close

Вход

Забыли?

вход по аккаунту

?

Microwave measurements on transition metal and weakly bound molecular complexes

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.
University M icrofilm s International
A Bell & Howell Information C om pany
300 N orth Zeeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
O rd e r N u m b e r 9 3 22664
M icrow ave m easu rem en ts on tra n sitio n m eta l and w ea k ly b ou n d
m olecu lar com plexes
R oehrig, M ark A ugust, Ph.D .
The University of Arizona, 1993
UMI
300 N. ZeebRd.
Ann Arbor, MI 48106
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MICROWAVE MEASUREMENTS ON
TRANSITION METAL AND WEAKLY
BOUND MOLECULAR COMPLEXES
by
M ark A ugust Roehrig
A D issertation su b m itted to the Faculty of the
D E P A R T M E N T O F CHEM ISTRY
In P a rtia l Fulfillment of the R equirem ents
For th e Degree of
D O C T O R O F PH IL O SO PH Y
In th e G rad u ate College
T H E U N IV ER SITY O F ARIZONA
19
9 3
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
As members of the Final Examination Committee, we certify that we have
read the dissertation prepared by
entitled
Mark August Roehrig_________________
Microwave Measurements on Transition Metal__________________
and Weakly Bound Molecular Complexes
and recommend that it be accepted as fulfilling the dissertation
requirement for the Degree of
Doctor of Philosophy______________
'2(e
_______ .
"Dr. S.G. Kykolich
Dr. W.Rj. Salzman
uJU
Date
^
Dr. L. Adamowicz
F. Burke
Dr. S.W. Buckner
Date
- X
Date
Date
Date
Final approval and acceptance of this dissertation is contingent upon
the candidate's submission of the final copy of the dissertation to the
Graduate College.
I hereby certify that I have read this dissertation prepared under my
direction and recommend that it be accepted as fulfilling the dissertation
requirement.
/f y y l
'-Dissertation Director
Dr. S.G. Kukolich
iL '
Date
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
/ Q c fi
3
STA TEM EN T BY T H E A U TH O R
T his dissertatio n has been su b m itted in p artial fulfillment of requirem ents
for an advanced degree a t th e U niversity of A rizona and is deposited in th e U niver­
sity L ibrary to be m ade available to borrow ers u nder th e rules of th e Library.
B rief quotations from th is dissertatio n are allowable w ithout special p er­
mission, provided th a t accu rate acknow ledgm ent of source is made. R equests for
perm ission for extended q u o ta tio n from or rep ro d u ctio n of this m anuscript in whole
or in p a rt m ay be g ran ted by th e head of th e m a jo r d epartm ent or th e D ean of th e
G rad u ate College when in his or h e r ju d g m en t th e proposed use of the m aterial is
in the interests of scholarship. In all o th e r instances, however, perm ission m ust be
obtained from th e au th o r.
SIGNED:
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
4
ACKNOW LEDGM ENTS
To o b tain one’s goals is not a solo effort. M any people have influenced m e
in m any ways and I would like to acknowledge th em all. In one form or an o th er they
have helped me achieve w h at I have set o u t to do in my life. From my dear friends
back hom e in M ichigan to all th e new friends I have m ade in Arizona, all to some
degree, have contributed in realizing my goals. For this I am eternally grateful.
In each endeavor th ere axe always a few people who have, upon reflection,
m ade them selves more available and therefore in stru m en tal in achieving one’s goals.
O ne of those persons is Steve Kukolich, whose guidance and criticism has helped
me shape my outlook in science and m ade m e a confident b u t realistic researcher. I
have also learned from Steve th e im portance of com pleting a ‘p ro jec t’ an d looking
forw ard to th e next one. I wish to take th is o p p o rtu n ity and th an k him for his
tim e and effort in helping me com plete my ‘p ro je c t’. I also wish to th a n k Ludwik
Adamowicz who has always had th e tim e an d patience to interact w ith me during
my stay here in Arizona.
O ver th e years th ere have been two people who have always su p p o rted me
unconditionally, even w hen I h ad ‘crazy ideas on th e shore of a lake’. Those people
are G erd an d G erti Rohrig, my parents. I do n o t know how to even begin to th an k
them . It is due to th em th a t I have h ad th e o p p o rtu n ity to atte m p t m any things
in my life and because of them I can look back a t w hat I have accom plished w ith
the sense of who I am an d where I have come from . Thanks.
In the end there is usually one person who has m ade the m ost of any
situation. My one person is Gill. By being th ere she has m ade my life m uch m ore
interesting and h er care an d su p p o rt have given me so much th a t I do n o t know
w hat to say. However, w h at I can say is, I look forw ard to sharing a life w ith her.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
5
TABLE OF C O N TEN TS
L IST O F T A B L E S ..............................................................................................................
LIST O F FIG U R E S
9
............................................................................................................. 12
A B S T R A C T ...............................................................................................................................13
C H A P T E R I: IN T R O D U C T IO N ....................................................................................... 14
1.1 T ransition M etal C o m p le x e s ...............................................................................15
1.2 Overview of R e s e a r c h ............................................................................................17
C H A P T E R II: PULSED BEA M F O U R IE R T R A N SFO R M MICROW AVE S P E C ­
TRO SCO PY
...........................................................................................................................19
11.1 P roperties of Supersonic E x p a n s i o n s .............................................................19
11.2 Polarization and Em ission of th e E xpanding G as
.................................. 23
11.3 M achine D iagram an d D etails of O p e r a t i o n ................................................26
11.4 G as and Sample H andling
.............................................................................. 30
11.5 P B -F T S S tark E x p e r i m e n t ...............................................................................33
C H A P T E R III: Q U A D R U PO LE C O U PLIN G IN NOC1 AND C1F3 .................... 36
III. 1 NOC1 S pectrum
III.l.i
I ll.l.ii
............................................................................................. 37
Synthesis of N O C 1 ..........................................................................................39
P ulsed B eam M easurem ents
....................................................................39
III.l.iii M olecular Beam M aser M easurem ents
Ill.l.iv Hyperfine S tru ctu re and A nalysis
..............................................41
........................................................ 44
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
6
TABLE OF C O N T E N T S - continued
111.2 CIF3 Spectrum and S tru ctu re
......................................................................48
III.2.i E x p e r i m e n t a l ...................................................................................................48
III.2.ii CIF3 Hyperfine A n a l y s i s ............................................................................ 49
III.2.iii D a ta A n a l y s i s ..............................................................................................50
III.2.iv S tru ctu ral A n a l y s i s ....................................................................................... 51
III.2.V Discussion of CIF3
....................................................................................... 53
111.3 Townes Dailey In terp retatio n of Q uadrupole D a t a .................................. 54
C H A P T E R IV: MICROW AVE M EA SU R EM EN TS O F CO BA LT T R I­
CA RBO N Y L N ITRO SY L, C Y C LO PEN TA D IEN Y L COBALT DI-CARBONYL,
AND CY CLO PEN TA D IEN Y L M A N G A N ESE T R I-C A R B O N Y L
IV .l N uclear Q uadrupole Coupling in C o(C O )3NO
IV .l.i P ulsed Beam M easurem ents
..................... 58
.......................................59
......................................................................61
IV .l.ii R otational an d H yperfine A nalysis
........................................................ 62
IV .l.iii S u m m a r y .........................................................................................................66
IV .2 C pC o(C O )2 : H indered R o to r S p e c t r u m .................................................... 67
IV.2.i E xperim ental C onsiderations
IV.2.ii D ata Analysis
......................................................................67
................................................................................................ 68
IV.2.iii Q uadrupole C o u p l i n g ................................................................................... 73
IV.2.iv R esults and Discussion
...............................................................................73
IV .3 M anganese Q uadrupole C oupling in C p M n (C O ) 3 .................................. 76
IV.3.i D a ta Analysis
................................................................................................ 76
IV.3.ii E xperim ental R esults
................................................................................... 81
IV.3.iii S u m m a r y ......................................................................................................... 84
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
7
TABLE OF C O N T E N T S - continued
C H A P T E R V: M ICROW AVE S P E C T R A O F C Y C LO B U TA D IEN E IRON T R I­
CA RBO NY L, C Y C LO H EX A D IEN E IRON T R I-C A R B O N Y L, AND BU TA DI­
EN E IRON T R I-C A R B O N Y L ............................................................................................85
V .l C yclobutadiene Iron t r i - C a r b o n y l ................................................................. 85
V .l.i E xperim ental
.....................................................................................................87
V .l.ii R esults and D i s c u s s i o n ...................................................................................89
......................................................................................................... 93
V .l.iii Sum m ary
V.2 C yclohexadiene Iron t r i - C a r b o n y l ................................................................. 94
V.2.i Microwave Spectrum an d A n a l y s i s .............................................................95
V.2.ii S tru ctu ral P aram eters
V.2.iii R esults
................................................................................... 96
......................................................................................................
101
V.3 B utadiene Iron tri-C arb o n y l K raitch m an A n a l y s i s ...........................
104
V.3.i K raitch m an E quations an d M olecular S tru ctu re
105
..........................
C H A P T E R VI: M ICROW AVE S P E C T R U M , S T R U C T U R E AND D IPO L E M O ­
M E N T O F T H E H CCH-CO C O M P L E X ..............................................................
VI. 1 E xperim ental
109
.............................................................................................
109
VI.2 Spectral A n a l y s i s ........................................................................................
113
V I.3 S tru ctu ral A n a l y s i s ....................................................................................
114
VI.4 D j, Force C onstant, an d B inding E n e r g y ............................................
117
V I.5 D ipole M om ent
........................................................................................
120
......................................................................................................
121
V I.6 Sum m ary
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
8
TABLE OF C O N T EN TS - continued
C H A P T E R VII: SUM M ARY AND C O N C L U S I O N S ........................................
V II. 1 F u tu re D irections
.................................................................................
A P P E N D IX A.I: C1F3 T R A N SIT IO N FR E Q U E N C IE S
...............................
123
125
127
A P P E N D IX A .II: T R A N SIT IO N FR E Q U E N C IE S F O R CO BALT T R ICA RBO N Y L N ITR O SY L, C Y C LO PEN TA D IEN Y L CO BALT DI-CARBO N YL,
AND C Y C LO PEN T A D IEN Y L M A N G A N ESE T R I-C A R B O N Y L
. . . .
132
A P P E N D IX A .Ill: TR A N SIT IO N FR E Q U E N C IE S F O R C Y C LO B U TA D IEN E
IRON TR I-C A R B O N Y L , C Y C LO H EX A D IEN E IRO N TR I-C A R B O N Y L, AND
BU TA D IE N E IR O N T R I - C A R B O N Y L ..................................................................
144
REFERENCES
151
..............................................................................................................
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
9
LIST O F TA BLES
Table II I .l T ransition frequencies for N0C1 obtained on th e
P B -F T S m achine. 40
Table III.2 T ransition frequencies for NOC1 obtained on th e
m aser
s p e c tro m e te r .............................................................................................................................. 42
Table III.3 Values of the m olecular p aram eters for NOC1 obtained from fittin g th e
tran sitio n frequencies in Tables III.2 and III.3 ................................................................. 47
Table III.4 Values of th e m olecular param eters obtained for CIF3 ..........................51
Table III.5 Effective stru ctu re calculation for CIF3
.................................................... 53
Table III.6 N itrogen quadrupole coupling stren g th for some typical molecules an d
NOC1.............................................................................................................................................. 55
Table III.7 C hlorine quadrupole coupling stren g th and unbalance of p electrons. 56
Table IV .l M olecular constants for C o (C O )sN O ............................................................64
Table IV .2 M olecular constants for C pC o(C O ) 2 ............................................................. 71
Table IV.3 Some N-fold p o ten tial barriers......................................................................... 75
Table IV.4 B est p aram eters to th e hyperfine transitions for C pM n(C O )3 . . .
79
T able IV .5 C om parison of th e distortion param eters D j for some tran sitio n m etal
com pounds....................................................................................................................................80
Table V .l Best fit values for C bFe(C O ) 3 .......................................................................... 89
Table V.2 C om parison of d istortion param eters for some o th er tran sitio n m etal
complexes w ith C bFe(C O )3 ................................................................................................... 91
Table V.3 R o tatio n al and distortion constants for C-hexFe(C O )3
..................... 96
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
10
LIST OF TABLES - continued
Table V.4 C om parison of m easured rotatio n al constants and calculated constants
from X -ray d a t a ......................................................................................................................97
Table V.5 S tru ctu ral p aram eters describing relative orientation of the ligands in
C -hexFe(C O )3 .......................................................................................................................... 98
Table V .6 Selected interatom ic distances from X-ray and optim ized X -ray coordi­
nates
....................................................................................................................................
101
Table V .7 Selected intram olecular angles from X-ray and optim ized X-ray coordi­
nates
....................................................................................................................................
103
Table V .8 C om parison of interatom ic distances and angles between K raitchm an and
stru ctu re fit a n a l y s e s .....................................................................................................
107
Table VI. 1 M easured ro tatio n al constants for norm al and isotopically labeled
H CCH-CO
.......................................................................................................................
Table V I.2 B est fit results for th e isotopom ers of HCCH- CO
......................
112
114
Table VI.3 M easured an d calculated ro tatio n al constants obtained from th e least
squares fit to m om ents of i n e r t i a ...............................................................................
116
Table VI.4 S tru ctu ral p aram eters for H C C H - C O ................................................
117
Table V I.5 C alculated k s, u3, an d e for the HCCH-CO complex
118
..................
Table V I.6 C om parison of H C CH -C O dynam ical param eters w ith o th er CO com­
plexes
................................................................................................................................
Table V I.7 O bserved stark shifts for H CCH-CO
................................................
119
120
Table A .I.l M easured an d calculated tran sitio n frequencies for 35C1F3
. .
128
Table A .1.2 M easured an d calculated tran sitio n frequencies for 37C1F3
. .
130
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
11
LIST OF TABLES - continued
Table A .II.l M easured and calculated frequencies for the 3=2—>3, K = 0 tran sitio n s
of C o(C O )3NO
..............................................................................................................
133
Table A.II.2 M easured and calculated frequencies for th e J = 3 —>4, K = 0 tran sitio n s
of C o(C O )3NO
..............................................................................................................
134
Table A .II.3 M easured and calculated frequencies for th e J = 4 —>5, K = 0 tran sitio n s
of C o(C O )3NO
..............................................................................................................
135
Table A .II.4 M easured and calculated frequencies for th e J = 5 —>6, K = 0 tran sitio n s
of C o(C O )3NO
..............................................................................................................
136
Table A.II.5 M easured and calculated frequencies for th e J = 3—>4, K = 3 tran sitio n s
of C o(C O )3NO
..............................................................................................................
137
Table A .II.6 M easured and calculated hindered ro to r tran sitio n frequencies for
C pC o(C O )2
.......................................................................................................................
138
Table A .II.7 M easured and calculated K = 0 hyperfine tran sitio n frequencies for
CPC o(C O )2
.......................................................................................................................
Table A .II.8 M easured an d calculated
CPC o(C O )2
139
0 hyperfine tran sitio n frequencies for
.......................................................................................................................
Table A .II.9 Hyperfine com ponents for the J = 3 —>4 tran sitio n of C pM n(C O )3
140
141
Table A .II.10 Hyperfine com ponents for th e J = 4 —>5 tran sitio n of C pM n(C O )3 142
Table A .II.11 H yperfine com ponents for th e J = 5 —>6 tran sitio n of C pM n(C O )3 143
Table A .II.12 H yperfine com ponents for th e J = 6—>7 tran sitio n of C pM n(C O )3 144
Table A .III.l M easured and calculated frequencies for C bFe(C O )3
. . . .
146
Table A .III.2 M easured and calculated frequencies for 56Fe C -hexFe(C O )3
147
Table A .III.3 M easured an d calculated frequencies for 54Fe C -hexFe(C O )3
150
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
12
L IST O F F IG U R E S
Figure II. 1 Block diagram of th e A rizona pulsed beam Fourier transform microwave
spectrom eter
.......................................................................................................................... 27
Figure II.2 Tim ing diagram for th e pulse s e q u e n c e ..............................................29
F igure II.3 Sample FID and F T sp e c tra o b tain ed by th e A rizona m achine . . 30
F igure II.4 S tark p late calibration p lot
..........................................................................34
Figure III. 1 S u b stitu tio n stru c tu re of NO C l .................................................................38
F igure III.2 F T -sp ectru m of th e NOC1 J = 0 —>1 hyperfine t r a n s i t i o n ..................... 40
F igure III.3 NOC1 m aser s p e c t r u m .................................................................................. 42
Figure III.4 Block diagram for th e m aser spectrom eter
Figure III.5 S tru ctu ral p aram eters for C1F3
F igure IV .1 S tru ctu re of C o(C O )3NO
...........................................44
.................................................................52
..........................................................................60
Figure IV .2 S tru ctu ral p aram eters for C pC o(C O ) 2 ....................................................72
Figure IV.3 S tru ctu re of C p M n (C O )3
Figure V .l S tru ctu re of C bF e(C O )3
..........................................................................81
.............................................................................. 87
Figure V .2 S tru ctu ral p aram eters for C -hexFe(C O )3
............................................... 99
Figure V.3 C onform ation an d o rien tatio n of th e C-hexFe(CO )3in th e principal axis
s y s t e m ................................................................................................................................
102
Figure V.4 M olecular stru ctu re an d fitted param eters as determ ined from K raitch ­
m an a n a ly s is
Figure VI. 1 V ibrationally averaged stru c tu re of HCCH-CO
104
..........................
115
Figure VI.2 P lot of A u vs. e2 for H C C H - C O .........................................................
121
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
13
ABSTR AC T
High resolution microwave sp ectra for th e tran sitio n m etal com pounds
cobalt tri-carbonyl nitrosyl (C o(C O )3NO ), cyclopentadienyl cobalt di-carbonyl
(C pC o(C O )2 ), and cyclopentadienyl m anganese tri-carbonyl (C pM n(C 0 )3 ) were
obtained for th e first tim e using pulsed beam Fourier transform spectroscopy. A n
oblate sym m etric top spectrum was m easured for C o(C 0 )3N 0 and th e first gas
phase value of th e cobalt nuclear quadrupole coupling p aram eter was obtained.
T he asym m etric top hindered ro to r spectrum for C pC o(C 0 )2 was m easured and a
barrier to in tern al ro tatio n was estim ated from th e spectrum . Analysis of the prolate
sym m etric to p hyperfine spectrum of C pM n(C 0 )3 yielded th e first gas phase m ea­
surem ent of th e ro tatio n al constant and th e M n nuclear quadrupole coupling. High
resolution microwave sp ectra for th e iron containing tran sitio n m etal complexes
cyclobutadiene iron tri-carbonyl (C bF e(C 0 )3 ), cyclohexadiene iron tri-carbonyl
(C -hexFe(C 0 )3 ) were obtained and a K raitchm an analysis of the isotopic sub­
stitu tio n d a ta for th e butadiene iron tri-carbonyl (B uF e(C 0)3 ) is also discussed.
S tru ctu ral param eters for HCCH-CO were obtained from various isotopom ers for
this com plex. A n analysis of th e d isto rtio n p aram eter D j yielded an estim ation of
th e binding energy for this weakly bo u n d complex. A nalysis of sp ectra for nitrosyl
chloride (N 0C 1) and chlorine tri-fluoride (C IF3 ) yielded th e first high resolution low
J d a ta sets for these molecules. T he quadrupole coupling d a ta are interpreted using
the Townes- Dailey model for quadrupole coupling an d an im proved ground sta te
stru ctu re for CIF3 was obtained. Microwave sp ectra reported here were obtained
using a pulsed beam Fourier transform microwave spectrom eter constructed a t the
U niversity of Arizona. T h e design is sim ilar to original Flygare-Balle ap p ara tu s
w ith m any m odifications for im proving signal sensitivity and d a ta aquisition.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
14
C H A P T E R I:
IN T R O D U C TIO N
Microwave spectroscopy has been used for m any years to o b tain very pre­
cise stru ctu ral m easurem ents for m olecular system s. To this day, it is considered to
be one of th e highest resolution forms of gas phase m olecular spectroscopy known
w ith linew idths as narrow as 2 kHz m easured from microwave m aser spectrom eters
[1]. N ot only do th e small linew idths con trib u te to th e overall precision of stru c­
tu ral m easurem ents, it has also co n tributed to th e m easurem ents of small hyperfine
interactions of nuclear properties of atom s contained w ithin the molecule.
Over th e years, large advances have been m ade in th e technology used in
microwave techniques. In the last ten years an in stru m en t was developed by Flygare
and Balle [2] th a t has brought new life in to microwave spectroscopy. By combining
the techniques of free je t expansions and F ourier transform spectroscopy, the study
of van der W aals molecules, which was pioneered by K lem perer in th e microwave
region in th e 1970’s, has exploded. C onditions in th e gas expansion are such th a t
van der W aals molecules are readily form ed an d stabilized long enough so rotatio n al
sp ectra can be obtained. Today m ore th a n 60 % of th e high resolution microwave
d a ta of these complexes has been o b tained by th e P B -F T S technique from groups
around th e world.
S tudy of van der W aals molecules continues today w ith experim ents de­
signed in microwave and other regions of th e electrom agnetic spectrum an d even
m any ab initio and semiempirical calculations have been stim ulated by th e large
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
15
am ount of d a ta obtained from PB -FT S. However, th e P B -F T S technique lends itself
to stu d y m any o th er areas of molecular spectroscopy as well. S tartin g in th e early
1950’s groups have been trying to obtain high resolution sp ectra of larger more
com plex molecules like transition m etal complexes. T he early studies resulted in
obtainin g m inim al am ounts of spectral inform ation, often obtaining only approxi­
m ations of ro tatio n al constants. The fundam ental obstacle was th e large m om ents
of in e rtia which produced closely spaced ro tatio n al transitions in th e sp ectra m aking
th e sp ectra difficult to assign.
C om bining P B -F T S w ith th e study of tran sitio n m etal complexes and other
large molecules has proven m ost successful as dem onstrated by th e results presented
in this dissertation. New inform ation on stru ctu re, bonding and nuclear interactions
have been o b tain ed for these com pounds and will add to th e overall understanding
of th e n a tu re of these com pounds.
1.1 T ran sition M eta l C om plexes
T ran sitio n m etal or organom etallic complexes combine organic com pounds
and tran sitio n m etals in a variety of ways. From binding of th e organic com pound to
a tran sitio n m etal center, th e reactivity or stability of th e ligand can be significantly
altered. For exam ple, free cyclobutadiene does not exist, b u t when com plexed to
F e(C 0)3 it is greatly stabilized and can be exam ined experim entally in m any ways.
T h e converse is also true, stable com pounds can combine w ith tran sitio n m etal com­
p ounds an d be activated to react [3] often in some predictable and beneficial way.
S ynthetic chem ists tu rn to organom etallic m ethods to exploit th eir special reactiv­
ity and selectivity th a t cannot be found in conventional m ethods. Therefore, there
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
16
is a precedence in obtaining new inform ation on tran sitio n m etal complexes which
could help in th e overall u n d erstanding of these com pounds and th eir properties.
Over th e m any years tran sitio n m etal chem istry has been studied, a num ber
of factors have been identified th a t seem to form a basis for a rationalism un d er­
lying the understanding of tran sitio n m etal reactivity. One of those factors is ‘th e
geom etrical restrictions to th e orientation of bonding of ligands in m any tran si­
tion m etal complexes. T his determ ines th e alignm ent of reacting species brought
together by th e m etal, so providing control of b o th chemical selectivity and stereos­
electivity’^]. Loosely tran slated , th e stru ctu re of these tran sitio n m etal complexes,
especially knowing how the ligands are oriented ab o u t th e m etal center, can im ­
prove insight into how th e reacting species interacts w ith th e complex. O btaining
very precise d a ta sensitive to th e stru ctu re of these molecules will aid in th e general
understanding of tran sitio n m etal complexes.
O ne p articu la r field of tran sitio n m etal chem istry can greatly benefit from
knowing detailed inform ation of th e geom etries of these complexes, catalysis. Syn­
thetic m ethods th a t are catalyzed by soluable tran sitio n m etal complexes of Co,
M n, an d Fe in th e form ation organic com pounds are used extensively in indus­
try ^ ].
M ost of th e stru ctu ra l inform ation on these complexes comes from solid
state X -ray diffraction studies. T hough very accurate in determ ining the locations
of th e large atom s in th e com plex, sm aller atom s of th e ligands such as hydrogens
are undeterm ined by X-ray diffraction. T he o rientation of th e hydrogens often indi­
cate some unusual bonding interaction betw een th e ligand and m etal center. Since
X-ray diffraction is m easured in th e solid state, th e crystal packing of th e solid pertu rb es th e geom etry of th e complex. Sometimes these affects are small b u t other
tim es they can be quite large. Therefore, gas phase m easurem ents of tran sitio n
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
17
m etal complexes is very desirable to provide inform ation on th e intrinsic geom etries
of these molecules.
1.2 O verview o f R esearch
T he studies of th e stru ctu res of some tran sitio n m etal complexes presented
here are th e beginnings of a research program involving th e use of pulsed beam
Fourier transform microwave spectroscopy (P B -F T S ) and tran sitio n m etal com­
plexes. T he goal was to o b tain high resolution sp ectra for these com pounds so th a t
more detailed an d perhaps new inform ation could be obtained on th e ir stru ctu res
as well as nuclear p roperties of several of the tran sitio n m etals involved. T h ro u g h ­
out th e research advances were m ade in th e handling of these com pounds so as to
provide a knowledge base for fu tu re studies of m ore reactive species like tran sitio n
m etal hydrides.
Studies involving th e interhalogen com pound CIF3 and th e n itrosyl NOC1
were m ade in o rder to o b tain a high resolution d a ta set of low J tran sitio n s w here
the com plicating effects of nuclear quadrupole coupling is som ew hat simplified. A
m odel in terp retin g th e quadrupole coupling d a ta in term s of an unbalance of b o n d ­
ing electrons was then applied to this high resolution d a ta set to acquire an accu­
ra te description of th e chemical bonding in term s of the in teractio n of th e nuclear
quadrupole m om ent an d overall rotation.
S p ectra acquired on th e HCCH-CO weakly bound com plex were m ade to
establish th e ground sta te vibrationally averaged stru ctu re of this com plex. A n
analysis of th e d istortion p aram eter D j was m ade to investigate th e weak interm olecular dynam ics of this system by estim ating th e weak bond dissociating energy.
This study com plim ents some of th e previous work done on weakly bo u n d m olecular
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
complexes in th e Kukolich lab o rato ry at th e U niversity of Arizona. T h e stu d y of
these complexes is ongoing w ith m ore an d m ore em phasis on com bining experim en­
tal and theoretical approaches. A n excellent review on polyatom ic weakly b o u n d
complexes can be found elsewhere [5,6].
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
19
C H A P T E R II:
PU L SED BEA M F O U R IE R TR A N SFO R M
M ICROW AVE SPE C T R O SC O PY
T he technique of pulsed b eam Fourier transform microwave spectroscopy
(P B -F T S ) h as been around for m ore th a n ten years [2]. T he m olecular system s
available to its scrutiny are are alm ost unlim ited w ith the determ ining facto r being
th a t of successfully introducing a sam ple in tact into the ap p aratu s. T he d a ta ob­
tain ed from these m achines is of unquestionable quality in term s of sensitivity and
resolution. In th is C h ap ter, illu stratio n of this technique will be achieved th ro u g h
a m oderate discussion of th e fundam entals of operation from supersonic free jet
expansions, po larizatio n of th e expanding gas, detection of th e m olecular signals
to d a ta aquisition and analysis. For details on th e construction of th e ap p ara tu s
please see reference [7] .
I I .1. P r o p e r tie s o f S u p erson ic E xp an sion s
To u n d erstan d th e basic properties of a supersonic expansion, it is w orth­
while to briefly co n trast it w ith th e m ore fam iliar properties of an effusive flow
through a nozzle source. To acquire effusive flow through a nozzle source, each
p article’s m otion m ust be independent of all oth er p article’s, th a t is, no collisions
occur betw een th e particles to alter th e ir energy in some way. For effusive flow to
exist thro u g h a nozzle of d iam eter D , th e condition is,
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
20
» 1
C1)
w here A' is the m ean free p a th of th e particle, and
is known as th e K nudsen
num ber. T he m ean free p a th is given by
A '= (2 ^tt p„d2)-1
( 2)
w here d is a m olecular diam eter. For an effusive source pn = p0, th e source or
stag n atio n m olecular density. For a typical m olecular diam eter of d « 4 x lO -8 cm
an d for 1 atm osphere, th e m ean free p a th is 4.3 x lO -5 mm. For a fairly stan d ard
nozzle diam eter of 1 m m used in pulsed beam m achines of today and stag n atio n
pressures ranging from 0.5 to 2.0 atm ospheres th e gas exiting from the nozzle under
these conditions is far from th e effusive lim it.
As we can see, the effusive flow
essentially retains the characteristics of th e stag n atio n reservoir, w ith th e difference
betw een th e gas in the reservoir an d th e gas exiting the nozzle being th e change
in M axwellian distribution of m olecular velocities, th e distribution is now only over
a single velocity com ponent of th e one coinciding w ith the exit hole of th e nozzle
source.
In the pulsed beam ap p aratu ses of today, th e gas exiting th e nozzle u n d er­
goes drastic changes in its p roperties w hen it is com pared to the reservoir conditions.
These changes can be a ttrib u te d to w hat is occurring in the nozzle as th e gas exits
the reservoir. T he gas does not leave th e reservoir unhindered, b u t rath er, u n d er­
goes m any collisions w ithin th e nozzle itself. These collisions have th e effect of
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
21
interconverting the stored in tern al energy in th e form of ro tations and vibrations
into m ass flow thus adiabaticlly cooling th e gas. T he effective ro tational and vibra­
tional tem p eratu res drop, th e velocity d istrib u tio n narrows and moves o u t along
th e flow velocity axis th ro u g h th e nozzle. As th e gas exits th e plane of th e nozzle,
it expands freely undergoing no m ore collisions w ith itself thus producing relatively
low tem p eratu res w ithin th e expansion[8] .
It is ap parent from th e ro tatio n al and vibrational cooling described above,
th a t to a high resolution spectroscopist, th e supersonic free je t expansion is an
indispensible an d powerful tool in th e aquisition of com plicated sp ectra of unique
m olecular species such as tran sitio n m etal and weakly bound complexes. T he ro ta ­
tional an d v ibrational congestion in th e m olecular sp ectra is dram atically reduced
by th e free je t expansion.
T h e discussion above was based on equations describing free jet expansions
in th e continuous or equilibrium conditions [8-12 ]. T he supersonic free je t incor­
p o rate d into th e P B -F T S is used in a pulsed fashion and th e duration of tim e th e
valve is actually open is very short an d th e gas is expanded into a high vacuum , th e
gas achieves equilibrium expansion pro p erties very rapidly under these conditions
an d these equations will hold. In addition, th e gas pulsed into the vacuum of th e
cham ber is m ixture of a few percent of sam ple gas in typically 1 to 2 atm ospheres of
a carrier gas such as argon or neon. Therefore th e characteristics of the expansion
are dom inated by the properties of th e carrier gas.
T h e m ajo r considerations of th e free je t expansion are the tem p eratu re of
the gas, its spatial d istrib u tio n during th e expansion and th e num ber of particles
entering th e vacuum. T he tem p eratu re an d sp atial distrib u tio n was described above
and th e num ber of particles entering th e vacuum is a vital factor th a t m ust be
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
22
considered as well since, it is obvious, th a t if th e num ber density of sam ple molecules
is not high enough during th e expansion th en detection of th e m olecular signal will
not be possible or a t least very difficult. T h e num ber of molecules released during
a gas pulse is given[ 2, 8] by,
N = poa0>ln<l,(0.31)
(3)
w here po an d ao are the source num ber density an d local speed of sound, A n is the
nozzle area, t v is th e tim e the valve is open. T he factor of (0.31) takes into account
for th e discharge coefficient for a m onatom ic gas an d th e flow velocity of th e gas
a t th e nozzle in term s of the M ach num ber [ 8]. Typical values for th e quantities
are, t v = 3 ms, A n = 7r m m 2 (for a nozzle d iam eter of 1 m m ), and for 1 atm and
300 K po = 2.69 x lO 19 m olecules/cm 3, there are 4 x lO 18 particles released from
th e nozzle into th e cavity. Considering th a t th e m olecules are travelling around 3.8
x lO 4 cm /s, there are ab o u t one te n th of this to ta l in th e beam w aist and since it
is a seeded beam there are usually only 1 to 5 percent of th e num ber in th e waist.
If we consider th e lim iting condition of form ing dim ers in th e expansion then this
num ber drops fu rth er still. T here has been m uch theory an d experim ent [13- 17] on
th e form ation of dim ers in an expanding gas and th e sim ple expression p 2D can be
used to predict th e am ount of dim er form ation in th e expansion, p an d D are the
pressure or num ber density a t th e nozzle and nozzle diam eter respectively. From
this expression and experim ental results [14, 15], dim er mole fractions as high as
0.1 have been reported. T his is sufficient for detection of m olecular signals using
microwave spectroscopy.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
23
T h e above lim iting case exam ple of dim er form ation was used to illu strate
th a t even u n d er th e conditions of th e free je t expansion, th e m olecular num ber
density o f unique m olecular species is sufficient for detection. A nd while m ost of
th e experim ents discussed here involve only single m onom er detection, th e effect
of dim er form ation is ever present to some degree. Therefore for every experim ent
th e conditions used in th e stag n atio n reservoir are very critical to th e successful
m easurem ent of th e m olecular spectra. M ore detail will be given to th e reservoir
conditions in Section Il.iv M odifications to P B -F T S .
I I .2. P o la riza tio n and E m ission o f th e E xp an d in g G as
A n u n d erstanding of th e free je t expansion p roperties is necessary to work
out tim ing schemes for pulsing th e microwave rad iatio n into th e F abry-P erot cavity
an d polarizing th e gas w ith th is radiation.
A m acroscopic polarization of a gas
containing a num ber of nondegenerate two-level q u an tu m system s w ith a static
electric field has th e form [2]
p (r, v ,f) = (pr (r, v, t) + ipi( r, v , t) ) ) e x p ( i u t) + c.c.
(4)
w here p r an d pi are real valued functions of th e six phase- space coordinates r ,
v , an d of t, an d ‘c.c.’ is the complex conjugate. Assigning each coordinate r of
th e gas w ith a unique velocity v ( r ) , th e polarization com ponents p r an d p, an d
th e two-level po p u latio n difference p er u n it volum e associated w ith the velocity v ,
A n ( r ,v ,i) , satisfy a set of B loch-type coupled p a rtia l differential equations [2]:
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
24
(Wt + v ’v) Pt + Au}pi + Pr^T2 = 0
(jft + V ' V) ?i ~
+ K2£(r’^ ~ j ~ +Pi/T2= 0
d
_ \ hAn
. .
h(An — A n n )
( a i + v ' v ) ~ 4 ~ - e ( r ’<) Pi +
4fi
~°
/
^
(6)
„ x
«
w here Aui = u Q —cj, th e difference between th e m olecular tran sitio n an g u lar fre­
quency u)0, an d th e carrier or stim ulating angular frequency u>, T\ an d T2 are phenom enologically introduced first- order relaxation term s, Ano is th e equilibrium
value of A n in th e absence of an electric field and is considered to be co n stan t,
k
= 2Ti~1\(a\fi,\b)\
and
(8)
w here (a\/j.z \b) is th e dipole tran sitio n m om ent connecting th e two levels.
T hese equations describe th e interaction of th e electric field or th e m i­
crowave pulse w ith th e gas pulse. T h e gas is pulsed into a F abry-P erot cavity w hich
is inherently a narrow bandw idth, high electric field in stru m en t, so solution to these
equations can be found by considering only the on resonance, Acu= 0, form of these
equations. W hen th e po larizatio n pulse length, r p, is short enough, th e m olecules
travel only a sh o rt distance when com pared to th e characteristic distance of varia­
tions in p r and pi, and A n , and m aking tp < < T j , T 2, th e term s involving v • V and
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
25
Ti and T2 m ay be neglected in th e above equations. T he solutions to th e resulting
equations are:
p r( r ,t ) = 0
P i ( r , t ) = - K^
n ° s in ( k
A n ( r , f ) = A n 0cos
I
(9)
£(r
e ( r , t ' ) d t ,S^
(10)
( 11)
T h e ap p ro p riate form of e ( r , i ) depends on the T E M 0og cavity m ode of th e FabryP erot cavity. T h e derivation of th e form for e (r, t) is given in detail in [2] an d only
th e results will be given here. Using th e results from [2] an d su b stitu tin g them into
the above equations, give a t th e end of the pulse du ratio n r p,
,
Pi ( r, t ) =
nhAno
----------------- —
( 12)
x s in ( k E 0tp
exp(—p2/uj2)cos(ky — 7r q /2)^
w ith a sim ilar expression for A n.
A fter th e polarizing rad iatio n has been removed, p r , pi, an d A n continue to
evolve as described by th e above equation w ithout the e ( r , t ) term s thus producing
an electric field w ithin th e F abry-P erot cavity.
Again, details of th e derivation
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
26
of th e em ission process are given in [2] and presented here are th e results of this
derivation. T he electric field produced by the em itting molecules can be w ritten in
the following simplified form:
e(t) = ex p (—t / T 2 )cos[u}0t + (w —o>0)]/(£)
(13)
w here t\ is th e tim e a t th e end of th e polarization pulse an d ui0 and u> are the
m olecular tran sitio n an d polarizing angular frequencies, respectively. T he function
I ( t ) contains all th e lineshape inform ation, while the rem aining term s provide an
exponential dam ping and center th e emission at u 0 [18].
T h e envelope function I ( t) accounts for am plitude variations of the em itted
signal characteristics of th e p articu lar cavity m ode, th e m olecular density d istrib u ­
tion in th e gas phase in th e gas pulse at tim e t and two types of signal dam ping.
T he first ty p e of dam ping corresponds to all fall off due to m otion of the molecules
out of th e cavity and has been shown to be small [ 2]. The second, and m ajo r type
of dam ping has been called D oppler dephasing and results from the m ovem ent of
molecules from a region w here they were polarized w ith one phase to regions w here
they would have been polarized w ith a different phase [19]. This is th en the m olecu­
lar signal being em itted in to th e Fabry-P erot cavity which in tu rn is detected, saved
and Fourier transform ed to yield th e spectral inform ation desired. D etails are given
in th e Section Il.iii.
I I .3. M ach ine D iagram and D e ta ils o f O peration
T h e essential theory behind th e operation of th e P B -F T S was given in the
previous sections an d presented in this section are the details of o p eratio n of the
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
27
Arizona M achine. A block diagram of the spectrom eter is shown in F igure II. 1.
T he gas or sam ple handling will be discussed in more detail in Section Il.iv an d
this section is concerned w ith only th e details of integrating th e gas pulses and th e
microwave stim ulation and subsequent detection and d a ta handling elem ents such
as storing the signal and Fourier transform ation.
SUPERSONIC
JCT
PIN 1
.
PIN 2
f a b r y - P e r o t C a v ity
□SC.
MIXER
20 M H z
MIXER 2
AMP,
“ ♦'moll
W ^ 7 SCOPE
IBM PC
+ 20 M H z
K1XER 3
AMP
Figure II.1 Block diagram of th e A rizona pulsed beam Fourier tra n s­
form microwave spectrom eter.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
28
A stim ulating frequency in th e 4-18 GHz range is generated by locking a
harm onic frequency generated in th e 800 to 900 MHz range by a H ew lett Packard
frequency synthesizer to a solid s ta te oscillating device. For frequencies generated in
the C -band range, 4 to 8 GHz, a YIG tu n e d tran sisto r oscillators (A vantek 7443 and
7553, 30 and lOOmW respectively) are used. In X- band, 8 to 12 GHz and K u-band,
12 to 18 GHz, th e oscillators are Y IG -tuned G unn diode devices (W atkins-Johnson
W J-5008-303F in X- band and W J-5041-303F in K u-band bo th rate d at 30 m W ).
A gas pulse is ad m itted to th e F abry-P erot cavity through through a solenoid valve
(G eneral Valve Corp. 9-181-100) resulting in a supersonic expansion w ith rotatio n al
and vibrational cooling. A fter a short period of tim e (75 fis to 3 m s), a 0.1 to 1.0 fis
d u ratio n stim ulating microwave pulse is coupled into th e cavity through a 1 /4 - wave
an ten n a m ounted in th e center of one of th e m irrors by opening a pin diode switch
(PIN 1) shown in Figure II.1. T he microwave pulse polarizes th e gas molecules as
described in Section Il.ii and th e d u ratio n of th e pulse, r p, is defined by how long
PIN1 is left open. This results in a tt/ 2 pulse and a superposition of rotatio n al
states will subsequently decay em ittin g rad iatio n a t th e microwave frequency into
the cavity in the form of a free induction decay or FID .
A fter P IN 1 closes there is tim e delay before opening PIN2 to allow th e
energy stored in th e cavity to decay away. This known as cavity ring and does
not include any signals from th e polarized gas which decays at a much slower ra te
th a n the cavity ring. PIN 2 th en opens coupling the FID out of the cavity through
the sam e 1 /4 -wave an ten n a and into a superheterodyne detection system . T he
tim ing sequence [ 7] of th e pulse valve an d P IN switches is shown in Figure II.2.
T he m olecular FID I'moiecuian w hich is in th e GHz frequency range, is th en m ixed
w ith a local oscillator which oscillates a t U[oc = ji'stim i 20 M H z | which produces
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
29
an interm ediate frequency z/Jn< = 20 M H z + |t/atim ± vmolecular\. uint is filtered,
am plified an d m ixed once m ore w ith a 20 M Hz frequency to produce an offset
frequency i/0f f = |r,stim — vmoiecuiar\ which is in th e 0 to 1 MHz range w hich is
am plified one m ore tim e to enhance signal sensitivity. T he u0f f is then digitized
into 4096 channels a t a 20 MHz sam pling ra te using a Physical D ata 522A transient
waveform recorder. The digital d a ta are transferred to an IBM P C where they are
averaged and saved onto a floppy diskette in th e form of th e digitized offset FID
signal v0f f - T he averaging sequence is such th a t gas pulses are applied on altern ate
microwave pulses an d altern atin g signals are added and su b tracted to produce the
average. T his has th e affect of rem oving m ost of th e spurious signals due to cavity
ring and switch noise. From th ere it is Fourier transform ed to yield th e relevant
spectral inform ation. A sam ple FID and F T sp ectru m is shown Figure II.3.
750us
GAS
PULSE
0.5 s
7 5 u s -3 n s
0.1-10 u s
PIN1
|-0.1-15us
PIN2
t=0
SEQUENCE
REPEATS
Figure II.2 T im ing diagram for th e pulse sequence. Taken from [ 7].
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
30
50.
100. ps
oo
F igure II.3 Sam ple FID and F T -sp ectru m o b tain ed from th e A rizona
P B -F T S machine.
II.4 . G as and Sam ple H and ling
T h e gas handling system which delivers th e carrier gas to th e pulsed valve is
described in m ore detail elsewhere [7]. D iscussed in this section are th e m odifications
m ade to p a rt of the gas handling system which allows th e stu d y of sam ples which
are not easily obtainable in th e gas phase, e.g. tran sitio n m etal complexes.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
31
M ost of th e tran sitio n m etal complexes discussed here are in a physical
sta te a t room tem p eratu re where obtaining sufficient vapor phase mole fractions is
negligible. These com pounds are usually in th e liquid sta te or, sometimes, in the
solid sta te a t room tem perature. Therefore to o b tain sufficient vapor pressure to
m easure gas phase spectra, one needs to h eat th e sam ple up in a reservoir prior to
pulsing into the cavity. However, ju st heating th e sam ple is n o t enough since once
coming into contact w ith cooler surfaces, th e sam ple will im m ediately condense out
of th e vapor phase due to its high boiling point. Therefore, a system is required in
which all th e surfaces th a t the sam ple m ay come into contact w ith, including the
pulse valve, be m aintained a t high enough tem p eratu res to prevent condensation of
the sam ple. T he basic is to heat th e sam ple an d pulse valve a t a constant elevated
te m p eratu re w ithin a stainless steel hot source. H eat is applied effectively through
h eat ta p e w rapped around and placed w ithin th e hot source and controlled by
several variacs. A n approxim ately 25 cm dia by 1 cm thick copper plate is placed
over th e top of th e hot source to help in m aintaining th e tem p eratu re. T h e sam ple
cells are m ade out of pyrex and are connected directly to th e pulse via Teflon Swedge
Lock fittings. T he construction of th e h o t source is based th e ideas and design of
S. Kukolich.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
32
W ith th e construction of th e h o t source, th e versatility of PB - F T S h as been
dem onstrated once more. T em peratures as high as 77 °C have been successfully
m aintained w ith little difficulty in the perform ance of th e pulse valve. However,
long exposure to elevated tem p eratu res seems to reduced th e lifetim e of th e pulse
valve com ponents requiring more frequent replacem ents of these com ponents. The
lim iting factor of this design seems to be th e high tem p eratu re failure of th e G eneral
Valve C orporation pulse valves. To obtain sp ectra a t higher tem peratures, perhaps
th e re- incorporation of th e once used fuel injectors m ay be necessary since these
devices are accustom ed to operating a t higher tem peratures. This is currently being
pursued.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
33
I I .5 P B -F T S Stark E xp erim en t
T h e S tark effect is a valuable tool to high resolution spectroscopists, since it
can be used to sort out ra th e r com plicated sp ectra by noting how certain tran sitio n s
behave w ithin an electric field, such as the first order S tark effect of a hindered ro to r
molecule. O btaining th e dipole m om ent fi of a molecule or weakly bound complex
will all ways be considered valuable inform ation because of how it relates to th e
fundam ental properties of th e molecule or complex. By incorporating S tark plates
into th e P B -F T S ap p aratu s one can potentially couple th e high sensitivity and
resolution of th e instrum ent w ith very precise m easurem ents of th e dipole m om ent.
S tark plates were constructed out of two 4 m m thick by 21.5 cm square
alum inium p lates where one side of each plate was polished to remove any surface
im perfections. T h e corners of th e plates were rounded w ith approxim ately a 2.5
cm radius so as to reduce th e potential of electrical arcing while in th e cavity of
th e spectrom eter. T he plates were suspended w ithin th e spectrom eter along the
centerline of gas expansion from the nozzle by m illed nylon suspension anchors
which a tta c h to th e same supports as the cavity m irrors do [ 7]. T he spacing of
the plates was chosen to be 10 cm so as to reduce th e p ertu rb a tio n of th e plates
w ithin th e electric field of th e F abry-Perot cavity. High voltage of up 6000 Volts
was applied to one of the plates so as to produce the S tark effect in th e molecules
during a gas expansion.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
34
T h e plates were calibrated by m easuring th e S tark shifts in 160 12C 32S
which has a very well know n experim ental value for its dipole m om ent. A calibration
plot is shown in Figure II.5 from which th e electric field and p late sep aratio n can
be determ ined.
0.20
N
a
a
<
0 .0 8
0.0 6
0.0 4
0.02
0.00
0
1040
2080
3120
4160
5200
Volt2
Figure II.4 S tark plate calibration plot for 160 12C 32S in th e P B -F T S
A rizona machine.
T h e results of th e S tark experim ent were below expectations. T he p e rtu r­
b atio n of th e electric field w ithin th e cavity due to th e presence of th e p lates proved
to be too great. S atisfactory sp ectra for molecules were h ard to o b tain since m ost
low J tran sitio n s for molecules w here th e S tark splitting is sim pler to analyze occurs
in the C -band range. O btaining sp ectra in th e C -band is inherently difficult w ithout
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
35
the presence of th e Stark p lates since th e long wavelengths of C -band produce a
lim ited nu m b er of m odes in which to tu n e the cavity. In the future, it is desirable to
change th e design of th e p lates so as to reduce this p ertu rb atio n w ithin th e cavity
and hopefully yield b e tte r results. A successful dipole m easurem ent was m ade on
the H CC H -C O weakly bo u n d com plex and those results are discussed in C h ap ter
VI.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
36
C H A P T E R III:
Q U A D R U PO LE C O U PLIN G IN NOC1 AND C1F3
Analysis of nuclear quadrupole coupling hyperfine sp ectra can provide in­
form ation ab o u t th e electronic environm ent around th e coupling atom w ithin a
molecule. T he Tow nes-Daily[20] m odel allows one to in terp ret th e quadrupole cou­
pling in term s of unbalanced p-electrons of th e coupling atom . T he electric field
gradient around an atom which gives rise to th e nuclear quadrupole coupling in
m ost molecules is due prim arily to an unequal filling of the valance p-orbitals in th e
coupling atom . T he results of this m odel are sum m arized below;
(UP) r =
~ ».
(Up), =
- n,
(14)
where n x, n y, and n z are the occupation num bers of their respective orbitals. (Up)g
is the unbalance of p-electrons along th e corresponding ‘g’ reference axis. Up is de­
fined such th a t a positive value corresponds to a p-electron deficit along its reference
axis and a negative value indicates a p-electron excess along the reference axis, thus
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
37
providing a sensitive tool in which to investigate th e electronic environm ents w ithin
molecules containing quadrupole coupling atom s.
I I I .l NOC1 Sp ectrum
T he hyperfine stru ctu re of nitrosyl chloride (N0C1) was first investigated
by Millen and Pannel[21] using a low resolution microwave spectrom eter. They
obtained isotopic su b stitu tio n d a ta which was used to determ ine the m olecular
stru ctu re of the molecule (see Figure I II.l). Chlorine quadrupole coupling constants
were m easured b u t hyperfine stru ctu re due to th e nitrogen quadrupole was not
resolved. F u rth er studies on N0C1[22] a t higher resolution obtained b e tte r values
for the chlorine quadrupole and new m easurem ents on nitrogen quadrupole coupling
were m ade. These previous experim ents utilized S tark m odulated spectrom eters,
and since then, there have been su b stan tial im provem ents in microwave technology
and m olecular beam techniques which resulted in orders of m agnitude im provem ent
in resolution an d sensitivity of m olecular m easurem ents in th e gas phase. T he recent
developm ent of th e pulsed beam Fourier transform technique (see C h ap ter II) has
resulted in obtaining ro tatio n al sp ectra w ith 10-12 kHz linew idth resolution which
can easily resolve the closely spaced J = 0 —> 1 hyperfine transitions in NOC1. The
J = 1 —> 2 transitions for NOC1 occur in th e K -band of the microwave region and are
out of range of th e current PB -F T S experim ental setup[23] . M easurem ents of the
J = 1 —►2 transitions were m ade w ith a m olecular beam m aser spectrom eter which
provides th e highest resolution available for m olecular spectroscopy w ith observed
linew idths of 2-8 kHz. M easurem ents of chlorine and nitrogen quadrupole coupling
were m ade using b o th techniques to o b tain a high resolution low J d a ta set for
NOC1.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
38
F igure III.l. S u b stitu tio n stru ctu re o b tained by Millen and Pannel[21]. r Nc i = 1.975(5)A, rw o = l-139(12)A , and Z0N C1=113.2(5)°
T he nitrogen quadrupole coupling can be used as a useful indicator of stru c­
tu re and bonding in larger molecules containing a nitrosyl groups[24] . A ccurate
quadrupole m easurem ents in sim ple com pounds such as N0C1 are im p o rtan t and
can aid in th e understanding of stru ctu re and bonding in nitrosyl groups of larger
molecules.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
39
I I I .l.i S y n th esis o f NOC1
N itrosyl chloride was prepared by bubbling gaseous HC1 through approxi­
m ately 20 ml of nitrosylsulfuric acid[25] a t room te m p eratu re u nder an atm osphere
of dry nitrogen. T h e nitrosyl chloride was a yellowish orange gas which was con­
densed in a cold tra p kept a t approxim ately -20° C by a dry ice ethanol b ath . T he
bubble ra te was controlled to approxim ately 200 to 300 bubbles p e r m inute. The
reaction was allowed to progress for approxim ately 2 hours or until no m ore yellow­
ish orange gas was observed. This usually resulted in 10 to 20 m l of relatively pure
condensed NOC1.
I l l . l . i i P u lsed B ea m M easu rem en ts
Seven hyperfine com ponents of th e J = 0 —> 1 tran sitio n were m easured
using using a Flygare-Balle type[23] P B F T microwave spectrom eter (details of the
spectrom eter are given in C h ap ter II). These tran sitio n s were observed in the 11104
to 11127 MHz range w ith approxim ately 12kHz linew idth (f.w .h.m ). An exam ple
spectrum of th ree of these tran sitio n s are shown in F igure III.2. NOC1 was delivered
to the spectro m eter via a connection to th e gas m anifold system from a sample cell
containing condensed NOC1 in a dry ice ethanol b ath . T he NOC1 was allowed to
w arm up to increase the vapor pressure of th e sam ple to 10 to 20 torr, and then
delivered to th e m anifold where it was d iluted w ith 1.0 to 1.1 atm ospheres of Ar
gas.
T h e resulting m ixture was th en pulsed through th e nozzle a t 0.5 Hz w ith
backing pressures ranging from 0.5 to 1.1 atm . T h e free induction decay signals
were averaged over several hundred gas pulses to im prove signal to noise.
hyperfine tran sitio n s m easured w ith th e P B -F T S are shown in Table 1.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
T he
40
0.8
!»
H
w
cn
Z
u 0.6
H
Z
tt
6:
§
w
0.4
0.2
0.0
146
291
437
583
728
874
1019
1165
O FFSET FREQ U EN CY (kHz)
Figure III.2. F T spectrum of th e J = 0 —> 1 hyperfine transition.
S tim ulating frequency= 11104.587 MHz
Table I I I .l T ransition frequencies for th e J = 0 —> 1, K = 0, and
A I — 0 m easured on the PB -F T S for 35C1N0.
21
2F
2F'
M eas.(M Hz)
M -Calc.(M Hz)
3
3
3
11104.010(2)
-0.007
5
1
5
1
5
1
11104.210(3)
11104.370(9)
-0.007
1
1
3
0.020
5
5
7
11116.324(8)
11116.414(14)
3
3
5
5
5
5
5
-0.005
3
11116.607(3)
11126.245(5)
0.021
0.006
-0.014
5
11104.210(3)
-0.007
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
41
I I I .l.iii M olecu lar B ea m M aser M easurem en ts
T he tran sitio n s in th e 22215 MHz to 22603 MHz range were m easured on
a m olecular beam m aser spectrom eter[26] operated in th e one cavity m ode and are
shown in T able III.2. T h e linew idths for th e m aser tran sitio n s were approxim ately
8kHz (f.w .h.m ) an d an exam ple of th ree of these transitions is shown Figure III.3.
Sam ple delivery to th e m aser was achieved by connecting th e sam ple cell containing
n eat NOC1 directly to th e nozzle. T he backing pressure of th e NOC1 ranged from
0.5 to approxim ately 1.0 a tm an d was controlled by th e placem ent of th e sample
cell above a dry ice-ethanol b ath .
A continuous m olecular b eam was produced
by th e expansion of NOC1 th ro u g h a nozzle source w ith a 0.15 m m exit hole[27]
. T he beam was then collim ated th ro u g h a liquid nitrogen cooled tra p followed
by a 0.6 cm a p ertu re placed 20cm down stream from th e nozzle. T he beam then
passes th ro u g h th e quadrupole focuser region where a p o ten tial difference of up
to approxim ately 15 kV was applied to altern ate stainless steel rods. T he large
rad ial electric field gradient produced by th e quadrupole focuses th e molecules w ith
positive S tark coefficients (usually molecules in u pper states) along th e sym m etry
axis of th e focuser w here th e electric field gradient is zero. Molecules w ith negative
S tark coefficients (usually molecules in lower states) are deflected away from the
m ain p a rt of th e beam . T h e beam th en enters a 10 to 20 cm long TMoio cylindrical
microwave cavity. T he stim u latin g frequency is then tuned to th e desired tran sitio n
frequency of th e molecule which also corresponds to th e cavity resonance mode.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
42
640.
Figure III.3.
kHz.
660.
680.
700.
720.
740
7 6 0 . kHz
F irst derivative m aser spectrum relative to 22227000
Table III.2 T ransition frequencies for th e J = 1 —> 2, K = 0, and
A I = 0 m easured on th e m aser spectrom eter for 35C1N0. Frequencies
in MHz.
21
2F
2F'
3
5
5
22215.324(6)
-0.014
3
1
1
1
3
3
22226.603(11)
-0.022
22227.414(6)
-0.001
1
5
9
0.010
1
1
5
22227.669(1)
22227.690(2)
1
1
5
3
5
5
22227.731(7)
22227.935(7)
0.009
0.013
1
5
5
22236.392(14)
-0.009
1
3
3
22236.597(6)
-0.009
1
3
22580.931(5)
1
3
7
9
22592.371(5)
0.010
-0.004
1
3
7
22592.987(4)
-0.003
1
5
1
22602.179(4)
-0.002
Meas.
M-Calc.
0.015
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
43
T he stim ulating frequency which induces molecular transitions in th e
TMoio cavity is produced by a phase-locked klystron. Signals are detected by a
superheterodyne microwave receiver w ith a phase-locked local oscillator followed by
lock-in detection near 200 Hz w ith a tim e constant of 15 s. T he q u artz crystal ref­
erence oscillator used earlier[27], was replaced by a variable frequency synthesizer
(V angaurd Labs model SG-100C) o p erated near 100 MHz. T he reference frequency
was th en slowly swept by a circuit external to th e synthesizer. This m odification
allowed a wide range of frequencies to be scanned w ithout th e need for changing
crystals. T h e stim ulating frequency klystron is phase locked to a m ultiple of th e
synthesized reference frequency which is phase m odulated by th e 200 Hz oscillator.
The m odulation signal also serves as th e reference for the phase-sensitive detector.
T he synthesized reference frequency is displayed on a frequency counter which is
referenced to th e 60 kHz radio transm ission of W W VB, Boulder CO. Frequency
m arkers are derived from th e counter and are displayed on th e chart recorder along
w ith the o u tp u t from th e phase sensitive detector. A block diagram of th e m aser
spectrom eter is shown in F igure III.4.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
44
BEAM
LOCAL.
□SC
LOCAL
LOCK
BOX
fDCUSSER
CAVITY
x2
STIM
□SC
S71M
LOCK
BOX
|N2
HLTER
L SYNC
MULTIPLIER
AMP
DETECT
D-SCDPE
CHART RCCDRDCR
Figure III.4. Block diagram of m aser.
I I I .l.iv H yperfin e S tru ctu re and D a ta A nalysis
The hyperfine stru ctu re was analyzed using th e coupled representation. The
lab fram e coupling scheme is J + lei = F i and F j + I^r = F. J is th e ro tatio n al
angular m om entum and le i and I at th e chlorine and nitrogen spin angular m om enta.
T he H am iltonian m atrix elem ents for the quadrupole hyperfine stru ctu re are
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
45
x S I X J ( F u I c u J ' , 2 , J , I Ci) + e Q q j ( N ) g ( J ) f ( I N ) ( - l ) J+I^ ^ F ^ F
(15)
x S I X J ( F , I n ,F[,2,Fu I n )
W here q j is the electric field gradient a t th e chlorine or nitrogen nucleus and is
related to th e com ponents qQQ along th e principal axes, a, b, and c by th e relation
—o V ' ~
< Jo>
qj - 2 2^9ao, Jj ((. TJ 4+. 1)
a
(16)
and
f(I) =
( 2 / + l ) ( / + l)(2 / + 3 ) p
1(2 1 - 1)
(17)
and
9(J)
J ( 2 J + 1 ) ( J + 1)
(2 J — 1)(2 J + 3)
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
(18)
46
Term s off-diagonal in th e ro tatio n al states were not included in th e analysis an d
m atrices for each value of F for each ro tatio n al s ta te were constructed an d diagonalized. T he ro tatio n al energies were described by :
H ROT = A J l + B J l + C J 2 -
D j ( J 2)2 -
D JK J 2J l
(19)
T h e differences betw een frequencies calculated w ith the above ‘first-order’
trea tm e n t an d calculations including term s off-diagonal in rotatio n al states were on
th e order of 3-4 kHz. T he m easured an d calculated, best fit, transition frequencies
are shown in Tables III.l and III.2. T h e s ta n d a rd deviation of th e fit was 14kHz.
T he deviations (last column, Tables I I I .l an d III.2.) betw een calculated and m ea­
sured frequencies are slightly larger th a n experim ental uncertainties for individual
transitio n s, b u t including oth er interactions, such as spin-rotation interactions, did
n o t significantly improve th e s ta n d a rd deviations for th e fits. T he values for A, D j
and D j k were determ ined m ore accurately in th e earlier work of reference[28] since
higher J tran sitio n s were m easured. T h e previous values of reference 2, A = 87374.46
MHz, D j = 0.0063 MHz and D j k = -0.0585 MHz were used as fixed param eters
in our analysis. T he m olecular p aram eters o b tain ed here are in reasonably good
agreem ent w ith th e previous values, as shown in Table III.3.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
47
Table III.3 Values of th e m olecular param eters o b tained by fitting the
observed tran sitio n s in Tables I II.l an d III.2. Two sta n d a rd devia­
tions are in parentheses, crFIT = 14k H z . Values in MHz.
P aram eter
P resentR esults
Previous Values0
B
5737.70(6)
5737.69(3)
C
5376.32(6)
5376.31(3)
eQqaa(Cl)
-49.05(4)
-49.18(35)
eQ q66(Cl)
30.0(4.0)
29.46(20)
eQ qaa(N)
0.98(6)
1.0(4)
eQ q6i(N)
-4.78(22)
-4.8(2)
“ see reference [22]
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
48
III.2 ClFs S p ectru m and Structure
T h e interhalogen com pounds have a variety of interesting and unusual
structures.
T he stru ctu re of CIF3
is unusual when com pared to o th er small
molecules. G as phase stru ctu res have also been o b tained for C1F an d CIF5. M i­
crowave spectroscopy[29] and x-ray diffraction work[30] , carried o u t 37 years ago,
obtained a planar, C 2 V stru ctu re for CIF3 w ith F -C l-F bond angles slightly less
th a n 90°. T he basic T -shaped stru ctu re w ith 90°F-C1-F bond angles is predicted
by the V S E P R descriptions of molecular bonding. Even th e observation of less
th a n 90°F-C1-F angles can be rationalized by th e larger lone pair-bond repulsion ef­
fects. A ccurate gas phase stru ctu ral param eters and quadrupole coupling strengths
can be o b tain ed using microwave m easurem ents. T he electronic stru ctu re inform a­
tion from th e quadrupole coupling com bined w ith b o n d lengths an d angles should
be useful for ab initio electronic stru ctu re calculations on this molecule. In th e
previous microwave work[29], five groups of rotatio n al tran sitio n s for 35C1F 3 and
37C1F 3 were m easured in th e 20-26 GHz range and fitted using a rigid rotor-first
order quadrupole H am iltonian. Some of th e hyperfine stru ctu re was only partially
resolved.
III.2 .i E xp erim en tal
T h e C IF3 spectrum was m easured using th e P B -F T S spectrom eter which
is described in d etail in C h ap ter II. A com mercial sam ple of C1F3 ( A ir P ro d u cts
UN 1749) was used w ithout fu th er purification. In spite of th e high reactivity of
CIF3 , sam ples could be handled for short periods of tim e in a stainless steel system .
M ixtures of 2% C1F3 in argon at 0.5 atm pressure were pulsed into th e F abry-Perot
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
49
cavity of th e spectrom eter a t a 1 Hz repetition rate. Signal-to- noise ratios ranged
from 5 to 0.1 for for a single gas pulse on the m easured transitions.
Forty-tw o tran sitio n s for 35C1F 3 and 37 transitions for 37C1F 3 are listed in
Tables A .I.l an d A.1.2 in A ppendix A.I. B oth Q -branches ( A j = 0) an d iZ-branch
(A j = 1), 6-dipole tran sitio n s were m easured. All transitions th a t were observed
w ith th e P B -F T S h a d K = 0 for either th e upper or lower state. T his is probably
due to the fact th a t th e ro tatio n al cooling has substantially reduced po p u latio n s in
K 7^ 0 states. T his would im ply efficient rotatio n al cooling an d beam ro tatio n al
tem peratu res less th a n 10 K for rath e r m ild expansion conditions.
III.2 .ii C lF t
H yperfin e S tru ctu re A nalysis
T h e chlorine hyperfine stru ctu re was analyzed using th e coupled represen­
ta tio n w ith th e to ta l angular m om entum F = I + J. T he nuclear spin I = 3/2 for
35C1F 3 and 37C1F 3 and the quadrupole coupling strengths were sufficiently large
th a t H am iltonian m atrices for each F value were constructed and diagonalized. T he
H am iltonian for ro tatio n al energies is given in equation (19) in section IIL l.iv and
the M atrix elem ents for quadrupole coupling have th e form
( I J ' F \ H Q\ I J F ) =
0.25eQq[(2I + 1)(2/ + 3)( J + 1 )/I(2J - l)]1/ 2
(20)
x [ J( J + 1)(2J + l) / ( 2 J - 1)(2 J + 3)]1/2
x S I X J ( F , I, J',2, J, I ) ( —1)J+I+F
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
50
where eQqd — 2eQ'%2qaa {Ja)/ J ( J + 1). T he differences between this diagonalization m eth o d and a ‘first- o rd er’ treatm en t of th e quadrupole coupling are on the
order of 30 kHz for CIF3 .
III.2 .iii D a ta A nalysis
A nonlinear least-squares fittin g routine was used to o b tain best fit values
for A, B , C , D j , D j K , e Q q aa, and eQqbb for b o th isotopic species of CIF3 . T he
calculated tran sitio n frequencies an d q u an tu m num ber assignm ents are given in
Tables A .I.l an d A .1.2. T he p aram eters obtained are shown in Table III.4. The
stan d ard deviation for th e fits are 17 kHz for b o th isotopom ers. T he d istortion
param eters D j and D j k were too sm all to be determ ined by th e fitting procedure.
U pper lim its can be placed on D j and D j k of 1kHz and 10 kHz from th e present
d a ta set.
T he ratio of quadrupole coupling strengths for th e two isotopes are
eQqaa (35C1/37C1) = 1.2686 an d eQqbb (35C1/37C1) = 1.2684. These are in agree­
m ent w ith th e very precise ratio 1.268877 obtained from atom ic beam resonance
experim ents [31] an d provide indication of th e accuracy of m easurem ents an d as­
signments.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
51
Table III.4 Values of th e m olecular param eters obtained by fitting
the observed tran sitio n s in Tables A .I.l and A.I.2. Two stan d ard
deviations are in parentheses, crF/T= 17kHz. U nits = MHz.
P aram e ter
35C1F3
37C1F3
A
B
13748.25(1)
4611.719(2)
13653.54(1)
4611.866(2)
C
3448.629(3)
3442.719(4)
eQqaa (C l)
82.03(3)
64.66(4)
eQqbb (Cl)
65.35(2)
51.53(3)
III.2 .iv S tru ctu ral A n alysis
T h e m easured ro tatio n al constants were used to obtain b e tte r stru ctu ra l
p aram eters for C1F3 . T h e in ertial defects A — I c — I a — lb were found to be
A = 0.19976 am u
A 2 for 35C1F3 an d A = 0.19962 am u A 2 for 37C1F3 . T hese are
typical values for chemically b o u n d p lan ar molecules and are a ttrib u te d prim arily
to vibrational effects.
Since only one isotope of fluorine is available, obtaining a su b stitu tio n stru c­
tu re is n o t feasible for this molecule. A n ‘effective’ stru ctu re can be obtained[20] by
fitting th e observed ro tatio n al constants. If A, B , and C were fit simultaneously, the
inertial defects w ould result in very large uncertainties. A preferred procedure[20]
is to fit pairs of ro tatio n al constants, assum ing a p lan ar structure. T he stru ctu ral
p aram eters are shown in F igure III.5.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
52
Cl-F
F-CI-F
Cl-F
F igure III.5. S tru ctu ral p aram eters for C IF3 .
The results of fitting th e stru c tu ra l param eters
0
f
-
ci
-
f
Z
ci- f
,
fci-F i
and
to A and B , or A and C, or B an d C values are shown in Table III.5.
T h e errors were much larger w hen fittin g B and C. A weighted average of th e fit
results is shown in the b o tto m line of Table 5. These should be approxim ate to
ro stru ctu ra l param eters, since su b stitu tio n of 37 Cl for 35Cl will n o t produce very
large changes in vibrational zero-point energies.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
53
Table III. 5. Effective stru ctu re for CIF3 calculations using leastsquares fits to pairs of ro tatio n al constants for 35C1F 3 an d 37C1F 3 .
Indicated errors are 2 stan d ard deviations. These values should be
very close to ro-type stru ctu ral p aram eters since th e effects of th e Cl
isotopic su b stitu tio n on vibration-averaged coordinates will be small.
From
rci-F(k)
^ f - ci - f
A,B
1.5984(5)
1.69990(6)
87.48(4)
A,C
1.5984(3)
1.70144(4)
87.48(2)
B,C
1.600(8)
1.6996(7)
87.7(6)
A vg“
1.5985(4)
1.70073(5)
87.48(4)
“W eighted average, weighted by inverse of stan d ard deviations.
III.2.V D iscu ssion o f ClF-t
T he m easured ro tatio n al constants an d quadrupole coupling stren g th s are
in very good agreem ent w ith earlier, less accurate, values obtained by Smith[29]
using a com pletely different set of transitions.
T h e earlier quadrupole coupling
stren g th s were slightly sm aller, possibly due to neglect of term s off diagonal in J.
An electron diffraction study of C IF3
was done recently[32] and th e stru ctu ral
p aram eters obtained are in excellent agreem ent w ith the present work.
A planar, T -shaped stru ctu re for B rF 3 was obtained previously by M agnuson[33] using microwave m easurem ents. T he B rF 3 stru ctu re is very sim ilar to th e
present CIF3 structure. In b o th molecules, th e axial X -F bond length is less th a n
th e equato rial X -F bond length. T h e F-X- F bond angles are b o th less th a n 90°,
w ith an F-C I-F angle of 87.5° for C1F3 and an F -B r-F bond angle of 86.2° for B rF 3.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
54
III.3 T ow nes D a iley In terp reta tio n o f Q uadrupole C oupling D a ta
From th e results of th e Townes-Dailey m odel of quadrupole coupling, sum ­
m arized in Section III. 1, one can discuss th e electronic an d bonding environm ents
aro u n d th e coupling ato m in th e molecule of interest. These results can also be
used to com pare th e environm ents betw een molecules containing th e same coupling
atom .
T h e m easured quadrupole coupling com ponents for N0C1 can be com pared
to values in o th er molecules. M etal nitrosyl complexes[34] show wide variations in
electronic stru ctu re and MNO bond angle. T he M NO or XNO bond angle and
m ore detailed inform ation ab o u t th e electronic stru ctu re can be related to nitrogen
nuclear quadrupole coupling strengths [24]. For linear bonding to NO the analysis is
fairly simple, since quadrupole coupling tensor will be collinear w ith th e NO bond
axis and will be cylindrically sym m etric (or very nearly cylindrically sym m etric).
T h e quadrupole coupling tensor for nonlinear XNO molecules will n o t be cylindri­
cally sym m etric and the principal axes will usually not be collinear w ith th e NO
bond.
Q uadrupole coupling stren g th s for some typical molecules and C1N0 are
listed in Table 8 . If these are in terp reted in term s of th e Townes-Daily model[20] ,
w ith n x , n y and n z representing th e approxim ate occupation of the valence atom ic
p orbitals on nitrogen, we can estim ate ( Up)z =
—n z^, th e difference in
occupation of parallel and perpendicular p orbitals w ith respect to the ‘z ’ axis. The
relation to th e quadrupole coupling along th e z-axis (a- axis for th e listed molecules)
is:
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
55
eQqzz = W ( n— ~ L _ n z ) M H z
(21)
= 10(Up)zM H z .
For N 2,(Up)z ta —0.25. This difference is reduced to ( Up) z ~ —0.19 for NO, pre­
sum ably by additional electron density in p x an d p y orbitals. This difference is
fu rth er reduced to nearly zero for NNO. C1NO an d FN O are b o th b en t w ith bond
angles of 113° an d 110° respectively. C F3NO is also b en t an d all th ree of these
molecules have fairly large positive quadrupole coupling term s perpendicular to th e
m olecular plane (eQ qcc). This indicates th a t th e occupation num bers of th e in-plane
p-orbitals on nitrogen are much larger th a n th e occupation num ber perpendicular
to th e m olecular plane. In spite of th e slightly different orien tatio n of th e a-axes for
C1NO an d F N O , th e nitrogen quadrupole coupling stren g th s are very similar.
Table III.6. N itrogen quadrupole coupling stren g th s for some typical
molecules and 35C1N0 and values of (Up)z o b tain ed from eQ qaa val­
ues. N ote th a t th e z axis is taken to coincide w ith th e a-axis for C1NO
and FN O . Values in MHz.
Molecule
eQ qaa
eQ q ti
(■UP)z
Ref.
n2
-2.52
1.46
-0.25
[20]
NO
-1.86
0.93
-0.19
[35]
N N *0
-0.27
0.135
-0.03
[36]
C1NO
FN O
C F 3NO
0.98
1.7
-4.78
-4.8
0.10
0.17
[37]
0.5
-6.0
0.05
[38]
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
56
T h e largest chlorine quadrupole coupling com ponent in NOC1 is negative
and lies along th e a-axis. Since the a-axis makes only an 18°angle w ith th e N-Cl
bond this would indicate a higher electron density in th e chlorine p 2 o rb ital along
the N-Cl b o n d th a n in th e perpendicular p x or p y orbitals. T h e difference, however,
is significantly less th a n for Cl atom (eQ q= -110 MHz) or CI2 molecule (eQ q = -109
MHz) or even HC1 (eQ q= -68 MHz).
M easured values for th e chlorine-35 quadrupole coupling for a few molecules
are listed in Table III.7. For a chlorine atom Up = 1 , so one can estim ate Up for
other molecules in Table III.7, using th e quadrupole coupling strengths. Up is 1.33
for C1F, indicating th a t n z is less th a n th e n eu tral ato m value of 1. T he fluorine
atom appears to pull electron density away from th e chlorine.
Table III.7. Chlorine ( 35C1) quadrupole coupling strengths and th e
unbalance of occupation num bers for th e p electrons, Up , for some
molecules.
Molecule
eQq (MHz)
uP
Ref
C1F
-145.87
1.33
C l(atom )
-109.74
1
Cl2
-108.95
0.99
[39]
IC1
-85.8
0.78
[39]
C1CN
-83.3
0.76
C H 3CI
-74.75
0.68
[40]
HC1
-67.6188
0.62
[40]
31F3 (066)
65.36(3)
-0.62
present work
21F 3 (qaa)
82.03(4)
-0.75
present work
[39]
[41]
[42]
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
For IC1, C1CN, C H 3CI, an d HC1, th e bonding atom s apparently contribute
additional electron density to p z(C l), resulting in n x > 1 and a sm aller unbalance.
For CIF3 , however, th e large positive values of e Q q in th e plane of th e molecule and
correspondingly large negative out of plane value (eQqcc =-147.4 MHz) indicate a
substan tial reduction in occupation of th e p orb ital p erpendicular to th e plane of
the molecule along w ith increased electron density in th e in-plane p orbitals.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
58
C H A P T E R IV:
M ICROW AVE MEAS. O F CO BALT TRI-CA RBO N Y L NITROSYL,
C Y C LO PEN TA D IEN Y L CO BA LT DI-CARBONYL, AND
CY C LO PEN TA D IEN Y L M A N G A N ESE TRI-CA RBO N Y L
M any tran sitio n m etal complexes function as catalysts in reactions im por­
ta n t in in d u stry and biology. For exam ple, cobalt is used in the catalytic production
of esters and alcohols[43] in in d u stry an d plays an im p o rtan t role in th e synthesis
of vitam in B 12 in some anim al species.
M anganese is crucial in th e conversion
of A D P to A T P in m etabolical processes as well as in photosynthesis[44]. Hence,
understanding th e roles tran sitio n m etal complexes play in these reactions m ust in­
clude an accurate description of th e ir overall geom etry and electronic environm ents
around th e tran sitio n m etal.
Microwave spectroscopy provides d a ta sensitive to
these considerations free from solvent and crystal effects which can disto rt th e tru e
n atu re of th e tran sitio n m etal complex.
U ntil recently, o btaining high resolution m olecular sp ectra of tran sitio n
m etal complexes was alm ost im possible. D ue to th eir relatively large size ( > 10
atom s) in terpreting th eir sp ectra can be very difficult since large m om ents of in ertia
produce closely spaced and often unresolved ro tatio n al transitions.
In addition,
unresolved nuclear hyperfine stru ctu re caused by large quadrupole m om ents found
in some tran sitio n m etals often com plicated the sp ectra so much th a t only estim ates
of m olecular p aram eters could be obtained. P resented in this C hapter are th e results
of the first high resolution microwave stu d y of C o(C O )3NO , C pC o(C O )2 , and
C pM n(C O )3 using P B -F T S .
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
59
I V .1 N uclear Q uadrupole C oupling in C oC C O aN O
T he stru ctu re of cobalt tricarbonyl n itrosyl is shown in Figure IV. 1. It is
a Czv oblate sym m etric top an d electron diffraction m easurem ents of th e gas-phase
stru ctu re of C o(CO )sN O were recently rep o rted by H edberg et al. [45] . T heir stru c­
tu re indicated a linear C o-N -0 group an d was consistent w ith C$v sym m etry for th e
complex. The num ber of stru ctu ra l p aram eters obtained from electron diffraction
d a ta is much g reater th a n from microwave m easurem ents, especially for a sym m et­
ric top molecule. However, th e accuracy relative to th e equilibrium stru ctu re is not
as high due to vibrational averaging effects an d possible overlapping of interatom ic
distances in the radial distrib u tio n function. Therefore it is desirable to o b tain b o th
electron diffraction and microwave d a ta for th e stru ctu re determ ination of this type
of molecule. This is particu larly tru e for sym m etric tops since only one rotatio n al
constant is o b tained from th e microwave d a ta for each isotopic species.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
60
Co
F igure IV .1. S tru ctu re of C o(C O )3NO taken from [45].
An early atte m p t by others to m easure the microwave spectrum of
C o(C O )3NO [46] ‘failed to give conclusive resu lts’. No tran sitio n frequencies were
rep o rted a t th a t tim e. It is likely th a t th e difficulties were due, in p art, to th e fairly
large and complex hyperfine splitting due to 59Co and 14N quadrupole coupling.
T h e nitrogen quadrupole coupling stren g th in C o(C O )3NO is slightly larger
th a n the value for CpNiNO and is consistent w ith values observed for o th er linear
M -N -0 complexes.
Since this is th e first gas-phase 59Co
quadrupole coupling
m easurem ent, com parisons of this p aram eter w ith oth er gas-phase values are not
possible. T he m easured value, however, should be useful for indicating valence shell
p-electron distribution. T he Co quadrupole coupling is sensitive to the electronic
stru ctu re n ear th e Co atom , an d can be used as a test param eter for electronic
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
61
stru ctu re calculations. Electronic stru ctu re calculations on sim ilar system s are cur­
rently being carried out, b u t are quite difficult due to th e large basis sets required
and relativistic effects associated w ith heavier nuclei.
I V .l .i P u lsed B eam M easu rem en ts
Microwave transitions in th e 6-13 GHz range were m easured using a pulsed
beam , Fourier transform spectrom eter described in detail in C hapter II. A 1-2%
m ixture of C o(C 0 )3N 0 in A r was pulsed a t room tem p eratu re w ith a pressure
behind th e nozzle ranging from 0.4 to 0.7 atm ospheres. T he Co(CO )3NO sample
was obtained from Strem Chem icals ( # 27-0500) and was used w ithout further
purification. T h e J = 4 —t- 5 tran sitio n s were p redicted to be near 10,791 MHz based
on th e m odel B, rg stru ctu re of Reference [45]. The first J = 4 —> 5 transitions were
found near 10,423 MHz. This m ay be an exam ple of th e ‘shrinkage’ effect in the
electron diffraction d ata. A 3.7% increase in ro tatio n al constants would correspond
to approxim ately a 1.8% shrinkage in coordinates if all coordinates are changed
by th e sam e fractional am ount.
W ith experim ental conditions optim ized, some
transitio n s could be detected w ith a single beam pulse a t 3 to 1 signal to noise. A
few weak lines due to J = 1 —> 2 transitions n ear 4170 MHz were detected in our
ap p ara tu s, b u t are not included in th e d a ta analysis.
T he 59Co nucleus has nuclear spin 1 = 7 /2 an d nuclear electric quadrupole
m om ent Q = 0.4 barns[47] so m any hyperfine tran sitio n s spread over m any MHz
were found. As a reference value, th e nuclear electric quadrupole m om ent of 79Br
is Q = 0.31 barns [47]. T h e additional hyperfine stru ctu re splitting due to U N
quadrupole coupling are also well resolved, resulting in rath e r complex hyperfine
stru ctu re p attern s.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
62
I V .l.ii R o ta tio n a l and H yperfin e A n a ly sis
C o(C O )3NO is an oblate sym m etric top w ith Cz v sym m etry. T he 12C 160
groups contain only bosons so th e p ro d u ct of th e nuclear spin wavefunction and
ro tatio n al w avefunction m ust be to tally sym m etric w ith exchange of the CO groups.
Since only th e to ta l 1=0 sym m etric spin function is possible for these CO groups,
only K = 0 or K = 3 n ro tatio n al states exist. T his simplified th e analysis of th e J =
2 —> 3 transitions, since only K = 0 states could co n trib u te to th e spectrum . K = 0
and K = 3 states were observed for th e J = 3 —> 4, an d 4 —> 5 ro tatio n al transitions.
T he hyperfine stru ctu re was analyzed using th e coupled representation. The
lab fram e coupling scheme is J + Ic 0 = F i and F i + I/v = F. J is th e rotatio n al
angular m om entum and I c 0 and Iat th e cobalt an d nitrogen spin angular m om enta.
T he H am iltonian m atrix elem ents for th e quadrupole hyperfine stru ctu re are
(IJ'F\H q \IJF) =
( - l ) J+Ic°+Fle Q q j ( Co)g ( J ) f ( I c o )
x S I X J ( F i , I Co, J \ 2, J, I c o ) + Ccc( I Co ■J )
( 22)
+ ( _ 1 y + i c o + i N+2 Fl+F x eQ q j ( N ) g ( J ) f ( I N )
x [(2Fx + 1 )( 2 F{ + 1 ) ] ^ 2 S I X J ( J \ f , I Co , Fu J, 2)
x S I X J ( F , I n ,F,2,Fu In )
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
63
T h e field gradient in a given ro tatio n al sta te (J,K ) for an ob late to p is re­
la ted to th e cobalt or nitrogen electric field gradient along th e c axis of th e molecule
by
3K 2
9J = ?“(1,7(7TT).)-1)
(23)
and
™
- K
^
&
+
r
+ T
<24)
and
„( tn
9 ^ ^
f J ( 2 J + !)(</ + 1)V /2
\ ( 2 J — 1 )(2 J + 3) /
CCo( J , K ) = C aa + ( C cc - C a a ) j ~ + ^
^
(26)
T h e K = 0 d a ta for J = 2 —>3 , 3 —►4, 4 —> 5 and 5 - ^ - 6 transitions
(Tables A II.l, A II.2, AII.3 an d A II.4 in A ppendix A .II) were fitted by constructing
H am iltonian m atrices including quadrupole term s off-diagonal in ro tatio n al states.
All d a ta , including the above sets, were fitted w ith a ‘first-order’ trea tm e n t which
did n o t include th e term s off-diagonal in ro tatio n al states. Differences betw een these
two types of calculations were less th an 2 kHz for th e listed transitions. T h e results
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
64
of fits to K = 0 transitions are shown in Table A .IV .5. T he decrease in ro tatio n al
constants can be used to determ ine th e distortion constant D j = 0.17(8) kHz. This
sm all value indicates ra th e r high stretching force constants for this molecule.
Table IV. 1. M olecular constants for C o(C O )3NO obtained from least squares to
th e tran sitio n frequencies given in A ppendix A.II. S.D. is th e s ta n d a rd deviation
for the fit. Values in MHz. Indicated errors are 2a. T he in d icated J value is 3iower
P aram e ter
A =B“
3—2
J= 3
.1561(7)
.1535(7)
J= 4
J= 5
.1503(4)
.1469(12)
Dj
F it R esult
.1590(4)
0.00017(8)
eQq c c 5 9 Co
35.37(8)
35.45(18)
35.79(16)
35.9(16)
DQ™ Co
C cc59Co
eQqCc 14 N
S.D.
35.14(30)
0.024(6)
0 . 010 ( 2 )
0.008(2)
0.008(2)
-1.57(7)
-1.53(9)
-1.65(6)
0.011
0.009
0.008
0.007(6)
-1.63(14)
0.0085(22)
-1.59(10)
0.014
“ T h e num ber preceding th e decim al point is 1042 for all of th e indicated transitions.
T he 59Co quadrupole coupling strengths ap p ear to be decreasing for the
higher J transitions. We can fit th e quadrupole coupling stren g th s to th e equation:
e Q q ( J ) = eQq 0 + D Q J ( J + 1)
(27)
T he p aram eters obtained are eQ q(59Co )= 3 5 .14(30) MHz and DQ = 0.024(6) MHz
for disto rtio n of th e quadrupole coupling.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
65
T he nitrogen quadrupole coupling values are reasonably consistent w ith an
average value eQ q ( 14N ) = -1.59(10) MHz. This can be com pared w ith th e value
eQ q = - 1.22 for CpNiNO[48] . T he rep o rted molecular param eters are based on
K = 0 d a ta only, since m any m ore K = 0 transitions th a n K = 3 tran sitio n s were
m easured an d assigned. T he fits were significantly im proved by including a spinro ta tio n constant C cc for 59Co . T he observed value of 0.009 MHz is fairly large
for a molecule w ith such small ro tatio n al constants.
K = 3 tran sitio n s were observed and identified for J = 3 —>4, 4 —>5 and
5 —> 6. T he m easured frequencies and best fit values for th e K = 3 , J = 3 —> 4
tran sitio n s are listed in Table A.IV.5. T he molecular param eters determ ined from
these transitions are also listed in Table IV. 1. T he ‘effective’ B value is
B ef f = B 0 — D
ji<K2
— 2D j ( J u p p e r ) 2
(28)
= B e f f ( K = Q ) - D J KK
2
C om parison of th is B ef f w ith th e K = 0 value yields a very small distortion constant
D j k ~ 0.6 kHz. T h e quadrupole coupling strengths and spin-rotation constant are
also quite close to K = 0 values, so no atte m p t will be m ade a t this tim e to ex tract
fu rth e r p aram eters, such as C aa from this data. The lim ited num ber of resolved K
= 3 lines for o th e r ro tatio n al transitions would not justify fu rth er analysis of these
tran sitio n s a t th e present tim e.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
66
I V .l.iii Sum m ary
T he absorption of tran sitio n s a t 4B, 6B, 8B, 10B and 12B frequencies along
w ith the observation of only K = 0 and K = 3 transitions confirm th a t this molecule
is of C i V sym m etry, sym m etric top. T he fairly large deviation between th e m easured
B value and th a t predicted from the electron diffraction d a ta is som ew hat surprising,
particularly since th e sm all m easured D j value (and D j k ) indicate a fairly rigid
structure.
Solid sta te values of eQ q (59Co
) in various tran sitio n m etal complexes
are ta b u late d in Lucken[49]. T he values were obtained from nuclear quadrupole
resonance an d range from 32 MHz in NH4[Co(N0 2 )(N H 3)2] to 172 MHz in
[(C5H 5)2Co]C104. O ur value of 35.1(3) MHz for C o(C 0 )3N 0 is n ear th e lower end
of this d istrib u tio n , indicating a fairly sym m etrical electronic charge distrib u tio n
around th e 59Co nucleus. T he basic stru ctu re of C o(C 0 )3 is very nearly te tra h e ­
dral so th a t th e observed quadrupole coupling m ust be due to the charge anisotropy
produced by differences in bonding of CO and NO ligands. A popular tex t [50] sug­
gests th a t NO is a ‘three electron d onor’ while CO is a ‘two electron donor’, b u t it
appears th a t th is rule would overestim ate th e observed charge anisotropy for this
complex.
T h e observed nitrogen quadrupole coupling e Q q cc (14N ) = -1.6(1) MHz
falls betw een th e value -1.22 MHz in Cp-Ni-NO [48] an d th e value -1.86 MHz for the
free NO radical[51]. This would indicate a larger unbalance in p* verses p x or p y
electron density th a n for CpN iN O, b u t sm aller th a n in NO or N 2 [48] (see C h ap ter
III for discussion of unbalanced p-electrons an d the Townes-Daily treatm en t).
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
67
IV .2 C d C o CCO^ : H in dered R o to r S p ectru m
A gas phase electron diffraction stu d y of C pC o(C O )2 by Beagley et al. [52]
provided values for m any of th e interatom ic distances. T he overall stru ctu re and
sym m etry was determ ined b u t the O C-Co-CO angle was not accurately determ ined
and the Co-CO bond peak was not resolved. Hopefully, th e com bination of electron
diffraction and microwave d a ta will allow a b e tte r gas-phase stru ctu re determ ina­
tion. T he stru ctu re shown in F igure IV.2 has th e cobalt atom on th e C 5 axis of
cyclopentadiene and th e carbonyls located such th a t th e OC-Co-CO p o rtion forms
a C 2 V stru ctu re. T h e present microwave d a ta is consistent w ith this basic structure.
Previous pulsed-beam m easurem ents have been m ade on th e related complexes CpNiNO [48] and C o(C O )3NO [53]. B oth of these cases are sym m etric top molecules
so only one p aram eter related to th e stru ctu re is available from m easurem ents on
a single isotopom er. Since C pC o(C 0 )2 is an asym m etric top, th e three principalaxis m om ents of in e rtia are m easurable an d independent. The correlation between
the ten fold b arrier to in tern al ro tatio n V10 and the A ro tatio n al constant is quite
high. O ther correlations are significant so ro tatio n al constants are n o t determ ined
to the usual microwave accuracy, b u t th e three to four significant figures should still
provide useful stru ctu re inform ation.
I V .2 .i E xp erim en tal C on sid eration s
M any hyperfine com ponents for 15 rotatio n al transitions in the 6-18 G H z
range were m easured w ith th is spectro m eter an d the details are given in C h ap ter II.
M ost of th e ro tatio n al tran sitio n s h ad te n or m ore com ponents spread over a 10-30
M H z range. T h e larger sp littin g were assigned to internal ro tatio n effects and the
sm aller sp littin g were in terp reted as 59Co quadrupole coupling hyperfine structure.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
68
T h e experim ental frequencies given in Table A .II.6 (in A ppendix A .II) are averaged
over the quadrupole splitting an d in some cases, are also averaged over th e in tern al
ro tatio n ‘m ’-state splitting.
T he C pC o(C 0)2 sam ple was purchased from Strem chemicals (pro d u ct #
27-0550) and vacuum distilled before use. T he sam ple was heated externally to in­
crease th e vapor pressure of C pC o(C O )2 and carried to a heated pulsed-expansion
nozzle a t the sam e tem p eratu re (28-320C ) by a flow of u ltra pure argon. Backing
pressures for th e experim ent ranged from 1 atm osphere down to 1 /3 of an atm o ­
sphere.
T he m olecular signal strengths were n o t very sensitive to th e choice of
backing pressure and strong transitions could be seen a t a 3 to 1 signal to noise
ratio for a single gas pulse.
IV .2 .ii D a ta A n alysis
T he averaged an d calculated tran sitio n frequencies are listed in Table IV.I.
T he hyperfine com ponents due to th e quadrupole coupling interaction of 59Co atom
were well resolved. The hindered internal ro ta tio n however, caused larger sp littin g
in the ro tatio n al spectrum .
In this section only th e hindered ro to r splittin g are
considered. T he quadrupole com ponents for a p articu lar internal ro to r tran sitio n
were averaged to give approxim ate line centers for th e internal ro to r spectrum .
T he H am iltonian m atrix for hindered rotation[54] was constructed in the
| J K m ) free-rotor basis and diagonalized. m is th e quan tu m num ber for ‘free’ ro ta ­
tion of th e ‘to p ’ (C p-ring) relative to th e ‘fram e’ (-C o(C O )2). The hindered ro to r
m atrix elements for a 10-fold p o ten tial b arrier are given by:
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
69
(J I < m \ H \ J K m ) = \ { B + C ) J ( J + 1)
£
(29)
+
K
( J K m \ H \ J , K ± 2, m) = -
2
+ F m 2 - 2 m A zK
B - C ) { [ J ( J + 1) - K ( I ( ± 1)]
(30)
x [ J ( J + 1) - ( K ± 1) ( K ± 2)]}1/ 2
( J K m \ H \ J K , m ± 10) = ~ V
(31)
10
w here A z is th e m om ent of in e rtia of th e fram ew ork alone, A z = tt,
l\lx
th e reduced m om ent of the two p a rts of th e molecule, F =
ift)
, an d F is
, -v. T he 10-fold
b arrier gives m atrix elem ents off-diagonal in m by 10 and the ^ -c o n sta n t for th e
to ta l I~ m om ent ( a =
was used as a fit param eter.
T h e selection rules[54-56] in th e free ro to r basis are A m = 0 and A K —
0, ± 2 , ± 4 ,...
T he A K = 0 tran sitio n s will be stronger th a n those w ith AA” =
± 2 , ± 4 ,... D ue to th e spin statistics of th e C pC o(C O )2 group (th e carbonyls are
bosons), there exists an additional restrictio n on th e allowed states for C pC o(C O )2
; only states w ith an even value of |m — K \ can exist. T his is sim ilar to restricting
K p values to even values for a rigid ro to r w here only bosons are exchanged by
a Ci ro tatio n ab o u t th e a-axis. T his simplified th e assignm ent of th e spectrum
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
70
since th e to ta l num ber of observed transitions is reduced by this constraint.
In
our calculations of the hindered ro to r energy levels th e m-values were tru n cated
at m — 44. At a 1 THz Vio barrier, this was sufficient to give tru n catio n errors
less th a n 1 kHz. A t Vio =40 THz or greater th e hindered ro to r program yielded
essentially rigid ro to r energy levels w ith very sm all in tern al ro tatio n splitting.
T h e fit p aram eters are given in Table IV.2. T h e fairly large deviations,
(M -C) values in Table A .IV .6, and sta n d a rd deviation values of th e fit p aram eters
are due, in p a rt, to the correlation effects and in p a rt, to averaging over m any
hyperfine com ponents caused by th e quadrupole coupling in teraction of th e cobalt
atom . Since th e hindered ro tatio n is around th e a-axis, th e A ro tational constant
is strongly correlated w ith Vio. Going from a ‘free ro to r’ to ‘high b arrier’ resulted
in A values ranging from 1440 M H z to 2170 M H z , respectively. The fit yielded
a reasonable set of ro tatio n al constants A, B , an d C , and a barrier to internal
ro tatio n , Vio, to be on th e order of 0.3 k J/m o l (0.82T H z ) .
These values for the
p aram eters were obtained by fixing th e top m om ent of inertia, I a = 118.3 am u A2.
This value of I a was calculated from th e stru c tu ra l d a ta obtained by th e electron
diffraction stu d y of Beagley et al. [52]. By allowing I a to be fit to the current
d a ta set, one obtains a slightly b e tte r ( S . T ) =S MHz ) overall fit. T he fitted value,
Ia =
122(3) am u A2, is not much different, w ithin experim ental error, th a n th e
I a o b tain ed from the electron diffraction[52] study. D ue to th e small im provem ent
in th e fit stan d ard deviation, we have chosen to fix I Q a t 118.3 am uA 2 and report
A, B , C and Vio as given in Table IV.2.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
71
Table IV .2. M olecular constants for C pC o(C 0 )2 obtained from a least
squares fit of th e H am iltonian to th e observed tran sitio n frequencies.
T h e repo rted error is 2a. T h e quadrupole coupling results are in Table
A.II.7. E rrors for eQq are 4 a. T he values are reported in MHz.
P aram e ter
Value
E rror estim ates
A
1625
±20
B
1257
±2
C
876
±2
0.82 T H z
Vio
± 0.20 T H z
&Qqaa
12
±4
&QQbb
132
±4
T he stru ctu re was param eterized (shown in Fig. 2) w ith R \ , the distance
from the Cp plane to Co, i ?2 the distance from Co to th e carbonyl carbon atom s
and 0, th e O C-Co-CO angle. The stru ctu re of the Cp group is fixed a t th e electrondiffraction results[52] w ith C-C distance =1.45
giving I Q =
118.3 a m u
A 2 . R \ , ?-2 and
6
A and C-H distance = 1.083 A,
are n o t linearly independent when used
to determ ine m om ents of in ertia since th e Co ato m is so close to th e center of mass
of the complex. R i was determ ined more accurately in th e electron diffraction th an
i ?2 or
0
, so we also fixed Ri a t 1.735
A, from the electron diffraction work. A least
squares fit to our m easured A, B and C values then yielded i ?2 =
1-69(5)
A and
0 = 98(3)°. Since there is a fairly large uncertainty in A due to internal ro tatio n
effects an d since
6
depends alm ost directly on A — B , we rep o rt
6
= 98(5)° which
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
72
is in agreem ent w ith th e electron diffraction result. O ur present R 2 value is also in
agreem ent w ith th e electron diffraction result.
Z
R2
R1
Figure IV .2. S tru ctu ral param eters for C pC o(C 0 )2 • R \ = 1.735
#2 = 1 .6 9 (5 ) A, 0=98(5)°.
A,
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
73
IV .2 .iii Q uadrupole C oup ling
T he 59Co quadrupole splitting of th e Jj< = 20 —> 3o, 30 —> 40, 40 —> 5o
transitio n s were analyzed to o b tain eQqaa and eQqbb• T h e K = 0 transitions were
chosen because th e relations betw een splitting an d eQq values are nearly indepen­
dent of internal ro tatio n effects for K = 0 and m — 0 and are described reasonably
accurately by a rigid-rotor hyperfine analysis[57]. T he results of th e least-squares
fit to these tran sitio n s are shown in Table A.II.7. T he stan d ard deviation for th e fit
was only 70 kHz b u t uncertainties in eQq values (Table II) are given as 4 MHz due
to possible sm all m odifications of {</„), ( J j ) and (J^) by internal ro tatio n effects.
T he eQqbb quadrupole value is significantly larger th a n th e value o b tain ed from th e
relatively sym m etrical C o(C O )3NO , b u t w ithin th e lim its of oth er m easurem ents
by N Q R in solid-state experim ents.
IV .2 .iv R esu lts and D iscu ssio n
T h e present results for Vio barrier can be com pared to earlier solidsta te N M R studies[58] of hindered ro tatio n of Cp in ferrocene, cobaltocene and
ruthenocene. In those cases th e barriers were a factor of 25 higher ranging from 7 to
9 k J/m o l. T his could be due to additional interactions w ith neighboring molecules
in the solid, or effects o th er th a n hindered ro tatio n which modify N M R linew idths.
Listed in Table IV .3 are potentials for some V3 and Vq barriers as well as
our current m easurem ent V10 for C pC o(C O )2 . T he V3 p o tentials for CH3OH an d
CH3C0C1 are relatively high, while th e V$ potentials for C H 3N 0 2 an d CH3B F 2 are
considerably lower. CH3N 0 2 and CH3B F2 were analyzed in the free ro to r basis and
provided reliable values for th e ir respective Vq barriers. Since these poten tials fall in
the ‘low b arrie r’ regim e for hindered ro tatio n , th e elem ents off-diagonal by m ± 6 in
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
74
the H am iltonian m atrix are small. T he p o ten tial in CF3N02[56] is relatively higher
so the m ± 6 elem ents are larger, therefore more m states m ust be included in th e
diagonalization of th e H am iltonian. M ore m ixing of th e free ro to r basis functions
occurs for increasing values of th e p o ten tial and also due to the asym m etry of the
molecule. Since th e m ultiplicity of th e p o ten tial is larger for CpCo(CO)2 relative to
CF3NO2, m values up to m = 44 were needed. A lthough th e stan d ard deviation for
the fit and th e statistical errors in A, B, and C are fairly large com pared w ith usual
microwave values, they are still sufficiently accurate to provide useful stru ctu ra l
inform ation. We believe th a t these large deviations are due to ‘interm ediate b a rrie r’
effects of hindered ro tatio n an d th e effects of large quadrupole coupling which are
only partially analyzed in th e present work. A nother hindered m otion of th e CO
groups could also cause shifts an d splitting. We believe th a t th e excellent fit to the
quadrupole com ponents (Table A .II.7) confirm the J, K , m assignm ents for those
transitions.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
75
Table IV .3. Some N —fold p o ten tial barriers (V}v) to internal ro tatio n
for various molecules.
M olecule
N-Fold
V /v(kJ/m ol)
CH3OHa
v3
4.477
CH 3COCI6
v3
5.422
c h 3n o 2c
V6
0.025
C F 3N 0 2d
v6
0.311
CH3B F 2
V6
0.058
C pC o(C O )2
V 10
0.3
0 see: [59], 6 [60], c [61], d [62]
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
76
I V .3 M anganese Q uadrupole C oup ling in C pM nfC O )a
Microwave m easurem ents in th e 4-14 GHz range were m ade using a PB -F T S
type microwave spectrom eter (see C h ap ter II). The spectrom eter was configured in
th e sam e m anner as it was for C pC o(C O )2 . A m ixture of 1-2% C pM n(C 0)3 in
argon was pulsed through a solenoid valve by heating a small sample cell containing
solid C pM n(C 0)3 to 72°C while flowing argon through the cell. The cell and pulse
valve were m aintained a t a co n stan t tem p eratu re of 72°C to keep th e valve from
sticking. T he C pM n(C 0)3 was purchased from S trem Chemicals ( # 25-0390) and
used w ithout fu rth er purification. T h e J =
5 —> 6 transitions were predicted to
be n ear 9918 M H z based on th e ro tatio n al constant B = 826.5 M H z obtained
from an earlier low resolution microwave experim ent[63] . T he first com ponents
of th e J =
5 —> 6 tran sitio n were m easured near 9930.6 M H z .
Good signal to
noise was o b tained by using backing pressures ranging from 0.6 atm for the low J
transitio n s to 0.4 atm for th e high J transitions. W hen optim izing experim ental
conditions, 3/1 signal-to-noise ratios could be obtained for a single gas pulse for
m any hyperfine com ponents of th e J =
3 —* 4, 4 —> 5 and 5 —* 6 ro tatio n al
transitions. M any com ponents corresponding to J — 7 —> 8 and J = 8 —> 9 at
13,248 MHz and 14,904 MHz respectively were also m easured, however, line centers
could not be obtained from th e d a ta due to th e large quadrupole interaction of the
m onoisotopic M n nucleus 1= 5/2. T h e presence of th e large quadrupole interaction
produced closely spaced hyperfine com ponents th a t could not be well resolved for
the transitio n s involving high J values.
I V .3 .i D a ta A n alysis
C pM n(C O )3 is a p ro late sym m etric top w ith C s symmetry, assum ing a
rigid stru ctu re for this molecule. T h e 12C 160 groups contain bosons while the Cp
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
77
ring contains fermions. T h e sym m etry p roperties of th e p ro d u ct of the ‘tw o-top’
spin wavefunctions and th e overall ro tatio n al w avefunction allow all the K -states
to exist. Therefore, m any hyperfine com ponents of th e ro tatio n al transitions were
present in the spectrum . T he m easured frequencies and th e ir assignments are listed
in Tables A.II.8- A .II.12 in A ppendix A.II. T he 96 reported frequencies were fitted
in a linear least squares fashion to a sym m etric to p H am iltonian including term s for
distortion and hyperfine stru ctu re due to th e quadrupole coupling and spin ro tatio n
effects of th e M n nucleus:
H — H rot + H q
(32)
where
H rot = B J 2 + ( A - B ) J a2 - D j ( J 2)2 - D j K J 2J l - D K ( J 2)2
and
J 2 = J 2 + Jl + J2
T h e H am iltonian m atrix elem ents for th e quadrupole and spin- ro tatio n interactions
are:
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
78
/ ( 2 I + 1 ) ( 2 / + 3 ) ( I + 1)
1(21 - 1)
J ( J + 1 )(2 J + 1 )\ *
(2 J - 1 ) ( 2 J + 3)J
(33)
x S I X J { F , I , J ', 2 , J , I } ( —1 ) J+I+F
4" CMn{lMn ' J')
where
C Mn { J , K ) = C bh + ( Caa - C bb)
^
T he field gradient in a given ro tatio n al sta te ( J , K ) for a prolate sym m etric top is
related to th e m anganese electric field gradient along th e a-axis of th e molecule by
q j = qaa (
3K 2
+
“ I)
(34)
T he quadrupole coupling was analyzed in th e coupled representation F = I + J,
w here I and J are th e spin angular m om entum for th e M n an d ro tatio n al angular
m om entum of th e molecule respectively, F is th e to ta l angular m om entum . All d a ta
were fitted by a ‘first order’ treatm en t of th e quadrupole coupling interaction which
did not include term s off diagonal in ro tatio n al states.
T he stan d ard deviation
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
79
for th e fit was 10 kHz. A second order analysis of th e d a ta was perform ed and
th e differences between these two analyses was on th e order of a few k H z for th e
tran sitio n s listed in Tables A .II.8-A.11.12 in A ppendix A.II.
T h e fit results to B , D j , D j k , eQqaa, and Cj& are shown in T able IV.4.
T h e ro tatio n al constant B = 828.0333(3) M H z is com parable to an earlier experimental[63] value of 826.5(5) M H z . T he d istortion param eters D j an d D j k are very
sm all b u t still well determ ined. D k cannot be determ ined from th e present d a ta
set. T h e disto rtio n p aram eter D j obtained in the present stu d y can be com pared
to values o b tain ed for o th er tran sitio n m etal complexes in Table IV.5.
Table IV.4. Best fit param eters to th e ro tatio n al hyperfine transitions
listed in A ppendix A.II. a p i T = 10 kHz. listed errors are 2 a.
P aram e ter
Value
B=C
828.0333(6) MHz
Dj
0.088(9) kHz
DjK
e Q q aa
C aa
-0.04(3) kHz
68.00(4) MHz
-5.5(4) kHz
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
80
Table IV.5. C om parison of d isto rtio n p aram eters D j for some tr a n ­
sition m etal containing complexes.
M olecule
D j (kHz)
CpN iN O
0.08(2)“
C o(C O )3NO
0.17(8)6
C pM n(C O )3
0.088(9)
a,b See [48] an d section IV. 1 respectively.
T h e hyperfine co n stan ts eQqaa = 68.00(4) M Hz an d Cbb= 5.5(8) kHz ob­
tained from th e fit are reasonable for a molecule containing an atom , such as M n w ith
large nuclear electric quadrupole (Q = 0.4 b arn s) and m agnetic dipole (/i = 3.468
nuclear m agnetons) m om ents. F ittin g b o th th e C aa an d Cbb com ponents of the
spin-rotatio n sim ultaneously resulted in p oorer fits, as d id fittin g only th e C aa
com ponent. T he best fit to th e d a ta set was o b tained from fittin g only th e Cbb
spin-rotation com ponent.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
81
IV .3 .ii E xp erim en tal R esu lts
T he basic geom etry of C pM n(C O )3 is shown in Figure IV.3.
T he Cp-
ring has C 5 axis th a t is collinear w ith the C3 of the M n(C O )3 group. T he overall
sym m etry of th e molecule is C s, however, th e two m om ents of in ertia Jj and I c are
equal, thereby producing a prolate sym m etric top. This was confirmed by observing
only tran sitio n frequencies a t 6B , 8B , 10B , 12B, and 1 4 5 .
Since no isotopic
su b stitu tio n was perform ed, stru ctu ra l param eters were not directly obtained from
th e present study.
Mn
Figure IV.3. S tru ctu re of C pM n(C O ) 3 .
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
82
T he small distortion p aram eters D j and D j k shown in Table IV.4 indicate
a fairly rigid stru ctu re for this molecule. A com parison of D j values for CpNiNO
and C o(C O )3NO w ith C pM n(C 0)3 (see Table IV .5) show th a t these molecules are
fairly consistent in th eir rigidity and undergo very little distortion due to ro tatio n
of the molecule.
This is the first gas phase m easurem ent of eQqaa for th e m anganese nucleus
so com parisons to other eQqaa values cannot be made.
However, a solid sta te
nuclear quadrupole resonance (N Q R) stu d y by Brill et al. [64] provided a value for
eQq = 64.29 M H z . It is ap p aren t th a t th e ir value corresponds to coupling along th e
sym m etry axis of C pM n(C O )3 which corresponds to th e a-axis in th e ro tatin g fram e.
C om paring our eQq = 68.00(4) MHz w ith th e solid state N Q R m easurem ent we see
a good agreem ent betw een these two values. T he 4 MHz difference could possibly
be a ttrib u te d to th e p ertu rb atio n s induced by th e lattice effects in th e solid sta te
whereas th e gas phase m easurem ent corresponds to the free molecule coupling. T he
large eQq for C pM n(C O )3 can be a ttrib u te d to th e effects of the ligands, especially
the Cp-ring. T he C p-ring is a five electron donor to the Mn an d lies along the
sym m etry axis of th e molecule. T he C p-ring appears to make a large contribution
to the axial (a-axis here) electron density and therefore contribute significantly to
the large field gradient and quadrupole coupling value eQq = eQqaa = 68.00 MHz.
In th e present results, we rep o rt only th e Cbb com ponent of the spin ro tatio n
interaction. P aram eter fits including th e Caa com ponent were m ade and resulted
in an im provem ent of th e overall fit by 0.041 kHz. However, th e value of th e sixth
param eter was Caa = —2.2 ± 3.4 kHz w ith th e 2a error lim its being larger th a n the
param eter. O ften th e values of th e spin ro tatio n com ponents in sym m etric tops are
very similar. For exam ple, spin ro tatio n com ponents C aa and Ccc for nitrogen in
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
83
am m onia are 13.6 kHz an d 11.4 kHz respectively, so one m ight expect th a t th e Cbb
and C aa com ponents in C pM n(C O )3 are also similar. W hen we exam ine equation
(2), we note th a t Cbb can be determ ined from K — 0 d a ta only, b u t to determ ine
(C aa — Cbb) we m ust use th e K ^ 0 d ata. T h e quality of K = 0 d a ta is usually
b e tte r an d m ore K = 0 tran sitio n s are accurately m easured due to stronger signals
so Cbb is much b e tte r determ ined th a n (C aa — Cbb)It is interesting to note th e possibility of C pM n(C 0)3 undergoing in ter­
nal hindered ro tatio n . Since th e effects of hindered ro tatio n are usually not seen
for sym m etric top molecules[65], except for sym m etric tops in excited vibrational
states, hindered ro tatio n effects could n o t be observed for this molecule. In fu tu re
isotopic su b stitu tio n studies of C pM n(C 0 )3, it could be determ ined if this molecule
undergoes internal ro tatio n by su b stitu tin g any one of th e CO ligand atom s. By this
substitutio n , th e sym m etry of th e CO group would be destroyed thus changing the
the m om ents of in ertia to give a slightly asym m etric top . T he resulting microwave
spectrum would th en reveal w hether or n o t C pM n(C O )3 is hindered rotor.
A n x-ray diffraction stru c tu re for C pM n(C 0)3 was obtained by B erndt and
M arsh[66]. T heir averaged stru ctu re (corrected for libration effects), can be used to
calculate ro tatio n al constants for com parison w ith th e present results. Using th eir
crystal stru ctu re d a ta we o b tain A = 1053.8 MHz and B = C = 821.3 MHz in
excellent agreem ent w ith our m easured value B = 828.033 MHz. This appears to
indicate th a t th ere is very little change in stru ctu re of the molecule in a crystal,
com pared w ith th e gas phase stru ctu re.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
84
IV .3 .n i Sum m ary
Microwave m easurem ents have been m ade for C pM n(C 0)3 confirming it as
a prolate sym m etric top. T h e d istortion param eters D j and D j k were well d eter­
m ined and com parison of D j values w ith some oth er transition m etal complexes
indicate th a t these molecules are rigid in th eir geom etries and undergo very little
distortion. T he hyperfine constants eQqaa and Cbb are th e first gas phase m easure­
m ents for these p aram eters for Mn. T he inform ation obtained in th e present study
should provide additional insight in to th e n atu re of molecules containing tran si­
tion m etal complexes and should aid theoreticians in semi-empirical and ab initio
calculations.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
85
C H A P T E R V:
M ICROW AVE S P E C T R A O F
C Y C LO B U TA D IEN E IRO N TRI-C A R B O N Y L,
C Y C LO H EX A D IEN E IRON TR I-C A R B O N Y L,
AND B U TA D IE N E IRON TR I-C A R B O N Y L
Iron containing tran sitio n -m etal com plexes are well known for th eir abil­
ity to catalyze various olefin reactions including hydrogenation, hydroform ylation,
isom erization, an d polym erization[67, 68]. In th is C h ap ter we look a t three olefin
containing iron carbonyl complexes an d investigate th e ir stru ctu ral param eters in
the gas phase using th e P B -F T S technique.
By determ ining how th e stru ctu re
and electronic properties of th e olefin are altered by th e m etal-olefin bonding, one
can b e tte r u n d erstan d how th e reactiv ity of an olefin is modified by its interaction
w ith a m etal atom . Therefore, precise s tru c tu ra l m easurem ents of m etal complexes
provide inform ation on bonding which can be related to their electronic structure.
V .l T h e M icrow ave S p ectru m o f C y clo b u ta d ien e Iron tri-carbonyl
C yclobutadiene is a p articu larly elusive com pound. T he com pound is only
stable as a ligand of low valent tran sitio n m etal complexes such as iron in cyclobu­
tadiene iron tricarbonyl. C yclobutadiene iron tricarbonyl (C bFe(C O )3 ) was first
synthesized by Emerson[69]. Since th en , th e re have been a num ber of speculations
and experim ents on the stru ctu re of unco o rd in ated cyclobutadiene (Cb) and Cb
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
86
com plexed as a ligand w ithin a molecule (CbM ). Infrared m atrix isolation studies[70] indicate th a t Cb is slightly rectangular w ith altern atin g single and double
bonds.
W hen cyclobutadiene is su b stitu ted w ith bulky te rt-b u ty l groups, x-ray
crystallographic data[71] show th a t th e su b stitu ted Cb is rectangular. However,
th e rectan g u lar geom etry is probably a direct consequence of packing effects and
th e bulky su b stitu en t ra th e r th a n th e n atu re of Cb itself. T he rectangular geom etry
observed in th e m atrix isolation studies[70] can be explained by th e pseudo-JahnTeller effect which leads to a change in geom etry tow ard lower sym m etry in system s
th a t have sets of degenerate orbitals th a t are n o t fully occupied [72]. Therefore,
it has been of interest, b o th experim entally an d theoretically, to investigate the
conditions for square vs. rectan g u lar geometry.
A lthough several experim ental atte m p ts have been m ade to characterize
th e stru ctu re of Cb in C bFe(C 0 )3 , none have clearly resolved th is issue. Electron
diffraction studies by O berham m er et al. [73] and Davis et al. [74] have determ ined
th e basic geom etry for C bF e(C 0)3 which is shown Figure V .l. However, this tech­
nique is insensitive to asym m etries in th e stru c tu re of th e Cb ring.
An earlier
microwave study[75] indicates th a t C bFe(C 0 )3 is a sym m etric top. However, this
previous stu d y reported signals w ith 20-60 MHz h alf w idth. These low resolution
m easurem ents were not capable of observing small asym m etry sp littin g which would
result from any d ep artu re from a sym m etric top (square Cb) stru ctu re. T he high
resolution available to P B -F T S will be useful in resolving th e square vs. rectangular
question of C bF e(C O )3 .
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
87
Figure V .l. S tru ctu re of C bFe(C 0 )3 . C oordinates o b tain ed from
reference [73]. The a —inertial axis is collinear w ith th e C 4 an d C3
axes of th e cyclobutadiene and Fe(C 0)3 respectively.
V .l .i E xp erim en tal
T he C bFe(C 0)3 was synthesized following th e basic procedure outlined
by P e ttit and Henery[76]. Cis-3,4-dichlorocyclobutene (0.02 mol) (Fluka, C at. #
35635) and iron nonacarbonyl (0.05 m ol) (S trem Chem ical, C at. # 26-2640) were
added to distilled benzene (20.0 ml), as a solvent, in a 100 ml round b o tto m flask.
R eactan ts and products were handled in th e in ert atm osphere of a dry box. The
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
88
reac tan ts were refluxed a t 55° C w ith continuous stirring for 1.5 hours. T he reac­
tion m ix tu re gradually changed from an orange to a dark green solution w ith a
noticeable evolution of carbon m onoxide gas. T he contents of th e vessel was then
filtered an d th e residue was w ashed w ith pen tan e and solvent was removed from
th e filtrate by enhanced evaporation under vacuum to afford th e p ro d u ct as a dark
green oil. F u rth er purification of th e p ro d u ct was achieved by distillation under
vacuum . T he dark green oil which contained C bFe(C 0 )3 and iron pentacarbonyl
w ith trace am ounts of triirondodecacarbonyl was th en transferred to a sam ple cell
which could be fitted to th e pulsed valve of our spectrom eter.
T h e pulsed valve and sam ple were m aintained a t a constant tem p eratu re of
60° C w ith argon flowing over th e sam ple an d pulsed th ro u g h th e valve as described
previously in C h ap ter II. The spectrum was m easured in th e 4-16 GHz range. The
J = 4 —> 5 tran sitio n was predicted to be n ear 9624 MHz based on th e previous
study[75]. U nder high resolution we observed a signal a t 9619.7 MHz w ith closely
spaced com ponents w ith less th a n 40 kHz separation.
These com ponents were
assigned to th e J = 4 —> 5 tran sitio n of C bFe(C O )3 w ith sp litting due to centrifugal
distortion.
Corresponding com ponents of th e J = 2 —> 3, through J = 7 —> 8
transitio n s were m easured an d are listed w ith th eir respective assignm ents in Table
A .III.l.
Signal strengths for single gas pulses ranged from 3 /1 (S /N ) for low J
tran sitio n s to 1/1 for higher J transitions. W eak FID signals were detected at 17315
MHz due to K com ponents of th e <7 = 8 —> 9 tran sitio n b u t were n o t included in
the present analysis due to uncertainties in th e line center m easurem ents.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
89
V .l .i i R esu lts and D iscu ssion
Spectroscopic p aram eters were obtained by fitting th e m easured frequencies
in Table A .III.l to p rolate sym m etric top tran sitio n frequencies given by
1/ = 2B ( J + 1) - 4D j ( J + l )3 - 2D j k ( J + l ) K 2
(35)
T he p aram eter values o b tain ed from this analysis are shown in Table V .l. Since
A '= 0 and A '=1 com ponents were not resolved for J — 2 —> 3,3 —> 4, 4 —> 5, and
5 —> 6 tran sitio n s, the calculated frequencies for K = 0 and A”= l were averaged
and fitted to th e corresponding observed line centers. T he uncertainty of B , D j ,
and D j k o b tain ed from these calculations is well w ithin th e expected precision of
microwave m easurem ents. These param eters are com pared to those of th e earlier
stu d y [75] in Table V .l. The high resolution Fourier transform m ethod employed in
this stu d y has enabled us to o b tain two orders of m agnitude im provem ent in the
A -rotatio n al constant.
Table V .l. B est fit values for adjustable p aram eters obtained using equation
1 and d a ta in Table A .III.l. T h e quoted errors are 2a for th e param eters and
th e a FIT — 7.6 kHz. Also shown are the the results obtained from previous
work [75].
p aram eter
B
present work
961.9856(8) MHz
Dj
0.184(8) kHz
D jk
1.20(3) kHz
previous work
962.41 MHz
0.41 kHz
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
90
T he centrifugal distortion param eter D j is approxim ately half of th e previ­
ously rep o rted value[75]. The difference can be a ttrib u te d to the ability to resolve
th e K com ponents in the current stu d y where as they were not resolved in th e p re­
vious study. T he previously m easured transitions were broadened by 20 to 60 MHz
half-w idth, thus skewing th e actu al line center m easurem ents due to th e unresolved
K com ponents.
D j can be used as a m easure of how strongly th e C b-ring is bound to th e
F e(C O )3 group. To a crude approxim ation, C bFe(C O )3 can be m odelled as a psuedodiatom ic where th e C b-ring an d Fe(C O )3 groups are regarded as p oint masses.
Since D j is th en a m easure of the susceptibility to centrifugal d istortion along this
internuclear axis, th e previous value of D j (see Table V .l) would in d icate a m ore
weakly bound complex th a n the current m easurem ents indicate. T he disto rtio n p a ­
ram eters observed for C bFe(C O )3 an d o th er tran sitio n m etal complexes[ 77 , 78, 79]
are sum m arized in Table V.2. U nfortunately, there is no x-ray d a ta for C bFe(C O )3
to com pare w ith th e current microwave results and older gas phase electron diffrac­
tion studies[73,74] give a wide range for the ro tatio n al constants calculated from
their stru ctu ra l param eters; 934.70 to 969.85 MHz.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
91
Table V.2. C om parison of d istortion param eters for some o th er tra n ­
sition m etal complexes w ith values for C bFe(C 0 )3 .
Molecule
D istortion
D j { kHz)
P aram eter
D j k ( kHz)
C bFe(C O )3
0.184(8)
1.20(3)
C o(C O )3NO
0.17(8)“
0.6
CpNiNO
0.08(2)6
2.70(6)
C pM n(C O )3
0.088(9)c
-0.04(3)
A j(k H z )
B uFe(C O )3rf
0.075(2)
A jA '(kH z)
0.16(1)
“ see C h ap ter IV
6 see [77]
c see C h ap ter IV
d see section V.4
B uF e(C 0 )3= butadiene irontricarbonyl.
A j and A j k sim ilar to th e sym m etric to p D j and D j k -
T h ere has previously been uncertainty w hether th e Cb ring is square or
slightly rectan g u lar when complexed[74]. T he present results, indicate th a t the Cb
ring is very nearly square w hen com plexed to Fe(C 0 )3. A rectangular stru ctu re
would make I& and I c different thus producing an asym m etric top spectrum . By
m odifying th e electron diffraction coordinates[73] for the Cb ring and assum ing no
change in th e Fe(C 0 )3 group, one can calculate th e effects of a slightly rectan g u lar
stru ctu re for th e Cb ring. By m aking a 1 mA difference in th e lengths of th e two
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
92
adjacent C-C bonds of th e Cb ring, B — C = 8 0 kHz and results in an asym m etry
splitting in th e spectrum of 230 kHz for th e J = 3 <— 2 transition. The observed
splitting is 30 kHz for this tran sitio n and is due to the small D j k value. Therefore,
th e vibrationally averaged stru ctu re of th e Cb ring appears to be square to b e tte r
th a n 1 mA. O th er stru ctu ra l anom alies in C bFe(C O )3 can also be tested such as
tilting th e cyclobutadiene ring relative to th e a-inertial axis. By calculating th e
effect of tiltin g th e C b ring by ± 2°, we can place an estim ated value of B —C = ±60
kHz which would correspond to a 100 kHz asym m etry splitting in the spectrum .
Therefore, we conclude th a t th e Cb ring is vibrationally averaged to w ithin less th a n
± 2° of being perpendicular to th e a-axis of the molecule. T he overall stru ctu re of
C bFe(C O )3 is shown in F igure V .l; th e C4 axis of the cyclobutadiene ring and the
C 3 axis of th e Fe(C O )3 are coincident w ith th e a-m olecular axis. T he b- and c-axes
are perpendicular to th e a-axis m aking / j and I c equal. This sym m etrical stru ctu re
is consistent w ith the observed spectrum .
T h e eclipsing of one of th e corners of the Cb ring w ith a carbonyl group
and locking it in to a fixed position would be detected only if the Cb ring was tilted
by this interaction. However, if a rigid non-tilted, eclipsed stru ctu re was present in
C bFe(C O )3 , th e m om ents of in e rtia w ould be th e same as those observed in the
present study. Therefore, a stru ctu re like this would still be consistent w ith the
observed spectrum .
Finally, th e possibility of hindered internal ro tatio n of the cyclobutadiene
ring relative to th e Fe(C O )3 group cannot be ruled out. Hindered ro tatio n effects in
sym m etric tops are not observed[80], since th e energy levels in a sym m etric top are
p ertu rb ed by hindered ro tatio n equally in each rotational sta te thus producing no
observable splittin g in th e spectrum . S p litting could be produced by asym m etric
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
93
isotopic su b stitu tio n and this m ight reveal w hether or not CbFe(C 0)3 undergoes
internal rotation.
V .l.iii Sum m ary
C bFe(C 0)3 is shown to be a sym m etric top molecule from th e analysis of
its high resolution microwave spectrum . Evidence has been given to indicate th a t
the Cb ring has a square geom etry in C bF e(C 0 )3 . A non-tilting eclipsed stru ctu re
cannot be distinguished from a staggered stru c tu re since th e respective m om ents
of in e rtia identical and consistent w ith th e observed spectrum . H indered internal
ro tatio n effects were not observed, however, evidence for such internal m otion should
be revealed by fu tu re isotopic studies.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
94
V .2 C vcloh exad ien e Iron tri-C arb on vl
In th e 1970’s, C o tto n and coworkers [81]investigated a series of su b stitu ted
cyclohexadiene iron tri-carbonyl com pounds using X-ray diffraction. R esults from
this stu d y yielded some stru ctu ra l p aram eters related to th e cyclohexadiene ring
itself as well as the Fe(C 0)3 group. However, cyclohexadiene iron tri-carbonyl (ChexFe(C O )3 ) u n su b stitu ted does n o t form a crystal at room tem p eratu re an d there­
fore stru c tu ra l d a ta could not be obtained on th e pure u n su b stitu ted C -hexFe(C O )3
. In 1989 Tam and coworkers[82]obtained X -ray d a ta for C-hexFe(CO )3 in a crys­
talline m atrix complex complex w ith thiourea. To date, this was the only known
stru c tu ra l d a ta on C-hexFe(C O )3 . However, th is stu d y did n o t determ ine how
m uch th e th e thiourea affected th e intrinsic gas phase stru ctu re for this molecule.
By analyzing gas phase ro tatio n al constants for two isotopom ers of C -hexFe(C 0)3
and relating them to stru ctu ra l p aram eters of th e molecule, th e affect of th e thiourea
interactio n w ith C-hexFe(C0)3 can be investigated.
A sample was obtained from S trem chem ical ( # 26-0850) and used w ithout
fu th er purification.
T he liquid sam ple was placed in th e hot source ap p aratu s
described in C h ap ter II and heated to 38 °C. This h eat was sufficient to o b tain a
high enough vapour phase fraction of C -hexFe(C O )3 in an A r carrier gas. T he gas
m ixture was m aintained a t approxim ately 160 Torr backing pressure and signal to
noise ratios of 20-30 to 1 were obtained for th e 56Fe species by averaging 50-100 gas
pulse F ID ’s. R otational tran sitio n s for th e 54 Fe species were m easured in n atu ral
abundance by averaging several hundred to a th o u san d gas pulses.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
95
V .2 .i M icrow ave S p ectru m and A n a ly sis
S p ectra obtained for 56Fe and 54Fe isotopes of C-hexFe(C O )3 contained
a — and c— dipole R -branch characteristics of a near p ro late asym m etric top. 76
transitio n s for J = 3 up to J = 9 and K p up to 6 were m easured for th e 56Fe species
(see Table A .V .2). The a— dipole transitions were 2 to 3 tim es stronger th a n the c—
dipole tran sitio n s. 14 a — dipole tran sitio n s were m easured for th e 54Fe isotopom er
(see T able A .V .3), which has 5.8 % n a tu ra l abundance.
B o th th e 56Fe and 54Fe sp ectra were fitted to W atso n ’s A -reduced Hamil­
tonian in th e I r representation[83] including term s for qu artic distortion,
H = H r o t + Hd
(36)
w here
H ro t = AJl
+ BJ2+
C J 2c
and
H d = - A j ( J 2)2 - A j K J 2J 2a
- 28 j K J 2{ J 2 - J 2) - 6K [ J2( J 2b - J \ ) + ( J 2b - J 2) J 2a ]
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
96
T he resulting fit to A, B , and C and th e four q u artic centrifugal distortion p aram ­
eters, A j , A j k ,
8
i< yielded stan d ard deviations of 3.1 kHz for th e 56Fe and 0.5
kHz for 54Fe isotopom er. T he A an d A k constants were highly correlated in th e
ro tatio n al fit. Therefore, A k was fixed a t zero an d was n o t determ ined by this
d a ta set. M easured and calculated tran sitio n frequencies are given in Tables A .III.2
and A .III.3 in A ppendix A .III. T he ro tatio n al co n stan ts an d d istortion param eters
obtained from these fits are given in Table V.3.
Table V.3. R o tatio n al an d centrifugal d isto rtio n p aram eters obtained
from least squares fits to frequencies in Tables A .III.2 an d A .III.3 in
A ppendix A .III. listed uncertainties are 2cr
P aram e ter
56C -hexFe(C O )3
54C -hexFe(C O )3
A(M Hz)
960.0298(4)
960.02(1)
B(M Hz)
681.8343(2)
682.0245(6)
C(M Hz)
659.3087(2)
659.4821(4)
A j(k H z )
0.027(2)
0.030(2)
A j A-(kHz)
0.051(9)
0.05(2)
0.0086(8)
0.008(2)
M kH z)
M kH z)
-0.59(6)
-0.4(1)
V .2 .ii S tru ctu ral P aram eters
From th e X-ray data[82] sum m arized in Tables V.4 and V.5 (stru ctu re I)
show th a t th e o rien tatio n of th e ligands to be th e sam e as th e num erous su b stitu ted
cyclohexadiene analogs rep o rted by C o tto n an d co-workers[81]. T he unique CO
ligand is o riented tow ard th e open side of th e diene an d th e cyclohexadiene ring is
b ent w ith th e m ethylene groups situ a ted out of th e diene plane (see F igure V.2).
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
97
However, th e X -ray coordinates of th e C-hexFe(CO )3 atom s show a large deviation
from th e expected x , z plane of sym m etry and rotatio n al constants calculated for this
stru c tu re yield significantly different values from th e observed microwave results.
T able V.4. R o tatio n al constants (obsd) from th e high-resolution mi­
crowave sp ectra of 56Fe and 54Fe C-hexFe(CO )3 isotopom ers com­
p ared w ith those calculated from th e X-ray stru ctu re (I and II) and
values calculated w ith optim ized ligand orientations (III).
P aram e ter
obsd
A (56Fe)
B (56Fe)
C (56Fe)
A (54Fe)
B (54Fe)
C (54Fe)
(M Hz)
960.030
681.834
659.309
960.020
682.024
659.482
I
972.704
680.594
664.298
972.710
680.780
664.475
8.655
II
972.644
679.719
664.583
972.650
679.906
664.583
8.654
III
960.025
681.828
659.301
960.025
682.030
659.490
0.007
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
98
Table V.5. S tru ctu ral p aram eters describing th e relative orientation
of th e CeHg an d CO ligands relative to th e iron atom (see Figure
V.2)a
P aram eter
a
microwave fit
X -ray d a ta
(deg)
71.9(3)
76.6
P (deg)
57.7(2)
52.06, 53.4
A X
(A)
0.860(4)
0.998
a T h e microwave fit values were determ ined form a least squares fit
to th e m easured ro tatio n al constants, w ith uncertainties rep o rted as
two stan d ard deviations. 6 T h e plane of sym m etry is not m aintained
in th e th io u rea com plex so two different X-ray values were obtained.
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission.
99
Fe
Figure V.2. S tru ctu ral p aram eters used in th e least squares fit to the
isotopic ro tatio n al constants.
P rin cip al m om ents of in e rtia an d th e corresponding ro tatio n al constants
were calculated from th e X -ray data[84] using a cartesian coordinate system es­
tablished w ith th e diene carbon atom s in th e x, y plane and th e 2 axis passing
through th e Fe ato m (see F igure V.2). The X-ray study did not determ ine the
hydrogen ato m coordinates so stan d ard sp2 and sp3 bond angles and bond lengths
of r(sp 2C — H )
=
1.080 A an d r (s p 3C — H )
=
1.099 A were used from th e
microwave results of [85, 86]. Elem ents of th e inertial tensor were calculated and
then diagonalized to o b tain principal m om ents of inertia, ro tatio n al constants, and
Reproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
100
th e orientation of th e principal in ertial axis system . Calculations were carried out
for the original X -ray stru ctu re (I), a sym m etrized stru ctu re (II) in which a plane of
sym m etry was im posed on the X -ray stru ctu re by replacing (x , y, z) and (a:, —y, z)
m irror coordinates w ith th eir average m agnitudes. The results are com pared w ith
the microwave m easurem ents in Table V.4. T he calculated ro tatio n al constants are
not very sensitive to the assum ed hydrogen atom coordinates, for exam ple, increas­
ing all th e diene C-H bonds by 0.01 A results in a decrease in the A and B rotatio n al
constants by 0.4 an d 0.9 MHz, respectively, an d an increase in C by 0.2 MHz. The
residuals suggest a loss of p lan ar sym m etry and fu rth er distortions in th e m olecular
stru ctu re due to th e interaction of th e th io u rea complex.
T he stru ctu ra l p aram eters in Figure V.2 are sensitive to deform ation from
various interactions in th e solid complex. By fitting various com binations of these
stru ctu ra l p aram eters to th e observed microwave d a ta and attem p ts to adjust the
m etal-diene separation from the solid sta te value (R
=
1.626 A) did not signif­
icantly improve th e value of th e calculated ro tatio n al constants. However, when
small adjustm ents in th e o rientation of th e ligands around the central iron atom
(see Tables V .6 and V.7) were m ade a significant improvement was m ade in th e fit.
Some selected interatom ic distances and intram olecular angles for th e optim ized
stru ctu re (III) are com pared w ith b o th th e original X-ray (I) and sym m etrized
X-ray (II) stru ctu res in Tables V .6 an d V.7. T he geom etry and orientation of ChexFe(C O )3 in th e principal axes is shown in Figure V.3. Figure V.3 shows th a t
th e stru ctu re is consistent w ith th e observation of only a— and c— dipoles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
Table V .6. Selected interatom ic distances (A) from the X-ray stru c­
tu re (I an d II) of C -hexFe(CO )3 com pared w ith those calculated w ith
optim ized ligand orientations (III).
distance
Fe-C l
Fe-C2
Fe-C3
Fe-C4
Fe-C5
Fe-C8
Fe-C9
O l-C l
02-C 2
03-C 3
C4-C5
C4-C9
C5-C6
C6-C7
C7-C8
C8-C9
I
1.799
1.798
1.794
2.030
2.104
2.100
2.034
1.140
1.140
1.142
1.409
1.398
1.507
1.540
1.501
1.423
II
1.799
1.795
1.795
2.032
2.102
2.102
2.032
1.139
1.144
1.144
1.416
1.398
1.503
1.538
1.503
1.416
III
1.799
1.795
1.795
1.968
2.125
2.125
1.968
1.139
1.144
1.144
1.416
1.398
1.503
1.538
1.503
1.416
V .2 .iii R esu lts
T h e microwave results are in agreem ent w ith the basic molecular stru ctu re
obtained from th e X -ray d a ta [82].
However, th e stru ctu re of C-hexFe(CO )3 is
deform ed by various interactions w ith th e th io u rea complex. T he large differences
are due to angular deform ation in th e o rien tatio n of th e ligands around th e central
iron atom as shown in Section V.2.ii. C om paring the angles C i-Fe-C 2, C j-Fe-C s,
and C2-Fe-C3 for th e gas phase molecule (stru ctu re III) w ith the original X-ray
stru ctu re (stru ctu re I, Table V.7) indicates th a t th e CO ligands are disto rted by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
repulsive interactions w ith th e thiourea resulting in a reduction of th e carbonyl
‘um brella’ angle of 2-3 °. T he asym m etry of th e X -ray d a ta w ith respect to th e
a — c plane, can be seen in Tables V.6-V.7.
From this analysis of th e X-ray and microwave d a ta it is clearly shown th a t
th e intrinsic gas phase stru ctu re of C-hexFe(CO )3 has been severely disto rted in
th e th io u rea complex. T he microwave d a ta is valuable in th e sense of indicating th e
m agnitu d es of these differences as well as some of th e inherent m olecular stru ctu ral
param eters.
T h e com bination of b o th X -ray an d microwave d a ta will result in
an overall b e tte r understanding of th e stru ctu re of this complex th a n using either
m eth o d independently.
F igure V.3. Conform ation and orientation of C-hexFe(C O )3 in the
principal in ertial fram e as determ ined from microwave spectra.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
Table V.7. Selected intram olecular angles from th e X -ray stru ctu re (I
and II) of C-hexFe(CO )3 com pared w ith those calculated w ith o p ti­
m ized ligand orientations (III).
angle(degree)
C l-Fe-C 2
C l-Fe-C 3
C l-Fe-C 4
C l-Fe-C 5
C l-F e-C 8
C l-Fe-C 9
C2-Fe-C3
C l-Fe-C 2
I
102.4
100.4
131.8
93.2
92.3
131.3
91.4
102.4
II
III
101.4
101.4
131.5
92.7
92.7
131.5
91.4
101.4
103.6
103.6
132.3
92.9
92.9
132.3
94.1
103.6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
V .3 B u ta d ien e Iron tri-C arb on vl K raitchm an A nalysis
T h e b u tadiene iron tri-carbonyl (B uFe(C 0)3 ) complex (shown in Figure
V.4) is one of th e earliest known 7r-bonded tran sitio n m etal complexes and was
described as early as the 1930’s [87]. B uFe(C O )3 did not a ttra c t much atten tio n
since it was first thought to be a m etalacycle stru ctu re w ith cr-bonding to th e m etal
atom .
T h e presently accepted 7r-bonded stru ctu re was proposed by H allam and
Pauson[88], based prim arily on th e stability of th e complex. E xperim ental results
on reactions w ith ozone and lithium alum inium hydride su pported th e 7r-bonded
stru c tu re B uF e(C O )3 . A large num ber of diene iron carbonyl complexes have been
reviewed by P e ttit and Emerson[89].
F igure V.4. M olecular stru ctu re and fitted param eters as determ ined
from the K raitch m an analysis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
T he microwave frequencies for the norm al and isotopically labeled forms of
B uFe(C O )3 have been reported[90, 91]. T ransition frequencies were fitted to a W at­
son’s A -reduced H am iltonian including term s for quartic d istortion (see equation
(2) section V.3.i). T he resulting rotatio n al constants were th en fitted to a m olecular
stru ctu re w ith th e param eters shown in Figure 4 using a least squares fitting rou­
tine. An alternative m ethod to ob tain a m olecular stru ctu re from this microwave
d a ta was pursued by using K raitch m an ’s equations for single isotopic substitution.
T he results o b tained from K raitch m an ’s analysis can easily be in terp reted in term s
of the m olecular stru ctu re an d can be com pared to th e least squares fit analysis.
V .3 .i K raitch m an ’s E q u ation s and M olecular stru ctu re
From K raitch m an ’s equations one can o b tain relative (relative to th e nor­
m al isotopic species referred to as the parent molecule) absolute coordinates of an
isotopically su b stitu ted atom w ithin a molecule. A detailed derivation and descrip­
tion of K raitch m an ’s equations is given in G ordy an d Cook[20] and shown here is
a sum m ary of those equations used in the analysis for B uFe(C O )3 . T he absolute
relative coordinate a in th e principal axis system is given by,
A Pa ( x + _ A P ^ \ /
a =
where n =
+
APZ
1/2
(37)
M is the to tal mass of th e paren t molecule an d A m is the
isotopic shift in m ass and
A P a = Q ) ( —A J a + A l b + A I C)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
and A I a = I'a ~ I a • APb and A Pc can be generated from a cyclic perm u tatio n of a, b,
and c in th e above equations. These equations were used w ith th e parent molecule
of norm ally labelled B uFe(C O )3 as well as th e triply su b stitu ted deuterium using
th e D 4 (term inal hydrogens su b stitu ted , see Figure V.4) as the paren t molecule.
T h e resulting a, 6, and c coordinates can be used w ith th e m olecular plane of
sym m etry bisecting the b u tadiene and containing th e unique carbonyl group C 5O 1.
N early enough stru c tu ra l param eters such as bond lengths and angles could be
obtained to describe th e overall geom etry of th e B uFe(C O )3 complex using only
th e K raitch m an ’s analysis. A fairly detailed description of th e butadiene fragm ent
could be o b tain ed since coordinates for all b u t two of the atom s (H3 and H4) could
be calculated. From th e least-squares stru ctu ral analysis, the bond lengths for the
term inal pro to n s H i—Ci an d H2—Ci were constrained to be equal. However, from
the K raitch m an ’s analysis, th e H i—Ci bond length was found to be 0.007 A longer
th a n the H2—C i bond.
T h e results of th e K raitch m an ’s analysis are listed in Table V .8. Only the
first four bond lengths an d th e first four angles were determ ined entirely from the
K raitchm an analysis data. O th er p aram eters required th e Fe coordinates in the
principal axis system from fit results as discussed below.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
Table V .8 Com parison of interato m ic distances and angles derived from the
K raitchm an analysis w ith th e Fe coordinates from th e stru ctu ra l fit and
com parable p aram eters determ ined entirely from th e stru ctu ra l fit.
K raitm an
H l- C l“
H2-C1“
C l-C 2 “
B ond lengths A
1.0958(2)
stru ctu ra l fit
1.089(5)
1.385(7)
1.087
1.087
1.385
1.409(1)
1.421
2.127
C2-Fe
C3-Fe
2.1270(3)
2.0870(2)
2.0870(2)
2.088
2.088
H l-F e
2.8230(1)
2.815
H2-Fe
2.6511(7)
2.650
Fe-C5
Fe-C6
1.7706(0)
1.771
1.7824(9)
1.783
Fe-C7
1.7824(9)
1.783
C2-C3“
C l-F e
B ond angles °
H l-C l-H 2 “
120.8(3)
117.0
H1-C1-C2“
H 2-C l-C 2“
112.1(4)
118.6(2)
C l-C 2-C 3“
118.4(2)
117.9
117.9
118.2
z-axis-Fe-C5
81.5(4)
82.4
z-axis-Fe-C6
z-ax is-H l-0 2
52.2(4)
20.5(2)
52.0
19.6
z-axis-Fe-03
xz- pl.-z-axis-0 2
52.2(2)
52.0
113.8(2)
113.3
xz- pl.-z-axis-03
113.8(2)
113.3
D ihedral angles °
H1-C1-C2-H3
11.3(3)
12.8
H2-C1-C2-H3
136.9(2)
136.5
“ For these p aram eters, th e K raitch m an colum n values were determ ined en ­
tirely from th e K raitch m an analysis; o th e r p aram eters required d a ta from
th e fit results or o th er m easurem ents. p l.= p lan e.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
T he dihedral angles betw een H i—C j—C2—H 3 and H 2—C i —C2—H3 atom s
were calculated by fixing th e H3 position in th e carbon skeletal plane at 120 °to th e
C2 —C 3 bond w ith a bond length of 1.090 A . These param eters also show differences
in the term inal hydrogen coordinates as discussed above.
T he stru ctu re of th e F e(C O )3 p o rtio n of th e molecule can be evaluated using
th e K raitchm an analysis as well. To accom plish this, th e location of th e Fe atom
relative to center of mass for th e molecule m ust be calculated using th e fit results
since th e Fe atom was not isotopically labeled. T he coordinates in th e (a, b, c)
fram e for Fe obtained from th e stru c tu ra l fit were used in th e K raitchm an analysis.
T he Fe ato m is so close to th e center of mass th a t th e uncertainties from th e fit
results should n o t introduce any significant additional error into these K raitchm an
derived coordinates. A dditional inform ation on th e term inal hydrogen positions can
be obtained from K raitchm an using th e D 4 su b stitu ted isotopom er as th e paren t
molecule and th e D 3 isotopom er for th e single hydrogen substitution. T he C i —Hi
or the C i—H 2 interatom ic distances cannot be determ ined from these d a ta sets.
To calculate these distances one m ust assum e a location for th e Fe atom in th e D4
(a, b, c) coordinate system . T he iron coordinates from th e D4 stru ctu ral fits were
used in a sim ilar m anner as for th e norm al isotopic K raitchm an analysis. Values for
Fe-Hj and Fe-H2 distances were 2.8304 and 2.6623 A , respectively. W hen com pared
to values given in Table V .8 , th e agreem ent between these two distances in th e D4
and H 4 analyses is rem arkably good. T h e results from th e K raitch m an ’s analysis
are in excellent agreem ent w ith th e results from th e stru ctu ral fits and b o th sets
are given in Table V .8.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
C H A P T E R VI:
MICROWAVE SPE C T R U M , STR U C TU R E,
AND D IPO L E M O M EN T F O R T H E H CCH -CO CO M PLEX
T h e infrared spectrum of acetylene-carbon monoxide van der W aals com­
plex was recently observed in th e 3p region by M arshall, P ritc h ard and M uenter[
92,93]. T he spectrum indicated a linear com plex an d la ter work[94] confirmed th a t
th e CO molecule was bound to th e acetylene th ro u g h a hydrogen carbon bond. A
num ber of oth er linear OC-HX complexes were studied previously including OCHF[95] , 0C-HC1[96] , OC-HBr[97] , OC-HCN[98] and OC-HI[99] . In all cases
hydrogen bonding to th e carbon atom of CO was observed.
In th e present work more precise values for ro tatio n and distortion con­
sta n ts for H CCH-CO an d H C C H -13CO were o b tain ed an d new results were ob­
tained for th e d eu terated isotopom ers H CCD -CO , D C CD -C O an d D CCH-CO . All
of th e above isotopom ers and isomers could be observed in a neon-helium ( ‘first run
neon’) expansion gas as well as in argon, b u t, only th e DCCH-CO isotopom er was
not found in th e argon expansion. T his could be due to the low barrier for ro tatio n
of the acetylene molecule ab o u t its center of m ass com bined w ith a facilitation of
this ro tatio n in argon. It is expected th e H CC D -C O isom er would be m ore stable
th a n the D CCH-CO isomer since m ost hydrogen bonds are stronger for D th a n for
H due to reduced ‘zero p o in t’ vibrational m otion of D relative to H.
V I. 1 E xp erim en tal
All m easurem ents were m ade using a Flygare-B alle type pulsed beam
Fourier transform microwave spectrom eter and th e reader is referred to C h ap ter
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
II for experim ental details of th e P B -F T S system .
S pectra for the five isotopic
species of th e H C C H -C O van der W aals complex were collected in the 4-18 G H z
range an d are shown in Table V I.1. Four of th e isotopic forms of th e H C C H -C O
com plex were observed by pulsing a 1-2% m ixture of acetylene and carbon monoxide
in 1 a tm of argon buffer gas into an evacuated (10~6 - 10- r to rr) Fabry-Perot cavity.
T he D C C H -12CO isotopom er was observed using th e sam e ratios of sam ple, b u t,
neon was used as th e carrier gas, since initial searches for this isotopom er in argon
yielded no observable signals. T h e carrier gas was changed to neon and a signal
belonging to th is isotopom er was found a t 7986.806 M H z . T he o th er isotopom eric
species were also observed in th e neon buffer gas.
Signals for th e H C C H -12CO and H C C H -13CO isotopom ers were unusually
strong for a complex consisting of one m onom er having no perm anent dipole mom ent
(acetylene) and one having a sm all dipole of 0.1D (carbon monoxide). T he signalto-noise ratios were on th e order of 100/1 p er gas pulse for all of the rotational
transitions. T he 13CO was purchased from Isotec Inc. ( # 83-70003).
T he d eu terated form s of acetylene, HCCD and D CCD, were m ade by drop­
ping a m ix tu re of 30% H2O an d 60% D 2O onto solid C aC 2 u nder atm ospheric
pressure. T he d eu terated species were trap p e d o u t and fu rth er purification was
accom plished by a trap -to -trap distillation u nder vacuum. A second m ethod for
synthesizing d eu terated acetylene was utilized by placing several ‘chunks’ of C aC 2
in an airtig h t container along w ith several small ice cubes m ade from 30% H2O and
60% D 2O solution. Once all reactan ts were placed inside, the container was sealed
and evacuated. Upon w arm ing, an increase in pressure was observed w ithin the
container. T he sealing flange was equipped w ith a ‘Swadge-lock’ valve and could
b e fitted directly to our gas handling system on th e spectrom eter. B oth m ethods
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I ll
of producing d eu terated acetylene worked equally well, however, th e la tte r m ethod
is appealing since th e sam ple is in ready to use form.
T he signals for the d eu terated species were n o t as strong as the H C C H 12C O ,-13CO isotopom ers p er gas pulse, signal- to-noise fts20/l. T he D C C H -12CO
complex signal strengths were considerably weaker even in neon, taking on average
several hundred gas pulses to o b tain reasonable signal- to-noise ratios.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
Table VI.1. Measured rotational transitions (J —> <7 + 1) for the five isotopomeric forms of the HCCH-CO van der Waals complex. Frequency in
MHz.
isotopom er
H C C H -12CO
HCCH —13CO
H C C D —12CO
D C C D - 12CO
D C C H —12CO
J
Frequency
1
5589.316(4)
2
8383.653(4)
3
11177.613(4)
4
13971.066(6)
5
16763.886(5)
1
5540.405(4)
2
8310.301(4)
3
11079.824(2)
4
13848.846(3)
5
16617.261(4)
1
5579.695(4)
2
8369.245(4)
3
11158.441(4)
4
13947.154(5)
5
16735.275(5)
1
5318.598(3)
2
7977.643(5)
3
10636.379(4)
4
13294.691(4)
5
15952.488(6)
2
7986.806(3)
3
10648.554(3)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
V I .2 S p ectra l A n alysis
T h e acetylene-carbon monoxide spectrum is th a t of a linear ro to r an d con­
tain s no observable hyperfine stru ctu re. T he presence of deuterium in some of th e
isotopom ers brings ab o u t th e possibility of m easuring th e deuterium quadrupole
coupling in this complex. However, sp ectra containing deuterium exhibited very
sm all sp littin g which were m ostly unresolved and no a tte m p t was m ade to o b tain a
quadrupole coupling constant for deuterium . All isotopom eric tran sitio n frequencies
were fitted to
v = 2 B ( J + 1) —4 D j( J + l )3
(38)
using linear regression. T he fit results to B an d D j are shown in Table VI.2 for all
five isotopom ers. All fits were well w ithin th e experim ental accuracy expected from
microwave m easurem ents. O nly two lines were m easured for D C C H -12CO so the
rep o rted B value was ob tain ed by using a D j value o b tained for th e H C C D -12CO
isotopom er. T h e B and D j values determ ined in th e present stu d y for H C C H -12CO
and H C C H -13CO species are in good agreem ent w ith those obtained by M arshall
et.al.[ 92, 94] using a m olecular beam infrared technique. T h eir values are also listed
in Table V I.2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
Table V I.2. F it results for th e five isotopom ers of H C C H -C O com pared w ith
results o b tain ed from references [92, 94].
isotopom er
B(M H z)
D j(kH z)
H CCH —12CO
1397.3705(6)
5.28(4)
HCCH —13CO
1385.1425(8)
5.15(6)
H C C D -12CO
1394.9635(3)
4.96(2)
D C C D -12CO
1329.369(1)
4.31(2)
D C C H —12CO
1331.1(1)
4.96(2)
HCCH —12C 0 6
1397.4(6)
5(2)
H C C H -13C 0 6
1385.4(6)
9(2)
“ value fixed to th e D j value for th e H C C D -12CO isotopom er.
6 ground s ta te constants o b tain ed from references [92, 94].
V I.2 . S tru ctu ral A n alysis
It is well known th a t van der W aals molecules undergo large am plitude
m otions ab o u t an equilibrium geom etry, which, in th e present case is linear. The
observed m om ents of in e rtia will th en be a function of th e stru ctu ra l param eters
induced by these v ibrational m otions. By taking into account these m otions, the
pseudodiatom ic m odel for van der W aals complexes can be m odified to include the
vibrationally averaged projections of th e m om ents to yield an effective m om ent of
inertia,
ir *"=
+
(i -
+ (i -
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
09)
115
where u PD = r™H^-<
' H7lJ ' Q \ an d R cm is the distance between th e centers of mass
\TnHCCH~'~TnCO >
of th e m onom er units.
T he bracketed quantities are th e average values for the
projections on th e R cm axis w ith vibrational m otion about th e linear equilibrium
geometry. Figure V I.1 shows th e stru ctu re of this complex. N ote th a t th e angle 0
represents th e v ibrational am plitude, not an equilibrium value.
Rcm = 5.018(6) A
Figure V I.1. T he vibrationally averaged stru ctu re obtained from the
stru c tu ra l fit to th e observed m om ents of inertia. T he angle 6 repre­
sents th e v ibrational am plitude of acetylene ab o u t R cm-
E xam ining equation (39), we see th a t upon isotopic su b stitu tio n of th e
m onom er un its there is a corresponding change in the centers of mass distance
R cm■ To account for this in th e fittin g algorithm we would calculate th e shifts in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
the center of m ass for th e respective m onom er u n its and then calculate th e new
centers of mass distance R cm. T he fitted results for each isotopom er are shown in
Table V I.3. T he reported value of R*m in Table VI.4 is th e corrected R cm distance
corresponding to th e paren t isotopic species H C C H -12CO. The angles
6
and <j>are
the angles of th e acetylene an d carbon m onoxide subunits respectively w ith R cmThese were assum ed not to be sensitive to isotopic su b stitution and no correction
factors for this were introduced into th e fittin g algorithm .
T he fit results to equation (39) in which R cm,
an d
sin2
$ are th e fit
param eters, are shown in Tables VI.3 an d VI.4. The R*cm distance of 5.018(6)
in good agreem ent w ith th e R cm distance of 5.011
92]. T he p aram eter 3i~
yields a value of
A is
A obtained by M arshall et.al.[
21° for the angle
6
and is consistent
w ith o th er linear hydrogen bound carbon monoxide complexes[ 95, 96, 97, 98, 99].
T he p aram eter SHLA was fixed to 0 since there was only one d a ta point which was
sensitive to this p aram eter, e.g. th e H CCH - 13 CO isotopom er. F its which included
this p aram eter did not yield a statistically significant value for ■
31” — .
Table VI.3 M easured and calculated ro tatio n al constants obtained from the
least squares fit to the observed m om ents of inertia. The param eter values
are listed in Table VI.4. Values in MHz. T he a f u was 0.583 MHz.
isotopom er
H C C H -12CO
H C C H -13CO
h c c d -12c o
d c c d - 12c o
D C C H -12CO
m easured
calculated
M-C
1397.370
1385.142
1394.963
1329.684
1331.225
1397.163
1385.336
1394.448
1329.770
1331.704
0.207
-0.194
0.515
-0.086
-0.479
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
Table VI.4. P aram eter values from th e fit to th e observed ro tatio n al con­
stants. Indicated uncertainties are 2 stan d ard deviations.
p aram eter
value
Rim
5.018(6) A
0.06(6)
V I .3. D t. F o rc e C o n s ta n t a n d B in d in g E n e r g y
R easonably accurate and consistent values for the centrifugal distortion
constant D j were obtained from analysis of the spectra. These D j values, com bined
w ith some stru c tu ra l d ata, can be used to calculate the stretching force constant
k s for changing th e separation of centers of m ass of th e monomers ( R cm) and to
estim ate th e binding energy. T h e norm al expression for D j for a diatom ic molecule
is modified[100] for a polyatom ic linear dim er to give
Dj=IFi1~B^~~
Ws V
R HCCH
Rco
I
(40)
where B is th e ro tatio n al constant of th e complex and the stretching frequency (for
the hydrogen bond) is such th a t
^
fJ'PD
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(41)
118
Using these relations k 3 values for th e various isotopom ers were calculated and listed
in Table V I.5.
Table VI.5. C alculated k 3, u3, re an d e values derived from m easured distor­
tio n constants D j and R cm values. T h e r e values are th e equilibrium centers
of m ass separation for the complexes.
isotopom er k 3(m dyne A *)
H C C H -12CO
H C C H -13CO
H C C D -12CO
D C C D -12CO
0.0171
0.0174
0.0185
0.0185
v3(cm *)
46.4
47.9
47.8
47.8
r e(-&)
e(cm 1)
5.0011
4.9788
4.9078
5.0006
299
301.5
311.5
323.4
T he L ennard-Jones p o ten tial has only two ad ju stab le param eters; the bind­
ing energy e and th e equilibrium internuclear sep aratio n r e. This potential has been
used, w ith pseudo- diatom ic m odel for complexes to estim ate th e binding energy
e, given th e derived force constant k 3 and in terpreting ro as the m easured value
for r cm. F irst an ‘equilibrium ’ ro tatio n al constant is obtained from the m easured
ro tatio n al constant for the complex B 0 using
and
2 tvu 3 = u>3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
119
r e is then th e equilibrium value for R Cm o b tained from B e using th e pseudo-diatom ic
m odel given earlier. T he well d epth is obtained from expanding th e Lennard- Jones
p o ten tial in a Taylor series ab o u t r e and identifying th e coefficient of the (r — r e)2
te rm w ith th e calculated force constant to o b tain an expression for th e estim ated
dissociation energy e
e = k sr 2j n
(43)
T h e values o b tain ed are listed in Table V I.6 . O u r present results for k 3 and e are
com pared w ith results for oth er linear OC-HX complexes in Table V I.6. We note
th e expected tren d of increasing binding energy an d force constant, and decreasing
r e as th e stren g th of th e Lewis acid HX is increased.
Table V I.6. C om parison of k s , r e and e for various hydrogen bound carbon
monoxide van der W aals molecules. r e is th e equilibrium centers of mass
separation.
com plex
H CC H -C O
IH-CO
BrH -CO
C1H-CO
FH -C O
fcs(m dyne
A-1)
0.0171
0.0171
0.0330
0.0446
0.108
r e(A)
5.0011
4.8925
4.5153
4.2260
3.047
e(cm“ l)
299
286
469
569
987
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
V I.4 D ip o le M om ent
S tark p lates were installed in th e P B -F T S ap p aratu s and details are given
in C h ap ter II. T he dipole m easurem ent was m ade by tracking th e J = 3 , m = 3 —►
J = 2 , m = 2 S tark tran sitio n as a function of applied electric field,
(44)
w here ftcomplsi is th e dipole m om ent of the complex, e is th e electric field across th e
S tark plates, and B compUx is th e ro tatio n al constant for th e complex. Table VI.7
an d F igure V I.2 show th e dependance of A u on e. T he Slope of th e plot in Figure
VI.2 yields a dipole m om ent fJ.compUx = 0.333(2) D.
Table V I.7 O bserved S tark shifts for HCCH-CO J = 3,rrij = 3 —> J =
2, m ; = 2 transition.
e(V /cm )
i/(MHz)
Ai/(M Hz)
0
8383.6526
178.737
8383.6608
0.0082
219.256
8383.6629
0.0103
252.881
8383.6662
0.0136
282.833
8383.6672
0.0146
0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
121
0 .0 1 8
0.016
0.014
0.012
0.010
0.008
0.006
0.004
0.002
0 .0 0 0
-
0.002
0
16 000
32 000
48 000
64 000
80 000
(Volt/cm )2
Figure VI.2 P lo t of A v vs. e 2 for th e HCCH-CO complex.
V I .5. Sum m ary
T he lack of any observable signals for th e D CCH -CO isom er in argon carrier
gas was quite unexpected, since signals for th e oth er isomers were easily observed
w ith a single beam pulse. It is likely th a t m ost of the D CCH -CO isomers form ed
were isomerized to th e lower energy H CCD-CO isom er in th e presence of th e argon.
This would require a low b arrier for ro tatio n of th e acetylene molecule w ithin th e
complex. We m ay speculate th a t since argon forms stronger com plexes w ith o th er
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
molecules th a n neon does, three-body collisions involving argon would facilitate th e
isom erization m ore readily th a n neon would. It would be helpful an d interesting to
have results of ab initio calculations on this complex w ith energies for various values
for 0, th e angle of ro tatio n of acetylene relative to r cm an d to com pare a calculated
dipole m om ent for this complex.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
C H A P T E R VII:
SUMMARY AND CONCLUSIONS
P ulsed beam Fourier transform microwave spectroscopy was used to m ea­
sure the ground sta te ro tatio n al, distortion and hyperfine constants for N0C1, CIF3
and the tran sitio n m etal complexes C o (C 0 )3 N 0 , CpC o(C O )2 , and C pM n(C 0)3
.
G round sta te param eters for th e iron containing complexes C bFe(C 0 )3 , C-
hexFe(C 0 )3 , and B uFe(C 0 )3 were also m easured using this technique as was th e
weakly bound H CCH-CO com plex. T he versatility of this technique has been clearly
dem o n strated by th e varied list of molecules studied here. T he experim ents p re­
sented here have been successful in obtaining new and im p o rtan t stru c tu ra l an d
nuclear hyperfine inform ation in th e gas phase for these com pounds as well as estab ­
lishing it as a viable technique in which to study complex molecules like tran sitio n
m etal complexes.
T he sp ectra obtained for NO Cl and CIF3 provide th e first high resolution
d a ta set for tran sitio n s involving low J rotatio n al states. More precise m easurem ents
for the nitrogen quadrupole coupling in N0C1 were obtained and com parisons to
other ‘sim ple’ nitrosyl com pounds were m ade using th e Townes-Dailey in te rp re ta ­
tion of nuclear quadrupole coupling d ata. Q uadrupole coupling d a ta for Cl in C1F3
were also in terp reted using th e Townes-Dailey model and m ore precise gas phase
stru c tu ra l inform ation was also obtained from the two isotopic sp ectra of 35 Cl and
37C1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
High resolution gas phase sp ectra were obtained for complexes involving
cobalt and m anganese.
New d a ta relatin g to th e stru ctu re of these com pounds
a n d th e first gas phase m easurem ents of cobalt an d m anganese nuclear quadrupole
coupling in these complexes were o b tain ed from the d ata. C pC o(C 0)2 was found
to be a hindered internal ro to r w ith a m oderately low b arrier to internal ro tatio n
of 0.82(20) THz (0.3 k J/m o l) estim ated from th e data. T he com plication of th e
large quadrupole m om ent for Co an d hindered ro tatio n precluded a m ore precise
determ in atio n of th is param eter. T h e possible hindered ro to r n atu re of C pM n(C O )3
can n o t be determ ined from this d a ta set since hindered ro tatio n effects are not seen
for sym m etric to p spectra.
T h e iron containing com plexes provided new stru ctu ral inform ation on the
interactio n of conjugated diene system s w ith th e tran sitio n m etal iron. Cyclobutadiene was found to be square w hen com plexed to th e Fe(C O )3 as opposed to
rectan g u lar as once p o stu lated . C yclohexadiene an d butadiene subunits were found
to be bonded to F e(C O )3 in a pi di — si g m a type fashion rath e r th a n in th e con­
ju g a ted sense of a p lan ar b u tad ien e su b unit. T here were no significant observable
hindered ro tatio n effects in th e sp ectra for these complexes indicating a fairly high
b arrier to in tern al ro tatio n of th e respective conjugated diene subunit relative to
th e tricarbonyl group.
G round sta te vibrationally averaged stru ctu ra l param eters for th e HCCHCO weakly bou n d complex were determ ined from th e microwave data. T h e stru c­
tu re was found to be averaged over a linear equilibrium geom etry w ith one of the
hydrogens of th e HCCH bo u n d to th e carbon of th e CO subunit. Force constants
an d stretch in g frequencies were o b tain ed from th e analysis of the d istortion constant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
D j . A binding energy of 300 cm -1 was estim ated from this d a ta as well. In tere st­
ing interm olecular dynam ics were observed for this complex when th e d eu terated
isotopom ers were studied. A n ap parent facilitation of th e interconversion betw een
th e hydrogen and deuterium bound species of th e complex was found to occur m ore
readily in an argon carrier gas th a n in a neon carrier gas. A dipole m om ent of
0.333 D was also obtained from th e stark experim ent and is com parable to a value
obtained sim ultaneously by th e G utow sky group [101] of 0.3112 D.
V II. 1 F u ture D irection s
As was stated in C h ap ter II th e only real lim itation of th e P B -F T S tech­
nique is th e ability to deliver a sam ple th ro u g h th e pulsed valve in tact into th e
F abry-P ero t cavity. Advances such as th e hot source an d th e incorporation of th e
S tark p lates have dem onstrated th e feasabilty of ad ap tin g the P B -F T S technique
to new experim ental challenges. T here still rem ains m uch work to be done w ith th e
tran sitio n m etal complexes and presented here are th e foundations in which fu tu re
studies m ay be based. T he sam ple handling techniques learned here will provide th e
investigator w ith valuable inform ation and tools in which to successfully m easure
sp ectra of th e next transition m etal challenge, hydrides. Transition m etal hydride
com plexes are very unstable and reactive species an d will require novel sam ple h a n ­
dling techniques and the inevitable design or radical m odification of existing pulsed
nozzle sources to accom m odate th eir u nstable n atu re. Some prelim inary testing of
new nozzle designs m ade from Teflon m aterial have already been m ade w ith some
success. A n a tte m p t to m easure a H M n(C O )s ro tatio n al tran sitio n was m ade al­
ready, b u t th e sam ple life in th e current spectrom eter configuration was too short
to have m ade an effective search for such a transition.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
O th er high tem p eratu re species may be m easured using P B -F T S by m od­
ifying th e new designs of todays fuel injectors used in th e autom obile industry.
These fuel injectors can w ith stan d very high operating tem p eratu re and could be
used in th e stu d y of high tem p eratu re species an d m ay even be used in forming
radicals in high enough concentrations. Since the exit nozzle of these fuel injectors
have four ‘pin-holes’ the am ount of m ass flux p er pulse m ay be high enough for
the subsequent generation of radicals while retaining the properties of supersonic
expansions. O ne of these fuel injectors has been m odified in our lab already and th e
signal arising from the Ar-HCl complex has been detected w ith greater th a n 20 to
1 signal to noise ratios thus showing th e retention of free je t expansion properties.
Incorporation of th e fuel injector m ay significantly im prove experim ents requiring
significant populations of these h a rd to get species like radicals.
T hrough th e course of these experim ents, unique situations an d questions
have arisen concerning the experim ental d ata. M ore often these questions can be
queried by eith er ab initio or sem iem pirical studies. T he knowledge obtained from
experim ental studies can be used to form ulate new ideas for theoretical studies and
also to guide th e m odification of existing theories in th eir continual development.
An ab initio stu d y [102] of the H CCH-CO complex is one exam ple of the interaction
of theory and experim ent. T he fu tu re will see m ore and m ore of such interactions
of theory and experim ent and perhaps the m odern day experim ental spectroscopist
will also have to be a com petent theoretician as well.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
A P P E N D IX A .I:
C1F3 TR A N SIT IO N FR E Q U E N C IE S
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
Table A.I.l Measured and calculated (least-squares fit) transition fre­
quencies for 35C1F3 . Frequencies in MHz. The standard deviation
for the fit is 17kHz. Units in MHz.
h < p,K0
2F
1i,i
5
2o,2
5
6859.947(8)
-0.010
ii.i
1
2o,2
3
6863.782(7)
0.022
3
2o,2
5
6876.331(9)
0.021
5
2o,2
7
6883.525(8)
0.003
1
2o,2
1
6887.322(8)
0.017
ii.i
3
2o,2
3
6893.172(10)
0.011
1i,i
3
2o,2
1
6916.707(4)
0.002
lo,i
1
1i ,0
1
10242.282(3)
-0.002
lo,i
3
li,0
1
10279.189(4)
-0.011
lo,i
5
11,0
5
10288.183(1)
0.028
1o,i
3
li,0
5
10308.665(50)
0.006
lo,i
5
ll.o
3
10325.016(4)
0.003
10,1
3
li,0
3
10345.530(5)
0.013
2o,2
1
2i,i
1
11527.025(7)
0.014
2 o,2
1
2i,i
3
11543.343(5)
0.011
2 o,2
3
2i,i
1
11550.550(6)
-0.006
2 o,2
7
2i,i
7
11555.460(2)
0.024
2o,2
3
2i,i
3
11566.884(3)
0.007
2o,2
7
2i,i
5
11571.814(3)
0.004
2 o,2
3
2i,i
5
11578.548(3)
0.024
2 o,2
5
2i,i
3
11583.727(3)
-0.001
2 q,2
5
2i,i
5
11595.389(3)
0.014
ii.i
2F'
Meas.
M-Calc
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
129
Table A.1.1 Continued.
J K P, K 0
2F
2F'
3 o,3
3
3 i ,2
3
13623.775(4)
-0.008
3 o,3
3
3 i ,2
5
13633.956(4)
0.007
3 o,3
9
3 i ,2
9
13641.558(6)
0.011
3 o,3
9
3a ,2
7
13651.754(6)
0.001
3 o,3
5
3l,2
5
13661.933(3)
0.002
3 o,3
5
3 i ,2
7
13666.701(5)
0.023
3 o,3
7
3 i ,2
9
13669.568(15)
0.022
3 o,3
7
3 i ,2
5
13675.000(3)
0.007
3 o,3
7
3 i ,2
7
13679.760(8)
0.006
2 i ,2
1
3 o,3
3
15677.957(4)
0.010
2 i ,2
3
3 o,3
5
15686.858(5)
0.011
2 i ,2
7
3 o,3
9
15691.215(3)
-0.016
2 i ,2
5
3 o,3
7
15700.060(3)
0.007
2 i ,2
5
3 o,3
5
15713.128(3)
-0.015
4(1,4
5
4 i ,3
5
16719.820(9)
-0.016
4 o,4
7
4 i ,3
7
16759.643(4)
-0.017
4 o,4
9
4 i ,3
7
16771.354(4)
-0.029
0o,o
3
ll .l
3
17183.757(3)
-0.031
Oo,o
3
ll ,l
5
17200.136(3)
-0.005
Oo,o
3
ll .l
1
17213.147(5)
-0.042
Meas.
M -Calc.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
Table A.1.2 Measured and calculated (least-squares fit) transition fre­
quencies for 37C1F3 . Frequencies in MHz. The standard deviation
for the fit is 17kHz. Units in MHz.
J/Cp,/C0
2F
J'
2F'
Meas.
M-Calc.
1
jl
5
2 o ,2
5
6945.428(7)
-0.007
*
>-1
3
2 o ,2
5
6958.340(8)
0.012
5
2 o ,2
7
6964.043(7)
-0.001
1
2 o ,2
1
6967.069(7)
0.026
3
2 o ,2
3
6971.633(8)
0.002
lo ,l
5
11 , 0
3
10230.832(5)
0.000
lo ,l
3
li,0
3
10247.000(5)
0.006
2 o ,2
1
2i,i
1
11454.733(7)
0.005
2 o ,2
1
2i ,1
3
11467.618(5)
0.018
2 o ,2
3
2i,i
1
11473.303(6)
-0.021
2 o ,2
7
2i,i
7
11477.198(3)
0.024
2 o ,2
3
2i,i
3
11486.210(3)
0.014
2(1,2
7
2i,i
5
11490.081(6)
0.002
2 o ,2
3
2i,i
5
11495.406(2)
0.021
2 o ,2
5
2i,i
7
11495.817(10)
0.035
2 o ,2
5
2i,i
3
11499.501(3)
0.001
2 o ,2
5
2i,i
5
11508.696(5)
0.008
3 o,3
3
3l,2
3
13565.239(5)
0.001
3 o,3
3
3 i ,2
5
13573.264(4)
0.017
3 o,3
9
3 i ,2
9
13579.298(4)
0.015
3 o,3
5
3 i ,2
3
13587.347(24) -0.030
3 o,3
5
3 i ,2
5
13595.385(2)
0.001
3 o,3
5
3 i ,2
7
13599.141(3)
0.016
1 ,1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
Table A.1.2 Continued.
2F
2F'
Meas.
M-Calc.
3 o,3
7
3l,2
9
13601.456(3)
0.023
3 o,3
7
3 i ,2
5
13605.721(6)
-0.008
3 o,3
7
3 i ,2
7
13609.475(4)
0.007
2 i ,2
3
3 o,3
5
15761.498(5)
0.005
2 i ,2
7
3 o,3
9
15764.987(5)
-0.006
2 i ,2
5
3 o,3
7
15771.888(4)
0.003
4, 4
5
4 i ,3
5
16686.710(9)
0.006
4 o,4
5
4 i ,3
7
16691.772(16)
0.003
4 o,4
7
4 i ,3
7
16718.198(3)
0.012
4 o,4
9
4 i ,3
7
16727.466(9)
0.027
4 o,4
9
4 i ,3
9
16729.263(8)
0.005
Oo,o
3
ll.l
3
17085.911(3)
0.024
0o,o
3
ll.l
5
17098.825(10)
0.003
Oo,o
3
l l .l
1
17109.079(2)
0.041
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
A P P E N D IX A .II:
TR A N SITIO N FR EQ U E N C IE S FO R
COBALT TR I-C A R B O N Y L N ITRO SY L, CY CLO PEN TA D IEN Y L
COBALT DI-CARBONYL, AND C Y C LO PEN TA D IEN Y L
M ANGANESE TR I-C A R B O N Y L
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
133
Table A.II.l Measured and calculated hyperfine transition frequen­
cies for the J == 2—>3, K=0 transition for Co(C0)3N0 in MHz. The
standard deviation for the fit was 11kHz. Units in MHz.
21
2F
21'
2F'
Meas.
M-Calc.
9
9
7
7
9
7
9
5
7
9
5
9
9
7
9
5
9
9
7
9
9
9
7
11
11
13
3
7
11
7
9
5
5
7
3
5
13
11
9
9
7
5
9
7
9
5
5
9
7
9
9
7
9
5
9
9
7
5
9
11
9
13
13
15
3
7
11
9
11
5
7
9
5
7
13
11
6249.592
6251.695
6251.851
6251.993
6252.112
6252.518
6252.570
6252.629
6252.867
6252.967
6253.500
6253.649
6254.039
6255.387
6255.529
6255.651
6257.027
6257.891
6258.073
-0.005
0.002
0.003
-0.020
-0.001
0.004
0.007
-0.005
0.005
0.004
-0.002
0.010
0.002
0.002
0.003
-0.006
0.024
-0.023
-0.004
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
Table A.II.2 Measured and calculated hyperfine transition frequencies
for the J = 3—>4, K=0 transition for 59 Co(CO)sNO in MHz. The
standard deviation for the fit was 9kHz. Units in MHz.
21
2F
21'
2F'
Meas.
M-Calc.
9
9
9
9
9
5
7
7
5
7
9
7
5
5
9
7
11
13
5
1
9
9
7
7
9
11
3
5
9
9
9
9
7
9
7
5
7
7
9
7
5
5
9
7
11
15
5
3
11
9
9
9
11
11
5
5
8335.891
8336.174
8336.292
8336.612
8337.706
8337.008
8337.106
8337.265
8337.326
8337.936
8338.183
8338.242
8338.962
8339.145
-0.003
-0.002
0.001
0.001
0.003
0.008
-0.005
0.003
0.004
-0.003
-0.022
0.009
-0.005
0.011
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
135
Table A.II.3 Measured and calculated transition frequencies for the J
= 4—>5 transitions for 59 Co(C0)3N0 in MHz. The standard devia­
tion for the fit was 0.008 MHz.
21
2F
9
9
7
9
5
7
9
9
9
5
7
9
7
9
7
9
7
13
15
11
19
17
13
11
7
7
9
5
7
13
21'
2F'
Meas.
M-Calc.
7
7
7
9
5
7
9
9
9
5
9
9
5
7
7
9
7
15
17
13
17
19
15
13
9
9
11
7
9
13
10419.863
10420.066
10421.088
10421.146
10421.216
10421.322
10421.356
10421.421
10421.804
10421.935
10422.202
10422.353
10444.448
10422.574
10423.035
0.003
-0.003
0.010
0.004
0.004
-0.009
0.004
-0.009
-0.005
-0.011
0.011
-0.002
-0.001
0.005
0.001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136
Table A.II.4 Measured and calculated hyperfine transition frequencies
for the J = 5—>6 transitions for 59 Co(C0)3N0 in MHz. The standard
deviation for the fit was 14 kHz.
21
2F
21'
2F'
Meas.
M -Calc.
9
9
7
9
5
9
9
7
15
17
17
19
5
9
5
9
9
9
7
9
5
9
9
7
15
19
19
21
7
11
7
11
12505.211
12505.493
12505.653
12505.653
12506.051
12506.140
12506.238
12506.434
0.001
-0.006
0.012
-0.002
-0.002
-0.022
0.012
0.006
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
Table A.II.5 Measured and calculated transition frequencies for the J
= 3—>-4 transitions for 59 Co(CO)3 NO in MHz. The standard devia­
tion for the fit was 0.01 MHz.
21
2F
21'
2F'
Meas.
M-Calc.
9
9
7
9
5
9
7
9
7
7
5
7
5
3
5
13
15
11
7
5
9
7
11
9
9
7
7
9
7
9
5
9
7
9
7
7
5
8
5
5
7
15
17
13
9
7
9
7
13
11
11
7
8330.883
8332.637
8334.378
8334.778
8334.819
8335.397
8335.683
8339.038
8339.232
8339.556
8339.742
8340.399
8340.613
-0.005
0.001
-0.006
0.005
-0.005
0.006
0.009
-0.001
-0.003
0.013
-0.014
0.011
-0.013
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
Table A .II.6 M easured an d calculated tran sitio n frequencies for the
hindered ro to r spectrum of C pC o(C O )2 . T he m easured frequen­
cies are averaged over m any hyperfine com ponents caused by the
quadrupole coupling interaction of th e cobalt atom and also possible
m -state splitting. S tan d ard deviation for the fit =10 kHz. Frequencies
in M H z .
m'
J'
I<'
m
J
K
M easured
0
1
2
0
1
0
1
1
0
1
1
0
2
0
0
3
3
3
4
4
4
4
4
5
5
5
6
6
7
8
0
1
2
0
1
2
3
-1
0
1
3
0
0
0
0
0
1
2
0
1
0
1
1
0
1
1
0
2
0
0
2
2
2
3
3
3
3
3
4
4
4
5
5
6
7
0
1
2
0
1
2
3
-1
0
1
3
0
0
0
0
6064.1
6087.1
6274.3
7806.8
7819.0
8478.0
8825.8
8841.8
9527.1
9527.1
1079Q3
112423
133967
129783
147381
Calc.
(M-C)
6053.2
6083.6
6268.5
7808.5
7816.5
8458.1
8822.3
8839.2
9524.8
9525.2
108040
112493
134056
12987.3
147335
10.9
3.5
5.9
1.7
2.5
19.9
3.5
2.6
2.3
2.0
13.7
7.1
8.9
9.0
4.8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
Table A .II.7 M easured an d C alculated (Calc.)
frequencies for
quadrupole hyperfine stru ctu re on K = 0, m = 0 for J —► J ' =
2 —> 3, 3 —* 4, 4 —> 5 transitions for C pC o(C O )2 . eQgaa = 12(4)
MHz and eQqbb= 113(4) MHz. S tan d ard deviation for the fit = 0.07
MHz. Frequencies in MHz.
J
2F
J'
2F '
M easured
Calc.
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
5
7
11
5
7
9
7
11
13
7
9
11
9
7
11
13
9
15
13
11
11
9
7
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
3
5
13
5
7
11
9
11
15
7
9
13
11
9
11
13
9
17
15
11
13
11
9
6053.13
6056.38
6060.42
6063.14
6066.39
6068.31
6072.42
6078.96
7808.43
7808.85
7812.98
7813.46
7817.15
7819.01
7820.21
7832.00
9522.70
9525.09
9527.75
9528.24
9530.58
9532.41
9532.86
6053.15
6056.40
6060.40
6053.19
6066.39
6068.28
6072.36
6078.97
7808.32
7808.85
7813.02
7813.45
7817.15
7819.00
7820.28
7832.02
9522.83
9525.06
9527.78
9528.20
9530.44
9532.33
9532.98
(M-C)
-0.02
-0.02
0.02
-0.05
0.00
0.03
0.05
-0.01
0.11
0.00
-0.04
0.01
0.00
0.01
-0.07
-0.02
-0.13
-0.03
-0.03
0.04
0.13
0.08
-0.12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
Table A.II.8 Hyperfine com ponents for th e J = 2 —> 3 rotatio n al tra n ­
sition and their corresponding assignm ents C pM n(C O )3 . Observed
transitio n s are F —» F ' and listed frequencies are in M H z units. The
o f i t — 10 k H z
I<
F
F'
Meas.
Calc.
2
1
1
0
0
1
0
1
2
2
9
3
9
7
9
7
5
5
5
7
U
3
11
9
11
9
7
7
7
9
4963.357
4964.982
4966.396
4966.749
4967.405
4969.160
4970.125
4970.866
4973.049
4976.460
4963.352
4964.991
4966.384
4966.741
4967.395
4969.170
4970.136
4970.863
4973.045
4976.455
M-C
0.005
-0.009
0.012
0.008
0.010
-0.010
-0.011
0.003
0.004
0.005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
Table A.II.9 Hyperfine components for the J — 3 —> 4 rotational
transition and their corresponding assignments for CpMn(C0)3 . Fre­
quencies in M H z .
K
F
F'
Meas.
Calc.
M-C
0
3
2
1
2
1
0
0
1
0
2
1
1
1
2
2
3
3
0
5
11
1
5
11
11
9
11
9
7
7
7
3
5
9
7
7
9
11
5
13
3
5
13
13
11
13
11
9
7
9
5
7
11
9
9
11
11
6618.920
6619.333
6619.370
6619.760
6621.787
6623.259
6623.339
6623.753
6624.141
6624.778
6625.241
6625.380
6625.715
6626.075
6626.565
6627.163
6630.174
6630.705
6635.767
6618.911
6619.347
6619.369
6619.753
6621.789
6623.254
6623.325
6623.742
6624.144
6624.776
6625.250
6625.377
6625.716
6626.077
6626.602
6627.178
6630.179
6630.699
6635.762
0.009
-0.014
0.001
0.007
-0.002
0.005
0.014
0.011
-0.003
0.002
-0.009
0.003
-0.001
-0.002
-0.037
-0.015
-0.005
0.006
0.005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
142
Table A.II. 10 Hyperfine components for the J — 4 —> 5 rotational
transition and their corresponding assignments for CpMn(CO) 3 . Fre­
quencies in M H z .
K
F
F'
Meas.
Calc.
M-C
3
2
4
0
2
3
3
3
2
1
0
1
1
0
1
1
0
2
3
2
0
1
3
3
3
4
2
4
3
5
13
9
7
13
5
13
13
13
13
11
3
9
9
5
7
7
7
9
11
11
11
9
9
11
13
9
5
5
15
9
7
15
7
13
15
15
15
13
5
11
11
7
9
9
9
11
11
11
13
9
11
13
13
11
8274.890
8275.394
8275.585
8276.615
8277.205
8277.486
8278.239
8278.643
8278.842
8279.682
8279.960
8280.018
8280.401
8280.465
8280.811
8281.351
8281.392
8281.443
8281.547
8281.839
8281.906
8282.027
8282.864
8283.186
8283.604
8285.398
8285.872
8286.063
8274.889
8275.406
8275.585
8276.593
8277.200
8277.493
8278.238
8278.641
8278.856
8279.673
8279.946
8280.013
8280.396
8280.459
8280.809
8281.349
8281.387
8281.458
8281.548
8281.859
8281.886
8282.019
8282.880
8283.203
8283.608
8285.388
8285.882
8286.057
0.001
-0.012
0.000
0.022
0.005
-0.007
0.001
0.002
-0.015
0.009
0.014
0.005
0.005
0.006
0.002
0.002
0.005
-0.015
0.001
-0.020
0.020
0.008
-0.016
-0.017
-0.004
0.010
-0.010
0.005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
Table A.II. 11 Hyperfine components for the J = 5 —> 6 rotational
transition and their corresponding assignments for CpMn(CO) 3 . Fre­
quencies in M H z .
IC
F
F'
Meas.
Calc.
M-C
1
2
5
2
4
3
3
4
3
2
3
1
1
0
0
1
1
0
1
1
3
4
5
2
1
4
3
4
4
5
5
9
7
15
9
15
5
15
7
9
5
7
15
13
15
11
5
11
9
9
7
13
9
9
13
13
1
11
11
11
13
11
9
7
17
9
17
7
17
9
9
7
9
17
15
17
13
7
13
11
11
9
15
11
11
13
13
15
13
13
11
15
13
9930.711
9930.584
9931.890
9932.310
9933.405
9933.574
9934.567
9934.652
9935.042
9935.386
9935.770
9935.913
9936.048
9936.074
9936.371
9936.471
9936.561
9937.000
9937.043
9937.078
9937.497
9937.914
9938.419
9938.543
9938.597
9938.801
9938.268
9939.762
9940.185
9940.459
9941.656
9930.699
9930.605
9931.893
9932.330
9933.398
9933.579
9934.570
9934.630
9935.049
9935.382
9935.773
9935.908
9936.044
9936.076
9936.370
9936.464
9936.582
9936.986
9937.044
9937.079
9937.514
9937.902
9938.418
9938.545
9938.587
9938.799
9938.273
9939.753
9940.190
9940.452
9941.655
0.012
-0.021
-0.003
-0.020
0.007
-0.005
-0.003
0.022
-0.007
0.004
-0.003
0.005
0.004
-0.002
0.001
0.007
-0.021
0.014
-0.001
-0.001
-0.017
0.012
0.001
-0.002
0.010
0.002
-0.005
0.009
-0.005
0.007
0.001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
144
Table A.II.12 Hyperfine components for the J = 6 —> 7 rotational
transition and their corresponding assignments for CpMn(CO) 3 . Fre­
quencies in M H z .
K
F
F'
Meas.
Calc.
M-C
1
1
0
1
1
1
3
4
17
15
13
13
11
9
11
15
19
17
15
15
13
11
13
17
11592056
11592102
11592336
11592478
11592839
11592899
11593260
11593682
11592049
11592098
11592344
11592480
11592838
11592900
11593270
11593672
0.007
0.004
-0.008
-0.002
0.001
-0.001
-0.010
0.010
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
145
A P P E N D IX A .III:
TR A N SIT IO N FREQ U EN C IES FO R
CY C LO B U TA D IEN E IRON TR I-CA R BO N Y L, C Y C LO H EX A D IEN E
IRO N TR I-C A R B O N Y L, AND BU TA DIENE IRON TR I-C A R B O N Y L
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14 6
Table A.III.l Measured and calculated J + 1 <— J transition frequencies for
cyclobutadiene iron tricarbonyl. Units in MHz.
J
2
2
2
3
3
3
3
4
4
4
5
5
5
5
6
6
7
7
7
7
K
0
1
2
0
1
2
3
0
1
2
0
1
2
3
0
4
0
2
3
4
Meas.
5771.893
5771.893
5771.865
7695.843
7695.843
7695.793
7695.759
9619.761
9619.761
9619.717
11543.668
11543.668
11543.601
11543.547
13467.545
13467.272
15391.408
15391.303
15391.224
15391.093
M-C
-0.004
0.004
0.002
0.002
0.012
-0.009
0.005
-0.006
0.006
0.002
-0.004
0.013
-0.013
0.006
-0.004
-0.006
0.012
-0.015
0.003
-0.003
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
Table A.III.2 Measured transition frequencies and least squares residuals
(M.-C.) for C-hex56Fe(CO)3. Units in MHz.
J k p ,k 0
J ' k ^K' o
3 i ,3
4l,4
3 o,3
4 0,4
32,2
42,3
3 3, i
4 3,2
32,1
42,2
3l,2
4l,3
3 o,3
4 i ,3
3 i ,2
42,2
4 i ,3
5 o,5
3 i ,3
42,3
4 i ,4
5 i ,5
4 o,4
5 o,5
42,3
52,4
4 i ,3
5 i ,4
42,2
52,3
43,1
5 3,2
4 3,2
53>3
44,0
54,1
32,1
4 3,i
32,2
4 o,4
43,2
5 i ,4
4 i ,3
52,3
5 i ,5
6 1 ,6
5 o,5
6 0 ,6
52,4
6 2 ,5
55,1
6 5 ,2
5 4,2
6 4 ,3
54,1
6 4 ,2
Meas.
5317.020
5351.691
5363.540
5367.055
5376.413
5406.840
5769.572
6184.592
6262.795
6300.294
6644.117
6680.680
6702.779
6755.745
6727.786
6710.729
6709.728
6708.620
6808.935
6815.138
7173.634
7505.538
7970.010
8005.089
8040.927
8050.118
8051.282
8051.324
(M.-C.)
-0.0004
0.0012
0.0008
0.0025
-0.0003
-0.0005
-0.0007
0.0017
-0.0012
0.0004
0.0005
0.0005
0.0030
-0.0012
-0.0008
-0.0008
-0.0005
-0.0082
-0.0011
-0.0014
0.0036
0.0011
-0.0014
0.0001
0.0011
-0.0011
0.0041
-0.0058
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
148
Table A.III.2 Continued.
j k v,i<a
J I K'P,K'0
Meas.
(M.-C.)
53,3
63,4
8052.777
-0.0003
53,2
63,3
8055.422
-0.0008
^2,3
62,4
8082.621
0.0019
5 i ,4
61,5
8102.518
0.0013
42,3
53,3
8161.329
0.0003
42,2
53,2
8143.252
0.0004
5 o,5
61,5
8595.464
-0.0031
62,4
7l,6
8716.581
0.0009
5 i ,4
62,4
8832.409
0.0001
6 1 ,6
7 i ,7
9294.673
0.0014
6 0 ,6
7 o,7
9325.627
0.0002
62,5
72,6
9377.789
0.0011
66 ,1
7e,2
9391.637
-0.0033
64,2
74,3
9394.598
0.0007
63,4
73,5
9396.065
0.0025
73,4
9401.932
0.0006
62,4
72,5
9439.950
-0.0019
61,5
7l,6
9446.473
0.0007
62,3
63,3
9470.890
0.0021
52,4
63,4
9511.328
-0.0020
7 l,7
81,8
10618.143
-0.0002
7 o,7
80,8
10643.546
-0.0004
72,6
82,7
10713.174
-0.0111
76,1
85,2
10734.039
0.0023
75,3
85,4
10735.506
0.0029
74,4
84,5
10738.099
-0.0022
74,3
84,4
10738.574
-0.0002
63,3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
149
Table A.III.2 Continued.
(M.-C.)
co
00
10739.374
-0.0012
83,5
10750.856
0.0084
7 l ,6
8 1 ,7
10786.850
-0.0006
7 2 ,5
82,6
10798.255
0.0014
8l ,8
9 l ,9
11940.531
-0.0026
00
Meas.
9 o,9
11960.158
- 0.0001
82,7
9 2 ,8
12046.967
0.0001
8 6 ,2
9 e ,3
12076.722
-0.0049
85,3
9 5 ,4
12078.852
0.0122
CO
J ' k ' p,K'0
ot
J k p ,k b
9 4 ,6
12082.322
-0.0016
83,6
9 3 ,7
12082.445
0.0020
81,7
9 l ,8
12122.886
0.0012
9i,g
IOi .io
13261.989
-0.0011
9 o,9
1 0 o ,io
13276.439
0.0022
9 2 ,8
1 0 2 ,9
13379.031
0.0027
93,7
103,8
13424.956
-0.0009
94,6
104,7
13427.064
-0.0021
94,5
104,6
13429.468
-0.0010
9 i ,8
10l ,9
13453.972
0.0006
93,6
103,7
13458.104
-0.0032
9 2 ,7
1 0 2 ,8
13511.508
-0.0010
00
7 3 ,4
0
7 3 ,5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
150
Table A.III.3 Measured transition frequencies and least squares residuals
(M.-C.) for C-hex54Fe(CO)3. Units in MHz.
J k p,i<0
J ' k ' p,K'0
Meas.
(M.-C.)
4m
5 i ,5
6645.883
-0.0004
4 o,4
5 o,5
6682.449
0.0002
42,3
5 2 ,4
6704.589
0.0004
42,2
52,3
6729.647
0.0007
4 l,3
5 l,4
6757.591
-0.0001
5 i ,5
61,6
7972.126
0.0005
5 o,5
60,6
8007.193
-0.0004
52,4
62,5
8043.095
-0.0004
52,3
62,4
8084.863
-0.0005
5 i ,4
61,5
8104.720
-0.0003
6 i ,6
7 i ,7
9297.133
0.0000
6 o,6
7 o,7
9328.065
0.0000
62,5
72,6
9380.314
-0.0001
61,5
7 l,6
9449.029
0.0005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
151
R eferen ces
1. H.A. Fry an d S.G. Kukolich, J. Chem. P hys. 76, 4387 (1982).
2. T .J. Balle an d W .H. Flygare, Rev. Sci. In stru m . 52, 33 (1981).
3. F .J. M cQuillin, D.G. P arker, an d G .R . Stepheson, T. M etal Organometallics
fo r Organic Synthesis, U niversity Press, C am bridge 1991.
4. G .W . P arshall, Homogeneous Catalysis, Jo h n W iley an d Sons, New York, 1980.
5. D ynam ics o f Polyatom ic Van der Waals Complexes, Eds. N. H alb erstad t and
K .C . Ja n d a , NATO ASI series. Series B: Physics v. 227.
6. Structure and D ynam ics o f Weakly B ound M olecular Complexes, Ed. A. We­
ber, NATO ASI series. Series C: M ath em atical an d Physical Sciences v.
212 .
7.
R. B um garner, Microwave Spectroscopy o f Weakly B ound Complexes and High
R esolution Infrared Studies o f the v§ and
Bands o f Formic Acid, Ph.D
D issertation, U niversity of Arizona, 1988, Chp. II.
8.
J.B . A nderson, Molecular Beams and Low D ensity Gas D ynam ics, Ed. P.P.
W egenerg (M arcel Dekker, New York, 1974).
9. H. P au ly an d J.P . Toennies, Methods o f Experim ental Physics, Eds. B. Bederson an d W .L. F ite (A cadem ic, New York, 1968), Vol. 7, p t. A.
10. J.B . A nderson, R.P. A ndres, and J.B . F inn, Advances in Chemical Physics,
Ed. I. Prigogine (Wiley, New York, 1974), Vol. 25.
11.
K. Bier an d 0 . H agena, Rarified Gas D ynam ics, E d. J.H . deLeur (Academic,
New York, 1966), Vol. II.
12.
H. A shkenas and F.S. Sherm an, Rarified Gas D ynam ics, Ed. J.H . deLeur
(A cadem ic, New York, 1966), Vol. II.
13.
O .F. H agena and W. O b ert, J. Chem. P h ys.56, 1793 (1972).
14.
D. Golomb, R.E. Good, A.B. Bailey, M .R. Busby, and R. D aw born, J. Chem.
P h ys.57, 3844 (1970).
15.
D. Golomb, and R .F. Brown, J. Chem. P h ys.52, 1545 (1970).
16.
.M. Calo, J. Chem. Phys. 62, 4904 (1975).
17.
T.A . M ilen, A.E. V andegrift, and F .T . G reen, J. Chem.
(1970).
P hys. 52, 1552
18.
A.C. Legon, A nn. Rev. Phys. Chem. 3 4 , 275 (1983).
19.
E .J. C am pbell, L.W . B uxton, T .J . B alle, and W .H . Flygare, J. Chem.
P hys.74, 829 (1981).
20.
W . G ordy an d R.L. Cook, “Microwave M olecular S p ectra” , John W iley and
Sons, New York (1984),Ch. 14.
21.
D .J. M illen an d J. Pannell, J. Chem . Soc., 1322 (1961).
22.
G. Cazzoli, R. Cervellati, and A.M. M irri, J. Mol. Spect. 56, 422 (1975); A.
G uarnieri an d P.G. Favero, Nuovo C im ento 3 9 , 76 (1965).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
152
23. R .E. B um garner an d S.G. Kukolich, J. Chem. Phys. 86, 1083 (1987).
24. S.G. Kukolich, J.V . R und, D .J. Pauley an d R.E. B um garner, J. Am. Chem.
Soc. 110, 7356 (1988).
25.
Colem an, Inorganic syntheses 1, 55 (1939).
26.
S.G. Kukolich, Mol. Phys. 29, 249 (1975).
27.
S.G. Kukolich and C.D. Cogley, J. Chem. Phys. 76, 1685 (1982).
28.
G. Cazzoli, R. Cervellati, and A.M. M irri, J. Mol. Sped. 56, 422 (1975); A.
G uarnieri and P.G. Favero, Nuovo. Cim. 39, 76 (1965) .
29.
D .F. Sm ith, J. Chem. Phys. 21, 609 (1953).
30.
R.D. B urbank an d F.N . Bensey, JC P , 21, 602 (1953).
31.
J.H . Holloway, P h.D . thesis, D epartm ent of Physics, M .I.T ., 1956; J.H . Hol­
loway, B.B. Aubrey, and J.G . K ing (unpublished results)..
32.
A.A. Ishchenko, I.N. M yakshin, G.V. Rom anov, V.P. Spiridonov, and V.F.
Sukhoverkhov, Dokl. Acad. Nauk. USSR 26 7 , 1143 (1982)..
33.
D .W . M agnuson, J. Chem. Phys.27,233 (1957)..
34.
R.D. F eltham and J.H . E nem ark, “Topics in Stereo C hem istry” , G.L. Geof­
frey, Ed. Vol 12, 155 (1981).
35.
W .L. M eerts, Chem. Phys 14, 421 (1976).
36.
K.H. C asleton and S.G. Kukolich, J. Chem. Phys. 62, 2696 (1975).
37.
K.S. B uckton, A.C. Legon an d D .J. Millen, Trans. Faraday Soc. 65, 1975
(1969).
38.
D.Coffey, C.O. B ritt and J.E . Boggs, J. Chem. Phys. 49, 591 (1968).
39.
L andolt-B ornstein- N um erical d a ta and functional relationships in science
an d technology (Springer, Berlin, 1974) Vol 6..
40.
41.
S.G. Kukolich and A.C. Nelson, J. A m . Chem. Soc.95, 680 (1973).
J.G . K ing and V. Jaccarino, Phys. Rev. 94, 1610 (1954).
42.
C.D. Cogley, H.A. Fry, an d S.G. Kukolich, J. Mol. Spedrosc.75, 447 (1979).
43.
J. M arch, Advanced Organic C hem istry, M cGraw Hill Book Co., New York
1977.
44.
A.C. Lehninger, B iochem istry, W orth Publishers Inc., New York, 1978.
45.
K. H edberg, L. H edberg, K. Hagen, R.K . R yan and L.H. Jones, Inorg. Chem.
24, 2771 (1989).
46.
A.P. Cox, L.F. T hom as an d J. Sheridan, N ature 181, 1157L (1958).
47.
Values from: W. G ordy and R.L. Cook, Microwave M olecular S pectra, 3rd
E d., Jo h n W iley an d Sons, New York (1984); 1 b a rn = 10-24 cm2..
48.
S.G. Kukolich, J.V . R und, D .J. Pauley and R.E. B um garner, J. A m . Chem.
5 o c .ll0 , 7356 (1988).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
153
49.
E.A .C . Luclcen, N uclear Q uadrupole Coupling C onstants, Ch. 13, pp. 328329, Academ ic Press, London (1969).
50.
F.A . C oton and G. W ilkinson, Advanced Inorganic Chem istry, Ch.27, In ter­
science, New York (1962).
51.
W .L. M eerts, Chem. Phys. 14, 921 (1976) (See discussion in Ref. 6).
52.
B.Beagley, C .T .P arro t, V .U lbrecht and G.G.Young, J. Mol.
(1979) 47 .
53.
S.G.Kukolich, M .A .Roehrig, S.T.H aubrich an d J.C .Shea, J. Chem. P hys. 94,
(1991) 191.
54.
55.
E.B.W ilson, C .C .Lin an d D .R.Lide, J. Chem. P h ys.2 3 , 136 (1955).
C.C .Lin an d J.D .Sw alen, Rev. Mod. Phys. 31, (1959) 841.
56.
W .M .Tolles, E .T annenbaum H andelm an an d W .D .G w inn, J. Chem. P hys.43,
(1965) 3019.
57.
S.T. H aubrich, M .A. R oehrig and S.G. Kukolich, J. Chem. P hys.93, (1990)
Struct.
52,
121.
59.
B .E M ann, In Comprehensive Organometallic C hem istry, G. W ilkinson,
F.G .A . Stone, E .W Abel, Eds; Pergam on Press: Oxford 1982, Vol 3 pp
89-171.
E.V .Ivash and D .M .D ennison, J. Chem. P hys. 21, 1804 (1953).
60.
K .M .S innott, J. Chem. P hys. 34, 851 (1961).
61.
T annenbaum , M yers and Gwinn, J. Chem. Phys. 25, 42 (1956).
62.
63.
R .E .N aylor Jr. an d E.B.W ilson Jr., J. Chem. P hys. 26, 1057 (1957).
J.K . T yler, A.P. Cox an d J. Sheridan, N ature 183, 1182 (1959).
64.
T.B . Brill an d G.G. Long, Inorg. Chem. 10 , 74 (1971).
65.
C.C. Lin an d J.D . Swalen, Rev. Mod. Phys. 31, 841 (1959).
66.
A .F. B ern d t and R.E. M arsh, Acta. Cryst. 16 (1963).
67.
F.A . C otto n , G. W ilkinson, Advanced Inorganic Chem istry, 5th ed., Jo h n
W iley and Sons, New York, 1988.
68.
J.P . Collm an, L.C. Hegedus, J.R . N orton, R.G. Finke, Principles and A p ­
plications o f Organotransition M etal C hem istry, U niversity Science Books,
Mill Valley, CA, 1987.; R.H. C rabtree, The Organometallic C hem istry o f the
Transition M etals, Jo h n W iley and Sons: New York, 1988.
69.
G .F Em erson, L. W atts, and R. P e ttit, J. A m .
(1965).
70.
T.H . Lowry an d K.S. Richardson, “M echanism and T heory in O rganic C hem ­
istry"; H arp er and Row: New York, 1987, 31-52 .
71.
G. M aier, S. Pfrien, U. Schafer, and R. M atusch, Angew. Chem. Int. Ed.
Engl. 17, 520-521 (1978).
58.
Chem. Soc. 8 7 , 131-133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
154
72. L. Salem, ‘T he M olecular O rbital Theory of C onjugated System s’, W .A. B en­
jam in: New York, 1966, C hap. 8 .
73. H. O berham m er and H.A. B rune, Z. Naturforsch 2 4 a, 607-612 (1969).
74. M .I. Davis and C.S. Speed, J. Organometallic Chem. 21, 401-413 (1973).
75. E.A. Lucken, R. Pozzi, an d K.R. R am aprasad, J. Mol. Struct. 18, 377-380
(1973).
76. R. P e ttit an d J. Henery, Organic Synthesis 50, 21-24 (1979).
77. S.G. Kukolich, J.V . R und, D .J. Pauley, and R.E. B um garner, J. A m . Chem.
Soc. 110, 7356 (1988).
78.
S.G. Kukolich, M.A. Roehrig, S.T. H aubrich, and J.C . Shea, J. Chem.
P hys.94, 191-194 (1991).
79.
S.G. Kukolich, M.A. Roehrig, D .W . Wallace, and Qi-Qi Chen, Subm itted to
Chemical Physics Letters.
80.
81.
C.C Lin and J.D . Swalen, Rev. Mod. Phys. 31, 841-892 (1954).
F.A . C o tto n and J.M . Troup, J. Organomet. Chem. 77, 369 (1974).
82.
W . Tam , D .F. E ato n , J.C . Calabrese, I.D. W illiams, Y. W ang, and A.G.
A nderson, Chem. M ater. 1, 128 (1989).
83.
J.K .G . W atson, Vibrational Spectra and Structure, J.R . Durig, Ed., Elsevier,
New York, 1977, Vol. 6, pp 1-89.
84.
J. C alabrese, E. I. du P o n t de Nemours and Co., W illm ington, DE, 19880,
private com m unication.
85.
86.
L.H. Scharpen, V.W . Laurie, J. Chem. P h ysA 3 , 2765 (1965).
D. D am iani, L. F erretti, and E. Gallinella, Chem. Phys. L ett.37, 265 (1976).
87.
H. Reihlen, A. G ruhl, G. VonHesting, and 0 . Pfrangle, Justus Liebiop A nn.
Chem. 48 2 , 161 (1930).
88.
B .F. H allam and P.L. Pauson, J. Chem. Soc. 1958, 642.
89.
R. P e ttit and G .F. Em erson, Adv. Organomet. Chem. F.G .A . Stone and R.
W est, Eds., I 1.
90.
S.G. Kukolich, M.A. Roehrig, G.L. Henderson, and Qi-Qi Chen, J. Chem.
P hys.97, 829 (1992).
91.
S.G. Kukolich, M.A. Roehrig, D.W . Wallace, and G.L. Henderson, J. A m .
Chem. Soc., in press.
92.
M .D. M arshall, D.G. P rich ard and J.S. M uenter, J. Chem. Phys.90, 6049
(1989).
93.
A.C. Legon, A.L. W allwork, J.W . Bevan, an d Z. W ang, Chem.
Lett. 180, 57 (1991).
94.
M.D. M arshall, J. Kim, T .A . Hu, L.H. Sun and J.S. M uenter, J. Chem.
P hys.94, 6334 (1991).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Phys.
155
95.
A.C. Legon, P.D. Soper and W .H. Flygare, J. Chem. P hys.74, 4944 (1981).
96.
P.D. Soper, A.C. Legon an d W .H. Flygare, J. Chem. Phys.74, 2138 (1981);
R .S.A ltm an, M.D. M arshall, W. K lem perer an d A. K rupnov, ibid. 79, 52
(1985).
97.
M .R. K eenan, T .K . M inton, A.C. Legon, T .J. Balle an d W .H. Flygare, Proc.
Natl. Acad. Sci., 77, 5583 (1980).
98.
F .J. G oodwin and A .C. Legon, Chem. Phys. L ett.87, 81 (1984).
99.
A. Suckley, Z. W ang, G. B andarage, R .R . Lucchese, A.C. Legon an d J.W .
Bevan, 46</l Ohio State University international Sym posium on M olecular
Spectroscopy, June 17-21, 1991. P ap er T F '8 .
100. D .J. Millen, Can. J. Chem 63, 147 (1985).
101.
T .C . G erm ann, S.L. Tschopp, an d H.S. Gutowsky, 4 7th Ohio State Univer­
sity International Sym posium on Molecular Spectroscopy, June 15-19, 1992,
talk: T H 0 3 .
102.
L. Adamowicz, Chem. Phys. Lett. 192, 199 (1992).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 114 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа