Characterization of Nanocrystalline Oxide Powders Prepared by C0 2 Laser Evaporation t

Today, a world-wide interest exists in producing nanosized ceramic powders. One of the suitable techniques is based on the evaporation of solid primary materials by lasers. Although this technique has been known for nearly two decades, the literature has hitherto only rarely reported effective powder yields (a few grams per hour). We produced nanocrystalline zirconia and alumina powders by evaporation of oxides in the focus of a C0 2 laser (transversal flow of the C0 2 gas; PL = 0.75 ... 4 kW) and by recondensation of the oxides in a carrier gas stream (air, oxygen, argon). Most experiments were carried out in the continuous wave (cw) mode, but earlier ones were also done by means of a pulsed laser (pw). The powder yield depends strongly on the laser power, on the focusing and on the velocity of the moving oxide target. Maximum yields of more than 100 g h- 1 were attained. The zirconia and alumina powders so prepared consisted of nearly spherical particles with diameters in the range of 10 to 100 nm. The particle distribution can be controlled by the para meters of the formation process. Features of the crystal phases and of the chemical composition of the produced powders are reported.


Introduction
Nowadays, there is a world-wide trend in the development of technologies for producing nanosized ceramic powders. Considering only the 3rd Conference of the European Ceramic Society (Sept. 1993, Madrid, Spain), we find eight contributions [la-g] concerning the preparation and special properties of such powders. There are, however, slight differences in the interpretation of the term ''nanosized' '.
The potential advantages and applications of such nanosized powders result from the extremely small dimensions of the particles and thus the extremely high surface. The densification rate d{!ldt and the strain rate d8/dt, depend very strongly on the grain size d during a ceramic sintering process. Assuming a relationship of the form d·P with p in the range of 2 to 4, it is to be supposed that d(!ldt and d8/dt will increase by factors of the order of 10 8 and 10 6 , respectively, if the grain size diminishes from 1 ftill to 10 nm [2].
This means that the sinter temperatures can decrease significantly. In this way, a co-firing of such ceramic materials with a second component of a lower heat stability such as reinforcement fibres may become possible. In addition, such nanocrystalline ceramics can possess unusual properties such as superplasticity or machinability [3]. A stabilization of metastable crystalline phases may be another consequence of the extremely small particle sizes [4] which can also lead to unusual ceramic properties.
Today, a lot of different techniques for producing manosized powders are under development. In addition to routes in the fluid phase such as sol-gel [1a], precipitation [5], hydrothermal [6] processes, gas-phase processes are tested or optimized for the prepartion of such powders. One type of such gas-phase technique which is similar to the process discussed here is the so-called (inert) gas condensation technique. By using usually UHV equipment, solid materials are evaporated in an inert or reactive atmosphere originally by means of heat [2,7], today also by sputtering with noble gas ions [8] or by an electron beam [9,10]. The vapour recondenses in the gas phase and deposits on cooled surfaces. In order to produce nanosized oxide particles in this way, frequently the metal or one of its suboxides is used as starting materials for the evaporation because of their higher vapour pressure in comparison with the stoichiometric oxide. However, in this case, the powder has to be reoxidized which can lead to coarsening or agglomeration of the nanosized particles. In principle, extremely fine particles with a maximum size of 10 nm are achievable in this way but the powder production rates are typically confined to a few hundred milligrams per hour.
Other promising gas-phase techniques for producing nanosized powders are plasma-chemical methods with different kinds of plasma excitation such as a d.c.arc [lc] or microwaves [1b]. In this case, gaseous primary materials are frequently needed. The production rates are significantly higher (several grams to several hundred grams per hour) than in the case of the gas condensation technique, but the size distribution functions become distinctly wider and extend to the range higher than 100 nm.
At present, many groups use the laser-assisted chemical vapour precipitation (LCVP) technique to prepare nanosized powders [1d-f, [11][12][13][14]. Although the same energy source, a laser, is used as in the technique described in this paper, there are important differences. In the former case, the laser is used to activate a chemical gas-phase reaction. Therefore, expensive gaseous primary materials are generally needed. Predominantly, the technique is used for producing non-oxide powders such as silicon carbide, silicon nitride or mixtures thereof. Very fine particles with diameters in the range of a few nanometers to a few tens of nanometers are obtained, but often the powders are inhomogeneous in their atomic structure and composition.
The most fully developed technique to produce nanosized powders in the gas phase is that of pyrolysis of metal halides in an oxyhydrogen flame. The method became known as the AEROSIL ® technique [15], since the most important nanosized material prepared by this technique is highly dispersed silica, which today is used industrially on a large scale for producing paintings and lacquers, reinforced polymers and rubbers, cosmetics and pharmaceuticals and catalysts. The technique is also used today for producing nano-sized alumina, titania, and zirconia.
Many of the particles are connected by interparticular solid bridges due to the heat treatment of the powders necessary to remove the HCl which is formed during the reaction and which is adsorbed onto the particle surfaces. Also, most of these powders show a very strong tendency to agglomerate due to the high content of surface hydroxyl groups which result from the preparation conditions. Although these powders 80 are cheap and easily available, there is little application of such powders in the field of advanced ceramics.
Within the context of all these gas-phase techniques, we want to discuss the advantages and disadvantages of the method of laser evaporation which we apply for producing nanocrystalline oxide powders. The method is distinguished by the possibility of using a solid material such as coarse powders, sintered ceramics or ceramic waste as the starting material. This material is evaporated in the focus of a laser beam, and the vapour recondenses in the stream of an inert (or reactive) carrier gas. This method was first described two decades ago by Kato [16], and in the meantime, Mordike and co-workers [17][18] have been engaged in this field. They applied this method for producing nanosized oxide particles by evaporating not only intrinsic oxides in an inert atmosphere, but also metals in an oxidizing atmosphere. To date, the advantages of the method, the possibility of producing extremely pure nanosized powders with size distributions which are controllable by the parameters of the evaporation process, were outweighed by the disadvantage of very low powder production rates. Mordike et al. applied Nd: YAG laser (400 W average power) and achieved rates of 3 to 4 g h-1 as best results for alumina and zirconia.
The aim of this research is to show how this rate can be increased by investigating correlations between laser parameters, the powder yield, the energetic efficiency as well as the particle characteristics. Our researches are not yet finished, and our paper has the character of a status report.

The experimental set-up
In order to achieve high vaporization rates for oxide materials we chose a C0 2 laser, because the wavelength of the emitted laser beam (,\ = 10.6 J.lm) is absorbed by oxides to a very high degree (in comparison with that of aNd: YAG laser). For this purpose, a transverse flow C0 2 laser GTL 480 [19] was available. It is characterized in the cw operation mode by the following parameters: • laser output power: 1 .. .4kW • diameter of the laser beam: 40 mm • electro-optical efficiency: 11 o/o • total electrical supply 70 kW (for resistance stabilized discharge) Due to the resonator configuration, the laser can be operated using a so-called Q-switch (quality switch) by applying a high-speed chopper disk which is mount-ed in the focal plane of an intracavity telescope [20]. The pulses have typical lengths of r""' 120 ns FWHM (full width at half maximum) and power peaks of nearly 100 kW at repetition rates of a few kHz. The averaged power can be varied by up to 1 kW as a function of the pulse tail.
The majority of the experiments were carried out in the cw mode, but early results using the pw mode conditions will also be reported. Figure 1 shows a schematic of the evaporation chamber which works usually with normal air, but oxygen or argon were also used. The laser beam enters through the opening E and is focused onto the surface of the oxide sample in the sample holder D. As oxide samples, we used coarse powders of different particle sizes (d 50 > 1plll) as well as sintered oxide ceramics of zirconia and alumina. In most of the experiments, we applied a coarse powder of unstabilized zirconia as primary material. During evaporation, the laser beam has to be moved on the sample surface in a well-defined manner in KONA No. 13 (1995) order to avoid the formation of deep craters and to ensure reproducible evaporation conditions. The movement is a superposition of translation and rotation and is caused by a controllable motor Mo.
The oxide vapour recondenses in the atmosphere above the sample. Due to the very narrow region at high temperature which is created by the laser, the nuclei migrate quickly to cooler regions of the reactor and the particle growth is stopped. The particles are transported by a carrier gas stream G which is caused by a pump at the end of the apparatus for all experiments in air. Whenever we used argon or oxygen atmospheres, the carrier gas stream is defined by the flow rate of the gases that stream in. In those cases, we used a modified evaporation chamber with a window for the incident laser beam and an additional gas inlet. The particles are collected by depositing on collision tins in a horizontal glass tube of about 2 m in length.

The powder yield
When applying this technique, the vaporization rates reported in the literature [18] are in the order of only a few grams per hour. Therefore, we investigated the process parameters which significantly influence the vaporization rate and the powder yield, respectively, with the intention of increasing them.
The vaporization rate depends strongly on the position of the focal plane of the laser relative to the oxide target surface. We obtained the optimum energy input and thus the highest powder yield when the focal plane coincided with the surface of the oxide sample. The laser beam was focused by means of a spherical mirror of focal length f = 150 em which created a focussed area of A F = 0.0078 cm 2 .
The dependence of the vaporization rate, dm!dt, of unstabilized coarse zirconia powder on the laser intensity, I, operating in the cw mode is shown is observed in the range of intensity from about 6-10 4 W cm· 2 (dmldt = 4 g h-1 ) to 4.5·10 5 W cm-2 (dmldt = 83 g h-1 ). The necessary energy coupling may be explained by multiple reflection in the vapour channel and by the formation of a laser-induced plasma in the channel [21]. At an intensity of 4.5-10 5 W cm-2 , there is no indication of a limit to the vaporization rate. The experimental feasibilities were limited by the maximum power of the available laser.
In the experiments described so far, the oxide target was rotated relative to the laser beam at about 0.75 r.p.m. This must be mentioned because it was noticed that significant changes in the yield of powder can be achieved by varying this target velocity. However, the correlation is complex because the yield is determined also by the laser intensity. This complex behaviour is illustrated in Figure 3. It shows the dependence of the rate of vaporization of zirconia on the median linear velocity of the oxide target, relative to the laser beam for three different laser intensities. Only in the case of the highest intensity ( 4.2 ·10 5 W cm-2) can an increase of dmldt be observed for velocities higher than 20 em s· 1 . This condition gave the best powder yield of about 130 g zirconia powder per hour at a velocity of 28 em s-1 . This powder yield value is higher by a factor of about 40 than the corresponding one previously reported in [18]. In this way, we produced amounts of 0.5 kg of zirconia powders from selected samples.

The energetic efficiency
By using the powder yield value of 130 g h-1 , we can estimate the energy which is necessary to produce 1 g of Zr0 2 ; if we take into consideration the 82 applied intensity I = 4.2-10 5 W cm-2 as well as the focal area Ap = 0.0078 cm 2 • The laser power factor of PL was 3.276 kW. It follows that 3.276 kWh of energy is necessary to produce 130 g of ultrafine zirconia powder or, in other words, an efficiency of about 40 g Zr0 2 /kWh was achieved. This value is higher by a factor of 2.5 than the corresponding reported value [18]. Moreover, in comparison with the efficiency factors of other gas-phase processes for producing nanosized oxides [10], this result is one of the best and is exceeded significantly only by the flame pyrolysis technique ( > 100 g/kWh).
A comparable value of the efficiency can be estimated from data which is reported [lc] for the synthesis of nanosized alumina/silica powders by a plasma chemical route. A plasma power of 104 kW was used to convert 5 kg of oxide per hour which corresponds to an efficiency factor of 48 g h-1 .
The calculated energy necessary to prepare 1 g of nanosized zirconia (2.5·10-2 kWh g-1 = 90 kJ g· 1 ) taking into account that (see 2 .1.) the mean electrooptical efficiency of a C0 2 laser is 11 o/o, seems to be 2 to 3 orders of magnitude higher than that which is necessary to prepare powders by mechanical grinding. However, it is well known that the mass specific energy for grinding depends strongly on the particle size achieved. Usually, energy factors in the order of 1 kJ g-1 correspond to grain sizes of a few micrometers.
Weichert [22], for instance, investigated the dependencies of the mean specific breakage energy on the particle size. His results for glass spheres of sizes in the range from 10 J.tm to 10 mm, can be extrapolated to fictitious particles of 10 mm. This dependence is characterized by a straight line a double logarithmic representation. Thus a fictitious mass specific energy of about 1000 kJ g· 1 is obtained for preparing an oxide powder with a mean particle size of 10 nm by means of grinding. Although such a powder cannot be prepared in this way, this extrapolation suggests that the mass specific energies needed for laser evaporation are not unacceptably high for the particle size obtained.

The particle size and shape
Zirconia particles of typical size and shape produced by laser evaporation are shown in Figure 4. The micrograph was taken by TEM (Philips CM 30). The powder was prepared with a laser power of 3.0 kW and a focussed area of 0.0078 cm 2 • The carrier gas was air and the sample rotated at a speed of 0.75 r.p.m. The vaporization rate achieved was 65 g h-1 .
It is seen that the powder consists of particles in the range from 10 to 100 nm. The particles are almost spherical in shape which is typical of powders from gas-phase processes. This zirconia sample has a specific surface area determined by BET, SBET of 18.9 m 2 g-1 . The cumulative size distribution by volume Q 3 is distinguished by d 10 = 32 nm, d 50 =58 nm and d 90 = 111 nm. The particle size distribution functions were determined by measuring the sizes of at least 1000 particles from electron micrographs. For example, Figure 5 shows the cumulative size distributions by number (Q 0 ) and by volume (Q 3 ) for another sample of unstabilized zirconia (PL = 2.7 kW, AF = 0.0078 cm 2 , SBET = 16 m 2 g-1 , dm/dt = 44 g h-1 ). From this figure, a value of d 50 (Q) = 66 nm results. There are vanous parameters of the evaporation process that influence the particle size distribution. We investigated the following ones: • the laser power, P L • the area of the laser focus, AF • the speed of the target relative to the laser beam • the velocity of the carrier gas stream • mode of operation of the laser • the pulse length and the repetition rate in the case of the pw mode. Figure 6 illustrates the influence of PL and AF on the surface area of the powders produced. It is obvious that the dependence on the power at a fixed focal area is only weak. Conversely, there is a significant dependence on the focal area at constant power. We interpret this behaviour in the following manner: If the laser power increases on a fixed focal area, the amount of evaporated oxide, i.e. the oxide vapour pressure above the sample surface, will increase. Simultaneously, the temperature of that vapour will also increase. Nucleation of new oxide particles (or droplets in the first state of their development) from the vapour requires a supersaturated vapour. However, since the supersaturation of a vapour is proportional to the oxide vapour pressure but decreases if the temperature increases, the degree of supersaturation must not be allowed to change significantly by increasing the laser power at a fixed focus area and, in this way, the conditions for nucleation and particle growth may remain comparable independently of PL.
A change of the focal area at constant power changes P L, the volume of the intrinsic process zone andwith that, the supersaturation as well as the temperature gradient of this zone-is changed. A decrease of the focal area at a fixed laser power causes an increase of the supersaturation and consequently of the nucleation rate. However, the process zone is  Fig. 6 The specific surface area SBET of zirconia as a function of the area of the laser focus, A F for two laser P L smaller so that the growth time of the particles is reduced. Consequently, a larger quantity by smaller particles develop. Another essential factor which influences the particle size distribution is the mode of operation of the laser. The experiments reported so far utilised the cw mode. Figure 7 illustrates the influence of the operating mode on the cumulative distributions, Q 0 , of the particle size of unstabilized zirconia. The laser intensity used for preparing sample 30 under cw conditions (1.3·10 5 W cm-2 ) was almost the same as the averaged intensity applied in the pw mode for producing sample 33A, (pulse lengths: 120 ns; repetition rate: 3.5 kHz; peak power about 100 kW). Significant differences in the size distribution of these samples can be seen. The d 50 values are 20 nm and 13.5 nm respectively, and the d 90 values are 47 nm and 32 nm. The surface area values are 19 m 2 g-1 for sample 30 and 30 m 2 g· 1 for sample 33A. Obviously, the growth time of the particles is curtailed or interrupted by the pulse mode so that smaller particles form.
We have also obtained comparatively high values of surface area of 30 m 2 g· 1 and more in the case of the cw mode provided that we also used a higher velocity (about 5 m s· 1 ) of the carrier gas stream. These experiments are still in progress and a further increase of the surface area is expected by combining the pw mode of operation with gas streams of higher velocity.
First attempts to convert these ultrafine powders into sintered ceramics were successful, and we observed a significant decrease in the required sinter temperature. It can be shown by dilatometry that the densification process is complete at a temperature of approx. 1200°C, whereas 1500°C and more is usually needed to sinter most of the commercially available zirconia powders.
Similar particle sizes were obtained by evaporation of mixtures of zirconia and alumina and of pure alumina, respectively, at process conditions used for zirconia. Figure 8 shows the number density distribution of a powder we produced from the evaporation of transformation-toughened ceramics consisting of alumina with 6 weight o/o Zr0 2 and 0.375 weight o/o Y 2 0 3 (I= 2.6·105 W cm-2 ). By comparing the broadening of the BRAGG peaks of alumina as well as of zirconia, we conclude that there are no significant differences in the particle sizes of either component. This mixed powder of alumina and stabilized zirconia is characterized by a specific surface area SBET = 44 m2 g· 1 , whereas pure alumina powders produced by laser evaporation show surface areas in the range of 50 to 60m 2 g· 1 .

Analysis of the crystal phases and the chemical composition of the powders
The nanosized powders prepared were studied by means of X-ray diffraction. In each case, crystalline phases were detected without indication of the existence of a significant amorphous portion. For instance, Figure 9 shows the diffraction patterns of three unstabilized zirconia powders, prepared at the same intensity (l = 6.6 ·10 4 W cm-2 ), but in different atmospheres (air, oxygen, argon).
These investigations with different atmospheres in  Dependent upon the preparation conditions, xy has values of 0.5 to over 0.9, although the monoclinic modification is stable at room temperature. Similar results have been reported previously [17,18,23]. Certainly, it is known [4] that the tetragonal modification can be stabilized by controlling the particle size. Garvie and Goss [4] discuss that phenomenon from thermodynamic considerations and conclude from the theory as well as from experimental results that zirconia particles with diameters of less than 10 nm should exist in the tetragonal form at room temperature. Lee et a!. [23] rely on these thermodynamic arguments to explain the high content of tetragonal particles in their laser-evaporated zirconia powders. However, in contrast to [4], we found tetragonal Zr0 2 particles with diameters of more than 100 nm [24].
Some experiments were repeated with a different atmosphere in order to test wether a possible nonstoichiometry of the oxide or an incorporation of nitrogen into the oxide lattice during vapour recondensation in air was being the culprits in hindering trans-KONA No.l3 (1995) formation from tetragonal-> monoclinic during the cooling process. However, the results in Figure 9 show that the intensity ratios remain unchanged if the atmosphere is varied while keeping the process conditions constant. In this case, we found a tetragonal part of about 55o/o independent of the atmosphere (air, oxygen, argon) employed. Thus, we have to conclude in agreement with Heuer and Ruhle [25] that this martensitic t->m transformation is kinetically dominated. We suppose that the delay in formation of monoclinic nuclei is due to the extremely short process time as well as to the very regular particle shapes in which structure defects, acting as possible nucleation centres, are rare. This hypothesis is supported by the observation that the amount of tetragonal zirconia can be decreased significantly, to less than 20o/o, by applying an additional tempering treatment for 30 min at 600°C, in the course of which SBET does not decrease significantly. This means that in spite of their small size, the majority of the zirconia particles tend to transform into the monoclinic form, but that during the process there is insufficient time to complete the transformation during the process.
In the case of alumina, also, particles of the most stable form (a-Al 2 0 3 ) are not obtained. We found a mixture of y-and o-Alz03 about one-third y-and twothirds a-alumina, although the powder is formed at temperatures significantly above the transformation temperature of a-alumina of approx. 1200°C. An almost identical diffraction pattern of alumina can be observ~d for powders prepared by the AEROSIL ® techniquL' [15]. In that case, too, the particle formation takes place at temperatures greater than 1200°C. Investigation of this transformation process is still in progress. By means of ICP optical emission spectroscopy (GBC Integra XM), it was possible to confirm that the powder preparation technique discussed here is an extremely clean process that does not contribute 86 impurities during the powder formation. In the case of alumina as well as of unstabilized zirconia, no differences in the chemical composition of the starting material and the prepared powders were observed. However, when evaporating Zr0 2 stabilized by CaO, MgO or Y 2 0 3 , clear composition changes are observed. The prepared powders are enriched and the residual material is depleted in stabilizers which have lower boiling points than zirconia (TB = 4300°C). Examples are magnesia (TB = 3600°C) and calcium oxide (TB = 2850°C). Conversely, the content of yttria (TB = -5000°C) is slightly reduced in the prepared powder. In the case of the evaporated mixture of alumina (TB = 2980°C) and zirconia, an enrichment of alumina is also found in the resultant powder. In future, we will try to diminish or avoid such effects by using a pulsed laser instead of a continuous one, since it is possible that a shortening of the process time will hinder the achievement of a local thermodynamic equilibrium.

Conclusions
The aim of this paper is to show that the technique of evaporating solid oxide materials in the focus of a laser is a competitive method for producing nanosized oxide powders in comparison with other gas-phase processes. It has been shown that this technique permits distinctly higher evaporation rates (in the order of 100 g Zr0 2 h-1 ) and better energy efficiency (40 g Zr0 2 kWh-1 ) than hitherto reported in the literature.
Evaporation rates and particle size distribution can be controlled over a wide range by adjusting various process parameters such as the laser intensity, the movement of the oxide target relative to the laser beam, the velocity of the carrier gas stream, and the mode of operation of the laser. The nanosized powders prepared have diameters in the range of 10 to 100 nm. The powders have specific surface areas up to 30 m 2 g-1 (Zr0 2 ) and 60 m 2 g-1 (Al 2 0 3 ), respectively. A further increase of these values seems to be possible.
The nanosized oxide powder prepared by laser evaporation are crystalline, but the existence of metastable modifications is obviously typical in such powders. Changes in the chemical composition are possible during the evaporation and recondensation of multi-component oxides, by applying the cw mode of operation of the laser. Since 1986 she is working as a research assistant at the Institute of Ceramic Materials of the Freiberg University of Mining and Technology with emphasis on joining and synthesis of technical ceramics.

Eberhard Miiller
Eberhard Muller received a diploma in physics in 1966 from the Friedrich Schiller University of ]ena. He got his doctor degree in 1970 at the same university in the field of crystal structure analysis by X-ray diffraction. In the following time he was engaged in structure analysis of surfaces and interfaces by electron diffraction as well as in theoretical investigations on the stability of surface structures. 1986 he defended his ''Doctor of Science'' thesis on the structure chemistry of crystalline interfaces. 1986 -1989 he was occupied with the preparation and characterization of highly dispersed oxides and with theoretical calculations of the surface energy of small particles. After a sabbatical at the Department of Material Engineering of the Federal University of Sao Carlos in Brazil (crystallization of glasses) he changed to the Freiberg University of Mining and Technology in 1990 and got a professorship on inorganic composites. Since that time he is engaged in the field of nanocrystalline ceramic powders as well as of ceramic fibres.

Christiane Oestreich
From 1975 to 1980 she studied physics. Since 1980 she works at the Technical University of Freiberg. Mrs. Oestreich obtained a doctorate in physics to luminescence of semiconductor materials in 1986. Since 1987 she is a research assistant in the department of material science. There she started with joining of ceramics. Her recent fields of research are physical vapour deposition and investigations of ultrafine ceramic powders by transmission electron microscopy.

Giinter Michel
Gunter Michel was born in Eisenach, Thuringen, Germany, on September 16, 1948. He received the degree of diploma in physics from the University of ]ena "Friedrich Schiller" in September 1972 in the area of quantum electronics. His dissertation concerned the characterization of perturbed arcs in turbulent axial gas flow. From 1973 to 1991 he worked at the Physical-Technical-Institute, ]ena, where he did research in the area of gas discharges and laser physics. In 1992 he transferred to the Technical-Institute of the University of ]ena. His present interests include High Power C0 2 -Lasers and their applications.
[ Author's short biography

Gisbert Staupendahl
Gisbert Staupendahl was born in Querfurt, Germany, on April 14, 1945. He received the diploma in Physics from the Physical Institute of the Friedrich Schiller University Jena. The graduation and the qualification for a lectureship he received from the same university in 1973 and 1981, respectively. In these years his research interests have included nonlinear optical effects in semiconductors, optical bistability and the development of high-power C0 2 lasers. From 1982 to 1987 he has been with the Feinmechanische Werke Halle, Germany, where he was engaged in the research and development of C0 2 lasers for industrial applications. Since 1987 he is a lecturer at the Technical Institute of the Friedrich Schiller University Jena. His present interests include the modulation of intense C0 2 laser radiation and the interaction of this radiation with matter.

Karl-Heinz Henneberg
After terminating his training as a precision mechanic (1964), firstly Karl-Heinz Henneberg studied control engineering (1967-70) and later he recieved a diploma in engineering for the construction of scientific instruments (1974) from the Friedrich Schiller University of Jena. Since 1974 he has been engaged in the field of research and development at the Institute of Physical Chemistry of the University of Jena. Topics in this field are the projecting and construction of devices for the preparation and characterization of highly dispersed and I or porous substances, specially oxides.