Chemical reactions at the surface and in the atmosphere of Venus
Item Type text; Dissertation-Reproduction (electronic)
Authors Sill, Godfrey Theodore, 1931-
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 07/10/2021 03:28:24
Link to Item http://hdl.handle.net/10150/565350 CHEMICAL REACTIONS AT THE SURFACE
AND IN THE ATMOSPHERE OF VENUS .. Godfrey Theodore Sill
A Dissertation Submitted to the Faculty of the
DEPARTMENT OF GEOSCIENCES
In Partial Fulfillment of the Requirements For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
1 9.7 6 THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
I hereby recommend that this dissertation prepared under my direction by ______Godfrey Theodore Sill ______e n title d Chemical Reactions at the Surface and in the Atmosphere
of Venus ______be accepted as fulfilling the dissertation requirement of the degree o f ______Doctor of Philosophy ______
Dissertation Director Date
After inspection of the final copy of the dissertation, the following members of the Final Examination Committee concur in its approval and recommend its acceptance:-
4 / 7/'
_ L ______
/ ______u ? / This approval ancjKy&cceptance is contingent ofi the candidate's adequate performance and defense of this dissertation at the final oral examination. The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination. STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrow ers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is m ade. Requests for permission for extended quotation from or re production of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgment the proposed use of the material is in the in terests of scholarship. In all other instances, however, permission must be obtained from the author.
SIGNED: DEDICATION
I am pleased to dedicate this work -to the fond memory of Dr. Gerard P. Kuiper, who first encouraged me to explore the mysteries of the planet Venus. ACKNOWLEDGMENTS
I appreciate the lengthy and encouraging discussions over the .
years with Dr. Bartholomew Nagy, who consented to direct this disser
tation after the death of Dr. Gerard P. Kuiper.
I also wish to thank Dr. Michael Drake for his incisive com
ments and discussions during the preparation of this work. I appreciate
the time and help given me by Dr. Bradford Smith, Dr. Paul Damon, and
Dr. Spencer Titley. I wish to thank Dr. Austin Long of the Department
of Geosciences for his data on the production rate of caliche in arid en vironments and Dr. Lloyd Wallace of Kitt Peak National Observatory for his interpretation of the UV. spectrum of Venus. I am most grateful for the inspiration and understanding of my wife, Dr. Laurel Wilkening.
This work was supported by NASA grant NGL 03-002-002.
iv TABLE OF CONTENTS
Page
LIST OF ILLUSTRATIONS ...... vii
LIST OF TABLES...... x
ABSTRACT ...... xi
1. INTRODUCTION...... ; ...... 1
Composition of the Atmosphere of Venus...... ' IX 1 Temperature and Pressure of the Atmosphere ...... : ...... 5 Albedo and Polarization ...... 10 Geochemical Considerations ...... 14
2. HOMOGENEOUS GAS EQUILIBRIUM REACTIONS ON VENUS ...... 28
Gaseous Constituents in the Atmosphere ...... 28 Terrestrial Volcanic Emanations as an Analog for Venus ...... SB.- Partial Pressures of Gaseous Components in the . . Atmosphere Determined by Homogeneous Equilibria . . . . ; 42 Carbon Equilibria . ...> ...... >...... '.-.^43- Sulfur Equilibria ...... , :.!.VV'46’ Nitrogen Equilibria ...... ' 66
3. HETEROGENEOUS EQUILIBRIUM REACTIONS ON VENUS ...... 76
The Surface Composition of Venus ...... 77 Carbon Dioxide .Buffer. Reactions ...... V 78 The Oxygen Fugacity of the Venus Atmosphere ;; .. . . . 86 Sulfur Buffers . ,...... '...... ' VO%/'... 92 Water Vapor Buffers'...... 4 ...... ■...... yWyLOl Hydrogen Halide Buffers ...... ^ ; vl03 Non-ideal Conditions...... 4 .... . H 05 ' CO 2 v Buffets ...... 4 ...... , ...... ;:il0 5 HgO Buffers ...... „ ...... 110 Halide Buffers ...... '. ., ...... 112
4. SULFURIC ACID IN THE VENUS CLOUDS ...... V ..... 117
Current Data on the Venus Clouds...... 117 Properties of Sulfuric A cid...... 118 H2O Vapor Pressure ...... 119 Index of Refraction ...... 120
v ' v l TABLE OF CONTENTS—Continued
Page
Spectral Properties ...... 121 Ultraviolet Spectrum ...... „ . ... .124 Mid-infrared Spectrum ...... 129' Chemical Production of the H 2SO4 Clouds ...... u . 131 Laboratory Simulation of H 2SO4 Production , ...... 140 Ultraviolet Dark Markings . . 141 Conclusions ...... 146
EVOLUTION OF THE VENUS ATMOSPHERE ...... 148
Surface Temperatures of Planets ...... , . . , . . . . . 149. Radiation Cooling Of a Planet's Surface! ...... 152 Hypotheses for the Evolution of the Venus Atmosphere. . . . . 155 Urey s Mode 1. * . . . * ...... * . . . . . 15 5•. Rasool and DeBergh's Model ...... 156 A New Hypothesis for the Evolution of. the Venus ; . ; Atmosphere ...... , ....160%:' The Problem of the Removal of Oxygen ...... 160 . Cold Trap on Venus ....•* ...... v yv.f. .;v ;.:r .164;:''' ■ Chemical Weathering by CO 2...... >•«..'...... 168 Water Vapor in the Evolving Atmosphere ...... ; . . . ■17d;;v ' Production of the Dense Atmosphere ...... ,...... 171 Summary ...... • . . 172
CONCLUSIONS ...... 174
APPENDIX I: REACTION RATES AND EQUILIBRIUM ...... 181
APPENDIX 2: DERIVATION OF EQUATION FOR EFFECTIVE TEMPERATURE OF A PLANET ...... 184
REFERENCES...... v . 186 LIST OF ILLUSTRATIONS
Figure Page
1. Temperature-pressure profile of . the Venus atmosphere ...... 9
2. Bond albedo of Venus from 0.2 to 4.0 pm ...... 12
3 . Evolution of the Venus atmosphere from hydrogen-rich to hydrogen-depleted state according to Dayhoff et_al. (1967)...... 19
4. Spectral reflectance of ferrous chloride dihydrate from 0.2 to 4.0 jum . . ^ . 26
5 . Equilibrium partial pressures of the gases in the reaction 2CO(g) = CO2 (g) + C(graph) as a function of temperature • from 300 to 750 K in the Venus atmosphere ...... 45
6. Equilibrium partial pressures of the gases in the reaction 82 (g) + 4002(g) = 2802(g) + 4CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere . . 49;
7. Equilibrium partial pressures of the gases in the reaction SO2(g) + C02(g) = SO3(g) + GO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere . 50
8 . Equilibrium partial pressures of the gases in the reaction s 2(g) + 2002(g) = cos(g) + S02(g) + 0 0 (g) as a function of temperature from 300 to 750 K in the Venus atmosphere ... i ...... '>• . . . . 52
9. Equilibrium partial pressures of the gases in the reaction COS(g) + 2C02(g) = 302(g) + 3CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere . . 53
10. Equilibrium partial pressures of the gases in the reaction 2COS(g) = 2CO(g) + 82 (g) as a function of temperature . from 300 to 750 K in the Venus atmosphere ...... i. . . . ' 54
11. Equilibrium partial pressures of the gases in the reaction 38 2(g) + 4002(g) = 4COS(g) + 2S02(g) as a function of temperature from 300 to 750 K in the Venus atmosphere .. 56
v ii v m
LIST OF ILLUSTRATIONS—Continued
Figure Page
12. Equilibrium partial pressures of the gases in the reaction 8 COS(g) = 8 CO(g) + Sg(g) as a function of temperature from 300 to 750 K in the Venus atmosphere . . , ...... „ 57
13. Equilibrium partial pressures of the gases in the reaction 3Sg(g) + IGGOgfg) = 16COS(g) + 0 88 2 (g) as a function of temperature from 300 to 750 K in the Venus atmosphere ...... 59
14. . Equilibrium partial pressures of the gases in the reaction 3S (s,i) + 2002(g) = 2COS(g) + SO2(g) as a function of temperature from 300 to 750 K in the Venus atmosphere . . 61
15. Equilibrium partial pressures of the gases in the reaction • H2S(g) + 0 0 2 (g) = COS(g) + H2 0 (g) as a function of temperature from 300 to 750 K in the Venus atmosphere ... 62
16. Equilibrium partial pressures of the; gases in the reaction .. H2S(g) + 0 0 2 (g) = COS(g) + H2 0 (g), when the partial pressure of H2O is fixed, as a function of temperature from 300 to 750 K in the. Venus atmosphere ...... 64
17. Equilibrium partial pressures of the gases in the reaction 1 v 2H28 (g) + 2002(g) = &2(g) + 2H2 0 (g) + 2GO(g) as a function of temperature of 300 to 750 K in the Venus atmosphere ...... 65
18. Equilibrium partial pressures of the gases in the reaction H2S(g) + 3002(g) = S 02(g) + H20(g) + 3CO(g) as a . function of temperature from 300 to 750 K in the Venus atmosphere ...... 67
19. Equilibrium partial pressures of the gases in the reaction 2NH3(g) + 3002(g) = N2(g) + 3H20(g) + 3CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere ...... 69
20. Equilibrium p'artial pressures of the gases in the reaction 2NH3(g) + 3COS(g) = N2(g) + SHgS(g) + 3CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere ...... 71
21. Equilibrium partial pressures of the gases in the reaction 4NH3(g) + 3802(g) = 2N2(g) +6 H2 0 (g) + 3S(s,i) as a function.of a temperature from 300 to 750 K in the Venus atmosphere ...... 72 lx LIST OF ILLUSTRATIONS - -C o n tin u ed
Figure Paige
22 o Equilibrium partial pressures of the gases in the reaction 2NH3 (g) 4- S0 2 (g) = N2(g) + 2H2 0 (g) + H2S(g) as a ; : function of temperature from 300 to 750 K in the Venus atmosphere . , , ...... 74
23. Equilibrium partial pressures of the gases in the. reaction 2NH3(g) + 3S (s,/) = N2(g) + 3H2S (g) as a function of temperature from 300 to 750 K in the Venus atmosphere . . 75
24. Partial pressure of CO2 in equilibrium with various minerals at 750 K ...... 6 . 108
25. Transmission spectrum of 88 % H2SO4 solution from 0.2 to 4 .0 jam...... 123
26. Bond albedo of Venus from 0.2 to 0.6 pm ...... 126
27. Transmission spectrum of HBr + Br 2 dissolved in sulfuric acid from 0.2 to 0 .6 pm...... 127
28. Transmission spectrum of SO 2 dissolved in sulfuric . acid from 0 .2 to 0 .6 pm ...... 128
29. Transmission spectrum of sulfuric acid and the power spectrum of Venus, 4 to 16 pm. . ^ ...... 130.
30. Pressure of H2O in the Venus atmosphere and the vapor pressure of sulfuric acid solutions ...... 133
31. Transmission spectrum of various species of bromine from 0.2 to 0,6 pm ...... 143
32. Vapor pressures of hydrobromic acid solutions and of ice at various temperatures...... 144
33 . Surface of Venus early in its history under 10 g of H2O and 10 g of 002 per cm2 ...... 167 LIST OF TABLES
Table Page
1. Spectroscopically determined composition of the Venus atmosphere. . ... „ ...... 30
2. Results of chemical analyses of the Venus atmosphere from Venera 4...... 31
3 . Analysis of gases from Kilauea Crater ...... 37
4. Log P (atm) of various species at equilibrium ...... 48
5. Chemical compositions of various rocks (weight'%)...... 79
6. Mineralogical compositions of various rocks (weight %) . . . 79
7. Abundance of gaseous species at the surface of Venus .... 100.
8 . Mixing ratios of hydrogen halides determined by mineral l^uffe rs ■ ...... # * ...... 115
x ABSTRACT
The atmosphere of Venus differs from that of Earth: mostly car bon dioxide at a pressure of 100 atm and temperature of 750 K. The temperature-pres sure profile of the Venus atmosphere is known;. certain minor constituents of the atmosphere have been identified, both by spec troscopy and by direct sampling conducted by spacecraft.
Chemical thermodynamics can be employed to determine the relative abundance of various gases in the presence of high-temperature, high-pressure CO2 . The sulfur gases , SO2 , COS , HgS, 82 , and Sg, are all stable in the Venus atmosphere. SO 3 is restricted . The pressure of ammonia detected by Venera 8 is compatible with COS and H 2S , but it is not compatible with certain partial pressures of SO 2 gas and sulfur liquid.
At the surface of Venus, heterogeneous equilibria with minerals determines the partial pressures of various gases. Carbon dioxide is probably buffered by silicate equilibria involving forsterite, diopside, and anorthite. The partial pressures of other gases are dependent on the oxygen fugacity at the lower atmosphere. The fugacity buffered by the magnetite-hematite equilibrium is 10 ~ ^ atm. Anhydrite in equilibrium with SO2 (10-5 atm) is stable under fQg = 10.-18 atm. If the oxygen is buffered by quartz-fayalite-magnetite, fQg ~ 10-25 atm. Under these conditions, pyrite or troilite in equilibrium with COS and H 2S (10-2 atm) is stable . H2O vapor is buffered by tremolite and certain sheet structure
x i silicates as phlogopite and chlorite „ Hydrogen halides are buffered by similar sheet silicates and possibly by plagioclase or carbonate.
The clouds of Venus are a dense layer 35 km thick. The author proposes that the clouds are composed of droplets of sulfuric acid solu tion of 86 percent by weight composition. Sulfuric acid satisfies cer tain criteria: refractive index, low vapor pressure of HgO, and infrared absorption properties. It is proposed that sulfuric acid is produced by the catalytic action of HBr: HBr is photolyzed by solar ultraviolet radia tion into Brg; Brg oxidizes sulfur compounds to SO 3 and is reduced to
HBr; SO3 reacts with water to form H2SO4 . The ultraviolet dark clouds may well be solutions of HBr and Brg that evaporate in the presence of
H2SO4 , a strong desiccant.
Evolutionary hypotheses for the formation of the Venus atmos phere are reviewed. It is difficult to have a high rate of H 2O dissocia tion in the presence of a massive CO 2 atmosphere; consequently, H 2O cannot be removed from Venus under these conditions. The author pro poses a new hypothesis . In the early history of Venus , CO 2 and H 2O freeze out on the cold night side of the planet. When the ice or frost covering on Venus is thick enough, liquid water will be produced in the daytime. CO 2 weathers rock efficiently in the presence of liquid water to form carbonates. After the ice cover on Venus has been photolyzed into H 2 (which escapes) and O 2 (which oxidizes surface minerals),
CO2 accumulates in the atmosphere. A runaway greenhouse effect then, causes all the carbonate deposits to decompose, releasing CO 2 back into the atmosphere. Eventually a massive, dry atmosphere surrounds
Venus. This is its present state. CHAPTER 1
INTRODUCTION
The purpose of this chapter is to review the work of those plan etary scientists who have contributed to our understanding of the planet
Venus. The review is partly historical. It emphasizes those properties of Venus that bear directly upon the chemical processes occurring at the surface and in the atmosphere of the planet. The chemical composition of the atmosphere is of prime importance. Our knowledge of the temperature-pres sure profile of the atmosphere was, until a few decades ago, only rudimentary. The albedo (reflectivity) and polarization proper ties of Venus' clouds severely restrict our choice of chemical substances ■ - > as candidates for the cloud particles. Finally, there is a short review of the work of geochemists who have concerned themselves with processes .. . . occurring at the surface of Venus .
Composition of the Atmosphere of Venus
Galileo's initial telescopic observations of Venus revealed little but the fact that the appearance of Venus mimicked the phases, of the moon. This led Galileo to announce that Venus moved in an orbit about the sun interior to the earth's orbit, observational proof of the Copemi- can heliocentric hypothesis.
Planets that orbit the sun within the earth's orbit on occasion pass directly between the sun and the earth. If such an inferior conjunc tion occurs at the nodes of the planet's orbit, the planet will be seen
I from earth as though projected against the bright disk of the sun. Tele
scopic observations of these "transits" of the planet Mercury indicated
the planet.was airless, because the edge of the planet's black disk is ..
sharply defined against the bright solar background with no evidence of
distortion due to an atmosphere. During the rarer transits of Venus, the
edge of the planet is seen encircled with a bright halo as Venus begins
to cross the solar disk, revealing the presence of a substantial atmos
phere, refracting sunlight around the rim of the planet. Other early tele
scopic evidence for an atmosphere on Venus is found in the "joining of
the cusps" at inferior conjunction. In this case , Venus passes above, or
below the sun at inferior conjunction, but instead of showing a thin
crescent along half of the planet's circumference (as the moon does at
new moon) the cusps of the crescent of Venus are joined together to com
pletely encircle the planet's disk. The phenomenon is caused by refrac
tion of sunlight through Venus' atmosphere.
Early telescopic observations of featureless, bright, yellow-
white image of Venus led early astronomers to believe they were observ
ing a planet enshrouded by dense clouds . No surface features were dis
covered, and detection of cloud structure was a tenuous observation at .1 . . best. Consequently, the rotation of the planet could not be determined.
Telescopic spectroscopy in the twentieth century finally led to
the discovery of atmospheric components on Venus, when carbon dioxide
was discovered by Adams and Dunham (1932). The COg band at 7800 A was noticeably more pronounced in the Venus spectrum than in a solar
spectrum. (A solar spectrum obtained at the earth's surface necessarily
includes features due to gases in the earth's atmosphere.) The absorption lines due to oxygen and water vapor appeared identical in the
Venus and solar spectra, indicating these two gases were less abundant . on Venus than on earth. Further investigation into the quantity of carbon dioxide on Venus was pursued by Kuiper and his associates (Kuiper,
1947, 1949, 1962). Using more sensitive bands of GO2 in the near infra red ( 1- 2.5. jam), they determined the quantity of CC >2 to be approximately one km-atm.
The presence of water vapor on Venus was first announced by a team of investigators at Johns Hopkins University, who utilized a spec trograph carried to high altitude by balloon (Strong, 1960; Bottema,
Plumer, and Strong, 1964). They reported the presence of 19 pm of pre- cipitable water above the Venus cloud deck. This result was at variance with data subsequently collected by Kuiper and Forbes (1968) in the NASA
990 Jet aircraft. They determined an upper limot of 2 pm. In the continu ing high-altitude spectroscopy program, Kuiper et al. (19 69) positively . identified Venus water vapor bands at 1.4 pm and 1. 9, pm wavelengths and determined the amount of water vapor to be equivalent to 5 pm of pre cipitable water. The volume mixing ratio, HgO/COg, is 1 x 10~®.
Apparently the quantity of water vapor visible on Venus is variable with time, according to observations made at McDonald Observatory by Schorn et al. (1969). Utilizing high dispersion techniques on the 8300 A line of water, they were able to separate telluric lines of water from the Doppler- shifted Venus lines . On one occasion they could not detect water vapor
(upper limit 16 pm of precipitable water), but on another occasion they determined the presence of 30-40 pm of precipitable water. The highest 4 resolution was obtained by Kuiper et al. (1969), so their results are
probably the most accurate.
Besides the main constituents in the atmosphere, namely, car
bon dioxide and water vapor, other gases were discovered telescopically
utilizing a new means of spectroscopic examination, a Michels on inter
ferometer. Connes et al. (1967) discovered hydrogen chloride and hydro
gen fluoride, with mixing ratios relative to CO 2 of 10~®• ^ and
respectively. These two gases in the upper atmosphere contain hydro
gen roughly equal in abundance to the quantity of hydrogen in water
vapor, perhaps a rather surprising result. Using the same instrument,
Connes et al. (1968) found the gas carbon monoxide with a mixing ratio
of lO- 4 °34. This value is not unsuspected due to the consideration of possible photolytic dissociation of carbon dioxide into CO and O. How
ever, the presence of atomic oxygen has not been demonstrated, nor
molecular oxygen of similar concentration; therefore, it leaves open to possible consideration the fact that the carbon monoxide may hot be photolytically produced but perhaps may be a constituent of the lower . atmosphere even be low the level of the visible clouds.
Besides the spectroscopic examination of Venus previously dis
cussed, there have been in situ measurements of certain constituents in the Venus atmosphere by means of various devices carried by the Venera
space probes, in particular Venera 4, 5, and 6. Some of these devices measured in a semiquantitative manner. For example, there was an.in
strument on board Venera 4 to determine whether oxygen was present. It consisted of a thin wire of tungsten heated to 800 C. It was presumed that the tungsten would be attacked by hot oxygen, analogous to the way in which an electric light bulb burns out. The wire tested as. burnt out;
therefore, a threshold value of 02 > 0.4 percent by volume was presumed.
The test is not exclusively valid or specific to oxygen; any number of
other materials would attack hot tungsten, such as hot oxidizing agents
like nitric and sulfuric acids , sulfur trioxide , and nitrogen dioxide , as
well as chlorine, fluorine, hydrogen chloride, hydrogen fluoride, and
hydrogen bromide . Therefore, one cannot really draw a conclusion about the presence or absence of oxygen from this test. It is even possible that the filament could have been broken before the test was made. Ex periments were performed by Venera 4 to determine the presence of other gaseous substances. Water vapor was measured by two devices and also the total amount of carbon dioxide was measured, as well as the inert
gases in the atmosphere. These tests will be discussed in the next
chapter.
Temperature and Pressure of the Atmosphere
Early attempts to measure the temperature of Venus resulted in values near 210 K, surprisingly low for a, planet so near the sun. The determination was performed by measuring with thermocouples the radi
ation in the infrared band from 5 to 15 jum and fitting the resulting inten
sities to a theoretical black-body curve. Modern measurements using
sensitive bolometers yield values around 225 K (Gillett, Low, and Stein,
1968). Analysis of the rotation and vibration bands of carbon dioxide in the 1-2 pm region of the infrared gave temperatures closer to 250 K
(Kuiper, 1969). All these measurements, however, are from that region of the planet accessible to these wavelengths of radiation, namely, to the top of the cloud deck. In order to see deeper into the atmosphere of the planet, it is necessary either to use wavelehgths that penetrate through the cloud layer or to use special "hot" bands produced by ex cited states of carbon dioxide. Spinrad (1966), analyzing the hot band of
CO2, 5 V 3, derived a rotational temperature of 450 K and a vibrational temperature of 400 K„ These temperatures indicate that the radiation produced in the deeper layers of the atmosphere is capable of penetrat ing outward or that there might be a mechanism for exciting CO 2 to these high effective temperatures, while the true thermal temperatures remain lower. -
Microwave measurements of the brightness temperature of Venus was first performed by Mayer, McCullough, and Sloanaker (1960). At
10.2 cm wavelength, the temperature was 600 + 65 K. Lilley (1961), us ing 21 cm radiation, measured 630 + 130 K. Mayer et a l . (1962) mea sured 548 + 55 K at 3.15.cm wave length, and Tolbert and Straiton (1964) measured 300 + 40 K at 0.32 Cm wavelengthi In general, the shorter the wavelength, the lower the temperature. Apparently, the longer wave lengths were emanating from the deep atmosphere or possibly from the surface. For a while it was uncertain whether the highe temperatures were indeed due to the deep thermal radiation from inside the planet's atmos phere or whether the high temperatures were indicative of an upper atmospheric thermosphere, similar to the earth's, which reaches tem peratures of over 1000 K. One way to resolve the problem is to.measure the microwave radiation from the planet with high spatial resolution, that is to compare the radiation from the limb with that from the center of the planet's disk. If the thermal radiation arises from the surface of the planet, then the center of the disk, as seen from earth or from a space
craft , would indicate the highest temperature because radiation from the
center of the disk would penetrate the least atmospheric mass, with the
least atmospheric absorption and scattering. Radiation from the limb of
the planet would suffer greater attenuation, and the temperature recorded
would consequently be less. This is the phenomenon known as limb
darkening. Conversely, if the radiation emanated from an upper atmos
pheric thermosphere, the exact opposite would obtain. A lesser quantity
of hot gas would be seen in the center of the disk and a greater quantity
observed at shallow or grazing angles of incidence. These effects give
rise to limb brightening. Radio telescopes on earth at that time could
not solve the problem because they could not resolve the center of the
disk from the limb. However, in December 1962, Mariner II flew by
Venus and good measurements were obtained with its microwave radio
meters as they scanned across the planet's disk. As reported by Barath
et al. (1964), the center of the disk had a temperature of 590 K and the
limb 460 K, at a wavelength of 1.9 cm. The pronounced limb darkening
indicated that the lower atmosphere or the surface was the origin of the high brightness temperatures . -
The general trend for measurements of temperature has. been that
as longer wavelengths of light were utilized to measure the temperature
of Venus, the brightness temperatures increased as the wavelength of the radiation increased; namely, the greater the penetration, the higher . the temperature. The final and perhaps definitive measurements of the temperature of Venus are those obtained in situ by the Soviet Venera probes, Venera 4, 5, 6> and 8 . The first probe, Venera 4, measured a 8
maximum temperature of 554 K before its radio transmitter failed (Avduev-
sky, Marov, and Rozhdestvensky , 1969). Veneras 5 and 6 gave readings near 600 K in the upper atmosphere at a height of 20 km, which extrapo
lated to the surface would yield a value near 750 K (Pravda, 1969) „ Final ly, Venera 8 (Marov, 1972) measured a true surface temperature of 747 K with a pressure of 90-95 atmospheres. The spacecraft survived for some
15 minutes on the surface of Venus. These measurements are undoubtedly the most accurate so far obtained. The temperature -pres sure profile of the Venus atmosphere, derived from measurements by Mariner 10 (Howard et al . , 1974) and Venera 8 (Marov et al., 1973), is presented in Figure 1.
In general, there are two extremes of temperature in the atmos phere , where probably most chemical reactions occur. The upper atmos phere, where the clouds are formed, has temperatures of around 210-250
K and is dominated by solar ultraviolet energy. In this case, most chem ical reactions will be those activated photolytically; whereas, in the lower atmosphere with a temperature of 750 K, thermochemical activation energy is available for many chemical reactions, such that the chemical products of a reaction can be determined quite reliably by thermodynamic calculations . It is most likely that the majority of chemical reactions that occur between the rock constituents of the surface and the atmos phere are those that are calculable through thermodynamic equilibrium relations, because at 750 K activation energy should be relatively easy to obtain. For example, the calcite-wollastonite equilibrium reaction has been extensively investigated by Barker and Tuttle (1956):
C aC 0 3(s) + S i0 2(s) = C aS i03(s) + C 0 2(g). (1,1) 9
8 0 -2.5
- 2.0 7 0 Top of clouds
60 -0.5 Mariner 10 50 0.0 E 6
+0.5 CD O
30 + 1.0 Bottom of clouds
20 Venera 8
+1.5
+ 1 .9 2 0 0 300 400 500 700 800600 Temp °K o Venera 4 measurements © Venera 8 NH3 measurements
Figure 1. Temperature-pres sure profile of the Venus etmos- phere
The top and bottom level of the clouds is indicated. Solid line i Mariner 10 profile; dashed line is Venera 8 profile. Venera 4 chemical experiments were performed at the two points indicate; Venera 8 measure ments of ammonia were performed at the points shown. 10
They found that equilibrium was obtained in 42 hours at 883 K and 1 hour at 1063 K, under a partial pressure of approximately 1 atm of H 2O vapor. .
Employing the Maxwell energy distribution function (see Appendix 1), one can calculate the activation energy for this reaction under the catal ysis of the water vapor. The energy required is approximately 1.80 eV
(42 kcal/mole). With this energy, the reaction at 750 K would take
2,160 times longer than at 1063 K, namely,. 90 days. Geologically speak ing, this is a fast reaction. Therefore, at least in this one example, the surface of Venus is hot enough for equilibrium to be obtained quickly.
The presence of certain gaseous components in the Venus atmosphere may facilitate many reactions. Even though water vapor does not appear in the stoichiometry of the above equation, its catalytic presence domi nates the speed of the reaction. Harker and Tuttle also measured the speed of the dry reaction at 1073 K and compared it to the wet reaction at the same temperature. The wet reaction was almost 1,000 times more rapid than the dry. The apparent activation energy of the dry reaction is
2.65 eV (61 kcal/mole). With this activation energy, the reaction at 750
K under dry conditions would take about 15,000 years to reach equilibrium.
Albedo and Polarization
Earth-based observations have furnished valuable data pertain ing to the chemical composition of Venus' atmosphere, its temperature, and pressure. The characteristic reflectivity of the planet has been studied telescopically as well. The first characteristic determined was
Venus' color, namely, yellowish. This is seen in detailed careful mea surements of the albedo of the planet at selected wavelengths. Work 11 that is still definitive in many respects is that of Irvine et al. (1968),
who.observed Venus in the visible range of the spectrum from 0.35 to 1.1 ,
pm. Rocket-borne observations have been performed by Evans (1967) and
Wallace, Caldwell, and Savage (1972). A composite of these observa
tions of albedo is found in Figure 2. It is interesting to note that the
planet has relatively strong absorptions in the ultraviolet, with increas-
ing reflectivity up through the visible spectrum, peaking at 95 % near
6500 A. Further discussion of this reflectivity or color curve will be
found in Chapter 3. Figure 2 does not stress gaseous absorptions, which have fine structure due to the vibrational-rotationak bands, but rather the broad spectral absorptional features produced by solids or liquids.
Only rarely in the visible spectrum does a solid possess sharp spectral features: for example, spectra of some manganese compounds show sharp: electronic transitions . But in general, the features in the visible spec tral range that are due to electronic transitions are very broad, usually extending over hundreds of angstroms. Furthermore, they are additive , that is, if several different compounds are present, their spectrum is the averaged sum of the spectra of each of the components. Venus may have more than one component in its clouds. If the ultraviolet absorp- . tions seen in Venus' spectrum were caused by gases, the fine structure of a gas would be present, especially in ultraviolet photographic spec troscopy. Such ultraviolet gaseous absorbers as sulfur dioxide and ozone, for example, appear to be missing in the Venus atmosphere. In stead there are only broad absorptions, indicating that these ultraviolet absorbers are present in either the liquid or the solid state. 12
IjO
Alb.
0.8
0.6 -
0.4
02
0.0 0.2 0.4 0.50.3 1.0 2.0 3.0
Figure 2. Bond albedo of Venus from 0.2 to 4.0 pm
Gaseous absorptions due to CO 2 are indicated. There is a scale change in wavelength at 0.5 pm. After Sill (1972). ' ' . ■ v 13 Another important property of cloud particles is the polarization
of light reflected from these cloud particles. The degree of polarization
that is produced by light reflecting from a planetary atmosphere and from
cloud particles is a function of a number of parameters, including par
ticle size, chemical composition of the particle (which gives rise to a
particular refractive index), and path taken by the light as it is scat
tered by the particle . The polarization varies with different types of
materials, with different values of refractive index, and with phase
angle. By measuring the amount of polarization and its sign at various
phase angles and different wavelengths, it should be possible to produce
unequivocal determinations of the size and refractive index of the cloud
particles. Initial measurements of the polarization of light reflected
from Venus were those of Lyot (1929) and later by Dollfus (1961). Ac
cording to the data obtained, they, found.that a possible fit to the Venus
polarization curve, at least in the visible part of the spectrum near 5000 0 A, was obtained with water droplets of 2.5 pm size. However, inspec
tion of Lyot's work comparing Venus' polarization data with the polar
ization data for water droplets shows that water, while being a fair
match, is not a perfect, match. Further work was needed to unambiguous
ly determine the refractive index and particle size. This work was pur
sued by Coffeen (1968). By considering the polarization properties at
wavelengths in the infrared, as well as the visible, Coffeen determined i the particles were spherical with a refractive index of 1.40-1.46 and
particle diameter of 2 .5 pm. Further work pursued by Hansen and *■ Hovenier (1974) determined that the refractive index was 1.44 + 0.015
and a particle radius was 1 + .05 pm. These values set strict limits on : ,/ ' 14 the chemical composition of the particles. They are not, however, prop
erties that can unambiguously be assigned to any one particular chem- ,
leal substance, because many chemical substances have similar refrac
tive indices, particularly water solutions of various types of solutes.
The refractive index measurement, therefore, cannot identify a substance,
but a substance that is proposed as a cloud particle and does not have a
refractive index in the permissible limit can be rejected. Any candidate
for the cloud material must have a refractive index within the range, per
mitted as well as absorption properties compatible with the Bond albedo
or spherical reflectivity. It must also be compatible with the other known
constituents in the Venus atmosphere. This approach will be discussed
in the chapter on the composition of the clouds of Venus.
Geochemical Considerations
Although carbon dioxide was discovered in the atmosphere of
Venus in the 1930's, it was not until two decades later that Urey (1952)
gave some consideration to processes on the surface of Venus that would
give rise to the carbon dioxide atmosphere. He held that the amount of
carbon dioxide present in the atmosphere could be due to the reaction of
silicates with carbon dioxide to produce carbonates and silica, or. the reverse reaction. Urey postulated that carbon dioxide in the earth's at
mosphere is removed by reaction with silicates, such as magnesium
silicate, producing carbonates, such as magnesium carbonate, thus ef fectively reducing the carbon dioxide in the earth's atmosphere to pres-.
sures of around 10 -4 atm, the equilibrium pressure at 300 K. In the case of Venus, Urey suggested that the very high pressure carbon dioxide . 15 atmosphere was perhaps due to nonequilibrium conditions. He considered that in its early history Venus, like the earth, removed carbon dioxide by reacting it with silicates in the presence of liquid water. In general, magnesium and calcium silicates would have absorbed the carbon dioxide at temperatures below 50 C and in the presence of liquid water produce large deposits of carbonates, such as limestone. Simultaneously, water was being decomposed in the upper atmosphere and hydrogen was escap ing, thus leading to a gradual depletion of water on the surface of the planet. Later in the planet's history, in processes similar to low and medium grade metamorphism on earth, these rocks were heated and de carbonated to produce the present high concentration of carbon dioxide in the atmosphere. In particular, Urey quoted the wollastonite equilib rium: C aC 0 3(s) + S i0 2(s) = CaSiC>3(s) + COatg) (1;1) calcite silica wollastonite carbon dioxide
Urey considered that the carbon dioxide atmosphere on Venus was simply the result of a process that is common on earth, namely, siliceous limestone decomposes in a metamorphic event to give rise to wollastonite and carbon dioxide . He presumed that this was not an equi librium reaction but one driven to the state of complete decomposition of limestone because of the great abundance of COg in the atmosphere. The above reaction is even now referred to as the "Urey reaction." Under the proper conditions, this reaction can also provide for the removal of carbon dioxide from the atmosphere. At 750 K the equilibrium pressure of the carbon dioxide in the wollastonite equilibrium is 100 atm, according to available thermodynamic data. These, conditions (750 K, 100 atm) are precisely those of the surface of Venus, and this reaction has been called on to regulate the surface pressure of C.O.2 on Venus, although this was not Urey's original intent. The reaction may.not really be apro pos to the removal of carbon dioxide from an atmosphere, if one consid ers that the prime weathering reactions would involve the reaction of z gases with minerals of igneous rocks. Since wollastonite is a metamor- phic mineral, produced by the decomposition of siliceous limestone, and is not generally found in igneous rock, one is forced to consider reac tions with other, more complex calcium silicates, such as diop side, a calcium magnesium pyroxene (CaMgSigOG)- Such minerals give rise to higher equilibrium GOg pressures than does wollastonite. The proper consideration of the wollastonite equilibrium should be as a process which produces CO 2 from siliceous limestone. As the surface heated to
750 K, the limestone decomposed to produce enough carbon dioxide to keep pace with the increase in the surface temperature.
Later, Mueller (1963), using the results of preliminary'work with radio emission from the planet and the new high surface, temperature corresponding to those data (700 K), began to examine the ratio of the partial pressure of carbon dioxide to carbon monoxide-end specified that this ratio probably corresponds to the center, of the magnetite field in the
Fe-O system. He did not overrule the possibility that iron could oxidize to the ferric state, hematite. He also considered the water-gas reaction,
G(s) + H 20 (g) = CO(g) + H2(g), (1, 2) as possibly limiting the amount of water in the known dry atmosphere of
Venus. If the ratio of CO 2 to GO is lO^, he concluded from the above . i - 17 reaction that the ratio of the partial pressures, H 2O/H 2, would be 102 -1 at 700 K.
The presence of free carbon on the surface of Venus could be evaluated by the reaction:
C 0 2(g) + C(s) = 2CO(g). (1,3)
If the ratio of CO2 to CO is lO^, a partial pressure of GO2 of 102.3 is implied by the above equilibrium reaction. Since this pressure of CO 2 is double the actual pressure, Mueller concluded that the atmosphere is too oxidizing for free carbon to exist. A check of Mueller's calculation, however, shows the pressure of COg is miscalculated. He should have obtained an equilibrium pressure of 10®*^ atm. His assumptions were:
PCO2 = 10 atm, PCO2/OCO = 103, hence PqO = 10-2; K= 10~0.6
- • f t
It is doubtful that any conclusion as to the oxidation state of the atmos phere can be derived from his uncertain COg and CO pressures.
Mueller also considered processes for stabilizing the amount of water in the atmosphere. Tremolite decomposing to diop side , enstatite , quartz and water vapor,
C a2Mg5Si8022 (0 H)2(s)= 2CaMgSi206(s) + 3 MgSiOaCs) + S i0 2(s) + H2 0 (g), (1,4) has an equilibrium partial pressure of 0.04 atm HgO. He concluded that tremolite could regulate the partial pressure of water vapor. He examined the equilibria of layer lattice silicates, such as talc decomposing to enstatite, quartz and water vapor,
Mg3Si4O i0(OH)2(s) = 3MgSi0 3 (s) + Si0 2 (s) + H2 0 (g), (1.5) ...... '\X:S:;v ■ ' 18 and calculated that the partial pressure of H 2O would be 2000 atm, great er than the total atmospheric presure of Venus, but he concluded that some micaceous minerals might be stable under Venus-type conditions „
Before the missions of Mariner 5 and Venera 4, Dayhoff et al.
(1967) considered possible constraints on the initial chemical composi tion of Venus due to simple gaseous reactions. The conditions used .for ' the surface of Venus were a temperature of 700 K and pressure of 50 atm.
The atmospheric composition assumed was derived from telescopic spec troscopy with the following composition by volume: Ng, 95%; COg, 2-
3%; HgO, 10_4-io-6; CO, T0- 6o In their scheme they considered a four-component reaction diagram in the form of a tetrahedron in which
C , H, and O constituted the triangular base of the tetrahedron and nitro gen the apex. For most cases, however, they concentrated on the base of C, H, and O. See Figure 3.
Dayhoff et a l. attempted to model the evolution of the Venus atmosphere by observing the effect of varying the initial C/O . The dia gram included separate phase spaces for graphite and asphalt (condensed aromatic hydrocarbons). Evolution of the atmosphere starts with a hydro gen-rich atmosphere at the H corner of the diagram and proceeds to the
C-O line by loss of hydrogen to space. The first model assumed a C/O value of 0.2 (Sue ss and Urey's value, 1956). Evolution of the atmos phere proceeds toward point A. Instead of terminating at point A where '
COg and Og would coexist, the evolutionary track would not cross the
HgO-COg tie line, rather, Dayhoff et al. argue, the oxygen produced at contact with the tie line would oxidize the surface rocks and the evolutionary track Would then follow the HgO-COg tie line to pure COg. 19
C
CO C /0=0.67 c
asphalt C /0 = 0 .4 4
C/O = 0.2 CH ^ graphite
Figure 3. Evolution of the Venus atmosphere from hydrogen- rich to hydrogen-depleted state according to Dayhoff et al. (1967)
Evolution of atmosphere begins at the H-rich corner and pro ceeds toward C—O boundary. Three C/O values are assumed with three evolutionary sequences , depicted by dashed lines and arrows. Dotted lines show paths of two sequences interrupted by intersection with the CO2—H2O tie line . After Dayhoff et al. (1967). 20
The second model assumes an initial C/O value of 0.44 (Cameron's,
1969, value). In this model the evolutionary track passes into, then out
of, the graphite phase space. Dayhoff et_al. state that graphite could
have been present on the surface of Venus at one time, provided the ac
tivation energy for the formation of graphite were supplied. Like model A,
this evolutionary track intersects the H 2O-CO2 tie line and proceeds to pure CO2 • The third model presumes an initial C/O value of 0.67 (an
other Cameron value). This track intersects the phase, space of graphite
and remains there, evolving to point C . No oxygen is absorbed by sur
face components and the atmosphere is rich in CO with graphite present
on the surface. This model is rejected, due to the low abundance of CO
in Venus' atmosphere. Dayhoff et al. conclude that it is probable that
Venus' atmosphere began its history with C/O = 0.5, and it evolved to
ward pure CO .2 by simultaneous loss of. hydrogen to space and oxygen to
the surface.
Mueller (1969), however, criticized these conclusions. He
took the new data obtained from Venera 4 of temperature and pressure at
the surface and considered various reactions that would give rise to the
present atmosphere of Venus. His main criticism of Dayhoff et al. was
that they did not consider equilibria with the surface. It is not valid,
Mueller contended, to take the various gaseous constituents and let
them interact with one another to produce a certain type of atmosphere.
Rather,. he said, the primary consideration in determining the atmosphere
is the constituents of the surface. For example, a reaction like
MgSiOa(s) + CaC03(s) + Si02(s) = CaMgSi20 6(s) + C 0 2(g) (1,6)
enstatite + calcite + quartz = diop side + carbon dioxide 21 is one that involves surface minerals of enstatite and calcium carbonate and diopside. The stoichiometry and thermodynamics of this particular reaction are affected by the presence of iron in iron silicates like pyrox ene, so that the equilibrium pressure may vary by a factor of 2, accord ing to Mueller. Some terrestrial metamorphic events produce CO 2 by the above reaction. The same reaction could control the pressure of CO 2.
Mueller contended that the high carbon dioxide pressure does not mean there is more carbon dioxide or carbon on Venus than on earth, but rather that on Venus more CO 2 was released to the atmosphere by reactions such as the wollastonite equilibrium. Without specifying details,
Mueller suggested that the overall oxidation state of the atmosphere was probably due to ferrous-ferric equilibria in the surface rock produc ing oxygen fugacities of around 10"3 atm. He also sugqested that water vapor partial pressure could be regulated by certain hydrated silicates and that gases, such as hydrogen chloride and hydrogen fluoride, could be produced by various equilibrium-type reactions. In regard to hydrogen chloride, for example, Mueller proposed that a reaction like
2NaCl(s) + Al2Si0 5 (s) + 5Si0 2 (s) + H2C(g) = 2NaAlSi308 (s) + 2HCl(g) (1,7) will give mixing ratios of HC1 to CO2 of about 10~3.7 z. using the Venera
4 temperature of 543 K and CO 2 pressure of 20 atm. Similarly, a reac tion such as
CaF2(s) + Si0 2 (s) + H2 0 (g) CaSiOsfe) + 2HF(g) (1,8) could regulate the hydrogen fluoride partial pressure. Mueller's point was that one must be cognizant of reverse reactions that remove the ex cess volatile constituents from the atmosphere, and that the phenomenon of removal of gases by chemical reaction with minerals is just as impor tant as degassing. Surface atmosphere heterogeneous equilibria are in compatible with certain current models of atmospheric evolution like those of Dayhoff et al. (1967).
It is of interest to note that some other investigators, while not considering the atmosphere of the planet Venus as such, nevertheless touched upon the composition of Venus as part of a general consideration of chemical reactions that would occur in various types of planetary at mospheres. So, Suess (1962), for example, considered the atmospheres of terrestrial planets in the process of losing hydrogen and photolyzing methane and water. He stated that with increasingly oxidizing conditions the gases pass through a phase in which certain organic compounds or elementary carbon (graphite) could be thermodynamically stable. By numerically considering equilibrium concentrations of hydrogen, methane, carbon monoxide, carbon dioxide, and water as a function of the total hydrogen pressure, he concluded that under certain conditions one could indeed form carbon species, hydrocarbons and graphite. He stated that the surface rocks of Venus might contain considerable amounts of carbon, even in the graphite state. Such a large abundance of carbon Would give rise to large amounts of carbon dioxide and also explain the virtual ah- sence of oxygen on Venus. In conclusion, Suess stated that if water is / • not a dominant constituent in the atmosphere, then it might be possible that other volatile substances, such as sulfur and sulfur compounds, are responsible for producing the clouds of Venus. At the time Suess wrote the article, there was no information available on the amount of water in the Venus atmosphere; therefore he was not able to judge how much 23
hydrogen escaped from Venus. The question of the contribution of sulfur
to Venus' clouds will be treated in greater detail in the fourth chapter.
The characteristic reflection of microwaves from the surface of
a planet is dependent on the dielectric constant of surface materials.
Pollack and Sagan (1965), examining the phase effect of microwave
pulses reflected from Venus, concluded that some possible surface con
stituents would be quartz, aluminum oxide, wollastonite, periclase,
enstatite , and limonite. Granite rock" itself was compatible with their analysis. Ferromagnetic minerals and aromatic hydrocarbons could not be excluded. These conclusions, as well as some of the conclusions of
Mueller, rested on the assumption that the surface pressure was only 10 atm. Many of these conclusions must be modified due to further knowl edge of the surface pressure.
Undoubtedly the most critical consideration of the surface con
stituents of Venus and also of atmospheric constituents has been made by Lewis (1968a, 1968b, 1970). Lewis (1970) discusses a great number of gaseous and surface-gas reactions that could operate on Venus. He considers three possible models Of the surface of Venus. In one model, the surface temperature is 632 K and pressure 190. bars; the second model has 593 K and 31 bars (based on Venera 4 data, a probe that did not reach the surface); and a third model at 747 K and a pressure of 120 + 20 bars (based on data obtained from Venera 6). The carbon dioxide mixing ratio is — 95%. Lewis considered some 19 carbon dioxide buffer reac tions. These are typical low-grade metamorphic type reactions in which a carbonate, sometimes alone but more often with a silicate, reacts to produce another type of silicate and carbon dioxide . These types of 24
reactions have been specified by workers in the field of metamorphic
geology (e.g., Barth, 1962). Barth considered 14 different metamorphic
reactions producing GOg gas, beginning with the lowest metamorphic
grade and proceeding to the highest metamorphic grade. In general, the
reactions involved a carbonate reacting with silica or a silicate, some
times in the presence of water, to produce other silicates and carbon
dioxide g a s . Lewis (1970) examined these and other similar types of
reactions to determine which reactions produce a pressure of COg at the
temperature that is appropriate to the surface of Venus. He found that
the wollastonite or the Urey equilibrium (equation 1,1) best satisfies
the conditions. Lewis considered six different carbon monoxide buffers.
The first one is the graphite reaction (equation 1,3). Another one in
volved iron in olivine:
3FeMgSi04 (s) + CO2 (g) = 3MgSi0 3 (s) + Fe 304(5) +.CO(g). (1,9)
Other buffer reactions involved iron oxides reacting with CO 2 to produce
GO and a more completely oxidized iron compound, such as:
2Fe304(s) + C 0 2(g) = 3Fe203(s) + CO(g). (1,10)
Lewis considered 10 water vapor buffers, most of which involve mica
minerals, talc (equation 1,5) and amphibole, tremolite (equation 1,4),
decomposing to produce other types of silicates and water vapor. Hydro
gen chloride buffers involve reactions with sodium chloride or a sodium
silico-chloride reacting with silica to produce other types of silicates
and HC1 gas, e.g. , equation (1,7). Hydrogen fluoride gas buffers in volve simple fluorine compounds like calcium and magnesium fluoride
(equation 1,8), and also some fluoromicas producing HF gas. Ten sulfur 25
buffer reactions are listed; some reactions produce reduced sulfur gas:
FeS (s) + M gSi03(s) + H20(g) = FeMgSi04(s) + H2S(g). (1,11)
Other reactions produce oxidized sulfur:
CaS0 4 (s) + C0 2 (g) = CaC0 3 (s) + S0 3 (g). (1,12)
Iron-bearing gas buffer reactions involved oxides of iron reacting with
HC1 to produce either ferric or ferrous chloride:
2Fe304(s) + 18HCl(g) = 6FeCl3(s) + 8 H2 0 (g) + H2(g). (1,13)
There are other types of reactions which Lewis has not considered which may have particular bearing on the surface composition of Venus and the gases in equilibrium with the surface. These reactions will be discussed
later.
This author collaborated with Kuiper (1969) in a paper on the
Venus clouds. In this work, the clouds were tentatively identified as ferrous chloride partially hydrate. Ferrous chloride was chosen as a
constituent of the Venus clouds because of its ability to match the Venus ultraviolet, visible and near infrared reflectance characteristics. It is difficult to match one particular substance with the spectrum of Venus.
This first attempt, using ferrous chloride, gave a near ultraviolet and . visible reflectivity that was a fairly good match to the Venus spectrum
(Fig. 4). In some particular^parts of the spectrum, however, ferrous .
chloride did not produce a good match with the spectrum of Venus,, par
ticularly near the iron absorption centered near 1 pm. The depth of the
absorption, roughly 45%, was much higher than the mild absorption
shown in the albedo of Venus. Likewise the absorption between 3 and 4 pm did not match the Venus reflectivity spectrum. On Venus, the reflec tivity between 3 and 4 pm is below 10%, whereas the water of hydration 26
VENUS 0.8
0.6
0.4 FeCU 2 Ftp
0.2
00 2a ^ l 02 03 0.4 0.5 2.0 30
Figure 4. Spectral reflectance of ferrous chloride dihydrate from 0.2 to 4.0 um
Dashed line: relative reflectance of FeClg' 2H2O versus a white standard (LiF). Solid line: albedo of Venus . Important differences in the spectra exist around 1.0 pm and between 3.0 to 4.0 pm. ' 27 band of ferrous chloride, centered at 3^im, is up to about 50% reflec tivity at 3 .6 jum. This great difference in reflectivity was one of the reasons why the author decided to examine other types of substances with a lower reflectivity in this region. The author investigated the . spectra of various iron compounds of sulfur, particularly hydrated fer rous and ferric sulfate. It was observed that the hydrates gave the re quired absorption band between 3 and 4jim, leading the author (Sill,
1972) to consider the possibility that another type of sulfate was respon sible for the Venus clouds, namely, sulfuric acid itself. This will be discussed in Chapter 4. CHAPTER 2
HOMOGENEOUS.GAS EQUILIBRIUM REACTIONS ON VENUS
The atmosphere of a planet is a dynamic entity, responding to the flux of solar radiation from above and to the flux of thermal radiation from the planet's interior. The atmosphere of Venus experiences a wide range of temperature and p ressu re, from 750 K and 100 atm at the sur face to 250 K and 0.05 atm at the cloud tops. Chemical thermodynamics is employed in this chapter to determine the. homogeneous (gas phase) equilibrium state of the Venus atmosphere, from near the surface to the
298 K level. While most of the reactions considered in this chapter are gas phase reactions, occasionally gases in the atmosphere react to produce solids. These reactions are also treated in this chapter.
Surface -atmosphere heterogeneous equilibria will be examined in Chap ter 3 and photochemically induced reactions in Chapter 4.
In order to discuss gas phase homogeneous equilibria, some of the components of the gas phase must be known, others can be inferred from thermodynamic considerations. Some components have been mea sured by telescopic spectroscopy,, others have been measured in situ by the Venera space probes. The existence of other gaseous components can be argued by analogy with terrestrial volcanic outgassing, in par ticular the existence of sulfur and its compounds. The equilibrium par tial pressures of gaseous components are determined for 18 chemical
• 28 29 reactions at temperatures and pressures that span the Venus atmosphere.
Carbon compounds appear in 15 reactions, sulfur in 16, hydrogen in 8 , and nitrogen in 5.
Gaseous Constituents in the Atmosphere
The abundances of gases that have been determined spectro scopically are given in Table 1. Many are only upper limits. The mea surements are telescopic and are valid for that region of the atmosphere accessible to observations of this type,, namely, the region,above the cloud tops. Of particular interest to note is the very low abundance of water vapor, roughly 1 ppm, much lower than that of earth.
Other measurements have been performed in situ by the Venera missions 4, 5, 6, and 8 . In these atmospheric probes, various chemical testing measurements were employed (Vinogradov, Surkov, and Florensky,
1969; Marov, 1972). Details of the measurements are given in Table 2.
One of the gases measured directly was carbon dioxide , detected pri marily by potassium hydroxide which absorbed carbon dioxide from the atmosphere. This would also absorb water vapor and acid constituents.
The quantity of gas was determined by measuring the pressure difference between two compartments in the space craft, one of which was simply a reference chamber filled with the gas from the atmosphere of Venus. The other chamber had the absorber present in it, and after a sufficient period of time the pressure difference between the two chambers was measured, this pressure difference being interpreted as a measure of the pressure of the gases absorbed. Therefore, since there was roughly a 95% pres sure decrease in the absorbing chamber, it was determined that the COg 30 Table 1. Spectroscopically determined composition of the Venus atmos phere
Gas Mixing Ratio Source
GO 2 0.90 Belton, Hunten, and Goody (1968); Gonnes et al. (1967)
CO 10-4.34 Gonnes et al. (1968)
HC1 . 10" 6.2 Gonnes et al. (1967)
HF 10- 8.2 Gonnes et al. (1967)
H2O 10- 6.0 Kuiper et al. (19 69)
CH4 < 10-6 Gonnes et al. (1967)
CH4CI < 10"6 Gonnes et al. (1967) c h 3f < IQ-6 Gonnes et al. (1967)
C 2H2 < IQ-6 Gonnes et al. (1967) HCN < IQ-6 Gonnes et al. (1967)
02 < 10-5 Belton and Hunten (1968)
03 < 10-8 Jenkins, M orton, and Sweigart (1969)
Br2 <10-7"3 Sill (1972)
S 0 2 <10-7 . 5 Cruikshank and Kuiper (19 67)
COS < 10-8 Gruikshank (1967); Kuiper (1969)
C 3O2 <10-6.3 Kuiper et al. (1967); Kuiper (1969) 00 O r-4 1 V h 2s ■ Cruikshank (1967) LO C^x i 0 1 — V n h 3 1 Kuiper and Gruikshank (1964); Kuiper (1969) Table 2. Results of chemical analyses of the Venus atmosphere from Venera 4—After Vinogradov et al. (1969)
Limits on Results of Analysis Component Type of Method of Measurements Measurements Conditions Determined Cell Measurement (% by volume) (% by volume)
First Group of Analyzers
Height, 26 km COg ' threshold heat conductivity ' threshold 1 >1 Pressure, 550 mm n n amplitude KOH absorption 7-100. Temperature , 25 C 2 90 + 10 n 2 amplitude preliminary absorption of .7-100 <7 0 2 , C 0 2 at 1000 C • 0 2 . threshold W 800 C outburning threshold 0.4 >0.4 . h 2o threshold electroconductivity of threshold 0.1 " > 0.1 moistened P 2O5
Second Group of Analyzers
Height, 19 km C02 amplitude KOH absorption 2-30 >30 Pressure, 1500 mm . Temperature , 90 C 2 . amplitude preliminary absorption of 2.5-50 < 2 .5 O2, CO2 at 1000 C 02(+H2Q) threshold . P vaporizing threshold 1.6 < i .6
h 2o "v , threshold electroconductivity of threshold 0.05 > 0.05 moistened P2Og
h 2o amplitude . CaCl 2 absorption threshold 0.7 < 0 .7 : 32
and whatever else would be absorbed by potassium hydroxide constituted
95% of the atmosphere. The inert ingredients were not directly measured; they could presumably be nitrogen gas and various inert gases, such as argon.
The water vapor measurements were made by two different de vices, one of which was a phosphorus pentoxide electrical conductor, i.e., a moist sample of P2O5 between two conducting platinum wires.
A change in conductivity of the moist P 2O5 indicated absorption of a minimum amount of water vapor. The device was a threshold indicator, giving a detectable increase in conductivity for concentrations of water vapor greater than 0.7 mg HgO per liter of gas. It is well, known that phosphorus pentoxide is a very strong desiccating agent, and the mea surements of the water vapor content are not completely unambiguous inasmuch as any cloud particle or droplet that was present in that cham ber would find itself desiccated by the P 2O5; therefore, the water mea surement that is made is not that simply of water vapor but also any water contained in a cloud particle.
Another type of device that was used was the calcium chloride detector. Again, this was a pressure differential measuring device with a sample chamber containing calcium chloride and a reference chamber with no absorber. Calcium chloride is typically used as a drying agent in many chemical applications. If any water vapor or any gas is ad sorbed by the calcium chloride, there will be a loss of pressure in that compartment and this could be interpreted as the amount of water vapor adsorbed by the calcium chloride. The measurement reported is < 0.7% by volume at or below the threshold of detection. However, there is 33 another phenomenon that could affect this result; namely, calcium chlo ride could also be used to generate hydrogen chloride gas simply by put ting on the calcium chloride a strong acid, such as sulfuric acid. In fact, one Of the simplest laboratory preparations for producing hydrogen chloride gas is the addition of concentrated sulfuric acid to a chloride salt. The gas produced is hydrogen chloride, and from the chemical equation for this reaction,
C aCl2 + H2SO4 = CaS04 + 2HC1, (2,1) one sees that for every mole of sulfuric acid consumed two moles of hydrogen chloride are produced. Therefore, if the gas entering this de tector is not purified of sulfuric acid by some absorbent, such as sodium hydroxide, then it is quite possible that the calcium chloride, while ab sorbing water from the air, could also produce an excess of gas. The lack of a pressure differential between the two chambers admits of am biguity.
Finally, there were measurements of ammonia performed by
Venera 8 (Surkov, Andreichihov, and Kalinkina, 1973), using a simple pH indicator, bromophenol blue, which changes from acid yellow to base blue at a pH of 4 . The indicator was absorbed on anhydrous silica gel, sensitive to gaseous ammonia as well as to other alkalis like NaOH.
(Solid alkalis are dismissed by these investigators as unlikely material in the atmosphere.) The ammonia was detected photometrically by mea suring the relative absorption of light in two chambers, a sample cham ber and a reference chamber. The sample chamber contained a light source, the translucent silica gel with its pH indicator, and a photo resistor to measure the intensity of the light passing through the silica 34 gel. The reference chamber utilized the same light source but with its own silica gel substrate and photoresistor; this chamber remained iso lated from the Venus atmosphere. At two set levels in the atmosphere
(where the pressure was 2 and 8 atm, respectively) the sample chamber was exposed to a known volume of ambient air which passed through the silica gel—pH indicator. A change in the relative signal strength me a- . siired by the two photoresistors was interpreted as a change in the color of the pH indicator in the sample chamber. The experimental results were reported as an ammonia abundance of 0 . 01% to 0 . 1% by volume.
The two measurements were performed while the landing probe was within the clouds. See Figure 1. It was not reported if the ambient gas from the Venus atmosphere was filtered to remove aerosol particles before the air entered the sample chamber.. Solid particles might produce a change in signal strength due to scattering of light, although the effect would probably be minor, since the silica gel substrate is described as translucent, namely a scattering medium. Surkov, Andreichikov, and
Kalinkina (1973) infer that the test is capable of measuring NH 3 gas even in the presence of acid or acid anhydride gases, such as HC1 and
CO2, because their identification of the clouds of Venus is based on a model employing ammonium carbonates (NH 4HCO3, (NH4)2C03) and am monium chloride (NH 4C I). The gaseous NH3 is produced by the decom position of these salts at the ambient temperatures of the two sampling intervals , 360 and 470 K:
NH4HCO3 (s) = NH3 (g) + H2 0 (g) + 0 0 2 (g) (2,2)
(NH4)2C0 3 (s) = 2NH3 (g) + H2 0 (g) + 0 0 2 (g) (2,3)
NH4C1(s) = NHsfe) + HCl(g). (2.4) 35
While the vapor pressure of NH4CI at 470 K is compatible with the mea
surements, the NH 3 vapor pressure at 360 K is two orders Of magnitude
less than the measured abundance (0.01% as the lower limit). The vapor
pressures of the two ammonium carbonates are far in excess of the mea
sured abundance which would imply that the clouds had totally evaporated
at the levels where the measurements were made. However, Lacis and
Hansen (1974) interpret the Venera 8 sunlight measurements as indicat
ing that the clouds extend down to the 10-atm pressure level on Venus,
below the levels of the ammonia measurements.
The abundance of ammonia can be interpreted also as due to the
decomposition of other ammonium compounds, such as ammonium sulfate:
(NH4)2S0 4 (s) = 2NH3(g) + H2 0 (g) + S0 3 (g) / " . (2 .5)
The ammonia partial pressure of the above reaction is such that at 360
and 470 K the mixing ratio in the Venus atmosphere would be 0.005% and
0.3%, respectively, not too dissimilar from the Venera 8 measurement of
0.01%—0.1%. If the clouds of Venus are sulfuric acid, then free am monia would be readily absorbed by the cold acid in the upper atmos phere :
2NH3(g)> H2S0 4 (aq) = (NH4)2S0 4 (aq). ' (2 ,6)
The ammonia detected in the lower atmosphere might be due to the ther mal decomposition of ammonium sulfate; therefore, the. amount of am monia discovered might simply express the relative concentration of the cloud droplets and the ammonium ions present in them. This measure ment, however, is extremely useful because it could provide some indi cation of the relative amount of acid droplets in the lower atmosphere where these measurements were made (40-28 km altitude). Both Moroz (1965) and L a d s (1975) deduced a particle density of 100-300 cm~3 m the atmosphere near the 50 mb pressure level. This density translates to a sulfuric acid mixing ratio versus carbon dioxide of 1 ppm; namely, the droplets of sulfuric acid dispersed or gasified in the GOg would give a mixing ratio of 10^6 versus the GOg. In the case of the measurement of ammonia, the mixing ratio is 10~3- i o - 4 . This therefore means that even if the cloud particles were totally composed of ammonium sulfate, the concentration of the cloud particles has gone up two or three orders of magnitude. Since the cloud droplets are not a 100% solution of ammo nium sulfate but are more likely to be sulfuric acid with some dissolved ammonium sulfate, it seems most reasonable to conclude that the cloud densities in the lower atmosphere where the measurements were made are much greater than 1 ppm, perhaps parts per thousand, parts per hundred, or greater.
Terrestrial Volcanic Emanations ■ as an Analog for Venus
In addition to those gases that have been directly measured, there are other gases that can be considered as probable constituents of the Venus atmosphere. This deduction is based on analogy with the gas eous constituents of the earth's atmosphere, namely, those that are out- gassed during volcanic activity. Common volcanic gases are listed in
Table 3.
The most abundant gas that is found from volcanic emanations is water vapor. In the case of Hawaii, this is undoubtedly due to the. fact that meteoric water seeps down into the region where hot magma causes it to vaporize and become the main constituents (>90%) of the 37
Table 3. Analysis of gases from Kilauea Crater—After Heald, Naughton, and Barnes (1963)
Mole Percentage as Mole Percentage Recalculated Collected to Exclude Air
Component Minimum Maximum Minimum Maximum
H2O 4.7 99.8 97.2 99.1
H2 0.0 0.43 0.00001 0.43 co2 0.00023 3.8 0.80 2.30 CO 0.0 0.086 0.0 0.087
CH4 0.0 0:047 0.0 0.023
h 2s 0.0 0.12 0.0004 0.67 so2 0.0 0.16 0.001 0.43 cs2 0.0 0.026 0.0 0.026
0 2 0.0 21
N-2 0.0 78 . ■ Ar 0.0 0.064
gas. The other abundant gases are the oxides of carbon, sulfur and its compounds, and halide compounds . Very often in Hawaiian volcanoes sulfur and calcium sulfate are deposited on the volcanic rocks surround ing a fumarole. The calcium sulfate is probably produced by the reaction of hot sulfur trioxide or sulfuric acid on calcium-containing rocks, such as basalts. Calcium sulfate does vaporize without decomposition at tem peratures around 1000 C and subsequently condenses near the fumaroles.
Hydrochloric and hydrofluoric acid gases are abundant volcanic or 38
fumarolic emanations, as observed in the Katmai Peninsula, Alaska,
where in the period of one year thousands of tons of HC1 and HP were
outgassed.
The redox state of certain volcanic gases are of particular in-
portance, because it is asserted that the source of their magmas is the
earth's mantle. Hawaiian tholeiitic basalt may have such an origin.
Nordlie (1971) examined the analyses of 18 gas samples collected at a
lava lake in the Halemaumau crater. He assumed that all the samples
had their origin in a deep magma and were in equilibrium with this magma
at depth. As the gas-charged magma rose to the surface, various con
taminants, such as meteoric water, invaded the magma. Furthermore,
at the surface of the lava lake and in the gas sampling process contami
nation with air was unavoidable. Correcting for air contamination as well
as for meteoric water (as far as possible), Nordlie arrived at one initial
composition of gas that was compatible with all the gas samples collected
at the surface. An optimistic viewpoint, he said, would hold that this
mixture of approximately equal proportions of HgO, GOg, and SOg was
the juvenile gas composition of the magma. While this composition ap
pears oxidizing, it is nevertheless compatible with such "reduced" min
erals as magnetite. The equimolar gas mixture of HgO, COg, and SO2
conforms to an oxygen fugacity buffered by quartz-fayalite-magnetite
(QFM). He further stated that reduced species, such as H 28 , H g, and
CO do not have to be dominant. The addition of water to the melt will
not cause it to be oxidizing unless the water concentration is "over
whelming" or unless hydrogen is lost by such as process as diffusion.
Indeed, the presence of hydrogen gas at burning fumes in Kilauea crater 39 has been demonstrated by Cruikshank, Morrison, and Lennon (1973). The estimated H 2 abundance was a few percent by volume
What happens to the gases that come from earth volcanic re gions? In general, most of them interact with either the atmosphere or the surface of the earth, producing various types of compounds of higher chemical stability. For example, hydrogen is oxidized to water and sul fur to sulfur dioxide and sulfur trioxide . SO 3 is a highly reactive acid gas and reacts with the surface rocks to produce sulfates. The most common insoluble sulfates formed are calcium sulfate , either hydrated
(gypsum) or anhydrous (anhydrite). Gypsum is an abundant constituent in earth sedim ents. Pettijohn (1957) estim ated that SO 3 has an overall com position of about half a percent in earth's sedimentary rocks. This quan tity may be a gross underestimate, however, due to the fact that deposits under shallow seas, as for example the Mediterranean, and deposits now found in dry, desert-type environments have not yet been examined to any particular depth for.their gypsum content. Occasionally, gypsum de posits crop out near the surface over large areas giving rise , for example, to the White Sands area of New Mexico. It is also of interest to note that in the Basin and Range province of the Southwest there may be vast undetected deposits of gypsum and other salts below the surface. In the area of Picacho, Arizona, test drillings by oil companies revealed that there were over 7,000 feet of anhydrite deposits in a.region covering many square miles. Around Litchfield Park, Arizona, many thousands of feet Of anhydrite and halite were discovered and the basement had not yet been tapped (Damon, Shafiqullah, and Lynch, 1973) . 40
Volcanic gases have been trapped in sedimentary deposits as carbonates, sulfates, and halides. Many volatile species have been trapped in the earth's oceans in solution. In addition to sodium and chlo ride ions, ocean water contains appreciable amounts of the metallic ions, magnesium, calcium, potassium, and strontium, and among the nonihe- tallic ions, sulfur usually as. sulfate, bromide , and sdihe carbonate.
These soluble salts are formed by the reaction of outgassing materials from magmas reacting with surface rocks . Some of the outgassed materi als are pure acids, for example, the halogen halides: HF, HC1, and pre sumably HBr and HI. Other gases are acid anhydrides, such as GO 2,
SO2 # and SO 3 . The acid anhydrides become acidic by reacting with water.
For example ,
H2OCO + C0 2 (g) -> H2C0 3 (aq). (2,7)
The carbonic acid that is formed is a well-known weathering agent of terrestrial rock. In a series of complex reactions, the acids in water solution react with the metal ions of rock, producing a salt of the metal . and often altering the silicate structure to a hydrated mineral. A simpli fied version of carbonic acid reaction with feldspar might be as follows:
CaAl2Si208 (s) + 2H2C0 3 (g) = CaO iC O sh (aq) + Al2Si0 5 (s) + 8102 (s) + H2OU) (2, 8 )
Al2Si0 5 (s) + S i0 2 (s) + 2H20(i) = Al2Si205(0 H)4(s). (2, 9 )
If the species produced is soluble in water, it is carried by fluvial processes to the oceans where it either remains in solution (e.g.,
Na and K ions) or in others cases, the species might eventually precipi tate (e.g., Ca ions). Therefore, a good measure of the amount of crustal 41
outgassing may be determined from the quantity of volatile and soluble
species now found in ocean water. There has been an abundance of HC1,
002/ 802, and HF outgassed from volcanic sources . The oceans also
contain appreciable quantities of bromide ion and iodide ion. The vari
ous halides should be present in the oceans in amounts comparable to
the ratio of their elemental abundances, with the exception of the fluoride
ion which is underrepresented in the earth's oceans due to the more in soluble nature of its principal compound , calcium fluoride.
The Venus atmosphere has most likely been influenced by sim
ilar acid magmatic emanations. Indeed, the most abundant constituent
in the Venus upper atmosphere is CO 2, followed by GO . There is spec troscopic evidence of HC1 and HF. Therefore, using the analogy of earth outgassing and considering the quantity of these compounds in the earth's oceans, it is likely that species of sulfur as well as hydrogen bromide and hydrogen iodide are present. All of these materials must, however, be in equilibrium with the rocks of the planet at the temperatures and pressures that obtain near the surface of Venus. For example, no HC1 is found in the earth's atmosphere due to the fact that, at the earth's low temperatures, metal chlorides are much more stable than HC1. Soluble halide compounds are found on earth principally in the oceans and. evap- orite deposits. On Venus there are no oceanic sinks for halogen com pounds. The high temperature of the surface, 750 K, favors the produc tion of hydrogen halide species in equilibrium with surface minerals.
Similarly, various species of sulfur, SOg and COS, should be present in the Venus atmosphere, again in equilibrium with their weathered pro ducts on the surface. Other gaseous species that should be present in 42 the Venus atmosphere are H 2 and O 2, produced by the decomposition of
HgO and COg.
Partial Pressures of Gaseous Components in the Atmosphere Determined by Homogeneous Equilibria
The abundance of various components in the atmosphere is de termined by a number of factors, three of which are: (I) the absolute abundance of the constituent elements near the planet's surface, ( 2) the / . chemical reactivity of the component in respect to the minerals of the planet's crust, and (3) the chemical reactivity of the component in re spect to the other gaseous components. For the first factor, the elemen tal abundance of carbon (GOg, 100 atm) and hydrogen (HgO, 0 . 1- 0 .7% by volume) have been established with some degree of certainty , although less so with water vapor, by the Soviet space probes . (Compare Figure 1 and Table 2.) Other elemental abundances are inferred by analogy with
Earth. The second factor will be treated in the next chapter. The third factor is critical; the coexistence of various gases depends on the oxi dation state of the atmosphere. A totally oxidized atmosphere would contain elements in their fullest oxidation state: GOg, HgO, SO 3, NOx , and O 2 . In the state of complete reduction, the gaseous species would be CH4 , H2O , H2S , NH3, and H 2 . Intermediate redox conditions would produce other species . From oxidized to reduced, carbon could exist in the following species: CO2, CO, G (graphite), aromatic hydrocarbons, aliphatic hydrocarbons, and CH 4 . Sulfur would exhibit these forms:
SO3, SO2, S2—Sg (gas, liquid, or solid, according to the temperature),
COS, and H 2S; nitrogen the following: N O 2, NO, N2O, N2, and NH 3 .
Some of the above species (oxides of nitrogen and many hydrocarbons) are not favored thermodynamically, since they have positive free ener gies of formation. They are not likely to be found in planetary atmos pheres unless either npn-thermodynamic processes prevail or very specific conditions exist. An attempt will be made to determine which gaseous species are compatible with the known quantities of GOg and
HgO in the atmosphere of Venus.
Carbon Equilibria
In gaseous homogeneous reactions, equilibrium not only may occur at the hottest, densest part of the Venus atmosphere, namely, at the surface, but also may occur in the upper part of the atmosphere where the temperature is still high enough for rapid chemical reactions .
One reaction that has been considered by many investigators (for ex ample, Lewis, 1970, and Holland, 1964), is the self-decomposition Of carbon monoxide. The reaction is
2CO(g) = 002(g) + C (graphite). (2,10)
The presence of CO in the atmosphere of Venus has been established by spectroscopic evidence with a molecular mixing ratio, CO/CO 2 =
10-4 *3. This measurement/however, is of the upper atmosphere and may not be valid for the lower atmosphere. It might also be true that this measurement is due not to some thermal type of reaction in the lower atmosphere where pressures and temperatures are high but is due rather to a photolytic decomposition of carbon dioxide at the very top of the
Venus atmosphere where solar ultraviolet radiation is present . The ratio, observed, 10~4•3, is in accord with theoretical calculations of McElroy,
S ze, and Yung (1973) and Prinn (1973). In the lower atm osphere, the 44
amount of carbon monoxide could very well be lower than this, or per
haps even higher. It is clear from the above equilibrium that if no carbon
is present, no carbon monoxide can be present. Therefore, it is possible
that the lower atmosphere contains no carbon monoxide at all. Consider
ing this equilibrium, Holland (1964) postulated a rain of graphite falling
on the surface of Venus. This reaction would be cyclic, involving the
formation of CO from GOg and graphite near the surface . The carbon
monoxide would then be carried to the upper atmosphere where it would
tend to decompose and produce the graphite rain. However, since COg
is thermodynamically favored relative to CO at lower temperatures, at
some point in the Venus upper atmosphere CO would effectively disappear.
Figure 5 is a graph of the partial pressures of CO and COg in reaction (2,10) versus temperature, as well as a plot of the GO pressure
at the spectroscopic mixing ratio of 10- 4 .,34. At the intersection of the
of the two CO plots is the lower temperature limit of the CO self-reaction,
550 K. In the upper atmosphere, the observed ratio of 10~4 . 3 higher
than the equilibrium ratio of 10" ^ . 5 (at 300 K). The ratio of 10~4 • 3
could be the "quenched" ratio produced at 550 K. However, it is not
known at the present, time if this reaction is indeed the proper one. The .
reaction seems to indicate that the pressure of CO would be much higher
near the surface, thus producing a more reduced oxidation state than one
would estimate from the upper atmosphere abundance. However, the
above reaction with CO could easily be overwhelmed by a reaction con trolled by more abundant species that react more efficiently than the CO
self-reaction would. There is, for example, the reaction with sulfur
sp e c ie s . 45
2 CO - CO2 + C (groph)
CO (equi
-2 CO (constant mixing ratio in ~ 4 .3 \ o> o
—6
-8
-10 300 400 500 600 700 800 Temp °K
Figure 5. Equilibrium partial pressures of the gases in the reac tion 2CO(g) = CO2(g) + C(graph) as a function of temperature from 300 to 7 50 K in the Venus atmosphere
Two curves of CO are shown: one is for the log PcO from the re action, the other indicates log Pq q presuming constant CO/COg = 10-4.3 . 46 Sulfur Equilibria
. The sulfur species to be considered are Sg, SOg, and COS. In respect to Sg, the predominant gaseous species of sulfur at 750 K, the following reaction could occur:
S2( g )+ 4 C 02(g) =-2S02(g) + 4CO(g). 2 (, 11)
The equilibrium constant K for gaseous reaction is equal to the product of the fugacities of the products of the chemical reaction raised to the power equal to their coefficient in the chemical equation divided by the product of the fugacities of the reactants raised to the power of their coefficient in the chemical reactions. In the above reaction, for example, the equilibrium constant, K WNsozlz [s2][co2]4 where square brackets denote fugacity. (Partial pressure and fugacity are approximately equal at low total pressures.)
It is interesting to speculate on the relative abundance of each of the above sulfur species in the Venus atmosphere. One might assume that all of the material is initially present as the species on the left- hand side of the equilibrium and that the species on the right-hand side of the equilibrium are formed from the reactants. In this example and all subsequent examples entailing sulfur, the initial carbon dioxide abun dance at the surface will be taken as equal to 100 atm and that the CO 2 pressure will fall off adiabatically as observed in the Venus atmosphere
(Figure 1). The partial pressure of the other gas, in this case , sulfur, will be given by either a measured or assumed mixing ratio. In unknown cases, the partial pressure will be assumed from analogy with the earth. 47 The analysis of Earth's sedimentary rocks gives an idea of the ratio of carbon to sulfur species, even though both carbon and sulfur are probably underestimated for many of the Earth's sediments. Pettijohn
(1957) list the value of G/S as 18 in sedimentary rock. Rubey (1951) . has a C/S value of 24 for oceans plus sedimentary rock. In equation
(2,11) at 750 K we assume that the CO2 pressure equals 100 atm and the
Sg pressure equals 2.5 atm, inasmuch as 82 contains two sulfur atoms.
The initial and equilibrium quantities are as follows:
82 +4002 = 2802 + 4CO
at start: 2.5 + 100 = 0 + 0
at equilibrium: (2.5 ^x) + (100-4x) = 2x + 4x.
The equilibrium constant, K = (2x)2(4x)4/(100-4x)4(2 . 5 -x) is solved by inspection and successive approximation. The thermodynamic data for many of these calculations is obtained from the tables by Robie and Waldbaum (1968) and also from the JANAF (1970) tables . The equi librium constant and the partial pressures, of the components are calcu lated from 750 K to 298 K in 100° increments . The results of the above reaction can be seen in Table 4. To make these results more easily visualized, the equilibrium partial pressures of the various gases have been plotted on a graph of log P versus temperature in Figure 6. The abundance of 82 decreases directly with the abundance of CO 2, while both CO and SO 2 decrease precipitously, at lower temperatures. How ever, at the lowest temperature on the graph (298 K), it is quite likely that equilibrium would never be obtained under these situations due to a lack of sufficient activation energy. Indeed, sulfur is quite stable in the presence of carbon dioxide from room temperature to. 300 or 400 K. 48
Table 4. Log P (atm) of various species at equilibrium
Log P Temperature Log Equilibrium °K Constant, K S2 C 0 2 8 O2 CO
.740 -18.572- 0.40 2.00 -1.85 — 1.55
700 -19.815 0.30 1.90 -2 .1 4 -1.84
600 -24.910 0.00 1.60 -3.32 -3 .02
500 -32.055 -0.50 1.20 -4 .7 6 -4.46
400 -42.779 -1.00 0.60 -7 .0 5 -6.75
. 298 -61.049 -1.60 0.00 -10.59 -10.29
Equilibrium will probably not be obtained throughout the whole Venus at mosphere but would be more probable at temperatures around 500. K.
Another sulfur equilibrium involves 802 and SO 3:
S 0 2(g) + C 02(g) = SOgfo) + CO(g) (2,12)
The initial species in the atmosphere .are carbon dioxide at 100 atm of pressure and sulfur dioxide at 5 atm. It can be seen in Figure 7 that the quantity of SO 3 and CO formed in.this reaction is vanishingly small.
The SO3 and GO are six orders of magnitude below SO 2 at the surface of
Venus and become even less abundant in the upper atmosphere. Sulfur trioxide will not be an abundant gas on Venus.
The third equilibrium involves three sulfur gases:
S2(g) + 2002(g) = COS (g) + 0S 2 (g) + OO(g). (2,13)
This equilibrium illustrates how a single species, gaseous sulfur, 82 ,
Reacts with carbon dioxide to produce three different species: an in S tion 0 t 70 i te eu atmosphere Venus the K in 750 to 300
2 Log P (atm) g + C2g = 82g + C() s fnto o eprtr from temperature of function a as 4CO(g) + 2802(g) = 4C02(g) + (g) -10 - 0 2 Figure 2 8 6 0 0 3 6 Eulbim ata pesrs fte ae i te reac the in gases the of pressures partial Equilibrium . 4C02 2S02+ 0 C 4 + 2 0 S 2 2= 0 C 4 + 2 S 0 0 4 500 Temp°K 06 0 7 0 0 0 0 8 49 50
S02+ C02= S03+ CO
CO
s o
•*—E o
-6
-8
-10 1— 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 Temp °K
Figure 7. Equilibrium partial pressures of the gases in the reac tion S 0 2(g) + CO2 (g) = SO3 (g) + CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere 51 oxidized form of sulfur, a reduced form of sulfur (COS), and carbon mon oxide. The conditions of the equilibrium demand that the COS, SOg, and
GO abundances have to be equal. Figure 8 shows that the S2 partial pressure is roughly 1.5 orders of magnitude greater than SO 2 and COS .
Furthermore, the carbonyl sulfide and sulfur dioxide do not fall off as drastically as in Figure 6, but nevertheless there are some five orders of magnitude separating their partial pressures from that of 82 at 300 K.
The relative stability of carbonyl sulfide is seen in the follow ing equilibrium:
COS(g) + 2C02(g) = S 02(g) + 3CO(g). (2,14)
If one assumes an initial partial pressure of 5 atm of carbonyl sulfide at the surface (750. K), there are three orders of magnitude difference be tween COS and SO 2, and the difference increases as the temperature de creases. The partial pressures plotted in Figure 9 imply that if COS should by any reason become the dominant gas in the lower Venus atmos phere, it should persist throughout the atmosphere.
Carbonyl sulfide can also undergo self-decomposition:
COS(g) = 2CO(g) + S2(g). (2,15) The equilibrium partial pressures are plotted in Figure 10 as well as the vapor pressure of sulfur (Sg). The sulfur vapor pressure is mostly com puted as Sg molecules; converted to S 2 vapor, the pressure would rise by a factor of 4i Again carbonyl sulfide is stable against self- decomposition, and the amount of sulfur produced is low enough so that it falls be low the vapor pressure curve; all the Sg would be in the gas eous state. At the surface (750 K) the 82 abundance is almost two orders of magnitude lower than COS and decreases even more at lower ue rm 0 t 70 i te eu atmosphere Venus the K in 750 to 300 from ture S tion
2 Log P (atm) -10 -6 -8 (g) + 2C02(g) = COS(g) + S02(g) + CO(g) as a function of tempera of function a as CO(g) + S02(g) + COS(g) = 2C02(g) + (g) 0 0 3 Figure 8 Eulbim ata pesrs fte ae i te reac the in gases the of pressures partial Equilibrium . S 0 0 4 2 + COS 2 C 0 2= COS + S 0 2+ CO 2+ 0 S + COS 2= 0 C 2 0 0 5 Temp °K 0 0 6 0 0 7 0 0 8 52 rm 0 t 70 i te eu atmosphere Venus the K in 750 to 300 from SO = 2C02(g) COS + (g)tion
Log P (atm) -10 -2 4 - -8 iue . qiiru pril rsue o h gss n h reac the in gases the of pressures partial Equilibrium 9. Figure 0 0 3 0 0 4 +20= C0 23C + 0 S 2002= S+ O C 2 () 3Og a a ucin f temperature of function a as 3CO(g) + (g) 0 0 5 Temp°K
0 0 6 0 0 7 0 0 8 53 54
2C 0S = 2C0 +5%
E o
CL cn o _J
-8
-10 3 0 0 0 5 0 0 6 0 040 0 8 0 070 Temp °K
Figure 10. Equilibrium partial pressures of the gases in the reac tion 2COS(g) = 2CO(g) + S2(g) as a function of temperature from 300 to 750 K in the Venus atmosphere
The vapor pressure of Sg is indicated by the dashed line. $2 would remain gaseous because its partial pressure is less than the vapor pressure of Sg. ' 55 temperatures . In both reaction (2,14) and (2,15), the carbonyl sulfide is much more stable than the other sulfur species.
The following equilibrium illustrates a reaction similar to reaction (2,13),
332(g) + 4002 (g) = 4003 (g) + 2302(g), (2,16) except that in this case the diatomic sulfur reacts with carbon dioxide to produce only carbonyl sulfur and sulfur dioxide . Figure 11 indicates that the partial pressures of the three sulfurspecies (32 , COS , and SO 2) re main within an order of magnitude. Higher in the atmosphere, the other quantities do not disappear at the usual precipitous rate. However, it should be borne in mind that the pressure of S 2 here is purely hypothet ical at this particular high mixing ratio because, as can be seen from the sulfur vapor curve , the 32 abundance is greater than the vapor pressure , , necessitating condensation of 82 to solid or liquid sulfur. Therefore, this reaction, while interesting in other respects, would have to be con sidered simply as purely hypothetical. As the 82 vapor pressure falls, the COS and SO 2 have to fall along with it at the same rate.
A self-decomposition of carbonyl sulfide more realistic than reaction (2,15) is:
8003(g) = 800(g) +88 (g). (2,17)
In the Venus atmosphere, it is more likely that the sulfur is present as the octatomic species rather than as the diatomic species . In general, the higher the temperature and the lower the pressure the more likely octatomic sulfur is to decompose, into S 5, S4, and 82 ; but at the pres sures and temperatures in the Venus atmosphere , most of the molecular species will be Sg. From Figure 12 one can see that at the surface the , 56
3S2+4C 02= 4C0S + 2S02
E o Q_ cn o -J
-8
-10 3 0 0 4 0 0 5 0 0 6 0 0 0 8 0 070 Temp °K
Figure 11 . Equilibrium partial pressures of the gases in the reac tion 382(g) + 4002(g) = 4COS(g) + 2802(g) as a function of temperature from 300 to 750 K in the Venus atmosphere
The reaction is hypothetical because of the assumed pressure of 8 2 (g) exceeds the vapor pressure of Sg (dashed line). 57
8C0S = 8C0+Se
E o CL cn " 4 o
-10 3 0 0 4 0 0 500 6 0 0 7 0 0 8 0 0 Temp °K
Figure 12. Equilibrium partial pressures of the gases in the re action 8COS(g) = 8CO(g) + Sg(g) as a function of temperature from 300 to 750 K in the Venus atmosphere
At 340 K, Sg(g) condenses to a solid because the partial pres sure exceeds the vapor pressure (dashed line). 58
Sg partial pressure is roughly two orders of magnitude lower than the
COS. Since the Sg molecule contains eight atoms versus one atom of S
in COS, the elemental sulfur abundance in Sg is about one order of mag
nitude less than in COS. As the temperature drops with altitude, the Sg
quantity does decrease rather precipitously. It should be noted that in
this equilibrium the quantity of Sg stays below the Sg vapor pressure
curve.; - all the Sg is in the gaseous state. Only where the two lines
cross over around 340 K would the solid form of sulfur begin to form
from the vapor.
The following equilibrium,
3Sg(g) + 16C02(g) = 16COS(g) + 8S02(g), (2,18)
and Figure 13 represent a more realistic version of reaction (2,16); here the sulfur is considered as the octatomic gas instead of the diatomic
species. As Figure 13 indicates, Sg is indeed a stable species versus
simultaneous oxidation and reduction to SOg and COS, respectively.
Since the equilibrium partial pressure of Sg is greater than the sulfur vapor pressure, Sg vapor is limited to the vapor pressure at any tempera ture. If Sg(g) is formed by the above reaction, excess Sg(g) would con dense, and precipitate as a rain Of liquid or solid sulfur . The removal of
Sg(g) by condensation would drive the equilibrium to the left, according to Le Chatlier!s principle. One may conclude that oxidized and reduced
species of sulfur (SO2 and COS, respectively) cannot coexist in the at
mosphere; elemental sulfur would be favored.
Would the equilibrium partial pressures of COS and SO 2 be radically affected if the sulfur species in reaction (2,18) were solid or
liquid S instead of Sg(g)? (Sulfur melts at 386 K.) 59
3S 8 + I6C02 = I6C0S + 8 S 0 2
CO
-2
-6
-8
-10 3 0 0 4 0 0500 6 0 0 7 0 0 8 0 0 Temp °K
Figure 13. Equilibrium partial pressures of the gases in the re action 3Sg(g) + 16CO2 (g) - 16COS(g) +8 SO2 (g) as a function of tem pera ture from 300 to 750 K in the Venus atmosphere
The reaction is hypothetical because the assumed pressure of Sg(g) exceeds the vapor pressure of sulfur (dashed line). 60
3S(s J) + 2C02(g) = 2GOS(g) + S 02(g). . (2,19)
Figure 14 shows that with sulfur at unit activity the partial pressures of
COS and SO 2 formed are less than are formed in reaction (2,18) due to
the fact that solid sulfur has zero free energy (the standard state), where
as, in the previous equilibrium, the Sq bctatomic gas has a positive free
energy and hence favors the formation of COS and. SO 2. Again one can
conclude that COS and SO 2 are mutually exclusive: elemental sulfur pre
dom inates.
In general, it appears that if one considers an equilibrium
without regard to the presence of sulfur-bearing minerals on the surface, -/
sulfur species remain relatively stable. COS resists self-decomposition
and reaction with CO 2 . S2, Sg, and S(s ,/) are relatively resistant to
formation of COS and SO 2 SO2 is relatively stable; it resists further
oxidation but can be reduced by such agents as carbon monoxide and 82 .
Another series of reactions involves reduced sulfur in the form
of hydrogen sulfide:
H2S(g) + C0 2 (g) = COS(g) + H20 (g). . (2,20)
This equilibrium illustrates the oxidation of B 2S by CO 2 to produce car
bonyl sulfide and water. The reaction is an interesting one. It is equi-
molar, hence insensitive to pressure changes, and the thermodynamics
do not particularly favor the formation of COS. In all cases , the equ.i- -
librium constant is less than unity, namely, AG is positive. Figure 15
is a plot of the partial pressures of the various gaseous species in re
action (2,20), assuming an initial partial pressure of 5 atm of hydrogen
sulfide and letting it react to form products. In this reaction, the assum ption was made that there was no initial water vapor present in the Venus rm 0 t 75 K n h Vns atmosphere Venus 7 the K in 50 to 300 from 3S (s action
Log P (atm) -2 -10 -4 -6 iue 4 Eulbim ata pesrs fte ae i h re the in gases the of pressures partial Equilibrium 14. Figure -8 0 0 3 ,}) 20 g = CSg +82g a a ucin f temperature of function a as 802(g) + 2COS(g) = (g) 2002 + Ss C2 2O + SOz + 2COS 2C02= + 3S(s) 0 0 4 CO 2 0 0 5 ep °KTemp O S 0 0 6 0 0 7 0 0 8 61 rm 0 t 70 i h Vns atmosphere Venus the K in 750 to 300 from cin H action
Log P (atm) -2 -6 -10 - 8 0 00 40 300 2 iue 5 Eulbim ata pesrs fte ae i te re the in gases the of pressures partial Equilibrium 15. Figure g + CO S + (g) H2S(g) 02(g) C + COS(g) H20(g) =+ 2 () CSg + H + COS(g) = (g) CO 2 500 e p °K Temp 0 2 (g) as a function of temperature of function a as (g) 600 700 00 80 63 atmosphere. It can be seen, however, that this reaction actually pro duces more water vapor than is presumed present in the lower portions of the Venus atmosphere. Water vapor mixing ratios, as measured by the
Venera space probes, are in the vicinity of 10-2-2—10-3 .0^ but at 750 K this reaction would give rise to a vapor pressure of water whose mixing ratio is about 10“1 •?, three times higher than the upper limit of the
Venera measurements. In all cases, it would seem that the hydrogen sulfide resists relative oxidation to COS. Another way to look at this particular reaction is to consider that not only is the carbon dioxide specified by the Venus atmosphere, namely, 100 atm near the surface, but also that the water vapor is established by other reactions on Venus .
The partial pressures are re plotted in Figure 16 with the water vapor fixed at 10~3 mixing ra tio . In this, c a s e , near the surface the COS is more abundant than the HgS by roughly one order of magnitude. Then the rela tive ratio GOS/HgS decreases until the two are equal at 480 K. At lower temperatures, the HgS becomes the predominant species, until at 300 K ■ the HgS is three orders of magnitude greater than COS. This reaction could have profound significance in the consideration of the source of sulfur for the Venus clouds.
Equilibrium (2,21),
2H2S(g) + 2C02(g) = S2(g) + 2H20(g)+2CO(g), (2,21) shows the relative stability of hydrogen sulfide to partial oxidation to sulfur. Figure 17 indicates that H2S is only slightly oxidized to S2.
Hydrogen sulfide is also stable in regard to decomposition into Sg.
Hydrogen sulfide is not easily oxidized to sulfur dioxide:
H2S(g) + 3CC2(g) = SC2(g) + H20(g) + 3CO(g). (2,22) 64
HgSfg) + C02(g) - COS(g) + H20(g)
E o
-8
-10 300 4 0 0 500 6 0 0 700 800 Temp °K
Figure 16. Equilibrium partial pressures of the gases in the re action H 2S(g) + CO2(g) = COS(g) + H20(g)/ when the partial pressure of H2O is fixed, as a function of temperature from 300 to 750 K in the Venus atmosphere
H2O is assumed fixed at a mixing ratio, H 2O/CO 2 = 10~3.0 e eprtr fo 30 o 5 K nte eu atmosphere Venus the K in 750 to 300 from temperature action action 2 -10 -6 Log P (atm) - 4 - H 8 iue 7 Eulbim ata pesrs fte ae i h re the in gases the of pressures partial Equilibrium 17. Figure 0 0 3 2 2H2S(g) 2 C02(g) + = S2(g) + 2H20(g) + 2CO(g) S + (g) 2 CO 0 0 6 0 0 4 2 () = (g) CO H2S 2 8 g + (g) 0 0 5 ep °K Temp 2 H 0 2 g + (g) 2 Og a a ucin of function a as CO(g) 0 0 7 0 0 8 65 66
The plot of partial pressures in Figure 18 indicates that the sulfur diox ide species is even less stable relative to hydrogen sulfide than is the
82 species. . ,
From the consideration of these twelve reactions, the most like ly sulfur species in the Venus atmosphere is the reduced species, either
COS or H28 . As seen in Figures 15 and 16 , which of the two species is : the more abundant would be strictly dependent on the water vapor content of the Venus atmosphere „ It is also interesting to note that the water vapor in the Venus atmosphere could be generated solely by means of hydrogen sulfide reacting with carbon dioxide.
Nitrogen Equilibria
Another element that might exist in the atmosphere of Venus in either a reduced or oxidized form is the element nitrogen. The reduced form is ammonia, NHg, and the oxidized form is nitrogen gas, N 2 . It was automatically assumed by most investigators that there Would be nitrogen in the atmosphere of Venus, and it was also assumed that the form would be the diatomic gas, N 2 . Therefore, it was with some sur prise that the Soviet Venera 8 reported a mixing ratio of ammonia of lO"^ to. 10~3 at 46 and 33 km above the surface (Surkov, Andreichikov, and
Kalinkina, 1973). The conditions corresponding to these measurements are at 46 km above the surface lO^ • 3 atm and 350 K temperature and at
33 km, 10^ -9 atm of pressure and 470 K. One obvious reaction that would serve to limit the quantity of ammonia is suggested by Goettel and Lewis (1974):
2NH3(g) + 3C 02(g) = N2(g) + 3 % 0(g) + 3CO(g). (2,23) eprtr fo 30 o 0 i te eu atmosphere Venus 7 the K in 50 to 300 from temperature cin g() SCOg + HgS(g) action
- Log P (atm) -10 4 - -2 6 0 0 5 0 0 4 0 0 3 Figure 18. Equilibrium partial pressures of the gases in the in gases the of pressures partial Equilibrium 18. Figure H2S(g) 2(g) 02(g) 0 + S 3C H20(g) + - + 3CO(g) (g) = HzS SO 2 (g) e p °K Temp SO + H + Q C 0 2 g + C() s fnto of function a as 3CO(g) + (g) 0 0 6 0 0 7 0 0 8 68
In this case the ammonia partial pressure was considered to be .2 atm
near the surface, a realistic estimate of the quantity of nitrogen out- gassed on Venus. With the initial conditions set at 2 . atm of ammonia
and 100 atm of C02 and considering the pressure of CO 2 to decrease
with altitude similar to the sulfur reactions, the partial pressures of N '2 /
H2O, CO, and NHg that are in equilibrium under these conditions are plotted in Figure 19. The partial pressure of NH 3 measured by Venera 8
is also plotted. The measured mixing ratio of 10~3 to 10-4 is less than
that predicted from the chemical equilibrium. If the Venera measurements
are accurate, they imply that either the reaction does not reach equilib rium or the total amount of nitrogen is less than that assumed. Since nitrogen gas, Ng, is notoriously difficult, to react with other species to produce ammonia, it is most likely that the partial pressure of ammonia
is low because of the inability of the back reaction to.produce the am
monia. It is clear from Figure 19 that the ammonia is thermodynamically
more stable at low temperatures and the nitrogen, Ng , is more stable at
high temperatures. It is not too likely that thermodynamic equilibrium
would be obtained at temperatures much lower than 500 K (cf. Haber process reaction kinetics in Cotton and Wilkinson, 1966), and so it is
not unlikely that the ammonia partial pressure found is less than would be predicted if equilibrium prevailed. The measured partial pressure of
NH3 corresponds to a "frozen" equilibrium established at 650 K. If the back reaction is kinetically hindered at low temperatures , it is possible to reconcile the. measurements of the Soviet probe Venera 8 with the par tial pressures thermodynamically calculated from equilibrium (2,23). 69
2NH3(g) + 3 C 0 2(g) - N2(g) + 3H 20(g) + 3C0(g)
-6
-10 300 400 500 600 700 800 Temp °K
Figure 19. Equilibrium partial pressures of the gases in the re action 2NHg(g) + SCOg(g) = Ngfg) + SHgOfg) + 3CO(g) as a function of temperature from 300 to 7 50 K in the Venus atmosphere
The dashed line encloses the range of P-T of NHg(g) measured by Venera 8. .70
Another reaction that might affect the abundance of ammonia is that involving sulfur-bearing gases, particularly carbonyl sulfide:
2NH3(g) + 3COS(g) = N2(g) + 3H2S(g) + 3CO(g). (2,24)
The carbonyl sulfide depletes ammonia to a greater degree than C0 2 d o e s, as Figure 20 indicates. In this reaction, the mixing ratio of NH 3 to C 0 2,
10~4 to 10-3, exists between temperatures of 330 to 470 K. Since mea surements were made near those temperatures , it can be seen that even in the presence of the sulfur compounds, the partial pressure of ammonia predicted by thermodynamic calculations is in accord with the observa tions . However, there still remains the problem of the activation of N 2 to produce ammonia in these particular temperature regimes. In the case of reaction (2 ,24), Ng is favored at high temperatures and a quenched equilibrium cannot be invoked as in reaction (2,23). Ammonia would, not be produced at these relatively low temperatures , unless one is pro vided with a very long time in a statically stable Venus atmosphere.
Stone (1975) estimates that the adiabatic lower Venus atmosphere has a mixing time scale of a few weeks, perhaps too slow for the nitrogen- ammonia equilibrium to be produced.
Another reaction with sulfur that would certainly limit the abun dance of ammonia would be the following:
4NHs(g) + 3S02(g) = 2N2 (g) + 7H20(g) + 3S(s,/>. (2,25)
The thermodynamics of this equilibrium greatly favors the right-hand side of the equation to the virtual exclusion of ammonia, as seen in Figure 21.
If S2(g) were produced instead of S(s,J), the equilibrium partial pressures of the other species would be similar, since S 2 would condense to the liquid or solid. Ammonia and sulfur dioxide are not mutally compatible 71
2NH3(g) + 3C0S(g) - N2(g) + 3H2S(g) + 3C0(g) 2
0
-2
-4 o>
-6
-8
-10 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 Temp °K
Figure 20. Equilibrium partial pressures of the gases in the re action 2 NH3 (g) + 3COS(g) = N 2 (g) + SHgS(g) + 3CO(g) as a function of temperature from 300 to 750 K in the Venus atmosphere 72 7 0 8 0 070 6 0 0 (g) (g) (s,/) + 33 as a function of 2 0 H 6 Temp Temp °K 5 0 0 4 NH3(g) + 3S02(g) = 2N2(g) + 6H20(g) + 3S(s) + + 6H20(g) 2N2(g) = 3S02(g) + 4NH3(g) 3 0 0 4 0 0 Figure Figure . 21 Equilibrium partial pressures of the gases in the re
-8 -10 -6
(ILIJD) d 6 0 1
action (g) 4NH3 + 3802(g) = (g) = 2N2 temperature from 300 to 750 in K the Venus atmosphere 73 with each other, and whichever one was more abundant would survive.
If the ammonia, therefore, is present in the atmosphere, there is an in dication that sulfur dioxide could not be present. A reaction similar to reaction (2,25) is reaction (2,26) where instead of the two gases form ing solid or liquid sulfur, hydrogen sulfide is formed:
2NH3 (g) + 802 (g) = N2 (g) + 2H2 0 (g) + H2S(g). (2,26)
From Figure 22, it is obvious that ammonia is no more stable under these conditions than in reaction (2,25), and it is abundantly clear that am monia and sulfur dioxide cannot coexist. Whichever one is more abun dant would rule to the virtual exclusion of the other.
Another equation of interest is the reaction of ammonia with solid or liquid sulfur:
2NH3'(g)+3S (s,i) = N2(g) + 3H2S(g). (2,27)
In this case, one would have to imagine that sulfur was being formed in one of the other sulfur equilibrium reactions and ask how it would affect the ammonia partial pressure. If equilibrium obtains. Figure 23 indicates that the ammonia partial pressure is too low. In general, ammonia is in compatible with.oxidized sulfur species or even with elemental sulfur. .
The only species with which ammonia could be present are either the re duced species of sulfur, COS or H 28 , or with carbon dioxide itself.
These three substances in the atmosphere permit ammonia to exist in the quantities measure by Venera 8 . The presence of ammonia reveals some thing about the overall state of oxidation of the Venus atmosphere, name ly, that the conditions are somewhat reducing. eprtr fo 30 o 0 i h Vns atmosphere Venus the 7 K in 50 to 300 from temperature cin N3g +82g = 2g + H0g + H + 2H20(g) + N2(g) = 802(g) + 2NH3(g) action -6
-10 Log P (atm) -2 iue 2 Eulbim ata pesrs fte ae i h r re the in gases the of pressures partial Equilibrium 22. Figure 4 - N 3(g) 2(g) N2(g) 0 2NH +S - H20(g)+ 2 H2S(g)+ 0 0 5 0 0 4 e p °K Temp 0 0 6 2 S(g) as a function of of function a as S(g) 03 0 7 0 0 0 0 8 74 75
2NH3(g) + 3S(s) — N2(g) + 3H2S(g) 2
0
-2
- 4
-6
NH -8
000 5 0040 6 0 0 7 0 0 8 0 0 Temp °K
Figure 23. Equilibrium partial pressures of the gases in the re action 2NH3 (g) + 3S(s,/) = N2 (g) + 3H2S (g) as a function of temperature from 300 to 7 50 K in the Venus atmosphere CHAPTER 3
HETEROGENEOUS EQUILIBRIUM REACTIONS ON VENUS
While homogeneous equilibrium reactions determine the partial . pressures of various gaseous species in Venus' atmosphere, it is impor tant not to neglect the heterogeneous equilibria that can occur between the various gases and the solid minerals of.the planet's crust. The abun dance of an element in the atmosphere will be set by the partial pressure of its molecular species in equilibrium with the surface. Heterogeneous reactions that limit the partial pressure of a gas are called "buffer" re actions . The chemical and mineralogical composition of the surface rocks will determine which buffer reactions are appropriate to the surface, of Venus. Since carbon dioxide is the most abundant gas in the atmos phere , the production and subsequent buffering of CO 2 partial pressure will be examined in detail. The partial pressure of many minor gaseous components is dominated by the oxygen fugacity of the atmosphere; like ly oxygen buffers are examined. Compounds of elements that can exist in various oxidation states , sulfur compounds , for example , will be buffered not only by the oxygen fugacity but also by various surface min erals. Beside sulfur buffers, a series of carbon, iron, and nitrogen buf- ers will be considered in connection with various oxygen fugacities. Likely wafer vapor and hydrogen halide buffers will then be investigated. Finally, the buffer reactions considered above were treated as ideal situations: the gases were treated as ideal and the minerals as. pure substances. The effect of pressure upon the free energy of formation of
76 .-■■■■ ' , - 77 carbon dioxide will be examined. Experimentally determined CO 2 and
H2O partial pressures' in certain heterogeneous equilibrium reactions will indicate how the presence of impurities in minerals serves to alter partial pressures of gases in equilibrium with these minerals.
The Surface Composition of Venus
The only direct measurement of the surface constituents of
Venus was that performed by the Soviet space probe, Venera 8. The space craft employed a gamma-ray spectrometer to measure the naturally occurring radiation due to uranium, thorium, and potassium-40 (Surkov, .
Kirnozov, and Vinogradov, 1.973). The elemental composition determined was U, 2.2 ppm; Th, 6.5 ppm; K,4%. Many earth granites have similar composition. Surkov, Kirnozov, and Vinogradov (1973, p. 679) postu lated that, like Earth, Venus has undergone differentiation in its crust:
"if processes leading to a change of the chemical composition of the primordial matter (whatever its nature had been) had taken place, they were, in the main, similar for the earth and Venus . " This statement ap pears reasonable, and for the purposes of modeling the chemical reac tions occurring at the surface of Venus, it will be assumed that the types of rock present on the surface of Venus are similar to the types of rock on earth, at least as far as the original igneous rocks are concerned.
Granitic rock was present at the land site of Venera 8, but just as granite (or rhyolite) is restricted to certain areas on the earth's sur face, it is probably so restricted on Venus. The greenstone belts in Pre- cambrian cratons are considered by Anhaeusser et al. (1969) to be the metamorphic remnants of lava flows ranging from ultramafic to rhyolitic. 78 An intermediate rock type, such as basalt or andesite, may represent the early crust of the earth, and a similar composition is assumed for Venus.
Typical elemental and mineralogical analyses of granite, intermediate, and basalt rocks are shown in Tables 5 and 6.
Earth has produced abundant sedimentary deposits due to the physical and chemical reactions of its atmosphere and hydrosphere on igneous rock. Venus should exhibit similar but not identical modifica tion of its surface igneous rocks. Earth sediments are dominated by shales (aluminous clays), sandstones (silica), and limestones (carbon ates). The depositional environment was characterized by aqueous solu tions and low temperatures (~290 K). Under the present conditions of temperature and pressure at the surface of Venus, sedimentary rocks de posited by the action of aqueous solutions cart be ruled out. However, high-temperature, high-pressure, "dry" weathering of igneous rock can occur on Venus, and chemical reactions appropriate to this type of regi men will be discussed in this chapter. In the early history of Venus, conditions of low temperature and pressure, as well as the presence of liquid water, could have given rise to weathering reactions similar to those of present-day Earth.
Carbon Dioxide Buffer Reactions
It was Urey's (1952) contention that the wollastonite reaction,
C aSi03(s) + C 0 2(g) = C aC 03(s) + S i0 2(s), (3,1) wollastonite carbon calcite quartz dioxide proceeding to the left, produced carbon dioxide in the atmosphere of
Venus, not necessarily under equilibrium conditions due to the escape Table 5 . Chemical compositions of various rocks (weight %)--After Nockolds (1954)
Alkali Granite Granodiorite Tholeiitic Basalt Alkali Basalt
S i0 2 73.86 66.88 50.83 46.77 A1203 13.75 15.66 14.07 14.65 CaO 0.72 3.56 10.42 12,42 Fe2°3 0.78 1.33 2.88 3.71 FeO .1.13 2.59 9.06 7.94 MgO . 0.26 1.57 6.34 6.82 T i02 0.20 0.57 2.03 3.00 N a2G 3.51 3.84 2.23 2.59 k 2o 5.31 3.07 0.82 1.07 h 2o 0.47 0.65 0.91 0.51
Table 6. Mineralogical compositions of various rocks (weight %)--After Larson (1942)
Granite . Granodiorite Gabbro Olivine Diabase
Quartz 25 21 --- ■--- ■ Orthoclase 40 15 --- Oligoclase 26 --- — T- : --- Andes it e --- . 46 ------Labradorite ------65 63 Biotite 5 3 1 — Amphibole 1 13 3 --- Orthopyroxene — — --- 6 --- Clinopyroxene ------14 ...■ 21 Olivine ------7 12 Magnetite 2 1 2 2 Ilmenite 1 — — 2 2 80 of CO2 - This is a typical metamorphic reaction in which siliceous lime stone produces carbon dioxide, which on earth generally Outgasses into the atmosphere. Urey considered the same thing to occur on Venus.
This reaction, however, is considered by Lewis (1970) and by many others to be the buffer reaction that limits the quantity of carbon dioxide on the surface, and indeed it is the only simple reaction that has the precise amount of carbon dioxide pressure corresponding to the surface temperature of Venus, namely, 100 atm at 750 K. Other reactions, such as the reaction with diop side,
CaMgSi206(s) + C0 2 (g) = CaGOsCs) + MgSiOgts) + SiOgXs) (3,2) diop side carbon ' calcite enstatite quartz dioxide produce CO 2 pressures that are higher than 100 atm. For this and all subsequent reactions at the surface, equilibrium partial pressures will be computed at 750 K, the surface temperature of Venus. The equilibrium partial pressure of CO2 in reaction (3,2) is 10^-40 atm. The following two reactions are other CO 2 buffers.
OaMgSigOgW + 2C02(g) = CaM g(C03)2 (s) + 2Si02(s) (3,3) diop side carbon dolomite quartz dioxide '
This reaction has an equilibrium CO 2 partial pressure of 103• 06 atm.
M gSi03(s) + C 0 2(g) = M gC03(s) + S i0 3(s) (3,4) enstatite carbon magnesite quartz dioxide
The equilibrium CO2 partial pressure in this reaction is 10^ -26 atm. In general, the pure calcium reactions produce lower CO 2 equilibrium par tial pressures than the pure magnesium reactions. This simply 81
demonstrates the fact that calcium carbonate is more stable versus its
silicate, than is magnesium carbonate .
Barth (1962) in treating contact metamorphism describes a s e
quence in which limestone, either of the calcitic or do lorn it ic variety,
undergoes gradual decomposition.under heat to produce various volatile s.
Reaction (3,5) demonstrates that the first product of metamorphism in a
mixture of dolomite , quartz, and water at around 475 K is talc, calcite ,
and some carbon dioxide .
CaMg(C0 3)2(s) + 4 S i02(s) + H20 (i) = Mg3Si4O10(OH)2(s) dolomite quartz. + water talc
+ 3C aC 03(s) + 3 C 0 2 (g) (3,5) calcite carbon dioxide
The second step occurs at higher temperatures around 500 K„
5Mg3Si4 0 io(OH)2(s) + 6C aC 03(s) + 4S i02(s) = talc calcite quartz
+ 3Ca2Mg5Si8 0 22(OH)2(s) + 6C 02(g) + 2H20(g) (3,6) tremolite carbon water dioxide
In this case, the talc that is formed reacts with the calcite and further
quartz, giving rise to tremolite, carbon dioxide, and water. At higher
temperatures (525 K), the tremolite formed reacts with further dolomite,
giving rise to forsterite, calcite, carbon dioxide, and water.
Ga2Mg3Sig022(0H)2(s) + UCaMg (C03)2(s) = 8Mg2Si04(s) tremolite dolomite • forsterite
+ 13C aC 03 (s) + 9 GO 2(g) + H20 (g) (3,7) calcite carbon water dioxide
These three reactions, which occur at relatively low temperatures, are a
sequence in which C0 2 gas is produced. The three equations may be 82 combined to eliminate the intermediate products, talc and tremolite, yielding equation (3,8).
2CaMg(C03)2(s) + S i0 2(s) = 2CaC0 3 (s) + 2C02(g) • . dolomite quartz calcite carbon dioxide
> Mg2Si0 4 (s) , (3,8) forsterite
This particular reaction is rather interesting because it indicates that it is possible to have a carbon dioxide-producing reaction without neces- arily involving the decomposition of calcite itself but only of the more easily decomposed dolomite. When dolomite plus quartz decomposes, reaction (3,3), the partial pressure of carbon dioxide produced is over a thousand atmospheres. In reaction (3,8), the partial pressure of car bon dioxide is much lower. It is interesting to note that the reaction produces in this case not the pyroxene, diop side, but rather the olivine, forsterite , and calcite. The minerals calcite and forsterite could exist at the surface of Venus, and as such these two materials could be a means for limiting the quantity of carbon dioxide in the atmosphere; that is , equation (3,8) would go in the reverse sense. The equilibrium partial pressure of carbon dioxide at 750 K is 102 • 34 f 220 atm. The un certainty in the calculations, which are mostly uncertainties in the AG of formation and the AH of formation, are + lO 1-1 •40, corresponding to a spread from 88 to 550 atm. Thus, beside the wollastonite equilibrium, this reaction is the only one that falls within the range of a possible
Venus carbon dioxide buffer reaction. These reactions occur, according to Barth's (1962) scheme., at temperatures below 540 K. The tremolite that was formed in reaction (3, 6) reacts with calcite and quartz to ■ 83 produce diop side, carbon dioxide and water, at 540 K„
Ca2Mg5Si8022(0H)2(s) + 3GaC0 3 (s) + 2Si02(s) = tremolite calcite quartz
5CaMgSi206(s) + 3C0 2 (g) + H2 0 (g) (3,9) diop side carbon water dioxide
Again, at somewhat higher temperatures, near 560 K, dolomite reacts with water to produce brucite, calcite, and carbon dioxide .
CaMg(C03)2(s) + H2OC0 - Mg (OH)2 (s) + CaC0 2 (s) dolomite water brucite calcite + 0 0 2 (g) (3,10) carbon dioxide
The formation of brucite, however, is probably of only minor importance and occurs over only a very limited span of temperatures because near
550 K brucite decomposes into magnesium oxide and water at 1 atm, thus effectively driving water out of the system. The overall products that are formed in this scheme are, first, talc, which subsequently de composes, tremolite, which in turn decomposes. Minerals that do not decompose at 750 K are: forsterite (reaction 3,7), diopside (reaction 3 ,9), and perhaps periclase (MgO) (from brucite, reaction 3,10). These species, along with calcite-woliastonite equilibrium (reaction 3,1), are important at 725 K, as Barth indicates. So the surface constituents that would be in overall equilibrium with the carbon dioxide atmosphere would be wol- lastonite , periclase , diopside, and forsterite . All of these components in the presence of calcite can be involved in equilibrium reactions to buffer the carbon dioxide pressure. The two most likely buffering reac tions are the wollastonite equilibrium (reaction 3 ,1) and the 84 forsterite-calcite equilibrium (reaction 3,8). It is logical to suppose . ■ ■ . - that the Venus atmosphere gradually produced an abundant carbon dioxide atmosphere through reactions (3,5—3,10) as well as the wollastonite equilibrium reaction, each phase occurring in its own proper temperature interval. As the carbon dioxide gas increased in quantity, the green house effect would cause the surface temperature to rise and allow each progressive step of metamorphism to occur, leading to outgassing of car bon dioxide and also possibly of water vapor.
There are other reactions besides those given above which would certainly help fix carbon dioxide in the early atmosphere of Venus when it was still cool and when liquid water was still present. The plagioclase feldspars are well known for their ability to weather to pro duce clays and calcium carbonate. This is the familiar reaction that occurs even in arid environments where deposits of caliche are formed from the decomposition of calcic plagioclase. The decomposition of anorthite could be schematically shown by equation (3,11):
CaAl2Si208 (s) + C0 2 (g) + 2H2 0 (/) = CaCOaXs). anorthite carbon water calcite dioxide
+ Al2Si205(0 H)4(s) (3,11) kaolinite
This equation would not be valid for regulating the quantity of GOg now present at the surface of Venus because the equilibrium partial pressure of CO2 is too high; furthermore , kaolinite would decompose . The above reaction would be very useful for producing carbonate precipitates under cold conditions . Another reaction that might possibly serve to fix carbon dioxide is a reaction with the other end member of the .plagioclase series, 85 albite, producing soda and kaolinite :
2NaAlSi30g(s) + C 02(g) + 2H20 (/) = N a2C 0 3(s) albite carbon water soda dioxide
+ Al2Si205(0 H)4(s) + 4S i02.(s) (3,12) . kaolinite quartz
The formation of sodium carbonate is not favored even at room tempera- ... ture „ The reaction proceeds to the right if the sodium carbonate is dis solved in solution and removed from the system. Even on earth, the re action is relatively rare, although soda deposits in alkaline lakes are known. Perhaps a more reasonable reaction to consider would be the decomposition of a typical plagioclase like labradorite containing both calcium and sodium to produce both carbonates.
Reactions (3,11) and (3,12) are proposed simply to indicate that carbonate may be accumulated at low temperatures for further decomposi tion later in the planet's history. Finally, to complete the picture there are reactions between other silicates and carbon dioxide. Equation
(3,13) is a typical reaction in which the silicate is fayalite and the Car bonate produced is siderite:
Fe2S i04(s) + 2C 02(g) = 2FeC03(s) + S i0 2(s) (3,13) fayalite carbon siderite quartz dioxide
The reaction proceeds readily to the right at 298 K. The equilibrium vapor pressure of C0 2 is ICP^.l atm.
A more realistic reaction involves olivine (containing Fe and
Mg) reacting with C 02 to produce the mixed carbonate breunnerite.
(Fe,M g)2Si0 4 (s) + 2C0 2 (g) = 2(Fe,Mg)C03(s) + S i0 2(s) (3,14) olivine carbon breunnerite quartz dioxide 86 Assuming equal mole fractions of Fe and Mg in olivine and breunnerite, the equilibrium partial pressure of GOg at 298 K is 10~^e® atm. A simi lar reaction involving enstatite (reaction 3,4) produces an equilibrium
CO2 partial pressure at 298 K of 10- 5.2 atm.
Equations (3,11)—(3,14) and (3,4) would indicate various ways of fixing initial carbon dioxide as carbonate on the surface of Venus as the primitive planet outgassed and CO 2 reacted with the minerals of ig neous rock. The presence of water is probably necessary to make these reactions occur in finite periods of time. In general, one would expect to find deposits of quartz, calcite, soda, and possibly breunnerite, as well as magnesite (equation 3,4), and dolomite (equation 3,3). Then, as the surface of Venus warmed up through time, the production of carbon dioxide occurred in reactions (3,5) —(3,10) as well as by the wollastonite equilibrium, reaction (3,1).
The Oxygen Fuqaclty of the Venus Atmosphere ■
The partial pressure of carbon monoxide at the surface of Venus has not been measured. The maximum amount of CO that can be present is given by the equilibrium reaction of the self-decomposition of CO in equation (2,10). Th equilibrium partial pressure of CO at the surface is
10~0.38 a^m or a go to CO2 mixing ratio of 10~2 • 38 _ This mixing ratio
(100 times upper atmosphere ratio) would have to be considered the. ab solute maximum of carbon monoxide present at the surface of Venus. The , minimum quantity of carbon monoxide present in Venus could be zero if and only if all other oxidizable species were completely oxidized to the highest extent. This does not seem likely; it would mean, for example. 87 that among the more common elements, all iron would have to be ferric iron and all sulfur would have to be in the 6+ oxidation s ta te , such as. in sulfur trioxide gas, SO 3, and in sulfates, such as CaSC> 4 . In order to determine what is the true oxidation state of Venus' atmosphere, it would be useful to consider the oxygen pressure or oxygen fugacity that would exist under various types of oxidizing conditions. Then with this calculated oxygen fugacity one may determine what is the overall state of various oxidizable and reducible species, such as oxides of carbon and oxides and reduced species of sulfur and nitrogen.
Carbon dioxide and water have probably been outgassed from the earth in the ratio H 2O/CO 2 = 4 by weight. This number is obtained by considering that the total mass of water outgassed from the earth is found mostly in its oceans (1.4 x 102 4 g ) and that the total amount of
GOg trapped as carbonates is about 70 atm, a weight of about 3.5 x
1023 g „ This is the assumption made by Rasool and DeBergh (1970). In their calculation concerning the greenhouse effect on Earth and Venus, they considered that water dissociated and the resultant hydrogen es- > caped from the planet Venus, while the residual oxygen was absorbed in . the crust. This has often been assumed by a number of different investi gators (Davhoff et al. , 1967; Sagan, 1968), but does require some anal ysis as to its possibility—just how much oxygen can be absorbed by the planetary crust? Considering that Venus now has in its atmosphere rough ly 0.10 of an atmosphere of water and that approximately 280 atm were produced, a great deal of oxygen left over from the dissociation of water must be absorbed by the crust in oxidizing various types of elements'. 88
The first and most common of the oxidizable materials is iron,
and by analogy with the earth, it is assumed that the crustal iron is not
metallic but rather has been already oxidized to the ferrous state in vari
ous types of silicates, such as olivine and pyroxene. The quantity of
oxygen released from dissociation of 280 atm of water vapor would be
1.9 x 105 g/cm^. If the average type of basalt is roughly 8 % FeO, 3 .1
x 10? g/cm^ basalt would be required to absorb the oxygen. With the
density of basalt averaging about 3.1 g/cm^, the basalt rock would have
to be weathered to about 70 km in depth, a substantial amount. Further
more, this calculation is done on the basis that the ferrous iron goes
completely.to the ferric state, a reaction like
FeO + 1/402 -» 1/2 Fe^Og. - (3,15)
To ease the burden on the quantity of iron required, it might be
useful to consider other substances that could be oxidized along with
iron, such as sulfur. The mineral troilite is oxidizable in two respects
and can absorb a considerable amount of oxygen as equation (3,16) indi-
‘c a te s .
FeS + 9 /4 0 2 = l/2Te20 3 + S 0 3 . (3,16)
The exact quantiy of sulfur on Venus is of course unknown, but from
analogy with earth sediments in which the weight ratio of CO 2 to SO3
is approximately 10:1, if there were a total of 120 atm of CO2 outgassed
from Venus, this would imply a total of 6.2 x 10^2 g SO3 being produced.
The amount of oxygen that would be absorbed in equation (3,16) would be
approximately 1.5 x 10^ g/cm2. This is less than a tenth of the total
amount of oxygen that would have to be absorbed from an ocean of water. 89 namely, 1.4 x 1()24 g water. Under the assumption that it is very diffi cult to completely oxidize all the iron in 70 km of crust of the planet, it seems more likely that the total amount of water that was outgassed from Venus was not equivalent to that of the earth but rather something much less, perhaps only equal to the amount of carbon dioxide or per haps even less than that. Rubey (1951) has estimated that the total mass of crystalline rock on earth that has been weathered is 1 . 1 .x 1024 g, or about 1 km, averaged over the earth's surface. If his estimate is true and Venus has had a comparable weathering process, the crustal rock could have removed oxygen from only 4 atm of water vapor. Another esti mate based on a later paper of Rubey (1955) indicates that earth sediments could remove the oxygen from 10 atm of water vapor. The total amount of outgassed water vapor on Venus is uncertain. Perhaps the water vapor quantity could be about equal to the CO 2 quantity, if Venus weathered its surface rock more extensively than the earth.
The absolute minimum oxygen fugacity would be given by the self-decomposition of carbon dioxide:
C 0 2(g) = C(s) + 02(g). (3,17)
At a temperature of 750 K, the partial pressure of 0 2 would be. 10-27.54 atm.
If iron is present in the ferrous state, as in olivine, the follow ing reaction could be suitable for buffering oxygen:
3Mg2Si0 4 (ol) + 3Fe2Si0 4 (ol) + 0 2 (g) = 6MgSi0 3 (py) forsterite fayalite oxygen . enstatite . ;. in olivine in olivine in pyroxene
. + 2Fe304(s). (3,18) magnetite 90
In this reaction, solid solutions of the minerals affect the equilibrium partial pressure of oxygen. In previous equilibria, the activities of solids were presumed to be unity. A typical olivine, however, has a mole fraction, X, of 0.7 for ste rite and 0.3 fayalite. The mole fraction of enstatite in pyroxene is assumed to be 0.7. The equilibrium constant for the following general reaction,
cC + dD = yY + zZ,
,s
where a is the activity of species. In equation (3,18), the activity of the solid will be assumed equal to its mole fraction, X, and the activity of the gas equal to its fugacity, f. The equilibrium constant for reaction
(3,18) is
K = feen)6 • (Xmg)2 _ (Q.7)6(i)2 (X^)6.(X^)6fo2 (0.7)6(0.3)6fo2
The exponent for the forsterite and fayalite mole fractions is 6 instead of
3 because each mole of forsterite contains two moles of magnesium and each mole of fayalite contains two moles of iron. The value of logK is determined from the thermodynamic tables, and is computed from the above expression. The value of fo2 at .750 K is 10-24.26 atm. Since the oxygen fugacity or partial pressure is greater than the minimum from re ac tion (3,17), reaction (3,18) is a possible oxygen buffer on Venus.
For pure fayalite
3Fe2Si0 4 (s) + 02(g) = 2Fe3Q4(s)_+3Si02(s), (3,19) fayalite oxygen magnetite quartz 91 the oxygen fugacity at equilibrium is 10~25.88 atm. A higher value of oxygen fugacity is obtained by further oxidation, of magnetite:
4Fe304(s) + 0 2 (g) = 6Fe20.3(5). (3,20) magnetite oxygen hematite
In this reaction, the equilibrium oxygen fugacity is 10*18.31 atm. Under these conditions, one can determine the CO pressure by means of the following reaction:
' - 2C02(g) = 2CO(g) +0 2 (g). (3,21)
With a surface pressure of CO2 at 100 atm and of O 2 at 10*18.31 atm, the equilibrium CO pressure is 10* 8 .99 atm or a mixing ratio, CO/CO 2, equal to 10* 5. 99 . This is also less than the maximum ratio of 10*2 .38 from reaction ( 2, 10), and it is also less than that observed in the upper atmosphere. This would indeed correspond to an oxidized state for the
Venus atmosphere. If, however, the amount of oxygen is that given by the olivine-magnetite equilibrium of reaction (3 ,18), the partial pressure of CO in reaction (3,21) would be 10*1.05 or a mixing ratio, CO/COg, equal to 10*3.05. jf iron is the dominant element in fixing the oxygen fugacity of the planet's surface, there are two main choices to take for . the oxidation state and the resultant carbon monoxide partial pressure: either the hematite-magnetite equilibrium or the olivine-magnetite equi librium. The boundaries for the mixing ratio, CO/CO 2, are 10*5 .99 for the hematite-magnetite equilibrium (3,20) and 10*3.05 for the olivine- magnetite equilibrium (3,18). The oxygen fugacity corresponding to these limits is, for the oxidized case, 10*18.31^ anc} for the reduced state,
10*24.26 atm. If all iron were completely oxidized, however, the oxy gen fugacity could exceed 10*18.31 afm. Sulfur Buffers
It is interesting to consider what these different conditions would, me an to the other gaseous species in equilibrium with such oxy gen fugacities. In the first stage, the oxygen fugacity will be set by the hematite-magnetite equilibrium. Under these conditions, tremolite oxidizes as follows:
4FeS(s) + 702(g) = 2Fe203(s)+ 4802(g). (3,22) troilite oxygen hematite sulfur dioxide
With an oxygen fugacity of 10”^ , the sulfur dioxide in equilibrium would be a 1,000 atm greater than the whole pressure of the atmosphere. All sulfur would immediately go into the gaseous state as sulfur dioxide.
But the sulfur dioxide would react with surface constituents to reduce its overall abundance. Such a reaction might be with wollastonite or with calcite to form calcium sulfate:
CaSiOs (s) + S02(g) + 1/20 2 (g) = CaS0 4 (s) + Si0 2 (s) (3,23) wollastonite sulfur oxygen anhydrite quartz dioxide or
CaC0 3 (s) + S0 2 (g) + l /202(g) = CaS0 4 (s) + C 0 2(g). ' (3,24) calcite sulfur oxygen anhydrite carbon - dioxide
In both of these equations, the amount of sulfur dioxide in equilibrium would be 10~5.04 atm, with fQ 2 fixed at 10~18 atm. The amount for both equations is the same because the calcite and wollastonite are in equi- . librium with the carbon dioxide at 750 K and 100 atm of CO 2 . This amount of sulfur dioxide is not overwhelming. The partial pressure of other sul fur species in equilibrium with this partial pressure of oxygen is not high, as seen in the following three equations: 93
2CaC0 3 (s) + S 2(g);+ 3 0 2 (g) = 2CaS0 4 (s) + 2002 (g) (3,25) calcite sulfur oxygen anhydrite carbon dioxide
CaC0 3 (s) + COS (g)+ 202(g) =C aS0 4 (s) + 2C 02 (g) ■ (3,26) calcite carbonyl oxygen anhydrite carbon sulfide dioxide
CaCOsXs) + H2S ( g ) +202(g) = CaS0 4 ( s ) + H 2 0 (g) calcite hydrogen oxygen anhydrite water sulfide + 002(g) . (3,27) carbon dioxide
In these equations, the partial pressure of the sulfur species is as fol- lows under equilibrium conditions: 83 = “15.47 atm, COS = 10"5 .88 atm,, and H 28 = io- 10*61 atm. Therefore, when sulfur species have to be in equilibrium with a sulfate, such as anhydrite, or other sulfates of mag nesium or sodium, the partial pressure of sulfur species is not great.
The most abundant species, as would be expected, is sulfur dioxide.
The partial pressure of sulfur trioxide due to decomposition of calcium sulfate is very small, only equal to 10~12.23 atm, the lowest of all the -- . sp ecies.
A fugacity of oxygen equal to 10~ 18 gravely restricts the par tial pressure of ammonia:
4NH3(g) + 302(g) = 2N2 (g) + 6H2 0 (g). (3,28)
The partial pressure of ammonia in equilibrium with this pressure of oxy gen is o n ly . 10“11 • 51 atm, an amount clearly undetectable by the devices used in the Venera 8 space probe.
The hydrogen halide species, however, are stable.
4HCl(g) + 0 2 (g) = 2H20 (g) + 2012(g). (3,29) 94
With PygO =: 0.1 atm and Pyci = 10“^ atm, the chlorine partial pressure is 10^15.60 atm, indicating very little decomposition of hydrogen chlo ride. .
The only species that have been detected spectroscopically that are compatible with this state of oxidation (fQg = 10-18 atm) are the hydrogen halide species of HC1 and HF. Carbon monoxide turns out to be almost two orders of magnitude smaller than the spectroscopic observa tions, and the amount of sulfur compounds (SOg for all purposes) is of such a small mixing ratio in the Venus atmosphere, namely, SO 2/CO 2 =
10-7, that it does not seem possible to produce clouds of sulfur com pounds. If the clouds are indeed sulfuric acid (as will be demonstrated in the next chapter), the concentration of sulfur in the upper atmosphere would appear to be at odds with the small mixing ratio in the lower at mosphere . But if the upper atmosphere were a sink for sulfur dioxide , perhaps through geological time sulfur compounds could be concentrated in the upper atmosphere. This would require increasing the SO 2/CO 2 to higher values, perhaps by 3 to 4 orders of magnitude. - No real con tact between the sulfur of the lower atmosphere and sulfur of the upper atmosphere would be possible, not a.realistic situation.
What conditions would prevail near the Venus surface if the more reducing situation prevails where the oxygen fugacity is equal to
10_24.26 atm? This corresponds to the oxygen fugacity set by equation
(3,18), the olivine-magnetite equilibrium. Under these reducing condi- . tions, the previous set of equations (3, 22)-(3, 27) undergo significant changes. In equation (3,22), for example, the equilibrium sulfur dioxide pressure is 10“^ * 11, extremely low in comparison with the previous 95 oxygen fugacity, down by approximately 10 orders of magnitude. Equa tions (3,23) and (3,24) would have a sulfur dioxide equilibrium pressure of 10~2.06 atm. However, this partial pressure of sulfur dioxide would be present only-if a sulfate, like calcium sulfate, were present at the
surface. It is of interest to see how calcium sulfate survives under these relative reducing conditions, and indeed the next three equations
(3,25), (3,26), and (3,27) indicate that the other sulfur species are much more plentiful. In equation (3,25), the equilibrium concentration of S2 rises to 10%«4! atm. In equation (3,26), the COS rises to 102*04 atm, and in equation (3,27) the H 2S pressure rises to lQl-42 atm. In all three cases, the partial pressure is greater than the abundance of sulfur; in the cases of S 2 and COS, greater than the Venus atmospheric pressure itself. This indicates that calcium sulfate would totally de compose in equations (3,25), (3,26), and (3,27), giving rise to the three sulfur gases, 82 being the most abundant.
For the reaction (3,28), the ammonia partial pressure rises con siderably but is still much less than that detected by Venera 8 . Under these more reduced conditions, the ammonia partial pressure becomes
10-7 .15 atm, and, of course, in equation (3,29) HC1 becomes even more stable than it previously was.
Since under these reducing conditions calcium sulfate is no ■ longer a stable species at the surface of the planet, some other species should be the regulating mineral that controls the overall sulfur abun dance in the atmosphere. Thermodynamic data are available on troilite, and under conditions where troilite is a stable species (on earth pyrite is the more common species), it might be the buffering mineral. The first .... 96 reaction to consider is
2FeS(s) + Si0 2(s) + .0 2(g) = Fe2Si0 4 (s> ■+_S2(g). (3,30) troilite quartz oxygen fayalite sulfur
In this case, there is no change in formal oxidation state of iron inas
much as it remains ferrous under both states, but oxygen is used to oxi dize the sulfur from sulfide to 82 gas. Under the reduced conditions of .
€>2 fugacity equal to '10- 24 atm, the 82 partial pressure is extremely low,
Iq-7 .22 atm. This reaction would virtually deplete the Venus atmos phere of sulfur, 82 . An equally depleting reaction is
6FeS(s) + 402(g) = 2Fe304(s) +3S2(g). (3,31) troilite oxygen magnetite sulfur
In this reaction the iron is in equilibrium between ferrous sulfide and magnetite. The equilibrium partial pressure of sulfur is only somewhat higher than in the previous equation, namely, 10~6.68 atm. In both cases ,. sulfur, 82 , would be a minor constituent.
The more reduced species COS fairs considerably better in the following equilibrium:
2FeS(s) + 8102 (s) + 2C02 (g) = Fe 28104 (5) + 2008(g). troilite quartz carbon fayalite carbonyl (3 >32) dioxide sulfide ’ .
In this case, as in equation (3,30), the formal oxidized state of iron is unchanged on both sides of the equation and the important concern is
with the change in sulfur to perhaps a somewhat more oxidized state,
although formally, sulfur stays reduced in both troilite and carbonyl
sulfide. The partial pressure of carbonyl sulfide at the surface would be 10-2.78 atm, 4 orders of magnitude higher than the 82 species. 97
If ferrous iron oxidizes to ferric iron, as in equation (3,31), the following reaction exists:
3FeS(s) + 3C0 2(g) + l / 2 02(g) = F e30 4 (s) + 3C0S(g). (3,33) troilite carbon oxygen magnetite carbonyl dioxide sulfide
In this reaction where iron is partially oxidized, the carbonyl sulfide partial pressure increases slightly to io~2-51 e
The partial pressure of hydrogen sulfide can be determined from the following two equations:
2FeS(s) + 8102 (s) + 2H2 0 (g) = Fe28104 (s) + 2H2S(g) (3,34) troilite quartz water fayalite hydrogen sulfide and
3FeS(s) + 3H20 (g) + 1/202 (g) = Fe304(s) + 3H.2S(g). (3,35) troilite . water oxygen magnetite hydrogen sulfide
In equation (3,34), the equilibrium partial pressure of hydrogen sulfide is 10~3.50 atm, and in equation (3,35), 10~3.23 atm, somewhat less than carbonyl sulfide.
Typical reactions involving SOg might be:
2FeS(s) + Si02(s) + 302(g) = Fe2Si04(s) + 2S02(g) (3,36) troilite quartz oxygen fayalite sulfur dioxide or 3FeS(s) + 502(g) = Fe30 4 (s) + 3S02(g). (3,37) troilite. oxygen magnetite sulfur dioxide
In these two reactions, the equilibrium partial pressure of sulfurdioxide is 10~6.88 atm in equation (3,36) and lO"^• 61 atm equation(3,37), again very low.
Since troilite is a mineral found mainly in meteorites, it is per haps not appropriate to use it in the equilibrium equations. The mineral 98 pyrrhotite, with the formula Fe^_XS, could possibly be considered as a mineral forming these equilibria, but it is relatively rare in comparison with the mineral pyrite. Iron sulfide equilibria using pyrite instead of troilite are seen in the following equations with the partial pressure of the sulfur species indicated.
2FeS2'(s) + S i0 2(s) + 0 2(g) = Fe2Si04 (s) + 2S2(g) . (3,38) pyrite quartz oxygen fayalite sulfur
Pgg = 1.93
3FeS2(s) + 2 0 2(g) = Fe30 4 (s) + 3S2(g) (3,39) pyrite oxygen magnetite sulfur "Pg. = 1Q-5.66
2FeS2(s)+Si02(s) + 4C02(g) = Fe2Si04(s) + 4COS(g) + 0 2(g) (3,40) pyrite quartz . carbon fayalite carbonyl oxygen dioxide sulfide
PCOS - IQ-2 - 14
. 3FeS2(s) + 6C 02(g) = Fe304(s) + 6COS(g) + 0 2(g) (3,41) pyrite . carbon magnetite carbonyl oxygen dioxide sulfide
P COS = 10-2 -00
2FeS2(s) + Si02(s) + 4H20(g) = Fe2Si04(s) + 4H2S(g)+ 0 2(g) (3,42) pyrite quartz water fayalite hydrogen oxygen sulfide
p H2S" = 10-2.73
3FeS2(s) + 6H20(g) = Fe304 (s) + 6H2S(g) + 02(g) (3,43) pyrite water magnetite hydrogen oxygen sulfide
Ph 2S = 10"2 -72 99
2FeS2(s) + 8 IO2 (s) + '50.2(g) = Fe2Si0 4 (s) + 4S 0 2 (g) (3,44) pyrite quartz oxygen fayalite sulfur dioxide
PS0 2 = 10-6.24
3FeS2(s) + 8 0 2(g) = Fe.304(s) + 6S02(g). (3,45 pyrite oxygen magnetite sulfur dioxide
PS02 = ID"6. 10
With the extra sulfur found in the solid mineral, it is not unexpected that
the quantity of sulfur compounds is increased in the equilibrium situation.
The reaction that yields the most abundant sulfur compound in the atmos
phere is that of equation (3 ,41) where carbonyl sulfide is in equilibrium
with magnetite and pyrite. The gaseous components carbon dioxide and
oxygen come into play in establishing the equilibrium quantity of COS.
So the most favorable mixing ratio, COS/CO2, equals 10“^"60. This is
not an abundant species, but it is nevertheless comparable to the mixing, ratio of carbon monoxide and thus could possibly be considered the fifth
most abundant species in the lower atmosphere after carbon dioxide,
water vapor, nitrogen, and carbon monoxide .
After considering these oxidizable and reducible species in the
Venus atmosphere, the author believes that using all the data previously
available and considered, such as the gas-gas equilibria reactions in
Chapter 2 and the data from Venera 8 , particularly the ammonia abun dance, and considering the presence of sulfur in sulfuric acid clouds,
it seems that the more reduced state, where the oxygen fugacity is 10^4 , is the most likely. The overall abundance of the common species is sum marized in Table 7. 100 Table 7 . Abundance of gaseous species at the surface of Venus
log Pressure Species (atm) Source Regulating Reaction co2 2.0 Venera 8
n 2 0 Venera 4, 5, 6 (3,27) .
h 2o - 1.0 Venera 4, 5, 6
CO -1.9 This paper (3,18)
COS - 2.0 This paper (3,40), (3,41)
H2S -2 .7 This paper (3,42), (3,43)
HC1 -4.0 Spectroscopic
HF - 6.0 Spectroscopic
n h 3 -7.1 This paper (3,28) s2 -5 .6 This paper (3,39) so2 —6.1 This paper (3,45) 00 CM °2 1 This paper (3,18) h 2 -3.22 This paper (3,46) 101 With the oxygen fugacity being 10~24z ^ interesting to note that hydrogen becomes moderately abundant in the Venus atmosphere, produced by the self-decomposition of water:
2H20(g) = 2H2(g) + 02(g). . (3,46)
With a water vapor partial pressure of 10~1 atm at the surface of Venus, the hydrogen pressure would be 10~3 .22 atm. Hydrogen would have some effect upon the stability of pyrite on the surface of the planet, as the following reaction indicates:
FeS2(s) + H2(g) = FeS(s) + H2S(g) . (3,47) pyrite hydrogen troilite hydrogen sulfide '
With the hydrogen partial pressure as given above, the hydrogen sulfide equilibrium partial pressure would be 10~2 .21 z three times more abun dant than in equation (3,43). The excess hydrogen sulfide would prob ably be used up in a reaction similar to reaction (3,34) or reaction
(3,35) to produce troilite. Generally, conditions on the earth tend to favor pyrite over troilite, but on Venus troilite could be the more stable species. If that is the case, the partial pressures of the sulfur species would be reduced by half an order of magnitude or more.
Water Vapor Buffers
The abundance of water in the upper atmosphere of Venus is very low. The mixing ratio H20/C02 is 10~6 for the upper atmosphere, with fluctuations as high as 10“4 . in the lower atmosphere, however, mea surements from the Venera space probes indicate a higher water vapor content with a mixing ratio of 1-6 x 10-3. In the following buffer reac tions, the H20 partial pressure is assumed to be 0.1 atm, corresponding 102 . I ■ ■ ■ ; to a mixing ratio of 10""3. The kinds of mineral buffers to be considered
are the inosilicates (amphiboles) and the phyllosilicates (mica). Thermo
dynamic data are available for some of these minerals, in particular,
muscovite , talc, and tremolite , and their decomposition reactions are:
KAl3Si3 0 io(OH)2(s) = KAlSi30 8(s) + A^OsCs)+ H20(g) (3,48) . muscovite sanadine corundum water
Mg3Si4 0 io(OH)2(s) = 3MgSiG3(s) + Si02(s) + H20(g) (3,49) talc enstatite quartz water
Ca2Mg5Si8022(0H)2(s) = 2CaMgSi206(s) + 3MgSi0 3 (s) tremolite diop side enstatite
+ Si02(s) + H20(g). (3,50) quartz water
In these three equations, the partial pressure of water in equilibrium at
750 K is: for equation (3,48), lO^*25 atm; for equation (3,49), 100• 34
atm; and for equation (3,50), 10~0-79 atm. Equation (3,50) is closest to
the value of 10- ^ from the Venera missions, although equation (3,49)
does not argue against the wetter models for the Venus atmosphere where .
the water vapor pressure is equal to 10® for a mixing ratio of IQ- ^.
Equation (3,48) is perplexing, however, because thermodynamic data predict an equilibrium water vapor pressure of some 10,000 atm, contra dicting empirically derived data from thermal decomposition of muscovite
where water vapor pressure of 10,000 atm is not obtained until a tem perature of 1,750 K, 1000 degrees higher than the thermodynamically
computed temperature (Barth, 1962).
According to Barth, the serpentine mineral .series, one member of
which is antigorite, Mgg8 i4 0 io(OH)g, would be unstable under Venus
conditions. Serpentine spontaneously decomposes near 500 C, and under 103 the. drier conditions of Venus it seems that serpentine would not be stable. It would be interesting, however, if there were thermodynamic data available on the mineral to compute its equilibrium partial pressure of water vapor. Another material for which thermodynamic data would be useful is the mica mineral phlogopite, KMgg AlSigOio (OH)2. This mica might be verging on stability around the surface of Venus. Possible If2.0 buffers on Venus include serpentine, the various micas, as well as talc and tremolite. Data from Deer, Howie, and Zussman (1962) indicate that the kaolinite clays are dehydrating in the temperature range 670—800 K and may be suitable buffers .
Hydrogen Halide Buffers
The abundance of the hydrogen halides in the upper Venus at mosphere has been given spectroscopically (Connes et al. , 1967).
Hydrogen fluoride has a mixing ratio of about 10~8 and hydrogen chlo ride a b o u t 10- 6. Possible surface buffering reactions are, where X =
Cl or F:
. 2NaAlSi30 g(s) + 2HX(g) = 2NaX(s) + AlgOgW alb it e hydrogen halite corundum chloride (fluoride) (villiaumite)
+ 6S i02(s) + H20(g) (3,51) quartz water .
C aC 03(s) + 2HX(g) = CaX2(s) + H20(g) + C 02(g) (3,52) calcite hydrogen hydrophilite water carbon chloride dioxide (fluoride) (fluorite) . •
V CaAl2Si208 ( s ) + 2HX(g)= CaX2(s) +A l2Si0 5 (s) anorthite hydrogen hydrophilite andalusite chloride (fluoride) (fluorite)
+ SiQ2(s) + H2 0 (g). (3,53) quartz water
In these reactions, the mixing ratio of the hydrogen halide is generally larger than that given in the Venus upper atmosphere „ In equation (3 ,51), for example, the hydrogen chloride equilibrium partial pressure is
10- 2.32 atm and the hydrogen fluoride 10"0.63 atm. Upper atmosphere mixing ratios, if valid for the surface, imply surface partial pressures of
10~4 atm for HCl and 10~6 for HF. In equation (3,52), the HC1 partial pressure would be 10~2.11 atm and the HF i 0- 3.66 atm. In equation
(3.53), the partial pressure is lO- * - 51 for HG1, 10“3.07 atm for HF.
These reactions do not appear to be suitable buffering reactions.
Since micaceous minerals can also contain either hydroxyl groups, fluoride ions, or chloride ions , a possible buffering reaction might be:
KMgaAlSijOio(OH)2(s) + 2HF(g) = KMggAlSigOioF^s) . phlogopite hydrogen f luorphlogop ite
+ 2H20(g) . (3,54) water
In this case, the equilibrium involves equimolar amounts of hydrogen fluoride and water vapor. The equilibrium partial pressure of HF is di rectly proportional to the partial pressure of H 2O; in reactions (3,51) —
(3.53) the partial pressure of HF is proportional to the square root of the
H2O pressure divided by the equilibrium constant. Thermodynamic data are available for an analogous reaction: . 105
Ca(OH)2(s) + 2HX(g) = CaX2(s) + 2H2G(g). . (3,55) portlandite hydrogen hydrophilite water chloride (fluoride) (fluorite)
In this reaction, the HF partial pressure is 10“7.12 atm, and the hydro gen chloride, 10*5.57 atm, lower pressures than assumed to exist in the Venus atmosphere. Nevertheless, they do show us that with equimolar reactions the quantity of hydrogen halide is less than in reaction (3,52). Furthermore, the ratio of HF to HC1 is around 10%, as is true of the upper atmosphere> A reaction like (3,54) both for HF and HC1 might well be the limiting buffer reaction at the surface of.Venus; that is, a micaceous mineral reacts with the halogen acids to produce water vapor and buffers the partial pressure of the acids to the values detected spectroscopically.
Non-ideal Conditions CO2 Buffers In all calculations performed so far concerning the equilibrium partial pressure of carbon dioxide in various silicate reactions , the free energy of formation of CO 2 has been taken from Robie and Waldbaufn (1968): -94.517 kcal per mole, or log k of formation 27,542. There are data from Hilsenrath (1960), however, that indicate that the free energy, or log k, of formation of carbon dioxide under pressure is not the same as the free energy of formation at one atm. The free energy change with pressure is tabulated in pressures running from 100 bars to 1,400 bars and from 400 K to 1,200 K. The values (AFp ^ - AF^) given at 750 K and 100 atm is 6,878 cal. This means, therefore, that the free energy of formation of CO 2 would be a -87.639 kcal with a log k of 25.538. If this value of log k is substituted into the gaseous reactions, carbon dioxide pressures would be reduced by one or two orders of magnitude. Using the data of Hilsenrath, one would solve the GOg buffer equations, and by a process of iteration, compute the pressure that cor responds to the free energy change of carbon dioxide proper to that par ticular pressure . For example, consider the wollastonite equilibrium
(3,1). It is first necessary to extrapolate the Hilsenrath tables to one atmosphere of pressure, and from inspection of the data at pressures above 100 atm, it appears as though the extrapolation of (AFp ^ — AF°) is linear from high pressures to 1 atm, where presumably the pressure correction would be zero; that is, the free energy of formation would be that given by Robie and Waldbaum. The iterative process first involves solving the equation as though the free energy of COg were the same as it was at 1 atm: the wollastonite equilibrium would give log PCC>2 ~ 2 .0 .
Using the Hilsenrath correction factor at 100 atm, one would obtain a log PcOg - -0.004 or approximately 1 atm. The correction factor is ob viously too large, as the correction was applied for 100 atm; the cor rected free energy value used in the equation yields only 1 atm. An in termediate pressure like 50 atm would yield a free energy intermediate, between 1 and 100 atm, with a consequent intermediate value for PcOg •
It is necessary through iteration to find a value of pressure and the Hil senrath correction that when used in solving the equation for the pres sure of COg gives a COg pressure that is equal to the pressure used in the correction factor. In the case of the wollastonite equilibrium, a pres sure of 10 atm gives a correction to the free energy of formation of COg that produces a PCOg of 10 atm. The same process was applied to the anorthite equilibrium (3,11), the diop side equilibrium (3,3), the enstatite equilibrium (3,4), and the forsterite equilibrium (3,8). The plotted values 107 are found in Figure 24 along, with the values obtained with a free energy corresponding to a pressure of CO 2 = 1 atm.
Beside these "Hilsenrath correction" factors, there are other more recent data obtained in regard to the experimental partial pressure of GO2 in the wollastoriite equilibrium (Barker and Tuttle, 1956). They discuss how Goldschmidt (1912) arrived at his values: CO 2 partial pres sure is approximately 100 atm for the decomposition of calcite and silica at 750 K. Using their new data, Barker and Tuttle found values for the pressure of CO2 at higher temperatures and pressures. Then utilizing the Nernst approximation formula:
log? = - r - ^ v + 1.75 log T + 3 .2 (AH = 23 .4 kcal/mole) 4 c o u / JL JL (in which the equilibrium pressure is a function of T and a constant, de termined empirically at higher temperatures)and also using the Van't
Hoff integrated form of the Clapeyron equation,
they were able to obtain a pair of univariant PT curves at lower tempera tures. They approached the equilibrium from both sides of the reaction.
In one case, they had quartz reacting with calcite to produce wollaston- ite and CO 2, and in other experiments, they reacted carbon dioxide with wollastonite to produce quartz and calcite. Two different pressures with two different extrapolated curves were produced. The value obtained at
750 K is approximately 25 atm, and the experimental variation is also shown as plotted in Figure 24—a low of 15 atm and a high of 33 atm .. 108
Minerol-COg Equilibria
2CoC03 + Mg2Si04 4- 2 CO2 C 2CaMg(C03)2 + Si02 O----
MgSi03 + C02r: MgC03 + Si02 -O- -e-
CoMgSi20 6 + 2C02X CaMg (C03)2 + 2Si02 O <3------k -•------
CoAI2Si208 + C02r; CaC03 + AI2Si05+ Si02 M------(
Co Si 03 + C02^ CaC03 4-Si02
VENUS
0 1 2 3 4 Log PCo 2
♦ Robie and Waldbaum B Hilsenrath CO2 free energy O Marker and Tuttle exper. Wollastonite CO2 pressure O Modified Hilsenrath CO2 free energy, conformed to Marker and Tuttle exper.
Figure 24. Partial pressure of CO2 in equilibrium with various minerals at 7 50 K
Pressures of CO2 are calculated or measured from data supplied by Robie and Waldbaum (1968), Hilsenrath (1960) # and Harker and Tuttle (1956). Arrows denote uncertainties in the value of the free energies of the compounds. Arrows with the Harker and Tuttle wollastonite equilib rium denote range of their extrapolated PCO 2 to lower temperature. 109
Also shown plotted in Figure 24 are the computed equilibrium pressures of three other reactions that would serve to buffer carbon di oxide:. the equilibrium reaction between anorthite and GOg producing calcite, andalusite and quartz; diop side reacting with GOg producing dolomite and quartz; enstatite reacting with CO 2 producing magnesite and quartz; and calcite and forsterite reacting with CO 2 to produce dolomite and quartz. In each of these four cases, as well as the wollastonite equilibrium, the diamond-shaped figures indicate the computed CO 2 p res sure , using the free energies given by Robie and Waldbaum (1968) that are proper to 1 atm of CO2 pressure. Arrows indicate the uncertainty in free energy of formation of all components. Plotted as squares are the
CO 2 equilibrium p ressu res, using the Hilsenrath (1960) CO 2 free energy corrections at higher pressure; in most cases they are 2 to 2 .5 orders of magnitude less than the pressures computed from Robie and Waldbaum.
Shown as open circles are the CO 2 pressures obtained, using a free energy of CO 2 derived from the Barker and Tuttle (1956) experimental results, and this change factor was applied to all the other reactions..
It can be seen that, if the experimental wollastonite data do indeed give a true picture of the free energy of CO 2, the most likely reaction serv ing to fix the carbon dioxide on Venus is that which involves diop side because the CO2 pressure comes closest to 100 atm. The anorthite and enstatite reactions produce CO 2 pressures that also fall within the un certainty of the free energies that were used to calculate the CO 2 p res sure. It appears as though the wollastonite equilibrium, if the experimen tal results on the free energy of CO 2 are to be believed, does not apply to Venus as a carbon dioxide buffer system. no HgO Buffers
Data on the free energy of formation of muscovite has been ob
tained by Huang and Keller (1972). They made measurements on five
common rock-bearing minerals to determine the -free energy of formation
of olivine , augite, labradorite , microcline, and muscovite. In their ex
periment, they used natural minerals of carefully analyzed composition
. which differed somewhat from the theoretical formulas that are found in
most tables on thermodynamic properties of minerals . Their experimen
tal method utilized the dissolution of these minerals in aqueous solutions.
The data obtained were applied to the actual mineral formulations em
ployed, as well as to the ideal structural formulas of the minerals, that
is, the usual stoichiometric formulas. Whereas the free energy of forma
tion of muscovite at 298 K is given by Robie and Waldbaum as -1,330
kcal/mole, the value for ideal muscovite computed by Huang and Keller
is -1,314 kcal/mole, and for the natural muscovite employed, -1,400
kcal/mole. Robie and Waldbaum's data on muscovite indicate a linear :
increase in free energy with rise in temperature. With a similar linear
relation, the value of free energy at 298 K (-1,400 kcal/mole) extrapo
lates to -1,254 kcal/mole at 750 K. With this latter value of free energy
in reaction (3,48), the water vapor equilibrium partial pressure, is lO- ^-^
atm at 750 K, a value not in accord with experimental results. Porn are v
and Ivanov (1972) report that the decomposition of muscovite into sani-
dine, corundum, and water vapor yields an equilibrium pressure of 1,000
kg cm~2 at a temperature of 580 C. This is about 1,000 atm at 850 K,
and following the usual decrease in pressure with lower temperature, at
750 K this would come to about 100 atm of pressure. Other experimental Ill values for the decomposition of muscovite, again decomposing into sani-
dine, corundum, and water vapor is given by Yoder and Eugster (1955) .
At 750 K, the equilibrium partial pressure of water vapor is approximate
ly 15 atm, this is too high for Venus . The maximum water vapor pressure is about 1 atm, that is, the upper limit of the Venera measurement at 1% by volum e.
There are other experimental data on the decomposition of talc and tremolite. The decomposition of talc into enstatite, quartz, and water vapor (reaction 3,49) reported by Bowen and Tuttle (1949) yields a decomposition water vapor pressure at 750 K equal to approximately 4 atm, just outside the Venus limits. The decomposition of tremolite into enstatite, diop side, quartz, and water (reaction 3,50) has been examined by Boyd (1959). His data, extrapolated to 750 K, give an equilibrium vapor pressure of water of 10“I atm, the probable lower limit of the
Venus pressure. The table for the high-pressure phase equilibria of
Clark (1966) shows no other simple dehydration water vapor partial pres sures that fall within the Venus range, between 0.1 and 1 atm, except for the decomposition of talc and tremolite. The only other mineral that has a decomposition temperature higher than that of tremolite is phlogo- pite, which decomposes as follows:
2KMg3AlSi3O10(OH)2 = 3Mg2Si04 + KAlSi20 6 + KAlSi04 phlogopite forsterite leucite kaliophilite
+ 2H'20 . (3,56) water
Data obtained from Yoder and Eugster (1954) give an equilibrium vapor pressure of water at about 10- 5.4 atm, much too small for Venus. This pressure of water vapor apparently conflicts with a new value for the ; 112 free energy of formation of phlogopite determined by Bird and Anderson
(1973), -1,389.5 kcal/mole at 298 K. Correcting for the usual increase in free energy with temperature, the author calculates that the equilib rium vapor pressure of water at 750 K in reaction (3,56) would be lO* atm .
Another type of layer-lattic silicate, the chlorites, may also be a suitable H2O vapor buffer. Yoder (1952) experimentally determined the decomposition H 2O vapor pressure of clinochlore , according to the fol lowing reaction:
5Mg5Al2Si30 io(OH)8 (s) = 3Mg2Si0 4 (s) + M g2Al4Si5 0 i 8 (s) clinochlore forsterite cordierite
+ 3MgAl204(s) + 20H2O(g). (3,5.7) spinel water
Extrapolation of this water vapor pressure curve to 750 K (Yoder and
Eugster, 1955) yields a value of. 10- 0 .5 atm, well within the Venus range.
Due to all the uncertainties in values of the free energies of formation and extrapolations of decomposition vapor pressures for lower temperatures, it seems that the suitable buffers for the partial pres sure of
H2O in the Venus atmosphere are phlogopite, clinochlore, talc, and trem- olite. Minerals that can be ruled out include muscovite, serpentine, hydroxides like brucite, kaolinite, and hydrates such as gypsum.
Halide Buffers
The data on the free energy of formation of phlogopite give one the ability to consider an equilibrium involving water in a micaceous
mineral in equilibrium with at least one of the halogen acids, namely. hydrogen fluoride, according to the following reaction:
KMg3(AlSi3Oio)(OH)2(s) + 2HF(g) = KMg3(AlSi3 0 io)F2(s) phlogopite hydrogen fluorphlogopite fluoride + 2H2 0 (g). (3,58) water
In this equilibrium at 750 K, the partial pressure of hydrogen fluoride is equal to 10~6 .2 atm, if the vapor pressure of water is 0.1 atm. This is the only reaction for which data are available on the relative stabilities of fluorine and hydroxyl type micas, and it gives an HF/CO 2 value of
10-8.2 close to the spectroscopic value of 10“8 . 0 . Since.no data are available on the chloride -type micas, it might, however, be possible that the HC1 is also of a lower ratio than that given in the previous re actions (3 , 51)-(3, 53).
Non-ideal conditions for both the a lb it e (reaction 3,51) and anorthite (reaction 3,53) equilibria would involve both minerals dissolved in the solid solution in plagioclase. Two possible halide buffer reac tions would be:
2NaAlSi3C8 (pg) + 2CaAl2Si208 (pg) + 6HX(g) = 2NaX(s) a lb it e in anorthite in hydrogen sodium plagioclase plagioclase halide halide
+ 2CaX2(s ).+ 3Al2SiOs(s) + 73102 (s) + 3H2 0 (g) (3,59) calcium andalusite quartz water halide
2NaA13i308(pg) + 6HX(g) = 2NaX(s) + Al2Si05 (s) +■ 5Si02 (s) albite in hydrogen sodium andalusite quartz plagioclase halide halide.
+ H2 0 (g). (3,60) water
The plagioclase is assumed to be labradorite, where albite:anorthite =
1:1,. and the activity of sodium halide, calcium halide, andalusite, and 114 quartz is unity; the fugacity of water is 0.1 atm; the mole fraction of al- bite and of anorthite in plagioclase is 0 .5. The equilibrium constants for the reaction are
K = ^aN aX^ ^ ^aCaX2^ ^aandal) ^ ^aq tz ^ ^H 2Q^ • (XPg)2CKPg)2(fHx)6 .
= i2 .12.13.17(0 al)3 (3,59) (0.5)2 (0.5) 2 (^ )6
K = (aNa%)^(aandal) (aqtz) ^ 2 _ 1'2 • 1«15(p (3,60) (Xgg)2(fHx)2 = (0--5)2«HX>6
The mixing ratio of the four hydrogen halide gases is found in Table 8 , along with the mixing ratio of the same gases determined by buffer reac tions (3,51) —(3 ,53) and (3, 58).
The buffer reactions that yield the lowest equilibrium partial pressure of the HX gases are as follows: (3,58), the phlogopite equi librium buffers HF at 10~6.2 atm, with a mixing ratio of 10“® *2; HC1 is limited in partial pressure either by a micaceous buffer similar to reac tion (3,58) or by the plagioclase buffer (reaction 3,60) with a mixing ratio io-4.03 (the pure a lb it e buffer, reaction 3,51, is ideal, only);
HBr is limited by the same buffer to a mixing ratio Of 10-4.61 and HI is buffered by the same plagioclase buffer (3,60) to a mixing ratio of 10-4.06. •
Is there sufficient iodine in the crust of Venus to produce a mixing ratio as high as 10-4.06? Using the earth for an analog, one might be able to answer the question by estimating the quantity of iodine outgassed from the earth's crust. Rubey (1951) considered such outgassed Table 8 . Mixing ratios of hydrogen halides determined by mineral buffers
Buffer Mineral Calcite Albite Anorthite Plagioclase Plagioclase Phlogopite
Equilibrium Reaction (3,52) (3,51) (3,53) (3,59) (3,60) (3,58) O CO CM log H F /C 02 -5.66 -2.63 -5.07 -4.13 -2.33 1 'sP 0 0 CO log HCI/CO2 -4.11 1 -3.51 -3.66 -4.03
log HBr/C02 -3.61 -4.91 -3.01 -3.52 -4.61 ^P O CO CO CO log HI/CO2 -2.91 1 -2.32 -2.87 1 116 species as "excess" volatile s. While he gives no value for excess io dine , he does list excess chlorine at 6 ,000 g / c m ^. Assuming that iodine is outgassed as readily as chlorine, one could compute the amount of excess iodine from the elemental ratio, Cl/l. Suess and Urey (1956) list
C l/l as 11,000; Cameron (1973) as 4,000. On earth, therefore, excess . iodine would be 0.53 g / c m 2 (Suess and Urey, 1956) or 1.50 g / c m ^
(Cameron, 1973). If Venus outgassed iodine similar to Earth, then the mixing ratio of HI in the Venus atmosphere has a maximum value between
10—4.9 ancj iQ-5.3^ Therefore, the quantity of HI in the Venus atmos phere is probably limited by its elemental abundance. CHAPTER 4
SULFURIC ACID IN THE VENUS CLOUDS
Current Data on the Venus Clouds
Substantial data exist on the spectroscopic and polarization properties of the. Venus clouds. Yet, for a long time the identity of the clouds was uncertain. Venus shows high reflection of visible light,
o with an albedo approaching 0.95 at 6000 A. There are strong absorption features in the ultraviolet and infrared, independent of the atmospheric absorptions caused by CO 2. Ultraviolet photography of the planet shows regions of variable darkness, attributed to variable cloud composition on the planet, revealing a dynamic system of production and destruction of a strong ultraviolet absorber, produced photolytically in the upper at mosphere. Figure 2 shows the albedo of Venus from 0.2 to. 4.0 pm, taken from Kuiper's (1969) article on the clouds of Venus, with the UV region modified by Wallace et al. (1972), and the 3-4 pm region modified to allow for the CO2 absorption band at 3.7-4.0 pm observed by Beer,
Norton, and Martonchik (1971). A double absorption near 3.8 pm, as sumed by these authors to be due to solid particles, is actually due to
CO2 gas (Burch et a l. , 1969).
Polarization studies by Coffeen (1968) led him to conclude that the particles in the clouds were probably spherical, with a refraction index (n) of 1.43-1.60 and a mean particle diameter of 2.5pm . Further work by Hansen and Arking (1971) and Coffeen and Hansen (1972) gives a
117 118 mean particle radius of 1.1 jam and n of 1 .46 + 0.02 at 0.35 jum, decreas ing to 1.43 at 1.0 ;im. A further refining of the data by Hansen and
Hovenier (1974) gives 1.44 + 0.02 as the refractive index at 0.55 ;um.
The particles are spherical and show other properties of spheres, the rainbow and possibly the glory.
Another important property of the Venus atmosphere at the level of the clouds, namely, at 50 mb pressure (lO- ^*^® atm) and 235-250 K, is the extremely low water vapor mixing ratio of 10~6.0 (Pj^O = 1.0- 4 . 30 mb). The vapor pressure of ice is higher than this value by orders of magnitude. If HgO is a constituent of the clouds, it must be present with a strong desiccating agent. Kuiper (19.69) in suggesting FeClg" ZHgO as a constituent of the Venus clouds made the cogent observation that the vapor pressure of water in FeClg ° 2H.20 at 250 K was compatible with the observed vapor pressure of water near the Venus clouds . This property is not unique to FeClg, however. Other possible desiccating agents can fulfill this function.
Properties of Sulfuric Acid
Is there any one substance that can account for all the above properties of the clouds and be compatible with the albedo of Venus over the spectrum. FeCl2 does a creditable job for some parts of the spectrum, but not all. FeCl 2 does not satisfy the polarimetric properties because it tends to form hexagonal platy crystals, not spheres.
The author was intrigued by the fact that FeSC >4 hydrate, as well as other hydrated sulfates, showed low reflectivity between 3.0 and 4.0 jum in the infraredi No other hydrates did as well in satisfying the low 119 reflectivity of Venus in this region. An obvious candidate for considera
tion as a cloud constituent was H 2SO4 , which also showed strong ab
sorption in the 3 to 4 pm region.
H2O Vapor Pressure
The first property to verify was the vapor pressure of water in
equilibrium with various H2SO4 solutions at 235 K. The International .
Critical Tables (Washburn, 1926-1930) give values of Pp^O over HgSO^
solutions from 10% to 95% H 2SO4 . The appropriate values seemed to be
in the high H2SO4 concentrations. At 235 and 250 K, the following were
calculated:
Log PH2Q of solution (atm) H2SO4 Percent by W eight 235 K 250 K
90 -8.05 -7.19
85 -7.15 -6.34
80 -6.52 -5.7
Since log Pp^O (atm) at the cloud level is -7.13, the 85% H 2SO4 solu tion is close to the observed value, with 86 % solution a good fit at
235 K; at 250 K the best fit is 89% solution. More exhaustive analysis of water vapor regulated by H 2SO4 was performed by Fink et al. (1972).
They found that H 2SO4 of about 80% composition can dry the.upper at mosphere to give good agreement with the measured abundance of H 2O. Index of Refraction
The second property to consider is refractive index n. The
International Critical Tables list refractive index for H 2SO4 solutions
(particularly 95%-96% solutions) at various temperatures and wave lengths. From 301 to 387 K, the refractive index decreases linearly with temperature increase. Extrapolating back to 235 K and correcting for o wavelength, n for 95%-96% H 2SO4 at 3500 A is 1.440. With the n value of 1.405 for H2SO4 • 2H2O (73%) at 20 G (Handbook of Chemistry and
Physics, Weast, 1970), the 235 K n value would be 1.419. The 86 %
H2SO4 solution would therefore be calculated as n =1.431 at 235 K.
This value is within the limits of the 1.44 + 0.02 of Hansen and Hovenier
(1974). Solutions of H 2SO4 < 95% are compatible with the polarization measurements. It well might be that the spherical droplets are actually frozen. This would raise the refractive index of 86 % H2SO4 to about
1 .453, well within the acceptable limits. H 2SO4 freezes in a complex manner, with various eutectics (H 2SO4-H 2O, H2S0 4 -2H2 0 , H2SC>4 *
4H2C) freezing at various temperatures, and mixtures of them freezing at lower temperatures. Pure H2SO4 freezes at 10.5 C, the monohydrate at 8 . 6, the dihydrate at -38.9, the tetrahydrate at -24.5.
The author has observed in the laboratory the freezing and melt ing properties of an 88 % H2SO4 solution. The frozen H 2SO4 seemed to have an ill-defined melting point; it appeared to just become less vis cous as it melted, behaving almost like a glass. A thin layer of the acid was frozen in dry ice and was perfectly transparent with a glassy sur face . A cold metal spatula seemed to dent the surface; no fractures were observed as is common with water ice. The cloud particles of Venus , 121 could therefore be of this "glassy" type H 2SQ4 , the droplets frozen into spheres, with a refractive index indicative of this denser state, namely, about 1.45.
Spectral Properties
The third property of H 2SO4 to compare with the clouds of Venus is spectral reflectivity. As of this time it has not been possible to ob tain reflection spectra from fogs of H 2SO4 . As a second best choice , transmission spectra of thin layers of H 2SO4 might simulate to a degree the complex absorption and scattering occurring in a fog of droplets. The thin layer of H 2SO4 was obtained by putting a drop of the acid between two plates and letting them seal together. Excess acid was usually squeezed Out around the edges of the plates. It is estimated that the layer of acid, is approximately 0 .05 mm thick. Two plates of optical quartz, each 2 mm thick were used in the transmission, spectra from 0.2 to 3.5 jum. AgCl plates, ground and polished, were utilized in the spec tral region 2.5 to 15.0 jum. Also used in this region, with less success, were drops of H 2SO4 sandwiched between layers of polyethylene and polyester. The absorptions of the plastics were added to H 2SO4, of course . This was partially cancelled in the dual-beam spectrophotom eter by placing two layers of plastic in the reference beam, but complete cancellation of the plastic film does not appear to be possible. Instru mental effects near the strong and sharp absorption features of the plas- . tic film produced spurious features in the H 2SO4 spectrum. The 0.2 to
3.5 pun spectra were obtained from a Beckman DK-2A dual-beam quartz spectrophotometer, and the 2.5 to.15.0 jum spectra from a Perkin-Elmer 122
137 NaCl spectrophotometer, both used by courtesy of Dr. B. Nagy of the University of Arizona's Laboratory of Organic Geochemistry.
Figure 25 shows the spectral transmittance of a thin layer of
88 % H2SO4 . The two quartz plates were used with an acid layer about
0.05 mm thick between the two plates. The reference beam was the air path. The spectrum needed only very small correction for the absorption of the two quartz plates. By comparison with the Venus spectrum (Fig. 1), it can be seen that there is a good match from about 1 to 3 jam. .
Further examination of the spectrum of Venus has been performed by Pollack et al. (1974). In their aircraft observations made with a wedge filter spectrometer from 1.2-4.1 Various laboratory spectra, as well as synthetic spectra of cloud par ticles , showed that there are some materials that have a very low reflec tance, such as Venus possesses. The synthetic spectra were generated in a computer program, utilizing, the known real and imaginary refractive indices of the various substances, with cloud particle size about 1 /am. Among the substances that had spectra similar to Venus were various concentrations of sulfuric acid as well as of hydrochloric acid. Other substances considered and ruled out were ferrous chloride, water, frozen water, mercury metal, ammonium chloride, nitric acid, carbon suboxide, and others. The sulfuric acid composition that seems to match the Venus low reflectance between 3 and 4 pm is a sulfuric acid composition great- . er than 75% and less than 95%. Hydrochloric acid droplets at concen trations near 12 molar, were also compatible with the Venus spectrum per se; however, in these concentrated solutions of hydrochloric acid the 123 Jr. 0.8 0.4 0.2 0.0 0.2 03 0.5 1.0 2.0 3.00.4 4.0 yUlTl Figure 25. Transmission spectrum of 88 % H2SO4 solution from 0.2 to 4.0 pm The sulfuric acid solution is 0.05 mm thick. Note scale change at 0.5 pm . 124 hydrogen chloride, HC1, vapor pressure is so very great that the low mixing ratio discovered spectroscopically for HG1, 10~6/ precludes the possibility of having such a concentrated hydrochloric acid solution at the level of the clouds. Therefore, it would seem that these observa tions of Pollack et al. (1974) confirm the identification of sulfuric acid. The author also wishes to note that similar identifications of the clouds of Venus as sulfuric acid were proposed independently by both the author (Sill, 1972) and by Young (1973). In Young's analy- , sis of the reflection spectrum of Venus, the vapor pressure of water, . and refractive index, he arrived at a composition of sulfuric acid that was about 75%. ■ Ultraviolet Spectrum Obviously, the ultraviolet and visible spectral region must be explained by some other substance (s). Among possible candidates for ultraviolet absorption is SOg dissolved in cold H 2SO4 . Since S is oxi dized to the 6+ oxidation state in H 2SO4 , the intermediate oxidation state of 4+ in SO 2 should be expected. The visible yellow color of the Venus clouds presents a problem as well. There are not too many gaseous substances that have a yellow tint. One is brown-yellow NO 2. But at the temperature of the Venus clouds, the dimer N 2O4 is highly favored, with - a consequent drastic lessening of color intensity. Furthermore, it may be extremely difficult to oxidize N in. the atmosphere of Venus. A better candidate is Br 2, a red-brown liquid or gas which, if diluted and dis solved in H 2SO4 to a moderate degree, produces a yellow solution; higher concentrations give a red-brown solution. The intensity of the , ; ' ' 125 color of Br2.varies with temperature. Bfg saturated in 88 % H2SO4 solu tion at room temperature produces a red-brown solution; when cooled to dry-ice temperaturethe frozen solid is yellow. The chemical aspects of Br2 and H 2.S.O4 will be discussed below. Figure 26 shows the inferred Bond albedo of Venus in expanded scale from 0.2 to 0 .8 jam. L. W allace (personal commun., 1972), from his geometric albedo in the rocket UV, estimates that the Bond albedo of o Venus is about 0.4 at 2500 A. Figure 27 shows the transmittance of Brg and HBr dissolved in H 2SO4; HBr was added to the Br 2 to dissolve more Br2 than water alone would dissolve. The (HBr + Br 2) solution was added to cold concentrated H 2SO4 to obtain a solution of 88 % H2.SO4 . Out- gassing of HBr from the solution was rather vigorous. Solution B was about five times more concentrated than solution A in Br 2 and showed more yellow color. Evidently, more coloring agent (Br 2), if added to the solution, would match Venus in the 0.3-0.5 pm regions, while the higher ultraviolet reflectance of Venus could be due to scattering . However, o Venus has a UV peak near 2500 A, whereas the Br 2 solution peaks at 2200 A and the UV scattering would tend to displace it to shorter wave - o lengths. The Venus spectrum shows a dip at about 2900 A, whereas the Br2 solution dips at 2700 A, which might also be slightly displaced to shorter wavelengths by UV scattering. . ' The SO2 spectrum is shown in Figure 28. Curve A is for SO 2 gas bubbled into cold 88 % H2SO4 and a layer about 0.1 mm thick was o observed through the quartz plates. The SO2 does show a peak at 2400 A and a dip centered at 2800 A, some what closer to features of the Venus albedo. To observe the effect of a greater amount of.dissolved SO 2 than 126 0.8 0.6 0.5 x 0.6^.m Figure 26. Bond albedo of Venus from 0.2 to 0 . 6 pm 127 1.0 Tr. 0.2 Figure 27. Transmission spectrum of HBr + Brg dissolved in sulfuric acid from 0.2 to 0.6 ^im Solution A: pale yellow color; solution B: yellow color, five times concentration of solution A. 128 0.6 0.2 0.3 0.4 0.5 X 0.6/xm Figure 28. Transmission spectrum of SC >2 dissolved in sulfuric acid from 0.2 to 0.6 >im Solution A: SO 2 - H2SO4 solution, 0.1 mm thick. Solution B: cell thickness, 1 cm; SO 2 0.005 of saturation. 129 could be held by the HgSO^ at room temperature, a cell of 88 % H2SO4 , 1 cm thick, was utilized with the SO 2 at about 1/200 of saturation (curve B). A similar cell with pure 88 % H2SO4 was used in the reference beam of the Beckman DK-2A. When Br 2 vapor is poured onto a SO 2—H2SO4 solution, immedi ate bleaching of the Br2 is observed, as well as a diminution of the ab sorption features of SO 2 . Depending on which constituent survives, the spectrum either resembles the SO2 or the Br2 spectrum more closely. Evidently the Br 2 oxidizes the SO 2 to H2SO4 and the Brg is reduced to HBr. Mid-infrared Spectrum Figure 29 shows the infrared spectrum from 4.0 to 15.0 jam of 88 % H2SO4 (solid line) with the left-hand ordinate of transmittance. The sample was one drop of acid squeezed between AgCl plates, with similar . plates in the reference beam. The dashed spectrum of Venus is that of Gillett et al. (1968) with the right-hand ordinate of log power at the detector. The dotted lines are (1) the reflection of sunlight from a body of unit albedo and (2) the black-body radiation curve at 225 K. If there is H2SO4 in the clouds cooler than the emitting surface, then indeed. H2SO4 should show in absorption. There are some similarities between the absorptions of the Venus spectrum 7-12 jim and the absorptions of 88 % H2SO4 , especially in the absorption feature centered at 11.2 jam, prominent in Hanel, Forman, and Stambach's (1968) Venus spectrum. Later, Young (1974) investigated the spectral behavior of su l furic acid solutions between 8.3 jam and 13.3 pm. He observed that the 130 1.0 Jr. 22 5 K block body 0.8 *"■ Albedo -4 0.6 to 0.4 -5 Cl 0.0 Figure 29. Transmission spectrum of sulfuric acid and the power spectrum of Venus , 4 to 16 jam Solid line: (left ordinate) transmission of 88 % H2SO4 solution; cell thickness 0.2 mm. Dotted line: reflection of body of unit albedo and blackbody emission of an object at 235 K. Dashed line: (right ordi nate) power spectrum of Venus . 131 solution of 75% sulfuric acid was in better accord with the observed spectrum of Venus than the 90% solution. However, Young did not take into account the absorption bands of carbon dioxide near 9.4 and 10.4 pm, and it is quite possible that the features that Venus shows near these wavelengths are those due to the wings of these carbon dioxide absorption bands rather than due to the particular hydration state of sul furic acid. Pollack et a!. (1974) stress this point in their article on the aircraft observations of Venus. Chemical Production of the HgSO^ Clouds The mode of production of H 2SO4 in the Venus atmosphere poses a complex problem. First of all, no sulfur compounds have been detected on Venus as gases. Upper limits to the mixing ratio of sulfur compounds are available, and their values are: SO 2 < COS< 10-®, and H28 < 10- 3.7 (Kuiper,: 1969). Two of these compounds can probably be ruled out because of their instability to ultraviolet radiation in the upper atmosphere of Venus. H 2S was exposed to the UV radiation of a quartz mercury pehlight (most effective radiation at the Hg wavelength 2536.5 A). Within minutes, notice able deposition of sulfur particles on the walls of the flask was observed as well as a fine smoke of S in the gas. Within an hour, decomposition was virtually complete. COS is almost as unstable to UV as H 28 , decomposing into CO and S . This decomposi tion also.approaches 100% in hours. Since the upper Venus atmosphere appears transparent to A > 2000 A, H 28 and COS can be ruled out. This leaves one with SO 2 and elemental sulfur as the most likely carriers of S to the upper atmosphere. 132 It should be noted that if sulfuric acid is the chief component of the Venus clouds, there will be a point in Venus' atmosphere where sul furic acid droplets falling downward will be consumed by other compo nents in the atmosphere. The first process would be that of simple evaporation in which the aqueous sulfuric acid solution of some 80%-90% would evaporate, first, by losing water vapor and, second, by having the sulfuric acid itself turn into vapor. As sulfuric acid loses water, the mixture would gradually become more concentrated until the azeotropic mixture of 98.3% by weight composition was obtained. The vapor at this point would consist of roughtly 10 parts sulfuric acid to 1 part water vapor by volume. The graph of the vapor pressure of sulfuric acid and of water vapor pressure both in equilibrium with different sulfuric acid compositions is shown in Figure 30. The top of the cloud deck is put at 250 K, and the bottom of the cloud deck at 450 K, as shown by the Venera 9 space probe (Marov et al. , 1973). The Venera space probe showed a sharp decrease in brightness measured through the cloud deck up to an altitude of 35 km, at which point, the flux density decreased less rapid ly from 35 km to the ground. This was interpreted by the Russian inves tigators as indicating that the cloud deck ceased to exist at altitudes below 35 km. Taking the usual lapse rate for Venus, this height of 35 km corresponds to a pressure of GOg of 10- 0-8 atm and a temperature of 450 K. In Figure 30, this is indicated by the vertical line at 450 K. A sulfuric acid solution near the azeotrope of 98.3% has a partial pressure of sulfuric acid and of water vapor that is close to the mixing ratio of water vapor at 450 K (10~3). while the intersection is not exact, it should also be recalled that the measurements of the water vapor 133 T°K 250* 300' 400° 450° 500° 0 - - 2 -=p Top of E clouds o CL - 4 CD O _J -6 Bottom of Cloud Deck -8 — -10 l/Tx 1000 Figure 30. Pressure of HgO in the Venus atmosphere and the vapor pressure of sulfuric acid solutions Vapor pressures of HgO and H 2SO4 for three concentrations are shown as a function of temperature . The partial pressure of H O in the Venus atmosphere is shown at two mixing ratios: x = 10"3 and x = 10“6 . The observed H 2O partial pressure on Venus corresponds to the vapor pressure of sulfuric acid solutions. 134 concentration have an uncertainty of approximately an order of magnitude. Therefore, it is certainly possible that the sulfuric acid simply evapo rates at this particular point, producing concentrations of H 2SO4 and water vapor such that their mixing ratios are 10~3. ^ But the sulfuric acid at this particular point is also subject to the chemical reactions outlined in the following equations: ' S(s,/) + 2H2S04 (/) = 2H20(g) + 3S02(g) (4,1) COS(g) + 2H2S0 4 (/) = 2H2 0 (g) + 3S02(g) + CO(g) (4,2) H2S (g) + 3H2S 0 4 m = 4H20(g)+ 4S02(g). (4,3) In reaction (4,1), sulfur which could possibly come about through de composition of carbonyl sulfide, reaction (2,17), would react with sul furic acid to produce water.vapor and sulfur dioxide. In reaction (4,2), the most likely sulfur species in the Venus atmosphere as computed in the previous chapter, carbonyl sulfide, reacts with sulfuric acid produc ing water, SO2, and CO. And, finally, hydrogen sulfide, which is com puted to be the second most abundant, species in the lower atmosphere, also reacts with sulfuric acid producing water and sulfur dioxide. There is also the further possibility that carbon monoxide could react with sul furic acid producing water vapor and 8 O2. At 450 K, each of these reac tions occurs with a negative change in free energy, that is, all the reactions readily proceed to the right. The log of the equilibrium con stant, log K, for reaction (4,1) at 450 K is 10.52, for reaction (4,2), 7.27, and for reaction (4,3), 19.50. Also at colder temperatures than this, they also would appear to go to the right; at 400 K each of these reactions has a very favorable equilibrium constant. 135 The question might arise as to why the clouds persist to this depth in the atmosphere where the temperature is 450 K. The. answer would reside first of all in the fact that the equilibrium constants were, computed for pure sulfuric acid in the liquid state and not for a solution, which is what exists in the Venus atmosphere. This solution stabilizes the reaction. For example, equation (4,1), using data obtained in the International Critical Tables, shows that for an aqueous solution of sul furic acid the above reaction has a log K greater than zero only at tem peratures higher than 430 K; whereas, computing the equilibrium constant with pure sulfuric acid at 430 K gives a value of log K equal to 10. How ever, the aqueous solution in the International Critical Tables is only 1 M, more dilute than 98.3%. A second factor may be the speed with which a reaction can proceed and the limited mixing ratio of the gases in question. Since apparently the sulfuric acid is as abundant as the carbonyl sulfide, the reaction will not run in the atmosphere until it has fallen through enough of the atmosphere to completely react with all the carbonyl sulfide. There are also kinetic effects that will limit the speed that the reaction can occur. It seems, however, to the author that it is not simply a co incidence that the vapor pressure of water from azeotropic 98% sulfuric acid matches so closely the mixing ratio of 10~3 that occurs at 450 K in the Venus atmosphere. It seems more likely that we are witnessing evaporation of the droplets. The sulfuric acid as it evaporates would react immediately with the sulfur species outlined in equations (4,1) — (4,3), producing sulfur dioxide. In effect, there is an overall oxidation of the sulfur species, particularly carbonyl sulfide and hydrogen sulfide 1 3 6 to sulfur dioxide, and a concomitant reduction of sulfuric acid, where sulfur is in the 6+ oxidation state, to the 4+ oxidation state of sulfur dioxide. The overall sulfur process appears to be a cyclic reaction of self-oxidation and reduction by different sulfur species. The oxidizing agent in the earth's atmosphere for converting SC>2 to SOg is O2. On Venus this is ruled out, and the most likely agents are molecular halogens , namely, F 2, CI2, Br2 , and I 2 F2 can be re jected because of the extreme chemical stability of HF. Likewise, CI 2 gas must be rejected because HC1 does not photolytically dissociate for o A > 2000 A. The spectra of HC1 solutions show no notable absorptions in the ultraviolet between 2000-4000 A. HBr and HI are strong absorbers. Both gases are readily dissociated into hydrogen gas and the molecular o halogen under Hg vapor irradiation (2537 A)., and a few hour's radiation brings about equilibrium. In the case.of HBr: h v . 2HBr(g)-.= H2(g) + B r2(g) (4,4) equilibrium quantities of the gases indicate that 7% o f HBr is decomposed in the gaseous equilibrium. The aqueous solution of HBr also shows a noticeable darkening due to formation of Br 2 under ultraviolet exposure. Mention was made of the bleaching effect of SO 2 gas dissolved in H 2SO4 when Br 2 fumes were poured onto the surface of the acid-S 02 solution. This apparently shows the ease at which Br 2 will oxidize SO 2 to H2SO4 in an acid medium. The thermodynamics also indicate a favor able reaction (Latimer, 1952): Br2 (i) + H2SO4 (aq) + H2O = 2Br~(aq) + 4H+(aq) + 8 0 ^" (aq) (4,5) E0 = +0.89 volts. 137 With a positive potential of 0.89 V, the reaction should and doe's pro- : ceed rapidly at room temperature. The two reactions can be summarized ’ as: 2HBr(g) = H2 (g) +Br2(g) (4,6) Br2(g) + S 02(g) + 2H20 = H2S04(/) + 2HBr(g). (4,7) The HBr is reconstituted in the process and, in effect, serves as a car rier catalyst for the reaction. The overall reaction therefore involves destruction of H20 into H 2 (g), which presumably escapes from Venus, and oxygen incorporated into H 2S04: S 0 2(g) + 2H20(g) = H2S04(/) + H2(g). (4,8) Reaction (4,8) is not favored thermodynamically and could never occur at the low temperature of the Venus clouds, but the intermediate reaction with Bi2 (and presumably the other halogen I2) make the process possible. None of the intermediate substances have been identified in the atmos phere of Venus, namely, SOg, HBr, and Brg . HBr is a difficult problem spectroscopically because its absorption bands at 2 and 4 jum fall into the heavy GOg"bands. Br 2(gas) was examined at the Lunar and Plane tary Laboratory with the helpful assistance of Allen Thomson and Thomas N. Gautier. Moderately strong absorption bands were found in the re gion of 5364 A. Sill, Gautier, and Kuiper (n.d.) attempted to detect gaseous Brg in the atmosphere of Venus , first with the 61-inch tele scope of the Catalina Observatory, then with the 107-inch telescope of the McDonald Observatory. Edwin S. Barker of the University of Texas, Austin, graciously aided us in obtaining high-re solution spectra at the Coude focus of the 107-inch telescope with its attendant echelle grating 138 spectrograph. Spectra of Venus, the sun, and the sun + Brg(gas) were o obtained in the region of 5364 A. The results were negative. From the spectra it is estimated that the smallest quantity of Brg that could be seen in Venus is about 0.13 mm-atm. An upper limit to the Venus mixing ratio Brg/GOg can be set at 10~7 • 3. This implies a very small quantity of Brg vapor, much less than the 4 mb vapor pressure of elemental Brg at 235 K. On the other hand, Brg is very soluble in cold H 2SO4, esp e cially in the presence of HBr, with a consequent lowering of its vapor p ressu re. Both Brg and SO 2 have mixing ratios with upper limits at 10“^. This may simply be a sign of their high reactivity with each other. The reaction is fast and involves reconstituting HBr. Since no upper limit on HBr has been established, the whole postulated mechanism cannot be proved at this time. The empirical data discussed above indicate that the visible and ultraviolet albedos of Venus may be consistent with Br 2 dissolved in H 2SO4 . The highly sensitive test for gaseous Br2 is not applicable to dissolved Brg as the sharp lines of the gas are smeared into the broad band of a liquid. Chemical calculations would lead one to suspect that HBr is a gaseous constituent of the atmosphere of Venus, chiefly because HF and HC1 are present. In general, one halogen implies the presence of the others, taking due account of their chemical reactivity. A.perfect case in point is the relative abundance of HC1 and HF in the Venus atmos phere, with mixing ratios of 10~6.0 ancj io-8.0/ respectively. The. Cl/F value in the atmosphere is 100, whereas the elemental abundance ratio Cl/F is 5.5 (Suess and Urey, 1956) or 1.8 (Cameron, 1973). Some : 139 process is making F less abundant jn the atmosphere of Venus. Sea water on earth shows a similar disparity with a Cl/F of 7 ,000. In this latter case, the explanation is due to the higher solubility of chlorides versus fluorides, plus the relative stability of fluoride micaceous miner als versus the chloride. As far as Venus is concerned, the relative solu bilities of chlorides and fluorides in water have no bearing on the case, but rather the chemical stability of chloride and fluoride minerals, in the presence of a hot COg atmosphere. Pertinent buffer reactions for the halides were discussed in the previous chapter: reactions (3 , 51) —(3 ,53), (3.58) —(3 , 60). The mixing ratios of the hydrogen halides were summa rized in Table 8 . Equilibria (3,52) and (3,53) have a Cl/F value of about 30. The phlogopite equilibrium (3,58) has a very low mixing ratio for HF, but data are not available for chlorphlogopite. Equilibrium (3,55), a pos sible analogy to hydrated micas, indicates HC1 is also about 30 times more abundant than HF. Table 8 also indicates that HBr mixing ratios are similar to those of HC1, within half an order of magnitude. It is possible that HBr might be more abundant than HC1 in the Venus atmosphere if equilibria with calcium minerals (3,52), (3,53), (3.59) are the dominant influence in atmospheric abundance. Chlorine in the earth's oceans amounts to 8.6 x 10^2 moles, whereas in.Venus' atmosphere there is about 1.3 x 10^ moles. Unless initial severe de pletion of Cl occurred on Venus, it should be retained in the surface minerals. Whe one considers Br on Venus, there might appear to be a problem of the absolute abundance of Br. The Cl/Br value in the earth's oceans is 660 (Suess and Urey, 1956., cosmic abundance ratio is also 660). If the absolute abundance of Br on Venus matches that in the earth. 140 there is more than enough Br to make the equilibrium mixing ratios of Table 8 applicable; and HBr could be more abundant in the atmosphere than HC1„ Hi abundance is limited to its absolute abundance, with a mixing ratio of perhaps 10- ^„ Laboratory Simulation of H 2SO4 Production Tests were conducted in the laboratory to see if HgSO^ could be formed by the irradiation of HBr and sulfur compounds. Acid solutions of HBr and various S gases, particularly SOg and COS,, were irradiated with the quartz Hg lamp. COS vigorously decomposed, producing large quantities of elemental sulfur. Some H 2SO4 was produced, which was analyzed by precipitation as BaSO^ on the addition of BaClg solution. The conversion of COS to H 2SO4 was about 3%. SO2 showed slightly more conversion to H 2SO4 . In 24 hours irradiation, about 5% SO 2 con verted. Without HBr, conversion to H 2SO4 by SO 2 was negligible. The amount of BaS 04 precipitated was minute . Even elemental Br 2 and solid sulfur mixed together in aqueous solution showed a high conversion to H2SO4 , but since this reaction occurred in the open air undoubtedly dis solved O 2 facilitated the reaction. Therefore, H 2SO4 is readily produced under postulated Venus atmospheric conditions . The lack of detectable sulfur compounds in the atmosphere of Venus has always been a problem. Whereas the atmosphere of Earth is dominated by O 2 and H 2O, the Venus atmosphere could be dominated by by the halogens. The lack of gaseous sulfur compounds is explained by postulating that the clouds of Venus are frozen droplets of 86 % H2SO4 , formed by the oxidation of SO 2 by elemental Br 2 in the upper atmosphere of Venus. 141 Ultraviolet Dark Markings The ultraviolet photographs of Venus taken by Mariner 10 have been described by Murray et al. (1974) as showing dark areas with vari able lifetimes . Some small areas disappear within minutes, whereas other general patterns may persist for hours or days. The dark area in the famous Y region seems to be a perennial source of ultraviolet dark material. What is required for the ultraviolet dark clouds seems to be some substance which forms at perhaps a lower level in the Venus at mosphere and then dissociates or evaporates at the upper level. The ab sorption of ultraviolet light could be caused either by an inherently ultra violet dark material or one that is photolytically decomposed to produce a new material with ultraviolet absorbing properties. The absorbing material need not be ultraviolet black in the re gion of the ultraviolet photographic filter (3400-400 A) but could have a non-zero reflectivity, varying with wavelength. The overall spectral re flectivity of Venus reproduced in Figure 2 shows a stronger absorption at o o 3000 A than at 4000 A, this despite Rayleigh scattering of the atmos phere, which would enhance the reflectivity of the planet at shorter wavelengths. The ultraviolet absorber ideally should show the same o • characteristic reflectivity as the planet: lowest near 3000 A, gradually » O increasing toward 4000 A with a minor peak near 2500 A. It should also show absorbance in the blue region of the visible spectrum to be com- ' patible with the overall yellow color of the Venus clouds. If an aqueous, solution of hydrobromic acid is irradiated, the solution acquires a yellow color, rapidly grading to reddish brown under further exposure. This is due to the formation of the polybromide ion. 142 whose formula can be expressed as Brg- , or Br^'Brg. The overall reac tion is hv 3HBr (aq) = HBr-Br2(aq) + H2(g). (4,9) Figure 31 shows the transmission spectrum of this acid solution (approxi mately 0.02 mm thick) obtained by a Beckman DK-2A dual-beam spec trophotometer. The ultraviolet absorption maximum occurs near 3000 A, o with a relative absorption minimum near 2400 A. This compares favor ably with the Venus ultraviolet reflectivity (Wallace et al. , 1972; Irvine et al. , 1968). By comparison, the spectra of hydrogen bromide gas and bromine vapor are also shown in Figure 31. The combination of these two gases in acid solution shows ultraviolet absorption properties different from the two gases due to Brg" ion formation and its consequent chargg transfer band with water. The ultraviolet dark markings on Venus have a limited lifetime and apparently are on the verge of stability. If these clouds are due to hydrobromic acid solution, then the droplets must form in equilibrium with their vapors: HBr(aq) = HBr(g) + H20(g). . (4,10) The relative vapor pressures of the two gases depend on the concentra tion of the aqueous HBr solution. In general, the more concentrated the acid, the higher the vapor pressure of the acid species and the lower the water vapor pressure. Figure 32 shows the vapor pressure of gaseous • HBr and H20 in equilibrium over acid solutions of various concentrations at various temperatures, obtained from data in the International Critical Tables. Also plotted on the graph are two points for the presumed vapor 143 100 HBr (gos) 80 oc b 60 h- o' VENUS/ Ij 20 FILTER 0.2 0.3 0.4 0.5 0.6 /xm Figure 31. Transmission spectrum of various species of bromine from 0.2 to 0.6 }im Illustrated are the transmission spectra of HBr gas, 1 cm-atm; Brg vapor, 1 mm-atm; and a solution of Brg in hydrobromic acid, liquid layer, 0.02 mm thick. The Bond albedo of Venus is shown by the dotted line. The band-pass filter of the Mariner 10 UV photographs is shown by the dashed line. 144 T(°K) 2 3 0 2 5 0 2 8 0 333 O 5 0 mb = P venus cn X E O 2m b E CL O 3 -2 - 3 - 4 .0 0 4 5 .0 0 4 0 .0035 .0 0 3 0 l/T Figure 32. Vapor pressures of hydrobromic acid solutions and of ice at various temperatures The vapor pressure of HgO and HBr in equilibrium with three dif ferent concentrations of hydrobromic acid is shown. At 230 K, the 2 mb pressure of Venus is plotted, and at 250 K (level of cloud tops) the pres sure of 50 mb. Also at 250 K is shown the 50 x 10~3 mb of water vapor pressure and the 50 x 10“^ mb of HBr pressure predicted in the Venus atmosphere. The 52% HBr solution has vapor pressures compatible with Venus. 145 pressure of HBr and HgO in the Venus atmosphere at 250 K, with mixing ratios of 10~4 and 1 , respectively. While the mixing ratio of HgO in the upper atmosphere is closer to 10~6, in the lower atmosphere the data from the Venera missions gave mixing ratios in the range of 10"2 to 10“3. The droplets would have to form in the lower atmosphere where the water vapor pressure is higher and gradually dissipate in the drier upper atmosphere. The evaporation of the liquid drops would explain the ephemeral nature of the ultraviolet dark clouds. The average mixing ratio HBr/CC>2 in Table 8 is approximately 10~4 . With mixing ratios of HBr at 10~4 arid HgO at 10~3, one can just barely make a solution of 52% HBr at 250 K, as Figure 32 indicates. The case is borderline, as perhaps it should be. The hydrobromic acid solution with dissolved bromine would form the sulfuric acid clouds, as discussed above. The process is ex tremely rapid. A reddish solution of HBr"Brg is immediately bleached white by the addition of sulfur dioxide gas. The resulting solution con tains sulfuric acid. Therefore, where the atmosphere of Venus contains . an excess of HBr, the clouds formed would have a yellowish ultraviolet absorber, HBr-Brg. Where SOg is present in excess, the clouds would be sulfuric acid with a different ultraviolet absorber present: dissolved SC>2. This latter solution does not have a visible absorption wing (see Figure 28). The differences in absorption could well explain the subtle differences in the ultraviolet coloring. The filter on the spacecraft, as o shown in Figure 31, favors absorptions in the 3400-4000 A region, where HBr• Br2 is more efficient. 146 A firm constraint on models of the clouds of Venus is the refrac tive index of the cloud particles. The refractive index at A = 0.55 jam is 1.44 + 0.02. Measurement of 52% HBr solution, extrapolated to 250 K, yields a value of 1.46, within the permissible limits of error. It is in teresting that sulfuric acid and hydrobromic acid have virtually the same refractive index. This cannot.be said for other substances postulated to .... be the ultraviolet darkening material, such as elemental sulfur, for ex ample , with a refractive index of 1.96. Furthermore, sulfur is solid, not a liquid, at 250 K. The value of the refractive index of Venus' clouds, determined by polarization, is considered so precise by Hansen and Hovenier (1974) that they reject the contribution of discrete solid par ticles or of solid particles nucleating the acid droplets. Apparently the ultraviolet darkening agent is either dissolved in the sulfuric acid drop lets or is in droplets of refractive index similar to sulfuric acid. Hydro- bromic acid solutions meet this requirement. Conclusions The halogen acids HC1 and HF are known constituents of the atmosphere of Venus and have been proposed as possible cloud materials (Lewis, 1972a; Hapke, 1972). The halogen acid HBr may be more abundant in the atmosphere than HC1. As such, it could form acid drop lets with water if the water mixing ratio is — 10"^. This acid is easily decomposed by ultraviolet light to produce a yellow or reddish-brown solution of HBr•Br 2 in water of approximately 52% by weight composition. Droplets of this solution would contribute to the overall yellow color of Venus, as well as to the ultraviolet darkening of the planet between 147 3000 and 4000 A. The droplets would have a temporary existence, evap orating slowly in the'upper, drier atmosphere of Venus where the water mixing ratio is 10- 6. The ultraviolet dark material, the Brg- ion, can also be destroyed by sulfur dioxide, the bromine oxidizing the SOg to sulfuric acid. The abundance of hydrobfomic acid droplets could be substan tial locally in the atmosphere, since the polarization measurements of the clouds of Venus yield a refractive index that is compatible not only with sulfuric acid of 86 % composition but also with hydrobromic acid of 52% composition. Probably the main constituent of the clouds is sulfuric acid, because it is compatible with the low vapor pressure of water, whereas the hydrobromic acid is out of equilibrium with water vapor in the upper atmosphere. The overall ultraviolet coloring of the planet is due to the interplay of bromine and sulfur dioxide. High-resolution spec troscopy down to 2500 A of the dark areas of the planet in comparison with the light areas would indicate whether SOg with its sharp feature o centered at 3000 A predominates in the light clouds and whether HBr• Br 2 o with its broad feature at 3000 to 4500 A dominates in the dark areas . CHAPTER 5 EVOLUTION OF THE VENUS ATMOSPHERE The planets Earth and Venus have similar masses and radii but differ substantially in many other respects. Earth's surface is dominated by the presence of liquid water, has a mean surface temperature of 288 K, and possesses an atmosphere of predominantly nitrogen and oxygen. Venus' surface is radically different in these respects: absence of liquid water, a surface temperature of 750 K, and a massive atmosphere of car bon dioxide. The planet Venus was formed closer to the Sun than the Earth. It has received twice the flux of solar energy that the Earth has received. What circumstances in the evolutionary history of the planets have produced such different surfaces? In this chapter, the effective temperature of a planet will be discussed. Since the surface of a planet may be covered by an atmos phere , the actual surface temperature can vary widely from the calcu lated effective temperature. Radiative cooling of the surface and atmosphere can occur during the nighttime, giving rise to the presence of cold traps on the planet's surface. The hypothesis of Rasool and De- Bergh (1970) on the evolution of the atmospheres of Venus and Earth will be discussed and its shortcomings delineated. Finally, the author will present a different approach to the evolution of the Venus atmosphere. 148 Surface Temperatures of Planets The effective temperature for an airless, rapidly rotating plan etary surface is given by 4 *—;-----:— TG ° Te = V Y/ p* (5-1) where Tq is the ground temperature, Te the effective temperature, $ is the solar flux, <5 is the Boltzmann constant, and Ap is the planetary al bedo. Appendix 2 discusses the derivation of the equation. Solving this equation with the average solar flux 1.4 x 10® erg cm- 2.sec'- ^ and an albedo of 0.07 gives the value of 275 K for the earth. The sub solar point would approach 389 K if the earth had this lunar-type albedo. The equiv alent temperatures for the surface of Venus would be an average planet surface temperature of 321 K and for the subsolar point a value of 454 K. These temperatures are only for a planet without an atmosphere, which is a blackbody radiator with a low albedo, as considered by Rasool and DeBergh (1970). The higher the albedo, the lower would be the tempera ture; e.g ., if Venus had an albedo of 0.80, the average temperature would be only 218 K. ■ At the present time, Earth has a moderately thick atmosphere of standard pressure, whereas the Venus atmosphere is 100 times as dense. This atmosphere affects the equilibrium surface temperature in various ways. First, because the solar illumination is strongest in the region of the visible spectrum, any atmosphere that is transparent to visible radiation will permit most of the solar illumination to arrive at the sur face of the planet. This is mostly the case on Earth. In the case of 150 Venus, although GOg is transparent to visible light,, nevertheless cloud particles and droplets show a very high albedo permitting only a small fraction of the illumination to reach the surface. For example, at the. surface of Venus, the estimated solar illumination is roughly 2% of that impinging at the top of the atmosphere (Lads and Hansen, 1974). Earth clouds are not as dense as Venus' and permit approximately 60% of sun light to reach the surface . There is another complication: the transparency of the atmos phere to the surface reradiation. Since the planetary surface is thousands of degrees cooler than the illuminating surface of the sun> most plane tary surfaces tend to radiate most efficiently in the far infrared around 10-30 pim. This radiation can be either trapped by molecules or aerosols in the planetary atmosphere and partially reradiated back to the surface. The main infrared gaseous absorber in Earth's atmosphere is water vapor with a small contribution from nitrous oxide with an absorption band near 5 and 6 pm. In the case of Venus, the chief infrared gaseous absorber in the atmosphere is GOg with its strong Tg band at 15 pm and 1^3 band at 4 pm. The chief aerosol infrared radiator in Earth's atmosphere is water in the form of droplets or ice particles; on Venus, its own very thick cloud deck is apparently responsible for much of the greenhouse effect. Therefore, even though the Venus surface itself receives only 2% of the incoming solar flux, the infrared radiation that is absorbed in the cloud aerosols and the COg under high pressure gives rise to a very high surface temperature of 750 K. In the case of Earth, the mean 151 effective temperature is only increased by 33 K by the presence, of the atmosphere. The upper Earth atmosphere is an absorber of ultraviolet radia- O ' tion at wavelengths shorter than .3600 A due to the formation of ozone at a height of 25 km. The dissociation of oxygen and of water vapor into atomic oxygen.gives rise to the ozone layer that effectively absorbs all the radiation of /\ < 3600 A ., The inability of ozone and the other common gases.at that temperature and altitude, namely, Og and Ng, to radiate at infrared wavelengths permits the upper atmosphere of Earth to rise to a high temperature. In the case of Venus, comparable amounts of ozone and oxygen have not been detected; nevertheless , the atmosphere is a o good ultraviolet absorber, especially around 3000 A. This could be due to the presence of hydrobromic acid solution. Venus also absorbs radia- o tion of X < 2000 A due to the presence of hydrogen bromide, sulfur diox ide, and hydrogen chloride. While Earth's atmosphere is not as efficient in absorbing these wavelengths, nevertheless radiation shortward of O 1600 A is absorbed by oxygen and water vapor as well as scattered from the upper layer of the atmosphere. At the present time, the surface of Earth is dominated by oceans of water and an atmosphere rich in Og. These two components are most efficient in trapping the products of volcanism and the chemical weather ing of rock. Halogen compounds are efficiently trapped in solution in the ocean, and other volcanic gases, such as oxides of carbon and sulfur, are trapped in the sedimentary deposits. Carbon compounds (except those sequestered from oxidizing effects of air) go to carbon dioxide and are trapped as insoluble carbonates of calcium and magnesium. Sulfur 152 oxide, SO 3 , is trapped primarily by calcium in the form of gypsum and anhydrite. The only element that is not efficiently trapped in this cold, wet environment is nitrogen gas, which constitutes the bulk of the at mosphere. It is also commonly assumed that the oxygen content of the atmosphere is due to photosynthesis, since Earth's crust is sub-oxidized; it is on the border between ferrous and ferric states. It is assume that all of the oxygen in the atmosphere would be removed by the oxidation of ferrous iron to ferric iron, if photosynthesis did not regenerate the oxy- . gen from water and CO 2. The cold, wet surface of Earth is contrasted with that of the planet Venus which is very hot and very dry. Under these conditions, trapping of volatile volcanic emanations is not as efficient as in the presence of liquid water; the chemical weathering actions proceed rather slowly and also at higher equilibria concentrations of the gaseous com ponents, as seen in Chapter 3. In the final state, an equilibrium partial pressure of CO2 approaches 100 atm and we find that residual water of roughly 0.1 atm pressure is present totally in the atmosphere as a gas, as well as the acid halide g ases, HC1, HE, and HBr. Under these same circumstances, the solid sulfur compounds are moderately stable at the surface and are in equilibrium with sulfur gases, particularly H 2S and COS, at partial pressures of 0.01 atm. Radiation Cooling of a Planet's Surface On Earth, the cooling of the planetary surface generally occurs at night and also in the regions of the terrestrial poles. In general,, the surface of Earth tends to lose the heat trapped by rocks, soil, and bodies 153 of water by infrared radiation with a peak at 10 jam; Most of this radia tion escapes through the 8-14 jam window from the planetary surface ex cept when it is trapped by water vapor or droplets in clouds; COg is un important. In general, the drier the climate, the cooler the nights. This is particularly true in the case of large dry areas like the Sahara Desert that may have diurnal variations of 30-40 C. In the case of Earth's poles where the absolute humidity is very low, there is also pronounced cooling by radiation in the long winter nights. The coldest temperature has been recorded in the vicinity of the South Pole, approximately -80 C. Planetary cooling is modified by the fact that Earth spins on its axis in 24 hours, a period short enough so that the radiation absorbed by the at mosphere during the day warms the surface at night, with the conse quence of pronounced cooling being seen only in the polar regions. Under present conditions obtaining on Venus, the thermal radia tion from the planetary surface is effectively trapped by two main con stituents, high-pressure GOg and the clouds. Venus rotates in a retrograde direction in 245 days. This gives rise to a day and night se quence oh Venus, each of which is 57 days long. With a nighttime dura tion of 57 days, it might be presumed that the planet could radiate a large quantity of heat through infrared radiation. However, this loss of heat is hindered by high heat capacity of the dense atmosphere. With these extremely efficient infrared trappers giving rise to the so-called "greenhouse effect," the surface never has a chance to cool down during the night. Venera probes that landed on the night side of Venus gave a temperature virtually identical to the daytime temperature, and the Mari ner atmosphere profiles obtained for nighttime and daytime exits and 154 entries, as well as entries and exits at various latitudes, indicate that the surface temperature appears to be variable only over tens of de grees,. if that. With very efficient convection in the atmosphere and in the presence of strong infrared absorbers, the whole planetary surface stays at the very high temperature of 750 K. The cold trap on Earth does serve to trap large quantities of water in the form of ice. This is the only volatile species preferentially absorbed in Earth's cold trap. Most of the ice is deposited on land masses, Greenland and Antarctica. The ice is limited in quantity by two effects . In the colder regions of Earth near the poles, the vapor pressure of water is only a few millibars, and this gives rise to an essentially arid type of climate with little snowfall. Furthermore, the land masses upon which the ice is deposited are surrounded by oceans which absorb glacial runoff. At present, there is equilibrium between the rate of depo sition on the ice caps and the amount of removal by the relatively warm ocean water. This serves to keep the cold trap from becoming planet- wide . During the winter, the amount of surplus snow and ice trapped at the poles is not available to raise the albedo of the planet and con sequently to reduce the solar insolation, because the polar regions are turned from the sun. On the other hand, when the poles turn toward the sun in the summer, the snow cover melts with a consequent lessening of albedo, and the ground absorbs heat more efficiently, evaporating the water. Under present atmospheric conditions, an equilibrium exists be tween the amount of snow deposited on the polar caps and the amount removed through evaporation and melting. The quantity of snow is 155 variable and depends on other factors: the amount of aerosols in the at mosphere, a change in solar output, or the amount of GO 2 in the atmos phere . At the present time, there is no effective cold trap on the planet Venus, except that which exists at the top of the atmosphere. However, an atmospheric cold trap does not hold large quantities of volatile ma terials but just the small amounts of water and sulfur compounds that are trapped in the sulfuric acid clouds. Under current conditions, there fore, there is nothing analogous on Venus to Earth's poles where large quantities of volatiles could be trapped, particularly water or GOg. This is distinctly different from the conditions on planet M ars, for example , where the polar ice caps appear to be a mixture of frozen CO 2 and water. Hypotheses for the Evolution of the Venus Atmosphere Urey's Model The evolution of the Venus atmosphere was a concern of Urey (1952), who considered that the beginning stages of the production of an atmosphere were not too different on Venus than on Earth. He considered that the volatile species initially outgassed from Venus / primarily water vapor and CO 2, were effectively trapped by reactions with the surface, giving rise to the formation of limestone and other carbonate deposits. These deposits eventually gave rise to the present high-pressure CO 2 atmosphere of Venus. Urey did not specifically consider the problem of the water vapor content of Venus because little was known at that time about the absolute abundance of water. Rasool and DeBergh's Model The greenhouse effect and the abundance of CO 2 on Venus were treated more fully by Rasool and DeBergh (1970). They assumed that the initial atmosphere of Venus contained CO 2 and water vapor , with the water vapor being four times as abundant as CO 2, by analogy with Earth. They used equation (5,1) to determine the effective temperature of a body rapidly rotating in an essentially airless environment to determine the initial temperature of the planetary surface. For Venus, the initial tem perature was computed to be approximately 330 K and for Earth 265 K, *' . . ■ ■ assuming a lunar albedo of 0.07. As the atmosphere accumulates, the CO2 and water vapor cause the opacity of the atmosphere to increase. The greenhouse effect begins to operate and the temperature of the planet rises. Assuming a certain optical thickness to the atmosphere, using an approximation of an average absorption coefficient for water and CO 2, they found that, for Earth, the accumulation of water vapor reached a point where its pressure intersected the gas-liquid phase boundary of water at 10-2 atm of H2O. At.that point, liquid water would begin to accumulate on the surface of Earth. This liquid water would dissolve CO2 and at the same time facilitate the reaction of CO 2 with typical ig neous minerals, such as pyroxenes, to; trap the CO 2 as carbonates of calcium or magnesium. With the water vapor content limited by its equi librium vapor pressure at the temperature in question, i.e ., near 280 K, and with CO 2 dissolving in the liquid water and also reacting with the rocks, it is evident that the atmosphere of Earth could never reach a high density and its optical thickness would remain small. The green house effect is moderate and does not run away because the density of 157 the atmosphere is limited to relatively small amounts of GOg and water vapor pressure (10-20 mb). 'The surface of Earth remains relatively cool, near 280 K. In the case of Venus, however, since the initial surface tem perature begins at 330 K, the initial outgassed CC>2 and HgO would go into the atmosphere and the pressure of water vapor would have to be 0.2 atm in order to accumulate an ocean. This point never arises be cause the increasing abundance of the gaseous species CO 2 and H 2Q gives rise to an increasing optical thickness producing a higher surface temperature; and this reaction escalates. Rasool and DeBergh call it the runaway greenhouse effect. The atmosphere would continue to accumu late until at 450 K the CO 2 gas would react with minerals, such as wol- lastonite, leading to the removal of CO 2 in the form of calcium carbonate. From this temperature-pres sure condition , the surface temperature and pressure would begin to rise with the partial pressure of CO 2 following the wollastonite equilibrium until it reaches its present state of 750 K and 100 atm of CO 2 pressure. Rasool and DeBergh solved for the ground temperature by using the Eddington gray atmosphere approximation: Tg4 = Te4(l + 3/4 To) (5,2) where Te is the effective temperature and T 0 is the optical thickness of the atmosphere defined as r OD- To = / KP dz J ° . where K is the average absorption coefficient, p is the density of the absorbing gas, and z is the altitude. 158 In this model for the evolution of Venus' atmosphere, it is nec essary to remove a vast amount of water, i.e ., some 400-500 atm. This is accomplished by dissociation of the water into hydrogen and oxygen by solar ultraviolet radiation. Hydrogen escapes from the planet and oxygen remains in the atmosphere where it reacts with oxidizable materials, such as iron. If the near-surface rock contains 8 % ferrous iron, oxidiz ing to ferric iron, surface rock to a depth of 100 km would have to have been processed by weathering in order to remove this quantity of oxygen. This amount of weathering oh Venus appears excessive. Earth has not exposed such a quantity of rock to weathering by the atmosphere. Donahue's Objection to the Rasool-DeBerqh Model. Serious exceptions can be taken to this model for the evolution of the Venus at mosphere. The ability to lose large amounts of hydrogen by dissociation of water vapor should not be presumed. Donahue (1969) undertook an analysis of the problem of removing hydrogen from the Venus atmosphere. He stated that the main problem in losing a hydrogen atmosphere is not the mechanism by which the hydrogen escapes but rather the rate of pro duction of hydrogen, presumably by photodissociation of hydrogen com pounds like water. The rate of production of hydrogen is dependent bn the mixing ratio of the water vapor in the atmosphere and the production rate of the dissociation of water into hydrogen atoms and OH radicals. O Photodissociation of water begins at wavelengths less than 1869 A. If other gases are present, there is competition for photons in the dissoci- O ation process. Oxygen begins to absorb at 1800 A and COg begins to O absorb at 1700 A. Some of the hydrogen atoms produced are removed by 159 recombination and the rest diffuse upward to the exosphere where they e sc a p e . If.Venus has lost as much water as is now present in Earth's oceans, the escape flux during the course of Venus' lifetime would be IQl! atoms cm“2 sec“V. On Earth, the rate of dissociation is 10^ atoms cm-2 sec- * . On Venus, there is sufficient solar flux to dissociate 2 x 10*2 water molecules cm-2 sec- *, if the water did not have to compete with other gases like Og and CO2 . The COg abundance is very high in Rasool and DeBergh's model of the Venus atmosphere and would absorb practically all photons of A <: 1700 A. This leaves the water with a band . '' o opening of only 160 A in which to absorb photons. Any oxygen produced by the dissociation of GOg and HgO would also serve to narrow the band further to 60 A. The dissociation of water would be an almost self- limiting process due to the ultraviolet absorption caused by Og. But if the oxygen is removed by reaction with surface minerals at a rate equal to its production in the atmosphere, the only absorption left would be that due to CO 2. This would not be negligible, and Donohue states that it is difficult to understand how the photodissociation rate could approach the level of 10** molecules cm -2 sec- * unless the atmosphere was free even of CO 2. He concludes that this process is possible only if the Venus atmosphere differed radically in the past from its present condi tion . Critical Dependence of Surface Temperature on Albedo. Another problem confronting the model of Rasool and DeBergh is that the effective temperature is critically dependent on the albedo. A value of 330 K cor responds to an albedo of 0 .07, typical of the airless planets, such as 160 Mercury and the moon & If one uses for this equation a value for the pres ent Venus albedo, 0.71, one Obtains a value for the effective temperature on Venus equal to 244 K. The current high albedo of Venus is due to its cloud cover. Planets with a thin atmosphere and minor cloud cover have lower albedos. (Mars, for example, has a thin atmosphere with an al bedo of 0.17.) What was the initial albedo of the planet Venus and did it change in the history of the planet? The albedo of 0.07 for Mercury and the moon is due to the darkening effect of solar ultraviolet radiation, solar wind ionization, and the production of small dark glass spheres due to micrometeorite bombardment. In the case of Mars, the albedo is higher because, while the atmosphere is virtually transparent to near ultraviolet and visible radiation, it is not transparent to micrometeorite bombardment and ionizing radiation and lunar-type darkening effects do not occur. In order to produce very high albedos (0.7 and 0.8), it is necessary to have a planet that is either totally cloud covered or con sists mostly of ices or other white compounds . The satellite lo, whose albedo is around 0.85, has its surface covered by a highly reflective ma terial, perhaps sulfur or a mixture of sodium sulfate and sodium chloride. A New Hypothesis for the Evolution of the Venus Atmosphere The Problem of the Removal of Oxygen Rubey (1951, 1955) pointed out that an initial dense atmosphere for Earth was very unlikely, especially one that retained the relative cosmic abundance of volatile components . For example , he mentioned the severe depletion of the noble gases in the present atmosphere of 161 Earth, as well as the relative depletion of water and gaseous carbon compounds. The present atmosphere and hydrosphere of Earth had to be outgassed from the interior of the planet. Rubey considered various hy potheses on the rate of planetary outgassing and concluded that it was more likely that the rate of outgassing has been relatively constant dur ing geologic time, without however denying the possibility of a more vigorous outgassing rate early in Earth's history. Rasool and DeBergh (1970) implicitly accepted Rubey 1 s model for the outgassing of Venus . The author generally follows this same model. , Apparently most of the water vapor produced by the volcanic outgassing on Venus has been dissociated into hydrogen and oxygen. This is a conclusion based on the relative amounts of water in the oceans of Earth (1.5 x 1C)24 g ) ancj in the atmosphere of Venus (5 x 10)20 g). Hydrogen would readily escape from Venus' exosphere, but oxygen would remain. Since the present atmosphere of Venus contains virtually no oxygen gas, the oxygen produced by the photodissociation of water vapor must have been trapped in surface rock by means of some oxida tion process. If iron is chosen as the reducing agent, being oxidized from ferrous to ferric, e .g ., in reaction (3,18) , a depth of 1 km of crustal rock of basaltic composition must be utilized to remove 4 atm of oxygen. In their model for the evolution of Venus' atmosphere, Rasool and De Bergh postulate the removal of 400 atm of oxygen dissociated from water. If iron is the reducing agent for the oxygen, weathering of rock to a depth of 100 km is implied. Nordlie (1971) investigated the composition of volcanic gases at Kilauea and concluded that the single magmatic gas mixture that is 162 capable of producing the collected volcanic gases has equal proportions of HgO, CC>2 and -SO' 2. If the volatiles outgassing from Venus were of this composition, another mechanism presents itself for removing oxygen from the photodissociation of water. Each mole of SOg could remove 1 mole of O atoms: ■ S 0 2(g) + I / 2 02(g) = SQ3(g). (5,3) Sulfur trioxide would readily attack surface minerals producing sulfates: CaMgSi20 6(s) + 2S03(g) = C aS04(s) + M gS04(s)- + 2S i02(s). (5,4) diops ide sulfur anhydrite magnesium quartz trioxide sulfate Since SO 3 readily hydrolyzes to sulfuric acid, the following reaction is also likely: CaMgSi20 6(s) + 2H2S 04 (aq) = CaS04 (s) + M gS04(s) diop side sulfuric anhydrite magnesium acid sulfate + 2Si0 2 (s) + 2H20 (f). (5,5) quartz water If the outgassed volatiles contain, equimolar HgO and SOg , all of the oxygen from dissociated water vapor can be trapped as sulfates, pro vided enough surface rock is weathered. There is competition for min erals containing Ca and Mg from carbon dioxide, which would also be removed from the atmosphere in the form of calcium and magnesium car bonates. In this scheme, half the Ca and Mg would form carbonates and half sulfates. How much rock must be weathered to remove all the volatiles? Assuming a crustal composition like that of tholeiitic basalt (Table 5) , each cubic centimeter of rock contains 0.31 g of CaO, 0.19 g of MgO, and 0.30 g FeO. The earth has trapped about 90 atm of CO 2 as 163 carbonates, requiring the removal of 93 kg of GOg/cm^. The amount of GaO and MgO in a 2.1 km layer of basalt can accomplish this. A similar quantity of 802 in the form of sulfate would require another 2.1 km layer of basalt. If GOg, H2O, and SOg outgas in equimolar proportions, 100 ' atm of GO2 outgassed from Venus (114 kg of C02/c m 2) implies 180 atm of SO3 (SO2 plus the O from H2O), or 205 kg of SO3/cm^, must be de posited as carbonate and sulfate, respectively. This requires a layer of basalt 5.0 km thick. If ferrous iron is also utilized to remove oxygen, the thickness of the layer is reduced to 4.2 km. Volatiles of equimolar CO2, H2O , and SO 2 composition would lead to surface deposits of sulfates and consequently to a more oxidized state for the present atmosphere of Venus; the oxygen fugacity wop Id be closer to the higher value discussed in Chapter 3, namely, 10 ~*8 atm . The presence of ammonia (detected by Venera 8 ) would be excluded by this oxygen fugacity. Since sulfates are relatively more stable than sul fides, as indicated in Chapter 3, gaseous sulfur species would be less abundant in the presence of surface sulfates unless the surface buffer involved magnesium sulfate: MgS0 4 (s) + CO(g) = MgCOs (s) + SO2 (g). . . (5 , 6). magnesium carbon magnesite sulfur sulfate monoxide dioxide If the CO partial pressure is 10~3• 99 atm (equation 3,21), the SO 2 par tial pressure would be io~0.14 atm/ 0r the SO 2/CO 2 mixing ratio would be 10-2.14. The Rasool and DeBergh model postulates a H 2O/CO 2 (weight) of 4/1. Mechanisms for removing the oxygen from 400 atm of water are hard to imagine. Even removing the oxygen from 100 atm of water is 164 difficult unless SOg is also an abundant outgassed volatile„ In the fol lowing model of atmospheric evolution, the author has varied the value for HgO/GOg from, l/l to 1/4 without producing notable differences in the final result. The "Nordlie" equimolar mixture has weight proportions of 1 :2 .4 :3 .6 for . These proportions are also compatible with the calculations performed. Cold Trap on Venus Let us assume that the initial outgassing of volatiles from Venus .was from a rocky surface with low albedo (0.10). Before the volatiles outgassed, the surface temperature was approximately equal to the effec tive temperature: 450 K at the subsolar point, gradually falling to a value of 164 K near sunrise and sunset (solar angle 1° above the horizon). Since the thermal inertia of rock is high, rapid cooling of the surface occurs after sundown. The surface of a slowly rotating Venus is probably similar to the lunar surface where the radiative temperature is 100 K at the anti-solar meridian and 90 K near the sunrise terminator (Mendell and Low, 1975). The initial products of outgassing would condense on the night side of Venus. Condensation of CO 2 and H. 2O would, release the latent heat of sublimation (600 cal/g for H 2O and 110 cal/g for CO2), raising the temperature of the surface layer. If the surface is. warmed to 120 K in this process, each, square centimeter of surface could radiate 1500 cal during the course of the Venus night (57 days' duration). This is the amount of heat that must be dissipated in order to condense 12.6 g of CO2 or 2 .3. g of H2O or 4.0 g of a 1:1 H2O-CO2 mixture. Since the heat energy radiated follows the Stefan-Boltzmann law and is proportional 165 to T^, the amounts of the. condensates would be doubled if the surface temperature were raised to 143 K. At sunrise, the evaporating CO 2 and H 2O would be advected back to regions of lower pres sure--pole ward and toward the night side. When the quantities of outgassed CO 2 and H 2O are each equal to 0.5 g/cm2, permanent condensation of HgO has occurred at the poles at latitudes greater than 85°. The amount of heat received from the sun during the day, 3885 cal, is sufficient to evaporate the 0.5 g of CO 2 but is only able to raise the ice temperature to 168 K, at which tempera ture its vapor pressure is negligible. The ice frost has changed the al bedo of the poles. With the change in albedo (ice frost is conservatively assigned an albedo of 0 .80) , the ice cap, begins to grow from the poles and toward the equator. The extent of the polar cap depends on the H 2O frost deposited during the night. As the ice cap migrates to lower lati tudes, evaporation of H 2O vapor occurs during the daytime and the cap is permanent only where the ice layer is of sufficient thickness . For example, the ice layer is permanent at 60° latitude only when the layer of ice exceeds 1.50 g/cm2 because solar radiation is able to evaporate 1.50 g/cm2 during the Venus day. (The temperature rises to 256 K in the early afternoon.) When the frozen layer of volatiles is 7 g/cm2 of H 2O frost and the same amount of CO 2, the ice cap can withstand evaporation down to 43° latitude. ' The polar caps cannot continue the expansion toward the equa- . tor beyond 43° latitude since the temperature at these latitudes exceeds the melting point of ice. The presence of liquid water causes the albedo to fall to low values, with consequent evaporation of the liquid. The 166 surface of the frost-free ground between 43° north and south latitudes becomes very warm (300-400 K), setting up a strong circulation of the gases. Hadley cell circulation should be present with warm air moving toward the equator, where it rises and flows northward to the poles. The descending polar currents would then flow equator ward along the surface. The Hadley cell circulation should help to transport some of the volatiles toward lower latitudes. This would prevent the polar caps from perma nently sequestering the volatiles. Liquid water may also be produced at places on the surface other than the edge of the melting polar caps at.43° latitude. If 10 g each of H2O and CO 2 are condensed all over the night side, even in the tropic latitudes, then, when these regions pass the sunrise terminator at dawn, the 10 g of CO2 rapidly evaporate by the time the sun is 20° above the horizon. The surface temperature rises until H 2O begins to evaporate at approximately 220 K. Evaporation and warming of the ice layer con tinues; at a solar elevation of 80° the temperature reaches 273 K, 7 g of ice having evaporated. The final 3 g of ice rapidly melt and subsequently evaporate in about 12 hours. At sunset, the ambient water vapor produced .by melting of the polar caps, as well .as the vapor produced in morning . evaporation at lower latitudes , may be sufficient to cause the formation of dew near sunset. The condensing dew may persist for 50 hours before it freezes . A temperature-pres sure profile at the surface of Venus is illus trated in Figure 33 for the above model with 10 g each of CO 2 and H 2O frozen on the night side of the planet. The band of liquid water produced by melting frost and the condensation of dew is indicated by stippling. | | LIQUID WATER Figure 33. Surface of Venus early in its history under 10 g of HgO and 10 g of CO 2 per cm2 Day hemisphere is from +90 to -90 longitude; subsolar point 0,0. The dashed lines are isotherms (in 0K, labeled at top). Pressures of CO 2 and HgO vapor at these temperatures are labeled at bottom. A band of liquid water is stippled. Albedo within the band of water is 0 . 2, outside the band 0. 8 . 168 Isotherms in degrees Kelvin are delineated on the upper half of the globe, and isobars in millibars on the bottom of the globe. The latitude and longitude lines are based on the subsolar point at 0, 0 . Note that con densation of 10 g of CC>2 is complete near the anti-solar meridian (180°). Circulation effects due to the changes in pressure as the volatiles evap orate are not considered in this model. In reality, the isobars should be shifted to indicate a smooth grading of pressure from the sunlit to the dark hemisphere. Temperatures within the stippled areas are based on an albedo of 0.20 (typical rock). Chemical Weathering by COg In the model presented in Figure 33, condensation Of CO 2 is complete on the dark hemisphere with the temperature reaching 100 K (vapor pressure of CO2 is 2 x 10- ^ mb) . As more CO2 is outgassed, more must be condensed on the night side, thereby raising the tempera ture . The warmer surface can radiate more heat and condense proportion ately more CO 2. (There is a limit to this process: when the CO 2 vapor pressure is high enough to produce an atmosphere opaque to infrared radiation, condensation on the surface ceases.) The average temperature of the polar regions and the night side rises, and the pressure of CO 2 b e comes equal to the vapor pressure of CO2 in the cold trap. This is the situation that exists on Mars where the CO 2 pressure is about 6 mb, equal to the vapor pressure of CO2 at 147 K, the temperature of the polar caps. The increased pressure of CO 2 on Venus would provide greater opacity to the atmosphere as well as carry heat from the daytime tropics to the poles and night side. The.temperature all over the planet would rise to the point where CO 2 would no longer condense. A runaway green house effect, as Rasool and DeBergh describe it, would follow. There is another process, however, that can remove CO 2 from the atmosphere so that the pressure never reaches the runaway value. On Earth, carbon dioxide dissolved in rainwater weathers rock in arid environments to produce abundant deposits of caliche, calcium carbon ate. The rate of production of caliche depends on the nature of the sur face rock as well as on the amount of precipitation. A. Long (personal commun., 1975) has dated terrestrial caliche deposits by the tech nique and estimates that caliche is formed at the rate of about 1 mm/5000 years where the annual rainfall is 4 inches . The thickness of the layer grows at 1 mm/1000 years- where the annual rainfall averages 8 inches, and it obtains a maximum rate of 1 mm/200 years where the precipitation is 15-20 inches per year. In regions where the annual rain fall exceeds 25 inches, caliche is not deposited due to the solubility of the weathered products. In arid environments, the soluble bicarbonate of calcium can decompose to the insoluble carbonate by loss of water and carbon dioxide: Ca(HC02)2(aq) = CaCOsts) + H2 0 (g) + 0 0 2 (g). (5,7) A similar weathering process can be postulated for the early history of Venus. The liquid water produced at the edge of the Venus polar cap and in other areas of Venus, as depicted in Figure 33, is anal ogous to the intermittent rainstorms of arid regions on earth. Since cali che formation occurs on earth where the partial pressure of CO 2 is 0.3 mb, the removal of CO2 from the Venus atmosphere, whose postulated partial pressure is 5-7 mb, should be efficient. What is the expected 170 rate of calcium carbonate formation on Venus? Five grams of HgO melt ing every Venus "day" is the equivalent of an annual precipitation on Earth of 6 inches . This precipitation rate on Earth corresponds to caliche production of 1 mm/2000 yr. Since the postulated Venus pressure of CO 2 is 20 times the Earth's partial pressure, the reaction should be 20 times faster, except that the temperature of the water on Venus is close to the melting point of ice, 25 degrees cooler than water from rainfall. This should decrease the rate of the reaction by a factor of 6 . The rate of caliche formation is estimated as 1 mm/600 yr. If 65 percent of the Venus surface (between lat 40° N. and lat 40° S.) is available for caliche formation, the time required to remove 100 atm of CO 2 (5.2 x 10^3 g) would be about 1 x 10^ years. The rate of formation of carbonate appears capable of removing CO2 as it is outgassed from the planet. The carbonate produced by weathering is able to withstand the high temperatures of the Venus afternoon. At 400 K, the equilibrium par tial pressure of CO2 from the wollastonite equilibrium (3,1) is 10- 3.3 atm or 0.45 mb, lower than the postulated ambient CO 2 pressure of about 6 mb. Dolomite (equation 3,3) is much less stable, having an equilibrium partial pressure of 28 mb at 400 K. Dolomite would be stable at tempera tures below 370 K, or poleward of 32° latitude . Water Vapor in the Evolving Atmosphere At the same time that carbon dioxide is being fixed as a car bonate in the surface rock, water vapor is being dissociated by the solar ultraviolet flux producing oxygen. If the rate of photodissociation of water molecules is IflU molecule s/cm / • sec on Venus , then 100 atm of . 171 HgO would be dissociated in about 1 x 10^ y e a rs. The ambient oxygen produced in photodissociation would reduce the theoretical maximum rate of photodissociation, 2 x 10^2 molecules/cm2 • sec (Donohoe , 1969). Due to chemical reaction with SO 2 and surface rock, Og would be gradu ally depleted from the atmosphere. When water has been substantially depleted from Venus, perhaps after 10.9 years, there is no longer a de posit of ice thick enough to produce liquid water during the Venus day. The ice layer evaporates before melting can occur. At this point, weath ering processes involving CO2 dissolved in water (e.g., caliche forma tion) would cease. Production of the Dense Atmosphere The Venus crust would continue to outgas GOg and HgO, but at at ,a rate assumed to be less than the initial outgas sing from the planet. If photodissociation of water vapor can remove H 2O as fast as it is pro duced, CO 2 will continue to accumulate, condensing on the night side of the planet. Increased condensation on the night side will raise the surface temperature, producing higher CO2 vapor pressure. When the night side cold trap has Warmed to 163 K, the vapor pressure of CO 2 will be 45 mb. The opacity of the atmosphere increases and the greenhouse effect will begin to warm the surface . As the surface temperature rises, • some carbonates will begin to decompose. Both magnesite (3,4) and dolomite (3,3) equilibrium reactions produce a CO 2 partial pressure of 45 mb at 408 K. These carbonates will begin to decompose and release CO2 to the atm osphere. The increased CO 2 content and the resulting greater opacity of the atmosphere cause the surface temperature to 172 r is e „ The CO2 equilibrium partial pressure of decarbonation rises rapidly- leading to a runaway condition. For example, the CO 2 equilibrium par tial pressure of the dolomite reaction reaches 0.5 atm at 457 K. Even tually , the more stable calcite would begin to decompose near the tropical regions , according to reaction (3 ,1). The final state of the Venus atmosphere would depend on the total CO2 outgassed from the original caliche (calcite) deposits and the analogous dolomite and magnesite deposits. The final surface pres sure is determined by the temperature at the surface, which, in turn, is determined by the opacity of the atmosphere to the infrared radiation. Any CO 2 outgassed beyond 100 atm of surface partial pressure would eventually be removed by one or another of the buffer reactions repre sented in Figure 24. The atmosphere in evolving to its high temperature- high pressure state has caused the surface rock to not only evolve CO 2 from carbonate deposits but also H 2O from clays and micaceous minerals. Some water trapped in hydrated minerals was isolated from photo dissoci ation and probably constitutes the source for the residual H 2O in the at mosphere. The other volatiles soon came into equilibrium with surface minerals through various reactions discussed in Chapter 3. Summary The hypothesis outlined above seems to answer some of the questions raised by shortcomings of alternate hypotheses to the evolu tion of the Venus atmosphere. The massive CO 2 atmosphere which hin ders photodissociation of water vapor is not present until most of the water has already been dissociated. The CO 2 abundance is kept in check 173 both by the reaction with surface minerals in the presence of liquid water as well as by condensation on the night side of the planet. The CO 2 pressure cannot rise until liquid water is no longer present on the sur face. Photodissociation of water produces oxygen,, which is removed by reaction with SOg to form sulfate deposits and by the oxidation of ferrous iron to ferric in surface minerals. A crustal thickness of 5-10 km of basalt composition is required to remove the CO 2., O2 from H2O, and SO 2 volatilized by Venus . The drastic depletion of water on Venus versus the earth (a depletion factor of 3000) is not due solely to the original inventory of the two planets but to the more rapid removal of . H2 from Venus through more efficient photodissociation of H 2O. The present dense atmosphere of Venus is not produced until most of the water has been photolyzed, perhaps after 10^ years of the planet's his tory . CHAPTER 6 CONCLUSIONS Earth and Venus are similar to each other in their masses and presumably in their general geology as fair as the solid surface of the planets is concerned, but two factors have led to a vastly different evo lutionary history for each. The two factors are the distance of the planet from the Sun and the rotational speed of the planet about its own axis. In general. Earth is dominated by a surface composed largely of water, the depth of the oceans being, sufficient to produce 400 atm of vapor if evaporated. The other gases in the atmosphere are inert nitro gen and oxygen, the latter produced mostly by the photo synthetic pro cesses of plants. The GOg partial pressure, 0.3 mb, is closely con trolled by the solubility of COg in water and the fixing of CO 2 both by living creatures that produce calcareous shells and by abiological pro cesses. The atmosphere is transparent to visible radiation and is rela tively opaque to far-infrared radiation due to the absorption by water vapor in the upper atmosphere as well as by solid and condensed water in the form of clouds. The oceans serve as the predominant temperature moderator in the planet, restricting large excursions between the night and day temperatures as well as moderating the effects of the seasons. The oceans themselves are also responsible for removing frozen water from the polar caps and distributing it more equitably around the planet. 174 175 In Venus, we have a planet which is essentially dry in compari son with Earth. Instead of some 400 atm of water vapor or condensed water, there is roughly 0.1 atm. The COg that on Earth is dissolved in sea water and fixed in carbonate deposits is mostly in Venus' atmos phere. The atmosphere itself serves as the great equalizer of tempera ture on the planet because of the huge heat capacity of 100 atm of COg. The temperature variation between night and day is less than 0.4 percent despite the 117-day duration of the solar day. The thick atmosphere also serves to effectively keep the poles at temperatures not much different from the sub-polar point of the planet. The clouds of Earth are dominated by the presence of water, and the atmospheric temperature profile is also very heavily dominated by the presence of water modifying and reducing temperature extremes as water condenses and evaporates . The clouds are responsible for most of the high albedo of the planet Earth, and the average albedo is about halfway between that of the clouds and the oceans with some contribution from continental m asses. The formation of clouds and their products of pre cipitation, particularly snow and frost, gives rise to the changes in the albedo of the planet which serve to modify the temperature , making win ters colder than would be expected and producing large cold sinks at the poles of the planet. However, on Venus the water vapor content is never high enough to give rise to clouds or to any condensation of water, either in the liquid or solid state. Nevertheless, there are abundant clouds present on the planet that are due to other gaseous and condensable species, namely, compounds of sulfur. On Earth, these volatile species are trapped in the ocean waters or sediments, and the chemistry of 176 sulfur and halogens is restricted usually to the surface of the planet, due to the fact that sulfur compounds, are oxidized to sulfates and halo gens are sequestered in sea water and sediments. ' In contrast, there is no sink on Venus for storing halogens and sulfur compounds except those that might operate near the surface of the planet where high-temperature equilibria can prevail between certain compounds of sulfur and of the halogens and atmospheric components. Most sulfur in the Venus atmosphere is in equilibrium with either troilite or pyrite, and the two abundant gaseous compounds are carbonyl sulfide and hydrogen sulfide. The halogens are in equilibrium with the various possible surface materials, fluorine and chlorine in equilibrium with micaceous or clay-type minerals and bromine and iodine in equilibrium with sodium or calcium bromide or iodide. The most stable form of these halogens in the atmosphere is that of the hydrogen halides. The sulfur and halogen compounds have sufficient abundance to be considered as minor or semi-major components in the atmosphere of Venus. Upon be ing brought to sufficient elevation, i.e ., where the pressure is around 50 mb, halogen compounds are able to be photolytically decomposed by o ultraviolet radiation (less than 2500 A) to produce sulfur dioxide and sul- fur trioxide. From these compounds and water vapor, the clouds of sul furic acid are produced. Desiccation by concentrated sulfuric acid reduces the water vapor content in the atmosphere to parts per million. At the same time, the acid produces a very opaque, upper layer in the atmosphere such that with the addition of the CO 2 of the atmosphere, the overall optical thickness of Venus has a value of 86 , serving to make 177 the surface temperature of the planet 750 K versus Its. effective tempera ture of only 243 K. " The overall state of Earth's atmosphere is oxidizing due to the presence of molecular oxygen. It is not known what the exact state of oxidation in the Venus atmosphere is because some CO found in the upper atmosphere may be due to photolytic dissociation of CO 2. A partially re duced state in the atmosphere of Venus, with oxygen buffered by the olivine -magnetite equilibrium, is considered likely, both because of the presence of ammonia, discovered by Venera 8 , and because of the higher partial pressures of sulfur compounds under these conditions . The sur face pressure of oxygen is set at 10~24 atm. in this case, the species in equilibrium with the surface are carbon monoxide, carbonyl sulfide, . hydrogen sulfide, hydrogen, and hydrogen halide, with some ammonia . This state of oxidation is chosen instead of the conditions along the magnetite-hematite boundary because of the discovery of ammonia or an ammonium compound in the lower atmosphere of Venus, as well as the difficulty of absorbing large quantities of oxygen from the primeval water of Venus. If Venus had as much water as Earth, it would require that ferrous iron be oxidized to ferric iron to depths of 100 km in the crust. Since Earth has not exposed this thickness of rock to surface weather ing, the author prefers a weathering of 5-10 km of crust, the oxygen from 100 atm of H2O vapor being reduced by iron and sulfur . The atmosphere of Venus in its current state is much different than the atmosphere of Earth because of the lack of water and the great abundance of CO 2. The evolutionary development of the planetary at mospheres of Earth and Venus was discussed, and it was concluded that 178 the main reason why Venus has its present atmosphere with low water vapor content is due primarily to the fact that the slow rotation on its axis, with the nighttime duration of 57 days, gives rise to a very effi cient cold trap on the night side of the planet. In the early history of the planet's outgassing, it is easy to see that water vapor can condense completely on the night side of the planet. Conditions may arise where by carbon dioxide itself can condense on the night side at temperatures near 145 K and at ambient pressures of 5-7 mb of COg. This sink for atmospheric volatiles serves to keep the atmosphere thin, prevents the formation of a thick CO 2 gaseous atmosphere, and produces a planet similar to Mars. ' Under these conditions, water will be photolyzed into hydrogen and oxygen near the surface of the planet. Hydrogen escapes from the - planet, and oxygen dissolves in the melting ice of the planetary surface when the icy coating is exposed to sunlight as the planet rotates . The freezing and melting during a daily Venus cycle would enable CO 2 and oxygen to be efficiently removed through aqueous reactions with the surface rock. The CO2 is removed from the atmosphere as carbonates, and oxygen as iron oxides or sulfates. The outgassed volatiles of the planet are kept in either the frozen state or as solids until practically all of the water has been eliminated by photolytic decomposition. With the present-day solar flux, photolytic decomposition of H 2O will be Complete after approximately a billion years of the planet’s history. Assuming that most of the volatile outgassing occurred early in its history, the water would be depleted to such an extent after a billion years that ice would not be able to survive the planetary daytime ( . x . 179 solar flux. The albedo of the planet would decrease as the ice melts, and the atmosphere would become warmer, reducing the effectiveness of the cold trap on the backside of the planet. At pressures near 1 atm, the nighttime temperature would be so high that.even water could no longer condense on the night side. As the atmospheric mass increased, carbonates, and hydrated minerals would begin to decompose releasing CO2 and H 2O, As more volatiles were driven into the atmosphere, the atmosphere would become more opaque to the infrared radiation of the planet. Water vapor and CO 2 would build up until an equilibrium was established between the rate of outgassing and the ambient pressure. This state is apparently reached at the present time, with pressures of 100 atm and a temperature of 750 K. The chief reason why Venus is as dry as it is is because that in its early history there was no thick atmosphere of COy. There is dif ficulty in photolyzing water diluted in a high CO 2 atmosphere. Rather, the CO2 pressure must be very low in order for photolysis to work effi ciently. In the early history of Venus the CO 2 pressure was restricted to a few millibars because CO2 partially condenses on the night side and is removed by aqueous chemical weathering on the day side. : The author hypothesizes a relatively early abundant H 2O out- gassing, comparable in magnitude to Earth's outgassing. This is in contradistinction to other models, e.g ., Lewis's (1972b, 1974), which postulate a low H2O abundance from the onset of Venus' formation. Us ing Cameron's (1969) model of an early solar system adiabat, Lewis finds that Earth condensed at a temperature of about 620 K and Venus at 900 K. He shows that Earth is at the edge of the stability field of 180 tremolite (Ca2Mg5Si8022(OH)2) and could have acquired its primordial water in the form of this mineral. In this model, Venus condensed be yond the field of stability of tremolite and consequently would have a low abundance of water. As indicated in Chapter 3, thermodynamic data on hydrated minerals seem not wholly reliable. The stability of hydrates varies from mineral to mineral and depends on the type and amount of "impurities" present. To base the primordial water abundance on a plan et on the theoretical stability of one hydrated minerals does not appear to be prudent. The author offers his hypothesis as an alternative solu- . tion for the problem of a "dry" Venus. . APPENDIX 1 REACTION RATES AND EQUILIBRIUM Thermodynamics is capable of determining the concentrations of the reactants and products at equilibrium. It also can predict how the concentrations of the reactants and products vary with temperature. How ever, nothing can be said of the rate of the reaction from thermodynamic calculations. Reaction rates are empirically determined and then fitted to a model that describes the behavior of the system. The order of the reaction is so determined. In many cases, the speed of a reaction and its order can also be predicted, with more or less success, by analogy to systems that have been empirically verified. For chemical reactions that involve oxidation-reduction, where chemical bonds are broken and re-formed, the reaction speed is critical ly dependent on the temperature of the system because chemical species —the activated complex—must be produced 'that possess a higher energy than the simple reacting species. Chemical bonds must be broken and re-formed through the activated complex. The energy required to produce the activated complex can come from a variety of sources, such as heat, light, or electric discharge. Photo lytic reactions in the upper atmos phere of a planet are activated by the energy of ultraviolet and visible light. For reactions occurring near the surface of a planet, such as Venus, where light energy is virtually nil, the activation energy can be only ther mal. The temperature of the surface of the planet and gas in contact 181 182 with it indicates the average energy of the molecules. But there is a distribution in the energies of the molecules such that a certain guantity of the molecules have less energy than average and others more energy than average. The more energetic are of interest in determining the speed of the chemical reaction because these are the molecules that will pro vide the activation energy for the reaction„ The percentage of molecules that possess a certain energy state in a gas at a given temperature is computed from the Maxwell distribution equation (Glasstone, 1949): ^ = ' 0?kT)3/2 S ^ 6 1/2 dC where n = number of molecules T = temperature (Kelvin) £ = energy (ergs) k = Boltzmann constant (1.3805 x lO- -*-^ erg °K~^). This equation gives the number of molecules with kinetic (translational) energy between € and £ + d €. The equation is most sensitive to the ex ponential term. The. number of molecules of a given energy range, between and € + d€ (which can be the activation energy), can be computed for various temperatures. Since T is in the exponential term, the number varies rapidly with small temperature changes. For example, in a mole of gas at 750 K, the number of molecules with an energy of 2eV to 3eV (46.to 69 kcal/mole) is 2.13 x 10^; at 760 K the number increases to 3.13 x 1012 „ This effect has been stated in the chemist's rule of thumb: raising the temperature 1 0 C doubles the rate of a chemical reaction. 183 The activation energy for a given reaction can be computed if the rate of a reaction is known at two temperatures, provided that the same mechanism is responsible for the reaction at both temperatures and provided that all other conditions are similar (in a gas-solid reaction, for example, the size of the solid particles could be critical) . Provided pressure remains constant, the reaction rate in a first-order reaction de pends on the population of molecules having the required activation energy. This is a function of temperature. If a first-order reaction reaches equilibrium in one hour at Tj and in 50 hours at Tg (a lower tem perature) , the activation energy can be computed from the Maxwell dis tribution equation by successive approximations; that is, by determining which activation energy gives a population of molecuibs > dn, of energy between c and e + de, which is 50 times greater at Tj than at Tg. This computed activation energy may then be used at other temperatures, T 3, for example, to compute a new dn population. The speed of the reaction at Tg is proportional to dn at Tg. APPENDIX 2 DERIVATION OF EQUATION FOR EFFECTIVE TEMPERATURE OF A PLANET The flux emitted by the sun at 1 A.U. is 2 cal cm- 2 min- *, or 1.4 x 10® erg cm~2 sec"l. This flux is subject to theinverse square law. A planet will absorb all the radiation presented by its cross- sectional area if it is a blackbody. If a planet, like Venus, reflects 71% of the solar radiation (albedo = 0.71), it only absorbs 29% of the flux and only this quantity heats the planet. Therefore, energy absorbed = (Trr^) (solar flux) (1 - albedo) (A2,1) or E = 7rrp2 jzf (1 - Ap). The heated surface radiates according to the Stefan-Boltzmann law: flux (energy cm ~2 se c - ^) = dT^ (A2, 2) where C is the Stefan-Boltzmann constant (5.67 x 10~® erg cm- ^ deg-^ . sec- !). The whole surface of a planet is 4Trp2, and it all radiates to space, so . energy radiated = 4irr 2 c T^ . (A.2,3) P Equating equations (A2 ,1) and (A2 ,3) and solving for T, we obtain the effective temperature , Te: For a surface normal to solar radiation (a part of the surface near the 184 185 sub-polar point .on a slowly rotating planet), the factor of 4 would be eliminated and the temperature could rise to this value: T6 = _Jl!ES. Similarly, since solar insolation in a first-order sense varies as the co sine of the latitude, the polar regions of a planet could have effective temperatures near absolute zero. REFERENCES Adams, W. S., and Dunham, T., 1932, Absorption bands in the infrared spectrum of Venus: Astron. Soc. Pacific Pub., v. 44, p. 243- 247. Anhaeusser, C. R., Mason, R., Viljoen, M. J ., and Viljoen, R. P., . 1969, A reappraisal of some aspects of PreCambrian shield geology: Geol. Soc. America Bull., v . 80, p. 2175-2200. Avduevsky, V. S., Marov, M. Y.., and Rozhdestvensky, M. K., 1969, model of the atmosphere of the planet Venus based on results of measurements made by the Soviet automatic interplanetary , station Venera 4, in Jastrow, R., and Rasool, S. I., eds., The Venus atmosphere: New York, Gordon and Breach, p. 79- 104. Barath, E. T., Barret, A. H ., Copeland, J ., Jones, D. E ., and Lilley, A. E., 1964, Mariner II microwave radiometer experiment and results: Astron.. Jour., v. 69, p. 49-58. Barth, P. F. W. , 1962, Theoretical petrology, ed. 2: New York, John Wiley & Sons. Beer, R., Norton, R. H ., and Martonchik, J. V., 1971, Absorption by Venus in the 3-4 micron region: As trophy s. Jour. Let., v. 168, p. L121-L124. Belton, M. J., and Hunten, D. M ., 1968, A search for O 2 on Mars and Venus: Astrophys .Jour., y. 1.53 , p. 963-974. Belton, M. J., Hunten, D. M. , and Goody, R. M., 1968, Quantitative spectroscopy of Venus in the region 8000-11000 A, in Brandt, J. C ., and McElroy, M. B., eds., The atmosphere of Venus' ■ and Mars: New York, Gordon and Breach, p. 69-97. Bird, G. W ., and Anderson, G. M ., 1973, Free energy of formation of magnesian cordierite and phlogopite: Am. Jour. Sci., v . 273, p. 84-91. Bottema, M ., Plumer, W ., and Strong, J., 1964, Water vapor in the at mosphere of Venus: Astrophys . Jour., v. 139 , p.. 1021-1022. Bowen, N. L ., and Tuttle, O. F., 1949, The system MgO-SiOg-^O: Geol. Soc. America Bull., v. 60, p. 439-460. Boyd, F. R., 1959, Hydrothermal investigations of amphiboles, in Abe Ison, P. H., ed.. Researches in geochemistry: New York,. John Wiley & Sons, p. 377-399. 186 187 Burch, D. E. , Grynak, D. A. , Patty, R. R ., and Bartky, C. E ., 1969, Absorption of infra radiant energy by CO 2 and H 2OV IV. Shapes of collision-broadened CO 2 lines: Optical Soc. America Jour. , v. 59, p. 267-280. Cameron, A. G. W ., 1969, Physical conditions in the primitive solar nebula, in Millman, P. M. , ed., Meteorite research: Dor drecht, Reidel, p. 7-15. ______1973, Abundance of the elements in the solar system: Space Sci. Rev. ,. v . 15, p. 121-146. Clark, S. P., 1966, High pressure equilibria, in Clark, S. P., ed.. Handbook of physical constants: New York, Geol. Soc. America, p. 346-370. Coffeen, D„, 1968, A polarimetric study of the atmosphere of Venus: un- publ. dissertation. University of Arizona, Tucson. ______, and Hansen, J. E., 1972, Polarization studies of planetary atmospheres, in Gehrels, T., ed. , Planets, stars and nebulae studied with photopolarimetry:. Tucson, University of Arizona Press', p. 518-581. Cbnnes, P., Connes, J., Benedict, W. S., and Kaplan, E. D ., 1967, Traces of HC1 and HE in the atmosphere of Venus: Astfophys. Jour., v. 147, p. 1230-1237. Connes, P., Connes; J. , Kaplan, L. D., and Benedict, W. S., 1968, Carbon monoxide in the Venus atmosphere: Astrophys. Jour., v. 152, p. 731-743. Cotton, F. A., and Wilkinson, G., 1966, Advanced inorganic chemistry, ed. 2: New York, Interscience. Cruikshank, D. P., 1967, Sulfur compounds in the atmosphere of Venus. II, An upper limit for the abundance of COS and H 28 : Lunar Planet. Lab. Commun. , v. 6,. no. 98, p. 199-200. ______, and Kuiper, G. P., 1967, Sulfur compounds in the atmosphere of Venus. I, An upper limit for the abundance of SO 2: Lunar Planet. Lab. Commun. , v . 6 , no. 97, p. 195-197. Cruikshank, D. P., Morrison, D ., and Lennon, K., 1973, Volcanic gases: hydrogen burning at Kilauea volcano, Hawaii: Science, v. 182, p. 277-279. Damon, P. E., Shafiqullah, M ., and Lynch, D., 1973, Geochronology of block faulting and basin subsidence in Arizona: Geol. Soc. America Abstracts with Programs, v. 5, p. 590. 188 Dayhoff, M. D. , Eck, R. V ., Lippincott, E. R., and Sagan, G., 1967, Venus: atmospheric evolution: Science, v. 155, p. 556-558. Deer, W. A., Howie, R. A., and Zussman, J ., 1962, Rock forming minerals, Vol. 3: London, Longman. Dollfus , A., 1961,. Polarization studies of the planets , in Kuiper, G. P., and Middlehurst, B., eds., Planets and satellites: Chicago, University of Chicago, p. 343-399. Donohue, T. M., 1969, The upper atmosphere of Venus: a review, in Jastrow, R., and Rasool, S. 1., eds.. The Venus atmosphere: New York, Gordon and Breach, p. 511-529. Evans, D. C ., 1967, Ultraviolet reflectivity of Venus and Jupiter, in Dollfus, A., ed., Moon and planets: Amsterdam, North Holland Publishing Co., p. 135-148. Fink, U., Larson, H ., Kuiper, G. P., and Poppen, R. P., 1972, Water vapor in the atmosphere of Venus: Icarus, v. 17, p. 617-631. Fornarev, U. I., and Ivanov, I, P., 19.72, Equilibrium of pyrophyllite and muscovite dehydration reactions at Ptot = PH 2 0 = 1000-8000 kg/cm2: Dokl. Akad. Nauk SSSR, v. 205, no. 3, p. 704-706; abstract in Chem. Abs ., v. 77 , no. 116657p. Gillett, F „ C ., Low, F. J ., and Stein, W. A., 1968, Absolute spectrum Of Venus from 2.8 to 14 microns: Jour. Atmos. S ci., v. 25, p. 594-595. Glasstone, S ., 1949, Theoretical chemistry: New York, Van Nostrand. Goettel, K. A., and Lewis, J. S., 1974, Ammonia in the atmosphere of Venus: Jour. Atmos. Sci., v. 31, p. 828-830. Goldschmidt, V. M., 1912, Die gesetze der gesteinesmetamorphOuse, Videnskapsselskapet Skrifter, I, Mat.. Naturv, Klasse, p. 12- 23. Hanel, R., Forman, M., and Stambach, G., 1968, Preliminary results of Venus observations between 8 and 13 microns: Jour. Atmos . Sci., v. 25, p. 586-593. Hansen, J. E., and Arking, A., 1971, Clouds of Venus; evidence for their nature: Science, v. 171, p. 669-672. Hansen, J. E., and Hovenier, J. W ., 1974, Interpretation of the polar ization of Venus: Jour. Atmos. Sci., v. 31, p. 1137-1160. Hapke, B., 1972, Venus' clouds: a dirty hydrochloric acid model: Science, y. 175, p. 748-751. 189 Harker, R. I ., and Tuttle, O. F., 1956, Experimental data on the Pc02""'^ curve for the reaction: calcite + quartz wollastonite + carbon dioxide: Am. Jour. Sci., v. 254, p. 239-249. Heald, E. F., Naughton, J. J ., and Barnes, I. L., 1963 , The chemistry of volcanic gases: 2. Use of equilibrium calculations in the in terpretation of volcanic gas samples: Jour. Geophys . Res., v. 68, p. 546-557. Hilsenrath, J ., 1960, Tables of thermal properties of gases; comprising tables of thermodynamic and transport properties of air.;, argon, carbon dioxide, carbon monoxide, hydrogen, nitrogen, oxygen and steam: New York, Pergamon Press . Holland, H. D ., 1964, On the chemical evolution of the terrestria! and Cytherean atmospheres, in Brancazio, P. J. , and Cameron, A. G. W ., eds., The origin and evolution of atmospheres and oceans: New York, John Wiley & Sons . Howard, H. T., Tyler, G. L„, Fjeldbo, G ., Kliore, A. J., Levy, G. S., Brunn, D. L ., Dickinson, R., Edelson, R. E., Martin, W. L ., Postal, R. B., Seidel, B., Sesplaukis, T. T., Shirley, D. L ., Stelzried, G. T. , Sweetnam, D. N . , Zygielbaum, A. I. , Espo- . sito, P. B., Anderson, J. D., Shapiro, I. I., and.Reasenberg, R. D., 1974, Venus: mass, gravity, atmosphere, and iono sphere as measured by the Mariner 10 dual-frequency radio system: Science, v . 183, p. 1297-1301. Huang, W. H ., and Keller, W. D., 1972, Standard free energies of formation calculated from dissolution data using specific min eral analyses: Am. Mineralogist, v. 57, p. 1152-1162.- Irvine, W. M., Simon, T ., Menzel, D. H ., Pikoos, C., and Young, A. T., 1968, Multicolor photoelectric photometry of the brighter planets. Ill, Observations from Boyden Observatory: Astron. Jour., v. 73, p. 807-828. JANAF, 1970, Thermochemical tables, ed. 2 (D. R. Stull and R. Prophet, project directors): U.S. Dept, of Commerce. Jenkins, E. B., Morton, D. G., and Sweigart, V. S., 1969, Rocket ul traviolet spectra of Venus and Jupiter: Astrophys . Jour., v . 157, p. 913-924. Kuiper, G. P., 1947, Infrared spectra of planets: Astrophys. Jour., v. 106, p. 252-262. ■______1949, Planetary atmospheres and their origin, in The atmos phere s and their origin, in Kuiper, G. P., ed ., The atmospheres of the earth and planets: Chicago, University .of Chicago Press, p. 306-405. 190 Kuiper, G. P., 1962, Infrared spectra of stars and planets. I. Photom etry of the infrared spectrum of Venus , 1-2 „ 5 microns: Lunar Planet. Lab. Commun., v. 1, no. 15, p. 83-118. _1969 , Identification of the Venus cloud layers: Lunar Planet. Lab. Commun., v. 6, no. 101, p. 229-250.. ______, and Cruikshank, D. P., 1964, Laboratory spectra for testing the presence of minor constituents in planetary atmospheres. I. CH4 , NH3 , N2O, CO, COS, region .1-2.5 microns: Lunar Planet. Lab. Commun., v . 2, no. 34, p. 141-165. Kuiper, G. P., and Forbes, P. P., 1968, High altitude spectra from NASA CV-990 jet. I. Venus, 1-2.5 microns, resolution 20 cm- *: Lunar Planet. Lab. Commun., v. 6, no. 95, p. 177-189. Kuiper, G. P., Forbes, P.P., Steinmetz, D. L ., and Mitchell, R. I ., 1969, High altitude spectra from NASA CV-990 jet. II. Water vapor on Venus: Lunar Planet. Lab. Commun. , v . 6, no. 100, p. 209-228. Kuiper, G. P., Sill, G. TV, and Cruikshank, D. P., 1967, The infrared spectrum of carbon suboxide. Part. I: region 1-2.5 microns: Lunar Planet. Lab. Commun., v. 6, no. 99, p. 201-203. Lacis, A. A. , 1975, Cloud structure and heating rates in the atmosphere, of Venus: Jour. Atmos. Sci., v . 32, p. 1107-1124. ______, and Hansen, J. P., 1974, Atmosphere of Venus: implications of Venera 8 sunlight measurements: Science , v. 184, p. 979- 982. Larson, E. S. , 1942, Approximate mineral compositions of principal types of plutonic igneous rock, in Birch, P ., Schairer, J. P. , and Spicer, H. C ., eds V Handbook of physical constants: New York, Geological Society of America, p. 3 . Latimer, W. M ., 1952, The oxidation states of the elements and their, potentials in aqueous solutions: New York, Prentice-Hall. Lewis, J. S., 1968a, Composition and structure of the clouds of Venus: Astrophys. Jour., v. 152, p. L79-L83. 1968b, An estimate of the surface conditions of Venus: Icarus , v . 8, p . 434-456. 1970, Venus: atmospheric and lithospheric composition: Earth Planet. Sci. Letters, v. 10, p. 73-80. 1972a, Composition of the Venus cloud tops in light of recent spectroscopic data: Astrophys. Jour., v. 171, p. L75-L79. 191 Lewis, J. S„ , 1972b, Metal/silicate fractionation in the solar system: Earth Planet Sci. Leters, v . 15, p. 286-290. 1974, The temperature gradient in the solar nebula: Science, v. 186, p. 440-443„ Lilley, A. E., 1961, The temperature of Venus: Astron. Jour., v . 66, p . 290. Lyot, B., 1939, Research on the polarization of light.from planets and from some terrestrial substances: NASA TT F-187, 144 p; Ann. Obs., Paris, section de Meudon, Trans., v. 7, no. 1. Marov, M. Y., 1972, Venus: a perspective at the beginning of plan etary exploration: Icarus, v . 16, p. 415-461. ______, Avduevsky, B. S . , Kerzhanovlch, V. V ., Rozhgestvensky, * M. K., Borodin, N. P., and Ryobov, 0. L ., 1973,. Venera 8 : measurements of temperature, pressure and wind velocity on the illuminated side of Venus: Jour. Atmos. S ci., v. 30, p. 1210-1214. Mayer, C. E ., McCullough, T. P., and Sloanaker, R. M., 1960, Ob servations of Vehus at 10.2 cm wavelength: Astron. Jour., v. 65, p. 349-350. ______1962, 3.15 cm observations of Venus in 1961: Soc. roy. sci. Liege Mem., serie 5, v. 7, p. 357-363. .. McElroy, M. B., Sze, N. D., and Yung, Y. L. , 1973, Photochemistry of the Venus atmosphere: Jour. Atmos. Sci., v. 30, p. 1437- 1477. Mendell, W. W ., and Low, P. J., 1975, Infrared orbital mapping of lunar features: 6th Lunar Sci. Conf., Proc., p. 2711-2719. Moroz, V. I., 1965, New observations of the infrared spectrum of Venus (A - 1.2-3 .8 p): Astron. Jour., v . 8, p . 566-572 . Mueller, R. P., 1963, Chemistry and petrology of Vehus: preliminary deductions: Science, v. 141, p. 1046-1047. ______1969, Planetary proble: origin of the Venus atmosphere: Science , v. 163, p. 1322-1324. M urray, B.C., Belton, M. J ., Danielson, G.E., Davis, M. E ., Gault, D, Hapke, B., O'Leary, B., Strom, R. B., Suomi, V., and Trask, N ., 1974, Venus: atmospheric motion and structure from Mariner 10 pictures: Science, v. 183 , p. 1307-1315. 192 Nockolds, S. R., 1954, Average chemical composition of some igneous rocks: Geol. Sbc. America Bull.., v. 65, p . 1007-1032. Nordlie , B. E ., 1971, The composition of the magmatic gas of Kilauea . and its behavior in the near surface environment: Am. Jour. Scii, v. 271, p. 417-463. Pettijohn, F. J ., 1957, Sedimentary rocks, ed. 2: New York, Harper Bros Pollack, J. B., Erickson, E. P., Witte born, F. C., Chackerian, C., Summers, A. L ., Van Camp, W ., Baldwin, B. J., Augason, G. C ., and Caroff, R. J. , 1974, Aircraft observations of Venus' near infrared reflection spectrum: implications for cloud composition: Icarus, v. 23, p. 8-26. Pollack, J, B., and Sagan, C., 1965, The microwave phase effect on Venus: Icarus, v. 4, p. 62-103. Pravda, 1969, June 4, reported in Aviation Week and Space Technology, June 23, 1969, p. 70-73. Prinn, R. G ., 1973, The upper atmosphere of Venus, in McCormac, B. M., ed ., Physics and chemistry of upper atmospheres: Boston, Reidel, p. 335-344. Rasool, S. I., arid DeBergh, C ., 1970, The runaway greenhouse and the accumulation of COg in the Venus atmosphere: Nature, v. 226? p. 1037-1039. Robie, R. A ., and Waldbaum, D. R., 1968, Thermodynamic properties of minerals and related substances at 298.15°K (25.0°C) and one atmosphere (1.013 bars) pressure and at higher tempera ture s: U.S. Geol. Survey Bull. 1359. Rubey, W. W ., 1951, Geologic history of sea water: Geol. Soc. Amer ica Bull., v. 62, p. 111-1175. ______1955, Development of hydrosphere and atmosphere, in Poldervaart, A., ed.. Crust of the earth: Geol. Soc. America Spec. Paper 62, p. 631-650. Sagan, C., 1968, Origins of the atmospheres of the earth and planets , in Runcorn, K ., ed ., International dictionary of geophysics: . London, Pergamon Press, p. 97-104. Schorn, R. A., Barker, E. S., Gray, L. D., and Moore, R. C., 1969, . High dispersion spectroscopic studies of Venus. II. The water vapor variations: Icarus, v. 10, p. 98-104. 193 Sill, G. T., 1972, Sulfuric acid in the Venus clouds: Lunar Planetary Lab. Commun., v. 9, no. 171, p. 191-198. ______Gautier, T. N., and Kuiper, G. P., n.d., unpublished data: Tucson, Lunar and Planetary Laboratory. Spinrad, H., 1966, Resolution of a CO 2 "hot band" in the Venus spec trum: Astrophys. Jour., v. 145, p. 943-945. Stone, P., 1975, Dynamics of the atmosphere, in Hansen, J. E., ed.. The atmosphere of Venus, Proceedings of a Conference at God dard Institute for Space Studies, New York, Oct. 15-17, 1974: NASA SP-382. Strong, J., 1960, Water in Venus' atmosphere: Sky and Tel., v. 19, p. 343. Suess, H ., 1962, ..Thermodynamic data on the formation of solid carbon and organic compounds in primitive planetary atmospheres: Jour, Geophys. Res., v. 67, p. 2029-2034. ______, and Urey, H. C ., 1956, Abundances of the elements: Rev. Mod. Physics, y . 28, p. 53-74. Surkov, I. A., Andreichikov, B. M., and Kalinkina, C. M.y 1973, Composition and structure of the cloud of Venus, in Space research XIV: Proceedings of the. sixteenth plenary meeting, Konstanz, West Germany, May 23-June 5, 1973: Berlin> Akademie-Verlag GMBH, p. 67.3-678. Surkov, I. A ., Kirnozov, F. F ., and Vinogradov, A. P ., 1973, Uranium, thorium and potassium content of Venus rock measured by Venera 8, in Space research XIV: Proceedings of the sixteenth plenary meeting, Konstanz, West Germany, May 23-June 5, 1973: Berlin, Akademie-Verlag GMBH, p. 679-687. ■ Tolbert, C. W ., and Straiton, A. W ., 1964, An investigation of 35 Gc. and 94 Gc. Cytherian radiation: Nature, v. 204, p. 1242-1245. Urey, H. C., 1952, The planets: their origin and development: New Haven,.Yale University Press,. . Vinogradov, A. P., Surkov, U. A., and Florensky, C. P., 1969, The chemical composition of the Venus atmosphere based on data . . . of the interplanetary station Venera 4, in Jastrow, R ., and Rasool, S. I., eds., The Venus atmosphere: New York, Gordon and Breach, p. 73-78. Wallace, L. , Caldwell, J. J., and Savage, B. D. , 1972, Ultraviolet photometry from the Orbiting Astronomical Observatory. III. Observations of Venus, Mars, Jupiter, and Saturn longward of 2000 A: Astrophys. Jour., v. 172, p. 755-769 . Washburn, E. W ., ed., 1926-1930, International critical tables: New York, McGraw-Hill, 7 vol. W east, R. C . , ed., 1970, Handbook of chemistry and physics, ed. 51: Cleveland, Chemical Rubber Co. Yoder, H. S., 1952, The Mg0-Al203~Si02-H20 system and the related metamorphic facies: Am. Jour. Sci. , Bowen vol., p. 569-627 . ______, and Eugster, H. P., 1954, Phlogopite synthesis and stability range: Geochim. et Cosmochim. Acta, v. 6, p. 157-185. ______, 1955, Synthetic and natural muscovites: Geochim. et Cosmo chim. Acta, v. 8, p. 255-279. Young, A. T., 1973, Are the clouds of Venus sulfuric acid?: Icarus, v . 18, p. 564-582 . ______1974, Venus clouds: structure and composition: Science, v. 183, p. 407-409.