Hypervelocity impact is one of the most common and important in physical and chemical processes affecting the evolution of terrestrial planets. When impactors such as asteroids or comets impact on the surface of planets, volatile components from surface and/or impactor are released due to high pressure and temperature states caused by passage of intense shock waves. This has been called the "impact-induced degassing". Degassed components (e.g., CO2 form carbonate, S-bearing gases from sulfate, and so on) may have played an important role in evolution of the surface environment of terrestrial planets (e.g., mass extinction, evolution of planetary atmosphere, and so on). The goal of this thesis is to understand the degassing mechanism at high shock pressure regions where the previous studies were not able to discuss intensively.

There have been many studies on impact-induced degassing. The degassing mechanism has been theoretically discussed and experimentally assessed. However, the shock pressures treated in these studies so far were relatively low (<~70 GPa). This is because breaking-up of sample container makes the experiments at higher-shock pressure regions difficult. In addition, the theoretical studies were not able to treat high-shock pressure regions because the assumption of conventional model may not be applicable to those regions. Thus, there has been discussed so far almost only the shock pressure required for "incipient decomposition" (i.e., beginning of degassing). Thus, impact-induced degassing at high-shock pressure region is not understood well.

The degassing mechanism may change from the conventional view on impact-induced degassing at adequately high-shock pressure. In the conventional model, it has been assumed that the minerals do not decompose at the moment when the shock wave passes but decompose after the pressure release. However, if the shock pressure is high enough, minerals might decompose even on Hugoniot (i.e., decompose at the compressed state before pressure release). This type of decomposition might lead intensive degassing because it does not require appropriate pressure release that the conventional mechanism requires. Thus, it is essential for the impact-induced degassing to study this new mechanism.

The change in degassing mechanism might occur when the thermodynamic state on Hugoniot becomes to satisfy the decomposition condition of minerals. That is, the shock pressure and temperature exceed the decomposition boundary of minerals on their P-T phase diagrams. Thus, we need to know the information on decomposition boundary of minerals at high pressure and temperature to determine the critical shock pressure where the decomposition occurs on Hugoniot.

However, in general, this is difficult because the temperature region of the phase diagrams of minerals reported so far was limited under ~2500 K. This is much less than the temperature achieved in planetary impacts, which is more than several thousand degrees kelvin. The phase diagrams were mainly used for the studies in the Earth's interior. The temperature at core-mantle boundary is estimated to be ~3000 K. Then, the phase diagrams over 3000 K are not necessary as long as we discuss the Earth's mantle. Therefore, we need to determine the decomposition boundaries at high-temperature (and pressure) region to study the impact-induced degassing mechanism.

In this study, therefore, first, we developed a new method to estimate the temperature and determined the decomposition boundaries of CaCO3 by using laser-heated diamond-anvil cell (LHDAC). Then, we discussed the mechanism of impact-induced degassing of CO2 from CaCO3. CaCO3 is one of the most important minerals for impact-induced degassing and evolution of the terrestrial environment. It has been known that CaCO3 decomposes at high temperature region:

CaCO3→CaO+CO2.

The decomposition boundary was experimentally determined for the regions on its P-T phase diagram lower than ~100 bar and ~1700 K. The decomposition boundaries theoretically estimated suggest that the decomposition temperature never exceed 4000 K even at 80 GPa. However, such estimates are based on unreasonable assumptions.

This thesis consists of three parts. In Part I, first we briefly explain the temperature measurement system and point out the problems of the method of temperature measurement in LHDAC experiments. Then, we propose more reliable temperature determination than before from the data with errors due to chromatic aberration. In Part II, we show the results of determination of the decomposition boundary of CaCO3 at high-pressure and high-temperature region. Finally, in Part III, we discuss the impact-induced degassing of CO2 from CaCO3 by using the decomposition boundary newly determined in this study.

The main results are summarized as follows In Part I, we proposed a new method of temperature measurement in LHDAC experiments in which the chromatic aberration is effective. One of the most important issues in LHDAC experiments is to estimate a reliable temperature. However, it has been known that use of refractive optics results in the problem of chromatic aberration if we measure the temperature by spectroradiometry, which is a typical temperature measurement method in LHDAC system. Therefore, we presented a new method to determine more reasonable temperature from the data obtained in the experimental system using refractive optics. We determined temperatures by the combination of spectroradiometric method and the method that uses, spectral intensity.

The advantage of this method is to use the spectral intensity at the focusing wavelength (630 nm for our experimental system) where the spectral intensity is a close to real one even though the chromatic aberration is effective. We found that the temperatures measured by spectroradiometry in high-radiance region are reliable because the chromatic aberration is not effective in this region. Thus, the high-radiance-region temperatures are fitted to the theoretical intensity-temperature relation for the 630-nm-wavelength light. Then, that relation is extrapolated to the low-radiance region. The temperatures in the low-radiance region were estimated by substituting the intensity at 630 nm to the extrapolated intensity-temperature relation.

In Part II, we developed the experimental method to determine the decomposition boundary of CaCO3 at high temperature and high pressure region. CaCO3 samples were pressurized and heated with LHDAC and were quenched. Temperatures were estimated by the method proposed in Part I. Then, the recovered quenched samples were analyzed with Raman spectroscopy and energy-dispersive X-ray spectroscopy (EDS) to identify CO2 and/or CaO that are the products of decomposition of CaCO3. We determined the decomposition boundary of CaCO3 up to ~10 GPa and ~5000 K. In the previous studies the decomposition boundary was experimentally determined for the region lower than ~100 bar and ~1700 K. The theoretical estimations of the decomposition boundary suggested that it never exceed 4000 K even at ~80 GPa. However, our experimental data suggest that the decomposition boundary locates lower-pressure and higher-temperature regions and that the decomposition does not occur even at ~5000 K and ~10 GPa. This means that CaCO3 is more stable against decomposition than previously thought.

In Part III, we discussed (i) the impact-induced degassing of CO2 form CaCO3 target at high-shock pressure region and (ii) impact induced CaCO3 melt generation using the results described in Part II.

We estimated the "Hugoniot decomposition pressure", which we define as the pressure on the intersecting point of Hugoniot with the decomposition boundary. Intensive degassing may occur above this pressure. The results indicate that the Hugoniot decomposition pressure is over ~115 GPa, which is much higher than the assumed pressure for complete decomposition in the previous studies (20-30 GPa). This indicates that much higher shock pressure is required for intensive degassing to occur than previously thought. Thus, the amount of degassed CO2 may be much smaller than that previously estimated. In the case of Chicxulub impact, the amount of degassed CO2 may be almost one order smaller than that previously estimated. Thus, the global warming caused by impact-induced degassing of CO2 may have little effect on the mass extinction at the Cretaceous-Paleogene (K-P) boundary.

The results of LHDAC experiments also show that the liquid field of the phase diagram of CaCO3 extends up to at least ~5000 K. This suggests that a large amount of CaCO3 melt is potentially produced when impacts on CaCO3-rich target occurs. This is consistent with the results of recent geological analysis of crater deposits.

The decomposition boundary of CaCO3 at high-temperature and pressure regions was experimentally determined for the first time. This makes it possible to discuss the impact-induced degassing of CO2 from CaCO3 at high-shock pressure. The experimental method developed in this study may be applicable to other important minerals for impact-induced degassing, such as sulfate, which is thought to play an important role in the mass extinction at Cretaceous-Paleogene boundary.

本論文はイントロダクション、3つの章、要約からなる。イントロダクションでは、衝突による炭酸塩岩の分解が二酸化炭素を放出することで地球表層環境変動に重要な役割を果たしうること、従来の研究においては分解の起こる条件について十分な検討がなされていないことが指摘され、本研究の目的が述べられている。

第1章は、本研究に用いたレーザー加熱ダイアモンドアンビル装置の概要と、その使用における最大の問題点である温度測定に関して述べられている。本論文ではその原因が、波長による焦点の違いによるが色収差であることが指摘されている。本研究においてはその問題解決のために、レーザービームの最強の波長は黒体輻射に従い、より長波長(低温)がそこからのずれを示すことをみいだした。高温の理想的な黒体輻射からのずれを補正することで、正確な収差のない正確な温度決定をする方法を提唱した。その結果、従来大きな誤差を含んでいた温度測定が高精度で可能となったことが示されている。この補正法の発見は、地球惑星科学の最先端研究分野の一つである、地球内部の物性を対象とする超高圧実験に多大な影響を与えるものである。

第2章は、実際に決定された炭酸カルシウムの相図について述べられている。レーザー加熱ダイアモンドアンビル装置を用いた高圧・高温実験を、炭酸カルシウムについておこない、その分解条件を決定した。炭酸カルシウムが分解した証拠として、ラマン分光による二酸化炭が検出されたことが示されている。ただし、CaOは観察されず、周囲の融解したCaCO3中にとけ込んでいる可能性が論じられている。また、本実験が熱平衡に達しているかどうかを、逆反応が成立することで確認した。いったん分解したと推定される試料をふたたび低温に置くと、融解していると考えられる丸い領域が消滅し、回収試料はもととは異なる組織を示すことにより、逆反応の成立が示されている。温度に関し2000K - 5000K、圧力に関し 1 GPa - 10 GPaの範囲で分解が起こるか否かの検討をおこない、それらの結果、分解曲線は2300K, 1GPaあたりから5000K, 10GPa あたりをとおる曲線であることを示した。すなわち、高温領域に向け、液相領域が大きく広がっているという、従来のモデル計算による狭い液相領域とはまったく異なる結果を与えている。この結果は、より定圧において実験的に決定された分解曲線の延長線上に位置し、その妥当性が示されている。これは世界で初めて、広い温度圧力条件において炭酸カルシウムの正確な相図を与えたと言う点において、きわめて高い価値をもつ。

第3章では、得られた結果を用い、衝突脱ガスのおこる条件を推定し、きわめて高温にならないとユゴニオ曲線にぶつからないため、衝突脱ガスがおこりにくいことが論じられている。従来炭酸カルシウムの衝突脱ガスは、高ユゴニオ曲線が分解曲線に交差することで進行すると考えられてきたが、本研究の結果は、広い条件において液相領域を通過することが示された。すなわち、発生する二酸化炭素量は圧倒的に少なくなることになる。その結果を白亜紀-暁新世境界でおきた巨大衝突に適用すると、生成される二酸化炭素量が従来の推定よりはるかに少なく、環境に与える効果が小さいことが指摘されている。また、最近その地域の地質試料中から発見され始めた溶融炭酸カルシウムの存在は、この仮説の正しさを支持している。天体の大規模衝突により発生する二酸化炭素は地球と生命の進化に重要な影響を与えると考えられてきたが、本研究によりその効果が従来考えられてきたよりはるかに小さい可能性が指示されたことになり、地球史の立場からみてもその学術的価値はきわめて高い。

なお、本論文第1章は八木健彦・岡田卓・松井孝典との、第2章と3章は八木健彦・松井孝典との共同研究であるが、すべて論文提出者が主体となって実験、検証、考察をおこなったもので、論文提出者の寄与が十分であると判断する。

したがって、博士(理学)の学位を授与できると認める。