Recently, owing to pollution of the Earth's atmosphere, the chemistry of commonly used chemicals has come under scrutiny. Such common chemicals include chlorofluorocarbons (CFCs), bromohydrocarbons and iodohydrocarbons, all of which have been widely used. Their impact on the chemistry of the atmosphere has most obviously been seen in the destruction of the ozone layer. In addition, the results of the large amount of combustion that occurs over the Earth's surface has been felt by the production of greenhouse gases, the production of acid rain, and the emission of harmful chemicals into the air we breathe. As we have realised the impact of our way of life on the environment, we have sought to understand the destructive chemistry of the chemicals we commonly use and design new ones, which we hope will be less harmful. In order to understand the chemistry of such complicated systems as the Earth's atmosphere and flames, many tools are needed: means of detecting what reactive chemical species are present, means of modelling the chemistry in laboratory environments, and data for input into computer modelling schemes. In the present proposal we continue the drive towards identifying spectroscopic signatures for reactive species, study the chemistry of reactive species in the laboratory, and by both of these means, produce data for input into computer modelling schemes.

Spectroscopic signatures are an important tool for the chemist in identifying what reactive species are present in a complicated mixture; this is extremely important, as the reactivity of the species excludes the ability to extract them and take them away for analysis. In addition, in situ monitoring of these species can give a real-time picture of the chemistry. Remote sensing is a powerful tool in the armoury of the atmospheric chemist, since it allows the monitoring of species high above the Earth's surface. Spectroscopic monitoring of reactive species in combustion systems, such as flames and car engines, is a very important way of identifying potentially harmful chemicals. In addition, understanding the chemistry of such systems ought to allow "cleaner" chemistry to be devised, by a judicious choice of the fuel and / or additives.

Laboratory modelling of atmospheric and combustion chemistry has long been a means of understanding the chemistry involved. In part, this requires the ability to produce the reactive species in isolation, and in sufficient quantities to be useful for further study. A part of the work in this proposal is to develop means of producing reactive species in the laboratory, and then to study the chemistry of these species under controlled conditions.

The results from these experimental studies, coupled with high-level quantum chemical studies, will provide a wealth of information which will be useful to our understanding, and be valuable input into sophisticated computer modelling programmes, increasing their reliability. Such information will include: reaction routes, thermodynamic quantities, reaction rate constants, and molecular parameters.

The social and economic benefit of the ability to understand and control atmospheric and combustion chemistry is high. In addition, this project would supply a number of well-trained and experienced postdoctoral researchers who would be able to work either in industry or academe, to address these issues.

Europe needs to develop an infrastructure of trained scientists capable of solving problems in atmospheric science and combustion in order to remain competitive with the U.S.A., and to address problems associated with atmospheric pollution and the efficient use of fossil fuels. The interdisciplinary skills needed to address fundamental issues in these areas include knowledge of laser-based spectroscopic methods, gas-phase kinetics, ion-molecule chemistry, high vacuum technology, modelling of gas-phase reactions and electronic structure calculations. The training of young researchers provided by this Network will produce scientists with a knowledge of all these areas, with a high degree of competence in several of them. These scientists will also have the confidence and flexibility to apply these skills to other areas such as plasma chemistry and materials science which are relevant to the electronics industry.

The overall objectives of the present research proposal are the development of methods of detection of reactive species in situ; their production in a laboratory environment, to allow further study of their chemistry, and the study of their reactions in the laboratory, in order to allow inferences to be made regarding their reactivity in more complicated systems. In order to allow more focused discussion of the proposed scientific work, the overall objectives have been divided into five key areas.

(i) Characterisation of Electronically Excited States of Reactive Intermediates

Small neutral radicals play central roles in both atmospheric and combustion systems. For example, the hydroxyl radical, OH, is an extremely important radical in tropospheric chemistry; and the CH and C₂ radicals are responsible for the green/blue colour of intense flames. Many of the radicals that may reasonably be inferred as being involved in the chemistry of the atmosphere, or a combustion system, have no known spectroscopic signatures, and therefore cannot be simply monitored. It is proposed to obtain more information about the electronic spectroscopy of such species, in order to identify such spectroscopic signatures. As a precursor to such studies, it will be necessary to develop means of production of the radicals in the laboratory, using wel1-established methods, and novel ones in some cases. Specific species that are of great interest to atmospheric chemisty are peroxy radicals (e.g. HO₂, and CH₃O₂) chlorofluorohydrocarbon radicals, (e.g. CFCl, and CFCl₂); halogen oxides (e.g. BrO, IO, and BrO₂, etc.); and metal containing reactive intermediates (e.g. AlOH, MgOH, and AlO₂). Specific species that are of great importance to combustion chemistry are hydrocarbon radicals (e.g. C₂H, CH₃C, tropyl, and polycyclic aromatic hydrocarbon, PAH, radicals). Most of these species will be studied as part of the proposed work.

(ii) Spectroscopy of Cations

Simple ionic species are dominant in the upper atmosphere, produced either via direct photoionization (the main production route during daylight hours), or via chemi-ionization reactions (the main route at nighttime). Ions can undergo very fast reactions with neutral molecules, because of the additional attractive charge/(induced-)dipole interaction compared to neutral-neutral reactions. Metal and simple metal compound ions are very common in the Earth's upper atmosphere during daylight hours, owing to the generally low ionization energies of such species; however, metal-containing ions are also reasonably common during the nighttime since they can be formed via chemi-ionization reactions, such as:

M+ O à MO⁺ + e^-

M+ O₂ à MO₂⁺ + e^-

MO + O à MO₂⁺ + e^-

An understanding of the energies with which cations may be formed both via photoionization and chemi-ionization is important, as this excess (rovibronic) energy can affect subsequent reactions.

The ionic nature of flames has been known for about 150 years. However, the route by which these ions arise is still not conclusively established, although it is known that thermal ionization is only one means by which they are produced. The most widely-accepted reaction is the chemi-ionization route:

CH + O à HCO⁺ + e^-

although other reactions such as

CH + C₂H₂ à C₃H₃⁺ + e^-

and

C₂ + CH₃ à C₃H₃⁺ + e^-

may play a role.

These reactions are of added importance, since it is believed that a series of ionic reactions occurs leading eventually, via polycyclic aromatic hydrocarbons (PAHs) to soot formation. Obviously, being able to control the chemi-ionization routes, and consequently also the formation of soot would have significant beneficial effects. In order to do this, it is necessary to know the energy content of the CH, C₂ etc. as they are formed in the flame. This is only possible via sophisticated laboratory experiments where isolated reactions can be studied under controlled conditions; or in flames, where the energy profile of the radicals may be examined from a spectroscopic probe experiment - it should be noted, however, that this requires a knowledge of the energy levels of the radical, which themselves are determined from laboratory spectroscopic measurements. There are a number of isomers of the C₃H₃⁺ species, and these will be investigated as part of this work; the subsequent reaction of C₃H₃⁺ with C₂H₂ can lead to C₅H₅⁺ and C₇H₇⁺, which will also be investigated.

The feasibility of ion-molecule reactions depends not only on thermodynamics, but also on kinetics, and so laboratory measurement of rate constants is of importance in trying to elucidate the importance or otherwise of a set of reactions. The reaction rate can vary dramatically with the internal energy of the reactants and the local temperature and pressure, and so measuring the reaction rate as a function of these parameters, although challenging, is an important step in the detailed understanding of complicated reaction schemes.

(iii) Reactive Intermediates Produced bv Photolysis and Pyrolysis

Many reactive species in the Earth's atmosphere are formed from photodissociation of stable species by sunlight. The species produced may be in a number of different electronic states, and each of these will have a certain distribution of internal rovibrational excitation. In addition, there may be more than one photodissociation route, with the branching ratio between them depending on the initial excitation of the parent molecule, and the wavelength of the photolysis source. A detailed knowledge of the product species, and their internal excitation, is very important for an in-depth picture of the profile of reactive species capable of undergoing further reactions. Of particular interest here is the photodissociation behaviour of fluorocarbons, which are increasingly being used as replacements for CFCs, although their photochemical behaviour is poorly understood. In addition, the photochemical behaviour of other common pollutants such as SO₂ and OCS will be investigated.

The trapping of chlorine atoms in compounds such as ClONO₂ is now known to have a dramatic effect on the ozone destruction rate. These compounds are rather facilely dissociated by sunlight, although the detailed dynamics are little understood; in addition, the photodissociation behaviour of the (ClO)₂ dimer will be investigated.

Azide compounds are used in a number of applications to produce molecular nitrogen in situ, such as in space vehicles, and in car airbags. They can also be used as a source of active nitrogen in chemical vapour deposition. Consequently, an understanding of azide decomposition is of importance to yield the most efficient sources of the appropriate nitrogen-containing species. It is planned to study the decomposition behaviour of a number of azide compounds in this work, to elucidate their decomposition routes, and so assess their efficacy.

(iv) Ion-Molecule Reactions as a Function of Pressure, Temperature and Size

Production of protonated water clusters in the upper atmosphere is thought to occur via the reaction:

NO⁺.(H₂0)_n + H₂0 à H⁺.(H₂0)_n + HONO

It may be seen from this reaction that a complete switch from solvated nitrosium cation chemistry to the chemistry of protonated water clusters has occurred. This is an example of an intracluster reaction that can dramatically alter the observed chemistry of the atmosphere. Such intracluster reactions also occur involving other cationic species, such as NO₂⁺ and O₂⁺.

The chemistry of a particular complex may be investigated by isolation of a particular complex size. This is important, as different sized complexes can behave in many different ways. For example, in the above equation, if n = 3 or smaller, almost no conversion to protonated water complexes occurs, whereas if n > 5, almost all of the nitrosium complexes convert to protonated water clusters.

The reaction of anions with atmospheric pollutants, such as fluorohydrocarbons, will also be addressed, since this is an area that has received little attention to date. Common atmospheric anions include O^-, O₂^- and OH^-, which can react with the fluorocarbons, extracting H atoms, and leading to reactive species; in addition, reactions between these anions and cations lead to neutral species, and so effectively leading to a cessation of the ion-molecule chemistry.

Consequently, it may be seen that ion-molecule chemistry is intricately involved with atmospheric chemistry, and in many cases is little understood. It is proposed both to develop means of characterising ion-molecule complexes spectroscopically, and also to investigate their chemistry under a wide range of variation of size, pressure and temperature.

(v) Heterogeneous Processes on Solid Surfaces and Clusters

The role of grains and ice particles in atmospheric chemistry has been highlighted in recent years. Effectively, the ice and grain particles serve to constrain the degrees of freedom in which reactant molecules can move, and so increase the likelihood of reaction. One can mimic the ice-particle reactions by studying the chemistry occurring in larger and larger water clusters. For the grain chemistry, the study of surface reactions is important.

The growth of clean films is important in the semiconductor industry. Often, the chemistry involved is arrived at in a "hit and miss" type approach; mainly as the plasma chemistry involved is little understood. The ability to detect and monitor, in situ, the species present in a plasma will be aided by knowledge of spectroscopic signatures. In addition, it is also proposed to study the dissociation behaviour of azides on surfaces, which can yield nitrogen-containing films.

The understanding of surface reactions via reactions on small metallic clusters is an important step in the development of, for example, catalytic converters. It is proposed to study various reactions on small size-selected metallic clusters, and to compare the results with similar reactions on bulk surfaces.

The main originality of this proposal arises from combining the expertise of eight major research groups to study a small number of scientific and technological problems associated with the identification and characterisation of excited electronic states of small reactive intermediates and clusters, and to study related ion-molecule reactions in a fundamental way. Each group has its own recognised area of expertise which is complementary to the expertise of the others groups. Five of the research groups involved in this project (in Amsterdam, Orsay, Crete, Lisbon and Southampton) are internationally recognised for their original contributions in these spectroscopic areas, and the other groups (Salisbury, Garching, and Bruker) have international reputations in ion-molecule chemistry.

Short-lived molecular fragments and radicals can be produced by a variety of methods from suitable precursor compounds. Thermal dissociation has proved to be a general method of making reactive intermediates, while microwave discharge methods have proved very suitable for the generation of halogen atoms which can subsequently be employed for the generation of transient species via rapid halogen atom-molecule reactions. Because of the advent of reliable high-intensity laser sources and efficient frequency-doubling methods, tunable uv radiation can be readily produced in the laboratory. Using such radiation, photofragmentation of precursor compounds leading to transient species which cannot be otherwise produced can be achieved. Moreover, the atomic and molecular fragments which are formed are often not produced in their ground states, but can show significant degrees of electronic, vibrational, and rotational excitation. At present, the use of photofragmentation methods for the generation of transients is a rapidly expanding field world wide.

Molecular clusters are commonly produced in a pulsed molecular nozzle beam. Molecular beam techniques are well-proven methods, and molecular clusters of greatly differing size and complexity can be generated in reasonable concentrations.

Conventional photoelectron spectroscopy (PES) has been applied to the study of transient molecules and clusters for a relatively long time, and with significant success. Drawbacks of the method are its relatively low sensitivity, its inherent lack of selectivity because it is a one-photon excitation method, the fact that the excitation wavelength is not tunable, and the fact that it is usually only applicable to studies of electronic ground states. In recent years, spectroscopic laser methods with sufficient sensitivity and resolution to characterise small transient molecules and clusters in detail have been developed. An important development has been the advent of multiphoton ionisation (MPI) methods which divide the photoionisation process into a two-step process where the intermediate state provides a significant enhancement of the overall probability of the process relative to a two photon non-resonant process. In the first step, an excited state is populated through absorption of one or more photons; in the second step, photoionisation from the excited state takes place through the absorption of more photons. This double-resonance scheme provides excellent chemical selectivity, and the method can be employed for the study of excited states of the chosen gas-phase species. The MPI process can best be detected through mass-resolved ion detection or through kinetic-energy resolved electron detection (MPI-PES). The latter method is extremely powerful in that it allows the detailed determination of the excitation energy in the final ionic state. Moreover, MPI-PES is unique in the sense that it gives direct experimental access to determining whether or not excited states of molecules possess mixed electronic character. A very recent development in the area of MPI-PES is Zero Kinetic Energy (ZEKE) electron detection which allows photoelectron spectroscopy to be performed with essentially laser limited sub-wave number resolution. The application of these sophisticated spectroscopic methods to transient species and molecular clusters has proved to be feasible, but has only been attempted in a very limited number of cases. The present proposal aims at extending these advanced techniques to a much broader range of radicals and clusters of importance in atmospheric and combustion chemistry than has previously been attempted. Since the electronic structure of most radical and cluster species which are atmospherically relevant have only been poorly characterised, there is a clear need for extending the above spectroscopic methods to these species.

An additional technique, which because of its sensitivity has proved its value under many different experimental circumstances, is that of dispersed laser induced fluorescence (LIF). This technique will also be employed in characterising the ground states of many of the species which are part of the present proposal.

The electronic structure calculations proposed as part of this project underpin the experimental programme – to study the electronic structure of reactive intermediates with uv photoelectron spectroscopy, to study the structure and reactivity of clusters and their ions, and to study molecular excited states by laser spectroscopy. With the expertise that has been developed in Southampton in the last ten years and the interaction that has been established with the experimental programme at CBD Salisbury, the application of electronic structure calculations notably to ionic clusters and ion-molecule reactions studied in Lisbon, Garching and Salisbury is clearly original. Of particular value will be calculations performed on reactions of ionic clusters which will be studied experimentally.

In each case, the proposed calculations begin with geometry optimisation, using analytical gradient methods on all species involved in a chosen chemical reaction. This will locate stationary points on the energy surfaces. Second derivative calculations at minima give the harmonic vibrational frequencies for the species studied. Also thermodynamic constants, such as standard enthalpy and entropy changes can be evaluated for the reaction considered. The results of such studies should allow experimental conditions to be specified which allow a particular ion to have a higher partial pressure than others, thus providing favourable conditions for spectroscopic and kinetic measurements.

The original element of the ion-molecule programme is that it is targeted at studying ion-molecule reactions of atmospheric importance over a wide pressure range. Reactions will be studied at ~760 torr using an Ion Mobility Spectrometer (IMS) and at ~ 10^-7 torr in an Ion Cyclotron Resonance-Fourier Transform Mass Spectrometer (ICR-FTMS). In addition, use will be made of a low energy electron attachment facility at 160 torr and a Selected Ion Flow Tube (SIFT) at 0.5 torr. At Porton Down, Salisbury, two major research themes will be pursued: (i) the reactions of stable molecules with negative ions; and (ii) positive ion reactions initiated by proton transfer to olefins and alcohols. In addition, there is an underlying desire to understand the influence of hydration on the reactivity of the ions studied; experiments will be designed to address this issue. At Garching, related experiments are proposed which use large ionised water clusters and a variety of hydrated ions as model systems for investigating heterogeneous reactions of relevance in atmospheric chemistry. An FT-ICR mass spectrometer has been modified to generate the required clusters, using electric discharge or laser vapourisation methods, and to study the reactions of size-selected clusters. Examples of reactions which illustrate the potential of the method will now be given. Using a corona discharge, water clusters of the type X(H₂0)_n can be generated where n ranges up to 100 and X is H⁺, O^- and OH^-. These clusters can be easily doped with acids of atmospheric importance, for example nitric or sulphuric acid. The FT-ICR method has the advantage that it permits the isolation and mass selection of any observed cluster which can then be reacted under well-defined conditions. The reaction rates can be measured and reaction products unambiguously identified. It is proposed to use reactions of large clusters of this type with ozone, hydrochloric acid and chlorine nitrate to gain a better understanding of stratospheric reactions. Also, sulphur-containing clusters can be easily generated by co-expanding the products of an electric discharge through for example SO₂ with water, or by reacting pure water clusters with SO₂ or SO₃. lt is also proposed to investigate the reactions of sulphur-containing clusters and the formation of sulphates and sulphuric acid using the FT-ICR method. Reactions of solvated metal clusters relevant to reactions occurring in the upper atmosphere can be studied in a similar way. The study of atmospherically-relevant cluster reactions in this fashion is highly original, and represents the current state-of the-art in the field.

In summary, the methods to be employed are state-of-the-art in many cases, with the senior researchers being world leaders. Consequently, this work will have a large number of man-years experience and knowledge behind it, ensuring the maximum chance of success in all areas. In addition, the group has already met, and has been seen to work well in our initial meeting to plan the proposal. Thus a powerful team is available to work on some very important chemical problems. The setting up of a Web site, frequent meetings and transfer of knowledge via the postdoctoral researchers will provide a steady flow of knowledge and expertise around the group. The scientific originality of this Network lies in the use of research expertise from different disciplines to address common objectives which are themselves key issues at the forefront of research in spectroscopy and dynamics, and ion-molecule chemistry.