Four entrance channels to the pentachlorodisilane (Si2HCl5) potential energy well, namely, SiCl2 + SiHCl3, SiCl4 + SiHCl, Cl3SiSiCl + HCl, and SiCl3 + SiHCl2, were analyzed in detail through transition state theory, Rice-Ramsperger-Kassel-Marcus (RRKM) theory, and solution of the multichannel master equation. The stationary points in the potential energy surface were optimized, and their vibrational frequencies and rotational constants calculated at the (U)B3LYP/6-31+G(d,p) level of theory; the (U)CCSD(T)/aug-cc-pVTZ level was then used for accurate estimation of activation energies. The pressure and temperature dependence of the rate coefficients of the channels related to Si2HCl5 stabilization/dissociation was determined along a wide range of conditions, for the first time. All channels showed strong pressure dependence in the four cases, at least at low-to-moderate pressure conditions. Each entrance channel leads to the formation of different products under different conditions, and the mechanism was analyzed in detail. The results indicated that at atmospheric pressure the reactions are in the falloff region, and therefore do not support the adoption of high-pressure limit rate coefficients in reaction models designed for simulation of systems at atmospheric or subatmospheric pressure conditions.

The dynamics of reactions of SiCl2 at Si(100) surfaces was investigated through the molecular orbital method at the B3LYP/6-31G(d,p) level of theory, with the surface being modeled using clusters of silicon atoms. The intradimer adsorption of a SiCl2 molecule proceeded with no energy barrier, and in the structure of the product of the adsorption reaction the Si atom of the SiCl2 adsorbate formed a triangular structure with the two Si atoms of the surface dimer, in agreement with theoretical predictions published recently in the literature for a small cluster. However, the dynamics reported in this work indicates that SiCl2 undergoes molecular adsorption at the silicon surface, in contrast with the dissociative adsorption suggested by some available kineticmodels. Intradimer adsorption of a second SiCl2 molecule, and interdimer adsorptions of a first, a second, and a third SiCl2 molecule were also seen to proceed without significant energy barriers, suggesting that the formation of the first additional layer of silicon atoms on the surface would be fast if the adsorption of SiCl2 were the only type of reaction proceeding in the system. The diffusion of the SiCl2 adsorbate over the surface and its desorption from the surface were found to have comparable activation energies, so that these reactions are expected to compete at high temperatures. (C) 2016 Elsevier B.V. All rights reserved.

As English-medium instruction (EMI) spreads around the world, university teachers and students who are non-native speakers of English (NNS) need to put much effort into the delivery or reception of content. Construction of scientific meaning in the process of learning is already complex when instruction is delivered in the first language of the teachers and students, and may become even more challenging in a second language, because science education depends greatly on language. In order to identify important pedagogical functions that teachers use to deliver content and to present different ways to realise each function, a corpus of lectures related to science and engineering courses was created and analysed. NNS teachers and students in science and engineering involved in EMI higher education can obtain insights for delivering and listening to lectures from the Online Corpus of Academic Lectures (OnCAL).

A process for obtaining trichlorosilane (TCS or SiHCl3) from hydrogenation of silicon tetrachloride (STC or SiCl4) through control of the temperature along a one-dimensional plug-flow reactor at atmospheric pressure was simulated using a reaction model that was built based on a literature review: Rate coefficients for reactions that occur in silicon CVD processes were adopted from published data. A compact model made available recently in the literature for simulating reactions occurring in silicon chemical vapor deposition (CVD) reactors was used as a basis, and more reactions from other sources were added for better description of the thermal decomposition of STC. The kinetic model built in this work comprised 63 elementary reactions occurring among 26 species. This model was applied to the calculation of the species concentration profiles along the reactor, which was divided into two parts: a STC hydrogenation step, proceeding at a high and constant temperature in the first, and a cooling step, where the gases are brought to room temperature, in the second part. When the hydrogenation temperature was set to 1273 K or higher, the first step was just a stage where the gases reached chemical equilibrium. More interesting reactions occur in the second step, where cooling prevents reactions from reaching equilibrium. This leads to the production of TCS from SiCl2 and HCl, with efficiencies that depend on the conditions. Among the conditions investigated in this work, hydrogenation of a mixture of molar composition STC:H-2 = 1:4 at 1473 K followed by cooling at a rate of -50.92 K cm(-1) gave the highest conversion efficiency from STC into TCS. (C) 2015 Wiley Periodicals, Inc.

In order to determine the effect of cluster size on the calculation of activation energies of reactions at silicon surfaces, intradimer adsorption/desorption reactions of H-2 and HCl with silicon clusters of various sizes and shapes were analyzed under the B3LYP/6-31G(d,p) level. Activation energies of reactions of H-2 and HCl with clusters composed of different number and orientation of dimers on the surface were estimated. For clusters having dimers parallel to each other, the activation energies converged satisfactorily at three (timers, but for clusters having dimers aligned in a row, activation energies did not converge sufficiently even at four dimers. The activation energies for clusters having three dimers parallel to each other on the surface were seen to be in good agreement with values published in the literature, but considering potential application of the cluster to model intrarow and interrow reaction mechanisms, reactions of H-2 and HCl with a cluster having three dimers parallel to each other and three (timers aligned in a row on the surface were also investigated, and the resultant activation energies were in even better agreement with published values. (C) 2015 Elsevier B.V. All rights reserved.

The OnCAL (Online Corpus of Academic Lectures) project was initiated in 2010 to support instructors who are non-native speakers of English (NNS) in preparing for delivering lectures in English. An interactive search interface was developed for lecture transcripts from MIT Open CourseWare (MIT OCW) and Stanford Engineering Everywhere (SEE). As of January 2015, 430 transcripts can be searched at the OnCAL site; the corpus comprises 3.5 million words, which corresponds to about 395 hours of recorded lecture time. The lectures cover many fields related to science and engineering: biology, chemistry, physics, mathematics, and computer science as undergraduate courses, and advanced mathematics, robotics, and artificial intelligence as graduate courses. Pedagogical functions have been identified, and in the present work, we show how teachers use language to connect the presentation of new content to contents presented previously. Here we present examples of pedagogical link-making devices, which are important for promoting continuity of ideas and helping students connect chunks of information presented over a series of lessons. Concrete examples can be observed via OnCAL to help NNS teachers deliver effective lectures and students listen to them.

In order to support instructors who are non-native speakers of English (NNS) in preparing for delivering lectures in an English-medium science and engineering program, we started the OnCAL (Online Corpus of Academic Lectures) project in 2010. Transcriptions of lectures delivered mostly by native speakers of English (NS) were obtained from MIT Open Courseware (MIT OCW) and Stanford Engineering Everywhere (SEE), and an online corpus was built. As of September 2015, the corpus comprises 430 lecture transcriptions, which contain 3.5 million words. This corresponds to about 395 hours of recorded lecture time. The lectures are from courses that cover a wide range of fields in science and engineering: biology, chemistry, physics, mathematics, and computer science as undergraduate courses, and advanced mathematics, robotics, and artificial intelligence as graduate courses. Pedagogical functions have been identified in the teacher utterances to allow OnCAL users to find examples of how teachers use language to realize specific pedagogical purposes. The examples of utterances found through OnCAL can thus be seen by NNS teachers as a reference for their own lectures. In the present work, we analyze how mathematical equations are explained in lectures. In this type of explanations teachers make use of and are aided by visual modes, but we believe that the spoken language that accompanies the other modes still have to be very accurate.

Oral communication skills are essential for engineers today and, as they are included in accreditation criteria of educational programmes, their teaching and evaluation deserve attention. However, concrete aspects as to what should be taught and evaluated in relation to oral communication skills have not been sufficiently established. In this paper, a method to aid the efficient teaching of oral presentation skills is proposed, from the presentation structure level to word and sentence level choices, through the use of JECPRESE, The Japanese-English Corpus of Presentations in Science and Engineering. As of June 2012, the corpus is composed of transcriptions of 74 presentations delivered in Japanese by students graduating from the Master's programme of various engineering departments and 31 presentations delivered in English, 16 by experienced researchers at an international conference on chemistry, and 15 by undergraduate engineering students of a mid-sized American university. The utterances were classified according to the specific moves (sections of the speech that express specific speaker intent) appearing in the presentations and frequently used words/expressions to express these moves were identified. © 2012 Copyright Taylor and Francis Group, LLC.

A modification of the CHEMKIN II package has been proposed for modeling addition of an arbitrary species at an arbitrary temperature to an arbitrary distance from the burner along a flat flame. The modified program was applied to the problem of addition of acetylene or benzene to different positions of a 40-Torr, phi = 2.4 benzene/O-2/40%-N-2 premixed flame to reach final equivalence ratios of phi = 2.5 and 2.681. The results obtained showed that acetylene addition to early positions of the flame led to significant increase in pyrene production rates, but pyrene concentrations were lower in the flames with acetylene addition in both the phi = 2.5 and 2.681 cases. Addition of benzene to the flame did not alter pyrene production rates in either the phi = 2.5 or 2.681 cases; however, for phi = 2.5, pyrene concentrations increased with benzene addition, while for phi = 2.681, pyrene contents decreased in comparison to the correspondent flames with no addition. Acetylene addition led to a significant increase in pyrene production rates, but the pyrene levels dropped due to increase in the flow velocity. Pyrene production rates were not sensitive to benzene addition, but pyrene contents increased with benzene addition when the flow velocity decreased. These results show that PAH concentration changes accompanying species addition to flames should be interpreted carefully, because an increase or decrease in the content of a PAH species does not necessarily reflect an effect on its formation rate or mechanism. (c) 2006 The Combustion Institute. published by Elsevier Inc. All rights reserved.

The oxidation reactions of phenyl and 1-naphthyl radicals through reaction with O-2 were investigated through 7 a quantum chemical molecular orbital method as potential sources of PAH species containing 5-membered rings in hydrocarbon flames. The reaction between phenyl radical and 0, was seen to produce phenoxy radical in a single elementary step. The same mechanism was followed in the reaction between 1-naphthyl radical and O-2. with formation of naphthoxy radical in a single step. In both phenyl and 1-naphthyl oxidation reactions, the variation of energy and atomic arrangement with the reaction coordinate was found to be similar and their rate coefficients, evaluated through transition state theory, to agree with each other. Thus, because the oxidation reaction of a species composed of one aromatic ring (phenyl radical) and that of a species composed of two aromatic rings ( 1-naphthyl radical) proceed through the same mechanism and have similar rate coefficients, the oxidation of species composed of an arbitrary number of aromatic rings can lie supposed to proceed through that mechanism and have the same rate coefficient. The mechanism of the subsequent decomposition of phenoxy radical into cyclopentadienyl radical and CO agreed with the mechanism reported in the literature, and the decomposition of naphthoxy radical was found to be similar to that of phenoxy radical. In summary, the routes
C6H4 + O-2 --> C6H5O + CO + O
C10H7 + O-2 --> C10H7O + O --> C9H7 + CO + O,
where C5H5 and C9H7 are, respectively, cyclopentadienyl radical and indenyl radical, proceed through the same mechanism. From these results it is suggested that the formation of species containing a peripheral 5-membered ring (aryl-5) through partial oxidation of a peripheral 6-membered ring of arbitrary aryl radicals (aryl-6), described generally as
aryl-6 + O-2 --> aryl-6-O + O --> aryl-5 + CO + O,
can be thought to proceed through the same mechanism. The rate coefficients of the first step are the same for any aryl-6 species, but they were found to he much smaller than values reported in the literature for phenyl as aryl-6 species. (C) 2002 by The Combustion Institute.

The sequence of hydrogen abstraction reactions from ethane to acetylene, which proceeds in fuel-rich methane flames, has been investigated by an ab initio molecular orbital method. First, the hydrogen abstraction mechanism was elucidated through an analysis of the variation in the electron spin density distributions through the reacting species. The H atoms holding the largest spin densities within a radical where seen to be those being abstracted from the species. The reaction rate coefficients for the four elementary abstraction steps from ethane to acetylene were determined through the transition state theory, and the results were compared with data published in the literature. The obtained rate coefficients were found to be lower than the published data for all four reactions examined. Especially, for two of the reactions, some published data seem to be too large, considering that the rate coefficients obtained through transition state theory tend to be larger than the true values.

The dynamics of the oxidation reaction of an acetylene molecule with an oxygen atom was investigated through an ab initio molecular orbital method. The three reactions (1) C2H2+O→HCCO+H, (2) HCCO+H→CH2+CO and (3) C2H2+O→CH2+CO, reported in reaction schemes, were confirmed as being the main elementary steps related to the acetylene oxidation but were found to be not single elementary steps, with each one forming intermediary species before the proposed products. These intermediaries are expected to decompose always and much faster than they form, so that those three reactions can be treated as elementary steps without problem. Under ideal conditions, only reaction (3) can be expected to proceed, but vibrations of the atoms in a structure formed during the process can cause its decomposition into HCCO+H, and later the so formed HCCO can meet another H atom and react to CH2+CO. Reactions (1) and (2) are then found to be, respectively, an early and a late step of reaction (3), but each of the three events can occur, depending on how is the collision between the O atom and the vibrating C2H2 molecule. The present results are compared with experimental data reported in the literature, and a qualitative agreement with the data of Miller and Bowman can be seen for reactions (1) and (3), and with the data of Warnatz for reaction (2).

The combustion mechanisms in premixed flames having a mixture of hydrogen and carbon monoxide as the fuel and the stoichiometric amount of air as the oxidizer were investigated by numerical simulation to elucidate the fuel-mixing effects on the burning velocity. Hydrogen and carbon monoxide have similar physical properties and their stoichiometric flames with air are known to propagate through the same type of mechanism, governed by active species diffusion. The flames of the mixed fuel hydrogen+carbon monoxide were found here to have their heat-release rate distributions changing with conformity in shape when the fuel composition is changed. The burning velocity was found to vary almost linearly with the hydrogen content in the fuel. Furthermore, the flux of chemical energy which is carried into low-temperature regions by hydrogen atoms diffusing from the flame front was also found to be roughly proportional to the burning velocity.

Fuel mixing effects on flame propagation were investigated with hydrogen and methane, whose flames, respectively, propagate through two different kinds of mechanisms. The calculated burning velocities did not show linearity with the hydrogen content in the fuel. The addition of only 5% of methane to hydrogen reduces the burning velocity by 20%, notoriously inhibiting the combustion of the mixture. This is caused by obstruction of the harmonious progression of two key chain-branching reactions, resulting in a decrease in hydrogen atom production at the flame front. However, the upstream flux, at the ignition point, of chemical energy due to diffusion of hydrogen atoms from the flame front and the integrated amount of heat released in the low-temperature region did show linearity with the burning velocity. Therefore, it was found that even for flames of mixed fuels whose components burn individually through different propagation mechanisms, the burning velocity can be expressed as a simple linear function of the upstream chemical energy flux or of the amount of heat released in the low-temperature region.

Premixed stoichiometric flames of acetylene, ethylene, and ethane with air were investigated using a chemical scheme composed of eighty-two pairs of elementary reactions to illustrate the dependence of the propagation mechanisms of C2 hydrocarbon flames on the heat release distribution. The general role of heat release distribution through flames was depicted by means of an analytical treatment of the governing equations with the thermal properties of the gases being constant and the temperature profile given by an appropriate function of the position in the flame. It was found that the amount of heat released by chemical reactions at low-temperature regions of a flame is of fundamental importance for its propagation and that the main trigger of exothermic reactions at those zones is hydrogen atoms, which diffuses from the flame front; in the acetylene flame reactions consuming hydrogen atoms at the low-temperature region were found to yield all the heat released at that zone. Heat release distribution extended down to low-temperature regions was found to be able to lead to large burning velocities even though the amount of heat released at each position was small.

In this work the combustion mechanism of silane-air flames is elucidated through a simulation employing a detailed chemical scheme. Two different pathways for the consumption of silane were found: SiH4 --> SiH3 --> SiH2O --> HSiO, which proceeds in low-temperature regions, and SiH4 --> SiH2 --> HSiO, occurring at high temperatures; this supports results previously reported in the literature. It was also found that some reactions occur over wide ranges of temperature and that the governing step for the whole silane combustion is the reaction of silane with hydroxyl radicals. A least square fitting to the burning velocities and adiabatic flame temperatures obtained for the four investigated mixtures, with equivalence ratios of 2.0, 1.43, 1.0, and 0.5, indicated that a mixture with equivalence ratio of about 1.2 would have the largest burning velocity and the highest flame temperature.

A detailed chemical reaction scheme and a full set of governing equations for fluid dynamics were used to elucidate the chemical and physical phenomena involved in a silane-air counterflow diffusion flame, where the fuel and air issue from the lower and upper nozzles, respectively. It was found that a stagnation surface, where the axial velocities fall down to zero, is kept in a position slightly shifted to the upper side, dividing the system into two zones: the upper air side and the lower fuel side; after leaving the nozzle silane begins to dissociate itself into SiH2 and H-2 when it reaches the region of rising temperature; oxygen penetrates into the fuel side by diffusion and reacts with H-2 forming OH radicals and H atoms. The produced SiH2 is further dehydrogenated to SiO via HSiO; comparing with premixed flames, the diffusion flames have large local equivalence ratios, so that the combustion reactions in the latter flames occur through a different mechanism.

A reaction scheme reproducing methane combustion by computer simulation under a wide range of stoichiometric ratio has been proposed and applied to flat methane-air premixed flames to elucidate factors determining their propagation mechanisms and the role of C2 reaction pathways. Main methane combustion reactions, which progress sequentially from the fuel to carbon dioxide, are inhibited at low temperatures due to retarded production of oxygen atoms which are necessary for the oxidation of methyl radicals. The role of C2 routes have been investigated by comparison of the results obtained by use of the C2 scheme and a simplified C1 scheme. Some of the methyl radicals are once recombined to ethane but most of the produced ethane is dissociated again to methyl via ethyl radicals. Then the C2 routes work as a temporary storage of species and hinder the main combustion reactions which go through C1 species.

A detailed model including a full scheme of combustion reactions and the governing equations of fluid mechanics was designed for two-dimensional hydrogen burning systems, and applied to a Burke-Schumann type hydrogen-air diffusion flame to elucidate its flame structure and combustion reaction mechanism. The same flame was also experimentally investigated and radial profiles of OH radical concentration and rotational temperature through the flame were determined by a conveniently improved line-of-sight absorption method. Simulation suggested that a well-developed diffusion flame occurs from heights about 5 mm from the burner mouth, in very good agreement with the experimental results. At each horizontal section of the well-developed diffusion region, the calculated rates of the chemical reactions showed considerable values within an annular zone about 5 mm thick, in which there is a point where H2 and O2 are simultaneously exhausted, and in whose vicinity the temperature becomes maximum. This result confirmed one of the Burke-Schumann's predictions, but radial displacements of about 1 mm between the peeks of the rates of different reactions and also between the maximum OH radical concentration and the maximum temperature were found in both experiments and calculation, showing that the reactions do not occur in an infinitely thin region, contradicting one of the Burke-Schumann's assumptions. At the lowest part of the flame, below the burner tip, the issuing air and the H2 diffusing downward meet and react like premixed gas mixtures
the whole flame was found to be held at the burner rim by large heat release rates at that region. The experimental results, in very good agreement with the simulation, showed that the maximum OH concentration through the flame occurs at a height around 2 mm, confirming the premixed-like intense combustion reactions taking place at the lower part of the flame. © 1991 Combustion Institute.

The combustion characteristics of diluted and preheated methane-air premixed and diffusion flames were analysed through flat-flame and counterfiow diffusion flame models. The results concerning to the premixed flame, or the flame in which the mixing rate of methane and air is very high, can be summarized as follows :1. In flames whose adiabatic flame temperature is higher than 2200 K, the fuel flux can be raised by diluting and preheating the metane-air mixture, independently of the air ratio. For the NO emission, in general, the greater the oxygen ratio the better, but there are optimum values for both the oxygen ratio and the preheating temperature.2. For flames whose adiabatic flame temperature is below 2000 K, both fuel flux and NO emission do not show strong dependence with neither dilution nor preheating. For these flames, preheating and dilution should be considered as the means to keep the condition of excess of enthalpy.For the diffusion flame, or the flame in which the mixing between methane and air is very slow, the following results were achieved :1. The NO emission level can be expressed as a function of the maximum flame temperature. In other words, neither the dilution nor the preheating of the air have any special effect in reducing the NO level.2. Therefore, the effect of dilution and preheating can not be achieved in flames in which the mixing between fuel and air is slow

平成9年度では,一連のH原子引き抜き反応とC_2H_2付加反応によってフェニルラジカルおよびナフタレンが生成することがわかったが,五員環の生成がみられなかった.平成10年度では,炭化水素火炎中の五員環生成を決定する主な因子を分子軌道プログラムを用いて求めた結果,以下のようにまとめられる.1. 火炎中で進行する六員環からの五員環生成反応として,フェニル(C_6H_5)ラジカル+O_2→シクロペンタジエニル(C_5H_5)ラジカル+CO+Oについての報告がある.その実験結果を,分子軌道計算によって再現できることをまず確認した結果,(1)O_2分子がフェニルラジカルに近づき,フェノキシ(C_6H_5O)ラジカル+Oとなる,(2)CC_6H_5Oラジカルが熱振動を起こす結果,五員環にCOが付加した化学種となる,(3)COが解離する,といった一連の素反応によって五員環化学種であるC_5H_5ラジカルが生成することがわかった.つまり,実験的に確認された反応を再現しただけではなく,その詳細な進行機構も明らかになった.2. 次に,六員環二つからなる1-ナフチルラジカルとO_2との反応を調べた結果,(1)フェニルラジカルの場合と同様の機構によって酸化反応が進行し,最後にCOが解離する結果,六員環と五員環一つずつからなるインデニルラジカルが生成する,(2)フェニルラジカルからのC_5H_5ラジカルの生成,1-ナフチルラジカルからのインデニルラジカルの生成,両反応過程におけるポテンシャルエネルギーも同様に変化することがわかった.3. 以上の結果から,六員環一つからなる化学種と二つからなる化学種が同様の機構によって酸化されて五員環に至ることから,任意の数の六員環からなる化学種の部分酸化によって五員環を含む化学種が生成すると予想される.これは,炭化水素火炎ではススが非常に多く生成する条件よりも,ススの生成が少ない条件の方がフラーレンの生成を促すという実験結果を裏付けるものであり,火炎中にある程度酸素が存在しなければ,ススとフラーレンの岐路を決定する五員環が生成しにくいことを示唆している

平成12年度においては,PAHに対するC_2付加反応による五員環および六員環の生成機構を調べた結果,以下のような結論を得た.1.C_2付加による五員環生成反応としては1-naphthylラジカルからのacenaphthalene生成および1-phenanthrylラジカルからのacephenanthrene生成,六員環生成反応としてはphenylラジカルからのnaphthalene生成,2-naphthylラジカルからのphenanthrene生成,4-phenanthrylラジカルからのpyrene生成およびbenzo[ghi]fluoranthylラジカルからのcorannulene生成を解析した.それらの反応機構はともにC_2が付加した後に閉環することからなり共通点が多いが,五員環が形成する場合は閉環反応の活性化エネルギーが大きく,六員環の方が形成しやすいことがわかった.2.また,corannuleneが生成する反応では六員環が形成するが,分子が平面構造ではなく立体構造となるため活性化エネルギーは五員環閉環反応のものよりも大きくなった.つまり,五員環が形成する反応と同様に,立体構造の分子を生成する反応の活性化エネルギーが大きく,高温でしか進行しないと考えられる.3.しかし,温度が高いほど付加したC_2が分解する傾向が強くなることも明らかになり,炭化水素火炎中でフラーレンが生成する条件では以前に解析したO_2との反応によるPAHの部分酸化は重要な役割を果たしていることを支持する結果が得られ,第38回燃焼シンポジウム講演論文集(2000,p.477)および28th International Symposium on Combustion,Abstracts of Work-In-Progress Posters(2000,p.314)に報告した

In order to offer language support to science and engineering researchers that are non-native speakers of those languages, research presentations delivered in Japanese by Japanese students, in English by American students, and in English by experienced researchers, were recorded and transcribed. After analysis of the presentation contents, a set of moves, which are a combination of a section and a step, were determined. Each utterance in the transcriptions was tagged properly according to the moves, and the tagged transcriptions were uploaded to JECPRESE(The Japanese-English Corpus of Presentations in Science and Engineering, http ://www. jecprese. sci. waseda. ac. jp/). Researchers from all over the world can now consult JECPRESE when preparing their presentations, through searches for words/expressions, or moves

シリコン結晶の成長反応機構を解明するために，気相中で生じる素反応および結晶表面で生じる反応を解析した．従来の研究に対象とされてこなかった，気相反応の圧力依存性を取り上げ，その速度定数を温度・圧力の関数として求めた．まず，SiCl2 + SiHCl3からの化学活性化反応を詳細に解析した．この反応が進行すると，Si2HCl5が中間体として生成し，圧力・温度の条件によっては複数の生成物に分解することがわかった．また，シリコン原子からなるクラスターを用いて結晶を近似して，以前から解析をしてきたHClやSiCl2の吸着反応の速度定数を求めた．

本研究では以前から開発し，公開した OnCAL (The Online Corpus of Academic Lectures, http://www.oncal.sci.waseda.ac.jp/) を活用して，MIT やStanford 大学の講義書き起こしを詳細に解析し，英語による講義における academic spoken English の言語特徴を解析した．また，英語による論文のコーパスを構築し，論文における academic written English の言語特徴を特定し，具体的に academic spoken English とどのように異なるかを解析した．OnCAL に約 350 万語，論文のコーパスには約 230 万語が含まれる．解析の結果，academic spoken English と academic written English との間に下記のような興味深い差異を特定した．１．教科書に頻度が高い academic words のリストとして知られる Academic Word List に含まれる単語を，OnCAL および論文のコーパス内の頻度は，大きく異なった．例えば，academic word である obtain は同義語の get との差を調べた結果，obtain は論文では 1,877 回出現するのに対して，OnCAL には 156 回しか出現しない．Get については，論文コーパスに 260 回，OnCAL には 15,919 回出現する．論文コーパスと講義コーパスの大きさは異なっていることを考慮しても，academic spoken English と academic written English との差ははっきり現れていると言える．２．単語（動詞，形容詞，名詞，副詞）レベルだけではなく，テキスト構造についても大きな差があることが明らかになった．論文は複数の section (Introduction, Methods, Results, Discussion, Conclusions) に分かれており，各 section は論理的に展開される．これに対して講義は，一番最初と最後の部分を除いて決まった section には分かれず，教員の教える意図は講義全体にわたる．３．論文作成の作業に研究者が時間をかけて表現を注意深く選ぶことがることができるのに対して講義の場合準備はできるが多くの不確定要素が多く入ってくる．英語非母語話者教員は講義を英語で行う場合，講義における academic spoken English の特徴を事前に把握できるように講義中の pedagogical functions を特定して，各 function 毎に典型的な表現を特定した．

本研究では，OnCAL (Online Corpus of Academic Lectures, http://www.oncal.sci.waseda.ac.jp) を開発している．2014 年度中には English Medium Instruction, Integration of Content and Language in Higher Education, English for Specific Purposes, Classroom language, Science Education など，幅広い分野における文献を調査し，その見知から構築したコーパスを分析した結果，11 種類の pedagogical functions を特定し，各 function によく使われる表現を特定した．Function 毎に高い頻度で出現した表現を OnCAL システムに登録し，ユーザが function を指定するだけでその function と関係する例文を簡単に参照することを可能とした．

目的　理工系の分野において，「論文」および「口頭発表」の形で研究者が情報発信を，英語によって行う必要がある．研究者はこの事実を強く意識しており，経験によってこれらのジャンルの言語特徴を把握し，適切に論文や口頭発表を行っている．近年，早稲田大学理工学術院においても，講義も英語によって行われることになった．ヨーロッパと異なり，日本においてはこの動きはまだ新しく，日本の大学教員は「英語による講義」の言語特徴を十分に把握できていない可能性は高い．そこで，本研究は「英語による講義」，「英語による論文」および「英語による口頭発表」の言語特徴にどのように異なるかを明らかにすることを目的とする．方法　「英語による論文」における言語特徴を特定するために Elsevier 社 やTaylor and Francis社，Institute of Physics, American Chemical Societyなどにある理工系の多分野にわたる多くのjournalに2013年に掲載された論文をダウンロードし，論文のコーパスを構築した．AntConcソフトウェアを用いて論文コーパスを解析した．　「英語による口頭発表」における言語特徴を特定するために以前に研究代表者が開発・公開したJECPRESE (The Japanese―English Corpus of Presentations in Science and Engineering)を用いた．　「英語による講義」の言語特徴を特定するために以前に研究代表者が開発・公開したOnCAL (The Online Corpus of Academic Lectures)を使用した．結果　論文，口頭発表および講義は，それぞれで取り上げられる内容は大きく異なるため，それぞれの言語特徴もそれによって異なるが，できるだけ似た内容を比較する必要がある．そこで，論文と口頭発表のIntroductionの部分は講義に出現する内容と最も近いと判断した．これは，Introductionではすでに知られていることを紹介する部分があるからである．その中ではknowという動詞に注目した．論文では，「It is well known that…」のような受け身形の動詞が含まれる表現は多く見られる．これに対して，knowという動詞を口頭発表で検索すると，「we all know that」に表現は変化し，「we」などの代名詞は主語になる表現は多く見られる．講義も話し言葉を用いるジャンルであるため，口頭発表と似た言語特徴を有する．しかし，講義では「don’t know」，「doesn’t know」および「didn’t know」の否定形はknowの全活用中に17%以上を占めていた．これは大変興味深い結果であり，ジャンル間に言語特徴が異なることを示している．結論研究者が研究成果を発信するためや講義にて教育効果を上げるためにそれぞれのジャンルにて言語を適切に用いることは重要であり，本研究で得られた結果がその事実を示している．

　金属表面を固める方法として真空浸炭が用いられており、炭素源となるのは acetylene (C2H2) や propane (C3H8)、ethylene (C2H4)、cyclohexane (C6H12) などの炭化水素分子である。真空浸炭は高効率であるため環境に優しい方法であるがすすを生成するという問題も生じる。本研究ではすすの前駆体である多環芳香族炭化水素（Polycyclic Aromatic Hydrocarbons、以降 PAH と呼ぶ）の挙動を解明するために実験および計算によって methane (CH4)、acetylene、propane および cyclohexane の熱分解過程を解析し、それぞれの条件で生成する PAH の濃度を求め、比較した。実験に小型の炉を用い、計算には文献に提唱されている化学反応モデルを組み合わせて使用した。得られた結果は以下のようである。　１．熱を加えても methane がほとんど分解されないまま高濃度で残る。Acetylene もかなり安定であるが benzene (C6H6) などの炭素原子が多い分子が生成する。　２．Propane および cyclohexane はほぼ完全に分解され、主な生成物は acetylene、hydrogen (H2)、benzene および methane である。　３．熱分解による主な生成物以外に加えて、diacetylene (C4H2), vinylacetylene (C4H4), butadiene (C4H6), toluene (C7H8), phenylacetylene (C8H6), styrene (C8H8), indene (C9H8), naphthalene (C10H8), acenaphthylene (C12H8), acenaphthene (C12H10) および phenalene (C13H10) の生成は実験的に確認され、実験的に求めたこれらの炭化水素濃度は計算の結果と定性的に一致した。　４．実験結果と計算結果との間に定性的な一致がみられたことによって採用した化学反応モデルは基本的には妥当であることがわかったが、diacetylene や styrene の濃度については差が大きかったことも明らかになった。すすの発生を抑える方法を探るためにPAH の挙動を解明する必要があるため、化学反応モデルを改良しなければならないことが判明し、今後の課題が明確になった。