Electrochemical microsystem technologies: from fundamental research to technical systems

Electrochimica Acta 44 (1999) 3605±3627

Electrochemical microsystem technologies: from fundamental research to technical systems J.W. Schultze*, V. Tsakova 1 AGEF e.V.-Institut an der Heinrich-Heine-UniversitaÈt DuÈsseldorf, D-40225 DuÈsseldorf, Germany Received 9 September 1998; received in revised form 25 November 1998

Abstract The interdisciplinary ®eld of electrochemical microsystem technologies (EMST) has a large impact on electrochemistry, electrochemical engineering, microengineering, material science, chemical analysis and its applications in biology and medicine. EMST include electrochemical reactions applied in microsystem technologies (MST) as well as MST applied in electrochemistry. Fundamental research refers to the scaling-down of reactions taking place in the mm range. Control of electrochemical microreactions, their advantages and applications are discussed. The technical requirement of mass production presumes the scaling-up of the microreaction to a multifold macroscopic process. This is demonstrated for electrochemical processes, e.g. the LIGA process, as well as for MST for electrochemistry, e.g. the Foturan1 technology. The ®nal industrial process requires the realization of multistep processes which is illustrated with the phosphating process and the printed circuit board manufacturing as examples. Electrochemical microsystems are presented and classi®ed by functionality and complexity. # 1999 Elsevier Science Ltd. All rights reserved.

1. Introduction During the last 30 years, electrochemical research became increasingly concerned with microscopic problems. With the progress of micromechanics and the improved sensitivity of electrical equipment, measurements on miniaturized electrodes became possible. Simultaneously, progress in mathematics and simulation techniques allowed an improved modelling of microscopic systems. The ®rst calculations on microelectrode reactions by Fleischmann [1] represent a typi-

* Corresponding author. Tel.: +49-21-1811-4750; fax: +49211-81-12803. E-mail address: [email protected] (J.W. Schultze) 1 On leave from the Institute of Physical Chemistry, Bulgarian Academy of Sciences, 1113 So®a, Bulgaria.

cal example. At ®rst, kinetic problems were solved and rates of di€usion and migration were calculated. Later on, microscopic problems of the electrode surface and its modi®cations became of interest for corrosion and chemical analysis. The papers presented at the Faraday Discussion on ``The solid/liquid interface at high resolution'' [2] illustrate this type of research which allows solving of the special problems of materials science, corrosion and reaction kinetics. Simultaneously, electrochemical analysis was pushed ahead in biology and medicine. Due to the special requirements of measurements at biological membranes, ultramicroelectrodes with a high selectivity for various ions were developed. Thus, microelectrochemical techniques are now widely applied in biology and medicine. The achievements of Neher [3] and Sakmann [4] obtained by the patch clamp technique are a typical example for successful application in biological research. Finally, electroche-

0013-4686/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 1 3 - 4 6 8 6 ( 9 9 ) 0 0 0 6 5 - 1

local analysis by reacting microelectrodes galvanic deposition of metals and alloys

Materials science, corrosion Electrode processes

Electrochemical microdevices and microsystems

local analysis by probing microelectrodes

Fundamental electrochemistry, biology and medicine

microengineering design

deposition of polymer and insulatring materials metal etching

Method

Field

Table 1 Microelectrochemistry in science and technology

ELMAS

micrelectrode arrays microdosimat

electrochemical micromachining microcell

Si-technology, printed curcuit boards polybithiophene on Si

magnetic recording media and magnetic heads LiGA

laser oxidation, pitting

amperometry, potentiometry, cyclic voltammetry, SECM

Application/examples

microanalysis in biology and medicine

fabrication of metal masks electrochemical reactions in microvolumes

metal structures for microelectronics and micromechanics Cu-interconnects in IC-chips through hole Cu-plating microstructures

determination of : concentration pro®les (e.g. pH) and transients, c(x, y, z, t ), kinetics of redox reactions kinetics of growth of oxide ®lms and corrosion processes Ni±Fe-magnetic alloys

Result

[33,34] [35,36]

[28±31] [32]

[27]

[23±26]

[11]

[20]: oxide on Al, [21]: Ti corrosion, [22]: pitting [12]

[15,16]: electroanalysis, [17]: pH, [18]: SECM, [19]: biocells

Refs.

3606 J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 1. Double-logarithmic plot of current I versus electrode surface area S. Typical ranges for di€erent processes and systems are indicated. EC=electrochemical, PCB=printed circuit boards, LIGA=lithographic galvanic process. Dashed lines indicate points of equal current density.

3607

mical analysis pro®ted from the progress in miniaturisation of sensors. In the ®eld of microelectronics and micromechanics, electrochemical techniques were applied for special deposition or etching processes [5,6]. The LIGA process [7±11] and the production of magnetic recording media [12,13] and magnetic heads [14] are typical examples. Only recently fully integrated electrochemical microsystems were developed. To illustrate the wide applications, Table 1 summarizes some achievements of microelectrochemistry in these di€erent ®elds and gives a few examples of methods, materials and systems. The ``International Symposium on Electrochemical Microsystem Technologies'' in DuÈsseldorfGrevenbroich, 1996, initiated a get-together of scientists from all these very di€erent ®elds. A lot of special know-how and ideas on electrodeposition, sensors, dielectrics, pharmaceutical problems and electrochemical systems could be collected and exchanged [37]. This discussion will be continued at the Second International Symposium on Electrochemical Microsystem Technologies in Tokyo, 1998. Since the term `electrochemical microsystem technologies' was used in di€erent ®elds with various meanings, we want to de®ne the term emphasizing each part of EMST:

Fig. 2. Fundamental relations for scaling-down: di€usion limited current Id, current density id, capacity C, Ohmic drop (IdR ) and RC-constant in dependence on the electrode radius r. (Schematic diagram in arbitrary units.)

3608

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

. Electrochemistry means all processes of ions or molecules at charged interfaces, i.e. faradaic and electroless processes. They take place usually at solid/liquid interfaces, but liquid/liquid interfaces [38] may be included, also. . Micro limits the systems to dimensions between mm and mm. The lower limit, however, is not sharp, since roughness in the nm-range or ultramicroelectrodes of nm dimensions are typical elements or aspects of microelectrochemical systems. . The term system means a combination of two or more parts or devices. A single microelectrode, for example, is essential but in EMST it is only a part of microsystems. . Technology ®nally limits the ®eld to processes and procedures with general relevance and good reproducibility. In contrast to the traditional electrochemical engineering, the scaling-down with a design of a microsystem becomes important [39]. This, however, is only the ®rst step. In a next step, the scaling-up to mass production becomes again a typical engineering problem. Even this comprehensive de®nition of the ®eld leaves us with two interpretations: `electrochemical technologies for microsystems' or `microsystem technologies for electrochemistry'. In fact, the experience shows that both topics are of great importance for EMST and we will discuss them both here. Based on this de®nition of the ®eld, it is obvious that up to now, there is no adequate description of principles of EMST. Therefore, we will describe the general features of EMST, discuss the scaling-down and scaling-up, the principles of technology and ®nally the development of electrochemical microsystems. 2. General features of EMST 2.1. Classi®cation by current and miniaturisation A rough classi®cation of EMST is shown in the double-logarithmic plot of current I versus surface S in Fig. 1. Relevant current densities decrease from MA (106 A) down to pA/cmÿ2 by 18 orders of magnitude. The dashed lines with the slope d log I/d log S=1 connect points of equal current density i. The range of common laboratory experiments at large surfaces is illustrated on the right side of the diagram. Continuous advances in miniaturisation shifted the accessible range of reactions to the left side into the ®eld of microelectrochemistry at small surfaces down to 1 mm2. In the upper part of this diagram, reactions with high current densities are shown. They are used for electrodeposition, e.g. in electroformation of conducting lines for the production of printed circuit boards (PCB) [25,26],

or mechanical structures in the LIGA process or etching, e.g. in electrochemical machining for mask production [28±31]. In electrochemical research, fast kinetic investigations are also possible at that range of very high current densities up to 1 A/cm2. On the other hand, the ®eld of analysis uses much smaller current densities usually in the range between pA and mA/cm2. Potentiometric analysis is done at negligible current densities. Especially for biological systems, voltammeters with high resistances of 1014 O are necessary to allow measurements at almost stable biological membranes. Also in the ®eld of surface analysis, corrosion, etc., very low current densities should be measured [39]. The measuring devices, operating nowadays down to the pA range, limit the region of modern electrochemical surface analysis. 2.2. Principle procedures of electrochemical microsystem technologies The broad ®eld of EMST involves very di€erent procedures like scaling down, scaling up, and combination and integration of various devices. This can be described by the development of multifunctional electrode arrays in the biological research as example. It started with the invention of glass electrodes of macroscopic dimensions, continued with the development and miniaturisation of speci®c electrodes based on the ion selectivity of ionophores incorporated into a membrane within a miniaturized glass capillary. Thus, the potentiometric determination of protons, magnesium, potassium and other cations was possible [40]. The smallest single electrodes were introduced into biology and medicine more than 10 years ago. In the meantime, the advances of glass technology allowed the combination of 2 or more capillaries to a multifunctional microelectrode array which is used as an advanced electroanalytic microsystem in the pharmaceutical research (see Fig. 14a). Similar developments can be demonstrated by the LIGA process and the production of thin ®lms or microreactors. In general, the procedures of research and development involve the following steps: . Scaling-down of systems down to the micrometer scale. . Scaling-up of miniaturised reactions or processes to a technical reaction in a multifold macroscopic system. . The combination of various reactions to an industrial process. . The integration of various modules to an electrochemical or other system. These steps will be used as a concept for the further review in Sections 3±7.

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3609

Fig. 3. Examples for single microelectrodes and electrode arrays: a) metal electrode for amperometric measurements [64], b) tip for electrochemical experiments in the STM, c) membrane electrode for potentiometry [17], d) microelectrode array for local analysis of glucose and other substances [75].

2.3. Reactions of EMST In principle, all electrochemical reactions involving an interface can be used for EMST. These include of course all Faradaic reactions as metal deposition or dissolution, oxide formation and anodic polymerization. Further, all reactions which are catalyzed or induced by cathodic or anodic reactions, e.g. the electrodeposition of paint [41,42] or phosphating [43] have to be considered. Finally, all electroless etching or deposition reactions consisting of competing anodic and cathodic partial reactions should be taken into account. With respect to their function, the electrochemical reactions used in EMST could be classi®ed in surface processing, microstructuring and functionalisation reactions. Surface processing starts with cleaning or pickling which removes organic deposits, tarnishing products and a small part of the substrate surface, too. Since most microstructuring reactions require a special topography, electropolishing or roughening are important

pre-treatment processes [44]. The anodic fabrication of porous silicon is an example of extreme roughening of microstructures [45,46]. Finally, an activation of a surface for following deposition reactions, e.g. the preparation of a conducting oxide on titanium layers for the LIGA process can be carried out [47]. Microstructuring reactions create positive microstructures by deposition, e.g. cathodic deposition of metals and alloys or anodic deposition of oxides or conducting polymers. On the other hand, negative structures can be obtained by anodic etching which is widely used for the preparation of both masks and negative silicon structures. Another class of processes consists of electrochemically induced reactions which take place as a consequence of electron transfer at the electrode surface. The cathodic or anodic electrodeposition of paint is an example which is well-known from car production, but applicable also for the microstructuring of sensors or the insulation of tunnel tips. Many microstructures need a surface ®nishing or functionalization as a ®nal process. The aim of such treatments may be manifold. Thus, the passivation of

3610

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 4. Various concepts of electrochemical microcells: a) open nl-droplet cell b) rigid or movable mask [85], c) capillary-cells [35,111], d) microcell [32], e) biological cell [19].

the surface is carried out to stabilize the microstructure against corrosion. Packaging consists of electrical contacting or insulation. Surface modi®cation for special functions may be obtained by deposition of thin surface ®lms, alloying etc. It should be noted that electrochemical processes used for technological purposes are optimized with respect to the ®nal quality of platings obtained. However, fundamental research distinguishes between the early nucleation and advanced growth stages in electrochemical deposition processes [48±50]. Thus further optimization of electrochemical processes should rely on a detailed understanding of the in¯uence of di€erent parameters on both nucleation and growth kinetics [51±54]. 2.4. Topographic, mechanical and chemical criteria of EMST As in MST, the quality of products of EMST is evaluated by engineering and chemical criteria. The aspect ratio A=height/width is a topographic criterion which is managed only by few technologies, and can be measured by few topographic techniques, too. High steepness (S=d(height)/d(width)) of microstructures, reaching about 908 can be another requirement. This criterion becomes important for etching, since for many systems the ratio of etching rates in depth and width is usually near to 1. However, this ratio may be increased up to 100 or more depending on many factors, e.g. crystal structure [55] or side wall passivation. Another important parameter is the roughness of walls which for series production with moulding tools or for optical applications should be in the nm-range. A second group of criteria refers to the mechanical and chemical properties of the microproduct. In general, they should be close to those of the macroscopic

bulk materials. However, due to special growth mechanism, the in¯uence of walls and di€erent concentration pro®les, anisotropy or chemical inhomogeneity can occur. Failure e€ects occurring during electrodeposition in microstructures have been analyzed for example by KuÈpper [56]. The preparation of gradient materials, e.g. Ni±Fe alloys is described in [11].

3. Scaling-down 3.1. Size e€ects of microelectrochemistry The kinetics of microelectrochemical reactions strongly depend on the size of the electrode and the rate of mass transport processes. This e€ect will be discussed here for disk electrodes in dependence on the radius r. As is known, at macroelectrodes the current increases linearly with r 2. For microelectrodes, however, a transition from planar to hemispherical di€usion pro®les takes place and modi®es the transport conditions. According to the theory the di€usion limited current (Id) at a microdisk electrode increases with r and the di€usion limited current density (id) decreases linearly with r [15,16]: Id ˆ 4nFcDr

…1†

id ˆ 4nFcD…pr†ÿ1

…2†

Microelectrodes present speci®c advantages such as the possibility to discriminate against double layer charging currents and to decrease the distortion from ohmic (IR) drop. Thus, for microdisk electrodes the ohmic resistance is given by: R ˆ …4sr†ÿ1

…3†

Localised signals

analysis of corrosion, oxide growth on Ti

metal deposition/dissolution

laser induced thermal reactions

steel corrosion

laser induced photoreactions

localised inhibition

W>30 mm

W>2 mm, A < 1

W>10±50 mm

W>3 mm W>10 mm, S 2 mm

[89±93]

[21,87,88]

[86]

[25,26]

[20,35], [32]

[84] [11], [85]

[25,26]

(continued on next page)

surface analysis

microdroplet, microcell

printed circuit boards polymer deposition on Si

W> 0.1 mm W>10 mm, S < 1 mm, 0 < A < 103, R < 0.3 mm W>100 mm, A < 1

Si-technology LIGA

®xed or movable mask

(electro) Chemical modi®cation passivation of copper through mask implantation

W>20 mm, S 1 mm, A < 1

printed circuit boards

Parameters: width W, sharpness Refs. S, aspect ratio A, roughness R

lithography, vis, UV, X-ray

Fields of application/examples

Geometric blocking

Model

Method

Principle

Table 2 Localization of electrode reactions: Parameters are given as mean typical values (not lowest limits) and correspond to the example mentioned

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627 3611

Localised transport

Principle

Table 2 (continued )

redox reaction, deposition, etching W>20 mm

W>60 mm

W>100 mm

[96±103]

[94,95]

[85]

Parameters: width W, sharpness Refs. S, aspect ratio A, roughness R

SECM

metal deposition

Fields of application/examples

studies of fast kinetics

Model

microjet

electric ®eld

Method

3612 J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3613

Fig. 5. Dimensions of electrochemical cells in a double-logarithmic plot of width (dy) vs thickness (dx). Dashed lines show aspect ratios 10ÿ3, 1 and 103, respectively.

and therefore the ohmic drop IdR does not depend on the microdisk radius. In contrast to microelectrodes, the term IdR increases linearly with r for macroelectrodes. The di€erent behaviour of macro- and microelectrodes with respect to current, ohmic drop, capacity and RC-constant is illustrated in Fig. 2. It should be emphasized that systems with high symmetry which allow exact mathematical treatment of the electrochemical kinetics are usually applied in fundamental research [1,15,16,57,58]. Examples for such systems are the spherical mercury droplet in polarography or the microdisk electrodes with 3- and 2D-symmetry, respectively, used for kinetic studies and analysis. However, the structures in EMST have usually a low (0- or 1-dimensional) symmetry. Moreover, the aspect ratio A of such structures can reach high negative or positive values as in recessed electrodes or pores and thus can in¯uence the mass transfer of electroactive species. Numerical calculations for given topographic pro®les are needed in order to account for the geometrical e€ects in these cases [59±63].

mechanical precision technologies can be applied down to the mm range for ¯at systems, they are limited to about 100 mm for high aspect ratios. Therefore, other techniques, e.g. heating of glass, etching of metals, ion mill techniques for tip preparation, have to be applied. Some examples of microelectrodes and microelectrode arrays are shown in Fig. 3. For amperometric measurements disk electrodes with a diameter of 10 mm can be melted in glass (Fig. 3a) [16,64]. For STM measurements, thin wires of about 25 mm are etched yielding a tip radius of about 20 to 50 nm [65]. They are insulated by electrodeposition of paint or nail-varnish which only leaves the utmost part of the tip uncovered (Fig. 3b). To measure the local concentrations in arti®cial microstructures, biological cells or corrosion pits, membrane electrodes which are selective for protons or other cations are used (Fig. 3c) [17]. They are widely applied in biology and medicine [66±71]. New developments refer to microelectrode arrays [33,34,72±81], e.g. for local analysis of glucose, oxygen or other substances (Fig. 3d) [75].

3.2. Microelectrodes

3.3. Localization of electrode reactions

The ®rst important technological step of EMST consisted of the miniaturisation of electrodes and cells from macroscopic to microscopic dimensions. While

The most important part of EMST consists of the localization of electrode reactions described above. As shown in Table 2, independent of the chemical process,

3614

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 6. Scaling-up of a microelectrode array to a technical process on a wafer [81].

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3615

localization of the reaction signal. Laser induced reactions [21,87±91,105] lead to a focussing down to 1 mm. High electric ®elds in vapor condensed ®lms allow even nanostructuring [106,107]. Local transport control is possible by the SECM [96±103], by microjet stream through a capillary [94,95] or by localized deposition of oxidizers [26]. 3.4. Miniaturisation of cells

Fig. 7. Technological sequence in the LIGA process [126].

various principles for the localization may be distinguished (see also Refs. [82,83]). Traditional techniques limit the reaction boundaries, for example, by geometric blocking. Photoresist techniques are widely used for metal deposition as well as etching, e.g. in the production of printed circuit boards. For special experiments, movable masks [85] or the wetting droplet technique [20,35] may be applied. While the photoresist and the wetting droplet may be used for small aspect ratios, the mask techniques are applicable for microstructures with high aspect ratios. In research, combinations of such techniques, e.g. the droplet on a photoresist, are useful for localized surface analysis [104]. Other techniques use the chemical modi®cation of the surface by localized adsorption of inhibitors [86] or the implantation of ions which allows the change of chemical or electronic surface properties. Thus, for example nitrogen implantation into silicon surfaces blocks photochemical reactions and, thus, allows for localized electropolymerisation [27]. Another principle of microstructuring consists of the

In the seventies, thin ®lm cells were developed for investigations of electrosorption by coulometry and spectroscopy [108±110]. In research, microelectrodes can be investigated in traditional macroscopic cells. Surprisingly, even AFM and STM experiments are carried out in macroscopic cells in spite of the atomic resolution which is achieved. As a result, there is no correlation between the STM-image and the microscopic picture of the samples investigated. For EMST and the localization of electrochemical reactions, on the other hand, the miniaturisation of the whole cell is often necessary. In Fig. 4, various concepts of microcells are compared. The photoresist concept for a microcell can be realized using electrolyte droplets with small volumes (less than 1 nl). This technique allows to minimize the parasitic e€ect of the photoresist and was successfully applied in local surface analysis of Ti electrodes [104]. In that case, even a combination with a thin ®lm layer was possible ®xing a glass window on a small distance wire [21]. Fig. 4(b) shows a simple arrangement realized by Karstens [85]. A microstructured (LIGA processed) honeycomb mask was used as a movable cell. The counter and reference electrodes were inserted by micromechanic manipulation. Applications of capillaries with an open droplet or closed area [111,112] are shown in Fig. 4(c). Finally, a microcell for optical measurements was developed recently which allows a local attachment of the cell on the investigated material, optical control of the surface, spectroscopic measurements and an adjustable electrolyte ¯ux through a small cell volume (Fig. 4d) [32]. This cell can also be used as an electrochemical microreactor with small changes of construction. A survey on the size of electrochemical microcells is given in the double logarithmic plot of Fig. 5. They can be characterized by their lateral (dy) and vertical (dx) dimensions. Lateral cells are characterized by a small aspect ratio, A < 1, and a large ratio of surface to volume. The cells shown in Fig. 4(b), (c) and (d) are up to now the smallest electrochemical cells with 3 electrodes. In STM experiments, e.g. with water adsorption ®lms [106,107] or the single ion cell [113], the electrochemical cells are much smaller, but contain only 2 electrodes. The overview of the microcell constructions in Fig. 4 demonstrates the important role of manufacturing

3616

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 8. Microsystem technologies for electrochemistry: fabrication of an electrochemical microcell in Foturan1 technology [32] and array fabrication in silicon planar technology [81].

technologies. Capillaries of glass or metals, silicon technology and new glass technologies like Foturan1 (see Section 4) are necessary for these developments.

4. Scaling-up 4.1. Dimensions of technical EMST processes and typical substrates In chemical engineering, scaling-up is usually achieved by increasing the dimensions of the apparatus. In contrast, scaling-up in EMST occurs through multiplication of the optimized microsystem. The great advantage of microtechnologies is the realization of mass production, i.e. the production of multiple microstructures on macroscopic substrates. A typical substrate is the silicon wafer due to its high mechanical

stability and smoothness. On the other hand, the production of printed circuit boards uses plates of epoxy resins as substrates which are covered by a thin copper ®lm. The LIGA process can be carried out on silicon wafers covered by a thin PVD ®lm of titanium. The wafer technology was also applied for the microstructuring of glass in the Foturan1 process [55,114,115]. With all these substrates, it is possible to carry out electrochemical processes with manifold repetition in periodic systems of macroscopic dimensions of 10 cm or more. The application in mass production, of course, requires both high reproducibility of microstructures on the whole substrate and high production yield. In general, the scaling-up of microreactions investigated in the mm-range in the laboratory challenges the engineers and the electrochemists as well. An example of a double-fold scaling-up is shown in Fig. 6 which presents a microelectrode array produced

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3617

Fig. 9. Development of packaging of microsystems: from the Dual Inline Package to Ball Grid Array and Flip Chip technologies [116].

by means of the silicon planar technology [81]. The ®rst scaling-up consists in the realization of many electrodes within a silicon chip of 1 cm2. The second scaling-up allows for the production of multiple chips in equal or di€erent design on a wafer with a size of 3 to 12 inch. A silicon wafer covered with an insulating SiO2 layer is used as a substrate. Then, electrodes, circuit lines and bond pads are prepared by PVD/ PECVD techniques applying photoresists, masks and RIE techniques well-known from the silicon technology. This microelectrode array was designed in order to be able to decide on the in¯uence of size and geometry of the microelectrodes on di€erent reactions. 4.2. The LIGA process A real electrochemical process for mass production is demonstrated by the LIGA technology [7±11]. It consists of the production of microstructures by deep etch X-ray lithography, electroforming and injection moulding/embossing (Fig. 7). The ®rst steps involve the fabrication of the mask which is used to prepare a photoresist pattern. The patterned resist layer,

obtained on a conducting substrate, is electroplated with nickel to form a mould insert. The latter is further used for the fabrication of plastic microcomponents or components with characteristic microfeatures by means of injection moulding or embossing. Then, after initial metallization, the plastic parts can be electroplated thus allowing to obtain corresponding metallic microcomponents. The LIGA process opens a way for a cost e€ective, large scale fabrication of microcomponents with special properties such as large height, precise wall geometry and additionally provides the possibility to use di€erent functional materials. The scaling-up is achieved in this case by the fabrication of multiple microstructures on silicon wafers. 4.3. Non-electrochemical technologies for EMST Figs. 4 and 6 have already shown electrodes and cells which are prepared by non-electrochemical technologies meeting the requirements of mass production of microstructures. Various steps of the silicon planar technology for the production of the microelectrode array of Fig. 6 are shown in Fig. 8.

3618

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 10. Lateral and vertical microtechnologies for electrochemistry in a double-logarithmic plot.

The planar technologies are preferential techniques for structures with low aspect ratios A < 1 wheras for high aspect ratios the glass technology in Foturan1 [55,114] is advantageous. The latter was used for the production of the glass cell shown in Fig. 4(d). 3 inch glass wafers of Foturan1 were processed by illumination through a mask and following etching. The di€erent etching rates v of non-illuminated and illuminated glass, vill/vinit>30, allow high aspect ratios to be obtained, but not such steepness as obtained by means of the LIGA-process. Using di€erent masks, a 3dimensional cell with specially designed volumes, channels and electrode connectors was produced. For the whole structure, about 20 glass chips are ®xed together by a glue or thermal bonding (for details see Ref. [32]). 4.4. Printed circuit boards and packaging Electrochemical reactions are widely applied in the production of PCBs. The principal technology is a negative microstructuring process with a low aspect ratio, A < 1. First, the epoxy substrate is covered by a thin ®lm of copper. The conducting lines are protected by a photoresist, while all other parts of the copper ®lm are etched. The typical width of copper lines is

now about 100 mm but this size should be diminished to about 20 mm in the near future. A special aspect of the PCB production consists in the through hole plating which means the plating of the side walls of drilling holes in the insulating material with A 1 5. This is a very special example of a sophisticated multistep microstructuring process which is described later (see Section 6.2). In contrast to the successful miniaturisation of devices to the nanometer scale, the packaging, i.e. the device connection, is nowadays limited by the miniaturisation down to 20 mm. The development of the packaging technology is schematically shown in Fig. 9 [116]. This shows the miniaturisation of a 64-pin device. Compared with the dual-in-line design of the 70's the geometric area of the package was reduced by a factor of 0.05 in the 90's. This is mainly done an arrangement of the contacts in rows and columns instead of two rows. The distance from pin to pin decreases from 2.5 to about 0.1 mm only. 4.5. Microtechnologies for EMST: overview In general, the scaling-up of electrochemical reactions and the production of electrodes, cells and other

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3619

Fig. 11. Scaling up and down in materials science with phosphating as example.

systems require a lot of di€erent technologies. An overview is given in Fig. 10 in dependence on the lateral and vertical dimensions of the system. Lateral and vertical techniques are separated by the aspect ratio A=1 which is indicated by the dashed line. As shown in Fig. 3 for di€erent electrodes, almost classical technologies like production of glass capillaries or etching of wires can be used to prepare nanostructures down to 100 nm. Silicon technologies are advantageous for planar processes, but can also be improved in the direction of 3D structures. Laser techniques are suitable for planar structures, while the LIGA process is the best for high aspect ratios. The accuracy of the Foturan1 technology is a little bit smaller due to the etching ratio (see 4.3). Finally, the technology of

packaging, i.e. insulating or contacting of microstructures, is very important because it is in fact often the limiting factor in micro- and nanoelectronics.

5. Scaling down and up in materials science In Sections 3 and 4 the scaling down and scaling up of technologies was discussed. Such procedures are also important for materials science. This will be explained by the process and research of electroless phosphating which is realized in the automotive industry with about 1000 km2 per year. Fundamental research of phosphate crystals was done with crystals carefully grown from the solution

3620

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

Fig. 12. Simpli®ed ¯ow diagram for the phosphating process [127]. Ucorr, corrosion potential; dox,, dphos-thickness of the corresponding layers; activation by TiPO4.

with ®nal size in the mm-range. The inset of Fig. 11 shows such a crystal with well oriented planes [117]. The crystal size and growth procedure used di€er essentially from the technological process taking place at the interface steel/water. Therefore, scaling down of materials science is necessary. Advanced technological research considers the growth of much smaller single crystals on a real steel surface under simulated technological conditions. Such crystals grow with preferential orientation (lateral or perpendicular to the surface) and can be studied by SEM, AFM and EDX. An AFM picture of these crystals (about 2±20 mm) is inserted in Fig. 11, also. The technical process in the car industry, however, again takes place at di€erent conditions, since the density of nuclei exceeds by some orders of magnitude the number density of crystals obtained by the technological research. This means that the concentration gradients which determine the growth of the crystals are di€erent in both the model experiments with microscopic crystals at the steel surface and the model growth of single crystals in solution.The upper insert in Fig. 11 shows a technical phosphating layer having about 107 crystals/cm2 with a size of 2 to 3 mm. Of course, the ®nal industrial process is carried out in the m-range. Similar e€ects can be illustrated by the study of Ti

crystals. It was only recently that the orientation of single grains of polycrystalline Ti could be determined by anisotropy micro-ellipsometry [118]. The possibility to isolate single grains in a technical material by means of a photoresist technique allowed to use these isolated grains as microelectrodes and to carry out single crystal experiments [21]. So far, this article concentrated on the aspects of EMST in the mm-range. Many materials such as the phosphate crystals described above have grain sizes of some mm and can be classi®ed as micromaterials. This, however, is not always the case, since in other systems with low nucleation work Ak the density of nuclei exceeds 108/cm2 and the ®nal size of the grains is in the nm-range. The nanomaterials are expected to have di€erent properties in comparison to the corresponding bulk materials due to their large surface area, possible anisotropy and the probability for quantum size e€ects to take place. Examples for anisotropy present the magnetic materials with longitudinal and transversal magnetization [119] and the conducting polymer materials grown in nanotubes [120]. In Fig. 11 the ®eld of nanomaterials follows above that of micromaterials. Experiments with single nanocrystals are carried out by means of the scanning tunnelling microscope and thus the scaling-down of surface science approaches

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3621

Fig. 13. Simpli®ed ¯ow diagram for the through hole plating process. v-velocity of epoxy resin loss, d(MnOx), d(polymer, d(copper) - thickness of the corresponding layers.

the nm-range. By analogy to the scaling-up in engineering, the production of nanomaterials is now also used in various ®elds of materials science. In fact, the scaling-down of experiments with single crystals is not yet a continuous one, but there is a gap between micromaterials and the STM experiments with single nanocrystals. (For instance, in the case of phosphating nucleation is initiated by the so called `activation' with TiPO4 whereas the hopeite crystals so far investigated are characteristic for an advanced stage of growth.) Correspondingly, the scaling-up of materials science from a single nanocrystal to the nanomaterials involves the jump in the production conditions. It can be expected that scaling-down in surface science and scaling-up in technology will be almost continuous in the near future.

6. Multistep processes for the industrial production The industrial production of a microsystem consists usually in a great number of technological steps including di€erent reactions. The process starts in general with cleaning, surface treatment and continues with activation etc. The main reaction of deposition or etching may be preceded, combined or followed by the pro-

cedure of microstructuring. Finally, the post-treatment follows. 6.1. Flow diagram of phosphating After the above discussion of the phosphating from the view point of materials science, we concentrate now on the technical process used in the car industry. In spite of the large-scale substrates of few meters, technical phosphating today is a highly sophisticated microstructuring process. The typical thickness and diameter of phosphate crystals should be about 2 mm and the density of the almost periodically arranged crystals is about 3  107 cmÿ2. The active, uncovered surface of the steel substrate should be less than a few per cent. This is necessary to allow the electrodeposition of paint in the following process. The phosphating itself consists of an indirect electroless process with a twofold function: the steel surface has to be etched and roughened and the phosphate layer has to be deposited. An electrochemical, cathodic ÿ reduction of the so-called accelerator (NOÿ 2 NO3 ) compensates the anodic dissolution of steel or zink. The deposition of the phosphate crystals is the result of a local pH shift at the interface. The industrial process consists of about 10 or more steps including clean-

Fig. 14. Examples for microelectrode arrays: a) multifunctional ion-sensitive electrode array in glass technology [19], b) monofunctional microelectrode array [75], c) monofunctional array from Panasonic1, d) monofunctional array used in biological research [34].

3622 J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

3623

Fig. 15. Examples for electrochemical microsystems: (a) microcell produced in Foturan1 technology [32], (b) electrochemical micro analysis system [36], (c) microdevice for electrophoresis produced in Si-technology [121] and (d) microdosimat [35].

ing, rinsing, activation, phosphating, and rinsing again. A simpli®ed ¯ow chart diagram is shown in Fig. 12. Parameters of the industrial process (temperatures, velocity of the electrolyte, concentration of accelerators, etc.) have to be distinguished from the system parameters like the coverage of activator, the thickness and coverage of the phosphate (for more details see Refs. [43,121]). 6.2. Through hole plating of printed circuit boards The through hole plating of PCBs is also an important industrial process which is now carried out with altogether 106 m2 per year. The crucial part of the reaction is the deposition of a conducting material

within a drilling hole of about 200 mm diameter with an aspect ratio of up to 5. It starts with a electroless oxidation of the epoxy resin by KMnO4 which only attacks the insulating polymer in the hole and simultaneously passivates the existing copper ®lm. Within the hole, a thin layer of about 100 nm MnO2 is deposited. This is used as a localized oxidant for the following production of the conducting polymer. Finally, a layer of 100 nm thickness of polyethylenedioxythiophene is produced which allows the following copper plating within the hole. Fig. 13 shows a simpli®ed ¯ow diagram for this process consisting of 5 steps including oxidation, rinsing, polymerisation, rinsing and electroplating [26]. During each of these di€erent processes, the technical parameters, like temperature, pH, velocity

3624

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

of the PCB have to be adjusted. The surface potential and the pH are regulated by the chemical composition of the solution. As a result, the system parameters, such as thicknesses of the manganese oxide, polymer and copper coatings change (see Fig. 13). Another example of a multistep process, the deposition of a conducting polymer within porous silicon, was described in [45]. 7. Application of EMST in microsystems 7.1. Microsystems with electrochemical production steps Microsystems now are widely applied in electronics, mechanics, chemical analysis, engineering and biology. In each of these ®elds the combination of `micro' with the name of the ®eld covers a range of highly sophisticated technical and scienti®c activities. A large number of systems is prepared with electrochemical processes as described above. Many ®nal systems such as magnetic heads or recording media, micromechanical devices etc. do not have a special electrochemical function. The description of these systems is out of the scope of the present article. 7.2. Microsystems with electrochemical functions For special microsystems, the electrochemical function i.e. the ¯ux of a current through an interface, is the process desired. The microsystems with electrochemical application can be classi®ed according to their role and functionality. Fig. 14 illustrates di€erent microdevices used for analysis in biology and medicine. The ion-sensitive electrode array produced by Schlue [19] is a multifunctional microtool allowing to detect simultaneously the local concentrations of di€erent ions. The microelectrode arrays used by Cammann [75] and Fromherz [34] are examples for monofunctional devices for local analysis of oxygen, H2O2 or glucose and neurocell activity, respectively. In these cases all electrodes are equal, have an equal distance and only a reference electrode is added on the chip. The commercially available Panasonic1 multielectrode array for local potential measurements in medicine can be also used only as a monofunctional device. In contrast, the multielectrode array described in Fig. 6 is intended as a multifunctional system which can combine various electrodes for di€erent reactions. Fig. 15 shows few examples of fully integrated electrochemical microsystems. The electrochemical microcell (Fig. 4d and Fig. 15a) produced by the Foturan1 technology can be used for electrochemical measurements in microvolumes combined with spectroscopic and optical investigations [32]. A completely di€erent

system is represented by ELMAS, the electrochemical microanalysis system (Fig. 15b) [36] which includes mechanical microparts, i.e. pumps and valves and electrochemical sensors. An interesting application of electrochemical microsystems is the micromachined Si-chip (Fig. 15c) used for mixture separation through capillary electrophoresis [122,123]. Fig. 15(d) shows the concept of a microdosimat for pH-regulation [35]. Future developments of electrochemical microsystems include the design of microreactors and microbatteries. Electrochemical microreactors are intended to realize electrochemical reactions with the advantage of microcells, e.g. good heat dissipation and short contact time of intermediates. While in the sixties, the Monsanto process was realized with the capillary cell by Beck [124,125], it is now carried out even better in electrochemical microcells. Another interesting ®eld of application is given by microbatteries which could be useful for electronic devices. In that case, however, the miniaturisation is limited by the total charge to be accumulated. Since 10 to 100 C have to be stored at least, only thin ®lm batteries have been developed up to now. For example, the smallest batteries used now in hearing aids have a diameter of 2 to 3 mm. Miniaturisation in this ®eld, however, can be useful for special purposes.

Acknowledgements The support of this work by the Ministerium fuÈr Wissenschaft und Forschung des Landes NordrheinWestfalen by the project ``Integrierte Mikrosystemtechnik fuÈr Fest/FluÈssig-Systeme'' and the ®nancial support of the Alexander von HumboldtFoundation (grant for VT) are gratefully acknowledged.

References [1] S. Pons, M. Fleischmann, Anal. Chem. 59 (1987) 1391A. [2] The liquid/solid interface at high resolution, Faraday Discuss 94 (1992). [3] E. Neher, Angew. Chemie 104 (1992) 837. [4] B. Sakmann, Angew. Chemie 104 (1992) 844. [5] L.T. Romankiw, D.R. Turner (Eds.), Electrodeposition Technology. Theory and Practice, Electrochemical Society Pennington, NJ, 1987. [6] N. Masuko, T. Osaka, Y. Ito (Eds.), Electrochemical Technology: Innovation and New developments, Kodansha and Gordon and Breach, Tokyo and Amsterdam, 1996.

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627 [7] E.W. Becker, W. Ehrfeld, P. Hagmann, A. Maner, D. MuÈnchmeyer, Microelectron. Eng. 4 (1986) 35. [8] W. Ehrfeld, in: H. Reichl (Ed.), Microsystem Technologies, Springer Verlag, Heidelberg, 1990, p. 521. [9] P. Bley, in: H. Reichl (Ed.), Microsystem Technologies, Springer Verlag, Heidelberg, 1990, p. 302. [10] W. Menz, P. Bley, Mikrosystemtechnik fuÈr Ingenieure, VCH, Weinheim, 1993 p. 189. [11] W. Bacher, in: N. Masuko, T. Osaka, Y. Ito (Eds.), Electrochemical Technology: Innovation and New developments, Kodansha and Gordon and Breach, Tokyo and Amsterdam, 1996, p. 159. [12] T. Osaka, in: N. Masuko, T. Osaka, Y. Fukunaka (Eds.), New Trends and Approaches in Electrochemical Technology, Kodansha and VCH, Tokyo and Weinheim, 1993, p. 13. [13] K. Ouchi, in: N. Masuko, T. Osaka, Y. Ito (Eds.), Electrochemical Technology: Innovation and New developments, Kodansha and Gordon and Breach, Tokyo and Amsterdam, 1996, p. 197. [14] O. Shinora, in: N. Masuko, T. Osaka, Y. Ito (Eds.), Electrochemical Technology: Innovation and New developments, Kodansha and Gordon and Breach, Tokyo and Amsterdam, 1996, p. 219. [15] R.M. Wightman, in: A.J. Bard (Ed.), Electroanalytical Chemistry, vol. 15, Marcel Dekker, New York, Basel, 1989. [16] J. Heinze, Angewandte Chemie, Int. Ed. Engl. 32 (1993) 1268. [17] E. Klusmann, J.W. Schultze, Electrochim. Acta 42 (1997) 3123. [18] A.J. Bard, G. Denault, Ch. Lee, D. Mandler, D.O. Wipf, Acc. Chem. Res. 23 (1990) 357. [19] W.R. Schlue, W. Kilb, D. GuÈnzel, Electrochim. Acta 42 (1997) 3197. [20] A.W. Hassel and, M. Lohrengel, Electrochim. Acta 42 (1997) 3327. [21] S. Kudelka, J.W. Schultze, Electrochim. Acta 42 (1997) 2817. [22] J.O. Park, M. Verho€, R. Alkire, Electrochim. Acta 42 (1997) 3281. [23] P. Andricacos, in: The Electrochemical Society Interface, 1998, p. 2. [24] P. Andricacos, in: IBM J. Res. Devel. Electrochem. Microfab. (special issue), September (1998). [25] E. Yung, L.T. Romankiw, R. Alkire, J. Electrochem. Soc. 136 (1989) 206. [26] D. Schattka, S. Winkels, J.W. Schultze, Metallober¯aÈche, 1997, p. 51. [27] T. Morgenstern, J.W. Schultze, Electrochim. Acta 42 (1997) 3057. [28] M. Datta, in: M. Datta, K. Sheppard, D. Snyder (Eds.), Electrochemical Microfabrication, The Electrochemical Society Proceedings, Pennington, New York, 1992, p. 61. [29] M. Datta, J. Electrochem. Soc. 142 (1995) 3801. [30] M. Datta, in: N. Masuko, T. Osaka, Y. Ito (Eds.), Electrochemical Technology: Innovation and New Developments, Kodansha and Gordon and Breach, Tokyo and Amsterdam, 1996, p. 137.

3625

[31] M. Datta, D. Harris, Electrochim. Acta 42 (1997) 3007. [32] A. Vogel, J.W. Schultze, M. KuÈpper, in: Proc. 2nd Int. Conf. on Microreaction Technology, 1998, p. 323. [33] H. Meyer, H. Drewer, J. Krause, K. Cammann, R. Kakerow, Y. Manoli, W. Mokwa, M. Rospert, Sensors Actuators B 18±19 (1994) 229. [34] P. Fromherz, Ber. Bunsenges. Phys. Chem. 100 (1996) 1093. [35] M. Lohrengel, Electrochim. Acta 42 (1997) 3265. [36] R. Rapp, W. Ho€mann, W. SuÈb, H.J. Ache, H. GoÈlz, Electrochim. Acta 42 (1997) 3391. [37] J.W. Schultze (Ed.), Electrochemical microsystem technologies, Electrochim. Acta 42 (20±22) (1997), 2981± 3406. [38] F. Silva, M.J. Sousa, S.M. Pereira, Electrochim. Acta 42 (1997) 3095. [39] J.W. Schultze, M. Schweinsberg, Electrochim. Acta 43 (1998) 2761. [40] R.C. Thomas, Ion-Sensitive Intracellular Microelectrodes. How to make and use them, Academic Press, London, 1978. [41] F. Beck, H. Guder, J. Appl. Electrochem. 15 (1985) 825. [42] K. Arlt, Electrochim. Acta 39 (1994) 1189. [43] A. Losch, J.W. Schultze, D. Speckmann, Metallober¯aÈche 46 (1992) 133. [44] Taschenbuch fuÈr Galvanotechnik, LPW-Chemie, Neuss, 1988. [45] J.W. Schultze, K.G. Jung, Electrochim. Acta 40 (1995) 1369. [46] M.J. SchoÈning, F. Ronkel, M. Crott, M. Thust, J.W. Schultze, P. Koerdos, H. LuÈth, Electrochim. Acta 42 (1997) 3185. [47] A. Michaelis, J.W. Schultze, A. Thies, DECHEMA Monogr. 125 (1992) 459. [48] A. Milchev, Contemp. Phys. 32 (1991) 321. [49] R. Winand, Electrochim. Acta 39 (1994) 1091. [50] E. Budevski, G. Staikov, W.J. Lorenz, Electrochemical Phase formation and Growth, VCH, Weinheim, 1996. [51] M. Kjuchukova, V. Tsakova, A. Milchev, J. Appl. Electrochem. 19 (1989) 819. [52] E. Michailova, I. Vitanova, D. Stoychev, A. Milchev, Electrochim. Acta 38 (1993) 2455. [53] E. Michailova, M. Peikova, D. Stoichev, A. Milchev, J. Electroanal. Chem. 366 (1994) 195. [54] M. Peykova, E. Michailova, D. Stoychev, A. Milchev, Electrochim. Acta 40 (1995) 2595. [55] T.R. Dietrich, W. Ehrfeld, M. Lacher, M. KraÈmer, B. Speit, Microelectron. Eng. 30 (1996) 497. [56] M. KuÈpper, J.W. Schultze, Electrochim. Acta 42 (1997) 3023. [57] M. Fleischmann, S. Pons, D.R. Rolison, P.P. Schmidt (Eds.), Ultramicroelectrodes, Datatech System, Science Publishers, Morganton, 1987. [58] in: M.I. Montenegro, M.A. Queiros, J.L. Dashbach (Eds.), Microelectrodes: Theory and Application, NATO ASI Series E 197, Kluwer, Dodrecht, 1991. [59] A.J. Bard, J.A. Crayston, G.P. Kittlesen, Th.V. Shea, M.S. Wrighton, Anal. Chem. 58 (1986) 2321.

3626

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627

[60] S. Mehdizadeh, J.O. Dukovic, T.T. Romankiw, H.I. Cheh, J. Electrochem. Soc. 139 (1992) 78. [61] S. Mehdizadeh, J.O. Dukovic, T.T. Romankiw, H.I. Cheh, J. Electrochem. Soc. 140 (1993) 3497. [62] R.V. Shenoy, M. Datta, J. Electrochem. Soc. 143 (1996) 544. [63] R.V. Shenoy, M. Datta, L. Romankiw, J. Electrochem. Soc. 143 (1996) 2305. [64] M. KuÈpper, J.W. Schultze, Electrochim. Acta 42 (1997) 3085. [65] H. Siegenthaler, in: R. Wiesendanger, H.J. GuÈtherodt (Eds.), Scanning Tunneling Microscopy II, Springer Verlag, Berlin-Heidelberg, 1992, p. 7. [66] T. Zeuthen, Curr. Top. Membr. Transport 13 (1980) 31. [67] J.W. Deitmer, W.R. Schlue, P¯uegers. Arch. Europ. J. Physiol. 397 (1983) 195. [68] W.R. Schlue, R.C. Thomas, J. Physiol. (London) 364 (1985) 327. [69] W.R. Schlue, J. Neurophysiol. 65 (1991) 736. [70] P. Hochstrate, Ch. Pile, W.R. Schlue, Brain Res. 696 (1995) 231. [71] K. Holtho€, O.W. Witte, Electrochim. Acta 42 (1997) 3241. [72] P. Connolly, G.R. Moores, W. Monaghan, J. Shen, S. Britland, P. Clark, Sensors Actuators B 6 (1992) 113. [73] G.C. Fiaccabrino, M. Koudelka-Hep, S. Jeanneret, A. van den Berg, N.F. de Rooij, Sensors Actuators B 18± 19 (1994) 675. [74] R. Kakerow, Y. Manoli, W. Mokwa, M. Rospert, H. Meyer, H. Drewer, J. Krause, K. Camman, Sensors Actuators A 43 (1994) 296. [75] H. Meyer, B. Naendorf, M. Wittkampf, B. GruÈndig, K. Cammann, R. Kakerow, Y. Manoli, W. Mokwa, M. Rospert, in: A. van den Berg, P. Bergveld (Eds.), Micro Total Analysis Systems, Kluwer, 1995, p. 245. [76] H. Meyer, H. Drewer, B. GruÈndig, K. Cammann, R. Kakerow, Y. Manoli, W. Mokwa, M. Rospert, Anal. Chem. 67 (1995) 1164. [77] G. Gross, B. Rhoades, H. Azzazy, M.-Ch. Wu, Biosensors Bioelectron. 10 (1995) 553. [78] K. Raimer, C. KoÈhler, T. Lisec, U. Schnakenberg, C. Fuhr, R. Hintsche, B. Wagner, Sensors Actuators A 46 (1995) 66. [79] G.C. Fiaccabrino, X. -M, T. ang, N. Skinner, N.F. de Rooij, M. Koudelka-Hep, Anal. Chim. Acta 326 (1996) 155. [80] H. Frebel, G.-C. Chemnitius, K. Cammann, R. Kakerow, M. Rospert, W. Mokwa, Sensors Actuators B 43 (1997) 87. [81] G. Buss, J.W. Schultze, M.J. SchoÈning, H. LuÈth, Electrochim. Acta, 44 (1999) 3899. [82] J.W. Schultze, T. Morgenstern, D. Schattka, S. Winkels, Electrochim. Acta, 44 (1999) 1847. [83] J.W. Schultze, M. Lohrengel, Metallober¯aÈche 51 (1997) 735. [84] L. Romakiw, Electrochim. Acta 42 (1997) 2985. [85] O. Karstens, J.W. Schultze, W. Bacher, R. Ruprecht, Metallober¯aÈche 48 (1994) 148. [86] K. Saurbier, J.W. Schultze, J. Geke, Electrochim. Acta 39 (1994) 1171.

[87] J.W. Schultze, K. Bade and, A. Michaelis, Ber. Bunsenges. Phys. Chem. 95 (1991) 1349. [88] K. Bade, O. Karstens, A. Michaelis, J.W. Schultze, Faraday Discuss. 94 (1992) 45. [89] R.J. von Gutfeld, E.E. Tynan, R.L. Melcher, S.E. Blum, Appl. Phys. Lett. 35 (1979) 651. [90] M. Datta, L.T. Romankiw, D.R. Vigliotti, R.J. von Gutfeld, J. Electrochem. Soc. 136 (1989) 2251. [91] M.H. Gelchinski, L.T. Romankiw, D.R. Vigliotti, R.J. von Gutfeld, J. Electrochem. Soc. 132 (1985) 2575. [92] M.E. BaumgaÈrtner, Ch.J. Raub, DECHEMA Monogr. 117 (1989) 191. [93] H.R. Khan, M.U. Kittel, Ch.J. Raub, in: Plating and Surf. Finish., 1988, p. 58. [94] J. Macpherson, S. Marcar, P.R. Unwin, Anal. Chem. 66 (1994) 2175. [95] J. Macpherson, P.R. Unwin, Anal. Chem. 69 (1997) 5045. [96] O. HuÈsser, D. Craston, A.J. Bard, J. Vac. Sci. Technol. B 6 (1988) 1873. [97] O. HuÈsser, D. Craston, A.J. Bard, J. Electrochem. Soc. 136 (1989) 3222. [98] Y. -M-Wuu, F. -R, F. an, A.J. Bard, J. Electrochem. Soc. 136 (1989) 885. [99] D. Mandler, A.J. Bard, J. Electrochem. Soc. 136 (1989) 3143. [100] D. Mandler, A.J. Bard, J. Electrochem. Soc. 137 (1990) 2468. [101] K. Borgwarth, D.G. Ebling, J. Heinze, Ber. Bunsenges. Phys. Chem. 98 (1994) 1317. [102] C. Heb, K. Borgwarth, C. Ricken, D.G. Ebling, J. Heinze, Electrochim. Acta 42 (1997) 3065. [103] Y. Zu, L. Xie, B. Mao, Z. Tian, Electrochim. Acta 43 (1998) 1683. [104] S. Kudelka, A. Michaelis, J.W. Schultze, Electrochim. Acta 41 (1996) 863. [105] T. Morgenstern, A. Thies, J.W. Schultze, Sensors Actuators A 51 (1995) 103. [106] H. Sugimura, T. Uchida, N. Kitamura, H. Masuhara, Appl. Phys. Lett. 63 (1993) 1288. [107] H. Sugimura, K. Okiguchi, N. Nakagiri, M. Miyashita, J. Vac. Sci. Technol. B 14 (1996) 4140. [108] E. Schmidt, H. Siegenthaler, Helv. Chim. Acta 52 (1969) 2245. [109] H. Siegenthaler, E. Schmidt, J. Electroanal. Chem. 80 (1977) 129. [110] H.D. Abruna (Ed.), Electrochemical Interfaces:Modern Techniques for in situ Interface Chracterization, VCH, Weinheim, 1991. [111] T. Suter and, H. BoÈhni, Electrochim. Acta 42 (1997) 3275. [112] T. Suter and, H. BoÈhni, Electrochim. Acta 43 (1998) 2843. [113] A.J. Bard, lecture at the Symposium Electrochemical Microsystem Technologies, Grevenbroich, 1996. [114] T.R. Dietrich, W. Ehrfeld, M. Lacher, B. Speit, F&M 7±8 (1996) 520. [115] A. Ruf, J. Diebel, M. Abraham, T.R. Dietrich, M. Lacher, J. Micromech. Microeng. 6 (1996) 254. [116] By courtesy of H. Meyer, Atotech, Berlin. [117] B. Middendorf, Diplomarbeit, 1989.

J.W. Schultze, V. Tsakova / Electrochimica Acta 44 (1999) 3605±3627 [118] A. Michaelis, J.W. Schultze, Thin Solid Films 274 (1996) 82. [119] T. Osaka, Electrochim. Acta 42 (1997) 3015. [120] C.R. Martin, R. Pathasarathy, V. Menon, Electrochim. Acta 39 (1994) 1309. [121] A. Losch, E. Klusmann, J.W. Schultze, Electrochim. Acta 39 (1994) 1183. [122] A. Manz, E. Verpoorte, C.S. E€enhauser, N. Burggraf, D.E. Raymond, H.M. Widmer, Fresenius J. Anal. Chem. 348 (1994) 567.

3627

[123] D. Raymond, A. Manz, H. Widmer, Anal. Chem. 66 (1994) 2858. [124] F. Beck, H. Guthke, Chem. Ing. Technik 41 (1969) 943. [125] F. Beck, in: Proc. IUPAC, Hamburg, 1973, vol. V, 1974, p. 111. [126] Prospectus `Microfabrication by the LIGA process', Kernforschungszentrum Karlsruhe, Institut fuÈr Mikrostrukturtechnik. [127] N. MuÈller, Dissertation, DuÈsseldorf, in preparation.