Ultrasound in materials chemistry Dietmar Peters Universitat Rostock FB Chemie Buchbinderstr. 9 D-18051Rostock Germany Ultrasound has applications in materials organometallic polymer and synthetic chemistry as well as materials and waste degradation. After a brief description of the fundamentals of sonochemistry an overview about current uses and future prospects in these fields is given. The first steps in sonochemistry were taken about 70 years ago,’ but the extended application of ultrasound as a tool for synthetic and materials chemistry did not begin until the 1980s. Since then many papers have been published and sonochemis- try has become a versatile approach in several areas (for some related reviews see ref. 2-8).. The subject is quite large and one had to select from a large number of papers..Thus this article will focus on applications to materials chemistry (metals solids polymers biological materials) sonochemistry involving metals and non-metallic solids materials degradation and related industrial and laboratory applications in these fields. Even with such restrictions a personal choice was unavoidable and some fields are only described in general and not in detail. Therefore this article attempts to give an overview of the scope and trends in the application of ultrasound in materials science and related fields in chemistry. Fundamentals The part of the sonic spectrum which ranges from about 20 kHz to 10 MHz is called ultrasound and it can be sub- divided in three main regions power ultrasound (20-100 kHz) high-frequency power ultrasound (100 kHz-1 MHz) and diag- nostic ultrasound (1-10 MHz)..The latter range is also often called high-frequency ultrasound. Acoustic energy is mechanical energy i.e. it is not absorbed by molecules. Ultrasound is transmitted through a medium via waves by inducing vibrational motion of the molecules which alternately compress and stretch the molecular structure of the medium.. Therefore the distances between the molecules vary as the molecules oscillate about their mean position. If the intensity of ultrasound in a liquid is increased a point is reached at which the intramolecular forces are not able to hold the molecular structure intact. Consequently it breaks down and cavitation bubbles are created..This process is called cavitation and the point at which it starts is known as the cavitation threshold. Two forms of cavitation are known stable and transient. Stable cavitation means that the bubbles oscillate about their equilibrium position over several refraction-com- pression cycles while in transient cavitation the bubbles grow over one (sometimes two or three) acoustic cycle to double their initial size and finally collapse violently. There are three different theories about cavitation the hot- spot the electrical and the plasma theory. But according to each theory there is no doubt that the origin of sonochemical effects is cavitation. Furthermore it has been shown experimen- tally that cavitational collapse creates drastic conditions inside the medium temperatures of 2000-5000 K and pressures up to 1800 atm within the collapsing cavity..Thus from a practical point of view the parameters which influence cavitation are important (Table 1).However it should be noted that there is often no simple relationship and an optimum can generally be found for all parameters. For details of theoretical aspects of ultrasound and cavitation refer to the From a practical point of view there are three possible reaction sites of a collapsing bubble the cavity interior the bubble vicinity and the bulk solution (Fig. 1).. The following beneficial sono- chemical effects can be observed (i) ligand-metal bond cleavage in transition-metal complexes to give coordinatively Table 1 Ultrasound parameters influencing ~avitation~,~.’-~~ parameter frequency intensity solvent bubbled gas external temperature external pressure effects low long cycle large bubbles low amplitude required to induce cavitation high short cycle high amplitude necessary increased attenuation weak or no cavitation in the MHz range (rarefaction cycle is too short to create bubbles) considerably higher intensity at high frequency is necessary to maintain the same cavitation as at low frequency indefinite increase limited by the material stability of transducer decoupling with the medium and a large number of bubbles (transmission barrier) the higher the vapour pressure the less violent the collapse (increased penetration of vapour into the bubbles) induction of cavitation is more difficult in solvents with low vapour pressure cavitation is easier in solvents with low viscosity and surface tension y =CJC should be high as the collapse temperature is proportional to (y -1) the smaller the thermal conductivity of the gas the higher the local heating during the collapse the greater the amount of dissolved gas the smaller the intensity of the shock wave dissolved gas acts as cavitation nuclei and leads to more facile cavitation temperature rise increases vapour pressure and collapse is less violent less intensity necessary to induce cavitation temperature near the boiling point of the solvent dramatically increases the number of bubbles which can act as sound barrier pressure rise decreases vapour pressure and collapse is more violent; higher intensity is necessary to induce cavitation optimum depends on frequency J.Mater. Chem. 1996 6(lo) 1605-1618 1605 bubble core (formation of ladicalslexcitcd mo!e+g bubble vicinity (pressudtemperaturcgradient elarical fields jetdshock waves onto meWsolid I surfaces or other phase boundaries)I \ -- - - -' bulk solution (solid or toluted reactants erpelled reactive intermediates) Fig. 1 Reaction sites of a collapsing cavitation bubble unsaturated species or modified complexes as well as complete removal of ligands to produce amorphous metals; (ii) dis- ruption of the solvent structure altering the solvation of reactants; (iii) sonolysis of molecules (homolytic fragmentation to radicals rupture of polymers generation of excited states cell disrupture); (iv) mechanical effects by cavity collapse onto metals and solids (shear forces jets and shock waves resulting in rapid mass transfer surface cleaning particle size reduction crystal defects and metal activation); (v) effects in liquid-liquid systems (improved mass transfer emulsification increase of the effect of phase transfer catalysts or even their replacement); (vi) effects in gas-liquid systems (degassing of liquids or melts atomisation of liquids in air); (vii) single electron transfer (SET) steps in chemical reactions are accelerated and if an ionic and an electron transfer pathway are possible the latter is preferred ('sonochemical switching').Metals Powders and other Particles Reactions and processes involving metals powders or other solid particles are probably the most successfully investigated fields in the application of ultrasound in chemistry and mate- rials science. Table 2 gives an overview of such fields. Decomposition of metal carbonyls The effects of ultrasound on metal carbonyls were investigated initially by Suslick and co-workers" in 1981. Upon irradiation of a decane solution of Fe(CO) under argon with ultrasound (20 kHz 100 W cm-') they observed an unusual Fe3(CO)12 cluster formation together with the formation of fine powdered iron (Scheme 1).The unique nature of the ultrasonic treatment can be seen by comparison with the effects of light and heat.Table 2 Applications of ultrasound to metals and non-metallic solids material applications of ultrasound metals production of amorphous nanosized metal powders and nanocolloids preparation of activated and supported metals used as catalysts agglomeration of metals formation of metal carbides electroplating and spray pyrolysis to form metal layers crystallisation metal welding machining soldering casting sonocleaning organometallic sonochemistry solids impregnation of catalysts on solid supports preparation of fine particles and colloids particle size reduction cavitation erosion surface treatment layered solids dispersion dyeing sieving heterogeneous sonochemistry rFe Scheme 1 Sonolytic thermolytic and photolytic decomposition of Fe(CO) (ref.6) Photolysis of Fe(CO)5 yields Fe3(CO)9 via Fe(CO) which reacts with unconverted Fe(CO)5 (Scheme 1).. Thermolysis of Fe(CO) results in decomposition to iron clusters and CO. The formation of amorphous iron depends on the sonication parameters influencing cavitation (solvent vapour pressure. gas).. The influence of the reaction conditions on the measured properties (e.g. particle size) of sonochemically produced amorphous iron has been described recently.12 Sonochemically produced iron is an amorphous pyrophoric powder having a surface area of about 150m2 g-'. It has a coral-like structure which is built up from nanosized clusters. DSC measurements show an irreversible crystallisation exotherm at 350°C reflecting the formation of normal fully crystallised u-iroa6 Using sonoluminescence as a spectroscopic probe of iron pentacarbonyl decomposition Suslick6 estimated temperatures of >5000 K and pressures of about 1700 atm for the hot-spot parameters of the cavitation event during a period of less than 100 ns..Therefore cooling rates of more than 10" K s-' are deduced for this process.. These rates are sufficiently high that the material can be frozen before crystallisation occurs and amorphous metals can be formed. The initially formed amorphous nanoscale metal can be trapped by a solid support. Upon adding a polymer [e.g. poly(vinylpyrrolidone)] or silica to the sonicated solution nanocolloids and supported metal catalysts respectively are made..These materials have interesting magnetic properties. Nanophase amorphous iron is a very soft ferromagnet having a magnetic moment between those of crystalline and molten iron. Nanocolloidal iron is superparamagnetic and nanophase iron supported on silica has similar magnetic properties. Other metal nanopowders using different transition-metal carbonyls as precursors are also available. Amorphous metals on supports are interesting catalysts as will be shown below. Recently sonochemical iron pentacarbonyl decomposition has been carried out under an air atmosphere instead of under argon.. The X-ray diffraction pattern of the produced fine powder showed the formation of a non-crystalline phase of magnetite.Annealing of this powder for 2 h at 770 K resulted in a polycrystalline phase of magnetite with the same diffraction pattern as a sample produced by milling of a magnetite single ~rysta1.l~ Sonoelectrochemistry producing metal nanopowders and metal layers In recent years ultrasound has been introduced into electro- chemistry. Beneficial effects are among others acceleration of mass transport cleaning and degassing of the electrode surface or disturbing the diffusion layer.14-16 A pplications besides the use in electrochemical synthesis and electroanalytical methods are for example the electrodeposition of metals or alloys (e.g. Cu Zn Cu/Zn Ni/Fe; for reviews see ref. 7 17 and references therein) the production of catalytically active powders (e.g.Cu Co Zn)" and the preparation of Si films." Improved physical and mechanical properties such as better hardness brightness and adhesion as well as a higher depos- ition rate an increased plating current lower internal stress of 1606 J. Muter. Chern. 1996 6(lo) 1605-1618 the coatings and a more uniform microstructure are the benefits of using ultrasound in electroplating.. The effects of 1.2 MHz and 20 kHz ultrasound applied at an intensity of 4 W cm-' have been compared in a study of the electrodeposition of Cu by Drake.,' Using the high-frequency ultrasound the diffusion layer is reduced from 200 pm to about 20-30 pm whereas 20 kHz ultrasound reduces it to 3.4 pm.The properties of porous silicon thin films produced by open-circuit stain etching are better under sonication compared to those of samples generated in the absence of ultrasound. The differences induced by ultrasound in surface morphologies (rougher and thicker films) and chemical stability (greater stability on exposure to water and organoamines) proved that ultrasonic irradiation is a useful t00l.l~ Pulsed ultrasound has recently found more interest in chem- istry.,' Reisse and co-workers" described the pulsed sonored- uction of metal salts in aqueous solution which resulted in the formation of metals and other solid powders (Cu Co Zn Ni Cr Ag Co/Ni MnO CdTe) with particles sizes of about 100 nm.. The sonotrode acts as ultrasound emitter and also as the cathode.. The 20 kHz high-intensity ultrasound pulse (pulse duration 100 ms followed by a 200 ms switch off) is superposed by an electrical current (pulse duration 300 ms).During sonication the cathode surface is cleaned the metal particles are expelled (preventing the growth of large particles) and the double layer is replenished with metal cations.. The metals obtained are fine crystalline powders with high surface areas and chemical purity.. The electrical yields are about 75-90%. Using a 1 cm diameter sonoelectrode amounts of 1 g h-' can be produced. Sonoelectrochemically produced Zn has been used as an effective catalyst in the Barbier-type reaction (Scheme 2). Sonication of metal powders solids and supported metals used as catalysts or reagents Owing to cavitational collapse sonication of metals or solids leads to microjet and shock-wave impacts on the surface which can result in particle size reduction interparticle collisions depassivation surface cleaning defects and erosion..The extent of erosion depends on the type of metal. Sonication of soft metals with hard oxides (e.g. Al Na Li) results in metal deformation which will damage the oxide layer. Hard metals are not plastically deformed by ultrasound but the surface is activated by the above-mentioned cavitation effects owing to the low cohesion of the oxide coating. If the particles are very small and/or in close proximity (e.g. in slurries) they will impinge (interparticle collision) agglomerate and even fusion can occur owing to the high temperatures of the hot spot.6 A recent study' was devoted to the effects of the particle size morphology of Cu and Pb sonicated in dilute HCl using a 38 kHz cleaning bath.It was found that the cavitational modification of the metals depended on the initial size of the solids. Large copper turnings or lead foil are greatly depassiv- ated by microjets. Small particles <ca. 100-150 pm (in the study:" Cu<63 pm and Pb ca. 150 pm) are not subjected to microjet impacts because a collapsing bubble will have an average diameter of about 150 pm at 20 kHz. Consequently Zn powder yield sonoelectroproduced 82% (70% isolated) commercial 27% Scheme 2 Application of sonoelectrochemical produced Zn powder" shock-wave impacts and interparticle collisions are dominant and only small differences to a stirred control probe were found.There are in principle two main groups of applications of ultrasound to metals and solids in sonochemistry and they are described below. (1) Sonochemical preparation of catalysts used in non-son-icated reactions. Applications include the preparation of acti-vated metals by reduction of metal salts (e.g. reduction with Li in THF to Rieke-type powders or with formaldebvde to Pd or Pt); the generation of activated metals by sonicntion; the precipitation of metal (Cr Mn Co) oxide catalysts and the impregnation of metals or metal halides on supports. The reduction of metal halides with Li in THF in a low- intensity cleaning bath gives rise to active metal powders comparable to Rieke-type powders.Sonication for less than 1 h (the conventional Rieke procedure requires about 8 h and stirring for up to 26 h) yields powders of Mg Zn Cu Ni Cr Co Pt and Pd. Platinum black produced by reduction of platinum halides under ultrasound has increased activity in the hydrogenation of alkenes up to a factor of three.23 Furthermore increases in magnetic susceptibility (98%) and surface area (62%) has been measured.. The most catalytically active platinum and palladium blacks have been obtained at 20 kHz (Pt) and 3 MHz (Pd).24 Similar activity enhancements have been found for Ni and Co obtained after precipitation of metal oxalates under sonication and subsequent reduction to the meta1.25s26 The sonication of ordinary commercial Ni (known as a poor catalyst) to obtain highly active metal powder has been investigated extensi~ely.~~~"~~Irradiation with ultrasound increases the catalytic activity dramatically reaching that of Raney nickel.Marked changes in the surface morphology have been observed the surface is smoothed by removing the crystallites and the oxide layer.. The particle size is hardly reduced. If metal slurries containing different metals are son-icated melted necks between two different metal particles can be observed.. Thus S~slick~,~' found that for example in the cases of Fe and Sn the neck is an alloy between the two metals i.e. high velocity collisions with sufficient energy to melt the metals have occurred.If the particles experience glancing collisions smoothing occurs. From investigations of different metals one can estimate the local conditions in the slurry. If Fe and Cr (which melt at about 1500 and 18OO0C,respectively) are irradiated tremendous agglomeration is observed. Sonication of Mo (mp ca. 2600 "C)still results in agglomeration and practically no smoothing is observed. In the cme of W ultrasound has no effect.. Thus a temperature limit of ca. 3000°C can be estimated. In the cases of Mo and W the production of metal carbides (e.g. Mo2C) has also been reported re~ently.~' Sonication of a mixture of MoC1,-SiCl and sodium-potassium alloy followed by annealing at 900 "C gives nanocrystalline MoS~,.~ Sonochemically obtained MoSi shows 50-70% higher microhardness and compression strength than the conventional coarse-grained MoSi,.Sonochemically obtained Ni powder is a highly hetero- geneous catalyst for hydrocarbon reforming and CO hydrogen- ation. In the hydrogenation of alkenes it is comparable with Raney Ni; however it is more selective as C=O groups are ~naffected.~"~'Ultrasonic precipitation of chromium molyb- denum oxide yields a catalyst with 15-20% increased dispersity and better activity for the oxidation of methanol to f~rmaldehyde.~~ In 1973 the sonication of aqueous suspensions of Cr2O3 Co,03 MnOz and alumina giving supported metal oxide catalysts of higher activity and dispersity for H,OZ decomposition was rep~rted.~' Thus the surface area of sup- ported Cr,03/A1,03 is enhanced from 108 mz g-' (normal preparation) to 135 mz g-' (sonochemical preparation). Impregnation of Pt on silica gel produced by sonoreduction J.Muter. Chem. 1996 6(lo) 1605-1618 1607 of ammonium hexachloroplatinate at 440 kHz yields a metal catalyst with an 80% increased surface area compared to that .~prepared by mechanical Ragaini et ~2 reported ~ recently on the dispersion of Ru on alumina. RuC1 (in water) and Ru(acac) (in toluene) absorbed on A1203 were subjected to hydrazine reduction under ultrasound. At low Ru content (0.5%) a higher dispersion (66% with RuC1,j was found compared to a non-sonicated run (37% with RuCl,). With a higher Ru content (5%) the dispersion was relatively low in both runs and was comparable in value (about 20%).. The applied power is critical as at high power the alumina support is affected.A further study investigated the bromination of aromatics with CuBr,/A1203 .3 Sonication of the reaction mix- ture with non-impregnated reagents led to a substantial rate improvement for the bromination of naphthalene compared to a non-sonicated run. However even better results were obtained when CuBr was sonochemically preimpregnated on alumina followed by conventional procedures. Sonochemically supported CuBr on Al,O showed reduced particle sizes and pronounced changes in the surface morphology. Multicomponent catalysts used in chemical technology can also been activated or reactivated Ziegler-Natta Ni/Mo cracking Ti/V oxide or Pd-alumina deni trification catalysts (see ref.7 p. 54ff j. (2) Activation of solid reactants (metals and non-metallic solids) in heterogeneous chemical reactions by ultrasound. Applications include the preparation of activated metal solu- tions; the preparation of organometallic compounds from main group or transition metals; sonochemical reactions involving metals via in situ generated organoelement species; and reac- tions involving non-metallic solids. There are some obstacles to reactions between a solid and a liquid (or in a liquid-dissolved) reactant in a heterogeneous system e.g. the small surface area of a bulk solid; the solid surface may be coated by oxide layers or impurities; species have to diffuse to and away from the solid surface; and deposition of products may inhibit further reaction.These difficulties can be overcome by the effects of ultra- sound due to cavitation as has been mentioned already in part shock waves (causing plastic deformations on soft mate- rials break-up of coatings) microjets (causing surface erosion defects and deformations enhancing the surface area) and microstreaming (improving mass transfer removing eroded particles and deposited impurities or products).. There are Organotin fluorides are insoluble in most solvents owing to their polymeric structures.. Therefore their reactivities are extremely low. Sonication of derivatives R3SnF (R =Bu Ph) with metal salts NaX (X=Cl Br OCN SCN) leads to the rapid formation of monomeric compounds R3SnX.66 An interesting effect of sonication on the reaction pathway has been observed in the reaction of solid KCN/A1203 with benzyl bromide (Scheme 6).39767.The course of the sonochem- ical reaction is completely changed as under stirring a different product is observed.. This so-called sonochemical switching (see Scheme 6 for examples) is explained by different reaction mechanisms. Under sonication an electron-transfer mechanism (leading in this case to a different product) is preferred whereas under stirring an ionic pathway dominates. As this chapter covers mainly synthetic chemistry aspects some potential future trends and questions will be given. At present sonochemistry is an established field in laboratory preparations.However an as yet unsolved problem is the scale-up on the one hand the scale-up of a laboratory synthesis to a 20-50g scale and on the other hand chemical synthesis on an industrial scale.. There are already examples reported in these two areas e.g.the preparation of perfluoroalkyl aldehydes on a 40g scale6* and the introduction of ultrasound in three steps of a complex 14-step steroid synthesis to desogestrel yielding the intermediates on a kg scale.69. Thus the existing sonochemical equipment allows in principle the synthesis on a kg scale and the realisation of higher bulk amounts should not be a technical problem.70. The production of kg amounts of compounds seems to be economical in the preparation of highly valuable compounds especially drugs in the pharma- ceutical industry.Further bulk applications in synthesis are still limited because of cost concerns. In laboratory applications the following trends will be followed synthesis of complex molecules and natural products; introduction of sonochemical steps in multistep syntheses; increased use of ultrasound in established fields of chemistry i.e. the combination of sono- chemistry with electrochemistry photochemistry high-pressure chemistry polymer chemistry and biotechnology; use of ultra- sound in reaction mechanism studies; and study of frequency effects and reactor design dosimetry of the sound field. Other applications to metals and non-metallic solids The use of ultrasound in this field of materials science is already extensive which can be exemplified by some further excellent reviews on heterogeneous son~chemistry.~.~,~~-~' selected applications and related recent publications crystallis- Therefore some recent papers have been selected as examples to illustrate the wide range of metals and solids used in synthesis.Highly active reducing agents can be prepared by the sonication of Mg in the presence of anthra~ene,~' K in toluene and Na in ~ylene.~ The alkali metals are obtained in colloidal solution. Similarly highly dispersed Hg emulsions are a~ailable.~ Organometallic derivatives are useful reagents in organic synthesis.. They are either prepared or commercially available or generated in situ.Schemes 3-5 show examples of reactions involving different main group and transition metals. The application of ultrasound to non-metallic solids has been reviewed extensively by Ando and Kim~ra.,~ Thus some .~~recent examples will be given. Goh et ~1 described the effect of ultrasound on sulfur-metal systems. Sonication of Cu Fe Zn or Mg with elemental sulfur in the presence of 2 mol dm-3 HC1 yields the corresponding sulfides (Table 3).. The increase of sulfide formation in the order water <hexane <CS2 corre- sponds to the solubility of sulfur in these solvents. Surprisingly iron is an exception giving the highest yield in water and only poor results in CS,. A similar reaction of selenium with alkali metals (Li K Na) giving the dimetaldiselenides M2Se2 has been published by.Thompson and Bo~djouk.~~ 1608 J. Mater. Chem. 1996 6(10) 1605-1618 ation and precipitation of metals alloys zeolites and other solid^,^'-^^ agglomeration of degassing of melts;78 spray pyrolysis to form thin or fine particle^;*^-^' treatment of solid surfaces; 92-94 dispersion of solid^;^',^^ prep-aration of colloids (Ag Au Q-sized CdS j;97-99 ultrasonic sieving,"' filtration'" and micromanipulation (transportation concentration fractionation) of small particle^;^^^^'^^ intercal-ation of guest molecules into host inorganic layered solid^;^,^ ultrasonic-aided development in advanced lith~graphy;"~ and electroless plating.105-109 Some of the items of this list will be described in more detail as the number of papers discussing these aspects has increased significantly over the last decade.Crystallisation and precipitation. In the treatment of liquid metals or alloys during solidification and of saturated solu- tions the crystallisation process is affected by ultrasound as follows inhibition of crust formation; intensification of heat transfer; increase in the nucleation and the growth rate; and affecting growth morphology.. Thus ultrasound induces changes in the formation of InSb crystals altering the crystal diameter changing the width of the facet region and the inclined angle of non-facet interfaces near the facet region for the (111) plane. For BiSb single crystals a strong decrease in Iithium 1441 sodium PhSe-SePh 1)) (Na) w 2 PhSeNa THF P\C(O) 5 min quantitauve 14% probe high power 50% 33% probe low power 66% 0% cleaning bath 100% 1471 potassium 1481 quantitative(E,Z) (E,E) = 8 162 U 83% Scheme 3 Sonochemical reactions involving alkali metals (ref.numbers in square brackets) the density of growth striations due to ultrasonic-induced convection currents in the melt has been reported.73 Non-ferroelectric but polar-oriented ceramics and films showing excellent piezoelectric or pyroelectric properties (Li2B407 Ba2TiSi208 Ba2TiGe20s) have received much atten- tion.. These materials can be prepared by surface crystallisation of glasses. However it is difficult to obtain well oriented crystallites by conventional crystallisation.Enhanced nucle- ation precipitation and oriented growth of desired crystals have been found when the glass surface is treated ultrasonically in suspensions of crystalline particles.. Thus surface-crystallised dense thin films of Li,B407 P-BaB,O Ba,TiSi,O on glass have been ~repared.~~.~' Ultrasonic irradiation also improves the precipitation of ceramic powders from solutions (Pb-Zr-Ti oxalate Pb oxa- late mullite-composition powder) and the hydrolysis of metal (Si Al) alkoxides. Accelerated precipitation and improvement of the homogeneity in the precipitates were observed.76 Ultrasonic spray pyrolysis. One of the most extensively used applications in materials science is ultrasonic spray pyrolysis (USP) which is used to produce either fine particles or thin films.Particles prepared by this ultrasonic method have the following interesting features spherical shape; uniform size distribution; adjustable particle size from micron to submicron range; and high purity.. The process requires a short preparation time and is continuous. Furthermore the size distribution is controlled easily by changing the solution concentration. In general aqueous solutions of metal salts (e.g.chlorides nitrates alkoxides carboxylates) are sprayed and hence the products are metal oxide particles. Studies on oxide ceramic preparations have been intensified since the report of high-T superconductors. Exam- ples of sonochemically produced systems are Y-Ba-Cu-0 and Bi-Ca-Sr-Cu-0 materials (e.g. YBa2Cu307-x Bi2CaSr2Cu20x Bi2Ca,Sr,Cu30,) having the beneficial properties described above and showing similar characteristics to superconducting particles sintered from the powders by conventional solid-state reactions.88 Other oxide-type examples are the preparation of Ti0,-SnO powder89 used for stable humidity sensors and Pb(Zr,Ti)O (PZT)90 which is a widely used piezoelectric and electro-optic material.Recently the preparation of non-oxide particles has been reported fine ZnS particles were produced from Zn(N03)2thiourea complexes. Deposition at low temperature (400 "C) gave amorphous particles whereas at around 800 "C a hexagonal crystalline phase was observed.. The latter particles ranging from 0.5 to 1.3 pm are spherical with a smooth surface.ZnS particles which were rough and included a zinc oxide phase were found at high temperatures (900 0C).9' Ultrasonic spray pyrolysis to form thin films has many advantages very small droplets which can be transported without heating of the carrier gas narrow droplet size J. Mater. Chem. 1996 6(lo) 1605-1618 1609 magnesrum Y W*COOMe quantitative indium HOEx + tin 0 PhGH NH2 92% THF-H20(1 5),ry % 30min 98% Scheme 4 Sonochemical reactions involving group 11 I11 and IV metals (ref. numbers in square brackets) distribution; solvent vaporises as it reaches the substrate; deposition under atmospheric pressure fast deposition rate; easy control of film composition and thickness; pure deposits free of contaminants; and a variety of precursors inexpensive raw materials (such as nitrates or chlorides) metal-organic compounds precursors with low volatility or low stability may be used.At present not only USP-prepared films made from relatively simple compounds (e.g. TiO,," SnO,," ZnOE4),but also more complex systems such as Y-Fe gernet calcia- stabilized zirconia La _,Sr,MnO La -,M,CrO and super- conducting Y-Ba-Cu-0 films have been described." Even multilayer films such as PbTi0,(001)/LaNi03( loo)/ MgO( 100) have been prepared." Recently metal-organic chemical vapour deposition (MOCVD) using USP has been used to form ferroelectric BaTi0384 and oriented thin TiO films." For the latter an improved method using pulsed injection in conjunction with USP and allowing good control over the film deposition rate (growth rate achieved 2.5 monolayers per pulse) has been developed..The excellent control over film growth and proper- ties is also a key feature in the preparation of SnO films.. The USP technique allows the fast preparation (2 nm s-l) of a film having photoelectrochemical properties and a surface appro- priate for a solar. The combination of ultrasound with an inductively coupled plasma (spray-ICP) allows the use of 1610 J. Muter. Chem. 1996 6( lo) 1605-1618 nitrate solutions as precursors. Applying this method perov- skite-type oxides have been prepared LaCr0,,80381 PbTiO LaTiO LaNiO LaCoO LaA10 (see ref. 80 and refer- ences therein).Sonication has also been proved to enhance diamond forma- tion on various substrates.. Thus ultrasound has been used to increase nucleation by seeding with diamond dust in the deposition of diamond on a carbon-carbon composite.86387 Treatment of solid surfaces and particles. Ultrasonic surface treatment (UST) can have different effects on the structures of solids e.g. increase in hardness of metals cavitational erosion enhancement of nucleation sites and generating and affecting crystal defects. Some examples are given below. The impact of ultrasound on a solid depends amongst other factors on the irradiation time. As already mentioned in the electrodeposition of metals sonication can enhance the hard- ness of metals.. Thus brief irradiation (5 min) of various metals caused a hardness increase of about 27% (Ti.Nb) to 150% (CU)."~On the other hand prolonged sonication leads to erosion of metals and alloys due to cavitation. Data on the resistance of such materials are of interest for applications where hydrodynamic cavitation can occur. Sonication of these materials can easily give such data as has been shown for aluminium alloys. Ultrasonic irradiation effects directly affecting the properties of point and extended defects on semiconductors are due to PhCH2CNthe stimulation of different processes generation of Frenkel pairs dissociation of point defect complexes and enhanced gettering of point defects by sinks such as dislocation grain 70% 12%boundaries or precipitates.Point defect gettering in silicon- based materials such as silicon wafers and solar-grade polycrys- talline silicon are examples. For the latter an improved minority carrier length and enhanced dissociation of Fe-B pairs have been found.92 PhH-trace 60% Carbon nanotubes can be opened selectively and filled with Scheme 6 Examples of sonochemical switching3 certain metals or metal oxides. Sonication prior to the oxidative opening with nitric acid has been found to enhance the number of acidic groups formed in this oxidation step.. These groups on the surface of the carbon nanotubes can bind palladium ions strongly.. Thus the amount of nanosized palladium crystal- lites deposited on the inner and outer surfaces can be increased by sonication..This effect is due to the creation of small local defects in the tubes such as buckling bending and lattice dislocations on the surface.93 The dispersion and impregnation of particles by ultrasound is another important application.. Thus metal matrix com-posites have been prepared by dispersion of ceramic particles in liquid metals. Normally the poor wettability of the particles by the metal segregating and clustering are problems in this process.. The obtained particle size is > 10 pm.. Thus sonication has been applied successfully in the production of Si3N,- Al-Mg. First Si,N (1 pm) was presonicated in acetone and then the particles were irradiated directly in the melt. As a result the particles were well wetted by the A1-Mg matrix a homogeneous distribution was achieved and no agglomeration was observed resulting in particle sizes <5 The impregnation of solid particles can be exemplified by dyeing of leather9' and the preparation of dielectric polymer composites by impregnation of dielectric BaTiO gel into pores of microporous polyethylene membranes.In the latter process a Ba( TiOPr'),-Pr'OH gel with immersed microporous poly- ethylene is sonicated for 15 min.. The obtained dielectric mate- rial has a significantly higher relative permittivity and higher losses than a conventionally produced BaTi0,-polyethylene fiim.96 Polymers The sonochemistry of polymers consists of three main fields the degradation of polymers the ultrasonically assisted syn- thesis of polymers and the determination of the polymer structure (for reviews see ref. 7,38,112-114)..The last area has been reviewed recentlyll' and will not be discussed here. Polymer degradation It has long been known that the irradiation of polymer solutions reduces their viscosity e.g. sonication (960 kHz 6.8 W cm-*) of an air-saturated 1% solution of polystyrene in toluene reduces the flow time from 23.6 s to 18.3 s after 2 h.38 As the early investigators concluded the decrease in viscosity is due to the degradation of the polymer chains. In polymer chemistry not only the rate and the yield of the polymerisation reaction as well as the structure of the polymer but also particularly the molecular mass and its range and distribution in the polymeric material are of interest.From the experimental data available some general conclusions concern- ing the rate and the extent of the ultrasonic degradation process can be made.. The rate of depolymerisation decreases with decreasing molecular mass of the polymer but below a limiting molecular mass there is no further degradation. Furthermore the degradation is affected by the ultrasonic parameters (frequency intensity) solution conditions (solvent gas content polymer concentration initial molecular mass) as well as external temperature and pressure (see Table 4).. The results of the influence of frequency intensity solvent and external temperature are in accord with the effects of exper- imental parameters on cavitation (see Fundamentals section) i.e.degradation is improved if cavitation is favoured. The effects of gas and external pressure are somewhat more complicated.. There is no doubt that the degradation is higher in the presence of a gas. As would be predicted polyatomic gases show lower degradation rates than diatomic ones. However the rates in the presence of monoatomic gases are between those for polyatomic and diatomic gases. Attempts to explain this abnormal behaviour discuss the gas solubility and 1612 J. Muter. Chem. 1996 6(lo) 1605-1618 Table 4 Parameters influencing polymer degradation7s3' parameter effects on degradation frequency increasing the frequency reduces the extent of degradation intensity increasing the intensity increases the rate and the extent of degradation there is an upper intensity limit due to the material stability of the transducer decoupling with the medium and a large number of bubbles (transission barrier) solvent the higher the vapour pressure the less violent is the cavitation collapse and the less is the extent of degradation cavitation occurs more readily in solvents with low viscosity and surface tension bubbled gas the solution should be saturated with a noble gas which has a low thermal conductivity resulting in higher local heating more violent cavitation collapse and increasing polymer degradation the greater the amount of dissolved gas the smaller the intensity of the shock wave i.e.the lower the gas solubility the higher the degradation external increasing temperature increases vapour temperature pressure and collapse is less violent resulting in decreasing degradation temperature near the boiling point of the solvent increases dramatically the number of bubbles which can act as sound barriers concentration decreasing the concentration of the polymer in solution increases the degradation process molecular a larger initial molecular mass of the polymer mass increases the degree of degradation there is a molecular mass limit below which there is no degradation its thermal conductivity.Contradictory results were found for the effect of pressure (for details refer to ref. 7). A recent study by Price et a1.'16 on the ultrasonic degradation of polystyrene clearly showed that by suitable variation of reaction parameters (temperature solution concentration sol- vent vapour pressure ultrasound intensity) extensive control over the molecular mass and the polydispersity of the resulting material is possible.There have been several attempts to explain the degradation and to outline appropriate reaction mechanisms.. The following causes have been considered frictional forces shear gradients and impacts due to cavitational collapse; hydrodynamic forces caused by shock waves; and chemical reactions caused by reactive intermediates. Many of the observed trends in polymer degradation can be explained by these theories. However why the polymers are broken in the middle of the chains can only partially be explained as reported by Glynn and co-w~rkers,~~~,~~~ who found a Gaussian distribution of the scission around the middle of a chain.Direct thermal degradation by the high temperatures of the hot spot seems to play a minor or even infinitely small role as this is mainly random. Polymer synthesis There are two basic groups of polymerisation reactions to which ultrasound is applied (1) sonication of a solution already containing a homopolymer and either a second homopolymer or a monomer; and (2) sonication of a solution containing only monomers (with or without initiator). The first group is related to the macromolecular radicals formed as a consequence of the polymer chain cleavage. Sonication of a mixture of two polymers or a homopolymer and a different monomer results in the production of block or graft polymers.However only combinations of two homo- polymers having similar macroradical formation rates lead to block copolymers. 0therwise only degraded homopolymers are observed.ii9 Adding a radicophilic compound to a son-icated polymer solution leads to end-functionalized polymers or copolymers e.g.to the introduction of a fluorescent group.'l2 Following this strategy polymers with modified properties (solubility elasticity thermal behaviour etc.) or for special uses can be obtained. In the second group of polymerisation reactions several papers concerning the radically initiated polymerisation of vinyl monomers appeared.Conventionally the initiating rad- icals are generated by thermal or photochemical decomposition of the pure monomer or of an initiator (AIBN DBP etc.). One major aim of polymer chemists is control over the molecular mass structure and properties of the resulting polymer. As will be shown in the following section sonication of aqueous solutions generates high HO- and H- concen-trations. On the other hand initiators like AIBN or DPPH are cleaved readily to radicals by ultrasound.. The initiation reactions are accelerated by several orders of magnitude.. The propagation and termination reactions of growing radicals are affected only slightly or not at all by ultrasound in contrast to the radical production. At lower temperatures the micro- structure of the polymer can be influenced ultrasonically e.g.the portion of syndiotacticity increases from about 55% at 60 "C to 74% at -10"C in the peroxide-initiated polymeris- ation of methyl methacrylate (MMA).92. There are several factors influencing polymerisation cavitation-affecting param- eters (solvent temperature intensity) and other reaction con- ditions (e.g. concentration and monomer type). For details refer to the above-cited literature. That cavitation is essential for polymerisation has been shown for example by Price,"2 who investigated MMA polymerisation. At room temperature a final conversion of 12% was reached after 6 h. A pronounced sound change indicated that cavitation stopped at this point resulting in no further conversion.In general polymerisations do not exceed about 20% due to the rapid viscosity increase which prevents cavitation. Besides radical-initiated polymerisation ultrasound has also been applied successfully to emulsion and suspension poly- merisation.. There are also some papers on ring-opening and organometallic catalysis-based polymerisations under ultra- sound. Scheme 7 shows some examples of the latter. ring opening polperisation Application of ultrasound to non-radical catalysed polymerisationsMaterials degradation Practically no solvent is inert under ultrasonic cavitation Chemical effects are either a consequence of the sonolysis of solvent vapour inside the collapsing bubble due to the harsh reaction conditions (cf.fundamentals section) or can be caused by secondary reactions (at different sites see Fig.1) resulting from reactive species (mostly radicals or radical ions) generated during the collapse. Sonolysis in aqueous systems It has long been known (since 1929)l2O that H,O is a product of water sonolysis:'21 2H20 220H- +2H* +H,02+H2 Studies and spin-trapping experiments of Riesz et gave evidence for HO and H as intermediates. From recent results123 of such studies temperatures of 2000-4000 K were estimated for the cavitational collapse hot-spot [from spin trapping with N-(tert-buty1)-a-phenylnitroneof H and D atoms formed in argon-saturated 1:1 H20-D,O mixtures] con-firming the temperatures firstly reported by Suslick et ~21.l~~ Henglein et ~1.'~~demonstrated the trapping of HO* and H.by D2to form HDO and HD respectively. However approxi- mately 80% of the OH and H atoms recombine (under argon and in the absence of scavengers).. The average peroxide formation rate is about 10-50pmol 1-' min-' using a probe system at 20 kHz and applying about 50 W cm-2.121 Irradiation of water in the presence of non-inert gases such as 02,N H CH or mixtures of H,-CO and N,-CO gives a variety of products.12' The radicals produced by the sonolysis of water or the generated H202 can also be trapped by organic compounds dissolved or dispersed in water. Several oxidation hydroxyl- ation and/or decomposition products have been detected (Table 5). However the reaction mixtures are rather complex owing to several secondary reactions especially in the presence of gases.For details refer to the literature (ref. 121 and references therein). In recent years the sonolysis of organic pollutants in water has become a developing field of research in environmental technol- ogy. Amongst the investigated compounds are chlorinated com- pound~,'~~-'~~ phenol^'^^-'^^chlorofluorocarbons (CFCS),'~'-~~~ and pesti~ides.'~~,~~~ The utilisation of ultrasound to convert environmentally hazardous substances into more benign sub- strates or better still to mineralise organics into carbon dioxide has been described recently in some review^.'^^-'^^ Some examples are given in the following paragraphs. Table 5 Products of the aqueous sonolysis of organic compounds' substrate main sonolysis products halogen compounds.Nagata et a1 studied the decomposition of 10 ppm aque- ous solutions of chlorinated hydrocarbons ( l,l,l-trichloro-ethane 1,1,2,2-tetrachloroethene and 1,1,2-trichloroethene) using 200 kHz ultrasound (6 W cm-'). After 10min irradiation the content was lowered to about 2ppm. The decomposition rates under argon are CHCl= CCl zCH,CCl >CCl =CCl In the presence of air the results are comparable but oxidation occurs CHCl=CC1,+2H20 "".2CO+3HC1+H2 CHC1=CC12 +2H20+O2 %2C02+3HC1 The sonolysis of CFC 113 CClF,-CClF has been studied recently Aqueous solutions of 25-1000 ppm Freon 113 were sonicated in a closed vessel under air or argon 1))CClF2-CClF2 +3H2O -1 2C02 +0 8CO +3HCl+ 3HF The decomposition depends on the type of gas as well as the gas/liquid volume ratio eg the concentration of CFC 113 is reduced from 100 ppm to about 55 ppm in a 105 ml vessel having a gas/liquid volume ratio of 45/60 and to about 15 ppm with a 70 4ml vessel and a ratio of 104/60 (time 60 min).The experiments showed that the decomposition is due to high- temperature pyrolysis inside the bubbles and is not caused by OH radicals or by combustion with 0 or air The sonolysis of phenol gives dihydroxybenzenes and quin- ones as primary reaction products which are further degraded with time to yield low molecular mass carboxylic acids 133 134 Ultrasound has also been used in rubber chemistry eg the sonocleavage of styrene-butadiene rubber,14' the degradation of butyl rubber142 and the devulcanization of waste rubbers143 have been reported Earthy-musty odorous compounds (e g geosmin) have become indicators of deterioration of drinking water Geosmin and other odorous compounds with related bicyclic structures have a low odour threshold Consequently its decomposition is of interest for water improvement.Thus the sonochemical degradation of geosmin was studied under different gas atmos- pheres144 It was found that geosmin is readily decomposed within 60 min For a 33 pmol dmP3 solution initial decompo- sition rates in the order Ar >O >air >N were detected Addition of tert-butyl alcohol as a radical scavenger suppressed the decomposition by about 60% indicating that the sonolysis is due to radical-induced (60%) and thermolytic degradation (40yo) The sonolysis of three chlorophenols (2- 3- and 4-chloro- phenol) was examined under pulsed sonolytic conditions (20 kHz 50 W) in air-equilibrated aqueous media z e under oxidation conditions.These phenols were transformed com- pletely to dechlorinated hydroxylated intermediate prod-ucts 126. The sonochemical oxidation technique shows strong similarities with conventional photocatalytic oxidation pro- cesses (direct photolysis flash photolysis UV/peroxide and irradiated semiconductor particulates like Ti02 and ZnO) Ultrasound is also able to assist these conventional methods of waste treatment As a model compound pentachlorophenol in the presence of 02% Ti02 has been investigated Conventional ultraviolet irradiation yields after 50 min 40% decomposition as the final result Combination of ultraviolet and ultrasound irradiation resulted in 60% conversion after 50min and the decomposition is quantitative after 2 h treatment 70 Sonolysis in non-aqueous systems A wide range of organic liquids have been sonicated Suslick et al found first that the sonolysis of higher n-alkanes leads 1614 J Muter Chem 1996 6(10) 1605-1618 to lower alkanes and alkenes It has been shown that the degradation process is similar to high-temperature pyrolysis and involves radical species following the principles of the Rice mechanism.145 On irradiation of diesel fuels the sonolytic reactions are on the one hand similar to the degradation of alkanes including cracking of ClO-C components and radical-induced polym- erisations and on the other hand comparable to processes occurring during long-term storage of fuels involving sediment formation and breaking down of longer alkane chains Sediment analysis gave similar data to storage under ambient temperature and no sonication in terms of molecular mass distribution (1000-10 OOO) nitrogen content and UV and IR spectra Using ultrasound sediment production seems to result from the same processes but the degradation is accelerated Thus 6 h high-intensity sonication produced an amount of sediment which is comparable to a storage at ambient tempera- ture for 16 months 146 Ultrasound has been applied recently as a first processing step in reducing the heteroatom content in upgrading of coal syncrude A syncrude from direct liquefaction of sub-bitumi- nous coal was first deasphaltated by ultrasonic disaggregation in n-hexane removing 39% of N 43% of S and 47% of 0 14' Another interesting application is the sonoisomerisation of alkenes mediated by either halogen radicals'48 151 or metal carbonyls ls2 Carbon-halogen bonds are cleaved easily by ultrasound to give halogenoradicals X..Thus if alkenes are sonicated in the presence of halogenated hydrocarbons the halogenoradical can add reversibly to the double bond via a transition state allowing free rotation Sonoisomerisation are reported for the maleic/fumaric acid148 and ester'49 system trans-dichloro- ethene'" in the presence of CHBr and bromoalkanes respect- ively as well as czs-and trans-vinyl sulfones mediated by CBrC1 15' A wide range of metal carbonyls (M =Cr Mo W Fe Ru Co) catalyse the isomerisation of pent-l-ene to czs-and trans-pent-2-ene 15' Sonication enhances the reaction rate by a factor of about lo5 compared to thermal isomerisation.The used metal carbonyls are sonochemically as effective as in photo- chemical isomerisations except for Ru (CO)' which is more active under sonication Biological materials The effects of ultrasound on biological systems depend on the intensity and frequency applied. Thus the following appli- cations are known non-destructive sonication of biosystems where the cell membrane remains intact and cell rupture (disintegration) to release the contents due to cavitation Sonolysisof biological materials Using high-intensity and/or focused ultrasound and depending on the conditions employed destruction of biological materials occurs cell walls are destroyed and the interior is expelled (ultrasonic extraction see section on Industrial and laboratory applications) into the surrounding bulk solution Further effects and uses are enzyme dea~tivation,'~~ 154 bactericidal treatment and dispersal or destruction of bacteria (eg in view of milk paste~risation,'~~.The combination 156 or food pre~ervation'~~) of ultrasound with other decontamination techniques (heating chlorination or extreme pH) seems to be particularly effec- tive 15' High-intensity ultrasound has also been used to activate antitumour agents'" or to induce hypothermia in living tissues for oncological treatment At this point it should be noted that diagnostic ultrasound used in medicine is far below the energy densities and intensities which are applied under sono- lytic or sonochemical conditions.Thus the cavitation threshold is not reached Non-destructive sonication of biological materials There are several papers on the influence of ultrasound on enzymes microorganisms and living tissues (e.g. ref. 153,161-164). Just three recent papers will be discussed in more detail. An interesting application deals with the increase in the permeability of human skin resulting in the possibility of transdermal drug de1i~ery.l~~ By irradiating the skin with 1 MHz ultrasound the transport of hydrophobic drugs could be enhanced..This is explained by the disorganisation of the membrane bilayers and the resulting formation of transpor- tation channels due to ultrasonically induced air pockets in the keratiocytes. Insulin y-interferon and erythropoietin diffuse at therapeutically useful rates through the skin on exposure to 20kHz ultrasound.. The skin change is reversible as the skin reverted to its impermeable state after switching off the ultra- sound. In uiuo experiments showed that the blood glucose level of normal and diabetic rats can be reduced by transdermal insulin delivery to the same extent as by insulin injection.It should be noted that ultrasound has already been used to alter polymer-membrane permeability to stimulate the release of polymer-coated encapsulated dr~gs'~~?'~~ and to enhance the dialysis separation of electrolytes through a cellophane membrane.16* The influence of sonication conditions on enzymes can be exemplified by a porcine pancreas lipase catalysed hydrolysis which has been published recently153 (Table 6). Performing the reaction in an ultrasonic bath (40 kHz 375 W electrical power input) resulted in a seven-fold increase in the reaction rate and the stereoselectivity remained unaffected compared to a non-sonicated control experiment.. The rate enhancement was explained by the authors to be due to the locally high pressure and/or the increase in catalytic surface.Under ultrasonic probe conditions (20 kHz 600 W electrical power input) and at a higher energy density the rate and stereoselectivities decreased probably due to higher local temperatures causing denaturation. Bioleaching of metals has become of increasing importance. Recently ultrasound-assisted microbial nickel leaching using Aspergillus niger has been re~0rted.l~~ Ultrasound enhanced the leaching rate of Ni which reached 95% after 30min sonication while only 25% is obtained by conventional in situ bioleaching (incubation time 14 d). Increasing the ultrasound intensity leads to a maximum and further intensity increases decrease the leaching rate. Longer sonication decreases the leaching rate slightly to 81% after 60 min.Using different frequencies leaching rates of 95% (20 and 43 kHz) and 86% (720 kHz) are obtained. Furthermore indications of substantial selectivity for nickel over iron were reported (under optimum conditions 95% Ni and 0.16% Fe). Industrial and laboratory applications There are several industrial applications of ultrasound. However ultrasonic plastic welding and sonocleaning are by far the most important uses. Other well established areas are ultrasonic soldering spraying metal welding machining and sonocleaning in the field of metallurgy and materials sciences cell disruption in biological sciences as well as dental scaling and ultrasonic nebulizers in medical therapy (for overviews see e.g. ref. 72,162,169-172).Applications to solids and melts Introduced commercially in 1963 plastic ~eldingl~~,~~~ is now a well established industrial process which is suitable for almost all thermoplastics with low thermal conductivity and melting temperatures (100-200 "C).Using a sonotrode a stand- ing wave is generated with its maximum amplitude at the contact surface of the two components to be welded. Process parameters are the sonotrode shape the contact pressure amplitude/power and irradiation time.. The ability of ultra- sound to propagate through elastic media allows welds at some distance from the ultrasonic horn (far-field processing). It requires rigid materials which are able to transmit vibrations with low attenuation e.g. polystyrene polyoxymethylene or styrene-acrylonitrile polymers.. Thermoplastics with higher mechanical damping need a smaller distance (<6 mm) to the ultrasonic tool (near-field processing)..The welding of polymer films and sheets is a typical application.. Thus PTFE sheets were joined using a 50 kHz flexural-mode transducer system.175 In ultrasonic welding the heat is generated inside the material using internal friction.. Thus the energy is limited to the welding zone resulting in fast welding (welds are typically performed in 0.2-1.5 s) and low part distortion. Other advan- tages are lower welding temperatures resulting in less material degradation and higher yields highly reproducible weld-seam quality high energy efficiency no adhesives or solvents or other additives as well as automatic processing in mass production.Special applications are the embedding of metal inserts in thermoplastics and ultrasonic bonding of synthetic fibres (especially polypropylene).. The latter has the major advantage of substantially lower energy consumption com-pared to thermal body welding. Plastic welding requires of course high power intensities which are in the kW (about 1000 W cmP2) range at 20 kHz. Ultrasonic welding cannot of course replace conventional metal welding. However it is suitable for special appli- cation~.~~,~~~In contrast to plastic welding where ultrasound is usually applied vertically metal welding uses lateral oscillat- ing horns inducing frictional heating between the surfaces.Surface oxides (like on aluminium) and other contaminants are broken up absorbed by the weld and finally the exposed metal surfaces fuse together under pressure. Ultrasonic metal welding is a form of low-temperature diffusion welding. Therefore brittle problems resulting from recrystallisation and the formation of intermetallic compounds are avoided. Ultrasonic metal welding is used for delicate joining of metals such as electrical grade aluminium and copper musical instru- ments or tiny pieces.. Thus it is applied in the semiconductor manufacturing industry for producing miniature semicon-ductor leads and chips as well as microbonding. The localised use of ultrasound at some distance from the horn and the relatively cold conditions allow the machining of hard and brittle materials such as ceramics glasses gem- stones and ferrites.Ultrasonic impact grinding or rotary abras- ive machining are common applications.. The cleaning effect of ultrasonic cavitation on surfaces is also the main principle in fluxless soldering. Ultrasound erodes the oxide layer of molten solder and exposes clean metal to solder.. This is especially applied to aluminium which is attacked by common alkaline or acid fluxes. The termination of aluminium cables and the soldering of mirror frames or heat exchangers are typical applications Other uses are the soldering of hard metals eg nickel Applications to liquids or solutions Ultrasonic cleaning7’ 177 was the first industrial application of ultrasound and has been applied since the early 1950s Commercial ultrasonic cleaners work in a frequency range of 20-60 kHz and with a power of 25-2500 W.They consist of stainless-steel tanks of 2-200 1 capacity and are usually equipped with temperature-controlled heaters However there are also huge tanks of several hundreds or even thousands of litres which are equipped with numerous transducers. The overall power of such systems reaches several kW Typical objects include glassware jewellery lenses spec- tacles medical instruments semiconductors circuit boards engine blocks and other machining parts Practically all mate- rials which are sound-reflecting (glasses ceramics metals plastics) can be used whereas materials such as rubber or textiles are cleaned less efficiently.Thus the ultrasonic cleaning of precision mass standards in ethanol has been reported recently to be the most efficient method for cleaning polished stainless-steel surfaces 178 Sonocleaning in the 0 8-1 0 MHz range (called megasonic cleaning) has been applied successfully to the removal of particles from silicon wafers 179 In ultrasonic cleaning the conventional brush is ‘replaced’ by cavitational bubbles doing the job Cavitation effects (micro- streaming high temperature/pressure jets and shock waves) at or near the surface ‘brush off’ the contaminants dirt or oxide layers which are either dissolved or (if insoluble) dispersed in the solution In combination with the ultrasonic effects there are also external heating detergents and the use of special solvents or water of a given pH value as bath liquid.The advantages of sonocleaning are less cleaning solvents needed (sometimes water can be used) use of parts with complex and non-regular shapes as well as materials having blind holes crevices or inner surfaces easy scale-up for large parts up to several metres and application under clean-room conditions to com- puter equipment Besides ultrasonic cleaning the dispersion of solids or liquids in liquids (suspension and emulsification respectively) and of liquids in air (atomisation) are the most common applications of ultrasound to liquids The preparation of dye pigments insecticides and magnetic oxides are examples of industrial uses in solid/liquid systems Some attention has also been paid to coal dispersions Ultrasound is applied in the food industry to generate fine emulsions Ultrasonic homogenisation is for example used in the production of tomato sauce mayonnaise or yoghurt Stable emulsions of immiscible liquids can be obtained ultrasonically with less or no surfactant Despite the well developed conven- tional surfactant-based methods and cost concerns there is currently growing interest in the food industry 157 166 18’ 184 Many advances have been made in ultrasound food technology (fields of interest mixing blending extraction crystallisation foam destruction particle/aerosol precipitation oxidation pro- cesses influencing enzyme activity sterilisation) in the last decade A more intensive use of ultrasound will depend on the availability of low-cost instrumentation that is shown to have significant advantages over current technologies Furthermore more fundamental research is required on the relationship between ultrasonic treatment (duration intensity etc ) and the effects on the properties of food materials There are also several reports on the ultrasonic extraction of organic compounds (eg soybean protein,lB5 saponin from ginsengIB6) from mainly vegetable or other plant sources.The beneficial effects of ultrasound result in (as already mentioned in the last section about Biological materials) the facile release of plant substances after destroying the epidermis or even the inner cell membranes.The advantages are the easy process the low temperatures the reduced damage to the extracts the short extraction times and the reduced loss of volatile components The application of ultrasound has become a widespread tool in analytical chemistry in the extraction of organic compounds from liquid or solid environmental samples Some recent examples are the extraction of polycyclic aromatic compounds (PAH) from soil suspensions,187 the recovery of herbicide residues from milk by ultrasonically breaking up fat globules,ls8 the removal of elemental sulfur from environmental samples by means of different reagents,lB9 the extraction of pentachloro- phenol in soil wood or water sampledg0 and the solubilization of margarine in hexane for tocopherols analysis 19’ The fine dispersion of liquids in air or other gases is a common technique in medical nebulizers for inhalation ther- apy and in sample injection in mass and atomic emission spectrometry.The spray pyrolysis of solid particle suspensions has already been mentioned Application of ultrasound in the 1-1 5 MHz range to induc- tively coupled plasma-atomic emission spectrometry (ICP-AES) can enhance the detection limits by a factor of 10 The efficiency of nebulization is so high that the solvent loading of the aerosol needs to be reduced by thermal desolv- ation to prevent cooling of the plasma 192 Ultrasonically assisted electrospray spectrometry (ESI-MS) has been applied successfully in LC-MS coupled analysis of proteins 193 Conventional ESI-MS has some severe limi- tations as the interface after liquid chromatography LC mobile phases which have high flow rates (>5 yl min-I) conductivities or surface tensions are normally not applicable eg the required mobile phase gradient MeOH-H,O in nucleoside separations does not meet this criteria By the application of an ultrasonic nebulizer LC-MS determinations of proteins over a wide range of methanol-water mobile phase conditions (0-100%) with different flow rates could be performed easily For a review on applications of ultrasonics in analytical chemistry see ref 194,195 I would like to thank Prof Dr Ralf Miethchen (Rostock) for numerous critical discussions and helpful comments in the preparation of the manuscript.The BMBF (project 03D0018) and the DECHEMA are acknowledged for financial support This work was effected under the auspices of the EEC COST organisation (Programme Chemistry D6) References 1 W T Richards and A L Loomis J Am Chem SOC 1927 49 3086 2 C Einhorn J Einhorn and J L Luche Synthesis 1989,787 3 Current Trends in Sonochemistry ed G J Price,. The Royal Society of Chemistry Cambridge 1992 4 T J Mason Practical Sonochemistry Ellis Horwood Chichester 1991 5 Advances in Sonochemistry ed T Mason JAI Press London VO~1-4,1991-1994 6 K S Suslick MRS Bull 1995,20 29 7 T J Mason Chemistry with Ultrasound Elsevier London 1990 8 Ultrasound-Its Chemical Physical and Biological Eflects ed K S Suslick VCH Wemheim 1988 9 F R Young Cavitation McGraw-Hill London 1989 10 T J Leighton,. The Acoustic Bubble Academic New York 1994 11 K S Suslick P F Schubert and J W Goodale J Am Chem Soc 1981 103 7342 K S Suslick S -B Choe A A Chichowlas and M W Grinstaff Nature (London) 1991,353,414 12 X Cao Y Koltypin G Kataby R Prozorov and A Gedanken J Mater Res 1995,10,2952 13 P Diadoti L Mirri C Petrillo and F Sacchetti Solid State Commun 1994,92,433 1616 J Muter Chem 1996 6(10) 1605-1618 14 T J Mason J P Walton and D J Lorimer Ultrasonics 1990 61 A J Fry G S Ginsburg and R A Parente J Chem SOC Chem 28,333 Commun ,1978,1040 15 T J Mason J P Walton and D J Lonmer Ultrasonics 1990 62 K Burger and B Helmreich Chem Ztg ,1991,115,253 .28,251 63 M S F Lie Ken and W L K Lam J Chem Soc Chem 16 T J Mason Practical Sonoelectrochemistry Ellis Horwood Commun ,1987,1460 Chichester 1991 64 N K Goh A T Teoh and L S Chia Ultrason Sonochem 1994 17 S M Kochergin and G Y Vyaseleva Electrodeposition of Metals in Ultrasonic Fields Consultants Bureau New York 1966 65 1 S41 D P.Thompson and P Boudjouk J Org Chem ,1988,53,2109 18 19 A Durant J L Delplancke R Winand and J Reisse Tetrahedron Lett 1995 36 4257 J Reisse Homogeneous sono- chemistry From physical studies to nanopowder preparation pre-sented at the DECHEMA Annual Meeting 1996,Wiesbaden R R Chandlerhenderson J L Coffer and L A Filessesler 66 67 68 P D Lickiss and R Lucas J Inorg Organomet Polym 1995 5,247 T Ando T Kawate J Ichihara and T Hanafusa Chem Lett 1984,725 A Miller C Zur D Peters and R Miethchen J Fluorine Chem J Electrochem Soc 1994,141 L166 in press 20 M P Drake Trans Inst Met Finish 1980,58,67 69 S Ring G Weber B Erhart P Mollmann and S Schwarz 21 22 23 A Henglein Ultrason Sonochem 1995,2 S115 C C T Alex N K Goh and L S Chia J Chem Soc Chem Commun ,1995,201 L Wen-Chou A N Maltsev and N I Kobosev Russ J Phys Chem 1964,38,41 70 Ultrasound in steroid chemistry Applications in complex hormone synthesis presented at the DECHEMA Annual Meeting 1996 Wiesbaden T J Mason Sonochemistry Current Trends and Future Prospects in Current Trends in Sonochemistry ed G J Price,.The 24 25 26 27 A N Maltsev Russ J Phys Chem ,1976,50,995 E Kowalska and M Dziegelewska Ultrasonics 1976 14 73 B N Tyutyunnikov and I T Novitskaya Maslob Zhir Prom 1959,25,13 K S Suslick and D J Casadonte J Am Chem SOC 1987 109 3459 71 72 73 Royal Society of Chemistry Cambndge 1992 R Snnivasan I Z Shirgaonkar and A B Pandit Sep Sci Technol 1995,30,2239 L D Rosenberg Physical Principles of Ultrasonic Technology Plenum New York 1973 G N Kozhemyakin and L G Kolodyazhnaya J Cryst Growth 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 E A Cioffi W S Willis and L S Svit Langmuir 1988,4 692 K S Suslick Solid State Ionics 1989,32/3,447 K S Suslick Science 1990,247 1067 K S Suslick,T Hyeon M M FangandA A Cichowlas Muter Sci Eng A 1995,204,186 T J Trentler R Suryanarayanan S M L Sastry and W E Buhro Muter Sci Eng 1995,204 193 T Popov D G Kissvrski K Ivanov and J Pesheva Stud Surf Sci Catal 1987,31 191 R Ranganathan I Mathur N N Bakhshi and J F Mathews Ind Eng Chem Prod Res Dev 1973,12,155 V I Shekhobalova and L V Voronova Vestn Mosk Univ Ser Khim 1986,27,327 C L Bianchi R Carli S Lanzani D Lorenzetti G Vergani and V Ragaini Ultrason Sonochem ,1994,1 S47 P Boudjouk Heterogeneous Sonochemistry in Ultrasound-Its Chemical Physical and Biological Efects ed K S Suslick VCH Weinheim 1988 T J Mason and J P Lonmer Sonochemistry.Theory Applications and Uses of Ultrasound in Chemistry Ellis Horwood Chichester 1988 T Ando and T Kimura Adv Sonochem ,1991,2,211 S V Ley and C M R Low Ultrasound in Synthesis Springer Berlin 1989 H Bonnemann B Bogdanovic R Brinkmann D W He and B Spliethoff,Angew Chem ,Int Ed Engl 1983,22,728 J L Luche C Petner and C Dupuy Tetrahedron Lett 1984 25,753 A J Fry and D Herr Tetrahedron Lett 1978,19,1721 J Einhorn and J L Luche J Org Chem ,1987,52,4124 C M R Low Ultrason Sonochem ,1995,2 S153 L Xu F Tao and T Yu Tetrahedron Lett 1985,26,4231 G Hugel D Cartier and J Levy Tetrahedron Lett 1989 30 4513 T S Chou and M L You J Org Chem 1987,52,2224 J L Luche C Petrier and C Dupuy Tetrahedron Lett 1984 25,753 S E de Laszlo S V Ley and R A Porter J Chem SOC Chem Commun 1986,344 H C Brown and U S Racherla Tetrahedron Lett 1985 26 4311 N Ishikara M G Koh T Kitazume and S K Choi J Fluorine Chem 1984,24,419 Y T Lin J Organomet Chem ,1986,317,277 R W Fitch and F A Luzzio Tetrahedron Lett 1994,35,6013 W H Binder R H Prenner and W Schmid Tetrahedron 1994 50,749 C Einhorn and J L Luche J Organomet Chem ,1987,322,177 S Tsunoi I Ryu H Fukushima M Tanaka M Komatsu and N Sonoda Synlett 1995 1249 M S A Parker and C J Rizzo Synth Commun 1995,25,2781 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 1995,147,200 Y Ding Y Miura and A Osaka J Am Ceram SOC 1994 77 2905 Y Ding A Osaka and Y Miura J Am Ceram SOC 1994 77 749 H L Choi N Enomoto and Z E Nakagawa J Muter Sci 1994,29,3239 J Nyvlt and S Zacek Cryst Res Technol 1995,30,1055 G I Eskin Ultrason Sonochem ,1995,2 S137 V A Versteeg C T Avedisian and R Raj J Am Ceram SOC 1995,78,2763 H Ichinose H Katsuki and M Nagano J Cryst Growth 1994 144,59 A Furusaki H Konno and R Furuichi J Muter Sci 1995 30,2829 K H Yoon and D J Nam J Muter Sci 1995,30,3415 I T Kim C H Lee and S J Park Jpn J Appl Phys 1994 33,5125 Y Harima Y D Wang K Matsumoto and K Yamashita J Chem SOC Chem Commun 1994,2553 Y C Kang S B Park andY W Kang Nanostruct Muter 1995 5,777 J M Ting J Muter Sci 1995,30,4049 J M Ting J Muter Sci 1995,30,4095 K Okuyama T Seto M Shimada and N Tohge J Muter Sci Muter Electron 1994,5,210 J H Lee and S J Park J Muter Sci Mater Electron 1993 4,254 H B Kim J H Lee and S J Park J Muter Sci Muter Electron 1995,6,84 N Tohge S Tamaki and K Okuyama Jpn J Appl Phys 1995 34 L207 S S Ostapenko L Jastrzebski J Lagowski and B Sopori Appl Phys Lett 1994,65,1555 S S Ostapenko and R E Bell J Appl Phys 1995,77,5458 R M Lago S C Tsang K L Lu Y K Chen and M L H Green J Chem Soc Chem Commun 1995,1355 L Q Ma F Chen and G J Shu J Muter Sci Lett 1995,14,649 T J Mason Sonochemistry Cavitation in the Service of Science and Technology World Congress on Ultrasonics Berlin 1995 P Kathirgamanathan J Muter Sci Lett 1993 12 1810 S A Yeung R Hobson S Briggs and F Grieser J Chem SOC Chem Commun ,1993,378 Y Nagata Y Watanabe S Fujita T Dohmaru and S Taniguchi J Chem SOC Chem Commun ,1992,1620 R A Hobson P Mulvaney and F Grieser J Chem Soc Chem Commun ,1994,823 Y Fang,D K Agrawa1,D M RoyandR Roy Am Ceram SOC Bull 1995,74,70 L Bjoorno S Gram and P R Steenstrup Ultrasonics 1978 103 M Takeuchi and K Yamanouchi Jpn J Appl Phys 1994 33 3045 59 R S Bates and A H Wright J Chem SOC Chem Commun 1990,1129 103 K Yasuda S Umemura and K Takeda Jpn J Appl Phys ,1995 34,27 15 60 A Tai T Kikukawa Y Sugimura T Inone T Osawa and 104 T Iwamoto H Shimada S Shimomura M Onodera and J Fugii J Chem SOC Chem Commun 1991,795 T Ohmi Jpn J Appl Phys ,1994,33,491 105 Y Y Zhao C G Bao R Feng and Z H Chen Ultrason 149 T Lepoint F Mullie N Voglet D H Yang J Vandercammen .Sonochem ,1995,2 S99 and J Reisse Tetrahedron Lett 1992,33 1055 106 C T Walker and R Walker J Electrochem SOC ,1974,124,661 150 T P Caulier M Maeck and J Reisse J Org Chem 1995 60 107 M C Hsiao and C C Wan Plat Surf Finish 1989,76,46 272 108 A Chiba and W C Wu Plat Surf Finish 1992,79,62 151 D Peters F Pautet H El Fakih H Fillion and J L Luche 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 Shinko Electnc Jpn Kokai Tokkyo Koho JP 5970756,1984 R Walker and C T Walker Nature (London) 1974,250,410 W J Tomlinson and S J Matthews J Muter Sci ,1994,29,1101 G J Price Ultrasonically Assisted Polymer Synthesis in Current Trends in Sonochemistry ed G J Pnce,.The Royal Society of Chemistry Cambridge 1992 G J Price Adu Sonochem ,1990,1,231 A M Basedow and K Ebert Adu Polym Sci ,1977,22,83 R A Pethrick Adu Sonochem 1991,2,65 G J Price P J West and P F Smith Ultrason Sonochem ,1994 1 S51 P A R Glynn B M E van der Hoff and P M Reilly J Macromol Sci A 1973,6 1653,7 1695 B M E van der Hoff and P A R Glynn J Macromol Sci A 1972,6,1653 S L Malhotra and J M Gauthier J Macromol Sci Chem A 1982,18,783 F 0 Schmitt C H Johnson and R A Olson J Am Chem SOC 1929,51,370 K S Suslick Homogeneous sonochemistry in Ultrasound-Its Chemical Physical and Biological Efects ed K S Suslick VCH Weinheim 1988 P Riesz D Berdahl and C L Chnstman Environ Health Perspect ,1985,64,233 V Misik N Miyoshi and P Riesz J Phys Chem ,1995,99,3605 K S Suslick D A Hammerton and R E Cline J Am Chem SOC,1986,108,5641 C H Fischer E J Hart and A Henglein J Phys Chem 1986 90,222,1954,3059 N Serpone R Terzian H Hidaka and E Pelizzetti J Phys Chem 1994,98,2634 K Inazu Y Nagata and Y Maeda Chem Lett 1993,57 H M Cheung M Bhatnagar and G Lansen Enuiron Sci Technol 1991,25,1510 Proc Pacifichem Conference '95 Honolulu 1995 Y Nagata K Hirai K Okitsu T Dohmaru and Y Maeda 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 J Prakt Chem ,1995,337,363 K S Suslick J W Goodale P F Schubert and H H Wang J Am Chem SOC 1983,105,5781 G L Lin and H C Liu Tetrahedron Lett 1995,36,6067 R T Roberts Chem Ind 1993,15,119 S E Jacobs and M J.Thornley J Appl Bacteriol 1954,17,383 G R Rubleson T M Murray and M Pollard Appl Microbiol 1975,350 G W Gould New Methods in Food Preservation Chapman and Hall London 1994 H S Lillard Food Technol 1994,48,72 S Umemura Jpn J Appl Phys ,1994,33,2846 N N Amso Ultrason Sonochem 1994,1 S69 K M Swamy L B Sukla R N Narayana R N Kar and V V Panchanadikar Acoust Lett 1993,17,45 J V Sinisteria Ultrasonics 1992,30 180 G L Lin H C Liu and S H Liu J Chin Chem SOC 1995 42,957 K M Swamy L B Sukla R N Narayana R N Kar and V V Panchanadikar Ultrason Sonochem 1995,2 S5 S Mitragotri D Blankschtein and R Langer Science 1995 269 850 S Miyazaki C Yokouchi and M Takada J Pharm Pharmacol 1988,40,716 J Kost K Leong and R Langer J Macromol Sci Macromol Chem A 1985 22 455 J Kost and R Langer Trends Biotechnol ,1992,10,127 H Li E Ohdaira and M Ide Jpn J Appl Phys ,1995,34,2725 B Brown and J E Goodmann High Intensity Ultrasonic Industrial Applications Van Nostrand Princeton 1965 A Shoh Industrial Applications of Ultrasound in Ultrasound-Its Chemical Physical and Biological Eflects ed K S Suslick VCH Weinhem 1988 P I Polyukhin,.The Use of Ultrasonics in Extractiue Metallurgy Technicopy Stonehouse 1978 C Elbaum J Alloys Compd 1994,212,20 A Shoh Ultrasonics 1976,14,209 Chem Lett 1995,203 174 A Puskar Use of High Intensity Ultrasound Elsevier 131 H M Cheung and S Kurup Enuiron Sci Technol 1994 28 Amsterdam 1982 1619 175 Y Watanabe Y Tsuda and E Mori Jpn J Appl Phys 1995 132 E A Stolyarov N G Orlova and L V Vedernikova Rub 34,2730 Termodin Kinet Khim Protsessov 1974,1147 176 A Hulst Ultrasonics 1972 10,256 .133 J Berlan F Trabelsi H Delmas A M Wilhelm and 177 R Pohlmann B Werden and R Marziniak Ultrasonics 1972 134 135 136 137 138 139 140 141 142 143 144 145 J F Petrignani Ultrason Sonochem ,1994,1 S97 C Petrier M F Lamy A Francony A Benahcene B David V Renaudin and N Gondrexon J Phys Chem ,1994,98,10 514 C Petrier M Micolle and G Reverdy Enuiron Sci Technol 1992,26,1639 A Kotronarou G Mills and M R Hoffmann J Phys Chem 1991,95,3630 Proc 4th Meeting Eur SOC Sonochem 1994 Blankenberge Belgien P Colarusso and N Serpone Res Chem Intermed 1996,22,61 T J Mason and S S Phull Sonochemistry in waste minimization in Chemical waste minimization ed J H Clark Blackie Glasgow 1995 Y Maeda Sonolytic decomposition of harmful chemical substances 5th Meeting of the European Society of Sonochemistry Cambndge 1996 R S Mauler F G Martins and D Samios Polym Bull 1995 35,151 V Dubey S K Pandey and N B S N Rao J Anal Appl Pyrolysis 1995,34 11 1 A I Isayev J Chen and A Tukachinsky Rubber Chem Technol 1995,68,267 Y Yoo N Takenaka H Bandow Y Nagata and Y Maeda Chem Lett 1995,961 K S Suslick J J Gawlenowski P F Schubert and H H Wang 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 10,156 R Schwartz and M Glaser Meas Sci Technology 1994,5,1429 A A Busnaina I I Kashkoush and G W Gale J Electrochem SOC,1995,142,2812 W Kowalski and E Kowalska Ultrasonics 1978 16 84 T J Mason Applications of Sonochemistry in Food Technology Pacifichem Conference '95 Honolulu 1995 Proc Ultrasound in Food Processing 25/26 09 1996 Leeds to be published D J McClements Trends Food Sci Technol 1995,6,293 K J Moulton J Am Oil Chem SOC 1989,66,896 H Fukase E Ohdaira N Masuzawa and M Ide Jpn J Appl Phys ,1994,33,3088 H Li E Ohdaira and M Ide Jpn J Appl Phys 1994,33,3085 U Hechler J Fischer and S Plagemann Fresenius J Anal Chem ,1995,351,591 A Lagana G Fago A Marion and B Pardomartinez Chromatographia 1994,38,88 J T Anderson and U Holwitt Fresenius Z Anal Chem 1994 350,474 A Besner R Gilbert P Tetreault L Lepine and J F Archambault Anal Chem ,1995,67,442 G Micah F Lanuzza and P Curro HRC 1993,16,536 T S Conver and J A Koropchak Spectrochim Acta Part B 1995,50,341 J F Banks S Shen C M Whitehouse and J B Fenn Anal J Phys Chem 1983,87,2299 Chem ,1994,66,406 146 G J Pnce and M McCollom Ultrason Sonochem ,1995,2 S67 194 F A Chmilenko A N Baklanov L P Sidorova and Fuel 1995,74,1394 Y M Piskun J Anal Chem 1994,49,494 147 I Fernadez M T Martinez A Benito and J L Miranda Fuel 1995,74,32 195 P Linares F Lazaro M D Luque de Castro and M Valcarcel J Autom Chem 1988,10,88 148 I E Elpiner A V Sokolskayer and M A Margulis Nature (London) 1965,208,945 Paper 6/02549H Received 1lth April 1996