NASA Technical Reports Server (NTRS) 19960048659: Photodissociation Cross Sections for the Production of C2 from C2H Using Laser Induced Hg Photosensitization and Tunable Ultraviolet and Visible Lasers

138 Pages·8 MB·English

by  NASA Technical Reports Server (NTRS)

NASA-CR-202203
i, __ _ _'_ • " ,,
•
Final Technical Report
NASA Grant Number NAGW-903
Principal Investigator: Dr. William M. Jackson
Title of Research: "Photodissociation Cross Sections for the
Production of C2 from C2H Using Laser
Induced Hg Photosensitization and Tunable
Ultraviolet and Visible Lasers"
Period Covered by Report: December 1, 1985 - November 30, 1995
Grantee Institution: University of California at Davis
Davis, California 95616
Continuation Information: This research is being continued under
NASA Grant Number NAGW-5083.
The NASA Technical Officer for this grant is Dr. Jay Bergstralh, NASA
Headquarters, Officeof Space Science, Code SL, Washington, DC 20546-0001.
X2E+ andthea31-I were produced since these states do not correlate to the electronic states of
g u
C2H which are initially excited. This lead us to the conclusion that non-adiabatic transitions must
be occurring as the molecule dissociates. Cometary observations near the nucleus of a comet
. ..... 2 + .
should reflect the lmtlal distribution of the X Z gif they are produced by the photolysis of C2H.
Recent observations of Comet Hyakutake may have confirmed this proposition.
Papers Published
1986, W. M. Jackson, and M. G. Prisant. A comparison of the high resolution IUE
.
observations of CS emission in comets Halley and Giacobini-Zinner, p. 545-551. In
European Space Agency SP-250, Proceedings of the 20th European Space Laboratory.
Symposium, Heidelberg, West Germany.
1987, M. G. Prisant and W. M. Jackson. A Rotational State Population Analysis of the High-
.
Resolution IUE Observation of CS Emission in Comet P/Halley. Astron. Astrophys.
187,489-496.
1986, P. D Feldman., M. C. Festou, M. E A'Hearn, C. Arpigny, T. S. Butterworth, C. B.
.
Cosmovici, A. C, Banks, R. Gilmozzi, W. M. Jackson, L. A. McFadden, T. Patriarchi, D.
G. Schleicher, G. T. Tozzi, M. K. Wallace, H. A. Weaver and T. N. Woods. IUE
observations of comet Halley: Evolution of the UV spectrum between September 1985
and July 1986, p. 325-328. In European Space Agency SP-250, Proceedings of the 20th
European Space Laboratory Symposium, Heidelberg, West Germany.
1987, M. G. Prisant and W. M. Jackson. A Rotational State Population Analysis of the High-
.
Resolution IUE Observation of CS Emission in Comet P/Halley. Astron. Astrophys.
187,489-496.
1989, R.S. Urdahl, Y. Bao, and W.M. Jackson. Laboratory Studies of Photo-dissociation
.
Processes Relevant to the Formation of Cometary Radicals. NASA Conference
Publication 3077, Proceedings of the 1st International Conference on Laboratory
Research for Planetary Atmospheres, Bowie State University, Maryland, p.266.
6. 1990, J. B. Halpern, L. Petway, R. Lu, W. M. Jackson and V. R. McCrary. The
Photochemistry of Cyano- and Dicyanoacetylene at 193 nm. J. Phys. Chem. 94:
1869-1873, 1990.
7. 1991 Yihan Bao, R.S. Urdahl, and W.M. Jackson. Detection of C2(B' 1Zg+) in the
multiphoton dissociation of acetylene at 193 nm. J. Chem. Phys.94:808-809
8. 1991, R.S. Urdahl, Yihan Bao, and W.M. Jackson. An Experimental Determination of the
2
Heatof Formation of C2 and the C-H Bond Dissociation Energy in C2H. Chem.
Phys.Letts. 178: 425-428.
1991 William M. Jackson. Recent Laboratory Photochemical Studies and Their Relationship
.
to the Photochemical Formation of Cometary Radicals. Kluwer Academic Publishers,
Comets in the Post-Halley Era, 1:313-332.
10. 1991, W.M. Jackson, Yihan Bao and R.S. Urdahl. Implications of C2H Photochemistry on the
Modeling of C2 Distributions in Comets. J. Geophys. Research., 96:17,569.
11. 1992 W. M. Jackson, Y. Bao, R. S. Urdahl, X. Song, J. Gosine and Chi Lu, Cometary
implications of recent laboratory experiments on the photochemistry of the C2H and
C3H 2 Radicals, Asteroids,Comets, Meteors 1991, Lunar and Planetary Institute,
Houston, 253-256.
12. 1995 V. R. Morris, Ke-Li Han, and W. M. Jackson, Time-Resolved IR Chemiluminescene
from reactive collisions between hydrogen atoms and SO 2, J. Phys. Chem., 99:10086-10091.
13. 1995 V. M. Blunt, H. Lin, O. Sorkhabi, and W. M. Jackson, Revised Molecular Constants for
the D 1Zu+ State of C2, J. of Molec. Spect., 174: 274-276.
14. 1996 W. M. Jackson, V. Blunt, H. Lin, M. Green, G. Olivera, W. H. Fink, Y. Bao, R. S. Urdahl,
E Mohammad, and M. Zahedi, Astrophysics and Space Science, 236:29-47.
3
Final Report
NASA 903
The principle goal of our research was to understand the formation of free radicals in
comets. To do this we compared laboratory results with cometary observations in attempt to make
sure that the cometary observations agree with what is known about the photochemistry of the
proposed parent molecule. Initially we concentrated on the CS emission in an effort to show that
the parent of this molecule was CS 2. From the cometary observations we knew that the parent of
CS had a very short photochemical lifetime. Two possible parents of the CS radical were COS
and CS 2. Of the two possible parents the CS 2 molecules has the shortest photochemical lifetime
which is consistent with the cometary observations. We also analyzed the rotational profile of the
high resolution cometary spectra to determine if it retained any memory of its formation. The
proposition was that the CS radical is produced with such a short scale length it might show a
rotational distribution that reflects the original distribution the radical is formed with. The
laboratory work shows CS radicals are vibrationally and rotationally excited when they are
produced photochemically. If the CS radicals are excited by a solar photon before they can be
quenched then there is a chance that the cometary distribution might reflect this. Rotational
analysis of the cometary spectrum showed that it could not befit with one Boltzmann
temperature and that the rotational profile required at least two different rotational distributions.
This work shows that although collisions and radiation tend to cool the CS radicals in the inner
coma, the radicals do retain some memory of the initial populations.
We then started to look into the problem of C2 formation in comets. Some time ago we
had postulated that the following mechanism could explain the origin of C2 in comets.
C2H 2 + hv ---> C2H + H (1)
C2H + hv ---> C2 + H (2)
Thus in comets it takes two solar photons to produce the C2 radical. We set out to to see if we
could measure all of the nascent distributions of the C2 products in the hope that they would be a
characteristic signature of the formation process. It is difficult to exactly simulate the conditions
in comets in the laboratory. Rather than do this we wanted to understand the photochemistry of
the intermediate since the C2 in comets will be characteristic of this photochemistry. The C2H
radical should be in it's lowest vibrational and rotational state in comets because it has a
permanent dipole and it will radiatively relaxto this state. Laboratory studies are further
complicated because the C2H is unstable and it must be produced while the study is going on.
The photochemistry of acetylene at 193 nm was used to produce the C2H which were then
photolyzed by light from the same laser. Vibrationally hot C2H radicals are produced in the
laboratory and absorption of the second 193 nm photon allows us to scan several excited
electronic levels that will be photochemically active in comets.
Using this method we were able to show that C2 in the following electronic states are
produced X2Z +g, A1yIu ,a31-Iu, B1Ag, and the B' 1Z+g" We were rather surprised to find that the
_.Plasma & L_a ..........."........ "
20th ESLAB SYMPOSIUM
on the
OF HALLEY'S COMET
EXPLORATION
Proceedings of the International Symposium
Heidelberg, Germany
27 - 31 October 1986
Organised jointly by
- Space Science Department of ESA, ESTEC, Noordwijk, The Netherlands
- Max-Planck-lnstitut fLir Kernphysik, Heidelberg, W. Germany
Sponsored by
- Inter-Agency Consultative Group (IACG)
- International Halley Watch (IHW)
Co-sponsored by
- Committee on Space Research (COSPAR)
- International Astronomical Union (IAU)
european space agency / agence spatiale europ_enne
8-10, rue Mario-Nikis, 75738 PARIS 15, France
545
A Comparison of the High Resolution IUE observations of CS
emission in Comets Halley and Giacobini-Zinner
William M. Jackson I*, and M. G. Prisant 2
IChemistry Department, University of California, Davis, Ca. 95616
2Chemistry Department, University of California, Berkeley, Ca. 94720
Abst_ct observations ofComets Giacobini-Zinner during theICE encounter andHalley's comet on
December 25and26, 1985. Various simulation procedureshave alsobeenused andwill
The high resolution spectra ofcomets Giacobini-Zinner andHalley have been meas- be compared with theinversion procedure. The extracted rotational distributions arethen
ured inthe region ofthe CS emission. Both spectra have essentially thesame shape, but
discussed in terms ofwhat should be expected from amolecular mode!.
theHalley spect_m has abetter signal to noise ratio. Attempts have been made to obtain
the rotational distribution ofthe radical in these cornets byusing an inversion program and
spectral simulation. Both procedures suggest that therotational diswibution isdominated Observations
byCSradicals with J" _5. Despite this itappears that another diswibutinn with higher J"s
Bogess [7]has previously described theIUE insaumentatlon and only theobserva-
is needed to explain the shape of the rotational distribution. Argumenta are presented
tional details and the results will be given inthis section. The CS portion Of the high
which suggest that this second distribution isdue to a balance between collisions and
resolution spectra which wasobtained withthe LWPcamera during observations ofcomets .
radiation inthe inner coma.
Giacobini-Zinner and Halley are shown inFigs. 1and2, respectively. The wavelength
scale isfrom theIUE guest observer tapeand has notbeen corrected fortherelative motion
ofthe comet andthe telescope. A 390 rain exposure was used forthe Giacobini-Zinnec
(LWP-6694) spectrum, obtained on30August "1985,while a720 rain exposure was used
fortheHalley spectrum (LWP 7383) observed onthe 25and26of December, 1985. The
data ispresented asis, with no smoothing andthesignal to noise ratio isestimated to be
12/1 forGiacobini-Zinner and 33/1 forHalley. The noise wasestimated from thenegative
excursion peaks inthe spectra intheregion where there should not be any CS rotational
Introduction lines. The CSspectra wasobserved intwo different orders inbothspectra butinthis paper
only the order that occurs near [he center of the camera will be discussed. The IUE
The CS molecule occupies aunique position in comets. Itisthe only molecule,
instrumental bandwidth isquoted to be 0.2A forapoint source and 0.8 Aforanextended
which isconsistently observed inthe spectra ofcomets that has ashortscale length in the
source. This isvaried intheanalysis because theimages ofthese comets completely fill the
coma [1,2,3,4]. Because of this we have speculated that it's specti'al signature can be a
slit. The true resolution issomewhere between these values•
monitor ofthelocal conditions inthis region of thecoma. The scale length has been shown
to beabout 600 km [4] which isstill inaregion where collisions could affect therotational
populations ofthe CSradicals. To evaluate thisidea amodel isneeded which would allow Thc_
one tomatch theobserved high resolution spectra with theor¢!ical profiles. Earlier avery
The high resolution emission spectrum ofboth comets obtained from lUE'shonld
simple model was used that assumed the rotational distribution ofthe CSradical could be
contain information about the local condition in the inner coma. The production scale
described byaBoltzmann distribution [3]. This amounts toassuming that theCSrotational
distribution isa thermometer for the collision region. Subsequently more theoretical lengths ofCShad previously been shown to beless than [3,4] 1000 km. Thus this species
isformed in theinner coma and may be aprobe ofthisregion. The rotational distribution
analysis and laboratory work [5] has suggested that amore sophisticated model isneeded.
ofthe CS radical in this region could possibly monitor thelocal conditions of the inner
Inparticular a model is required that could extract the rotational distribution from the
coma. Todetermine whether this isthecase itisnecessary to see iftheobserved spectra
observed spectra, which could then be compared with the expected distributions based
can befitted with models that require collisions.
upon amolecular model. Recently Prisant andZare [61have described an inversion proce-
dure that can be used toobtain therotational distribution from anobserved spectra. Inthis There areseveral approaches that can betaken to match theobserved spectra with a
model spectra. Earlier work [31had calculated thehigh resolution spectra assuming that the
paper this method .will be adapted to analyze high resolution spectra obtained in IUE
rotational distribution could befitted with aBoltzmann distribution ofrotational levels. In
thepresent work that approach will be used aswell asthe inversion method [61which has
*Guest observer with the International Ultraviolet Exnlorer Satellite observatory beendeveloped for inverting thedistribution floratheobserved spectra.
Proc. 20th ESLAB Symposium on the Exploration of Halley's Comet, Heidelberg, 27-31 October 1986, ESA SP-250 (December 1986)
W.M. Jackson et al.
2.0 j | i _ _ J
¢/)
t--
1.0
t_
v
>..
I---
o9 0.O
z
ILl
F---
Z
-1.01
t T f I | ? • f
2572 2576 2580 2584 2588
WAVELENGTH (A)
Fig. 1 The CS region of the LWP6694 Giacobini-Zinner spectrum. This is a 390 min exposure and the data is as it is provided to the
guest observer on tape. It was extracted from the tape image using ANA software supplied by Gibor Basir.
4.0 , , , , ' ' '
cO
.1...i 't
o_ 3 o-0
2.0
t_
t_
v
>-
l- 1.0
cO
Z
w
0.0
Z
-1.0 t t f ! f f f
2572 2576 2580 25_,4 2588
WAVELENGTH
Fig. 2 The same spectral region for comet Halley, but this is a 720 rain exposure. The data were taken on 12/25 and 12/26, 1985.
The above equation canbe written inmatrix form asfollows,
Inthe analysis itwill beassumed that theCSradical isproduced byphotodissocia- I =M N (8)
tion ofCS2. Ample evidence has beenpresented that suggest thatthis isthe most likely
Inthisequation the IisaLdimensional column vector with Lbeing thenumber ofpoints
source ofCS [4]. Photodissociation ofCS2 isknown to lead to the production ofhighly
measured inthespectrum, and Nisan Idimensional column vector with Ibeing thenum-
vibrationally androtationally excited CS radicals [5]. The radicals appear tobe formed
ber"ofground state rovibrational levels.
with internal energy uptothethermodynamic limit, sothat at193 nm theycan be formed
with upto 12 quanta of vibrational energy andwith substantial amountS of rotational The rotational levels can be further parameterized, in terms ofJmax where Jmaxis
energy. Once theradicals are formed in'thecometary environment itisunlikely thatthey defined by Emax= BJ"max (J"max+ 1). Let us define anew variable X, where,
wili survive inthese highly excited vibrational levels. The natural radiative lifetime ,"o_a X= -1+(2J/Jmax) (9)
CSradical radiating intheinfrared region viathe following process,
The population for agiven X, P(X), can.be expanded inaLegendre Series, sothat
CS(v") _ CS(v"-l) +hv (I) (10)
N(X) =__.qPq(X)+alq
isoftheorder ofafew milliseconds- This ismuch smaller thanthesolar excitation rate and where, aq=(Pq[ P)=l-'lPqPdt X . (11)
thecollision rate, so thatitisunlikely that vibrationally excited radicals will survive long with (PpIPq)=_1 (12)
enough to beelectronically excited and detected. Similar considerarioas forthe highly ex"
The aqarethemomentS ofthedistribution.
cited rotational levels lead tothe same conclusion for theupper rotational levels, with large
therefore,
spacings. TheexaCt number ofrotational levels that will have to beconsidered will depend I1=,y_,qaq,Y_,iPq(Xi)_jkSijSjk,Oij.O3jkO(,oij)¢[_(__..3.1) (13)
upon radiative relaxation time and thecollision time between the CSradical and thegas in
since,
thecoma. ll=qM (14)
With theabove considerations in mind amodel canbe developed to tryto match the
SO,
observed high resolution spectra. The overall process can bedescribed bythe following
]v[,lq=,"i'_,ijkPq(Xi)SijSjk'Uij'U3jkO('uij)_b(_"j_'_t't)
series ofreactions,
(15)
CS(XI_,v",J '') +hu _ CS(Al[],v',J') (2)
and,
CS(AI[],v, j,) _ CS(XI_,v".J '')+ h_ (3) q={Aql (16)
First, label all ofthe initial levels ofCS(XI,_.) with the index i,the intermediate levels of
The aq arethe moments of thedistribution and may be independently extracted from the
CS(At 1-[)with the index j,and thefinal levels ofCS(XI_) with the index k. The levels i
least-square's analysis ofthespectrum [8].. .
arethepopulated levels inthecoma. thelevels jarc thelevels thatareoptically populated by
absorption of solar radiation, and the levels kare thelevels that areoptically coupled by Results and Discussion
emission fromj. Itisthis latter emission that ismeaSured bythe IUE telescope.
A careful comparison ofFigs. 1and 2suggest that the two spectra were identical
The emission intensity, ljk0-t), from agiven jlevel to aklevel at wavelength 7-tis
except for the signal to noise ratio. Because ofthis, it was decided to only analyze the
proportional to, Halley corner spectra since ithaSamuch better signal tonoise ratio.
ljkO_l)=_jkNjSjk_j3jk_(_.Xi ) (4)
•A synthetic spectra was first derived assuming various Bohzmann temperatures.
where, _(kjk-_.l) is the instrument function of the telescope, "Ojkisthe frequency ofthe Some oftheresults ofthis simulation procedure areshown in Fig. 3, along with theHalley
transition between thejandthe klevels, Sjk isthe natural line strength, kjk isthe Doppler comet spectra for comparison. The 29K spectrum has the right width but it has two
_hifted transition wavelength, and Njis the population of the intermediate level j. The distinct peaks corresponding tothe Rbranch andthe overlapping PandQ branches. The
natural line strength isgiven by, resolution of theIUE instromenr isnot high enough to resolve these latter two branches.
Sjk =q2v,v.p.2jkSjk (5) The 7 Kspectrum has the righ_ shape, but itistoo narrow to match the spectrum that is
observed. Increasing thetemperature does notimprove the match because the two peaks
in which qev" istheFranck-Condon factor, p.isthetransition dipole, andSistherotational
remain and eventually become too wide.
line strength. The sum intheintensity function isover alloptically coupled levels inemis- A simulated spectra was them computed using auniform distribution, such tl_at,
sion and absorption. P(J"i) = P(J"k) for all J"i and J"k < J"max, and is equal to zero outside ofthis range.
The population of the intermediate level j can be determined from thefollowing
These results are shown in Fig. 4, which show that the shape is better fitted with a
relationship, distribution that only employs levels with low J. "me spectra calculated from diswibutions
lqj=_ilqiSil)ijO(1)ij) (6)
with higher J'shave toomany peaks and evcmually become toowide.
where, O('oij) isthe intensity ofthe solar radiation atthe Doppler shifted absorption fre- Fig. 5shows, abetter shape can beobtained byemploying arotational distribution
quency uijand allofthe other symbols are thesame. The equation for theemitted light function that monotonically decreases from I" = 0to J"= 5, Withaslope ofone. The
intensity IIcan now be rewritten as, width of thesimulated spectrum, however, isstill too narrow.
Several bimodal distributions have been tried. The best fitthat wehave been able to
II=EiNi(._jkSijuijU3jkO(Uij)_(_.jk._.l) ) (7)
obtain thus far with abimodal distribution isgiven inFig. 6. This calculated spectrum is
The light intensity ata given line islinearly dependent upon the population ofthe initial wider than the spectra obtained with asingle distribution and has thecorrect shape. Itis
states responsible for that line. Iftheinstrument function issuch that allofthelines arere- still, however, not aswide astheobserved spectrum.
solved then the population of the initial rotational levels are completely defined. Infact Several spectral inversions were attempted using the theory that was described
since there arethree branches, the P,Q, and Rbranches, we have more equations than uu- earlier. All of them gave results similar tothose shown in Fig. 6,which looks very much
knowns. like some of theresults obtained bydirect simulation. Inthis case ithasthe undesirable
properties that both thewidth andshape arein poor agreement with the observed spec0mm.
548 W.M. Jackson et al.
P(J')"I FOR O<_J'._<lO
THEORY BOLTZMANN DIST. P(J')"O FOR J'_lO
T--29K
, _ t'll f
i Ie..I_
I-- o.4 I-" e_4
•Z
tlJ
I,- e.z _z
Z
o.a o.e
a_tz. =_tz.
WAVELENGTH(h) WAVELENGTH(,_)
1.1
t.2
e.4
ttl e.e
I-- LU
Z 14
Z
,.l=e:•_z I Z=_i'4, i Z_7i6, i Z=71i," i l_ll.i il_e; _:tz. " 2574. zs7(. 2=7e. ==ee. izsaz.
WAVELENGTH(,_)
WAVELENGTH(,_)
P(J')'I FOR O_-d'...<'3
THEORY BOLTZMANN DIST.
'_ P(J')'-O FOR J'>3
t.o
m T-_7 K m
c l.l e.8
.o
¢1
I-- e.4 +._ >I-,. e.,i
uJ o.z
I--
z J
tt_tl, trot4. i i=t_.r 1578:. l =:eli. .Il:.'z. , , . ,, • , . ,
Z=t4. isis. 2_tll, =_el. I_IIZ.
WAVELENGTH(A)
WAVELENGTH(,_)
Fig. 3 Two simulated spectrum using a Boltzmann distribution Fig. 4 Two simulated spectra using a uniform distribution.
of rotational levels. The Halley spectrum is included for comparison. The Halley spectrum is included for comparison.
p(O)=lO
p(jO)=p (J'- 1)12;1_-J'-_5
1.0 _ ' ' _' ' ' '
in e.e
e.6 ERVED
C l'e I
o.¢
e.4
e.z
Z LA
e,t
_: ..zstz. t Z:T4.i 2571G. 2S78. 2aee.t i zssz. 0•e57S.I 0 " _ ,w'2_5.A.76V=E.LeENGTH'(A) 22_s7_:7.11. zsrs.e - zsT_,.a"
WAVELENGTH(A)
t.Z
t.o ' ' d
r."
d_ e.6
c_
>-
--, e.e _ERVED
p- e.z
z
._., u u n n t n i i z _,,J LCUL TE
lztz. z_74. Zst6. 2_78. z_e.
2575.0 ZS?6 .ll 2S7r .e 2_;,e.e 2._79.0
WAVELENGTH(A)
WAVELEN GTH(,_,)
Fig. 6 Simulated Spectraobtained using bimodal, distributions, Thetopfigure hasrelative
Fig. 5 Asimulated spectrum using adnear distribution.
populations of 10, 10, 10, 10, 5,0.5, 0.5; 0.5, 0.5, 0.5, 0.2, and 0.2 for theJ" =
O,1,2, 3,4, 5,6,7, 8,9, 10and 11respectively. The bottom figure has arelative
l_pulation of 10, 15,20, 10, 5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.2 and 0.2 respectively.
i i 1
t.O
09
0.8
e"
OBSERVED
0.(;
k,.
V
>-
I-
09
Z
111
I-- e.z
Z
0.0 2_76._ I 2577.0I 2578.0 2579 .O
2575.0
WAVELENGTH(A)
Fig. 7 An inversion using the theory described inthe text.