Hydrogen labeling of some heteroaromatic carbon acids

Material Information

Hydrogen labeling of some heteroaromatic carbon acids
Jacobson, Harvey Lewis, 1946-
Publication Date:
Copyright Date:
Physical Description:
ix, 92 leaves. : illus. ; 28 cm.


Subjects / Keywords:
Acidity ( jstor )
Buffer storage ( jstor )
Buffer zones ( jstor )
Carbon ( jstor )
Iodides ( jstor )
Ions ( jstor )
Kinetics ( jstor )
Mathematical constants ( jstor )
Protons ( jstor )
Reactivity ( jstor )
Chemistry thesis Ph. D
Deuterium ( lcsh )
Dissertations, Academic -- Chemistry -- UF
Organic acids ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )


Thesis -- University of Florida.
Includes bibliographies.
Additional Physical Form:
Also available on World Wide Web
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000577609 ( AlephBibNum )
13995919 ( OCLC )
ADA5306 ( NOTIS )


This item has the following downloads:

Full Text






To my Father -

with respect, admiration
and love.


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.




LIST OF TABLES.....................................

LIST OF FIGURES.....................................

ABSTRACT ......................... .................


1. INTRODUCTION ...............................








Results ..................................

Preliminary Experiments .................

Deuteroxide Ion Catalysis...............

Catalysis by Other Buffers ..............

NMR Spectra of 4-Alkyl-1-methylpyri-
dinium Iodides in Liquid Ammonia ........

Discussion .................................

The Brdnsted Correlation.................

Transition State Structure ..............


4. EXPERIMENTAL ...............................

Instrumentation ............................

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


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


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
4-Isopropyl-1-methylpyridinium lodides
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


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



Harvey Lewis Jacobson

August, 1973

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
5 6
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.




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-


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,

Scheme 1.

Scheme 1

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-

tonation reaction.

k = k D[OD ] + kB[B] (2)

Preliminary Experiments

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

concentrate on.

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
Deuteroxide Ion.


Calcium Deuteroxide


pyri dine






-1 -1
koD,M sec

4-Methyl 4-Isopropyl

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
8 9
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-]
-14 (10)
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.


0 1

oC --

r--- r, o ,.- ,

0) i r o C- Co
*r- 0 r-- -- .- r0 r-
"o oc


r- CO LO r
*.. I CO C) CM CO C)

C 1 co 0 0 0 0l

I -- "- r-- r--



0 LnV n


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 :- ,

0n0-. 0C
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 -

E m

*I OJ O 0 -

0- 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-

U4- 0
+ 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

buffer runs.

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.
1 _I
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



E r-

r- X

E- I

E -

.- r0

X C)

1 4-

o 0
o co

*r a /) I

*-C 0X

\ O




o o a
4U c

\ 0 1 --a
a 4- 4-'

04- C)
o ( -


\) 0 C a2


Iz D-

L n3 Lii C -
\ 3-1
\ 3



rr- E

r--- 00

O C" I
S .,*-

0 ='-

C U)

-- "O

C 0
.) -E


C L1 4-)

S- U) 4-'
o- o c

i T

4-- "a
O 0 C

*- I 0

.r U

C *4- --

e- 0 C
\.. -


O 0


LC n -
CD C:)
zro i
rin +

\ o=
\ o-
\ 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

" 0
C) CO Co C)
0 "m



- I X X X X
>f3 --- I -- t) U3

-- o 0 0 00 0
>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 o



S- (U
4-- .r--


(o II IU I- cI

I < I X X X.

-U o N*
5.-a, 1

0 O0 0 0 0 0

.- aC xxxx

a) .- . Q

1- i^ M- n -




a) Lo)
S*- 4- rI


*r- CO r-
C -C

(0 Q"
s- -

o, 0

O *

r-- m

. S.

4- a)

aU 4- -

5- I

0- C-


01 <
o +-"

>J O

*4- w

I L 3
I( I

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
and 13.

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.

log kOD





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.

Table 5.

Thermodynamic Constants for 1,4-Dimethyl- and
4-Isopropyl-l-methylpyridinium lodides Reacting
with Deuteroxide Ion.

4-Methyl 4-Isopropyl

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-






Acetic Acid


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-









1 .59


1 .81


1 .00

1 .87






aUncorrected for deuteroxide ion catalysis and acetate ion

b[acetate] = 0.100 M.

C[acetate] = 0.090 M.

dpKa = pD + log [BD+]/[B].

Table 6.

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
_2 -44
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-


2,2,2-Tri fluoroethanol


Imidazol e


Acetic Acid

- 1a

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].










1 .39

1 .59


1 .81


1 .00

1 .00


X 10 -3
x 10
x 10
x 10
x 10
x 10
x 10
x 10
x 10


Table 7.

methylpyridinium Iodide in Selected Buffers at 75.00.

[BD+ ],M

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

pKa ,obsd









i -1
kB,M- sec

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



I -.



:- 0

I )


r-- o
- -

r-- )
o (1)





4- U


-- >,
C e

I- *r-'

Qo U


t/ (

o *o


0o 0o


L- L

rl- -)-



0 c
Ll .

I CJ 0
I (\J

S- CO.




S ( CO

C\ C\J

i I

-- .-- rn


Co L mc

co LO C\M





1 0

.- >1
E C-


- C\J C

,a- me


- >1
r- -U

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

r- c\j

0 0 0
r--- e- e-


Lnl C co

u3 C) 'z-


I a0















( IJ.

4- 0



(U J

c- Q

n3 j=
> +->





S r-
L >

3 O

5- D

a E
U S-

L .L
3 0

> a1
s.- (

(a )

S.- -e
-0 II


0 <1





r-- -
fL 0


0 l

. S-

a..- 4 --
I =3

K- cc


0- .<-




\ O'-r-

< Ln a4-

S .-

Oc o

\\ co (1 5

\\J \ ')- L
\0 -
\ \ c \ \ 'Q-
\ \ -^ ^
\ \ a '

d\\ LO

co \ \ <
0\<1 ,

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.





a, R=H
b, R=CH

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

















1 .87







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.

C,6HS C6Hs




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
significant deviation.
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
Br0nsted relationship.

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
2 4
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


c-C6H 1NH
at 500c,


at 300d


CsHs5CI2C! 3


aReference 3.

Deuterated substrate.

CReference 25.

dReference 26'.


of amide ion catalyzed dedeuteration to vary over four
powers of ten.


4 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


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
29 30
substituted carbon acid was observed. '

--3 2 3 / 3 2 2 3 H3 C

16.5 /

127 /

Similar results were obtained for deprotonation in
31 32
a series of alkyl ketones in aqueous hydroxide.

(CH3) 2CO / (CH 3CH ) CO / (CH 3) CH) CO

119 /

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


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



/ 1



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
5 6
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
35 36
indicate. '


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

+ N


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
40 41
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

inductive effect.

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

can dominate.

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.


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)
i- Ii

+ 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.



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)

S\C H3

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

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

hydrogen phthalate.


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.


Pyridinium iodides

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.

Pyridazinium iodides

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

(mp 156-1580).

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.

-2.3[log(A/Astd)t log(A/Astd)t
1 2
k = tl t2 (16)

pD Measurements

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

reported values.

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

8 59
25 14.869 0.41 9

50 14.103 0.38b
C 60
75 13.526 0.35

aThese values are uncorrected for salt effects which
are expected to be small.

bFrom interpolation.

cCalculated from reported data.89

Values for the respective buffer pKadeterminations

were obtained from the pD measurements by the formula

pKa = pD + log[BD+]

Control Runs

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,











0 c


Co 1


I -


C 0


cO I
C.) u -



" )












r C





C )



**r- C'.

4- O-
4 0

4- (

0) E
4-, 4-3

) Co
L- 0)

0) 0
3 U


C0 '
CO 7
5- O

(U O

3 -->



c *r-

4-) 4-)

S -)
-" 4-)

4- 3
4-) 4-)


O 3
4- 4-)



4- U
*r- X


o c

"j E



-o -

cn '4

) 3


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,
522 (1924).

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,

" Cornell
Chapter 10.

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.
Soc. ,

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.

M. C.

C. R.
Soc. ,

25. A. Streitwieser, Jr.
Chem. Soc., 84, 249

26. J. E. Hofmann, R. J.
J. Amer. Chem. Soc.,

27. A.

R. Kendal
94, 1250

Kendal 1 ,
94, 1254

and D. E.


1, and I. D. Reingold,

and T. A.


Van Sickle, J. Amer.

ler, and A. Schriesheim,
3002 (1963).

I. Shatenshtein and H. A. Gvozdeva, Tetrahedron,
,2749 (1969).

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,
6998 (1972).

31. R. P. Bell and H. C.
636 (1946).

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
(1 973).


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,
212 (1943).

36. G. W. Wheland and J. Farr, J. Amer. Chem. Soc., 65,
1433 (1943).

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,
1329 (1962).

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
gratefully acknowledged.

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
York, 1958.


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).


2 1 3

a, X=O
b, X=S

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,
331 (1971).

Table 1. Deuterated 2-Furoic and 2-Thiophenecarboxylic
Acids Prepared by Hydrogen-Deuterium Exchangea

Acid Method T, C

% Yield
Time %3-D %5-D Acid

6 hr.

-- 95 52

250 45 min. 19 28 26



5 hr.


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
2 4
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.

Hydrogen-Deuterium Exchange

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
3 4
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.


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-
5 ""
ments in its reactions with nucleophiles such as ROO .

Scheme 1

|+ CH3COOAr "+ArO

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

value. )

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.

Ka -2
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.

('4 CM4
I I -

x x x
0 0 0

<~D 0 00 ^t-

+1 +1 +1 +1

LO :- co 0
m co Q0 C)

;Z;- c\j ifl


4-4 -








Oz c

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

Figure 1.)

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



log k2








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-

ous suppliers.

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,
2622 (1968).

(6) D. D. Perrin, "Dissociation Constants of Organic
Bases in Aqueous Solution," Butterworth and Co.,
London, 1965.

(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,
2577 (1962).