摘要:

Some of the history of the development of UHV technology, and its effects on surface physics and low‐energy collision studies, is presented. The author’s role in these developments is suitably over‐emphasized.

ISSN:0022-5355    

DOI:10.1116/1.568957

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

Electrochemical reactions at semiconductor surfaces depend on transport of electrons or holes between the surface and the bulk. This is very similar to transport in metal‐semiconductor Schottky barriers and leads to differences betweenn‐ andp‐type materials and to photochemical effects. Reaction at the surface may involve either decomposition of the semiconductor itself, or exchange of charge with an ion in the electrolyte. Photochemical decomposition of the crystal is common among binary compound semiconductors having pronounced ionic character. In an elementary model for this reaction the negative ion is identified with the valence band and the positive ion with the conduction band. Ions of one sign are neutralized, forming either a gas or a solid deposit on the surface. Minority carriers are needed to complete the reaction and these are generated by the light. Covalent semiconductors can also undergo reactions that are limited by the supply of minority carriers. Reactions with ions in solution can be understood by regarding the ions as localized states for electrons. Different ionic species have different energy levels for electrons and can react with carriers from either the valence band or the conduction band.

ISSN:0022-5355    

DOI:10.1116/1.568806

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

Organic polymer surfaces are metallized in two general ways: (i) thin film metallization in which metals are deposited by vapor deposition, sputtering, etc. and (ii) relatively thick metal coatings are deposited by means of electroless deposition followed by electrolytic deposition. In the present review paper, various applicable mechanisms—mechanical or interlocking, weak boundary layer, chemical, and electrostatic—of adhesion, relevant properties of metals and polymers, and techniques of controlling adhesion in metallized polymers are discussed, but the thin metallization of polymers has been emphasized. Evidence and mechanisms for the charge transfer across the metal–polymer interface are reviewed and the electrostatic component of adhesion is discussed. Apparently, there is a linear relationship between the charge transferred across the metal–polymer interface and the work‐function difference between the metal and the polymer. It is concluded that the electrostatic component of adhesion may have some contribution toward the over‐all experimentally determined adhesion. The adhesion can be improved by modifying the surfaces of polymers by subjecting them to glow discharge, grafting certain functional groups, etc. Formation of and evidence for chemical‐bond formation at the metal–polymer interface is discussed.

ISSN:0022-5355    

DOI:10.1116/1.568850

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

Aluminum thin films have been in common use as interconnections on Si devices since the early 1960’s. However, during the past decade alloying additions have been made to Al to improve some aspect of interconnection performance. For example, Si additions minimize metal penetration at ohmic contacts during heat treatment1; Cu additions greatly reduce the mass transport of Al due to electromigration.2It is widely recognized that Al interconnections will corrode in the field if not adequately protected by hermetic packages or a glass passivation system.3However, there has been little discussion of the corrosion of Al films during the fabrication of devices (i.e., in‐process) when there is typically little protection for the film, yet much exposure to corrosive reagents and environments. It is the purpose of this paper to focus on in‐process corrosion.The corrosion of Al and Al alloys in bulk metallurgical form has been studied for many years and is now reasonably well understood.4For example, it is known that the Cu‐bearing Al alloys are more subject to intergranular and pitting corrosion than pure Al. However, one very important difference between bulk and thin film observations is that superficial surface corrosion in bulk material can be equivalent to total destruction of a thin film which is typically 1 μm in thickness.When Al–Cu alloy (4.5% Cu, balance Al) was first introduced to the manufacturing of integrated circuits, the periodic presence of black spots on the interconnection patterns was noted. The level of the problem was low; less than 1% of the integrated circuits on a wafer would exhibit one or more black spots. At first, it was thought that the spots were of cosmetic importance, a consequence of poor photoresist removal or contamination of the surface. However, the scanning electron microscope soon revealed that the problem was one of ’’missing aluminum’’ in which a large part of the metallic cross section was absent. Therefore, the phenomenon was of great importance with regard to long term reliability of the product. The defect was too fine and infrequent to effectively screen with visual inspection.The use of electron microscopy and diffraction techniques showed the problem to be intergranular corrosion occurring in the processing sequence after subtractive etching, but before sputtered SiO2deposition. The corrosion was traced from the surfaces of the conductor along grain boundary paths. The subsequent use of ultrasonic cleaning magnified the defect by causing the separation of granular masses of Al. Typically, the associated corrosion products were chlorides, oxychlorides, and oxides of Al and Cu.The prime source of corrosion was found to be chlorinated hydrocarbons (Freon and trichloroethylene) which had decomposed in wafer cleaning baths. The decomposition came from the unintentional contamination of the chlorinated solvents with water and/or alcohol which caused the generation of free chloride ions in the bath. Chloride ions act as strong corrosion accelerators for reactive metals such as Al.5Characterization of alloy Al–Cu films showed large compositional inhomogeneities in the as‐deposited films, with anomalously large concentrations of Cu near the bottom of the film. After the heat treatment required to form good ohmic contacts, the Al–Cu becomes metallurgically overaged. Large particles of ϑ‐phase (CuAl2) deplete the Al alloy of Cu adjacent to grain boundaries. The Al‐rich regions in contact with the higher Cu regions establish microscopic galvanic couples which accelerate the dissolution of the Al‐rich regions. The result is intergranular and pitting corrosion.6Some effort was made to optimize the structure of the alloy to minimize corrosion. A more uniform distribution of Cu in the vertical profile, a reduction in tensile residual stress, and the sandwiching of the alloy between thin sacrificial layers of pure Al (thin film Alclad) were all steps in the right direction. However, optimization of the film structure could only modify the mode of the corrosion process, not eliminate it.The removal of chlorinated solvents, wherever possible, was the most effective means of eliminating ’’missing Al’’ even in nonhomogeneous, overaged alloy.However, once sensitivity to the problem was established, other promoters of the corrosion were also identified: fluoride ions and highly polar alcohols.It had been recognized, even with pure Al films, that the buffered HF used to etch via holes in sputtered SiO2passivation will rapidly attack Al when the etchant is greatly diluted. The danger period is not immediately at the end point of the etch cycle when the buffered HF mixture impinges directly on Al in the bottom of the via, but in the subsequent wash cycle in deionized water. As the HF is diluted with water, it passes through a low concentration level at which the etch rate of Al is maximized.7The postetch washing process must pass through this peak quickly to avoid corrosion.Molded CTFE (chlorotrifluorethylene) wafer carriers having trace impurities of free fluoride, a result of polymer decomposition during molding, were also found to be a cause of corrosion.During the recent petrochemical shortage, a much more subtle corrodant was discovered. The substitution of the more readily obtained methanol for isopropyl alcohol triggered a mild outbreak of in‐process corrosion in Al–Cu. It was then confirmed that a highly polar alcohol such as methanol can also be mildly corrosive to Al and should be avoided in thin film processing.

ISSN:0022-5355    

DOI:10.1116/1.568867

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

The progress in solid state technology is largely determined by our ability to control and modify certain materials parameters in a well defined manner. Semiconductors are a typical example, where minute traces of impurities in the order of parts per million and less will influence the electrical properties markedly.The concept of introducing such dopants into semiconductors by means of high energetic particles was discussed many years ago. In 1954, Shockley1submitted a patent describing the ’’Forming of Semiconductor Devices by Ionic Bombardment.’’ As the semiconductor technology matured, developing shallower and shallower device structures, the control aspect became vitally important. And it is precisely in the field of control that ion implantation offers major advantages over standard diffusion technologies, e.g.,(a) Accurate control of the total amount of impurity transferred to the wafer by measurement of the accumulated charge during implantations.(b) A high degree of areal uniformity across the wafer obtained by mechanical or electrical scanning (uniformity better than 1%).(c) Accurate control of the depth distribution by a well defined accelerating potential. Thus a wide range of dopant profiles for device configurations is possible.In contrast to standard diffusion profiles, which are frequently close to error‐function types (with the maximum concentration at the surface), the impurity profiles in implanted structures are of full‐Gaussian or truncated‐Gaussian type with the maximum concentration at a mean projected rangeRPand with a standard deviation ΔRP. Substantial progress has been made in recent years in understanding the physical phenomena concerning the interaction of high‐energy particles with target atoms. Investigations by various authors2–6have given rise to theoretical models enabling one to make reasonably accurate predictions of implant profiles as a function of ion species, target material, implant energy, and dose. These models also permit some estimate of the radiation damage and its distribution. The radiation damage is sometimes a less desirable side effect of this new technology and, in most applications, requires post‐implant annealing. This effect is strongly dependent on species, dose, energy, and certain process details. A typical example is discussed by Mader and Michel7for high‐dose As implantations into Si as required for emitter, subcollector, or source–drain application. In these experiments, As was predeposited into a shallow surface layer (∠500 Å) by an implant step and subsequently diffused into the Si. Pronounced differences in the resulting defect patterns are found for implants into bare silicon and through screen silicon oxide films. In the latter case, oxygen atoms are injected into the silicon by knock‐on effects between the high energetic As projectiles and the oxygen. The high‐temperature heat treatment8,9serves a multiple purpose: to ’’activate’’ the implanted species electrically, that is to permit impurities like As, B, or P to be incorporated into substitutional lattice sites; to anneal out the radiation damage; and, in special cases, even to diffuse the impurities away from a residual radiation‐damaged zone deeper into the bulk region, a consideration particularly important for emitter processes in bipolar devices. Excellent electrical junction characteristics have been obtained under such conditions.10Ion implantation is expected to be strongly directional and to result in considerably less lateral spread than standard diffusions in a planar technology. Furukawaetal.11have studied these effects theoretically, and experiments by Pan and Fang12have confirmed their calculations on narrow‐gate field effect transistor (FET) devices by use of self‐alignment concepts.In fact, it was in the area of FET’s that ion implantation was first used successfully to manufacture active silicon devices. Since they require low doping concentrations and shallow distributions, FET’s are ideally suited to take advantage of the features of ion implantation.Ion implantation opened up the possibility of selective channel‐doping and individual threshold adjust, thus enabling one to build enhancement and depletion devices on the same chip. It further permits the reduction of parasitic capacitances, increasing the speed performance of FET’s to the point where they can compete favorably with bipolar transistors. Fang and Crowder13have reported on a microwave FET structure, which has been operated up to 14 GHz. By combining ann‐channel enhancement‐driver device with a depletion‐load FET, Fang and Rupprecht14have measured turn‐on delays as low as 115 ps per stage for an 11‐stage ring oscillator.It was mentioned before that ion implantation is normally accompanied by radiation damage. Some examples follow to illustrate how, in certain cases, these effects can be utilized beneficially. For many years, FET devices have been known to be very susceptible to the gate insulator integrity. Large threshold variations can be measured because of fast surface states at the insulator–silicon interface, or because of mobile charges (e.g., Na ions). Normally, the surface‐state density can be effectively reduced by a hydrogen annealing step. In the presence of a silicon nitride–silicon oxide gate structure, relatively high temperatures are required to diffuse the hydrogen through the nitride layer. Various workers in the field therefore suggested the use of implantation. An interesting side effect, according to Ku,15was the simultaneous reduction in mobile ions. The present assumption is that the mobile ions become pinned on traps in the silicon oxide, which are created by the radiation damage rather than by a chemical effect. This model is supported by the observations of Goetzberger,16who found a similar phenomenon by implanting sodium directly into the gate oxide. Another example in which radiation damage has been utilized is the gettering of fast‐diffusing metals to improve the electrical junction characteristics. Bucketal.17for instance, reported on striking reductions in the density of bright spots on silicon photodiodes.Summary: This paper has concentrated on the advances made in the semiconductor field by the use of ion implantation, this being the area in which the greatest impact has been felt so far. There are, however, many other disciplines in solid state technology—such as metallurgy, magnetic bubbles, insulators, and integrated optics—where ion implantation has just started to make contributions and where a wide open field is to be explored.

ISSN:0022-5355    

DOI:10.1116/1.568825

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

It is shown for the special example of silicon films on sapphire and spinel that the ’’interface’’ between both single crystals is more complex than the simple superposition of two different lattice planes. The structural, mechanical, chemical, optical, and electrical aspects of the interface are discussed. It is shown that, depending on the kind of aspect, we can describe it as a two‐dimensional interface, a three‐dimensional transition layer, or a three‐dimensional intermediate layer. All the quoted properties depend both on the kind of preparation of the surface, like polishing or heat treatment, and on the epitaxial deposition process itself, e.g., on temperature and growth rate. A model for this special example of heteroepitaxy will be presented.

ISSN:0022-5355    

DOI:10.1116/1.568832

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

A most common technique used in the fabrication of optoelectronic devices involves the growth of one or several epitaxial layers of a givenAIIIBVsemiconductor compound over a single crystal substrate. The choice of the epitaxial compound composition (binary, ternary, or even quaternary compounds are used) is usually guided by the desire to achieve a given band gap. As a result, a lattice mismatch is generally introduced between the epitaxial layers and substrate. The lattice mismatch parameterf, which is a function of the film composition, and the temperature of the layer‐substrate structure, is an important parameter for achieving control over the film quality and the device performances. The misfit‐induced defect generation and the role of these defects on the device performances are analyzed. The results presented which cover experiments on Ga1−xAlxAs1−yPy–GaAs double heterojunction structures and GaP homojunction structures are general enough to be applied to other types of mismatched epitaxial structures.The buildup in misfit strain energyEfduring the growth of a mismatched epitaxial layer is released by elastic and/or plastic deformation of the lattice. The introduction of misfit dislocations, one form of the film plastic deformation, takes place in several stages which are functions of the magnitutde ofEfand the structure temperature.1,2Substrate defects such as dislocations, dislocation loops, and precipitates are playing a major role in the first stage of the misfit dislocation formation since they originate from these substrate defects. During the first stage of formation an interesting asymmetrical distribution of misfit dislocations along a 〈110〉 direction is also observed for {100} Ga1−xAlxAs1−yPy–GaAs structures.1,2Use of this asymmetrical behavior is made to obtain defect‐free double heterostructures. The second stage of formation corresponds to the generation of ’’60° type’’ dislocation half loops to accomodate the misfit strain in the other 〈110〉 direction. The third stage results in an increase in the density of interfacial dislocations with Burger’s vectors in the interface plane.In the case of GaP epitaxial layers3with small misfit values, several defect‐formation stages are observed during the film growth. The first one corresponds to a large increase (10–100 times) in the epitaxial‐layer dislocation density over that of the substrate. These new dislocations which originate from substrate dislocation loops near the interface are not efficient in relieving the misfit strain. The other stages are analogous to those described above for the formation of misfit dislocations in Ga1−xAlxAs1−yPy–GaAs structures. The role of dislocations on the device efficiency and degradation behavior1,4is analyzed. Evidence for a stimulated dislocation motion and multiplication process upon minority carrier injection conditions is presented. This dislocation motion is attributed to the recombination‐enhanced mobility of point defects in these materials and to the efficient point‐defect sink behavior of the dislocations.5

ISSN:0022-5355    

DOI:10.1116/1.568895

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

What is normally referred to as ’’adhesion’’ of a thin film to a substrate is, in reality, the sum of five forces. There are three possible attachment forces: electrostatic, van der Waal’s, and chemical bonding. The detachment forces are stress in the film and at the film–substrate interface and any externally applied forces (e.g., scratching, abrasion, peeling, etc.). Virtually all of the so‐called ’’adhesion tests’’ that have been described should really be called ’’service use tests,’’ in that they test for failure in a particular fracture mode. As a result, quantitative information regarding interfacial bond strength cannot be derived. Further, common to all these tests is the requirement that the film surface be disturbed (e.g., by attaching something to the surface, scratching it, etc.). The very act of disturbing the surface could propagate mechanical damage to the interface and further cloud the results when quantitative information about the bond strength is desired.The bond strength of a thin film to a substrate can be measured quantitatively, with no contact to or disturbance of the film surface prior to measurement by a laser spallation technique. The technique involves impinging a pulsed, high‐energy, Nd‐glass laser beam (1.06 μm) onto the back surface of a substrate (made opaque in the case of transparent substrates). The explosive evaporation of absorbing material on the back side of the substrate generates a compressive shock wave in the substrate directed toward the film–substrate surface. The magnitude of the shock wave can be varied to establish a threshold for spallation or detachment of the film from the substrate. With appropriate corrections for shock wave damping in the substrate and stress in the film, the technique can yield quantitative information about the bond strength of the film to the substrate.Ideally, the film should be patterned into dots of a size such that the entire dot can be detached with one laser shot. This avoids the problem of measuring the sum of the film–substrate bond strength and the tensile strength of the film. Further, it is possible in some cases to examine the back side of a spalled dot and the front side of the substrate from which the dot was spalled by various surface analytic techniques to gain insight into the mechanisms by which interfaces are formed.

ISSN:0022-5355    

DOI:10.1116/1.568925

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

Mössbauer spectra of thin, unenriched Fe and Fe–Si films were obtained from conversion electrons in a backscatter geometry. The electron detector was a windowless, channel electron multiplier constructed in our laboratory. It operates in vacuum with a relatively high efficiency for counting electrons, enabling the measurement of spectra from thin films deposited on conventional, thick substrates and eliminates the necessity for thin transparent absorbers. A comparison between common transmission Mössbauer spectroscopy and the more rarely employed backscatter methods shows that backscatter conversion‐electron Mössbauer spectrscopy results in larger signal‐to‐noise ratios as well as significant reduction in counting times when thin films are involved. The magnitude of the effects theoretically expected in films will be discussed and compared with that measured experimentally.

ISSN:0022-5355    

DOI:10.1116/1.568932

出版商:American Vacuum Society    

年代:1976

数据来源: AIP

摘要:

Since XPS samples only a few surface layers of the solid for most materials, the signal from these first few layers makes a dominant contribution to the total signal. For the inorganic azides available, cross sections indicate that in most cases the surface layers contain less than the stoichiometric amount of azide (nitrogen). Since most of the metals are reactive, the excess metal is predominantly in the form of compounds, e.g. oxides. Carbon and oxygen are always observed as surface contaminants and in many cases the amounts of C and O detected are of the same order of magnitude as the metals and azide. The lack of stoichiometry and C and O contamination may be important for electrical properties involving the surface, e.g., electric‐field initiation to detonation of explosive azides. The surface films may tend to provide protective coatings, so that twenty‐year‐old lead azide does not show significant bulk decomposition. The decomposed surfaces can be reactivated for most azides by exposure to hydrazoic acid. Short exposure of copper metal to HN3vapor produces a thin film of copper azide.

ISSN:0022-5355    

DOI:10.1116/1.568943

出版商:American Vacuum Society    

年代:1976

数据来源: AIP