Boiling is highly utilized in the industrial applications of large scale, such as nuclear power plant and steel industry due to its good heat transfer characteristics. In the past several decades, a large number of studies on boiling heat transfer have been carried out. However, because of complexity of boiling phenomenon, some mechanisms remain not clear. One of the recent major targets of the boiling heat transfer is the cooling of electronic devices such as CPUs. The recent rapid increase of the power consumption of electronic packages gives rise to the major issue of the cooling of that. The need for a miniaturization of cooling unit drives application of cooling by flow in small passages, which has large heat transfer coefficient even in single phase. Boiling heat transfer in small passages is expected to achieve higher heat flux than single phase heat transfer. Bubbles confined in a small passages, however, cause large pressure drop to flush them out and result in a dryout of the heated channel walls to damage the components.

In response to this, the author applied microbubble emission boiling to boiling in small passages. Microbubble emission boiling is a peculiar mode of boiling that is specific to highly subcooled boiling at high heat flux. In this mode, numerous microbubbles are injected from heat transfer surfaces due to violently rapid shrinkage of vapor bubbles with condensation. It leads not only to the fluctuation of bubble surface to generate microbubbles, but also to the good heat transfer characteristics by bringing cooler liquid above to the heat transfer surfaces. Experimentation of highly subcooled flow boiling in small channels was conducted to keep small pressure drop at the same level of single phase flow at up to high heat flux near 4.8 MW per square meter (Tange et al. (2004) and Tange et al. (2005)). The physics of flow boiling is more complex than that of pool boiling because boiling bubbles emerge in forced convective flow and phenomena in upstream affects those in downstream. Most of boiling experiments at high heat flux use copper blocks with large thermal capacities because of its toughness against of heating. In order to predict boiling in a small cooling packages, however, knowledge about boiling with small thermal capacities should be needed.

In order to survey the mechanisms and the occurrence condition of microbubble emission, the author conducted two series of experiment: (1) highly subcooled pool boiling on heated metal wires, and (2) highly subcooled pool boiling on artificial surfaces.

(1) Highly subcooled boiling on heated wires

A platinum wire with a diameter of 300 micron was submerged in a stagnant water pool in which liquid temperature was controlled and it was heated by direct electric current.Heat flux and average temperature of the wire was estimated from power consumption and resistance. Boiling curves of various subcooling conditions was constructed, however, there is no clear jump of temperature between ordinary nucleate boiling regime and microbubble emission boiling as reported in the system of heater blocks with large thermal capacities. As the increase of the subcooling, critical heat flux increases under low subcooling conditions and the tendency of the increase decreases over about 30 K.

From the photographic observation with high-speed video camera, it was found that primary bubbles on a wire did not depart from the wire and rapidly collapsed to generate several microbubbles with the deformation of the top of the primary bubbles under high subcooling condition.

Analysis of heat transfer was conducted to estimate the contribution of microbubble emission to the heat transfer qualitatively. Near saturated condition, boiling bubbles detached from the heat transfer surface and heat removed by the vapor bubble is not negligible. As the increase of the subcooling, maximum diameter of bubbles on the wire become small and heat transfer completes within a thin area around the wire.

Under more highly subcooled condition, however, it was found that heat transfer did not complete in the thin area. This indicates that emission of microbubbles and induced flows by collapses of the mother bubbles contribute the heat transfer.

(2) Highly subcooled boiling on artificial surfaces

Microbubbles came to the world with violently rapid deformation of a mother bubble on the heated surface in the heated wire experiment. According to the boiling experiment on a heated wire, maximum diameters of boiling bubbles depends on the position of nucleation sites even in the same heat flux and the same subcooling. Relatively large bubbles of more than 500 micron diameter deform largely, although smaller bubbles remain spherical during growth and shrinkage process. This indicates that the waiting time till the nucleation and the growth of thermal boundary layer varies according to the specification of the nucleation site. Steady heating experiment can not avoid the effect of preceding bubbles and adjacent bubbles on the motion of the bubble in question.

Subsequently, in order to examine the behavior of a single boiling bubble in an ideal situation imitating an actual boiling, careful experimentation was conducted. The objective of this experimentation is to generate a single vapor bubble in an ideal situation. Artificial surfaces were fabricated with MEMS technique to realize arbitrary thermal boundary layers and control the bubble nucleation. Nakabeppu and Wakasugi (2006) proposed new artificial surface fabricated with MEMS technique and performed an experiment on boiling bubbles generated at a single site with steady heating. This surface has miniature thermocouples and a hydrogen trigger for the nucleation. In this experimentation, the author employed the original Nakabeepu's surfaces and upgraded them by adding the thin film heater to make it possible to heat the surface transiently.

A new MEMS chip as artificial heat transfer surface is a silicon chip square in shape, 20 mm in height and width and 0.5 mm in thickness. It has seven thermocouples for temperature measurement and hydrogen trigger for control of bubble nucleation on a front side, and thin film heaters for transient heating on a back side. Each component on the chip was fabricated with sputtering and liftoff techniques.

A test section with the chip was submerged into stagnant pool in which test fluid, distilled and degassed water, is maintained at a desired temperature. The thin film heaters were heated by direct electric current to generate thermal boundary layer on the chip. Hydrogen trigger generated liquid-gas interface playing the role of the nucleus. It is the special feature of this chip that nucleation of a bubble is separated from heating of the surface. Temperature distribution and its time variation of the surface were measured by thermocouples.

At various heating time and heat flux, the bubble was generated and observed, and it was categorized into four patterns. Under the conditions of highly subcooled boiling at high heat flux, the growing bubbles largely deformed and generate tiny bubbles from the top part of it as same as that on heated wires.

Both on heated wires and on artificial surfaces, it was observed that microbubbles were produced from largely deformed mother bubble touched on the surfaces. And only relatively large bubble produced microbubbles. To employ microbubble emission boiling to boiling in small passages, height of the channel needs to be large enough to allow bubbles on the surface to grow and deform largely.

Although microbubbles emerged, boiling on heated elements with small thermal capacities did not reproduce the heat transfer characteristics of ordinary MEB, a stable regime in the temperature range of transitional boiling, as reported in the systems of heater blocks. This is because it is difficult for a small heated element to maintain the temperature in a range of transitional boiling once it partially dry out.

The main body of the thesis, which contains four chapters, treats the investigation on a single vapor bubble conducted against the background described above with experimental approaches. Chapter 1 introduces boiling phenomena in the context of applications and fundamentals and open problems on it in order to familiarize the reader with the terms in boiling research and to show the motivation of the investigation. Experimentation of subcooled boiling on heated wires is described in Chapter 2. Experimentation of subcooled boiling on artificial surfaces is described in Chapter 3. Conclusions and remarks are summarized in Chapter 4.

本論文は「Highly subcooled boiling on heated elements with small thermal capacities(微小発熱体上の高サブクール沸騰)」と題し, 全4章からなる. 論文は微小流路内強制流動沸騰を用いた電子機器の冷却を背景に,微小発熱体上における高サブクール沸騰の伝熱特性と気泡挙動を解明し,冷却デバイスの設計に指針を与えることを目的としている.

近年の電子機器や光学機器の発熱量増加を背景とし,比較的小さなシステムの冷却に沸騰を利用することが試みられている.高効率化や小型化への要請から,沸騰は微小流路内における強制流動沸騰の形で利用されるが,微小流路内では,沸騰気泡が充満することにより,圧力損失が増加するだけでなく,伝熱面上に乾き面ができることで冷却機器を焼損させる恐れのあることが指摘されている.そこで,気泡微細化沸騰という現象を微小流路内沸騰に適用することを提案し,実際の冷却デバイスを理想化した実験系で先行実験を行った.気泡微細化沸騰とは,高熱流束,高サブクール条件で発生する沸騰で,伝熱面上の気泡が激しく凝縮することによる気泡の微細化を特徴とし,気泡の凝縮が伝熱面上に直接サブクール液を供給し高い熱伝達が得られることも知られている.実験の結果,高い除熱性能を保ちつつ,気泡の微細化によって気泡の閉塞およびドライアウトを回避できる可能性を示した.しかし,流動沸騰における沸騰気泡は流れ場の影響を大きく受けるため,気泡の微細化がどのような温度場で起こるのか特定することは難しく,また,実際の小型冷却システムのように伝熱面の熱容量が小さい場合に沸騰現象が異なるかどうかを知る必要がある.

これを踏まえ,本論文では,微小発熱体における伝熱特性と沸騰気泡挙動との関係と,気泡が微細化される条件の解明を目的として,加熱細線上の高サブクールプール沸騰実験を行った.続いて,そこで得られた結果を元にして,より理想的な状況での単一気泡の挙動解明を目的として,MEMS技術を用いた人工伝熱面で微細気泡が発生する条件を実験的に探索した.

第1章は「Introduction(序論)」であり,研究の背景として沸騰の産業的応用,沸騰熱伝達の特徴,及び沸騰気泡の描像について概観し,気泡微細化沸騰に関する関連研究をまとめている.また,気泡微細化沸騰の有用性を示すために行った微小流路内高サブクール沸騰の実験結果を,研究を行う動機として示してから,本研究の目的を述べている.

第2章は「Subcooled boiling on heated wires(加熱細線上のサブクール沸騰)」であり,サブクール液中における加熱細線上の沸騰実験の結果と考察を示している.サブクール度の増加とともに限界熱流束は上昇するが,高サブクール度領域においてその上昇傾向が鈍ることを明らかにした.先行研究によって報告されている銅ブロック系のように,沸騰曲線に核沸騰と気泡微細化沸騰の境界を示す兆候は確認されなかったため,その原因について伝熱面の熱容量と限界熱流束機構の点から考察をしている.また,高速カメラを用いた観察により,微細気泡の射出メカニズムが,一次気泡が球形を保ちつつ成長し気泡の上部が大きく変形することによるものであることを捉えている.微細気泡の生成,射出が伝熱に与える影響を定量的に評価するため.マクロな実験データから細線周りの熱伝達について解析を行い,微細気泡が確認される領域において微細気泡の担う熱伝達が大きいことを明らかにしている.

第3章は「Isolated bubbles during subcooled boiling(サブクール沸騰における孤立気泡)」であり,沸騰を模擬した理想的な環境における単一気泡の生成を人工伝熱面によって実現し,気泡の成長崩壊を観察して微細気泡の射出が起こる条件を探索している.まず,薄膜ヒータと水素トリガーにより過熱液層と気泡核の生成を独立に制御できるよう設計された人工伝熱面について説明をし,実験手順を示している.次に.気泡生成直後における壁面温度分布の測定結果を示し,気泡底部での三相界線の移動を捉えたことを述べている.また,様々な条件での実験結果を発泡形態のマップとしてまとめ,高サブクール条件における発泡の様子を4種類に分類して過熱液層の厚さとの関係を考察している.特に,高熱流束で加熱時間が長い場合には,気泡がすばやく大きく成長しサブクール液中に蒸気塊を取り残すことで,細線上沸騰で確認されたのと同様の微細気泡が発生することを捉え,気泡挙動を示している.

第4章は「Conclusion(結論)」であり,本研究で得られた結果を総括し,微小流路内高サブクール沸騰の利用に対し設計指針を与えている.

先に述べたような背景から,高サブクール沸騰において気泡の微細化がどのような条件で起こりうるのかを示し,その気泡の微細化がどのようなメカニズムで起こるかを明らかにしたことの意義は大きい.特に,孤立気泡の発生制御を可能にする人工伝熱面を設計し,サブクール度・熱流束・発泡までの加熱時間,という基本的なパラメタによって微細化条件を実験的に明らかにしたという点で非常に優れた論文である.

よって本論文は博士(工学)の学位請求論文として合格と認められる。