/

...

....."'. -

---..;....-_._ •.

TIME-RESOLVED TUNABLE DIODE LASER DETECTION pF

THE PRODUCTS OF THE INFRARED MULTI PHOTON DISSOCIATION

) OF CF2 HCl, (CF3)2CO, AND CHC1 3

•

by

JOHN JOSEPH ORLANDO, B.Sc .

A Thesis

Submitted to the School of Graduate Studie~

in Partial Fulfilme~t of the Requirements

•

, -

•

,@

for' the Degree

Doctor "'of Rhilosophy

~.

"'­McMaster University

1987

..

~\

\

.." ./

l"j

:

~.

TUNABLE DrODE "LASER DETECTION

OF THE PRODUCTS OF IRMPD

,

,

\

--.'-"

, '

, ..,

DOCTOR OF PHILOSOPHY (1987)(Chemistry)

..

.'~

McMASTER UNIVERSITYHamilton, Ontario

\

, \

TITLE: Time-resolved'Tunable Diode Laser Detection of t~e

Products of the Infrared Multiphoton Dissociation.'~f CF2HCI, (CF3)2CO, and CaCl 3

of

AUTHOR: John Joseph Orlando, B.Sc. (McMaster University) ,

\

(

ii

i

,.,i

--..J

ABSTRACT

./ -In the, experiments"descr'ib~d in thesis,' tunable

u

diode lasers are used to detect infrared absorption by the\

transient species produced in the infrared multiphoton

disltociation (IRMPO) of CF 2HCl, /CF3)2CO, and CHCl 3·.

The time-resQlved infrared absorptions ,of CF 2 and

HOI are detected following' the IRMPD of CF 2HCl. /

,em

of the HCl allows one to det~rmine aI I

(9+'2)L ~ X 10-21,Quantification

\

infrared-labsorption. Iinestrength,

-1 rmolecule for the .93(9) line. From this linest.re~th,

the CF2

VI bandsirengt~ is found to be (3.4+0.8) ~ 10-18,

cm molecule-I. The rate c~nstant for CF2

recombination to

form C2

F4

is found to be (2.3~O"7) X.10-~4 cm3 molecuie-1

-1s

Time-resolvetl .detection of· CF3 , C2F6 , anCL,_CO</ •

following the IRMPD of (CF3)2CO establishes that the.

dissociation mechanism involves the production of one

-,

~s

em

found

From, this stoichiometry,

rr.- ~ 3

~ilculated 'to be :(8+2) X 10-17

rate constant for CF' recombination3

then1S

. The

molecule of co and two of CF3

"

h " ~h f h rt e 11nestrengt 0 t e R16(~0) CF 3 transition is

'-" 0 =-1to be (1.4+0.3) X 10 - 'cm' molecule and the

bandstrength

-1molecule.

iii

•

, "

..

10 - 12 cm3· -1-1measured to be (.z-:.2+0.Sf X molecule s at or

near the high pressure- limit, while the ra.te constant for.('

-1s

reacting

and

respectively.

wi th 02 .;.nd NO are

('"2 • 8~0 " 7 ) X 10 - 2 9"-.

f ouod to be . ( 2 .1 +0 • S )

cm 6 . molecule~2

x

~he use of TDLs in determiping the isotopic

selectivity of IRMPD processes is· then analysed. The

sensitivity of the t~chniqti~ is demonstrated in the

detectioQ of the DCI ~roduced in' the IRMPD of - natural

com~lications due to _

monitoring the DCI and HC 1 on a' short timesca Ie (tens

Cbundance CDCI] in CHCl 3 " However,

wall adsorption and desorption effects necessitate..,.

of

~s) to obtain meaningful quantitative information. It 1.s.

shown that the CC1 4 and much of the HCI produced 1.n the

photolysis of mixtures is ~he result of

j.sotopically non-selective reactions involving Cl ,aio,?

\-

\:iv

. 1

..

"

ACKNOWLEDGMENTS

I' would like to thank the following people for~

their contribution to this thesis :' •

(( Qr. Don Smith, for his guidance, throughout this

work;

Dr. JolJn Re{d, for lending his expertise in the

area of TDL analysis and'for helpful discussions regarding

,

all aspects of this work;

'"Atomic Energy of Canada Ltd., for the. loan of the

high repetition rate ,C0 2 \laser; ..

Dr. JoHn Johns of N.R.C., for confirming the

, method used in calculating the CFi ba~dstrength;

, N.S.E.R.C.• , ......for financial support in .the form of a•

post-grad~ate scholarship;

Mr. Carl Brown,' for his collaboration in the

development of the transient detection technique for CF 2

(Section 3.2) and CF 3 <Section 4.2) and in the measurement

of th~ CF3 recombina.tion r~te.-constant (Section 4.4.5),

his general ass~stance throughout the course of this work

~ and for making the lab an enjoyable place to work.

-0 Mr. Paul Beckwith, for his collaboration ~n the4

, development of the transient detection technique for CF,

v

\.

(S~~tion 3.2), the development of D&l detection and r in

particular, for collecting the data presented in figurer

5.4.

support;

My<J

parents and brother for ~heir continbal

The many friends who made my graduate life

that much more. rewarding;

manuscript and for her constant support and friendship .•

Brian fahie,

especially. Joan fahie,

for his advice and'~friendship, . and~

-lfor her help in proofreading this ..

?-

-",,",

vi

., •

J

p

TABLE OF CONTENTS

ABSTRACT

ACKNOWLEDGEMENTS

LIST OF FIGURES

LIST OF TABLES

CHAPTER

iii

v

x

. xx

1 INTRODUCTION

1.1 General

1.2 The IRMPD Process

1.3 Theory of Infrared Absorpt~on

1 . 4: SUmmary

1

1

5

24

28

,

2' EXPER"rMENTAL

2.1 Introduction

2.2 Gases ~nd Gas~Handling Tech~iques

30

30

30

2.3 Photolysis and Stable Pro~uct Analysis 31Techniques.

3.4 Tim~esolved Detection of Transients with 40TDLs'

2."5 Summary 47

vii

fI

~.

S. I

3 IRMPD OF CF 2HCl

3.1 Introduction

48

48

3.4.3 CF2 Kinetics

3.4 Results and Discussion

3.2 .E~periinental

3.3 }hotolysis Mechanism

59

5S

66

52

55

48

(

1.r

CF2 and HCl Vibrational Relaxation ,.... . -

LineJtrength and BandstrengthCalculations

3.4.2

3.4.1,

3.5 Summary 68

tIRMPD OF HEXAFLUOROACETONE 70

4.1 Introduction 70

4.2 Experimental<!:l

4.3 P¥ev10us Work

70

73

4;4 Results and Discussion 75

4.4.1 Photolys~~ Mechanism

4.4.2 Vibrational Relaxation of CF 3

4.4.3 Photolysis Yield as arFunctionof Fluence

75

80

80 .

4.4.4 CF3 .Linestrength and BandstrengthCalculations

81

.'4.4.5 CF

3Kinetic stud1es

4.5 Summary

85

92

viii. c.

5.4.3 Time-Resolved TDL Detection~ofDCI/RCl ' ,

5.4.1 Effect of Additives on ProductYield in ~DCl3 IRMPD

5.4.2 ;DCl/HCI Detection Following IRMPDof Chloroform

.,

95

9.5

96

98•104

104.".~\

110

117

121

C--123

I

\

5 • 2 Exper imEmta 1

5.4 Results and Discussion

5.3 Previous Work

".5.5 Summary

6 CONCLUSIONS

" 5 IRMPD O~ CDCl3

/CHCl3

5.1 Introduction

\.....

~6.1 Introduction 123.

6.2 Ti~e-Resolved~Detectionwith TDLs

6.3 Determination of the IRMPD Mechanis~

124

125

6.4 Linestrengthand Bandstren~th calculatio~ 127

6.5 Kinetic~studies of Transie~t Species 12~-~ I ~

6.6 Laser Isotope Separati911 131,

6'.7 Summary 132

REFERENCES ' ,J:

APPENDIX A f

- ).,

,

133

'148

}

ix

NO.

2.1

LIST OF FIGURES

Experimental dlesign for static . sample'\

photolyses. The CO2 laser beam is

focussed by means of a 25 cm BaF2 lens

PAGE

32

into the cell. Polyethylene sheets are

2.• 2

.. used as necessary to obtain the desired

CO 2 laser pulse energy .•

Schematic diagram of TDL stable product

analysis apparatus. The TDL beam ~s

~ .focussed into the cell by means of LI ,

aqd then focussed onto ~he IR detector

using L2 • The TDL beam is cnopped· by the

36

mechanical chopper. Detection 4S done

2.3

using a. lock-in amplifie~ whose phase

matches that" of the chopper.

Calibration plots ~or CO dnd C,O-CHCl 3

mixtures~- Line A, shewing detection of

p~re CO, gives a.linestrength for R<8l in

38

agreement with the literature. Line B '\

shows the effects of pressure broadening

x ...

~ ,...

....

on the co absorption coefficient. The

data are consistent with a pressure­

broadening coefficient of 8.0 X 10-5 cm- 1

-1Torr ' as measured in our lab. Line C,

with detection done on R(6), shows that

the measured absorption coefficien{ (by

either direct or second harmonic

detection) is linear over a ra\1ge of co

pressures of greater than thre~ orders of

magnitude. The inset defines the

parameters I and I 0'used in the

measurement of (l .0

2.4 Schematic

apparatus.

~~ .of TDL transient detection

Lens L1 is ~sed to focus the

41

The two beams are

TDL beam, while lens L2 collimates the

-CO2 laser beam.

combined on beamsplitter M1 and focussed

on L 3 into the capillary cell. The two

beams are then separated on a diffr~ction

grating, with the TDL beam then' being

focussed onto the infrared detector,. and

the CO 2 laser beam being focussed onto an

energy meter.

xi

2.5 Idealized

(

schematic of the detector ..output in a transient detection

experiment. In each modulation cycle,

the absorption feature ~ sampled twice,

once in each direction.

2.6 Demonstration of .the advantages of 46

background subtraction in TDL transient

spectroscopy. Note that the subtraction

3.1

removes the precursor absorptions and.

enhances the CF 2 signal. The CF 2 is ~

produced by photolysis of 1.5 Torr of

C2

F3cl, and both the signal and

background represent the average of 50

scans.

FT-IR spectrum of 2 Torr of CF2HCl.~, The

pathlength is 10 cm, while the FT-IR

resolution- is set at i em-1

50

3.2 Transient TDL aQsorption by HCl. The 51

absorption feature monitored is P(9) at

2677.732 -1em The HCl is produced by

photolysis of 800 mTorr of CF 2HCl..:J

". '';'

xii

3.3 . )' spectrum

-1cm The

of CF 2 in the region of 1243

figure shows the absorption

53

lines created in a microwave discharge by

Davies ~38 and in lab for~. , our

compar~son. The lower trace shows an

expanded view of a portion of the

sP~7m.."

t

3.4 TDL transient absorption } t showing 58spec rum

the presence of the CF 2 triple~ near 1243..-1

cm CF2 is produced by irradiation of

~OO mTorr of.CF2 HCI. The TDL sweeps ovec

three CF2 line~ at a modulation freguen~y

of 40 kHz. The strongest absorption line

corresponds to about 20% absorption .. The

growth of the two outer lines relative to

the central line of the triplet is

attributed to rotational relaxation. The

overall growth of the·, triplet is

3.5

attributed to cascading of vibrational I y \(::f

excited CF 2 into-the vibrational ground

f state.

,

CF2

and Hel yields (in terms of the

xiii

61

\

absorption per cm, B ) as a function ofo

the average CO2 laser fluence. The

CF2 HCl pressure 19 800 mTorr.

3.6 Partial pressure of CF2 and C2 f 4 as a

•function of time f~llowing photolysis of _

800 mTorr~ of CF2HCl at an average CO2

laser fluence of 17 J cm- 2

"'---(

1.7 second order plot (the inverse of the CF2 67

linecenter absorption coefficient, B ,o

versus time) showing the decay of CF~ via

recombination. The data were obtained by

photolys~s of 800 mTorr of CF2 HCl with a

. -.?peak CO 2 laser fluence of 17 J cm -

4.1 FT-IR spectrum of ,,\.5 Torr of 72

-recorded at a resolution of

pathl~ngth of 10 ern.

\

over a

-is

-1'2 em

spec~rumThehexafluoroaeetone.

4.2 Transient tunable diode laser absorption 77

signal from CF3 created in the IRMPD of

hexafluoroaeetone. The detection i~ done

xiv

on the

1264.739

r R16

(20) CF3

absorption line

-1cm

at

4.3 Transient absorption near 1263-1

cm 78

showing CF3

decay immediately following

the CO2 .laser pulse, and the growth of

C2

F6 at later times.

CO(+) and C2 F6 (O) yields from IRMPD of 82

600 mTorr hexafluoroacetone as a

",

..4.5

function of the average incident CO2

lase~ fluence. The irradiations are done

with the lOR(~02' laser line.

Plot of CF3 absorption per em at

lineeenter, ao ' as a function of CF3

pressure. a for CF3is -rtfeasured at a

0

variety of ,CO2 laser f 1uences, -,and the

83

yields measured at each of these

pressure is obtained from the CO and

fluences and a~ "

knowledge of the

l 4.6

dissociation stoichiometry.

second order decay plot (the inverse of

xv

90

'.

4.7

the absorption coefficient per cm versus

time) for the recombinatio~ of CF 3 . CF3

is created from the IRMPD of 600 mTorr of

hexafluoroacetone. The 10R(14) laser

'line, with a fluence of 34 J cm- 2 is

emplo'yed.

plot 4f the inverse CF3

half-life (ms- i ) t:. 91

versus the concentration of added °2 or

NO (Torr) • CF 3 is created by the IRMPD

of 600 mTorr of hexafluoroacetone at a,

,c'rn-2CO 2 laser fluence of 34 J (on

10RnO) ).

5.1

11

An FT-IR spectrum of a mixture of 2.5

Torr CDC13

and 2.5 Torr CHC13

, recorded

in a 10 cm long Pyrex with a resolution

of 4 cm- l

97

5.2 plot of the produc;t yield of CCl 4 and 106

dissociation yield of CDCl 3 following the

photolysis of 1 Torr CDCl 3 in the

presence of additives. The irradiation

is done on lOP(48), with a'peak focal

. -2fluence of 10 J cm • Errors are +10.

xvi,<

o

'..

Line A: C~14 yield fol~owing irradiation-,

CDCl 3

tJJ

yield

Torrof 1irradiation

B:Line

of 1 Torr CDCl 3 with 1 Torr Ar ..Upper limit of CCl 4

~ith 1 Torr C2F4 •/

\"-

'\ .'"" following

"'-..,

Line C: CCl4 yield following irradiation

of 1 Tor~ 'CDCl] with 1 Torr Cl 2 .

Line 0:

)

CCl 4 !ield following irradiation

of 1 -Torr CDCl 3 with 5 Torr Cl 2 •

Line E: Average dissociation yield of

CDCl 3 ~in the above ~xperiments A through

D. The difference in the dissociation.,

yield in the four cases was not

. significant.

5.3 Relative DCI yield as a fundtion ·of\

. pho~olysis fr~quency fol~owing IRMPD of 1

Torr CDC1 3 in a 10 cm Pyrex cell. The

112

peak -2focal fluence is 10 J em ) in all

.'cases. The error bar represents ~10 •

5.4 Successive scans taken of a 5 Torr sample

~. of natural c'l,loroform after ,k~diatiOn

with 100 pulses of 10P(38) . an~10P(48)

-114 .

xvii

.". )

.1

I.-radiation (about 10 J -2cm )" The •

increase in DCI concen~ration after

irradiation with 10P(48) is attributed to

IRMPD of naturally occurring CDCl 3 in the

sample. The noise level 1S equivajent to

'- ~ about 0.8 ppm DC I.

5.5 Dissociation yield of CDCl 3 (lower line)

and total (HCI + DCI) yield following the

irradiatLon of a mixture of 0.1 Torr--

116

CDCl 3 and 1.9 Torr CHCI 3 . The photoiysis

1S performed at a peak focal fluence of

10 J-2cm with the 10P(48) laser

line. Th~ 'points labelled 1,2, and 3

represent the CDCl3dis~ciation yield 1n

three separate experiments, while the

points labelled a,b, and c represent the

(HCl and DCI) yield in the same three

"experiments.

5.6 Transient HCI yield following the single 120

pUl~~adiation of an equimolar mixture

of CDCI3

:CHCl3 :Ar (Line A) or

CDCl 3 : CSCl 3 : Ci F4 (Li1ne B) • The

irradiation is performed with tht:: 10P(38)

xviii.

I

•

.,

CO 2 laser line at ~peak focal fluence of

16 J cm-2 E 1rrors are + o.1;

I

//

/I

.'

xix