HYDROGEN LABELING OF SOME
HETEROAROMATIC CARBON ACIDS
HARVEY LEWIS JACOBSON
A DISSERTATIO'! PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENiT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To my Father -
with respect, admiration
The author is indebted to his research advisor,
Dr. John A. Zoltewicz, for his guidance and endless
patience during the course of this work.
A special debt is owed the author's wife, Cindy,
not only for her support and understanding during the
later stages of this work, but also for providing the
incentive to finish it.
The author would also like to thank his fellow
graduate students and all those others who made these
few years a unique and rewarding experience.
Financial support from the Chemistry Department
of the University of Florida is gratefully acknowledged.
TABLE OF CONTENTS
LIST OF TABLES.....................................
LIST OF FIGURES.....................................
ABSTRACT ......................... .................
1. INTRODUCTION ...............................
2. ALKYL GROUP HYDROGEN-DEUTERIUM EXCHANGE
Preliminary Experiments .................
Deuteroxide Ion Catalysis...............
Catalysis by Other Buffers ..............
NMR Spectra of 4-Alkyl-1-methylpyri-
dinium Iodides in Liquid Ammonia ........
The Brdnsted Correlation.................
Transition State Structure ..............
3. ALKYL GROUP HYDROGEN-DEUTERIUM EXCHANGE
IN 1,3,6-TRIMETHYL- AND 3,6-DIISOPROPYL-
4. EXPERIMENTAL ...............................
Chem icals .............................. ....
Stock Solutions........................... 57
Nucleophiles ............................. 58
Substrates ............................... 59
Preparation of Solutions................... 61
Kinetic Procedure for H-D Exchange........ 61
pD Measurements ........................... 64
Control Runs ......... .................... 67
BIBLIOGRAPHY ...................................... .. 71
PREVIOUSLY PUBLISHED INVESTIGATIONS................. 75
Convenient Preparations of Mono- and Dideuterated
2-Furoic and 2-Thiophenecarboxylic Acids........... 76
Nuclpophilicities of Compounds with Interacting
Electron Pairs. Diazine-Catalyzed Ester
Hydrolysis ................................ ......... 82
BIOGRAPHICAL SKETCH................................. 92
LIST OF TABLES
1. Rate Constants for H-D Exchange of 1,4-Dimethyl-
and 4-Isopropyl-l-methylpyridinium lodides by
Deuteroxide lon.................................... 9
2. Kinetic Data for H-D Exchange of 1,4-Dimethyl-
pyridinium Iodide in Phenol Buffers at 75.0 ...... . 13
3. Kinetic Data for H-D Exchange of 4-Isopropyl-l-
methylpyridinium Iodide in Phenol Buffers
at 75.0 ........ .................................... 14
4. Kinetic Data for H-D Exchange of 1,4-Dimethyl-
pyridinium Iodide in 4-Amino-2,6-dimethyl-
pyridine Buffers at 75.0 .......................... 18
5. Thermodynamic Constants for 1,4-Dimethyl- and
Reacting with Deuteroxide Ion...................... 22
6. Kinetic Data for H-D Exchange of 1,4-Dimethyl-
pyridinium Iodide in Selected Buffers at 75.00..... 24
7. Kinetic Data for H-D Exchange of 4-Isopropyl-
1-methylpyridinium Iodide in Selected Buffers
at 75 .0 .. .................. ......................... 26
8. Summary of Rate Constants and pKa Values for
1,4-Dimethyl- and 4-Isopropyl-l-methyl-
pyridinium lodides Reacting in Selected Buffers
at 75.00............................................ 29
9. NMR Spectra of 1,4-Dimethyl- and 4-Isopropyl-
1-methylpyridinium Iodides and Their Conjugate
Bases in Ammonia................................... 34
10. Relative Rates of Hydrogen Isotope Exchange
of Alkylbenzenes at the a-Position ................. 41
11. Dissociation Constants for D20 and pH to pD
Conversion Factors at Selected Temperatures ........ 66
12. Solution Composition Data and Results for
Proteo Control Runs .............................. .. 69
LIST OF FIGURES
1. Plot of Kinetic Data for H-D Exchange of
1,4-Dimethylpyridinium Iodide in Phenol
Buffers at 75.00............................... 16
2. Plot of Kinetic Data for H-D Exchange of
4-Isopropyl-l-methylpyridinium Iodide in
Phenol Buffers at 75.00.......... ................ 17
3. Plot of Kinetic Data for H-D Exchange of
1,4-Dimethylpyridinium Iodide in 4-Amino-
2,6-dimethylpyridine Buffers at 75.00 ........... 19
4. Arrhenius Plot of log k versus 1/T for
H-D Exchange of 1,4-DimRehyl- and 4-Iso-
propyl-1-methylpyridinium lodides ............... 21
5. Bronsted Plot of log k versus pKa for H-D
Exchange of 1,4-Dimethyl- and 4-Isopropyl-
1-methylpyridinium Iodides in Selected
Buffers at 75.0 ................................. 30
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
HYDROGEN LABELING OF SOME
HETEROAROMATIC CARBON ACIDS
Harvey Lewis Jacobson
Chairman: John A. Zoltewicz
Major Department: Chemistry
Rates of hydrogen-deuterium exchange in the alkyl
group at the 4-position of 1,4-dimethylpyridinium iodide
and 4-isopropyl-l-methylpyridinium iodide in buffered
D20 solutions at 75.0 + 0.10 were obtained by the use
of nmr spectroscopy in order to determine the effects
of methyl groups on carbon acid acidity. Deprotonation
was observed to take place by general base catalyzed
reactions. An excellent Bronsted correlation (p = 0.75)
was obtained for the deprotonation of each carbon acid
using a series of structurally unrelated bases. Two
methyl substituents were found to exert very little rate-
retardina effect on the kinetic acidity. The largest
reactivity ratio, found for deuteroxide ion catalysis, shows
the methyl acid to be more acidic kinetically than the
isopropyl acid by a factor of 6.69. Other bases and
reactivity ratios are: phenoxide (2.71), 4-aminopyridine
(2.42), acetate ion (1.98), and imidazole (1.20). The
small effect of the methyl substituents on the kinetic
acidity is explained on the basis of a transition state
almost planar in structure with substantial charge
delocalization into the heterocyclic ring.
Attempts to measure the relative equilibrium acidity
of the two molecules in liquid ammonia by nmr spectroscopy
were unsuccessful due to the instability of the conjugate
base of the 4-methyl acid. It was determined, however,
that the 4-methyl acid cannot be substantially more
acidic, thermodynamically, than the 4-isopropyl acid.
Rates of hydrogen-deuterium exchange at the 6-position
of 1,3,6-trimethylpyridazinium iodide and 3,6-diisopropyl-
l-methylpyridazinium iodide in buffered D20 solutions at
75.0 0.1 were briefly examined. The isopropyl carbon
acid is substantially less acidic kinetically than the
methyl acid. The large reactivity difference is believed
to be the result of steric inhibition of resonance in the
deprotonated form of the acid.
Great interest has been shown in the acidity of weak
carbon acids and many investigations have been conducted
1 2 3
on the effect of structural changes on that acidity.
A carbon acid is an organic compound which when
treated with a suitable base, donates a proton to that
base by the breaking of a carbon-hydrogen bond. Since
most organic compounds contain carbon-hydrogen bonds,
most compounds are potential carbon acids. The study of
the effects on the acidity of a carbon acid is, therefore,
a study of widebpraed significance.
The acidity of carbon acids has been studied from two
different approaches; kinetic acidity, dealing with the
rate of the proton transfer reaction and thermodynamic
acidity, dealing with the position of the equilibrium
between the acid and its conjugate base.
For weak carbon acids, the rates at which protons are
transferred from Ia.rbon can be measured much more easily
than equilibrium constants, the most commonly used method
beinq base-catalyzed hydrogen isotope exchange. Using this
technique, kinetic aciaities of a wide variety of carbon
acids have been studied including arenes, sulfides, sul-
fones, carbon! compounds, halo and cyano compounds, and
nitroalkanes. Much of this work has involved attempts to
elucidate the relationship between structural changes in
a series of compounds with the effects of these changes on
the rate of the proton transfer reaction. In the present
study, the effects of methyl substituents on the acidity
of some alkyl-substituted heteroaromatic carbon acids were
The effect of the methyl group on carbon acid acidity
is an interesting subject since at first glance it might
seem that its behavior is quite erratic. Methyl groups
usually decrease both kinetic and equilibrium acidities
of carbon acids. The retardation effect of one methyl
group on the rate of proton transfer has been reported to
be as high as 700. It has also, however, been reported
to be so low as to have practically no effect at all. '
And in some s-ystems, for example the nitroalkanes or
fluorenes, a methyl substituent even acts to increase the
carbon acid equilibrium acidity.
In order to determine the effect of methyl substituents
on acidity in the 4-alkyl-l-methylpyridinium iodide system,
hydrogen isotope exchange experiments were carried out with
a number of different bases in buffered D20 solutions and
the rates of the second-order reactions were determined by
nmr spectroscopy. The relative kinetic acidities of 1,4-
dimethylpyridinium iodide and 4-isopropyl-l-methylpyridinium
iodide were determined. The magnitude of the relative
acidities of the two compounds is explained on the basis
of the structure of the transition state of the proton
transfer reaction. Attempts were also made to obtain
information on the equilibrium acidities of these two
molecules by studying their nmr spectra in liquid ammonia.
The kinetic acidities of 1,3,6-trimethylpyridazinium
iodide and 3,6-diisopropyl-l-methylpyridazinium iodide
were also briefly examined.
ALKYL GROUP HYDROGEN-DEUTERIUM EXCHANGE IN
The kinetics of H-D exchange in the 4-alkyl group,
(exchangeable protons underlined), of 1,4-dimethyl-
pyridinium iodide, I, 4--ethyl-l-methylpyridinium iodide,
II, and 4-isopropyl-l-methylpyridinium iodide, III, in
buffered D20 solutions of 1.0 M ionic strength (KC1 added)
were compared at 75.0 0.10. The exchange reactions were
followed by measuring the change in the integrated area
of the appropriate nmr signals relative to that of a non-
exchanging internal standard present in the reaction mix-
ture. The standard chosen was either tetramethylammonium
bromide or sodium acetate.
C 1H3 C HzC H3 CH(CH3)2
+ I- + I + I-
CH3 CH3 CH3
I II III
In the isotope exchange reactions of these compounds,
deprotonation could take place by either or both of the
following pathways employing catalytic bases OD and B,
C-H + OD~ C + HOD -D0 C-D + OH
C-H + B C + BH D> C-D + B
Since the reactions are carried out in D20, (110 M
in D), and the substrate concentration is never more than
0.6 M, it is reasonable to assume that the concentrations
of both HOD and BH are small enough to make the exchange
reaction effectively irreversible.
The rate of deprotonation, therefore, can be expressed
by equation 1,
rate = koD[C-H][OD ] + kB[C-H][B] (1)
where kOD is the second-order rate constant for deprotona-
tion by OD ion and kB is the second-order rate constant
for deprotonation by buffer base.
The reactions were carried out in buffered D20
solutions. Since base is not consumed, the concentrations
of OD and B are constant and the reaction is pseudo-first-
order, i.e., only the H content of the substrate changes
with time. Therefore, equation 2 can be written for the
pseudo-first-order rate constant associated with the depro-
k = k D[OD ] + kB[B] (2)
In order to approximate the reactivity differences
between the three compounds I III, they were studied
by pairs in the same buffer solutions. First, both the
4-methyl compound, I, and the 4-ethyl compound, II, were
dissolved in separate portions of the same 10:1 bicarbonate-
carbonate buffer solution and heated at 75.0 0.10. Since
both compounds were subjected to the same conditions, the
assumption was made that a relative reactivity ratio could
be obtained from the ratio of the respective reaction half-
lives, the reactivity ratio being the inverse of the half-
life ratio. Comparison of the compounds in this manner
indicated the 4-methyl protons of compound I are 1.8 times
more acidic than the methylene protons of the 4-ethyl group
of compound II.
The 4-ethyl compound, II, was then compared to the
4-isopropyl compound, III, by means of a 2:3 bicarbonate-
carbonate buffer solution at 75.0 0.10 and comparison
of the half-lives for these two exchange reactions
indicated the methylene protons of the 4-ethyl compound,
II, are more acidic than the methine proton of the 4-
isopropyl compound, III, by a factor of 5.
Combining these two ratios, the relative acidities
of the protons of the three different groups is obtained;
methyl / ethyl / isopropyl is 9 / 5 / 1. It appears that
the rate retarding effects of the methyl groups are
approximately additive, one methyl group decreasing the
rate of deprotonation by a factor of about 4. Considering
the relatively small reactivity difference between the
methyl and isopropyl groups, it was decided to concentrate
on these two compounds and discontinue further study of
the 4-ethyl compound.
Deuteroxide ion Catalysis
The value of k for H-D exchange in a 4-alkyl-l-
methyl-pyridinium iodide is expected to be dependent,
equation 2, on both the deuteroxide ion concentration and
the buffer base concentration, i.e., the reaction is
expected to be general base catalyzed. If the exchange
proceeded by specific base catalysis, i.e., either no
buffer base catalysis, kB=O, or there were no buffer base
present, [B]=0, the expression for the pseudo-first-order
rate constant would simplify to equation 3
k = koD[OD ] (3)
and a value for kOD could easily be obtained by measurement
of the observed rate and the pD of the solution.
For this purpose, a saturated Ca(OD), solution was
employed. This material has been shown to be an effective
alkaline pH standard. Due to its inclination to super-
saturate at higher temperatures, however, it was necessary
to carry out the reactions in solutions with a small
amount of solid Ca(OD)2 present to insure constant base
It was decided to obtain kOD values not only at 75.00
but also at 25.00 and 50.00 so that values of the energies
and entropies of activation for the 4-methyl and 4-isopropyl
compounds could be calculated. On performing the kinetic
runs, however, it was found that at 25.00, the 4-isopropyl
compound was too unreactive and at 75.00, the 4-methyl
compound reacted too rapidly. As a result, second-order
rate constants could not be obtained in these two cases.
The values that were obtained are listed in Table I.
Since it was not possible to obtain a value of kOD
for the 4-methyl compound in Ca(OD)2 at 75.00, it was
necessary to try a less basic buffer and apply equation
3 to obtain a value for kOD and also a value for kB, the
buffer base catalysis constant.
A phenol buffer was chosen and several runs were
carried out o-n both the 4-methyl and the 4-isopropyl com-
pounds using different buffer concentrations and ratios.
In order to obtain values for kOD and kB, equation 2 was
rearranged to the standard form of an equation defining a
straight line, equation 4.
:__ k [B] (4)
[OD] = B [B] + k (4)
[OD[ D] OD
Use of this equation to obtain values for kOD and
kB required the construction of a graph with axes of
k /[OD ] and [B]/[OD ]. Determination of the slope of the
line obtained by plotting these two quantities provided
the2 -1 -
the value of kB. The values obtained, 1.27 x 10 M sec
Table 1. Rate Constants for H-D Exchange of 1,4-Dimethyl-
and 4-Isopropyl-l-methylpyridinium lodides by
8.68 x 10- ---
7.30 x 10-2 7.78 x 10-3
4.00 x 10 1
6.55 x 10-
6.5 x 10 2
75.0 4.75 x 10 1
_3 _1 _1
for the methyl compound and 4.68 x 10 M sec for the
isopropyl compound, indicate the methyl compound to be
more reactive toward phenoxide ion by a factor of 2.7.
The intercept provided the value for kD. The values
-1 -1 -1
obtained, 4.00 x 10 M sec for the methyl compound
2 _1 _1
and 6.5 x 10- M sec for the isopropyl compound, indi-
cate the methyl compound to be more reactive toward
deuteroxide ion by a factor of 6.15.
It should be noted that considerable difficulty was
encountered in obtaining a consistent value for the
experimental pKa of phenol from the relationship pKa =
pD + log[BD ]/[B] when this was applied to reaction mixtures
containing the buffer. Close examination of this problem
led to the conclusion that the source of the error was
in the pD measurement.
The high concentration of iodide ion together with
silver ion from the electrode electrolyte solution slowly
clogged the porous electrode reference junction causing
substantial drifts in pD measurements. Elimination of this
drift was accomplished by the use of a thiosulfate wash
of the electrode but deviations of the pKa values could not
be eliminated, due possibly to interaction between sub-
strate and phenol buffer. As a consequence, it was.
necessary, for this one buffer, to determine a pKa in the
absence of substrate and to use this value to calculate
hydrolysis corrections and pD values in the manner now
Any buffer can undergo hydrolysis reactions which
serve to change the initial buffer ratio. A buffer acid
can dissociate into D and its conjugate base and,
similarly, buffer base can react with D20 to generate its
conjugate acid and OD The extent of such reactions for
a particular buffer is determined by the solution acidity
and the buffer concentration. For alkaline solutions,
a measure of the extent of hydrolysis is given by the
concentration of OD-. This concentration may, therefore,
be used to correct initial concentrations for hydrolysis.
In order to calculate hydrolysis corrections, equa-
tions5 and 6,
[D+][OD-] = KD20 (5)
[BD] = Ka (6)
where [B] and [BD ] are the equilibrium concentrations of
buffer base (phenoxide ion) and conjugate acid (phenol)
respectively, were combined to give equation 7.
Ka/K 0 O LBD1] (7)
Equilibrium concentrations of B and BD+ are obtained
from equations 8 and 9
[B] = [B]o [OD-] (8)
[BD+] = [BD+]o + [OD0] (9)
where [B]o and [DD ]o are the initial concentrations of
buffer base and conjugate acid, respectively.
Substitution into equation 7 of the experimentally
determined equilibrium constant, of a calculated value for
KD20 at 750, and of equivalent quantities of [B] and
[BD ] as given by equations 8 and 9 results in equation
1.26 x 10 [B]o [OD-]
2.98 x 10 [OD-]([BD+]o + [OD ])
Equation 10 has the form of a quadratic equation (11) where
ax + (ab + 1) x c = 0 (11)
1.26 x 10 10
a = 2.98 x 10-" b = [BD ]o, c = [B]o, and x = [OD-].
This equation was used to obtain [OD-] directly. This
value, when applied to equations 8 and 9, gave the
equilibrium concentrations of the species B and BD In
all cases, the correction for [B] was less than 5 percent.
Of the eleven corrections for [BD+] involving both the
methyl and isopropyl compounds, the correction in all but
three cases was also less than 5 percent. Of the three
remaining cases, two corrections of 13 percent were made,
one for each compound. In one case involving the 4-
isopropyl compound under the most basic conditions employed,
a correction to [BD ]o of 50 percent was necessary.
Buffer concentrations and kinetic results for the two
compounds in phenol buffer are reported in Tables 2 and 3.
r--- r, o ,.- ,
0) i r o C- Co
*r- 0 r-- -- .- r0 r-
X X X X X
r- CO LO r
*.. I CO C) CM CO C)
C 1 co 0 0 0 0l
I -- "- r-- r--
X X X X X
0 LnV n
"C CO C ) CO C CO
o r-- C -- r--
4- 0 0 .0 0 0
C 00 0000
--- r >- r-- r-
-/ III I
0 O C CO 0C) 0
X --2 r- r ---
I, coI I :- ,
I C-I CCJ CD 0 m LA
4-) LA LA CD m ((
1.0 m) :Z .
Co ri rCL
*r 3 a a i" f 5n
(-) OC ro
-OI C II O 0
I- c] (0 -j ro ro f
r- I I I I I
"0 "< 0 0 0 0
O I r r r -
X X X X
*I OJ O 0 -
3 I C 0 0 C 0
, ::: x X X X X X
4- M 0C\J CYJ co C- UO
i Ca co C) cn LO *- cn
x 17 C:) C) C) C) Cn) C)
L- 0 00 0 0 0 0 0
*r' l- X X X X X X
t-4 a) r- C m O O
O4- Co Un cN U,. o
O) c r r-- - L r-
4 -- ),
C C) C) C) C) C
W .- I I I - I O-
+ I1 X X X X X X
IOO c ) r 1 0 0 U
-- o 00 J L o x
a'. C) 0 .C4 C) 0)
.0 0) 00 0 CO O cn m
4- 0 C.)
+I Ln Ir Ln I I C)
c cin CO o o oc0
*r- a' 01 i- O i-
Graphical treatment of the data in the manner previously
described gave the results shown in Figures 1 and 2.
Comparison of the value of koD for the 4-isopropyl
compound obtained from the phenol buffer runs (6.5 x 10
M sec ) with the value obtained from Ca(OD)2 solution
2 _1 _
(6.55 x 10 M sec ) shows a more than satisfactory
agreement. Since only one determination of the value of
kOD for the 4-methyl compound has been obtained, however,
it would be appropriate to use another buffer to verify
1 1 1
the value of 4.00 x 10 M sec- derived from the phenol
For this purpose, 4-amino-2,6-dimethylpyridine was
chosen as a buffer. Buffer solutions of this compound
are not as basic as the phenol buffer solutions but the
two methyl groups ortho to the pyridine nitrogen would
be expected to favor reaction by deuteroxide ion by steric-
ally hindering reaction by the buffer base.
Four rates were measured using the 4-amino-2,6-dimethyl-
pyridine buffer. The conditions employed and results
obtained can be found in Table 4. Graphical treatment of
the results in a manner similar to that used for the
phenol buffer runs is shown in Figure 3.
The value of kOD obtained for the 4-methyl compound
using the pyridine buffer is also included in Table 1.
This value, 4.75 x 10 M- sec and the value obtained
-1 -1 I,
using the phenol buffer, 4.00 x 10 M sec were then
X 1 0
*r a /) I
o o a
\ 0 1 --a
a 4- 4-'
o ( -
\) 0 C a2
L n3 Lii C -
O C" I
C L1 4-)
S- U) 4-'
o- o c
O 0 C
*- I 0
C *4- --
e- 0 C
LC n -
\ U~ 0
\ < C t
0 L M 0 0
1 I I I I
r- X X x x
(L I C\ CO 0 CO
C) CO Co C)
- I X X X X
>f3 --- I -- t) U3
-- o 0 0 00 0
- S" CD CD) D 0
a, r- -
*l- .4> 04 (' CM 0
by I I I
Sk0 0 0 0
-- <) + X X X X
4-4- CC U ci -
oC LJ U)n M UO O'i
CD CU CM CM'
I I I
(o II IU I- cI
I < I X X X.
-U o N*
0 O0 0 0 0 0
.- aC xxxx
a) .- . Q
1- i^ M- n -
S*- 4- rI
*r- CO r-
aU 4- -
I L 3
averaged. Deviation of the two values from the average
1 _1 1
of 4.38 x 10 M sec is less than 9 percent.
As a check on the values of kOD obtained at the three
temperatures, an Arrhenius plot of log kOD versus the
inverse of the temperature was constructed using the values
of kOD obtained in Ca(OD)2 solution and, for the 4-methyl
compound, the average of the two values obtained at
75.00 from buffer studies. The linearity of this plot,
Figure 4, strongly suggests that the deuteroxide ion
catalytic constants have been determined correctly.
Energies of activation and entropies of activation
for the two compounds were calculated by equations 12
log k2 log k = (E/2.303R)(T2-T (12)
AS /2.303R = log k 10.753 log T + (2.3 3RT) (13)
Results are reported in Table 5. Pearson and Dillon
have compiled a list of activation energies and entropies
for slow proton transfers from 6-diketones and nitro-
alkanes. In comparison with those results it can be con-
cluded that the deuteroxide ion catalyzed deprotonation
of the two pyridinium iodides shows typical behavior for
slow proton removal from weakly acidic species.
I I I I I I 3
2.90 3.00 3.10 3.20 3.30 3.40 1/T x 10
Figure 4. Arrhenius Plot of log kO) versus 1/T for H-D
Exchange of 1,4-DimethyT O)- and 4-Isopropyl
(A )-l-methylpyridinium lodides.
Thermodynamic Constants for 1,4-Dimethyl- and
4-Isopropyl-l-methylpyridinium lodides Reacting
with Deuteroxide Ion.
Eact. 16.3 kcal/mol. 19.2 kcal/mol.
AS a -15.4 e.u. -11.1 e.u.
aCalculated at 50.00.
Catalysis by Other Buffers
With the value of kOD known, kB can be easily cal-
culated for any buffer base once the pD is measured and
k is obtained for any given kinetic run.
Rates were measured using six different buffers:
2,2,2-trifluoroethanol, 4-aminopyridine, imidazole,
2,6-dimethylpyridine, pyridine, and acetic acid. Buffer
concentrations and kinetic results are given in Tables
6 and 7.
In the cases of four of these buffers (2,2,2-tri-
fluoroethanol, 4-aminopyridine, imidazole, and acetic
acid), kinetic runs were carried out as competition
experiments. Both the 4-methyl and the 4-isopropyl
compounds were run in the same solution. In this way,
both the rate of reaction and the relative reactivity of
the two compounds could be measured without regard to
differences or changes in individual solutions.
In the cases of 2,6-dimethylpyridine and pyridine
buffers only one of the substrates was used. It was not
possible to study the 4-methyl compound in a 2,6-dimethyl-
pyridine buffer due to methyl group overlap in the nmr
spectrum. The 4-isopropyl compound was not studied in
a pyridine buffer due to the lack of a suitable internal
standard for nmr analysis. Exchange catalyzed by acetate
ion, the usual internal standard employed in runs with the
4-isopropyl compound, is too competitive with exchange
catalyzed by pyridine to allow reliable rate measurement.
Kinetic Data for H-D Exchange of 1,4-Dimethyl-
5.488 x 10
1.364 x 10
1.357 x 10-
6.462 x 10
2.500 x 10-
9.627 x 10
7.063 x 10-
aUncorrected for deuteroxide ion catalysis and acetate ion
b[acetate] = 0.100 M.
C[acetate] = 0.090 M.
dpKa = pD + log [BD+]/[B].
pyridinium Iodide in Selected Buffers at 75.00.
+ d -1 1
[BD+],M [OD ]/[B] pKa,obsd k ,M sec
-2 -1 -1
1.71 x 10 1.55 x 10 10.95 2.78 x 10
2 4 _4
1.90 x 10- 4.29 x 10 8.44 5.40 x 10
1.90 x 10 4.33 x 10 8.44 5.60 x 10
-2 _6 _5
1.94 x 10 7.67 x 10 6.70 2.60 x 10
1.00 x 10- 1.32 x 10 6.65 2.15 x 10
_2 -7 _6
1.94 x 10 4.47 x 10 5.46 4.97 x 10
-2 -7 -6
1.91 x 10 2.77 x 10 5.25 3.39 x 10
catalysis when present.
Kinetic Data for H-D Exchange of 4-Isopropyl-l-
7.098 x 10-
6.548 x 10
4.810 x 10
4.907 x 10
4.882 x 10
1.88 x 10-
3.19 x 10
3.505 x 10
aUncorrected for deuteroxide ion catalysis and acetate
b[acetate] = 0.090 M.
c[acetate] = 0.100 M.
dpKa = pD + log [BD+]/[B].
X 10 -3
methylpyridinium Iodide in Selected Buffers at 75.00.
8.98 x 10-
1.71 x 10
1.90 x 10
1.90 x 10
1.94 x 10
1.00 x 10
1.02 x 10
1.91 x 10-2
1.91 x 10
3.06 x 101
1.55 x 10
4.29 x 10
4.33 x 10-
7.67 x 10
1.32 x 10-6
9.29 x 10
2.77 x 10
2.97 x 10-
3.08 x 10
2.10 x 10
2.42 x 10
2.20 x 10
1.72 x 10-5
2.58 x 10
1.72 x 10
catalysis when present.
Also, use of the ring protons of the 4-isopropyl compound
as a standard is prohibited by overlap of the pyridine
ring protons in the nmr spectrum of mixtures of the two
Attempts were made to obtain rate constants for D20
acting as a base by use of 0.1 M DC1 solutions but they
proved unsuccessful. Both compounds appeared to degrade
in the acidic solution. The methyl compound degraded
approximately 10 percent after one week of heating; the
isopropyl compound degraded about 10 percent after two
weeks of heating. No exchange, as evidenced by the
broadening of the 4-methyl singlet or the emerging of a
singlet between the 4-isopropyl gem.-dimethyl doublet,
could be detected in the nmr during this period.
Average values of the rate constants obtained in
all buffers used along with the respective buffer pKa
values are listed in Table 8. They are graphically
represented by the Bronsted plot found in Figure 5.
The equilibrium constants for D20 and imidazole are
statistically corrected to reflect the two acidic
centers in each acid, i.e., pKa + log 2 values are
used. Both the equilibrium constant and the rate con-
stant for acetic acid are statistically corrected to
reflect the two basic centers of the acetate ion, i.e.,
pKa log 2 and log k l log 2 values are used.
Proteo control runs were carried out to verify the
stability of the two pyridinium iodides in selected
CO C C\
I CJ 0
I I I
I I I
X X X
S ( CO
C\J O L
-- .-- rn
X X X
LO D rC
Co L mc
co LO C\M
- C\J C
L r -r-
o o CD 0
C N *,-- *r- U
*,-- 3 Q *r-
E I r --
C *i- L O
I E > u
LO LD r 0 co mn
co C\J m
O- m C0
0 0 0
r--- e- e-
X X X
Lnl C co
u3 C) 'z-
a..- 4 --
< Ln a4-
\\ co (1 5
\\J \ ')- L
\ \ c
\ \ 'Q-
\ \ -^ ^
\ \ a '
co \ \ <
buffers. The control runs on the 2,2,2-trifluoroethanol
buffer showed substantial pH changes and new nmr peaks on
heating. As a result, the values reported in Table 8 for
this buffer are considered uncertain and are excluded
from the Br0nsted plot in Figure 5. Details of all the
control runs are given in the experimental section.
The least squares lines calculated from the data in
the Br0nsted plot of Figure 5 give equation 14 for the
4-methyl compound and equation 15 for the 4-isopropyl
compound. The uncertainty is expressed in terms of the
standard deviation. The correlation coefficient (r)
is also given. The Bronsted slopes are the same for
log k2 = (0.76 + .12) pKa 9.62 + 0.96
r = .991. (14)
log k2 = (0.75 + .07) pKa 9.86 + 0.58
r = .998 (15)
the two compounds.
NMR Spectra of 4-Alkyl-l-methylpyridinium lodides in Liquid
In order to obtain some knowledge of the equilibrium
acidities of both 1,4-dimethyl- and 4-isopropyl-l-methyl-
pyridinium iodides, nmr spectra of the two compounds in
liquid NH3 were recorded. It was hoped that observable
amounts of the conjugate bases of the two compounds (IVa
and IVb) would form. If so, the amount of the conversion
could be used to provide some measure of the relative
acidity of the two carbon acids in the basic solvent.
The compounds were added to nmr tubes along with about
one milliliter of liquid NH The tubes were then sealed,
warmed to room temperature, and their nmr spectra recorded.
A solution of the 4-methyl compound in ammonia was
opaque and dark green. The nmr spectrum, however, showed
the presence of nothing other than the 4-methyl pyridinium
iodide. Standing overnight resulted in no change in the
spectrum although the formation of some solid precipitate
in the tube was noted.
A solution of the 4-isopropyl compound was a clear
orange. The nmr spectrum indicated that signals of the
starting material and additional up'ield signals attribut-
able to the conjugate base were present. Integration showed
approximately 10 percent of the pyridinium iodide had been
converted to the conjugate base.
Ammonia solutions of the two compounds were prepared
again and this time solid KOH was also added before the
tubes were sealed.
The nmr spectrum of the opaque, dark green solution of
the 4-methyl compound in ammonia with KOH added at -35
showed approximately 20 percent of the material had been
converted to the conjugate base. Upon warming the solution
to room temperature, the amount of this form increased so
as to become the predominate form. However, the conjugate
base apparently is unstable; signal strength slowly decreased
until all nmr signals disappeared.
The nmr spectrum of the clear orange solution of the
4-isopropyl compound in ammonia with KOH added at -350
showed approximately 30 percent of the material had been
converted to its conjugate base. Upon warming to room
temperature, the only signals observed were those corres-
ponding to the deprotonated form. After standing overnight,
the spectrum was unchanged with no loss of signal.
Solutions of the pyridinium iodides in ammonia were
then prepared with a benzene internal standard so that nmr
chemical shifts could be assigned. Periodic integration
of the 4-methyl solution indicated a 10 percent loss of
substrate signal within an hour and a 20 percent loss
after standing overnight.
The nmr assignments are reported in Table 9. Depro-
tonation of each carbon acid results in up-field shifts
for the ring protons of the conjugate base. It is assumed
that H-2,6 is at lower field than H-3,5 in the conjugate
base, the same order as in the acid.
Table 9. NMR Spectra of 1,4-Dimethyl- and 4-Isopropyl-
1-methylpyridinium lodides and Their Conjugate
Bases in Ammonia.d
ac values with benzene, T
J = 7 Hz.
CPresence of several peaks
= 2.60, as internal standard.
at high field precludes
It might at first appear that the 4-isopropyl compound
is the more acidic of the two iodides in ammonia. However,
due to the instability of the conjugate base of the 1-4-
dimethylpyridinium ion, it cannot be conclusively stated
that this is in fact the case. Nevertheless, on the basis
of the results obtained, it can be stated that if the
4-methyl compound is the more acidic of the two carbon
acids, the relative acidity ratio will be small.
Due to possible variation in the amount of KOD in
solution, the amount of water present, and other possible
complicating factors, the equilibrium acidities of the two
compounds can only be properly compared by having both
compounds in the same tube. As can be seen from the chemical
shifts listed in Table 9, however, signal overlap makes this
It has previously been shown that the conjugate base
of 1,4-dimethylpyridinium ion is unstable. Ethereal
solutions of the material, obtained by quickly extracting
highly alkaline solutions of the pyridinium ion, rapidly
degrade. Substitution in the alkyl group of electron
withdrawing substituents, however, greatly enhances sta-
bility. For example, the conjugate base of 4-(a,a-
diphenylmethyl)-l-methylpyridinium ion is a stable solid.
The Brdnsted Correlation
The Br0nsted plot in Figure 5, derived from the bases
numbered 3, 5, 6, 8, and 9, shows remarkable correlation
considering the different types of bases involved. Phen-
oxide ion, 3, two pyridines, 5 and 8, acetate ion, 9, and
imidazole, 6, all fit the derived Brdnsted line without
Behavior such as this is not unique in a Br0nsted
relationship. Strictly speaking, the Br0nsted equation
is expected to hold for a series of related bases with no
significant structural or electronic differences such as
a series of carboxylate anions or a series of structurally
similar amines. Good correlations can be found, however,
for bases of different types in some instances.
For example, the base-catalyzed H-D exchange of
isobutyraldehyde-2-d has been studied extensively. Although
the results for different classes of amines could not be
represented by a single Br0nsted line, both pyridine
and phenoxide ion bases can be included in the same
As a better example, in a study of the base-catalyzed
enolization of acetone, a reaction which like H-D exchange
involves the removal of a proton as the rate determining
step, it was found that pyridines, carboxylate anions,
and amines all fit a common Brdnsted plot.
As previously stated, the correlation of the different
types of bases in the Bronsted plot of Figure 5 is quite
good. Mention should also be made, however, of those bases
whose corresponding points do not fit the Br0nsted
The most obvious deviation can be seen in the points
(1) corresponding to deuteroxide ion catalysis. From the
graph, it appears that the deuteroxide ion catalyzed
reaction is slower than expected by a factor of 360 for
the 4-methyl compound and by a factor of 1200 for the
4-isopropyl compound. This anomolous behavior, however,
is not uncommon in a Bronsted relationship and although
is not generally understood, it is discussed extensively
18 19 20
in the literature. '
The deviation of the points for 4-amino-2,6-dimethyl-
pyridine (4) and 2,6-dimethylpyridine (7) can be readily
understood as examples of decreased reactivity due to
steric hindrance of proton transfer by the ortho methyl
groups of the buffer. Steric hindrance of this type has
been well documented for reactions involving rate-limiting
16 21 22
proton abstraction from carbon. In the base-
catalyzed H-D exchange of isobutyraldehyde-2-d, 2,6-
dimethylpyridine has been found to be less reactive than
expected by a factor of almost 150 while for the base
catalyzed deprotonation of 2-nitropropane, the observed
rate is less than expected by a factor of only 5. The
steric effect of the ortho methyl groups retards the rate
of H-D exchange in 1,4-dimethylpyridinium iodide by a
factor of 5 and by a factor of 40 in the 4-isopropyl
pyridinium iodide. The greater effect on the isopropyl
compound than on the methyl compound is quite consistent
with the idea of steric hindrance.
The slight deviation from the Bronsted line observed
for the imidazole catalyzed exchange of the 4-methyl
compound, while not appearing too significant, is not
unprecedented. In the base-catalyzed H-D exchange of
isobutyraldehyde-2-d, N-methylimidazole is three times
less reactive than expected from a consideration of the
reactivity of unhindered pyridines. This decrease in
the effectiveness of N-methylimidazole as a catalyst
is attributed to the changes in internal geometry of the
imidazole ring resulting from protonation of the molecule.
In the general base-catalyzed dehydrochlorination
of 9-fluorenylmethyl chloride to give dibenzofulvene,
both imidazole and N-methylimidazole are off the Brgnsted
line established by tertiary amines by an amount equivalent
to a fifty-fold reduction in catalytic effectiveness,
a result which would seem to support the idea that imida-
zoles do not correlate well with other bases. However,
in a paper dealing with the amine catalyzed elimination
from a -acetoxy ketone, both imidazole and N-methyl-
imidazole correlate quite well with the Bronsted line
determined for tertiary amines.
In light of the fact that no plausible argument
has been put forth to explain the apparent inconsistency
in behavior of imidazole and its N-methyl analog, it
cannot be determined whether or not the small deviation
observed for the imidazole point obtained in this study
is in any way meaningful.
Transition State Structure
The striking feature of the exchange results listed
in Table 8 and illustrated in Figure 5 is that although
the reactivity order is the expected one on the basis
of the electron-releasing character of a methyl group,
the reactivity difference between the 4-methyl and
4-isopropyl compounds is quite small. The largest
effect the substitution of two a-methyl groups for two
protons has on the rate of exchange is observed in the
case of deuteroxide ion catalysis. The reactivity ratio
for the methyl compound relative to the isopropyl compound
is 6.7. For the other bases, the ratio ranges from
2.7 to 1.2. Although the ratio decreases as the catalyst
becomes more weakly basic, changes are small and the ratio
for imidazole (1.2) appears to be unusually small.
This rate-retarding effect of methyl groups ranks
among the smallest known. The small size, however,
provides considerable information about the structures
of the transition states of the deprotonation reactions.
Information from the literature on the magnitude of
rate-retarding effects of methyl groups on other reactions
involving deprotonation of carbon acids makesit clear
that the effect of the methyl groups in the case of the
pyridinium ion carbon acids is among the smallest on
The examples now considered include arenes, sulfides,
sulfones, various carbonyl compounds and nitroalkanes.
Where necessary, the reactive center is underlined.
The rates of base catalyzed hydrogen isotope exchange
at the a-position of alkyl benzenes has been examined
utilizing different base and solvent systems. These
include potassium amide/ammonia, potassium cyclohexylamide/
cyclohexylamine, and potassium tert-butoxide/dimethyl-
sulfoxide-t. The reactivity of the methyl group of
toluene is greater than that of the isopropyl group
of isopropylbenzene by factors ranging from 35 to 125.
Table 10 records the results.
The very large effect of the methyl groups on the
acidity of some thioalkylbenzenes in ammonia causes rates
Table 10. Relative Rates of Hydrogen Isotope Exchange of
Alkylbenzenes at the a-Position.
at 10 a ,b
of amide ion catalyzed dedeuteration to vary over four
powers of ten.
CHsSCD3 / CHsSCD2CH, / CHsSCD(CH3)2
10 / 10 / 1
The rates of deuteroxide catalyzed H-D exchange
in DO2-dioxane at the non-benzylic a-position of a
series of alkylbenzylsulfones also are highly influenced
by the presence of methyl groups. Relative rates are
indicated. It is suggested that this relative reactivity
C6HsCH SO2CH3 / CHsCH2SOCH2CH, / CH 5CH2SOCH(CH,3)
104 / 10 / 1
may be inflated by as much as two powers of ten due to
the presence of internal return. But that still leaves
a relative reactivity of at least 10 / 10 / 1.
The rates of methoxide catalyzed H-D exchange in
methanol-0-d of a series of a-substituted methyl acetates
were determined. Again, a reactivity difference of
two powers of ten between an unsubstituted and dimethyl
substituted carbon acid was observed. '
CH3CO CH, / CH CH2COCH3 / (CH )CHCO2CH
--3 2 3 / 3 2 2 3 H3 C
Similar results were obtained for deprotonation in
a series of alkyl ketones in aqueous hydroxide.
(CH3) 2CO / (CH 3CH ) CO / (CH 3) CH) CO
Rates of proton abstraction in methoxide/methanol
of another pair of ketones show that a single methyl
group retards the rate by a factor of 35.33
C6H5CHC1COCH2CH / CH5CHC1COCH(CH )2
Similar results are indicated for deprotonation
reactions of 1,3-dicarbonyl compounds in aqueous hydroxide.
The rate-retarding effect of a single methyl group varies
from a factor-of 68 to 136.3
(CH CO) CH2 / (CHCO)2 CHCH3
CH COCH2CO2 C2H / CHOCOCH(CH )CO 2C2H
CH, (CO2CH3)2 / CH CH(CO2CH3)2
Finally, nitroalkanes have been studied in some
detail. Two methyl groups retard the rate of deprotona-
tion by deuteroxide ion by a factor of 87.
CH3NO / CH 3CH2NO2 / (CH3) CHNO2
87 / 16 1
However, when the base is acetate ion or water,
two methyl groups have essentially no rate-retarding
effect. That is, the unsubstituted and the dimethyl
substituted nitromethanes react with these two bases
at essentially the same rate. This methyl group
effect clearly ranks among the smallest for carbon acid
The rate retarding-effect of a methyl group on
deprotonation at carbon has been rationalized in terms
of the electron-releasing effect of the group. Negative
charge builds up on a carbon as the transition state
for deprotonation is approached; a methyl group by its
inductive effect destabilizes the developing negative
charge and thereby retards the rate of the reaction.
In order to understand the variation in the magni-
tude of the effect of the methyl groups with changes
in the basicity of the catalyzing base, it is necessary
to consider the effect of methyl groups on the equilibrium
acidity of nitroalkanes. Methyl groups increase the
acidity of nitroalkanes, as the following pKa values
CH3NO, CH3CH2NO2 (CH3)2CHN02
pKa 10.2 8.5 7.7
The principal factor governing the pKa changes
in the nitroalkanes is the resonance stabilization of
the carbanions. Two important resonance structures
can be drawn for a nitronate ion.
0 R 0-
R-C-N < C=N
R O- R/ \0-
Because the negative charge can be contained on
the electrone-gative oxygen atoms, the structure with
a CN double bond is a more important contributor to
the resonance hybrid than that with a single bond. The
stabilizing effect of the methyl group on a double bond
becomes more important than the inductive effect of the
methyl group and so stabilization results.
The interesting variation in the methyl group
effect found for nitroalkanes reacting with a series
of bases is said to reflect changes in the extent of
CH bond cleavage in the transition state. In the
presence of a strong base like hydroxide ion, the struc-
ture of the transition state for deprotonation is
more reactant-like and therefore the reactivity is
determined by the inductive effect. With a weaker base,
the structure of the transition state is closer to that
of the product, the anion. As a result, the stabilizing
effect of the methyl groups (as seen in the pKa values)
becomes increasingly important.
This traditional interpretation has been challenged
as a consequence of new results dealing with the kinetic
and equilibrium acidities of arylnitroalkanes. No
evidence was found to support the idea of variable transi-
tion state geometries with variable base strength when
the rates of deprotonation of arylnitroalkanes by a variety
of bases were determined. Relative rates of deprotonation
varied only by a small amount, regardless of whether a
strong or weak base was employed as a catalyst.
It is not clear, however, whether the older inter-
pretation for nitroalkanes really is invalidated by
the new results. It is not clear if, in fact, a more
product-like transition state requires more negative
charge to reside on carbon. It is possible that in the
arylnitroalkanes, a larger fraction of the negative
charge may be borneby the nitro group leaving the amount
of charge on carbon about the same regardless of varia-
tion in the structure of the transition state.
With this background it now is possible to interpret
the kinetic results for the pyridinium ions. The key
observation is that the rate-retarding inductive effect
of the two methyl groups is unusually small, regardless
of the identity of the catalyzing base. This requires
that the amount of charge on the carbon being deprotonated
be small. Two interpretations are possible.
According to the first, the kinetic effect is small
because the extent of CH bond cleavage in the transition
state is small and the transition state structure closely
resembles that of the reactants. This interpretation can
be rejected because (a) it is not consistent with the large
Br0nsted B value of 0.75 which implies significant proton
transfer, (b) the deprotonation reactions are expected
to be endothermic and therefore the transition state
should not resemble reactants in structure, and (c) the
enormous activating effect of the heteroatom is not
consistent with this view. The substrates "aza-p-xylene"
and "aza-p-cymene" are isoelectronic with p-xylene and
p-cymene, yet they are enormously more reactive than
their hydrocarbon counterparts. This large difference
in reactivity is strongly contradictory to a small amount
of CH bond cleavage.
The second interpretation, more consistent with the
small kinetic effect of the two methyl groups, is that
in the transition state, there is substantial cleavage
of the CH bond and a substantial fraction of the negative
charge is delocalized into the heterocyclic ring. Charge
neutralization involving the positively and negatively
charged centers of the conjugate base is expected to
play a very important role in stabilizing both the transi-
tion state and the conjugate base as illustrated by the
resonance structures for the conjugate base. It should
R -R R ,R
CH- C H
be noted that the uncharged structure IV is a kind of
enamine with an olefinic carbon atom as part of the
The small methyl group effect admirably supports
this proposed transition state. Largely off-setting the
inductive deactivating effect of methyl groups, which
serves to make the isopropyl substrate less reactive
than the methyl compound when the side-chain carbon
bears a negative charge, is the well-known stabilizing
effect of methyl groups bonded to an olefinic center.
Because of the olefinic character of the reactive center
in the transition state, the destabilizing effect of the
methyl groups is substantially attenuated. Perhaps the
carbon atom at the reactive center has the geometry of
a flattened pyramid, i.e., the ligands bonded to the
carbon atom are approaching a state in which they lie
in a plane defined by the heterocyclic ring.
It does not necessarily follow from the above
description of the transition state that the equilibrium
acidity of the isopropyl acid will be greater than that
of the methyl acid. Methyl groups need not increase
equilibrium acidities as in the case of the nitroalkanes.
In fact, methyl groups usually decrease both kinetic and
equilibrium acidities of carbon acids. Carbon acids
containing cyano and sulfonyl groups show behavior
of this type. Whether or not the isopropyl acid is more
acidic than the methyl acid depends on how much negative
charge resides on the side chain. The more the conjugate
base resembles an olefin, the more likely it will be
that the isopropylpyridinium ion will be the stronger
Although the equilibrium acidities of the two pyri-
dinium ion carbon acids were not determined, a pKa value
has been obtained for the 1,2-dimethylpyridinium ion
N+ CH3 a C HI
using an acidity function determined by a DMSO-water
solution containing tetramethylammonium hydroxide. This
value is reported to be 20.0.42 It seems reasonable to
assume that the pKa for the 1,4-dimethylpyridinium ion
is greater than 20 from the following two considerations.
First, the kinetic acidity of 1,2-dimethylpyridinium
iodide has been examined and was found to be greater
than that for the 1,4-dimethyl compound. 3 It is unlikely
that the equilibrium acidity order would be inverted.
Second, when deprotonation takes place on a side-chain
bonded to the heterocyclic ring, the equilibrium acidity
of a 2-substituted pyridinium ion has been found to be
greater than that for a 4-substituted one as long as no
complicating steric factors exist.4" For methyl substi-
tution, the steric effect should not be significant.
In as much as the acid-strengthening (equilibrium)
effect of methyl groups is uncommon, it is well to review
another example, 9-alkylfluorenes. A number of molecular
I L 0 0 0 + H-'
explanations have been advanced; they will be reviewed.
The equilibrium acidities of a series of 9-substituted
fluorenes in dimethylsulfoxide-water were determined.45
It was found that contrary to the expectation that alkyl
groups are electron releasing and des tdbilize carbanions
in solution, 9-methy!fluorene is more acidic than fluorene.
Also noting that the acidity of the 9-alky1fluorenes
followed the order of CH3 > C2Hs > CH(CH3)2, it was con-
cluded that the acidity order could best be explained
on the basis of an anionic hyperconjugation.
In a study by other workers, it was suggested
that alkyl group stabilization of carbanions occurs by
dispersion interactions, a sort of internal van der Waals
or London electronic correlation effect. Still others
attempted to explain the acidity order by suggesting a
methyl group somehow increases charge delocalization.
In the latest study of the acidities of 9-alkyl-
fluorenes in cyclohexylamide/cyclohexylamine, results
have been obtained consistent with other workers. The
explanation proposed, however, is far more convincing.
It is suggested that the increased acidity of
9-methylfluorene is due to a stabilization by the methyl
group resulting from a sigma bond strength change. In
the fluorenyl anions, the negative charge is extensively
delocalized and the deprotonated carbon is sp hybridized.
In 9-methylfluorene, this would mean a change from
3 3 3 2
Csp -Csp in the hydrocarbon to Csp -Csp in the carbanion.
In fluorene itself, the comparable bond change is Csp -H
to Csp -H. There are abundant analogies to show that
putting more s character into a carbon-carbon bond provides
greater stabilization. This sigma bond stabilization
effect is large enough to override the counteracting methyl
This argument was generalized in some later work
in a form quite applicable to the results of this present
study. It was stated that in conjugated carbanions where
only a partial negative charge is associated with the
substituted carbon, the stabilizing effect of methyl
substituents on trigonal carbanions dominates and alkyl
substituents enhance acidity. In proton transfer transi-
tion states the central carbon is still pyramidal and
hybridization stabilization is reduced. Similarly, less
charge is delocalized than in the product carbanion.
The more charge is concentrated at the central carbon,
the more the inductive effect of alkyl substituents
In summary, the kinetic acidities of 1,4-dimethyl-
pyridinium iodide and 4-isopropyl-l-methylpyridinium
iodide were determined and it was found that the effect
of methyl substituents on the acidity was quite small.
This small effect was explained on the basis of the olefin-
like structure of the transition state for the proton
transfer reaction and the stabilizing effects of the
methyl groups on a double bond in contrast with the desta-
bilizing inductive effect of the methyl groups on a negative
charge. It would be of interest to have the equilibrium
acidities of these two molecules measured so as to con-
clusively establish the role played by the methyl groups
in determining the stability and structure of the conjugate
bases of these two molecules.
ALKYL GROUP HYDROGEN-DEUTERIUM EXCHANGE IN 1,3,6-TRIMETHYL-
AND 3,6-DIISOPROPYL-1-METHYL-PYRIDAZINIUM IODIDES
The kinetics of H-D exchange in the 6-alkyl group of
1,3,6-trimethylpyridazinium iodide (V) and 3,6-diisopropyl-
1-methylpyridazinium iodide (VI) in buffered D20 solutions
of 1.0 M ionic strength were compared at 75.0 + 0.10. The
exchange reactions were followed by measuring the change
in the integrated areas of the appropriate nmr signals.
CHs C H(C H)
+ CH C 3
SCH (C H3)
In order to approximate their reactivity difference,
the two compounds were dissolved in bicarbonate-carbonate
buffer solutions and compared by nmr. In a solution of
0.42 M in bicarbonate ion and 0.20 M in carbonate ion,
the acidity of the 6-methyl group of V was sufficient to
cause immediate and complete exchange simply on mixing.
No signal for the 6-methyl group could be observed in the
nmr. The acidity of the 6-isopropyl group of VI was much
less and the exchange reaction in a 0.10 M bicarbonate -
0.15 M carbonate buffer solution, pD = 10.75, had a half-
life of approximately 200 minutes.
The two compounds were next compared in a 0.075 M
D2PO4 0.15 M DPO, buffer solution, pD = 7.34. In
this buffer solution, the 6-methyl group of V was less
reactive but not dramatically so. Exchange could be
observed taking place in the nmr probe at 350 during the
course of recording the spectrum. The 6-isopropyl group
reactivity of VI had decreased to the point where the
exchange reaction at 75.00 would no longer take place at
a convenient rate.
In an effort to approximate the 6-methyl group reac-
tivity, compound V was next studied in a 0.25 M formic
acid 0.25 M format ion buffer, pD = 4.15. In this buffer
at 75.00, the exchange reaction proceeded with a half-
life of approximately 300 minutes.
Assuming the exchange reaction to proceed by specific
base catalysis, the reactivity ratio of the two compounds
can be calculated from the buffer pD's and reaction half-
lives. Treating the kinetic data in this fashion, it is
found that the 6-methyl group of 1,3,6-trimethylpyridazinium
iodide is more acidic than the 6-isopropyl group of 3,6-
diisopropyl-l-methylpyridazinium iodide by a factor of 3 x 10.
Obviously, if the reaction is general base catalyzed
and catalysis by the buffer base makes a substantial
contribution to the exchange rate of the methyl compound,
the above acidity ratio would be decreased. It seems
reasonable to assume, however, that this ratio would not
be reduced to a point of decreased significance.
Although a thorough investigation has not been con-
ducted, it would appear the primary reason for this large
difference is steric inhibition of resonance. The
negative charge on carbon in the transition state for
deprotonation can be delocalized in the case of the methyl
compound by donation to the ring and the positively
charged nitrogen. Such resonance stabilization is shown
for the intermediate resulting from deprotonation.
CH C H
In the isopropyl compound, the steric interaction
of the ortho methyl and isopropyl groups prevents the
isopropyl group from achieving the geometry necessary
for maximum orbital overlap and charge delocalization.
Again the intermediate is shown.
C H(C H3)
HC C \CH3
C H(C Ha),
Nuclear magnetic resonance spectra were recorded on
a Varian Associates Model A-60A instrument. Melting
points were obtained with a Thomas-Hoover Unimelt melting
point apparatus. Measurements of pD were determined on a
Beckman Model 1019 Research pH meter equipped with a
Corning (476050) semi-micro combination electrode. Both
measurements of pD and kinetic runs were carried out in
a Lauda/Brinkmann Model K-2/R constant temperature
All common laboratory chemicals, unless specified
to the contrary, were reagent grade and from various
suppliers. Deuterium oxide (99.8 percent) was obtained
from Columbia Organic Chemicals.
Stock solutions of dilute DC1 were prepared by diluting
:,cnce:ntrated HC' with DC and standardized by potentiometric
t tr ti on is I I standard 7ized NaCH.
Stock soi'.uc :;ns of dilitle potassium deuteroxide were
prepared by dissolving weighed quantities of reagent grade
KOH in D20. The solutions were standardized by potentio-
metric titration using primary standard grade potassium
Aldrich Chemical Company gold label grade 2,2,2-
trifluoroethanol and Mallinckrodt reagent grade sodium
acetate were used directly. Pyridine obtained from
Mallinckrodt Chemical Works, was dried over sodium and
distilled from zinc powder (bp 114-116; lit51 115.50).
Eastman Organic Chemicals 2,6-dimethylpyridine was likewise
dried over sodium and distilled from zinc powder (bp 142-
1430; lit51 1430). Imidazole, purchased from Matheson
Coleman and Bell, was recrystallized from hexane (mp
89-91; lit5 900). Reilly Tar and Chemical Corp.
4-aminopyridine was purified by vacuum sublimation and
recrystallized from benzene (mp 157-160; lit51 1580).
The method of Evans and Brown2 was used to prepare
4-amino-2,6-dimethylpyridine which was purified by
successive vacuum sublimations (mp 190-1910; lit52 191-192 ).
The purification of phenol was accomplished by adding benzene
to phenol, liquified reagent, obtained from Matheson
Coleman and Bell, and distilling first a benzene-water
azeotrope, then excess benzene and finally the phenol at
reduced pressure and under nitrogen (bp 900/25 torr; lit53
900/25 torr). Calcium hydroxide was prepared according
to the procedure of Bates, Boi r, and Smith by heating
well-washed calcium carbonate in a platinum crucible at
approximately 10000 C with a Meeker burner for one-hour
intervals until a constant weight is obtained. The
freshly prepared oxide was then slowly added to water, the
solution heated to boiling, cooled and filtered. The solid
was then oven dried and crushed to a finely granular state
for use. Calcium deuteroxide was prepared by dissolving
calcium hydroxide in D20.
1,4-Dimethylpyridinium iodide.--4-Picoline, obtained
from Matheson Coleman and Bell, was distilled from zinc
powder and dissolved in methanol. Methyl iodide was slowly
added. The mixture was then refluxed for one hour, evap-
orated on a rotovap and the resultant solid recrystallized
from absolute ethanol (mp 153-154; lit 153-153.80). The
compound was stored under vacuum.
4-Ethyl-1-methylpyridinium iodide.--The compound was
prepared by dissolving freshly distilled 4-ethylpyridine,
obtained from Aldrich Chemical Company, in ethanol, adding
methyl iodide and refluxing one hour. The resulting salt
was recrystallized from an ethanol-ethyl acetate mixture
(mp 1090; lit54 109-1100). The compound was stored under
4-Isopropyl-l-methylpyridinium iodide.--The liquid
4 -isopropyI!pyridine, purchased from K and K Laboratories,
Inc., was first fractionally distilled, the portion dis-
tilling at 181-1820 collected. The distillate was then
dissolved in methanol and treated as above with methyl
iodide. After reflux the solution was cooled and the excess
methyl iodide and methanol were evaporated with as little
heating as possible to facilitate the evaporation. The
crude salt was then recrystallized by dissolving in excess
ethanol at room temperature and then inducing crystalliza-
tion by slowly adding small portions of ethyl ether
(mp 125.5-128.50; lit 117-1200 dec). The compound was
stored under vacuum. Analysis: Calcd. for C9H14NI:
C, 41.08; H, 5.36; N, 5.32; I, 48.23. Found: C, 41.09;
H, 5.38; N, 5.28; I, 48.24.
1,3,6-Trimethylpyridazinium iodide.--The procedure
of Overberger, Byrd, and Mesrobian was followed for the
synthesis of 3,6-dimethylpyridazine. The pyridazine and
methyl iodide were added together neat and the resulting
solid recrystallized from acetone (mp 119.5-120.50; lit
3,6-Di i sopropyl --methylpyridazini um iodide.--The
compound 3,6-diisopropylpyridazine (mp 75-76.5) was
obtained from White. The pyridazine was dissolved in
methyl iodide and gently refluxed for two hours. The
resultant salt was recrystallized from acetone/ether
Preparation of Solutions
Either three or ten milliliters of solution were
prepared for each run in an appropriate sized volumetric
Substrate was weighed on an analytical balance along
with a corresponding molar amount of internal standard and
transferred to the volumetric flask.
Accurate volumes of stock acid or base were delivered
by means of Hamilton Microliter syringes.
Depending on the intended concentration, solubility,
or physical characteristics, the appropriate nucleophile
was either accurately weighed on an analytical balance
and transferred directly to the flask or first dissolved
in D20 to make a stock solution from which the proper
volume was then withdrawn and transferred by syringe. For
the 4-amino-2,6-dimethylpyridine runs, not only was it
necessary to make a D20 solution first, but it was also
necessary to add an equivalent of DC1 to get the solid
to dissolve easily. The DC1 was then later neutralized
with KOD solution.
Weighed amounts of potassium chloride were added to
each flask to obtain an ionic strength of 1.00 M and the
solutions were finally diluted to mark with D20.
Kinetic Procedure for H-D Exchange
Kinetics were obtained by two methods. The first
method, by which a majority of the work was done, involved
the use of three milliliters of solution. Upon completion
of the solution preparation, approximately one milliliter
was withdrawn and transferred to an nmr tube which was
then flushed with nitrogen and sealed. The remainder of
the solution was stored in the flask, under nitrogen, for
later comparison and pD measurements.
A proton nmr spectrum of the solution in the sealed
tube was recorded. The nmr tube was then immersed in a
constant-temperature circulating bath which had been
previously set at the desired temperature using a National
Bureau of Standards Certified thermometer. Periodically,
the nmr tube was removed from the bath, immediately
quenched by immersion in ice water, and the proton nmr
spectrum of the solution was recorded.
In mixtures with a high deuteroxide concentration,
it was apparent that the temperature of the nmr probe
was sufficient to maintain the exchange reaction. For
these cases, a second method was employed. From ten
milliliters of stock solution, two milliliters were with-
drawn by syringe and stored, under nitrogen, for pD
measurements. The remaining solution was immersed in the
constant-temperature bath in a 10 ml volumetric flask
that was fitted with a rubber septum. Periodically,
0.9 ml of solution was withdrawn by syringe and injected
into a test tube containing 0.1 ml of a 1.2 M DC1 quench
solution. This neutralized solution was then transferred
to an nmr tube and its proton nmr spectrum recorded.
Reactions were followed a minimum of 1.5 half-lives
by measuring the change in the integrated area of the nmr
signal of the proton(s) of interest with respect to that
of a non-exchanging proton in the reaction mixture. The
integrals of proton signals were measured in a minimum
of five successive sweeps and the average value was
In practically all the runs, an internal standard
external to the substrate was used. For runs involving
the 1,4-dimethyipyridinium iodide, tetramethylammonium
bromide was added as an internal standard. If 4-isopropyl-
1-methylpyridinium iodide were present, it was necessary,
due to peak overlap, to change to sodium acetate as an
internal standard. Although acetate ion was not an ideal
standard since it does promote exchange, it does so
slowly and once a rate constant was obtained for this
exchange, it could be easily calculated out of the par-
ticular reaction kinetics.
In the case of 2,6-dimethylpyridine, it was found
necessary to use the ring protons of the substrate as an
internal standard since catalysis by acetate ion was
greater than by 2,6-dimethylpyridine. Substrate ring
protons were also used as internal standards for all
runs involving 3,6-dimethyl- and 3,6-diisopropyl-l-
methyl-pyri daz; nium i odide.
For each kinetic run a plot was made of the quantity
[log(A/Astd)t-log(A/Astd)t] versus time where (A/Astd)t
is the ratio of the integrated area of the reacting
proton(s) to the integrated area of the internal standard at
a given time, t, and (A/Astd)t is the ratio of the two
areas at the start of the run, i.e., to. A pseudo-first-
order rate constant was then calculated from each plot
by visually fitting the best straight line through the
points and applying equation 16.
k = tl t2 (16)
Measurements of pD were performed on all solutions
employed in the various kinetic runs. NBS standard
buffers were prepared as described by Bates.58
For pD measurements at 75.0 0.10, the electrode
was first allowed to equilibrate in 4 M KC1 at 75.0 + 0.10
for a minimum of 20 minutes. The meter was then standard-
ized at pH 6 852 against the NBS phosphate buffer by
adjusting the standardization control on the meter.5s
When the pD of an alkaline solution was being measured,
the meter was linearized at pH 8.905 against an NBS
borax buffer by adjusting the temperature control on the
meter.5 When the sample solution being measured was
acidic, the meter was linearized at pH 4.145 against an
NBS phthalate buffer. Standardization and pD measurements
were carried out without allowing the electrode to cool
by rinsing and storing of the electrode in distilled
water at 75.0 0.10 between actual measurements.
For pD measurements at 50.0 0.10, the procedure
was exactly the same with the meter being standardized
against an NBS phosphate buffer value of pH 6.833 and
linearized against a borax buffer value of pH 9.011.58
No acidic pD values were measured at this temperature.
At 250, no temperature equilibration was necessary
for the electrode. Once again phosphate (pH 6.865) and
borax (pH 9.180) buffers were used for standardization
and linearization, no acidic pD measurements being made.58
Since the pH meter was standardized and linearized
against standard proteo buffers, it was necessary to add
a correction to the meter readings obtained for the various
samples to arrive at accurate pD values. For pD measure-
ments at 250,-the pD value is reported by Bates to be
obtained by adding 0.41 to the meter reading. For pD
measurements at 750, this correction factor is reported
to be 0.35.60 For pD measurements at 500, a value of
0.38 is obtained by simple interpolation for the above
The concentration of deuteroxide ion was calculated
using the relationship pOD = pKw pD. The values used
for pKw the dissociation constant for deuterium oxide,
as well as the factors for converting pH meter readings
to pD, may be found in Table 11.
Table 11. Dissociation Constants for D20 and pH to pD
Conversion Factors at Several Temperatures.
T, C pKwD pHpD
25 14.869 0.41 9
50 14.103 0.38b
75 13.526 0.35
aThese values are uncorrected for salt effects which
are expected to be small.
cCalculated from reported data.89
Values for the respective buffer pKadeterminations
were obtained from the pD measurements by the formula
pKa = pD + log[BD+]
Although the presence of an internal standard,
agreement of pD measurements on original and recovered
solutions, and the linearity of the pseudo-first-order
kinetic plots indicated the absence of important compli-
cating factors, control runs were carried out to determine
the stability of both the 4-methyl- and 4-isopropyl-
pyridinium iodides under various conditions.
The two pyridinium iodides were first dissolved
in 0.10 M DC1 solutions with an acetic acid internal
standard and heated at 750 to determine their stability
and, if possible, measure any exchange catalyzed by
D20 acting as the buffer base. No exchange, as evidenced
by the broadening of the 4-methyl singlet or the emerging
of a singlet between the 4-isopropyl gem.-dimethyl doublet,
could be detected in the nmr. These nmr spectral changes
are a more sensitive indication of initial deuterium
substitution than change in the integral ratios.
The solutions were heated until the change in the
integral ratios of substrate to internal standard reached
10 percent. In neither case were there observed the
above-mentioned spectral changes indicative of exchange.
The appearance of a precipitate was also noted in both
solutions. The change in the integral ratio was, therefore,
attributed totally to degradation of substrate. For the
4-methyl compound, heating for a period of seven days
produced the 10 percent degradation while for the 4-iso-
propyl compound, heating for a period of fourteen days
was required to produce this same percent change.
Proteo control runs were then carried out in three
different buffers to verify the stability of the two
pyridinium iodides in basic solution. Previously used
buffer solutions were duplicated using H20 in place of
D,0 and the mixtures were heated for the equivalent of
ten half-lives. For each buffer, the most basic con-
ditions previously employed were the conditions duplicated
for the control runs. Although kinetic runs were never
carried out with the 4-isopropyl compound in 4-amino-2,6-
dimethylpyridine buffer, a control run using this buffer
was carried out for comparison purposes. Details of
these control runs, the solution compositions, heating
times, and observed pH changes are contained in Table 12.
Degradation of substrate as measured by loss of the
signal for the 4-alkyl group relative to the signal of
acetate ion internal standard, was less than 10 percent
in all cases. The pH changes were also small (.035 or
less) for all but the 2,2,2-trifluoroethanol buffer.
For this buffer, the pH change was substantially larger,
C.) u -
being 0.196 for the 4-methyl compound, and 0.728 for the
4-isopropyl compound. In addition, the nmr spectra of
these solutions, although indicating less than 10 percent
change in the integral ratio values, also showed unidenti-
fied peaks similar to and emerging 10 to 20 Hertz downfield
from the expected nmr signals. As a result, the values
of the rate constants obtained using this buffer are
1. A. Streitwieser, Jr. and J. H. Hammons, Prog. Phys.
Org. Chem., 3, 41 (1965).
2. D. J. Cram, "Fundamentals of Carbanion Chemistry,"
Academic Press, New York, 1965, Chapter 1.
3. A. I. Shatenshtein, Advan. Phys. Org. Chem., 1, 155
4. H. Shechter, M. J. Collis, R. Dessy, Y. Okuzumi,
and A. Chen, J. Amer. Chem. Soc., 84, 2905 (1962).
5. R. P. Bell and D. M. Goodall, Proc. Roy. Soc., [A],
294, 273 (1966).
6. 0. Reitz, Z. Phys. Chem., [A], 176, 363 (1936).
7. R. G. Bates, V. E. Bower, and E. R. Smith, J. Res.
Nat. Bur. Std., 56, 305 (1956) R. P. 2680.
8. A. K. Covington, R. A. Robinson, and R. G. Bates,
J. Phys. Chem., 70, 3820 (1966).
9. G. S. Kell, J. Chem. Eng. Data, 12, 66 (1967).
10. J. F. Bunnett, "Technique of Organic Chemistry," VIII-1,
Interscience Publishers, New York, 1961, pp. 200-201.
11. R. G. Pearson and R. L. Dillon, J. Amer. Chem. Soc.,
75, 2439 (1953). -
12. J. A. Zoltewicz and L. S. Helmick, J. Org. Chem.,
38, 658 (1973).
13. W. Schneider, K. Gaertner, and A. Jordan, Ber., 57,
14. G. E. Tschitschibabin and S. W. Benewolenskaya, Ber.,
61, 547 (1928).
15. J. Hine and J. Murders, J. Org. Chem., 32, 2200
16. J. Hine, J. G. Houston, J. H. Jensen, and J. Mulders,
J. Amer. Chem. Soc., 87, 5050 (1965).
17. M. L. Bender and A. Williams, J. Amer. Chem. Soc.,
88, 2502 (1966).
18. R. P. Bell, "Acid-Base Catalysis," Oxford University
Press, London, 1941, pp. 91-95.
19. R. P. Bell, "The Proton in Chemistry,
University Press, Ithaca, N. Y., 1959,
20. M. Eigen, Angew. Chem. Internat. Ed., 3, 1 (1964).
21. C. D. Gutsche, D. Redmore, R. S. Buriks, K. Nowotny,
H. Grassner, and C. W. Armbruster, J. Amer. Chem.
Soc., 89, 1235 (1967).
22. E. S.
Lewis and L. H. Funderburk, J. Amer. Chem.
89, 2322 (1967).
23. T. A. Spencer,
J. Amer. Chem.
24. D. J. Hupe, M.
J. Amer. Chem.
25. A. Streitwieser, Jr.
Chem. Soc., 84, 249
26. J. E. Hofmann, R. J.
J. Amer. Chem. Soc.,
Kendal 1 ,
and D. E.
1, and I. D. Reingold,
and T. A.
Van Sickle, J. Amer.
ler, and A. Schriesheim,
I. Shatenshtein and H. A. Gvozdeva, Tetrahedron,
28. F. G. Bordwell
Soc., 93, 6303
and M. D.
Wolfinger, J. Am. Chem.
29. J. Hine, L. G. Mahone, and C. L.
Chem. Soc., 89, 5911 (1967).
Liotta, J. Amer.
30. J. Hine and P. D. Dalsin, J. Amer. Chem. Soc., 94,
31. R. P. Bell and H. C.
32. R. P. Bell, G. R. Hil
Street, J. Chem. Soc.
Longuet-Higgins, J. Chem. Soc.,
J. W. Mansfield, and D.
33. F. G. Bordwell and J. Almy, J. Org. Chem., 38, 575
34. R. P. Bell and J. E. Crooks, Proc. Roy. Soc., [A],
286, 285 (1965).
35. D. Turnbull and S. Maron, J. Amer. Chem. Soc., 65,
36. G. W. Wheland and J. Farr, J. Amer. Chem. Soc., 65,
37. F. G. Bordwell, W. J. Boyle, Jr., and K. C. Yee,
J. Amer. Chem. Soc., 92, 5926 (1970).
38. F. G. Bordwell and W. J. Boyle, Jr., J. Amer. Chem.
Soc., 93, 511 (1971).
39. F. G. Bordwell and W. J. Boyle, Jr., J. Amer. Chem.
Soc.,94, 3907 (1972).
40. F. Hibbert, F. A. Long, and E. A. Walters, J. Amer.
Chem. Soc., 93, 2829 (1971).
41. F. G. Bordwell, R. H. Imes, and E. C. Steiner, J.
Amer. Chem. Soc., 89, 3905 (1967).
42. M. J. Cook, A. R. Katritzky, P. Linda, and R. D. Tack,
J. Chem. Soc., 1295 (1972).
43. J. A. Zoltewicz and V. W. Cantwell, unpublished
44. J. A. Ber.son, E. M. Evleth, Jr., and Z. Hamlet, J.
Amer. Chem. Soc., 87, 2887 (1965).
45. K. Bowden, A. F. Cockerill, and J. R. Gilbert, J.
Chem. Soc., [B], 179 (1970).
46. C. D. Ritchie and R. E. Uschold, J. Amer. Chem. Soc.,
89, 1721 (1967).
47. A. G. Evans, M. A. Hamid, and N. H. Rees, J. Chem.
Soc., [B], 2164 (1971).
48. A. Streitwieser, Jr., C. J. Chang, and D. M. E. Reuben,
J. Amer. Chem. Soc., 94, 5730 (1972).
49. A. Streitwieser, Jr. P. C. Mowery, and W. R. Young,
Tetrahedron Letters, 3931 (1971).
50. L. P. Hammett, "Physical Organic Chemistry," 2nd Ed.,
McGraw-Hill, New York, 1970.
51. "Handbook of Chemistry and Physics," 47th Ed., The
Chemical Rubber Publishing Co., Inc., Cleveland, 1966.
52. R. F. Evans and H. C. Brown, J. Org. Chem., 27,
53. "Dictionary of Organic Compounds," 4th Ed., Oxford
University Press, New York, 1965.
54. E. M. Kosower and J. A. Skorcz, J. Amer. Chem. Soc.,
82, 2195 (1960).
55. C. G. Overberger, N. R. Byrd, and R. B. Mesrobian,
J. Amer. Chem. Soc., 78, 1961 (1956).
56. R. L. Letsinger and R. Lasco, J. Org. Chem., 21, 764
57. R. M. White, Ph.D. Dissertation, University of Florida,
1972. Receipt of this compound from Dr. White is
58. R. Bates, "Determination of pH. Theory and Practice,"
John Wiley and Sons, Inc., New York, 1964.
59. A. K. Covington, M. Paabo, R. A. Robinson, and R. G.
Bates, Anal. Chem., 40, 700 (1968).
60. R. E. Cross, Ph.D. Dissertation, University of Florida,
61. H. S. Harned and B. B. Owen, "The Physical Chemistry
of Electrolytic Solutions," 3rd Ed., Reinhold, New
PREVIOUSLY PUBLISHED INVESTIGATIONS
The material contained in this section has been
separated from the main text as it has already been
published. It consists of two parts. The first,
"Convenient Preparations of Mono- and Dideuterated
2-Furoic and 2-Thiophenecarboxylic Acids," has been
published in the Journal of Heterocyclic Chemistry and
is presented here exactly as it appears in the litera-
ture. The second part, "Nucleophilicities of Compounds
with Interacting Electron Pairs. Diazine-Catalyzed
Ester Hydrolysis," was published in Tetrahedron Letters.
In as much as Tetrahedron Letters does not publish
experimental sections, the original article will be
presented along with experimental details.
Convenient Preparations of Mono-
and Dideuterated 2-Furoic and 2-Thiophenecarboxylic Acids*
2-Furoic la and 2-thiophenecarboxylic lb acids and
their derivatives are useful starting materials for the
preparation of furans and thiophenes containing various
side-chains, including compounds of biological interest.
We wish to report convenient and very simple prepara-
tions of mono- and dideuterated forms of these two acids.
Deuterium labeling was achieved by hydrogen exchange
reactions either at position 5(2) or at positions 3, 5 (3)
of each acid (Table 1).
D COOH \COOH D COOH
2 1 3
The following conditions were found to be optimum for
monodeuteration (method A). The appropriate carboxylic
acid was heated at 1650 in a deuterium oxide-carbonate
buffer (pD -10). The monodeuterated product 2 was
obtained on cooling and acidifying the reaction mixture.
Published in the Journal of Heterocyclic Chemistry, 8,
Table 1. Deuterated 2-Furoic and 2-Thiophenecarboxylic
Acids Prepared by Hydrogen-Deuterium Exchangea
Acid Method T, C
Time %3-D %5-D Acid
-- 95 52
250 45 min. 19 28 26
2 hr. 32 32 63
aNmr analyses of percent deuteration have about a 3%
uncertainty, H-4 being used as an internal standard.
b>90%D in the COOD group initially.
Nmr analyses revealed that in each case the H-5 signal
of the acid had almost completely disappeared (>95
percent D); the remainder of the spectrum was that of
a simple AB system. Note that the chemical shift order
(decreasing T values) for la is H-4>H-3>H-5 but for
lb it is H-4>H-5>H-3. Mass spectral analysis indi-
cated the formation of less than 6 percent dideuterated
acid. While it was not possible to determine clearly
the position of the second deuterium atom, the results
given below suggest that it is position 3.
Deuteration at the 3,5 positions was conveniently
effected by heating the dry acid containing the COOD
group at 2500 (method B). Deuterium was introduced
into the carboxyl group by recrystallizing 1 from
deuterium oxide. The amount of deuterium in the carboxyl
group was determined by nmr analysis of a methylene
chloride solution. In the case of 3b the amount of deu-
terium introduced into the 3,5 positions was that expected
for a statistical distribution of deuterium among these
two positions and the carboxyl group. Deuteration was
not statistical in the case of 3a, the 5-position under-
went more exchange than the 3-position. Statistical
distribution of deuterium was not observed since a
shorter reaction period was necessary due to the exten-
sive decarboxylation of la at the temperature employed.
Although higher degrees of deuteration could be
achieved in method B, no attempt was made in this
direction. By relabeling the carboxyl material, addi-
tional hydrogen-deuterium exchange would result.
Materials.--2-Furoic acid (la), m.p. 133-134,
Matheson Coleman and Bell and thiophene-2-carboxylic
acid (lb) m.p. 127-128 (Aldrich Chemical Co.) were
used as received. Deuterium oxide (>99 percent) was
supplied by Columbia Organic Chemicals Company. A
Parr Instrument Company Monel Bomb was employed.
Method A. Exchange at H-5.--Deuterium oxide (16 ml.)
was added to an equimolar mixture of 0.008 M of la or lb
and sodium carbonate. The solution having pD-10 was
heated in a bomb at 1650. After cooling, the reaction
mixture was acidified with dilute hydrochloric acid and
the precipitate was collected. Recrystallization from
proteo water gave the corresponding carboxylic acid-5-A.
Nmr analyses were obtained on methylene chloride solu-
tions. Results are summarized in Table 1. Mass spectral
analysis of 2a showed d0=8.0 percent, d1=86.2 percent,
and d2=5.8 percent; 2b showed do=3.6 percent, d,=95.0
percent, and d,2=.4 percent.
Method B. Exchange at H-3,5.--2-Furoic acid-0-d
or 2-thiophenecarboxylic acid-0-d (2.0 g) was heated in
a bomb at 2500. The product obtained from the cooled
bomb was dissolved in methylene chloride for nmr analysis.
Prior to nmr analysis of 2a, the solid was gently warmed
to remove furan formed by decarboxylation. Results are
given in Table 1. The dideuterated products were
recrystallized from proteo water before mass spectral
analysis: 3a showed d0=59.9 percent, di=34.6 percent,
d2=5.5 percent; 3b showed d0=46.2 percent, d1=43.6 per-
cent, d2=10.2 percent.
(1) A. P. Dunlop and F. N. Peters, "The Furans,"
Reinhold Publishing Corp., New York, N. Y.,
(2) S. Gronowitz, Advan. Heterocycl. Chem., 1, 2
(3) Sadtler Standard Spectra, NMR No. 633M, Sadtler
Research Laboratories, Inc., Philadelphia, Penn.
(4) Ibid., NMR No. 523M.
Nucleophilicities of Compounds with Interacting Electron
Pairs. Diazine-Catalyzed Ester HydrolysisA
Pair-pair electron repulsion has been suggested to
be an important factor responsible for the abnormally
high reactivity of nucleophiles such as ROO- toward some
electrophiles. Recently, it has been suggested that
widely separated electron pairs may interact strongly.
Thus, molecular orbital calculations and photoelectron
spectroscopy indicate that the unshared electron
pairs of the diazines pyridazine (I), pyrimidine (II)
and pyrazine (III) interact strongly. Interactions are
transmitted both through space and through bonds.
I II III IV
This recent evidence for electron pair repulsion
prompted us to determine whether the diazines and a
benzolog, phthalazine (IV), would show an enhanced
reactivity toward 2,4-dinitrophenyl acetate (DNPA);
these compounds are expected to act as nucleophilic
catalysts for the hydrolysis of this ester. A
* Published in part in Tetrahedron Letters, 189 (1972).
representative hydrolysis pathway is shown in Scheme 1.
This ester was selected for study because it was expected
to react with the compounds of interest at convenient
rates and because it is known to show large rate enhance-
ments in its reactions with nucleophiles such as ROO .
|+ CH3COOAr "+ArO
+ CH3COOH + H
cIScoN + ArO~
+ CH3COOH + H+
The approach adopted is a standard one. The reac-
tivities of I-IV were estimated from their pKa values
using an established Bronsted reactivity-basicity
correlation. The estimated reactivities then were
compared with experimental reactivities obtained under
similar experimental conditions. Differences between
observed and estimated nucleophilicities provide a
measure of rate enhancements.
The reference Brdnsted correlation was established
using known rate constants for the reactions of DNPA
with 4-methylpyridine (V), pyridine (VI) and nicotinamide
(VII) in water at 25.00 and 1.0 M ionic strength. (In
order to check this method, the reactivity of nicotinamide
toward DNPA was determined. The second-order rate con-
stant obtained is only 6 percent less than the reported
The reactivities of I-IV toward DNPA at 25.00 were
measured spectrophotometrically at 400 nm in 1:1 acetic
acid-acetate ion buffers (3.2-20 x 10- M, total buffer)
maintained at 1.0 M ionic strength with KC1. The con-
centration of DNPA was varied over the range 1-15 x 10 M.
Pseudo-first-order rate plots were linear over at least
4 half-lives and second-order rate constants, k2, were not
dependent on the initial concentration of DNPA, showing
that the reverse of the first step in Scheme 1 is
kinetically unimportant. Rate constants were calculated
according to equation 1.
k = k [B] Ka + 3.4 x 10 [CHCO2
k[t [H] + Ka
+ 1.2 x 10 [H20] (1)
Corrections for acetate ion and water catalyzed ester
hydrolyses were made using known rate constants; they
were < 13 percent of kp. The concentration of nucleophile
in the free base form was calculated from a knowledge
of the total concentration of nucleophile, [B]t, its Ka
and a measured pH. Titrations were used to obtain pKa
values for I and IV at 25.00 and 1.0 M ionic strength;
values for II and III are taken from the literature.
Results are summarized in Table 1.
I I -
x x x
0 0 0
<~D 0 00 ^t-
O CO C
+1 +1 +1 +1
LO :- co 0
m co Q0 C)
;Z;- c\j ifl
S^- ro I'
Figure 1 shows the Br0nsted plot of nucleophilic
reactivity versus pKa established by pyridine nucleophiles.
The results for diazines II and III lie on this line and
do not show an enhanced reactivity toward DNPA. But
diazines I and IV show rate enhancements by a factor of
12. (Rate and equilibrium constants for the diazines are
statistically corrected to reflect reaction at two equiva-
lent nitrogen atoms, i.e., k2/2 and 2Ka are used in
It is clear from our results that I and IV can show
reactivities which exceed those predicted by their basi-
cities but it is curious that no special nucleophilicities
are found for II and III. It will be of interest to
determine whether rate enhancements can be demonstrated
for II and III toward other electrophiles.
Instrumentation. Ultra-violet absorption spectra
were obtained on a Zeiss Model PMQ II spectrophotometer.
Constant temperature in the cell holder was maintained by
connection to a Lauda/Brinkman Model K-2/R constant
temperature circulator. Temperature in the cuvettes was
checked by an NBS certified thermometer and found to be
+ 0.50. Measurements of pH were determined on a Beckman
Model 1019 Research pH meter equipped with a Corning
(476050) semi-micro combination electrode. Melting points
I I I I I I
1 2 3 4 5 6 pKa
Figure 1. Bronsted plot of pKa versus log k2 for diazines
I-IV and pyridines V-VII reacting with DNPA.
were obtained with a Thomas-Hoover Unimelt melting point
apparatus and are uncorrected.
Chemicals. All heterocycles used as nucleophiles
were commercially available from various suppliers and
were used as received with the exception of phthalazine
which was first recrystallized from ether mp 900 (lit
mp 90-91). The procedure of Bender and Nakamura was
used for the synthesis of 2,4-dinitrophenyl acetate.
The ester was recrystallized from ethyl acetate/petroleum
ether, mp 70.5-71.5 (lit mp 720). All common laboratory
chemicals were reagent grade and were obtained from vari-
Kinetics of Acetylation of 2,4-Dinitrophenyl Acetate.
Pseudo-first-order rate constants, k for reactions
between nitrogen heterocycles and 2,4-dinitrophenyl
acetate in aqueous solution at 25.00 and 1.0 M ionic
strength were obtained by monitoring the formation of
2,4-dinitrophenol at 400 nm in the ultra-violet spectrum.
Buffer solutions were of two types. One type, used
for phthalazine and pyridazine, consisted of a 2:1 molar
mixture of heterocycle and HC1 in a solution made 1.0 M
in ionic strength by the addition of KC1. For the weaker
bases pyrimidine and pyrazine, and in some instances
phthalazine and pyridazine, solutions consisted of
heterocycle in a 1:1 acetic acid-acetate ion buffer
solution that was also made 1.0 M in ionic strength by
the addition of KC1. The KC1 solutions, with and without
acetate buffer, were also employed as optical blanks.
Reactions were initiated by syringing a measured
amount of the ester, in a water-acetonitrile mixture
(4:1 by volume), into a cuvette thermostated inside the
The observed rate constant, kp, was experimentally
determined by applying the equation
[A -A ]
kpt = 2.303 log o-Q
-log [Ao-A] = k2 t log [A,-Ao
S t 2.303
where A. is the absorbance at infinite time, At is the
absorbance at.any time, t, and A is the absorbance at
time zero. The observed rate constant, kt, is obtained
merely by plotting -log[Ao-At] versus t, the slope of the
line being k/2.303.
The observed rate constant was corrected to eliminate
reaction by water and acetate ion by multiplying the
known second-order rate constants for both water and
acetate ion by their respective concentrations and
subtracting from the experimentally determined rate
Measurements of pH were made on all solutions and
the concentration of unprotonated heterocycle was
calculated by multiplying the total concentration of
heterocycle in solution by the fraction Ka/([H]+Ka).
Second-order rate constants were then obtained
by dividing the corrected ki by the concentration of
unprotonated heterocycle. (See equation 1.)
(1) K. M. Ibne-Rasa and J. 0. Edwards, J. Am. Cnem. Soc.,
84, 763 (1962); J. D. Aubort and R. F. Hudson, Chem.
Commun., 937 (1970); K. Tsuda, J. B. Louis and R. E.
Davis, Tetrahedron, 26, 4549 (1970).
(2) R. Hoffmann, Accts. Chem. Res., 4, 1 (1971).
(3) R. Gleiter, E. Heilbroner and V. Hornung, Angew.
Chem. Internat. Ed. Engl. 9, 901 (1970).
(4) For a summary of results dealing with photoelectron
spectroscopy see, S. D. Worley, Chem. Rev., 71, 295
(5) W. P. Jencks and M. Gilchrist, J. Am. Chem. Soc., 90,
(6) D. D. Perrin, "Dissociation Constants of Organic
Bases in Aqueous Solution," Butterworth and Co.,
(7) "Dictionary of Organic Compounds," 4th Ed., Oxford
University Press, New York, 1965.
(8) M. L. Bender and K. Nakamura, J. Am. Chem. Soc., 84,