-' -_ '“wh

VW-‘

 

 

 

KINETECS— OF THE THERMAL DECOMPOSI'HON
OF DIMETHYLMERCQRY

Thesis for the Deqru of M. S.
MICHEGAN STATE UNWERSITY

Kenneth Brant Yerrick
1962

' LIBRARY

Michigan Stat:
University

 

ABSTRACT

KINETICS OF THE THERMAL DECOMPOSITION
OF DIMETHYLMERCURY

by. Kenneth Brant Yerrick

Previous studies concerned withthe thermal decomposition of
dimethylmercury have produced disagreement as to the kinetic order of
the reaction and the mechanism by which it proceeds.

The method used to study the reaction for the present research
was a static system. Since the reaction was studied in the gas phase,

a conventional vacuum system was employed. The gaseous reaction
products were analyzed qualitatively by gas chromatography. The

- kinetic studies have been made for the temperature range 275-3300C
and for the concentration range 0,. 3 x 10'6 to 7.0 x 10"6 moles cc‘l.

On the basis of the experimental data obtained, the following empirical

rate expression has been postulated:

d(D)
dt

 

= k'a(D) + .kb(D)2

-In this, expression (D) is the concentration of dimethylmercury; 1%; and

kb are first and second order rate constants; respectively. These rate
constants have been evaluated by a method of successive approximations.
. Least squares analyses have yielded the following Arrhenius rate

equations:

w
II

a! 2.6 x109 exp (-39,400/RT),. (sec. -‘)

9.5 x1026 exp (-70, GOO/RT), (cc. mole'l sec. '1)

w
0.;
ll

Kenneth Brant Y errick

The gaseous reaction products analyzed by gas chromatography
were found-to be (excluding methane) ethane, ethylene, acetylene,
propane, propene, n-butane, l-butene and Z-methylpropene. ‘Some
preliminary studies of the carbonaceous deposit which is formed on
the reactor walls are discussed.

The lack of experimental data has prevented the postulation of a
plausible mechanism for the decomposition reaction. On the basis
of the observations made in this study, further experimental studies

have been propos ed.

 

 

KINETICS OF THE THERMAL DECOMPOSITION
OF DIMETHYLMERCURY

BY

Kenneth Brant Yerrick

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemistry

1962

 

To My Wife Carole

ii

To My Wife Carole

ii

To My Wife Carole

ii

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation to
Dr. Morley E. Russell for his guidance, assistance and encourage-
ment which he so generously gave throughout the investigation and
preparation of this thesis. Appreciation is gratefully extended to
the author's parents whose understanding and patience guided him
through his early years. Thanks is also given to Mr. F. E. Hood
for his assistance in the development of the apparatus; and to
Drs. C. H. Brubaker and R. B. Bernstein who kindly furnished
the dimethylmercury.

****>:<>§<>'.<******

iii

TABLE OF CONTENTS

Page

INTRODUCTION ......................... 1
I. Historical Background . ........... . . . . . 1

II. Objectives of Present Study .............. 7
EXPERIMENTAL ........................ 8
I.Apparatus.............. .......... 8

A. Vacuum System ................. 8

B. Reactor . . . . . ................ 9

C. Gas Chromatography . ......... . . . . 12

II. Materials . ....................... 12

A. Dimethylmercury ............. . . . 12

B. Carbon Dioxide .............. . . . 13
III.OperatingProcedure.................. 15

A. Typical Dimethylmercury Experiment ..... 15

B. Experiments with Carbon Dioxide . . ..... 16

IV. Analysis of Products ...... . ........ . . . 17
RESULTS ............................ . 18
I. Kinetic Data . ................... . . 18

II. Analysis of Products . . . . . . . .......... . 25
DISCUSSION OF RESULTS ................. . . . 27

I.KineticData....................... 27
A.RateExpression-................ 27
B. Comparison of Data of Other Investigations . . 32
II.ReactionProducts................... 37

A. Gaseous Products ........... . . . . . 37

B. Solid Product . ................. 38

III. Mechanism ....................... 38

IV. Further Experimental Studies ............. 38
BIBLIOGRAPHY ......................... 40
APPENDICES . ......................... 41

iv

TABLE

II.

III .

IV.

VI.

VII.

VIII.

X.

XI.

LIST OF TABLES

Temperature Profile of Reactor .......... . .
Thermal Decomposition of Dimethylmercury at 2750C.
Thermal Decomposition of Dimethylmercury at 2890C.
Thermal Decomposition of Dimethylmercury at 3030C.
Thermal Decomposition of Dimethylmercury at 3190C.
Thermal Decomposition of Dimethylmercury at 3300C.

Temperature Dependence of the Rate Constants ka
and kb 0 O O O O O O O O O O O ..... O O O O O O O O 0

Comparison ofdand ac at 3300C. . . . . . . . . . . .
Comparison of tc and t at 319°C ....... . . . . .

Summary of First Order Rate Data. . . . . . . . . . .

Comparison of ac and (1 at 342°C. . ........ . .

Page

11

19

19

20

20

21

24

29

30

33

36

LIST OF FIGURES

FIGURE

1.

2.

3.

4O

DetailsofReactor..............

Spectral Comparisons . . . . . . . . . . . .

Gas Sampling Bulb ....................

ArrheniusPlotforka . . . . . . . . . . . .

.ArrheniusPlotforkb . . . . . . . . . . . .

Dependence of First Order k on (Do) . . . .

. Gas Chromatograms of Gaseous Products . .

Effect of (Do) on Arrhenius Activation Energy for First

Order................... .

vi

Page
10
14
17
22
23
26

31

34

LIST OF FIGURES

FIGURE
1.DetailsofReactor....................
2.SpectralComparisons. . . . . . . . . . . . . . . . . .
3.GasSamplingBulb....... .....
4. Arrhenius Plot for ka .......... . . . . . . . .
5. Arrhenius Plot for kb ............ . . . . . .
6. Gas Chromatograms of Gaseous Products . . . . . . .

Dependence of First Order kon (Do) . . . . . . . . . .

Effect of (Do) on Arrhenius Activation Energy for First
Order.. ..........

vi

Page
10
14
17
22
23
26

31

34

APPENDIX

LIST OF APPENDIC ES

Page
Thermocouple Calibrations ........... . . 42
Derivation of Rate Expression ...... . . . . . 43
Evaluationofkaandkb............... 45

Relationship of ka and kb to First Order Rate
Expression ............. . ....... 46

vii

INTRODUCTION

I. Histo ric a1 Bac kg round

The interest in the decomposition of dimethylmercury has stemmed
from the extensive use of this compound as a source of methyl radicals
which are formed by the photolytic or thermal scission of the mercury-
carbon bond. A review (to 1953) of the photolytic decomposition appears
in a monograph by' Steacie (1). The most important steps of various

proposed mechanisms are thought to be:

Hg(CH3)z-}l-"—> HgCH3- + 01-13- _1_
HgCH3r ~——> Hg + CH;.,- _2_
2CH3- ——-> C2H6 3
CH3- + Hg(CH3)z ———> CH4 + CH3\HgCH2- 1

At room temperature the gaseous product is primarily ethane
with the production of methane increasing as the temperature increases.
The methane presumably arises from a process such as reaction 4.

The photolytic decomposition of dimethylmercury is fairly well
understood; especially with regard to the general mechanism presented
above. On the other hand, there is a great deal of controversy regarding
the thermal decomposition. In general, there is good agreement among
those who have investigated the thermal decomposition with respect to
steps _1_, _2_, and 3 presented above. The disagreement arises concern-
ing the remainder of the mechanism; especially the methane producing

step.

Cunningham and Taylor (2) made a cursory examination of the
pyrolysis reaction in a static system at temperatures of 302 and 348°C.
They found no measurable amount of decomposition up to 2900C. They
reported methane and ethane as the gaseous products with increasing
amounts of methane as the temperature was increased. In contrast to
the photolytic decomposition, they reported that a carbonaceous deposit
was formed on the walls of the reaction vessel which they postulated to
be of a (CI-i2)n nature. Their basis for this postulation stems from the
experimental observations that less than 60% of the carbon from the
decomposed dimethylmercury could be found in the product gases and
also no free hydrogen was detected in the gaseous products. Cunningham
and Taylor made no attempt to determine the rate of the pyrolysis or
its kinetic order.

To verify the formation of methyl radicals, which are postulated
in all mechanisms, Lossing and Tickner (3) decomposed dimethylmercury
in a flow system at 850°C. They determined the presence of methyl
radicals by use of a mass spectrometer. The product gases from the
decomposition were allowed to leak from the reactor through an orifice
into the ionization chamber of the spectrometer. They were able to
detect a measurable amount of methyl radicals, thus supporting the
postulate that methyl radicals are formed during the thermal decompo-
sition. . Lossing and Tickner reported an activation energy for the
overall reaction of 42 kcal/mole in the temperature range 611-8500C
and pressures of dimethylmercury of 2. 8 and 8. 1 microns.

Gowenlock, Polanyi and Warhurst (4) employed a flow system at
temperatures of SOD-600°C and reported first order kinetics. They
found that there was no appreciable inhibition due to the addition of
toluene to the system. They also reported appreciable amounts of
ethylene in the gaseous products. In contrast to Cunningham and Taylor,

who reported more methane than ethane at 300°C, Gowenlock gt :1.

reported ethane in a greater abundance than methane in the temperature
range SOD-600°C. When Gowenlock it a_._1. interpreted their data, they
related kobs. to k1, which is the rate constant for the initial rupture of
the carbon-mercury bond. They reported an activation energy of 51. 5
kcal/mole for the temperature range 500-6000C and total pressures of
approximately 10 mm Hg. On the basis of their observations, they

proposed the following non-chain mechanism:

Hg(CH3)z———->- CH3' + HgCH3- l
HgCH3- ———>- CH3- + Hg 2
2 CH3. ——-> CZH(> _3_
CH3- + Hg(CH3)Z ——> CH4 + CH3HgCHz° f_l_
CH3- + CH3HgCHz- ——s\ CszHgCH3 5
2 CH3HgCHZ' —> (CH3HgCHZ)z 6

Gowenlock .e_t al. reported no experimental evidence for reactions 2
and 6 and they concluded that reactioné plays only a minor role in the
decomposition process.

Using a flow system, Price and Trotman-Dickenson (5) examined
the pyrolysis reaction over approximately the same temperature range
(465-6080C) as Gowenlock e_t al. The data of Price and Trotman-
Dickenson agree very well with that reported by Gowenlock it a}.

The total pressure employed by Price and Trotman-Dickenson was
approximately 20mm Hg and they reported an activation energy of 50. 1
kcal/mole. Since they employed the toluene-carrier technique, Price

and Trotman-Dickenson assumed a somewhat different mechanism.

 

 

H8(CH3)2 ——> HgCH3o + CH3- _1_
HgCH3° —-——> Hg + CH3- 2
2 CH3- s> c2146 3
CH3- + C6H5CH3————-> CH4 + C6H5CH2~ i
2 C6H5CHZ' > (C6H5CHZ)Z _s

Yeddanapalli, Srinivasan and Paul (6) used a static system
(305-3420C) to study the pyrolysis reaction. They reported that at the
lower temperatures it was difficult to choose between first order and
three-halves order kinetics but at the higher temperatures the data
pointed decidedly to three-halves order kinetics. Therefore
Yeddanapalli 3t 31. selected three-halves order for the entire tempera-
ture range of 305-3420C. They also observed that methane and ethane
were the main reaction products with methane being in greater abundance.
They reported small amounts of hydrogen, ethylene and acetylene in the
gaseous products. Yeddanapalli e_t it}. also observed the carbonaceous
deposit on the reactor walls. They obtained an overall activation energy
of 44. 7 kcal/mole in the temperature range 305-3420C and at total
pressures of approximately 40 mm Hg. . In order to account for the

experimental data observed, they postulated the following mechanism.

Hg(CH3)z————>HgCH3- + CH3- _1_
HgCH3- ———> Hg + CH3° 3
2 CH3- -—> can, 3
CH3. + Hg(CH3)z———>- CH, + CH3HgCHz' g
CH3HgCHz- ——> HgCHZ! + CH3° g

 

They proposed that the small quantities of hydrogen, acetylene and
ethylene might be due partly to the decomposition of the HgCHz:
radical in reaction 2 to give CH2: radical which would undergo dispro-
portionation and polymerization reactions to give (CHx)n and hydrogen.
Ganesan (7) recently recalculated the data of Yeddanapalli 3t 31. using
a novel method and obtained an activation energy of 48. 2 kcal/mole which
agrees better with the activation energy reported by other workers.
Laurie and Long (8) reinvestigated the pyrolysis over approxi-
mately the same temperature range (294-343OC) as that of Yeddanapalli
_e_t 31. Laurie and Long also used a static system to study the decom-
position; however, they used a "clean" reactor (reactor cleaned thoughly
before each experiment) whereas other investigators have employed a
- "conditioned" reactor (several decompositions carried out before taking
kinetic data). Laurie and Long reported first order kinetics near 3000C
but they observed that the order increased to nearly three-halves order
at 3430C. They reported an activation energy of 51. 3 kcal/mole for
the pyrolysis in the temperature range 294-333OC. at total pressures of
approximately 70 mm Hg. , Long (9, 10) proposed that the principal
hydrocarbon product, methane, is produced by methyl radicals abstract-
ing a hydrogen atom from hydrocarbons such as ethane. . Long proposed

the following mechanism.

Hg(CH3)z ——-> HgCH3° + CH;,- _1_
HgCH3e ———> Hg + CH3° _2_
2CH3- -——-> Cali, 3
CH3. + CZH6 ——-> CH4 + Cszo :1.
CH3- + CZHS- ———> CH, +C2H, g

CH3° + Csz' é C3H8 é-
CZHS ° + (:sz ' —-—'> C2H6 + Cal-I4 l
C2H5° + Csz' ——'> C4Hlo é

Cattanach and Long (11) reinvestigated the decomposition reaction over
approximately the same temperature range (303-3320C) as Laurie and
Long (294-343OC). Cattanach and Long studied the effect of added
nitrogen, ethane and azomethane in order to confirm the mechanism
proposed by Long.

Russell and Bernstein (12, 13) studied both the inhibited and un-
inhibited thermal decomposition of dimethylmercury in a static system
from 265-3500C- Although they made no attempt to determine the order
of the uninhibited reaction, their data agrees with the observations of
the other investigators. In the cyclopentane-inhibited reaction they
observed that the decomposition exhibited first order kinetics with
methane being the major product. Due to the inhibition from cyclopentane
and from the isotope effect data, they postulated that at least short
chains are formed in the decomposition reaction. On the basis of their
experimental observations, Russell and Bernstein proposed the follow-

ing mechanism for the uninhibited reaction.

Hg(CH3)2 —-> CH3' + HgCH3° _1_
HgCH3° ——-> CH3' + Hg 2
2 CH3- —---—-> C2H6 3
CH3- + Hg(CH3)z-—-—> CH4 + CH3HgCHZ° i

R:

I-..

CH3. + CH3HgCH2' _9 CszHgCH3 é
CH3HgCHZ- ———> HgCH3- + CH7} 1
CH2: + Hg(CH3)Z ——9 CZHngCH3 8

2 CH2: ——‘9 Cal—I4

lxo

II. Objectives of Present Study

Since there has been some disagreement as to the kinetic order
of the thermal decomposition of dimethylmercury, it seemed possible
that this inconsistency could be better evaluated if the reaction was
studied under a wider range of experimental conditions. Previous
investigators who employed a static system observed a variation in the
order of the reaction as the temperature was increased (6, 8, 12).

Since the previous work employing static systems was conducted in the
vicinity of 300°C, it appeared advisable to investigate the reaction over
a wider temperature range. Furthermore, earlier investigations of
this pyrolysis have shown some tendency for the observed rate constant
(calculated on the basis of first order) to be a function of the dimethyl-
mercury pressure (6, 8, 12). This pressure dependence was studied
previously for dimethylmercury pressures. less than 80 mm Hg. With
this in mind, it seemed advisable to study a wider pressure range in
order to establish the order of the reaction. Since a detailed qualitative
investigation of the gaseous products had not been conducted previously,
it appeared this would perhaps be beneficial in proposing a more com-

plete mechanism for the decomposition reaction.

EXPERIMENTAL

1. Apparatus

The experimental work was conducted in a conventional Pyrex high
vacuum system. All stopcocks and joints which were exposed to
dimethylmercury were lubricated with Dow Corning high vacuum sili-
cone grease. The important details of the decomposition apparatus

will be discussed in succeeding sections of this thesis.

A. Vacuum System

 

The vacuum system was maintained at a high vacuum by a 2-stage
mercury diffusion pump in conjunction with a Welch Duo-Seal rotary
vacuum pump. The pressure in the system was measured qualitatively
by an RCA type 1946 thermocouple vacuum gage and associated circuit
which had been calibrated with a McLeod type vacuum gage. The operat-
ing pressure within the system was always maintained at less than one
micron.

Due to the toxicity of dimethylmercury; two manifold traps, one
on each side of the diffusion pump, were cooled withliquid air to prevent
dimethylmercury from escaping to the atmosphere. Upon the completion
of an experiment, any dimethylmercury in the first trap was allowed to
distill to the second trap which was situated between the diffusion pump
and the rotary pump. The second trap was then removed to an efficient
hood and the contents allowed to distill to the outside atmosphere. The
dimethylmercury remaining after decomposition was stored in a storage

bulb ("used" storage bulb) which was kept at -780C.

One section of the vacuum system was designed for purification
operations. This section contained a U-tube filled with P205 and was
in series with several other U-traps which were employed in trap to

trap distillations .

B. Reactor

A schematic diagram of the reactor is shown in Figure 1. The body
of the reactor Was constructed of 77 mm Vycor tubing and contained a
7 mm Vycor axial thermocouple well. The inlet was constructed of
Pyrex capillary tubing connected to the reactor with a 13 mm graded
seal. The surface area of the reactor was estimated to be 630 cm2‘ and
its volume was found to be 870 i 1 cc.

The Vycor vessel was encased in two concentric aluminum tubes
which extended approximately 3 cm above and below the body of the
vessel (Figure 1). The reactor was heated by five independent windings
of Nichrome resistance wire. Winding W1 was wound the entire length
of the aluminum shell and was connected to a Leeds and Northrup
Electromax Controller Model 6251-7. This instrument controlled the
temperature of the reactor to better thani 0. 2°C at 2750C for a period
of four days. Windings W2 and W3 were wound around the upper one-half
and lower one-half of the aluminum shell respectively, and were
separated from W1 by a thin layer of asbestos. Winding W4 extended
from the reactor inlet to the stopcock and was covered with approxi-
mately one inch of asbestos. Winding W5 was coiled at the lower end of
the reactor inside the heavy envelope of asbestos which encased the
entire reactor. All windings were powered by Central Scientific Company
"Powerstat" variable voltage transformers. Ten Chromel-Alumel
thermocouples were used to measure the temperature of the reactor
relative to an ice bath at 00C. The average value of thermocouples
one through eight was recorded as the reactor temperature. A typical

temperature profile of the reactor may be found in Table I.

10

 

 

 

 

 

 

 

 

 

 

 

 

 

w. \
W2 0
o
o
o
w, o
w. 4;
o
0
\°
w5 \ O o X
w1 \Js

    

 

 

 

 

 

 

 

 

Control

Thermocouple
Well

 

V

 

 

 

 

 

Legend:
- Aluminum 0 Nichrome Resistance Wire
Asbestos W Winding

x Thermocouple

Figure 1. Details of Reactor.

11

Table 1. Temperature Profile of Reactor (Experiment 11-6-33).

_ — —
==== J J.» ___m =====

Number ' Potentiometerl reading (mv) Temperaturez (0C)

 

Main the rmoc ouple s

 

 

 

1 13.471 330.2
2 13.473 330.3
3 13.473 330.3
4 13.474 330.3
5 13.474 330.3
6 13.470 330.2
7 13.471 330.2
8 13.464 ' 330.1
Average 13.471_t0.002* 330.3.+_0.1*
Auxiliary thermocouples
9 12.990 318.9
10 6.035 147.9

 

lRubicon Type B Potentiometer.
z"Values taken from Reference Tables for Thermocouples (14).

:1:
Average Deviation.

12

Since the reactor deposits a polymer-like material on the reactor
walls (2), several preliminary decompositions were performed to

"condition" the reactor befOre recording any kinetic data.

 

C. Gas Chromatography

The gas chromatography apparatus used for the analysis of the
gaseous reaction products was a Perkin-Elmer Model 154D Vapor
Fractometer utilizing a thermo-conductivity detector. The fractometer
was employed in conjunction with a Fischer Recordall electronic strip-
chart recorder and associated circuit. Helium was used as the carrier
gas in the fractometer.

The gas-liquid partition column was constructed of 7 mm. o. (1.
stainless steel tubing which measured 3. 85 m in length. The packing
material used in the column was "dimethylsulfolane" (2, 4-dimethyl-
tetrahydrothiophene-l, l-dioxide) on Chromosorb and was obtained from
the Perkin-Elmer Corporation. The column was operated at room
temperature (Ix/23°C). Further details of the experimental procedure
for handling the gas samples will be discussed in the portion of this

thesis entitled Experimental, section IV.

11. Materials

A. Dimethylmercury

 

The dimethylmercury was prepared by allowing methylmagnesium
iodide (prepared from iodomethane) to react with mercuric chloride
according to the method of Gilman and Brown (15). The dimethylmercury
was kindly furnished by R. B. Bernstein and C. H. Brubaker. .The
material received from C. H. Brubaker was fractionated through an 8
inch column packed with glass-helices. The fraction boiling from

89-920C (740 mm Hg) was collected and mixed with the material from

13

R. B. Bernstein. The two fractions were mixed and fractionated again
A through the same column and the 91. 5-92.00C (740 mm Hg) fraction
collected which was used for the experiments. It was stored at room
temperature in a Pyrex container kept in the dark. When the dimethyl-
mercury was admitted to the vacuum system, it was further purified
by passage through P205 and a trap to trap distillation from -380C to
~194OC. The middle fraction (about 80% of the total) was retained from
this distillation and stored under vacuum at ~780C. The infrared
spectrum of the dimethylmercury appears in Figure 2 (Spectrum A) and
was obtained with a Perkin-Elmer Model 21 double beam spectrophotom-
eter with sodium chloride optics and a 0. 107 mm. pure liquid cell with
sodium chloride windows. The spectrum agrees quite well with the
data reported by Gutowsky (16). The peak at 4. 3 |J., however, does not
appear in the liquid spectrum of reference 16 but it does appear in the

gas spectrum.

B. Carbon Dioxide

 

In certain experiments the total pressure of the system was
increased by the addition of carbon dioxide. The carbon dioxide was
Matheson "bone dry" with a stated purity > 99. 9% and it was used with
no further purification. An evacuated bulb was filled with carbon
dioxide and connected to the vacuum system. The carbon dioxide was
then transferred to a U—trap at -1940C and the volatile substances
vented to the manifold. The carbon dioxide was degassed several
times by allowing the trap to warm to -780C and recooling to -1940C
followed by venting to the manifold. A trap to trap distillation from
-780C to -1940C was performed and the carbon dioxide sent to a storage

bulb kept at ~780C.

14

NH

.mGOmCmQEoO 3300mm

Anson 38v Aymcofiofimg

OH m

.N oudmfm

 

 

_ _ _ _

Stet:
.me Eonw osgmom u m 8:30on

2anva .. 4 guuoomm

 

 

 

 

 

 

 

 

 

 

0
fl

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

uotsstuxsuexl was led

15

111. Operating Proc edure

A. Typical Dimethylmercury Experiment

 

An evacuated sample bulb was thoughly cleaned before weighing
and the dimethylmercury, stored at -780C, was warmed up and a
quantity of it frozen at -1940C into the previously weighed sample bulb
and the bulb weighed again. The sample bulb was then connected to
the vacuum system and a quantity of dimethylmercury was vaporized
into the reactor. The excess dimethylmercury in the inlet was frozen
at -l94oC into a small U-trap in the inlet system. The U-trap was
warmed to 00C and the excess dimethylmercury frozen into the sample
bulb which was weighed again to determine the amount of sample in the
reactor. The excess dimethylmercury from the introduction of the
sample was then returned to the storage bulb at -780C. The sample
bulb was again evacuated and weighed as a check against the original
weighing. The temperature of the reactor was recorded several times
during the period of decomposition.

After a predetermined period of time, the reactor stopcock was
opened and its contents frozen into a U-trap at -194OC. The product
volatile at -1940C, assumed to be methane (4, 8, 10), was passed
through a ~194OC trap and vented to the manifold. The U-trap was
warmed to 00C and the volatile products frozen into another trap at
-194OC. This trap was then warmed to -970C and the volatile products
were either vented to the manifold or removed by means of a Toepler
pump to be retained for gas chromatographic analysis. The residual
material from the -970C trap was warmed up, transferred to the sample
bulb, and weighed as dimethylmercury. The infrared spectrum of this
material for an experiment carried out to 16. 2% decomposition appears

as spectrum B in Figure 2.

16

For certain experiments which required high pressures of
dimethylmercury, heating tapes were used to heat the sample bulb and
the inlet independently (sample bulb, ~90°C; inlet, ~120°C). The
same procedure of introduction and removal of dimethylmercury was
followedin these experiments except for the heating of the sample bulb
and the inlet.

Experiments were conducted to determine the loss of dimethyl-
mercury incurred during the course of an experiment due to the
absorption of dimethylmercury by the stopcock grease. The reactor
was cooled to 230°C at which temperature no measurable decomposition
occurs and the procedure described previously for the introduction and

'removal of dimethylmercury was carried out for various concentrations
ofvdimethylmercury. The loss of dimethylmercury was observed to be
a function of the total amount of dimethylmercury used. For example,
using a total dimethylmercury sample of 600 mg there was a 1. 2 mg

— loss (0. 2% loss). The loss was only slightly greater in the experiments

in which the sample bulb and inlet were heated (less than 0. 3% of the

total sample). Each sample was therefore corrected for this deviation.

B. Experiments with Carbon Dioxide

 

The dimethylmercury was introduced using the same procedure
described previously (see Experimental, sec. III-A). After the dimethyl-
mercury had been introduced into the reactor, the carbon dioxide was
admitted to the reactor at its own vapor pressure at -780C. The total
pressure (dimethylmercury plus carbon dioxide) was measured by means
of a manometer in the inlet system. The same procedure was followed
for the removal of the products as was described in a previous section

of this thesis.

17

IV. Analysis of Products

A. Gaseous Products

 

As mentioned previously, the gaseous products were analyzed
gas chromatographically using a gas-liquid partition column packed
with "dimethylsulfolane" on Chromosorb. A gas syringe was used to
inject the products into the fractometer. Figure 3 illustrates the gas

sampling bulb.

Standard Taper Joint

*1“ Gum Rubbe r Plug

Figure 3. Gas Sampling Bulb

The gaseous products which were retained from the -970C trap
(see Experimental, sec. I-C') were frozen into the gas sampling bulb
at -194OC.. Since the products were all gases at room temperature,
the sample was removed from the bulb by injecting the needle of the
gas syringe through the gum rubber plug. After a quantity of the
products had been admitted to the gas syringe, the needle was extracted
until only the tip remained in the rubber plug. The plunger of the
syringe was then depressed slightly to minimize the amount of air
rushing into the syringe upon complete extraction of the needle from
the rubber plug. This sample was then injected into the carrier gas

stream of the fractometer.

 

RESULTS
I. Kinetic Data

The following rate expression was used to correlate the data
obtained for the thermal decomposition of dimethylmercury.
d(D)

- -—d-t— : ka (D) + 1210(1))z

This empirical expression can be integrated to give (see Appendix 2):

(Do) k + k (D)
; [(D) ka + kb (D0) a
In this expression (Do) and (D) are reSpectively, the initial concentration
of dimethylmercury and the concentration at any time t. The data for
the thermal decomposition of dimethylmercury is summarized in Tables

II, III, IV, V, and VI. The following expression was used to evaluate

ka and kb (see Appendix 2).

 

- 1 k
(D0) = a z ' a '"a-
kat (k t) k
klot [1 '2':— 3?‘ ' ' ° ] b
where
a = (09)
(D)

A least squares analysis of the data at the temperatures studied gave

the following Arrhenius rate equations for the decomposition reaction

k 2.6 x109 exp (-39,400/RT), (sec.'1)

a

9.5 x 1026 exp (-70,600/RT), (cc. mole"1 sec,-1)

kb
The Arrhenius plots for ka and kb appear in Figures 4 and 5, respectively.
Table VII summarizes the least squares data for ka and kb for the

temperatures at which the decomposition was studied.

18

19

Table II. Thermal Decomposition of Dimethylmercury at 2750C.

 

 

 

 

Exp. T(°C) [Do] x 106 t (sec.) 1
(moles, c071,)
11-66 275.5 0.526 313,680 0.170
11-61 275.8 1.636 217, 620 0.122
11-62 275.3 3.848 220, 680 0.163
11-63 275.3 5.454 187, 200 0.149
11-64 275.3 6.786 185,760 0.160
(D)

>1<
Extent of decomposition, f = 1-
(Do)

Table III. Thermal Decomposition of Dimethylmercury at 2890C.

 

 

 

===L - m
Exp. T(OC) [Do] x 106 t (sec.) f
(moles cc?!)

II-83 289. 3 0. 728 125,220 0.146
11-76 289. 2 1. 206 97, 080 0.127
11-82 289.2 1.736 106,200 0.158

- 11-77 289.2 2.746 76, 200 0.133
11-80 289. 2 3. 601 93, 900 0.154
11-81 289.2 4.825 80, 520 0.151
11-78 289. 3 5. 865 70, 260 0.152
mm 289. 2 7. 052 63, 900 0.153

 

20

Table IV. Thermal Decomposition of Dimethylmercury at 303°C.

 

 

 

Exp. T(°C) [Do ] x 106 t (sec.) f
(moles coal)

1-12 303.7 0.312 33,180 0.153
1-15 303.8 1.488 26,100 0.139
'1-16 303.8 1.734 26,100 10.158
1-14 303.8 1.799 26,100 0.148
1.22 303.5 2.915 22,800 0.162
1-25 303.8 3.486 20,160 0.152
1-23 303.7 3.862 19,800 0.168
1-20 303.6 6.683 14,700 0.157
1-24 303.8 7.305 13,380 0.158

 

Table V. Thermal Decomposition of Dimethylmercury at 3190C.

 

 

 

Exp. T(°C) [Do] x 106 t (sec.) f
(moles cc".1})

11-67 319.2 0.746 9,240 0.124
11-69 319.3 2.416 7, 260 0.185
11-71 319.3 2.488 6,000 0.166
11-73 319.2 2.812 5,400 0.162
II-70 319.3 2.834 5,880 0.157
11-74 319.2 4.117 3,960 0.159
‘11-72 319.3 5.416 3,360 0.162

 

21

 

 

 

Table VI. Thermal Decomposition of Dimethylmercury at 330°C.

Exp. T(°C) [Do] x 106 t (sec.) f
(moles CC'Tl)

II-39 330.3 ' 0.220 11,160 0.149
11-37 330.3 0.255 10,800 0.170
11-36 330.4 0. 270 8, 440 0.122
II-35 330.3 0.380 7.,440 0.134
II-34 330.3 0.434 7,440 0.163
11-53" 330.0 0.638 7,200 0.184
11-54* 330.0 0.699 6,600 0.189
II—32 330.3 0.727 6, 960 0.167
II-33 330.3 0.845 5,820 0.153
II-30 330.2 1.631 3,660 0.135
II-31 330.2 1.792 4, 200 0.162
II-41 330.3 2.742 2,460 0.137
11-40 330.3 2.821 2, 940 0.165
11-38 330.2 3.108 2,820 0.166
11-42 330.1 3.401 2, 220 0.146
II-45 330.3 3.822 2,100 0.154
11-43 330.3 3.917 2,040 0.147
II-44 330.3 4.626 1,800 0.152
11-46 330. 3 4. 942 1, 740 0.149
II-47 330.3 5.459 1,680 0.162
-II-48 330.2 5.907 1,440 0.150
II-49 330.3 6.378 1, 380 0.147
II-50 330.3 6.436 1,440 0.162

 

Approximately 6. 7 x 10-5 moles cc. -1 of C02 added (~580 mm. Hg)

22

.mM .HOH uoHnH msficoann/w .w onfiwfih

 

 

, ml
:kov MB
on; me; 2.; be;
_ a _ n
1 0.0
..1
0
an
n.
1 m.m e)
S
x
O o
r
O
I on
_ _ _ _

 

 

.nrx ROM worm mgsoflnhdn .m madman.»- ,

 

23

 

H.
A0311.
a o no”
.2 ms.n os.H no.2
_ _ _ -
no.7
o
m
on
x.
q
m
o lo W
m.
8
_t
S
3
b
.1
o 1e;
o
_ _ _

 

 

23

om.

.nx MOM uoHnH msfisoAnuifl .m ohdmwh -

H
70:11
a one

H mp4 ow;

mo;

 

 

_ _ _

 

 

(1' 'oss I_91011.1--:):>)q>{ 801

23

hr eon SE sadness..- .m spawns .

AMVB
Toma
ow.~ m>.H ON.H

mo;

 

 

_ a _

oJ:

 

 

(1' was 1-910111'3310131 BOT

24

Table VII. Temperature Dependence of the Rate Constants ka and kb.

 

L
L

 

 

T(°C) ka x 106,7(sec.'1) kb, (cc.mole"l sec.‘l)
275 0.547 0.0644

289 1.14 0.221

303 4.39 1.24

319 8.01 9.22

330 13.2 18.2

 

25

II. Analysis of Products

As mentioned previously (see Experimental, sec. IV), the
gaseous products were analyzed by gas chromatography. Three gas
chromatograms for experiments-conducted at different temperatures
are represented in Figure 6. a The symbols X4 and X1 inFigure 6
refer to the attenuation of the signal from the detector to the recorder.
In all three experiments represented in Figure 6, the methane was
removed (see Experimental, sec. III) before the chromatographic
analysis. Chromatogram A represents an experiment conducted at
3190C and 16. 2% decomposition; whereas, Chromatogram B represents
an experiment at 3030C and approximately 50% decomposition. The sub-
stances represented in Chromatogram C were obtained by combining
the products of two experiments at 289°C with an average decomposition
of 15. 2%. The combined products from these two experiments at
2890C were submitted to a trap to trap distillation in an attempt to
separate the less volatile substances which were present in small
concentrations. The gaseous products were identified by matching their
elution times with those of known hydrocarbons analyzed under the
same conditions. The products were found to be ethane, ethylene,

acetylene, propane, propene, n-butane, l-butenewand 2-methylpropene.

26

 

 

 

le

\u
l\

 

X4

 

mammoumafioau M
0:33;

ocofwuoom

oceans .. c

odomonm

mammoum

onogguo

 

 

11.1

0:830

has

 

J

 

 

 

 

 

 

 

oncommom

25

20

15

10

Time (min. )

Gas Chromatograms of Gaseous Products.

Figure 6.

DISCUSSION OF RESULTS

1. Kinetic Data

A. Rate Expres sion

 

As previously mentioned, the following empirical rate expression

was employed to correlate the data obtained during the present study:

d(D)
' dt

 

= ka(D) + khan)?-

This expression integrates to give:

ka, . 1,, (93.). , Isa+_k_b<_m_
(D) ka + kb(Do)

To facilitate the evaluation of ka and kb this equation can be written as

(s ee Appendix 2):

 

a - 1 o. k
(D0) = kat (kat)z " E:
kbt 1' ”_2'. + 3'. - o o o

The procedure for the evaluation of ka and kb can be found in Appendix 3.
The parameter din this expression is related to f, the extent of decom-

position, by the relation:

rle
Ha

I
9|?

An arbitrary restriction was placed on d for the evaluation of ka and kb.

The only experiments used in the calculations were those which satisfied
the restriction that o. .= .3 i 0. 04 (where '6' is an average value of a.

and was approximately 1. 180 at all temperatures) which corresponds to

f = f i 0.03. This restriction was placed on the values of 0. because

 

28

of the method of evaluating ka and kb. In order to test this restriction,

values of a were calculated from the following expression:

k + k (D)
1 = k t _ 1n _a____b___

where
(Do)
(D)

The calculated values of 0., denoted as o. are compared in Table VIII

c,
with the experimental values of o. for a series of experiments carried

out to various extents of reaction at 3300C. Since the average value of

a for all temperatures was approximately 1. 180 and from the comparison
of a. and ac in Table VIII, it appears that the restriction placed on o.

(a = E i O. 04) is reasonable for the evaluation of ka and kb.

The validity of the empirical rate expression was investigated

using the following form of the integrated rate equation:

 

t - _1_ In (Do) . M
C ‘ ka (D) ka + kb (Do)

From this equation, the calculated time of reaction, tc, was evaluated
for the experiments performed at 319°C. The calculated time of
reaction, tc, and the experimental time of reaction, t, are compared
in Table DC.

For all of the temperatures investigated in the present study there
appeared to be a. linear dependence of the apparent first order rate
constant as a function of dimethylmercury concentration. This is
shown in Figure 7 for the temperature 303°C. For small extents of
reaction, this linear dependence can be shown to be consistent with
the empirical rate expression adopted in the present study. The pre-

sentation of these arguments appears in Appendix 4. As a test of

29

Table VIII. Comparison of o and 9c at 3300C.

 

 

 

 

Exp. (Do) x 106 (D) x106 ac a t (sec.)
(mole~ CC'I) (moles- cc'l)
II-55 3.303 3.064 1.080 1.078 1,140
II-60 3.247 2.904 1.119 1.118 1,680
II-57 3.218 2.738 1.174 1.175 2,340
11-59 3.151 2.447 1. 277 1. 288 3,360
II-56 3.200 2.396 1.329 1.336 4,200
II-58 3.135 2.195 1.420 1.428 5,400
Average deviation of a from o. 0. 005.

C

30'

Table IX. Comparison of tC and t at 319°C (0.: 1.182).

 

 

 

Exp. (D0) x 106 (D) x 106 tc (sec.) t (sec.)
(moles cc'l) (moles cc'l)
11-67 0.746 0.653 9, 220 9, 240
II-69 2.416 1.968 7,300 7,260
II—71 2,488 2,075 6, 270 6,000
11-73 2.812 2.358 5, 540 5,400
II-70 2. 834 2. 388 5, 650 5, 880
mm 4.117 3.461 4,050 3, 960
11-72 5.416 4.540 3,290 3,360

 

Average deviation of t from tC

123 sec.

31

4on do x nopHO “warm mo monopcomom .m. onswwh

$100.,mofiocflv 80H on Ash:

Nd. 4.0 o.m wé 0.4 N.m «SN 04 m6

 

 

 

_ a a _ _ _ _ A _
I N
6688 spasm sadness o
Genoa a: 984 use 333.0
1 e

(.3096) 901 x I>1

 

 

 

32

these arguments, a least squares analysis was performed on the
apparent first order rate constant as a function of dimethylmercury
concentration at each temperature. These values of the first order
rate constant were compared with those calculated from experimental
data. The results of these comparisons appear in Table X.

From the experiments with carbon dioxide, it appears that an
inert gas has no appreciable effect on the rate of the reaction. . This
indicates that the reaction was not studied in the region where the

apparent first order rate constant "falls off. "

B. Comparison of the Data of Other Investigations

 

A comparison of the data reported by other investigators employ-
ing static systems shows that the data obtained in the present study are
consistent. . Laurie and Long (8), who reported first order kinetics
for the pyrolysis, investigated a dimethylmercury concentration range
of 0.8 x 10"6 to 2.2 x 10'6 moles-cc?1 at 305°C. The data of the
present study were calculated on the basis of first order kinetics in an
attempt to illustrate the effect of dimethylmercury concentration on the
apparent first order rate constant. These data are compared in
Figure 7 with the data of Laurie and Long. The scattering of the data
of the investigators does not seem to warrant the postulation of first
order kinetics.

Furthermore, the concentration of dimethylmercury appears to
influence the Arrhenius activation energy of the reaction. This can be
illustrated if the activation energy of the assumed first order reaction
is studied as a function of dimethylmercury concentration. Using the
data of the present study, first order rate constants were calculated
for three different dimethylmercury concentrations. These data are
presented in the form of Arrhenius plots in Figure 8. The solid circles

in Figure 8 represent the same concentration range used by Laurie and

33

Table X. Summary of First Order Rate Data.

 

 

 

 

 

T(°C) Per Cent (11
275 .
289 4. 3
303 4. 9
319 2.
330 3.4

I ~-

ZImI

Per Cent (1 = L x 100

kL = first order k from least squares

k1 = experimental first order k

n = number of experiments treated

Per Cent (1 = per cent average deviation

34

 

 

 

.noUuO «muwh mo ewmnocm £039,304 23 so Ash: mo woommm .w onsmwh
.H.
A -M II
a CV an:
ow; ms; on; me;
_ _ _ _
/
1 lo.
0 o
O
O
0
1 9 O I o .m
01/
O
AToonoHoErbH on some n 28 e o/
$.09 moHoEveuoH x mood .1. 2n: O 9
I r..oo_.no387.2 x omnd u an: o I or.
_ _ a of

 

 

 

(Poss) {>1 801-

35

Long to obtain the value of 51. 3 kcal.mole'l for the activation energy of
the pyrolysis reaction. From Figure 8, for the same concentration
range, a value of 53. 0 kca1.mole'l was obtained for the activation
energy based on first order kinetics. At a concentration of 0. 350x 10'6

l, the activation energy from Figure 8 is 43. 3 kcal.mole'1;

moles cc. "
at 6. 200 x10"6 moles cc. '1, it becomes 60. 5 kcal.mole7l.

Yeddanapalli e} a_:_l. (6) also investigated the influence of dimethyl-
mercury on the reaction over a range of 0. 307 x 10"6 to 1.678 x 10'6
moles cc'l (11. 79 - 64. 37 mm. Hg). They correlated their data on the
basis of three-halves order kinetics. Their experiments on the effect
of dimethylmercury concentration were conducted at a higher tempera-
ture (3420C) than any of the data in the present study. For this reason
an extrapolation of the data of the present study was necessary in order
to compare the two investigations. The values of ka and kb obtained
by extrapolation of the Arrhenius plots (Figures 4 and 5) to 342°C were
used to calculate values of 9c for the individual experiments of
Yeddanapalli e1: a_l. The comparison of 0., the experimental value
obtained by Yeddanapalli e} a_.l.; and ac, the value calculated using ka
and kb from the present study, is shown in Table XI. Based on this
comparison it appears as though the data are consistent.

. It is of some interest here to again note the observations of
previous investigators with respect to the effect of temperature on the
kinetic order‘of the reaction. Yeddanapalli e} 31. (6) reported that at
305°C and 323°C there was little to choose between first and three-
halves order kinetics. On the other hand, at the highest temperature,
342°C, the three-halves order rate constants were decidedly more con-
sistent than the first order constants. Hence, they reported three-
halves order for the entire temperature range 305-3420C. Laurie and
Long (8), who reported first order kinetics, observed that the reaction
appeared to be first order in the range 294-3320C but at 343°C the

order apparently increased to three-halves.

 

36

Table XI. . Comparison of c1 and 9c at 3420C (Data from reference 6).

 

 

 

 

pol(mm Hg) pz(mm Hg) f t (sec.) (1 (18
11.79 9.91 0.160 3,600 1.189 1.184
18.32 14.74 0.195 3,600 1.243 1.233
24.84 19.00 0.235 3,600 1.307 1.287
39.31 29.24 0.256 3,600 1.344 1.349
64.37 44.89 0.303 3,600 1.435 1.456

ka = 2.77 x10'5 (sec'l)
kb = 1.69 x10"6 (mrni'1 sec‘l)

1initial dim ethylme rcury pre 5 sure

zfinal dimethylmercury pressure
k + k
ka+ kb P0

average deviation of a from ac: 0.012.

3In QC: kat - 1n

37

These observations can be shown to be qualitativelyconsistent»
with the empirical rate expression adopted in the present study. This

expression was presented earlier as

 

- = ka (D) + kb (D)2

In this expression ka is the rate constant for a first order process whose
activation energy was found to be 39.4 kcal.mole"l. The activation

energy for kb, the second order process, was found to be 70.6 kcal.mole'1.
Considering these activation energies for the empirical rate expression,
the second order process would be expected to become more important

as the temperature is increased. The apparent order of the reaction

would then increase with increasing temperature.

II. Reaction Products

A. Gas eous Products

 

Previous investigators have not concerned themselves with a
detailed study of the gaseous reaction products. Laurie and Long (8)
reported methane, ethane and small amounts of ethylene with methane
being in the greatest abundance. Yeddanapalli e_:_t_a_1. (6) reported
methane present in larger quantities than ethane as well as trace
amounts (less than 1% of total) of ethylene, acetylene and hydrogen.
Russell and Bernstein (12) analyzed the gaseous products by mass
spectroscopy and reported methane, ethane, ethylene, propane, propylene,
C4H8* and C4H10. * The gaseous reaction products observed in the
present study were identified by gas chromatography and were found to
be ethane, ethylene, acetylene, propane, propene, n-butane, 1-butene

and 2-methylpropene. The methane was always discarded and therefore

 

5::
No attempt was made to distinguish the isomers.

38

was not detected by the gas chromatography analysis. Due to the method
of analysis in the present study it was not possible to obtain the concen-

trations of the various product gases.

B. Solid Product

 

Some preliminary experiments were conducted in an attempt to
characterize the carbonaceous deposit (2, 6, 8, 11) which is formed on the
walls of the reactor. An attempt was made to dissolve the substance in
several organic solvents but it appeared to be insoluble. However, the
deposit was soluble in concentrated nitric acid and the resulting solu-
tion gave a positive test for mercury. Furthermore, several decompo-
sitions were conducted and the amount of mercury collected was con-
siderably less than the amount that should have been produced based on
the amount of dimethylmercury decomposed. These preliminary
experiments suggest that perhaps the carbonaceous deposit contains

mercury.

III. Mechanism

A reasonable mechanism could not be postulated for the pyrolysis
of dimethylmercury in view of the data obtained in the present study.
It is apparent that several more aspects of the reaction should be investi-

gated before a plausible mechanism can be postulated.

IV. Further Experimental Studies

On the basis of the preceding discussion of the pyrolysis reaction,
a number of experiments suggest themselves as useful in the postulation

of a plausible mechanism:

 

39

(1) Careful studies of the heterogeneous nature of the reaction
should be conducted by packing the reaction vessel to increase the
surface/volume ratio.

(2) The reaction should be investigated using larger concentrations
of dimethylmercury.

(3) The methane producing step should be clarified; perhaps by
adding deuterated ethane to the reaction mixture to determine whether
methane is produced by hydrogen abstraction from compounds such as
ethane or by hydrogen abstraction from dimethylmercury.

(4) A carbon/hydrogen ratio and a quantitative mercury balance
might be conducted on the carbonaceous deposit in an attempt to
characterize it, if possible.

(5) The reaction might better be followed by intermittent sampling
of the reaction mixture in conjunction with a quantitative analysis of the

gaseous products .

 

5.

10.

11.

12..

13

14.

15.

16.

17.

BIBLIOGRAPHY

E. W. R. Steacie, Atomic and Free Radical Reactions, New York,
Reinhold Publishing Corporation, 1954, p. 386.

 

. J. P. Cunningham and H. s. Taylor, J. Chem..Phys. g, 359 (1938).

. F. P. Lossing and A. w. Tickner, J. Chem. Phys. _2_9_, 907 (1952).

B. G. Gowenlock, J. C. Polanyi, and E. Warhurst, Proc. Roy..Soc.
A218, 269 (1953).

S. J. W. Price and A. F. Trotman-Dickenson, Trans. Faraday Soc.
_5_3, 939 (1957).

..L. M. Yeddanapalli, R. Srinivasan and V. J. Paul, J. Sci. Ind.

Research (India) 13B, 232 (1954).
R. Ganesan, J. Sci. Ind. Research (India) 20B, 228 (1961).

C. M. Laurie and L. H. Long, Trans. Faraday Soc. _5_1_, 665 (1955).

.. L. H. Long, Trans. Faraday Soc. _5_1_, 673 (1955).

L. H. Long, J. Chem. Soc., 3410 (1956).
J. Cattanach and L. H. Long, Trans. Faraday Soc. _5_6_, 1286 (1960).

M. E. Russell and R. B. Bernstein, J. Chem. Phys. 1!, 607
(1959). ’

.. M. E. Russell and R. B. Bernstein, J. Chem. Phys. _3_0_, 613

(1959).

H. Shenker, J.. I. Lauritzen Jr., R. J. Corrucini and S. T.
Lonberger,. Reference Tables for Thermocouples, Nat. Bur. Stnds.
Circular 561, 1955, p. 26. V '

H. Gilman and R. E. Brown, J. Am. Chem. Soc. _5_2, 3314 (1930).
H. S. Gutowsky, J. Chem. Phys. _1_7, 128 (1949).

E. W. Washburn et a1. , International Critical Tables, New York,

 

.MeCraw-Hill, lne'T, '1926, p. 53, Vol. 1.

4O

 

APPENDICES

41

APPENDIX 1

Thermocouple Calibrations

Zinc, c.p. grade, was covered by powdered carbon in a porcelain
crucible and was melted with a Chromel-Alumel thermocouple in good
thermal contact with the melted zinc. After cooling, the temperature
became constant and the e.m.f. was read using a Rubicon Type B
Potentiometer. The e.m.f. recorded was 17. 212 mv. From Reference
14:

o
17. 212 my = 419. 3 C

melting point of Zn = 419.4OC (17)

The same procedure was carried out using '.cad1nium,'1.c. 1).-’- grade,

for the calibration

potentiometer reading = 13. 070 mv = 320. 80C (14)
melting point of Cd = 320. 9°C (17)

No corrections were applied to the temperature readings.

42

 

and rearranging this expression in an integrable form:

APPENDIX 2

Derivation of Rate Expression

 

o. - 1 k
[Dd = kat (hat)z ' ° k:
kbt 1- 2! + 3'. - ...]

Assuming the empirical expression:

H12)

dt ka(D) + kb(D)’-

ID d(D) J.t dt

Do Ika+kb(D)](D) = ‘ o

integ ration gives:

or

expansion of e

_1_. 1,, 112.2.._a_b__k+k(Do) -t
ka (Do) ka + kb(D)

(D) ka + kgmo) z e- kat
(Do) k3, + kb(D)

let (1 = (Do)/(D)

k t '
a gives:

-kt (k t)2 (k 1:)3
e a =1-kat+—a—Z, - 4—3, +---

substituting (1. 5) and (1. 6) in equation (1.4) gives:

43,

(1.1)

(1.2)

(1.3)

(1.4)

(1.5)

(1.6)

9'; '. "m w

44

ka + kb(Do)
aka '1' kb1D0)

(kit)z _ (kit)3

2'. 3'. +'°° (1'7)

 

l-kat+

. Upon expansion and collection of terms the following expression is

obtained:

 

2
(1t- 0.) =[Gka+k‘b(D0)][-1+ 252%:1- £315?L)-+ o..] (1.8)

and after rearranging equation (1. 8) becomes:

 

- o.-— (1.9)

[D1=
o kbt|:1- 123—fi- + (1392- ...] kb

 

 

APPENDIX 3

Evaluation of ka and kb

The rate constants ka and kb were evaluated from the following

expression by successive approximations using the method of least

 

 

squares. FM
11
a 1 :ka ;
(D0) = k t (k 11F - 01:}; (1.1) .

A least squares analysis was carried out for a plot of (Do) vs. l/t which 1

yielded approximate values of ka and kb. The values of t used in this
expression were the times of reaction, calculated on the basis of first
order, required to obtain the average extent of reaction. Using these
values of ka and kb, a second approximation was performed by calculating

values of H and performing a least squares analysis for (Do) vs. H .

 

l
H = ”“9
k t (k t)2
kbt [1- f + f - .0.

After three such approximations ((D0) vs. H ), an extrapolation was
made for t to the time required for the average extent of reaction, tr,

which involved the use of the following expression:

= _1_ 1.23,"- kb (D) .(DQ)
1:r a 1n [ka + kb (Do) (D)] (1. 3)

The values of ka and kb used in this expression were the values obtained

 

W

from the previous approximations. These values of tr were then used
in equations (1. 1) and (1. 2) and the successive approximations performed

in the same manner as previously discussed.

45

 

APPENDIX 4

Relationship of ka and kb to First Order Rate Expression

The apparent first order rate expression can be expressed as

 

 

klt = In 130) = - 1n (1-f)
(Do) _ _1_
where (D) - l-f

For small values of f the term -ln(1-f) can be expanded to give
f2

f+-2—

f
klt- i(l+-Z-)

From the present study

d(D)
" dt

 

= kam) + kbm)2
For small extents of reaction, i. e. , small f, this can be expressed as

_ At?) = _2_____(D) ' (D) = 1.35) + kb(fi)’-

 

 

At t
where (D) =' (p0);- (D):
(.29.):‘121 = [ka+kb('f)) ]t
(D)
(D0) ' (D) =.[ ka+ kb(fi)] t
(D9) + (D)
2
since (D0) = % , the above expression reduces to

46

 

'v‘ftlc

47

f

f = [k3, '1' kb (EH 1'.
1":

 

For small values of f

——f-—’-—‘—’f(1+£2-)

f
1':

[ka+kb(D)]t= £(1+-fi)

For small extents of reaction, this exPression is similar to the first

order rate expression

 

Z

From this

k, = ka + kb (13)
k1: k3. + kb [gig—241.91]
since (D) = (Do) (1 - f)

, f
k1=ka+kb(Do) (1-3)

which shows the dependence of k1 on (Do).

CHEMISTRY LIBRARY

 

 

 

III
I
III
11
I
I
II
11..
III
I
III
I
l
1.1
III
I11
I