- Email: [email protected]
The fluorescence of solar ionizing radiation
Planet.Space Sci.1965. Vol.
13, pp. 947 to 957. Pergamon Press Ltd. Printed in Northern Imland
THE FLUORESCENCE
OF SOLAR IONIZING
A. DALGARNO*
RADIATION
and M. B. McELROYt
(Received 22 June 1965)
Abstract--The mid-day dayglow intensities arising from the fluorescence of solar ionizing radiation are calculated. The predicted overhead intensities above 120 km of the O+(*P - *D) lines at R1731%7330, the Meinel band system of NZ+, the first negative system of N,+, the first negative system of O,+, the Hopfield emission system of OZ* and the second negative system of O,+ are respectively 500 R-l kR, 9 kR, 600 R, 2 kR, 600 R and 400 R. 1. INTRODUCTION
When solar ultraviolet radiation ionizes an atmospheric constituent, the positive ion may be left in an excited electronic level which can radiate. The fluorescent emission may be detectable using rocket-borne or ground-level instrumentation, thereby providing an optical method for the investigation of the ionosphere. fn particular, it may be possible to monitor the solar ionizing radiations by observations in the visible.(r) 2. ION PRODUCTION RATES The absorption of ionizing radiation by atomic oxygen can lead to a significant population of several excited states of the resulting positive ion. Thus the ejection of an outer shell electron from atomic oxygen proceeds according to O( ~~3~32~~)3~
+ e
(1)
O( 1sS2s22p4)3P+ hv -+ o+(ls%~2p3)~D -f- e
(2)
_t hv -+
0~(l~22~z2~}*~
O(ls22s22p4)3P + hv --+ O+(ls22~22p3)2P+ e,
(3)
the spectral heads of which are located at, respectively, 910 A, 732 8, and 663 A. ejection of an inner shell electron from atomic oxygen proceeds according to
The
0(ls22.s22p4)3P + hv -+ 0+(1~~2s2~~)~~+ e
(4)
0(1~22~22~*)3~+ hv -+ 0+(~~z2~2~4)2~+ e,
(5)
the spectra1 heads of which are located at, respectively, 434 A and 310 A. The configuration of molecular nitrogen in the ground state is (KKa,2s2s,2s2rr,2p2a,2pqlC,+. The first ionization potential corresponding to the transition
of N, is equivalent
N,(XS,+)
in energy
to a wavelength
+ hv --+ N2+(X2C,+) + e
of 796 A, (6)
in which a a,2p electron is ejected. At 732 A, a rr,2p electron can be ejected, corresponding to the transition N,(XlC,+) + hv -+ N2+(A211,) + e (7) * Department of Applied Mathematics,The Queen’s University of Belfast. t Kitt Peak National Observatory, Tucson, Arizona, U.S.A. 947
948
A. DALGARNO and M. B. MCELROY
and at 661 A a 0~2s electron
can be ejected, corresponding
to the transition
N2(X1Ecg+) + hv * Nzf(B2&+) The location
of the spectral
head for the ejection N,(XlE,+)
+ e.
(8)
of a 0~2s electron
corresponding
to
+ hv + N2+(21;,+) + e
(9)
is uncertain and we have assumed arbitrarily that it is 350 A. The configuration of molecular oxygen in the ground state is (KKq,2s20,2s2ug2p2~,2p4rr,2p2)~&,-. At 1026 A, a rr,2p electron
can be ejected corresponding 02(X3C,-)
At 767 A, a ~,2p electron
+ hv --f 02+(X211,) + e
can be ejected corresponding 02(X3Z8-)
and at 729 A, a rr,2p electron
A 0,2p electron
+ e
(11)
+ hv ---f 02+(A211,) + e
(12)
can be ejected by radiation 02(X3C,-)
and a ~~2s electron
by radiation
to
shorter
than perhaps
+ hv --f 02+(b4Z;,-)
shorter
02(X3E,-)
(10)
to
+ hv * O,+(a4II,)
corresponding
02(X3C,-)
to
+ e
than 504 A corresponding
+ hv + 02+(c41;,-)
682 A corresponding
+ e.
to (13)
to (14)
There are other discontinuities in the photoionization cross section yielding O,+ in unidentified doublet and quartet C, and C, states. We have assumed arbitrarily that the 4E states are energetically accessible to radiation shorter than 430A and the 2C states to radiation shorter than 310 A. The photoionization cross sections for the individual transitions are available for calculations. (12) For molecular oxygen and molecular atomic oxygen from theoretical nitrogen we have assumed, following earlier work, (3) that whenever the photon energy is such that a multiplicity of ionizing transitions can occur the probability of a particular transition is proportional to the statistical weight of the product positive ion states and we have employed the measured total cross sections of Huffman, Tanaka and Larrabee,c4) Cook, Ching and Becker’@ and Sampson and Cairns .@) Schoen’7) has recently given estimates of the percentages of photoionizing transitions in molecular oxygen and nitrogen which terminate in particular states. His estimates are uncertain but they suggest that for N,f the production rate of the A2fII, state should be increased by 50 per cent and the production rate of the B2E ,+ state should be decreased by 75 per cent, the production rate of the ground state being altered so that no change in the total N,+ production rate occurs. rates of the &II, and A2fI, states should be They suggest that for 0, f the production decreased by 50 per cent and the production rate of the b4F, state should be increased by 100 per cent, the production rate of the ground state being altered so that no change in the throughout total O,+ production rate occurs. We have made these changes consistently the calculations. For the distribution of neutral particles we have used the analytical representations
THE
FLUORESCENCE
OF SOLAR
IONJZING
949
RADIATION
TABLE 1. ATMOSPHERICNUMBER DENSITIES(cm-“) Altitude W,)
(km)
n(0)
n(O*)
650 600 550 500 450 400 350 300 275 250 225 200 190 180
1.79(5) 5.08(s>
6.44-l) 5.19 4.31(l) 3.69(2) 3.26(3) 2.98(4j 2.81(5) 2.74(6) 8.68(6) 2.78(7) 8.98(7) 2.96(8) 4.81(8) 7.91(8) 1.32(9) 2.26(‘9) 4.04(9) 7.83(9) 1.78(10) 6.00(10)
1.46(6) 4.29(‘6) 1.27(7) 3.85(7) 1.18(8) 3.70(8) 6.58(8) 1.18(g) 2.12(9) 3.86(9) 4.94(9) 6.36(9) 8.29(g) l.lO(lb) 1.51(10) 2~20(10) 3.62(10) S.OO(lO)
170 160 150 140 130 120
453(l) 2.82(2) 1.80(3) 1.18(4) 7.94(4) 5.51(5) 3.93(6) 2.89(7) 7.92(7) 2.19(8) 6.12(8j 1.74(9) 2.67(9) 4.12(9) 6.47(9) 1.04(lb) 1.74(10) 3.14(10) 6.59(10) 2.00(11)
n(He)
n,
435(5) 5.64(5) 7.36(5) 9.63(5) 1.26(4) 1.67(6) 2.21(6) 2.94(6) 340(6) 3.93(6) 4.56(6) 5.33(6) 5.70(6) 6.12(6) 6.63(6) 7.28(6) 8.19(6) 9.64(6) 1.24(7) 2.00(7)
8.91(4) 1.10(5) 1.51(5) 2.09(5) 2.88(5) 3.98(5j 5.25(5) 646(5 j 6.57(5) 5*89(5) 4.68(5) 3.63(5) 3.31(5) 3.02(5) 2.88(5) 2.66(5) 2.54(5) 2.29(5) 2.00(5) 1.74(5)
introduced by Bates@) and tabulated by McElroy. (g) Our calculations refer to the model atmosphere with an exospheric temperature of 750”K, a temperature gradient of 20°K km-r at an altitude of 120 km and the particle distributions listed in Table 1. This atmosphere is in harmony with that derived by Hinteregger, Hall and Schmidtkeu”) from an analysis of solar ultraviolet absorption. w For the incident flux of solar photons we have used values Hall and Schmidtke,o”) Hall, Schweizer and reported by Hinteregger, (12) Hinteregger, and Hall, Schweizer, Heroux and Hinteregger.u4) The resulting ion Hintereggero3) production rates are presented in Tables 2, 3 and 4. TABLE 2. O+ PRODUCTION RATES (cm-3) \ State Altitude (km) \
20
650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
1.50(-2) 4.27( -2) 1.23(-l) 3.60(-l) 1.07 3.23 9.89 3.05(l) 5.35(l) 9.31(l) 1.59(2) 2.56(2) 3.03(2) 3.49(2) 3+38(2) 4.13(2) 4.13(2) 3.84(2) 3.33(2) 2.65(2)
=P 5.39(-3) 1.53(-2) 4.41(-2) 1.29(-l) 3.83(-l) 1.16 3.54 1.09(l) 1.92(l) 3.33(l) 5.68(l) 9.19(l) 1.09(2) 1.25(2) 1.40(2) 1.49(2) 1.50(2) 1.41(2) 1.24(2) 9.92(l)
4P
=P
7.66(-5) 2.18(-4) 6.27( -4) 1.83(-3) 5.45(-3) 1.65(-2) 5.04(-2) 1.56(-l) 2.74(-l) 4.80(-l) 8.30(- 1) 1.39 1.68 2.01 2.36 2.70 2.98 3.12 3.00 2.47
4.01(-5) 1.14(-4) 3.28( -4) 9.60( -4) 2.85(-3) 8.62(-3) 264(-2) 8.18(-2) 144(-l) 2.54(-l) 4.40(- 1) 7.49(-l) 9.17(-l) 1.11 1.33 1.57 1.81 2.02 2.11 1.91
950
A. DALGARNO TABLE
and M. B. MCELROY
3. N,+ PRODUCTION
RATES (cm-a)
State Altitude (km) \ 650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
TABLE
Altitude (km)\ 650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
State \
ATI”
B%,+
TX,+
1.29(-5) 8.04-5) 5.13(-4) 3.36(-3) 2.26( -2) 1.57(-l) 1.11 8.09 2.17(l) 5.85(l) 1.53(2) 3.78(2) 5.25(2) 7.06(2) 9.07(2) 1.09(3) 1.34(3) 1.15(3) l-00(3) 9.66(2)
9.43(-7) 5.85(-6) 3.75(-5) 2.45( -4) 1.65(-3) 1.14(-3) 8.12(-2) 5.87(-l) 1.58 4.25 1.11(l) 2.75(l) 3.80(l) 5.01(l) 6.62(l) 8.000) 8.88(l) 8.82(l) 8.08(l) 8.05(l)
6.72( -7) 4*18(-6) 2.67(-5) 1.75( -4) 1.18(-3) 8.16(-3) 5.8q-2) 4.23(-l) 1.15 3.12 8.40 2.22(l) 3.25(l)
4. OS+ PRODUCTION
a411” 7.25( -8) 5.80( -7) 4.84(-6) 4.14(-5) 3.66(-4) 3.34(-33) 3.14(-2) 3.02(-l) 9.45(-l) 2.92 8.90 2.57(l) 3.83(l) 5.50(l) 7.70(l) 1.02(2) 1.25(2) 142(2) 1.52(2) 1.70(2)
AaII U 3.46(-g) 2.79(-7) 2.31(-6) 1.98(-5) l-75(-4) 1.60(-3) 1.50(-2) 1.45(-l) 4.51(-l) 1.40 4.26 1.23(l) 1.83(l) 2.65(l) 3.69(l) 4.90(l) 6.05(l) 6*95(l) 7.50(l) 8.50(l)
::::c; 9.54(l) 1.32(2) 1.79(2) 2.32(2) 2.82(2)
RATES (cm-“)
b’C,-
&&-
1.27(-7) 1.03(-6) 8.54( -6) 7.30( -5) 6.46(-4) 5.88(-3) 5.54( -2) 5.34(-l) 1.66 5.16 1.57(l) 4.54(l) 6.74(l) 9:78(l) 1.37(2) 1.83(2) 2.28(2) 2.66(2) 2.96(2) 340(2)
3.17(-g) 2.56( -7) 2.12(-6) 1.82(-5) 1.61(-4) 1.47(-3) 1.38(-2) 1.33(-l) 4.16(-l) 1.30 4.01 1.20(l) 1.83(l) 2.74(l) 404(l) 5.79(l) 8.03(l) 1.07(2) 1.35(2) 1.63(2)
THE
FLUORESCENCE
OF SOLAR
3. EXCITATION
REMOVAL
IONIZING
951
RADIATION
PROCESSES
3.1 Atomic oxygen ions The excited transition
O+PP state produced
by (4) will decay rapidly
according
to the allowed
o+( lSZ2S2P4)4P + o+( ls22s22p3)4S + hv
(15)
with the emission of a photon of wavelength 833 A which can be absorbed by atomic and molecular oxygen in the transitions (1) and (10) respectively. The excited state produced by (5) will also decay rapidly either through o+( lS22S2P4)2P -+ o+( lS22S22P3)2D + hv with the emission
of a photon
of wavelength
(16)
537 A or through
o+( lS22S2P4)2P -+ o+( lsZ2s22p3)2P + hv with the emission of a photon of wavelength 581 A. According the relative probability of occurrence of (16) to (17) is 3.27. photon produces further ionization and the rates of population Of are slightly increased by the cascade processes. The radiative lifetime of the 2D3,2 metastable state of Of state, 2 x lo4 sec.(16) They can be deactivated by electron 2D5/2
(17)
to Cohen and Dalgarno,(15) In both cases, the emitted of the metastable states of is 5 x lo3 set and of the impact
e + 0+(2D) + e + 0+(4S), the rate coefficient of which is about particles such as ion-atom interchange
3 x lOpa cm3 set-l
(18) 06) or by collisions
with neutral
O+(2D) + N, + NO+ + N
(19)
and charge transfer 0+(2D) + N, + 0 + N2+ O+(2D) + 0, --+ 0 + 02+(411, 211).
(20)
0mholt(17) has pointed out that if the N2+ ion . in (20) is produced in the Y = 1 vibrational level of the A2H, state, the process is in close resonance and Hunten has presented evidence that the 1-O and l-2 bands of the Meinel system may often have anomalously high intensities during auroras. The O+(2D-4S) multiplet at 113729-3726 has been observed in auroras by several groups. os) It is variable in occurrence and was not positively identified until the great red aurora of 11th February 1958 when Wallace(20) resolved the two lines. Its weakness in ordinary auroras suggests that deactivation is severe and there seems little possibility that significant radiation can occur in the normal dayglow. However if (20) is responsible for suppressing the emission there may occur an enhancement of the u’ = 1 progression of the Meinel band system, a possibility to which we shall return in $4. The radiative lifetime of the 2P112metastable state is 5 set and of the 2P3,2 metastable state 4 set and significant radiation may arise from the 2P states,(l) the computed overhead intensity at noon in the absence of deactivation being 1.5 kR. The 2P states will undergo some deactivation by impacts with the ambient electrons e + 0f(2P) + e + 0+(2D, 4S). We adopt
the electron
density
distribution
listed in Table
1 and a rate coefficient
(21) tc for
A. DALGARNO
952
and M. B. MCELROY
(21) of 1 x IO-’ cm3 secf1.06) The effect of electron impact deactivation on the luminosity altitude profile of radiation emitted by 0+(2P) is shown in Fig. 1. Electron impact deactivation reduces the overhead intensity from 1.5 to 1.3 kR. The 2P-2D multiplet at 217319-7330 has been observed in auroras(17,20,21) and Chamberlain has argued that the observations indicate some deactivation, occurring presumably through reactions similar to (19) and (20). The effects of a deactivation coefficient of lo-lo cm3 set-l in collisions with N, on the altitude profiles of radiation emitted 600-
100 10-z
I IO“
I I
I IO
I 102
I IO"
FIG. 1. THE ALTITUDE PROFILE OF FLUORESCENT EMISSION AT NOON FROM THE 0+(2P-sD) TRANSITION. FROM RIGHT TO LER, THE CURVES SHOW THE INCLUSION OF ELECTRON IMPACT DEACTIVATION AND MOLECULAR NITROGEN IMPACT DEACTIVATION.
from the 2P states are illustrated in Fig. 1. The emission at low altitudes is suppressed and the peak intensity occurs at 200 km instead of 150 km. The overhead intensity is reduced from 1.3 to 0.5 kR. Because of the uncertainty in the deactivation coefficient, accurate predictions of the intensities of the four lines of the 2P-2D multiplet are not possible. The intensities given in Table 5 correspond to no deactivation by N, and to a deactivation coefficient of lo-lo cm3 se&. They are consistent with an earlier estimate.(l) 3.2 Molecular nitrogen ions The excited N,+(2C,+) molecules produced by (9) will, if stable, decay rapidly to the B%;,+ and ASIA, states. The resulting emission is concentrated in the ultraviolet region of TABLE 5. INTENSITIESAT NOON
Transition
2P11,-2D312 =PI,B-~&z =PsIB-~D~IE =Ps/a-=Dm
Wavelength
1329.9 7318.6 7330.7 7319.4
OF THE O+PP-BD)
LINES
Intensityin rayleighs No deactivation N, deactivation 280 190 280 520
100 70 120 230
THE FLUORESCENCE
OF SOLAR IONIZING
RADIATION
953
the spectrum and it will be reabsorbed by 0, and 0,. We assume arbitrarily that the cascading is primarily to the A2111, state in which case its population is increased by about 12 per cent. The A2111, and B%,+ states both decay rapidly by allowed transitions giving rise respectively to the infra-red Meinel system and the first negative system. Deactivation is unimportant and the rates of population shown in Fig. 2 are also the rates of emission of 600-
500 -
E x
400-
:: =I c 2
300 -
200~-
‘O&e
I 10-s
I 10-a
I 10-3
I 10-Z
I IO“
I
I
I
IO
I
I@
I IO’
FIG. 2. THERATESOFPOPULATION AT NOONOFTHEUPPERLEVELS AZ& AND B*C,+ OFRESPECTIVELY THE MEINEL AND THE FIRSTNEGATIVEBAND SYSTEMS OF NITROGEN.
the two systems. The overhead intensities are shown as functions of altitude in Fig. 3. Above 120 km the intensity of the Meinel system is 9 kR and the intensity of the first negative system is 600 R. It is not possible to predict with confidence the distributions of intensities within the band systems beacuse a significant contribution to photoionization may come from autoprinciple. ionizing processes”) which do not necessarily conform to the Franck-Condon If autoionization is not important the vibrational distributions are governed mainly by the Franck-Condon factors(22) and the intensity distribution in the first negative system should approximate that of Table 6. The intensity distribution in the Meinel system is less well determined but perhaps 3 kR will appear in the (2,0) band at 7850A, l-5 kR in the (3,l) band at 808OA and 750 R in the (4,2) band at 8321 A. 3.3 Molecular
oxygen
ions
The unidentified doublet and quartet states of O,+ will decay rapidly by allowed transitions in the ultraviolet or they may dissociate into 0 and Of. The cascading transitions will increase the populations of the lower lying levels of O$ but in no case will the increase exceed 10 per cent and we exclude the high lying states from further consideration. The c4&- and b4E,- states produced respectively by (14) and (13) will radiate by allowed transitions. The former gives rise to the Hopfield emission bands c4XU- - b4C,in the ultraviolet(23) and the latter to the first negative band system b4X,- - hIIl, in the
A. DALGARNO
954
and M. B. MCELROY
region of 6000 A. Deactivation is unimportant and the rates of population shown in Fig. 4 are also the rates of emission of the two systems. The overhead intensities are shown as functions of altitude in Fig. 5. Above 120 km, the intensity of the Hopfield emission bands is 600 R and of the first negative system 2 kR. The vibrational distribution in the first negative system which occurs if the FranckCondon principle is applicable (see however Schoen”)) is given in Table 7.
Fro.
3. THE OVERHEAD
TABLE
INTENSITIES OF THE MEINEL NITROGEN.
AND
FIRST NEGATWE
BAND
SYSTEMS OF
6. OVERHEADINTENSITIES AT 12Okm OF THE FIRST NEGATIVE SYSTEM OF N,+
DUE
TO FLUORESCENCE Intensity
Band
Wavelength
(rayleighs)
090 OJ OF2 093 1,o 191 192 133 194
3914.4 4278.1 4709.2 5228.3 3582.1 3884.3 4236.5 4651.8 5148.8
720 100 25 5 25 20 20 15 2
The A211, state produced by (12) decays to the ground state with the emission of the second negative band system in the ultraviolet. The luminosity profile is illustrated in Fig. 4 and the overhead intensity in Fig. 5. Above 120 km the overhead intensity is 420 R. The &III, state produced by (11) and by cascading from the b4Z,- state is metastable. It is probably severely deactivated by electron impacts and by collision processes such as atom-atom interchange O,‘(d’II,) + O(V) -+ O(V) + o,+(xW,) (22)
THE
FLUORESCENCE
OF SOLAR
IONIZING
RADIATION
955
FIG. 4. THE RATES OF POPULATION AT NOON OF THE UPPER LEVELS &,,-,bPC,- AND APIIu OF RESPECTIVELY THE HOPFIELD AND THE FIRST AND SECOND NEGATIVE BAND SYSTEMS OF OXYGEN.
FIG. 5. FROM RIGHT TO LEE THE CAPTIONS SHOW THE OVERHEAD INTENSITIES OF THE FIRST NEGATIVE, THE HOPFIELD AND TKE SECOND NEGATIVE BAND SYSTEMS OF OXYGEN.
and little emission is to be expected. With our (arbitrary) assumptions about the individual photoionization cross sections, the integrated rate of production of metastable molecules above 120 km is l-5 x lo9 cm-2 set-l compared to a total O,+ production rate of I.3 x lOlo cm-2 set-l. Accordingly metastable rPn, O,+ molecules may play a significant role in determining ionospheric composition in the E- and F-regions. 4. OTHER
EXCITATION
MECHANISMS
The excited ionic states may be populated by other mechanisms in addition to the direct Collisions between the fast photoelectrons produced by absorption of solar radiation.
A. DALGARNO
956
and M. 3. MCELROY
TABLE 7. OVERSIEZ~ INTENSITIESAT 120 km SYsmhi OF o.+ Band 090 OJ 092 120 1,l 132 2,Q 2s 292
Wavelength 6026 6419 6856 5632 5973 6351 5296 5.598 5926
OF THE FIRST NEGATIVE Intensity (rayleighs)
-
180 200 150 420 30 45 7.5 60 45
photoionization and the neutral atmospheric particles constitute a source of excitation which may be significant. Detailed calculations will be required in general but an estimate of the electron impact contribution to the first negative system of nitrogen can be readily obtained using the known efficiency with which electrons absorbed in air produce 3914A emission.@) Electron impacts are responsible for about 450 R above 120 km. It may be shown that as in aurorasoO) electron impact with the positive ions gives a negligible cont~bution despite the large threshold cross sections. Resonance scattering of solar radiation by the positive ions must be a major source of excitation if the upper state is accessible by an allowed transition from the ground state. Thus resonance scattering by Nzf ground state ions contributes to the Nz+ first negative and Meinel band systems and resonance scattering by O,+ ground state ions contributes to the O,+ second negative system. Resonance scattering by O,fdiII, metastable molecules may contribute to the O,+ first negative system. The O-O band at 3914 A has been observed in the day airglow by Wallace and Nideycz5) and by Zipf and Fastie,(26) who have shown that with plausible assumptions about the N2+ density distribution the measured intensities of 5 kR at a solar zenith angie of 76’ 25 and of 7 kR at a solar zenith angle of 60’ tz5)can be explained by resonance scattering. The predicted intensity from fluorescent scattering and from electron impact is several hundred rayleighs for the solar angles obtaining during the measurements but their contributions are detectable in principle by a detailed analysis of the luminosity profile low in the atmosphere where there are few Nz+ ions and by the Swings effect.(27,Z5) Zipf and Fastie(26) have concluded from their observations that the efficiency with which photoionization of nitrogen produces a 39148 photon is at most 5 per cent. Our predictions at 120 km, based on Schoen’s data(‘) and assuming no contribution by downward transitions from the %,+ state, are equivalent to an efficiency of 2.7 per cent for photoionization and an efficiency of 3-7 per cent for the associated electron impact ionization. The un~rtainties in the composition of the neutral atmosphere are such that there is not necessarily any discrepancy. Specific collision processes by which an ion is converted into a different species may lead to selective enhancements of spectral features. Only one appears to be of interest for ionic emissions. If all the 0+(20) ions were removed by 0+(20) + N,(XlC,+, u = 0) -+ 0(3P) + Nz+(A211,, 21= 1)
(23)
4 kR of radiation would occur in the Meinel bands originating in the v = 1 level, comparable to that anticipated from fluorescence. The altitude distributions of the two sources
THE FLUORESCENCE
OF SOLAR IONIZING
would differ significantly, fluorescence tending (23) tending to follow the oxygen distribution.
to follow
RADIATION
the nitrogen
957
distribution
and
Acknow[edgments-Contribution 114 from the Kitt Peak National Observatory operated by the Association of Universities for Research in Astronomy under contract with the National Science Foundation. The work was uartlv suouorted bv the GCA Coruoration under contract with the National Aeronautics I and Space Adr&nist;atid~. * REFERENCES 1. A. DALGARNOand M. B. MCELROY,Planet. Space Sci. 11, 727 (1963). 2. A. DALGARNO,R. J. W. HENRYand A. L. STEWART,Planet. Space Sci. 12,235 (1964). 3. A. DALGARNO,M. B. MCELROYand R. J. MOFFETT,Planet. Space Sci. 11,463 (1963). 4. R. E. HUFFMAN,Y. TANAKAand J. C. LARRABEE, Disc. Faraday Sot. 37, 159 (1964). 5. G. R. COOK, B. K. CHINGand R. A. BECKER,Disc. Faraday Sot. 37, 149 (1964). 6. J. A. R. SAMSONand R. B. CAIRNS,J. Geophys. Res. 69, 4583 (1964). 7. R. I. SCHOEN,J. Chem. Phys. 40,183O (1964). 8. D. R. BATES,Proc. Roy. Sot. A 253, 451 (1959). 9. M. B. MCELROY, Contribution from the Kitt Peak National Observatory, No. 55 (1964). 10. H. E. HINTEREGGER, L. A. HALL and G. SCHMIDTKE Space Research V (editor P. Muller). North Holland, Amsterdam (1965). 11. M. B. MCELROY,Planet. Space Sci. 13, 403 (1965). 12. H. E. HINTEREGGER, J. Geophys. Res. 66, 2367 (1961). 13. L. A. mu, W. SCHWFIZERand H. E. HINTEREGGER, J. Geophys. Res. 70, 105 (1965). 14. L. A. HALL, W. SCHWEIZER,L. HEROUXand H. E. HINTEREGGER, Astrophys. J. 142, 13 (1965). 15. M. COHENand A. DALGARNO,Proc. Roy. Sot. A 280,258 (1964). 16. M. J. SEATONand D. E. OSTERFIROCK, Astrophys. J. 125, 66 (1927). 17. A. OMHOLT,J. Atmos. Terr. Phys. 10, 324 (1957). 18. D. M. HUNTEN,Ann. Gkophvs. 14, 167 (1958). 19. J. W. CHAMBERLAIN, Phys’ici of the Au&a and Airglow Academic Press, New York (1961). 20. L. WALLACE,J. Atmos. Terr. Phvs. 17. 46 (1959). 21. M. DUFAY, knn. geophys. 15, 1$4 (19i9); k. V~LLANCEJONES,Canad. J. Phys. 38,453 (1960). 22. R. W. NICHOLLS,Ann. geophys. 20, 99 (1964). 23. F. J. LE BLANC,J. Chem. Phys. 38,487 (1963). 24. A. DALGARNO,Ann. geophys. 20, 67 (1964). 25. L. WALLACEand R. A. NIDEY,J. Geophys. Res. 69, 471 (1964). 26. E. C. ZIPF and W. G. FASTIE,J. Geophys. Res. 69, 2357 (1964). 27. A. VALLANCEJONESand D. M. HUNTEN,Canad. J. Phys. 38,458 (1960). Pe3IOM+BaTOti pa6OTe J(aIOTCFIBbIVEZCJIeHHR AHTeHCI%BHOCTH IIOJIf~eHHOFO CBe=IeHHR He6a, Bo3HHKaIoIqme OT @Ioypec~eHqm comevHofi Hom3i4pymueti paEnaIxm. npeABEIaeHHafI 3eHlXTHaFI I'IHTeHCHBHOCTb Ham 120 KM JllJHHi'i0+(2P-2D) IIpIi ill, 7319-7330, cncTeMa nonoc MeikHem Nz+,nep3a~ HeraTmHaB cmTeMa Nz+,nep~aa HeraTHBHaR CMCTeMa o,+, BMACCROHHBR CRCTeMa rOII$UIbj.(a o,+, -COCTaBJIFIlOT, CooTBeTcTBemro, 500 R-l kR, 9 kR, 600 R, 2 kR, 600 R and 400 R.
13, pp. 947 to 957. Pergamon Press Ltd. Printed in Northern Imland
THE FLUORESCENCE
OF SOLAR IONIZING
A. DALGARNO*
RADIATION
and M. B. McELROYt
(Received 22 June 1965)
Abstract--The mid-day dayglow intensities arising from the fluorescence of solar ionizing radiation are calculated. The predicted overhead intensities above 120 km of the O+(*P - *D) lines at R1731%7330, the Meinel band system of NZ+, the first negative system of N,+, the first negative system of O,+, the Hopfield emission system of OZ* and the second negative system of O,+ are respectively 500 R-l kR, 9 kR, 600 R, 2 kR, 600 R and 400 R. 1. INTRODUCTION
When solar ultraviolet radiation ionizes an atmospheric constituent, the positive ion may be left in an excited electronic level which can radiate. The fluorescent emission may be detectable using rocket-borne or ground-level instrumentation, thereby providing an optical method for the investigation of the ionosphere. fn particular, it may be possible to monitor the solar ionizing radiations by observations in the visible.(r) 2. ION PRODUCTION RATES The absorption of ionizing radiation by atomic oxygen can lead to a significant population of several excited states of the resulting positive ion. Thus the ejection of an outer shell electron from atomic oxygen proceeds according to O( ~~3~32~~)3~
+ e
(1)
O( 1sS2s22p4)3P+ hv -+ o+(ls%~2p3)~D -f- e
(2)
_t hv -+
0~(l~22~z2~}*~
O(ls22s22p4)3P + hv --+ O+(ls22~22p3)2P+ e,
(3)
the spectral heads of which are located at, respectively, 910 A, 732 8, and 663 A. ejection of an inner shell electron from atomic oxygen proceeds according to
The
0(ls22.s22p4)3P + hv -+ 0+(1~~2s2~~)~~+ e
(4)
0(1~22~22~*)3~+ hv -+ 0+(~~z2~2~4)2~+ e,
(5)
the spectra1 heads of which are located at, respectively, 434 A and 310 A. The configuration of molecular nitrogen in the ground state is (KKa,2s2s,2s2rr,2p2a,2pqlC,+. The first ionization potential corresponding to the transition
of N, is equivalent
N,(XS,+)
in energy
to a wavelength
+ hv --+ N2+(X2C,+) + e
of 796 A, (6)
in which a a,2p electron is ejected. At 732 A, a rr,2p electron can be ejected, corresponding to the transition N,(XlC,+) + hv -+ N2+(A211,) + e (7) * Department of Applied Mathematics,The Queen’s University of Belfast. t Kitt Peak National Observatory, Tucson, Arizona, U.S.A. 947
948
A. DALGARNO and M. B. MCELROY
and at 661 A a 0~2s electron
can be ejected, corresponding
to the transition
N2(X1Ecg+) + hv * Nzf(B2&+) The location
of the spectral
head for the ejection N,(XlE,+)
+ e.
(8)
of a 0~2s electron
corresponding
to
+ hv + N2+(21;,+) + e
(9)
is uncertain and we have assumed arbitrarily that it is 350 A. The configuration of molecular oxygen in the ground state is (KKq,2s20,2s2ug2p2~,2p4rr,2p2)~&,-. At 1026 A, a rr,2p electron
can be ejected corresponding 02(X3C,-)
At 767 A, a ~,2p electron
+ hv --f 02+(X211,) + e
can be ejected corresponding 02(X3Z8-)
and at 729 A, a rr,2p electron
A 0,2p electron
+ e
(11)
+ hv ---f 02+(A211,) + e
(12)
can be ejected by radiation 02(X3C,-)
and a ~~2s electron
by radiation
to
shorter
than perhaps
+ hv --f 02+(b4Z;,-)
shorter
02(X3E,-)
(10)
to
+ hv * O,+(a4II,)
corresponding
02(X3C,-)
to
+ e
than 504 A corresponding
+ hv + 02+(c41;,-)
682 A corresponding
+ e.
to (13)
to (14)
There are other discontinuities in the photoionization cross section yielding O,+ in unidentified doublet and quartet C, and C, states. We have assumed arbitrarily that the 4E states are energetically accessible to radiation shorter than 430A and the 2C states to radiation shorter than 310 A. The photoionization cross sections for the individual transitions are available for calculations. (12) For molecular oxygen and molecular atomic oxygen from theoretical nitrogen we have assumed, following earlier work, (3) that whenever the photon energy is such that a multiplicity of ionizing transitions can occur the probability of a particular transition is proportional to the statistical weight of the product positive ion states and we have employed the measured total cross sections of Huffman, Tanaka and Larrabee,c4) Cook, Ching and Becker’@ and Sampson and Cairns .@) Schoen’7) has recently given estimates of the percentages of photoionizing transitions in molecular oxygen and nitrogen which terminate in particular states. His estimates are uncertain but they suggest that for N,f the production rate of the A2fII, state should be increased by 50 per cent and the production rate of the B2E ,+ state should be decreased by 75 per cent, the production rate of the ground state being altered so that no change in the total N,+ production rate occurs. rates of the &II, and A2fI, states should be They suggest that for 0, f the production decreased by 50 per cent and the production rate of the b4F, state should be increased by 100 per cent, the production rate of the ground state being altered so that no change in the throughout total O,+ production rate occurs. We have made these changes consistently the calculations. For the distribution of neutral particles we have used the analytical representations
THE
FLUORESCENCE
OF SOLAR
IONJZING
949
RADIATION
TABLE 1. ATMOSPHERICNUMBER DENSITIES(cm-“) Altitude W,)
(km)
n(0)
n(O*)
650 600 550 500 450 400 350 300 275 250 225 200 190 180
1.79(5) 5.08(s>
6.44-l) 5.19 4.31(l) 3.69(2) 3.26(3) 2.98(4j 2.81(5) 2.74(6) 8.68(6) 2.78(7) 8.98(7) 2.96(8) 4.81(8) 7.91(8) 1.32(9) 2.26(‘9) 4.04(9) 7.83(9) 1.78(10) 6.00(10)
1.46(6) 4.29(‘6) 1.27(7) 3.85(7) 1.18(8) 3.70(8) 6.58(8) 1.18(g) 2.12(9) 3.86(9) 4.94(9) 6.36(9) 8.29(g) l.lO(lb) 1.51(10) 2~20(10) 3.62(10) S.OO(lO)
170 160 150 140 130 120
453(l) 2.82(2) 1.80(3) 1.18(4) 7.94(4) 5.51(5) 3.93(6) 2.89(7) 7.92(7) 2.19(8) 6.12(8j 1.74(9) 2.67(9) 4.12(9) 6.47(9) 1.04(lb) 1.74(10) 3.14(10) 6.59(10) 2.00(11)
n(He)
n,
435(5) 5.64(5) 7.36(5) 9.63(5) 1.26(4) 1.67(6) 2.21(6) 2.94(6) 340(6) 3.93(6) 4.56(6) 5.33(6) 5.70(6) 6.12(6) 6.63(6) 7.28(6) 8.19(6) 9.64(6) 1.24(7) 2.00(7)
8.91(4) 1.10(5) 1.51(5) 2.09(5) 2.88(5) 3.98(5j 5.25(5) 646(5 j 6.57(5) 5*89(5) 4.68(5) 3.63(5) 3.31(5) 3.02(5) 2.88(5) 2.66(5) 2.54(5) 2.29(5) 2.00(5) 1.74(5)
introduced by Bates@) and tabulated by McElroy. (g) Our calculations refer to the model atmosphere with an exospheric temperature of 750”K, a temperature gradient of 20°K km-r at an altitude of 120 km and the particle distributions listed in Table 1. This atmosphere is in harmony with that derived by Hinteregger, Hall and Schmidtkeu”) from an analysis of solar ultraviolet absorption. w For the incident flux of solar photons we have used values Hall and Schmidtke,o”) Hall, Schweizer and reported by Hinteregger, (12) Hinteregger, and Hall, Schweizer, Heroux and Hinteregger.u4) The resulting ion Hintereggero3) production rates are presented in Tables 2, 3 and 4. TABLE 2. O+ PRODUCTION RATES (cm-3) \ State Altitude (km) \
20
650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
1.50(-2) 4.27( -2) 1.23(-l) 3.60(-l) 1.07 3.23 9.89 3.05(l) 5.35(l) 9.31(l) 1.59(2) 2.56(2) 3.03(2) 3.49(2) 3+38(2) 4.13(2) 4.13(2) 3.84(2) 3.33(2) 2.65(2)
=P 5.39(-3) 1.53(-2) 4.41(-2) 1.29(-l) 3.83(-l) 1.16 3.54 1.09(l) 1.92(l) 3.33(l) 5.68(l) 9.19(l) 1.09(2) 1.25(2) 1.40(2) 1.49(2) 1.50(2) 1.41(2) 1.24(2) 9.92(l)
4P
=P
7.66(-5) 2.18(-4) 6.27( -4) 1.83(-3) 5.45(-3) 1.65(-2) 5.04(-2) 1.56(-l) 2.74(-l) 4.80(-l) 8.30(- 1) 1.39 1.68 2.01 2.36 2.70 2.98 3.12 3.00 2.47
4.01(-5) 1.14(-4) 3.28( -4) 9.60( -4) 2.85(-3) 8.62(-3) 264(-2) 8.18(-2) 144(-l) 2.54(-l) 4.40(- 1) 7.49(-l) 9.17(-l) 1.11 1.33 1.57 1.81 2.02 2.11 1.91
950
A. DALGARNO TABLE
and M. B. MCELROY
3. N,+ PRODUCTION
RATES (cm-a)
State Altitude (km) \ 650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
TABLE
Altitude (km)\ 650 600 550 500 450 400 350 300 275 250 225 200 190 180 170 160 150 140 130 120
State \
ATI”
B%,+
TX,+
1.29(-5) 8.04-5) 5.13(-4) 3.36(-3) 2.26( -2) 1.57(-l) 1.11 8.09 2.17(l) 5.85(l) 1.53(2) 3.78(2) 5.25(2) 7.06(2) 9.07(2) 1.09(3) 1.34(3) 1.15(3) l-00(3) 9.66(2)
9.43(-7) 5.85(-6) 3.75(-5) 2.45( -4) 1.65(-3) 1.14(-3) 8.12(-2) 5.87(-l) 1.58 4.25 1.11(l) 2.75(l) 3.80(l) 5.01(l) 6.62(l) 8.000) 8.88(l) 8.82(l) 8.08(l) 8.05(l)
6.72( -7) 4*18(-6) 2.67(-5) 1.75( -4) 1.18(-3) 8.16(-3) 5.8q-2) 4.23(-l) 1.15 3.12 8.40 2.22(l) 3.25(l)
4. OS+ PRODUCTION
a411” 7.25( -8) 5.80( -7) 4.84(-6) 4.14(-5) 3.66(-4) 3.34(-33) 3.14(-2) 3.02(-l) 9.45(-l) 2.92 8.90 2.57(l) 3.83(l) 5.50(l) 7.70(l) 1.02(2) 1.25(2) 142(2) 1.52(2) 1.70(2)
AaII U 3.46(-g) 2.79(-7) 2.31(-6) 1.98(-5) l-75(-4) 1.60(-3) 1.50(-2) 1.45(-l) 4.51(-l) 1.40 4.26 1.23(l) 1.83(l) 2.65(l) 3.69(l) 4.90(l) 6.05(l) 6*95(l) 7.50(l) 8.50(l)
::::c; 9.54(l) 1.32(2) 1.79(2) 2.32(2) 2.82(2)
RATES (cm-“)
b’C,-
&&-
1.27(-7) 1.03(-6) 8.54( -6) 7.30( -5) 6.46(-4) 5.88(-3) 5.54( -2) 5.34(-l) 1.66 5.16 1.57(l) 4.54(l) 6.74(l) 9:78(l) 1.37(2) 1.83(2) 2.28(2) 2.66(2) 2.96(2) 340(2)
3.17(-g) 2.56( -7) 2.12(-6) 1.82(-5) 1.61(-4) 1.47(-3) 1.38(-2) 1.33(-l) 4.16(-l) 1.30 4.01 1.20(l) 1.83(l) 2.74(l) 404(l) 5.79(l) 8.03(l) 1.07(2) 1.35(2) 1.63(2)
THE
FLUORESCENCE
OF SOLAR
3. EXCITATION
REMOVAL
IONIZING
951
RADIATION
PROCESSES
3.1 Atomic oxygen ions The excited transition
O+PP state produced
by (4) will decay rapidly
according
to the allowed
o+( lSZ2S2P4)4P + o+( ls22s22p3)4S + hv
(15)
with the emission of a photon of wavelength 833 A which can be absorbed by atomic and molecular oxygen in the transitions (1) and (10) respectively. The excited state produced by (5) will also decay rapidly either through o+( lS22S2P4)2P -+ o+( lS22S22P3)2D + hv with the emission
of a photon
of wavelength
(16)
537 A or through
o+( lS22S2P4)2P -+ o+( lsZ2s22p3)2P + hv with the emission of a photon of wavelength 581 A. According the relative probability of occurrence of (16) to (17) is 3.27. photon produces further ionization and the rates of population Of are slightly increased by the cascade processes. The radiative lifetime of the 2D3,2 metastable state of Of state, 2 x lo4 sec.(16) They can be deactivated by electron 2D5/2
(17)
to Cohen and Dalgarno,(15) In both cases, the emitted of the metastable states of is 5 x lo3 set and of the impact
e + 0+(2D) + e + 0+(4S), the rate coefficient of which is about particles such as ion-atom interchange
3 x lOpa cm3 set-l
(18) 06) or by collisions
with neutral
O+(2D) + N, + NO+ + N
(19)
and charge transfer 0+(2D) + N, + 0 + N2+ O+(2D) + 0, --+ 0 + 02+(411, 211).
(20)
0mholt(17) has pointed out that if the N2+ ion . in (20) is produced in the Y = 1 vibrational level of the A2H, state, the process is in close resonance and Hunten has presented evidence that the 1-O and l-2 bands of the Meinel system may often have anomalously high intensities during auroras. The O+(2D-4S) multiplet at 113729-3726 has been observed in auroras by several groups. os) It is variable in occurrence and was not positively identified until the great red aurora of 11th February 1958 when Wallace(20) resolved the two lines. Its weakness in ordinary auroras suggests that deactivation is severe and there seems little possibility that significant radiation can occur in the normal dayglow. However if (20) is responsible for suppressing the emission there may occur an enhancement of the u’ = 1 progression of the Meinel band system, a possibility to which we shall return in $4. The radiative lifetime of the 2P112metastable state is 5 set and of the 2P3,2 metastable state 4 set and significant radiation may arise from the 2P states,(l) the computed overhead intensity at noon in the absence of deactivation being 1.5 kR. The 2P states will undergo some deactivation by impacts with the ambient electrons e + 0f(2P) + e + 0+(2D, 4S). We adopt
the electron
density
distribution
listed in Table
1 and a rate coefficient
(21) tc for
A. DALGARNO
952
and M. B. MCELROY
(21) of 1 x IO-’ cm3 secf1.06) The effect of electron impact deactivation on the luminosity altitude profile of radiation emitted by 0+(2P) is shown in Fig. 1. Electron impact deactivation reduces the overhead intensity from 1.5 to 1.3 kR. The 2P-2D multiplet at 217319-7330 has been observed in auroras(17,20,21) and Chamberlain has argued that the observations indicate some deactivation, occurring presumably through reactions similar to (19) and (20). The effects of a deactivation coefficient of lo-lo cm3 set-l in collisions with N, on the altitude profiles of radiation emitted 600-
100 10-z
I IO“
I I
I IO
I 102
I IO"
FIG. 1. THE ALTITUDE PROFILE OF FLUORESCENT EMISSION AT NOON FROM THE 0+(2P-sD) TRANSITION. FROM RIGHT TO LER, THE CURVES SHOW THE INCLUSION OF ELECTRON IMPACT DEACTIVATION AND MOLECULAR NITROGEN IMPACT DEACTIVATION.
from the 2P states are illustrated in Fig. 1. The emission at low altitudes is suppressed and the peak intensity occurs at 200 km instead of 150 km. The overhead intensity is reduced from 1.3 to 0.5 kR. Because of the uncertainty in the deactivation coefficient, accurate predictions of the intensities of the four lines of the 2P-2D multiplet are not possible. The intensities given in Table 5 correspond to no deactivation by N, and to a deactivation coefficient of lo-lo cm3 se&. They are consistent with an earlier estimate.(l) 3.2 Molecular nitrogen ions The excited N,+(2C,+) molecules produced by (9) will, if stable, decay rapidly to the B%;,+ and ASIA, states. The resulting emission is concentrated in the ultraviolet region of TABLE 5. INTENSITIESAT NOON
Transition
2P11,-2D312 =PI,B-~&z =PsIB-~D~IE =Ps/a-=Dm
Wavelength
1329.9 7318.6 7330.7 7319.4
OF THE O+PP-BD)
LINES
Intensityin rayleighs No deactivation N, deactivation 280 190 280 520
100 70 120 230
THE FLUORESCENCE
OF SOLAR IONIZING
RADIATION
953
the spectrum and it will be reabsorbed by 0, and 0,. We assume arbitrarily that the cascading is primarily to the A2111, state in which case its population is increased by about 12 per cent. The A2111, and B%,+ states both decay rapidly by allowed transitions giving rise respectively to the infra-red Meinel system and the first negative system. Deactivation is unimportant and the rates of population shown in Fig. 2 are also the rates of emission of 600-
500 -
E x
400-
:: =I c 2
300 -
200~-
‘O&e
I 10-s
I 10-a
I 10-3
I 10-Z
I IO“
I
I
I
IO
I
I@
I IO’
FIG. 2. THERATESOFPOPULATION AT NOONOFTHEUPPERLEVELS AZ& AND B*C,+ OFRESPECTIVELY THE MEINEL AND THE FIRSTNEGATIVEBAND SYSTEMS OF NITROGEN.
the two systems. The overhead intensities are shown as functions of altitude in Fig. 3. Above 120 km the intensity of the Meinel system is 9 kR and the intensity of the first negative system is 600 R. It is not possible to predict with confidence the distributions of intensities within the band systems beacuse a significant contribution to photoionization may come from autoprinciple. ionizing processes”) which do not necessarily conform to the Franck-Condon If autoionization is not important the vibrational distributions are governed mainly by the Franck-Condon factors(22) and the intensity distribution in the first negative system should approximate that of Table 6. The intensity distribution in the Meinel system is less well determined but perhaps 3 kR will appear in the (2,0) band at 7850A, l-5 kR in the (3,l) band at 808OA and 750 R in the (4,2) band at 8321 A. 3.3 Molecular
oxygen
ions
The unidentified doublet and quartet states of O,+ will decay rapidly by allowed transitions in the ultraviolet or they may dissociate into 0 and Of. The cascading transitions will increase the populations of the lower lying levels of O$ but in no case will the increase exceed 10 per cent and we exclude the high lying states from further consideration. The c4&- and b4E,- states produced respectively by (14) and (13) will radiate by allowed transitions. The former gives rise to the Hopfield emission bands c4XU- - b4C,in the ultraviolet(23) and the latter to the first negative band system b4X,- - hIIl, in the
A. DALGARNO
954
and M. B. MCELROY
region of 6000 A. Deactivation is unimportant and the rates of population shown in Fig. 4 are also the rates of emission of the two systems. The overhead intensities are shown as functions of altitude in Fig. 5. Above 120 km, the intensity of the Hopfield emission bands is 600 R and of the first negative system 2 kR. The vibrational distribution in the first negative system which occurs if the FranckCondon principle is applicable (see however Schoen”)) is given in Table 7.
Fro.
3. THE OVERHEAD
TABLE
INTENSITIES OF THE MEINEL NITROGEN.
AND
FIRST NEGATWE
BAND
SYSTEMS OF
6. OVERHEADINTENSITIES AT 12Okm OF THE FIRST NEGATIVE SYSTEM OF N,+
DUE
TO FLUORESCENCE Intensity
Band
Wavelength
(rayleighs)
090 OJ OF2 093 1,o 191 192 133 194
3914.4 4278.1 4709.2 5228.3 3582.1 3884.3 4236.5 4651.8 5148.8
720 100 25 5 25 20 20 15 2
The A211, state produced by (12) decays to the ground state with the emission of the second negative band system in the ultraviolet. The luminosity profile is illustrated in Fig. 4 and the overhead intensity in Fig. 5. Above 120 km the overhead intensity is 420 R. The &III, state produced by (11) and by cascading from the b4Z,- state is metastable. It is probably severely deactivated by electron impacts and by collision processes such as atom-atom interchange O,‘(d’II,) + O(V) -+ O(V) + o,+(xW,) (22)
THE
FLUORESCENCE
OF SOLAR
IONIZING
RADIATION
955
FIG. 4. THE RATES OF POPULATION AT NOON OF THE UPPER LEVELS &,,-,bPC,- AND APIIu OF RESPECTIVELY THE HOPFIELD AND THE FIRST AND SECOND NEGATIVE BAND SYSTEMS OF OXYGEN.
FIG. 5. FROM RIGHT TO LEE THE CAPTIONS SHOW THE OVERHEAD INTENSITIES OF THE FIRST NEGATIVE, THE HOPFIELD AND TKE SECOND NEGATIVE BAND SYSTEMS OF OXYGEN.
and little emission is to be expected. With our (arbitrary) assumptions about the individual photoionization cross sections, the integrated rate of production of metastable molecules above 120 km is l-5 x lo9 cm-2 set-l compared to a total O,+ production rate of I.3 x lOlo cm-2 set-l. Accordingly metastable rPn, O,+ molecules may play a significant role in determining ionospheric composition in the E- and F-regions. 4. OTHER
EXCITATION
MECHANISMS
The excited ionic states may be populated by other mechanisms in addition to the direct Collisions between the fast photoelectrons produced by absorption of solar radiation.
A. DALGARNO
956
and M. 3. MCELROY
TABLE 7. OVERSIEZ~ INTENSITIESAT 120 km SYsmhi OF o.+ Band 090 OJ 092 120 1,l 132 2,Q 2s 292
Wavelength 6026 6419 6856 5632 5973 6351 5296 5.598 5926
OF THE FIRST NEGATIVE Intensity (rayleighs)
-
180 200 150 420 30 45 7.5 60 45
photoionization and the neutral atmospheric particles constitute a source of excitation which may be significant. Detailed calculations will be required in general but an estimate of the electron impact contribution to the first negative system of nitrogen can be readily obtained using the known efficiency with which electrons absorbed in air produce 3914A emission.@) Electron impacts are responsible for about 450 R above 120 km. It may be shown that as in aurorasoO) electron impact with the positive ions gives a negligible cont~bution despite the large threshold cross sections. Resonance scattering of solar radiation by the positive ions must be a major source of excitation if the upper state is accessible by an allowed transition from the ground state. Thus resonance scattering by Nzf ground state ions contributes to the Nz+ first negative and Meinel band systems and resonance scattering by O,+ ground state ions contributes to the O,+ second negative system. Resonance scattering by O,fdiII, metastable molecules may contribute to the O,+ first negative system. The O-O band at 3914 A has been observed in the day airglow by Wallace and Nideycz5) and by Zipf and Fastie,(26) who have shown that with plausible assumptions about the N2+ density distribution the measured intensities of 5 kR at a solar zenith angie of 76’ 25 and of 7 kR at a solar zenith angle of 60’ tz5)can be explained by resonance scattering. The predicted intensity from fluorescent scattering and from electron impact is several hundred rayleighs for the solar angles obtaining during the measurements but their contributions are detectable in principle by a detailed analysis of the luminosity profile low in the atmosphere where there are few Nz+ ions and by the Swings effect.(27,Z5) Zipf and Fastie(26) have concluded from their observations that the efficiency with which photoionization of nitrogen produces a 39148 photon is at most 5 per cent. Our predictions at 120 km, based on Schoen’s data(‘) and assuming no contribution by downward transitions from the %,+ state, are equivalent to an efficiency of 2.7 per cent for photoionization and an efficiency of 3-7 per cent for the associated electron impact ionization. The un~rtainties in the composition of the neutral atmosphere are such that there is not necessarily any discrepancy. Specific collision processes by which an ion is converted into a different species may lead to selective enhancements of spectral features. Only one appears to be of interest for ionic emissions. If all the 0+(20) ions were removed by 0+(20) + N,(XlC,+, u = 0) -+ 0(3P) + Nz+(A211,, 21= 1)
(23)
4 kR of radiation would occur in the Meinel bands originating in the v = 1 level, comparable to that anticipated from fluorescence. The altitude distributions of the two sources
THE FLUORESCENCE
OF SOLAR IONIZING
would differ significantly, fluorescence tending (23) tending to follow the oxygen distribution.
to follow
RADIATION
the nitrogen
957
distribution
and
Acknow[edgments-Contribution 114 from the Kitt Peak National Observatory operated by the Association of Universities for Research in Astronomy under contract with the National Science Foundation. The work was uartlv suouorted bv the GCA Coruoration under contract with the National Aeronautics I and Space Adr&nist;atid~. * REFERENCES 1. A. DALGARNOand M. B. MCELROY,Planet. Space Sci. 11, 727 (1963). 2. A. DALGARNO,R. J. W. HENRYand A. L. STEWART,Planet. Space Sci. 12,235 (1964). 3. A. DALGARNO,M. B. MCELROYand R. J. MOFFETT,Planet. Space Sci. 11,463 (1963). 4. R. E. HUFFMAN,Y. TANAKAand J. C. LARRABEE, Disc. Faraday Sot. 37, 159 (1964). 5. G. R. COOK, B. K. CHINGand R. A. BECKER,Disc. Faraday Sot. 37, 149 (1964). 6. J. A. R. SAMSONand R. B. CAIRNS,J. Geophys. Res. 69, 4583 (1964). 7. R. I. SCHOEN,J. Chem. Phys. 40,183O (1964). 8. D. R. BATES,Proc. Roy. Sot. A 253, 451 (1959). 9. M. B. MCELROY, Contribution from the Kitt Peak National Observatory, No. 55 (1964). 10. H. E. HINTEREGGER, L. A. HALL and G. SCHMIDTKE Space Research V (editor P. Muller). North Holland, Amsterdam (1965). 11. M. B. MCELROY,Planet. Space Sci. 13, 403 (1965). 12. H. E. HINTEREGGER, J. Geophys. Res. 66, 2367 (1961). 13. L. A. mu, W. SCHWFIZERand H. E. HINTEREGGER, J. Geophys. Res. 70, 105 (1965). 14. L. A. HALL, W. SCHWEIZER,L. HEROUXand H. E. HINTEREGGER, Astrophys. J. 142, 13 (1965). 15. M. COHENand A. DALGARNO,Proc. Roy. Sot. A 280,258 (1964). 16. M. J. SEATONand D. E. OSTERFIROCK, Astrophys. J. 125, 66 (1927). 17. A. OMHOLT,J. Atmos. Terr. Phys. 10, 324 (1957). 18. D. M. HUNTEN,Ann. Gkophvs. 14, 167 (1958). 19. J. W. CHAMBERLAIN, Phys’ici of the Au&a and Airglow Academic Press, New York (1961). 20. L. WALLACE,J. Atmos. Terr. Phvs. 17. 46 (1959). 21. M. DUFAY, knn. geophys. 15, 1$4 (19i9); k. V~LLANCEJONES,Canad. J. Phys. 38,453 (1960). 22. R. W. NICHOLLS,Ann. geophys. 20, 99 (1964). 23. F. J. LE BLANC,J. Chem. Phys. 38,487 (1963). 24. A. DALGARNO,Ann. geophys. 20, 67 (1964). 25. L. WALLACEand R. A. NIDEY,J. Geophys. Res. 69, 471 (1964). 26. E. C. ZIPF and W. G. FASTIE,J. Geophys. Res. 69, 2357 (1964). 27. A. VALLANCEJONESand D. M. HUNTEN,Canad. J. Phys. 38,458 (1960). Pe3IOM+BaTOti pa6OTe J(aIOTCFIBbIVEZCJIeHHR AHTeHCI%BHOCTH IIOJIf~eHHOFO CBe=IeHHR He6a, Bo3HHKaIoIqme OT @Ioypec~eHqm comevHofi Hom3i4pymueti paEnaIxm. npeABEIaeHHafI 3eHlXTHaFI I'IHTeHCHBHOCTb Ham 120 KM JllJHHi'i0+(2P-2D) IIpIi ill, 7319-7330, cncTeMa nonoc MeikHem Nz+,nep3a~ HeraTmHaB cmTeMa Nz+,nep~aa HeraTHBHaR CMCTeMa o,+, BMACCROHHBR CRCTeMa rOII$UIbj.(a o,+, -COCTaBJIFIlOT, CooTBeTcTBemro, 500 R-l kR, 9 kR, 600 R, 2 kR, 600 R and 400 R.