if not noted
otherwise. Gas-phase basicities from ref 25. c pKa values of
b
The absolute pKR values were calculated as in the
previous papers14,20 by minimizing the sum of squares u
conjugate acids of corresponding bases from refs 24 or 23 if not
noted otherwise. Reference 17. e Reference 26. f Reference 27.
d
i
of differences between directly measured ∆pKR values
and the assigned pKR values while keeping the pKR value
of triethylamine constant and equal to 12.5.
g
h
i
Reference 28. Reference 29. Anchor of AN scale; value taken
from ref 22.
AN14 (Table 4) have minor corrections (up to 0.03 pKa
units) due to the new measurements.
nm
u ) (∆pKi - (pK (HB+A-) - pK (HB+A-))) (15)
∑
R
R
2
R
1
i)1
Discu ssion
p KR Va lu es of Im in op h osp h or a n es. Unsubstituted
PhP1(pyrr) is a strong base with basicity (pKR ) 15.9)
between those of TMG and DBU. By substitution of the
phenyl ring, the basicity can be varied over a wide pKR
range. In this work the pKR values of substituted PhP1-
(pyrr) range from 7.5 (2,6-dinitro-) to 17.1 (4-(dimethyl-
amino)-), that is - by almost 10 orders of magnitude. The
influence of substituents in the phenyl ring on the
basicity of the phosphorane is easily predictable giving
the possibility to conveniently “tune” the basicity of the
phosphorane.
Alkyliminophosphoranes are significantly stronger bases
than aryliminophosphoranes. Thus, EtP1(pyrr) is by ca.
5.5 and t-BuP1(pyrr) by ca. 4 orders of magnitude
stronger than PhP1(pyrr). The inductive effect and some
delocalization of the lone electron pair of the imino
It should be stressed that the absolute pKR values of
bases given in Table 2 are not as accurate as the relative
pKRs. One could anchor the scale to any other absolute
pKR value, and the relative basicities will remain the
same. Precision s of the measurements was calculated
as in the previous papers:14,20
u
- nc
s )
(16)
n
x
m
The number of measurements is nm ) 83; the number
of pKRs determined nc ) 43. For our results, s ) 0.10 (for
pKip, s ) 0.09). Some previously published pKa values in
(21) Streitwieser, A.; Kim, Y.-J . J . Am. Chem. Soc. 2000, 122,
11783-11786.

1880 J . Org. Chem., Vol. 67, No. 6, 2002
Rodima et al.
nitrogen into the aromatic ring (see ref 14 for discussion
on this topic) are most probably the reasons.
possible uncertainties of these measurements. On the
UV-Vis measurements we did not observe noticeable
∆pKip dependence (see Table S1 in Supporting Informa-
tion) on concentrations while changing the concentration
of Et3N over the wide range (from 4.5 × 10-5 M to 2.3 ×
10-2 M) and keeping the 2-Cl-C6H4P1(pyrr) concentration
10-4 to 10-5 M.
The disagreements can be due to the formation of
aggregates of 1:1 ion pairs at higher concentrations,
especially at those used in the NMR method. Due to this
concentration dependence, the NMR measurements in-
volving bases of small size were not included in the scale
(Table 2).
As can be expected, the P2 phenyliminophosphoranes
are stronger bases than the corresponding P1 phenyl-
iminophosphoranes. The difference is ca. 4-5 pK units.
For alkyliminophosphoranes the same difference is ca. 6
pK units in acetonitrile.19 We do not have data on P2
alkyliminophosphoranes in THF, so direct comparison is
not possible.
The relatively good predictability of the basicity to-
gether with suitable spectral properties in the UV range
make the phenyliminophosphoranes convenient neutral
indicators in the medium to high basicity range. The
choice of neutral indicators in the high basicity range is
currently very limited.
We used the Streitwieser method30 to estimate the
mean aggregation numbers of the ion pairs. For the UV-
Vis data these were around 1, indicating that no signifi-
cant aggregation of the ion pairs was taking place during
the measurements. In principle the method also permits
to find the mean aggregation constants; however, due to
the very low extent of aggregation we could not get
reliable estimates for the aggregation constants. Also, it
was not possible to apply that method to the NMR data,
because it is necessary that one of the protonated bases
would be in the solution only in the form of a monomeric
ion pair. This condition is not met at the concentrations
used for the NMR measurements.
An ch or in g of th e Sca le. These disagreements cast
some doubt on the suitability of triethylamine as anchor-
ing point for our data: the pKR value of triethylamine
was determined by the Morris group using NMR mea-
surements (concentrations were in the range of 0.02-
0.07 M).10 The other two compounds common in this work
and ref 10, N,N-dimethylaniline (pKR ) 6.0)10 and TMG
(estimated pKR ) 15),10 are not as suitable because they
are, respectively, either at the very bottom of the scale
or have a pKR value that has only been estimated, not
measured. They are of similar size to triethylamine, so
that similar concentration dependence problems can be
anticipated. The pKR values of N,N-dimethylaniline and
TMG from this work are 4.9 and 15.3, respectively. Thus,
in the case of N,N-dimethylaniline there is a disagree-
ment, but the general picture would remain the same if
one of these two compounds would be used as the
anchoring point. To the best of our knowledge, besides
the work of the Morris group,10 there is no other absolute
basicity data in THF available in the literature that could
be used to anchor our scale.
Com p a r ison of Ba sicities in THF w ith Th ose in
Oth er Med ia . Correlation of pKip and pKR values in THF
with pKa values in acetonitrile yields the following
equations: pKip(THF) ) (-2.68 ( 0.55) + (0.83 ( 0.03)‚
pKa(AN); n ) 39; r2 ) 0.959; s ) 0.89. pKR(THF) )
(-5.08 ( 0.39) + (0.92 ( 0.02)‚pKa(AN); n ) 39; r2
)
0.983; s ) 0.63. Correlation of pKip and pKR values in THF
with pKa values in water yields the following equations:
pKip(THF) ) (1.78 ( 0.64) + (1.08 ( 0.08)‚pKa(H2O); n )
17; r2 ) 0.926; s ) 1.05. pKR(THF) ) (-0.31 ( 0.49) +
(1.14 ( 0.06)‚pKa(H2O); n ) 17; r2 ) 0.960; s ) 0.80.
Correlation of pKip and pKR values in THF with gas-phase
basicity pKa values (pKa(GB) ) GB (kcal‚mol-1)/1.364
(kcal‚mol-1)) is poor and yields the following equations:
pKip(THF) ) (-53.29 ( 12.52) + (0.39 ( 0.08)‚pKa(GB);
n ) 15; r2 ) 0.676; s ) 2.57. pKR(THF) ) (-61.34 ( 12.48)
+ (0.43 ( 0.08)‚pKa(GB); n ) 15; r2 ) 0.718; s ) 2.56.
From these correlations it appears that the differen-
tiating ability of THF for basicities is between water and
AN. One can see that the transfer of this reaction series
from THF to AN increases slightly its sensitivity toward
substituent effects both in case of pKip and pKR whereas
the transfer from THF into water leads to the opposite
result.
It is interesting to note that in all cases the correlation
is better with the pKR than with the pKip values. This
result indirectly validates the method of correction for
ion-pairing.
Con cen tr a tion Dep en d en ce of p Kip Va lu es. If we
assume that no larger associates than 1:1 ion pairs exist
in the solution then the pKip values should not show any
concentration dependence. The concentrations in the
NMR measurements are intrinsically higher than in
UV-Vis measurements, and the agreement between
these two methods serves as a good indicator. According
to the data in Tables 2 and 3, the results of the two
methods agree well for the ∆pKip values of arylphospha-
zenes. With bases of smaller size, however, disagree-
ments are observed. When comparing the results of the
measurement of the same equilibrium carried out by
different methods, the following is observed: the ∆pKip
between DBU and 4-Me2N-C6H4P1(pyrr) according to the
UV-Vis measurements is 0.83, according to NMR it is
0.48; according to the UV-Vis measurements DMAP is
by 0.54 pKip units stronger base than 2-Cl-C6H4P1(dma),
whereas according to the NMR measurements DMAP is
by 0.38 units weaker. The situation is even more serious
with triethylamine: according to the UV-Vis measure-
ments it is by 1.58 pKip units stronger base than 2-Cl-
C6H4P1(dma), according to the NMR it is by 0.87 pKip
units weaker. These discrepancies are larger than the
Con clu sion s
Relative ion-pair basicities ∆pKip of 25 substituted aryl
and alkyl iminophosphoranes and 20 other N-bases
(22) Coetzee, J . F.; Padmanabhan, G. R. J . Am. Chem. Soc. 1965,
87, 5005-5010.
(23) Tables of Rate and Equilibrium Constants of Heterolytic Organic
Reactions; Palm, V. A. Ed.; VINITI: Moscow-Tartu, 1975-1985.
(24) Perrin, D. D. Dissociation Constants of Organic Bases in
Aqueous Solution; Published as a supplement to Pure and Applied
Chemistry, Butterworth: London, 1965.
(25) Hunter, E. P. L.; Lias, S. G. J . Phys. Chem. Ref. Data 1998,
27, 413-656.
(26) Schwesinger, R. Nachr. Chem. Technol. Lab. 1990, 38, 1214-
1226 and references therein.
(27) Kolthoff, I. M.; Chantooni, M. K., J r; Bhowmik, S. J . Am. Chem.
Soc. 1968, 90, 23-28.
(28) Leffek, K.; Przuszynski, P.; Thanapaalasingham, K. Can. J .
Chem. 1989, 67, 590-595.
(29) Hibbert, F. J . Chem. Soc., Perkin Trans. 2. 1974, 1862-1866.
(30) Kaufman, M. J .; Streitwieser, A. J . Am. Chem. Soc. 1987, 109,
6092-6097.

Acid-Base Equilibria in Nonpolar Media
J . Org. Chem., Vol. 67, No. 6, 2002 1881
(various pyridines, amines, amidines) have been mea-
sured in THF medium using the UV-Vis and/or 13C NMR
methods. The ∆pKip values were corrected for ion pairing
using the Fuoss equation to obtain relative ionic basicities
∆pKR. Based on these measurements, a basicity scale
ranging from 2-methoxypyridine to EtP1(pyrr) and having
a total span over 18 pK units has been created. The scale
has been anchored to the pKR value of triethylamine
(pKR ) 12.5). It is practically impossible to give any truly
absolute pKip values at this time, because absolute pKip
data of neutral bases in THF is almost nonexistent in
the literature.
The predictability of the basicity together with suitable
spectral properties in the UV range make the phenyl-
iminophosphoranes convenient neutral indicators in the
high basicity range where the choice of neutral indicators
is very limited.
Ack n ow led gm en t. This work was supported by
grants 4376, 3992, 3366, and 3370 from the Estonian
Science Foundation.
Su p p or tin g In for m a tion Ava ila ble: Table (Table S1) of
detailed experimental data (includes data for all the UV-vis
equilibrium measurements: concentrations, acid used, calcula-
tion method) and 13C NMR spectra of selected compounds
(Figures S2). This information is available free of charge via
the Internet at http://pubs.acs.org.
The results are compared to pKa values in various other
solvents and in the gas phase. The pKR values give better
correlations than the pKip values, thus indirectly validat-
ing the procedure of correction for ion pairing.
J O016185P