Atomic layer deposition(ALD), a chemical vapor deposition technique based on sequential self-terminating gas–solid reactions, has for about four decades been applied for manufacturing conformal inorganic material layers with thickness down to the nanometer range. Despite the numerous successful applications of material growth by ALD, many physicochemical processes that control ALD growth are not yet sufficiently understood. To increase understanding of ALD processes, overviews are needed not only of the existing ALD processes and their applications, but also of the knowledge of the surface chemistry of specific ALD processes. This work aims to start the overviews on specific ALD processes by reviewing the experimental information available on the surface chemistry of the trimethylaluminum/water process. This process is generally known as a rather ideal ALD process, and plenty of information is available on its surface chemistry. This in-depth summary of the surface chemistry of one representative ALD process aims also to provide a view on the current status of understanding the surface chemistry of ALD, in general. The review starts by describing the basic characteristics of ALD, discussing the history of ALD—including the question who made the first ALD experiments—and giving an overview of the two-reactant ALD processes investigated to date. Second, the basic concepts related to the surface chemistry of ALD are described from a generic viewpoint applicable to all ALD processes based on compound reactants. This description includes physicochemical requirements for self-terminating reactions,reaction kinetics, typical chemisorption mechanisms, factors causing saturation, reasons for growth of less than a monolayer per cycle, effect of the temperature and number of cycles on the growth per cycle (GPC), and the growth mode. A comparison is made of three models available for estimating the sterically allowed value of GPC in ALD. Third, the experimental information on the surface chemistry in the trimethylaluminum/water ALD process are reviewed using the concepts developed in the second part of this review. The results are reviewed critically, with an aim to combine the information obtained in different types of investigations, such as growth experiments on flat substrates and reaction chemistry investigation on high-surface-area materials. Although the surface chemistry of the trimethylaluminum/water ALD process is rather well understood, systematic investigations of the reaction kinetics and the growth mode on different substrates are still missing. The last part of the review is devoted to discussing issues which may hamper surface chemistry investigations of ALD, such as problematic historical assumptions, nonstandard terminology, and the effect of experimental conditions on the surface chemistry of ALD. I hope that this review can help the newcomer get acquainted with the exciting and challenging field of surface chemistry of ALD and can serve as a useful guide for the specialist towards the fifth decade of ALD research.

Surface chemistry of atomic layer deposition: A case study for the trimethylaluminum/water process

The silicon-based microelectronics industry is rapidly approaching a point where device fabrication can no longer be simply scaled to progressively smaller sizes. Technological decisions must now be made that will substantially alter the directions along which silicon devices continue to develop. One such challenge is the need for higher permittivity dielectrics to replace silicon dioxide, the properties of which have hitherto been instrumental to the industry's success. Considerable efforts have already been made to develop replacement dielectrics for dynamic random-access memories. These developments serve to illustrate the magnitude of the now urgent problem of identifying alternatives to silicon dioxide for the gate dielectric in logic devices, such as the ubiquitous field-effect transistor.

Alternative dielectrics to silicon dioxide for memory and logic devices

Atomic layer deposition (ALD) is gaining attention as a thin film deposition method, uniquely suitable for depositing uniform and conformal films on complex three-dimensional topographies. The deposition of a film of a given material by ALD relies on the successive, separated, and self-terminating gas–solid reactions of typically two gaseous reactants. Hundreds of ALD chemistries have been found for depositing a variety of materials during the past decades, mostly for inorganic materials but lately also for organic and inorganic–organic hybrid compounds. One factor that often dictates the properties of ALD films in actual applications is the crystallinity of the grown film: Is the material amorphous or, if it is crystalline, which phase(s) is (are) present. In this thematic review, we first describe the basics of ALD, summarize the two-reactant ALD processes to grow inorganic materials developed to-date, updating the information of an earlier review on ALD [R. L. Puurunen, J. Appl. Phys. 97, 121301 (2005)], and give an overview of the status of processing ternary compounds by ALD. We then proceed to analyze the published experimental data for information on the crystallinity and phase of inorganic materials deposited by ALD from different reactants at different temperatures. The data are collected for films in their as-deposited state and tabulated for easy reference. Case studies are presented to illustrate the effect of different process parameters on crystallinity for representative materials: aluminium oxide, zirconium oxide, zinc oxide, titanium nitride, zinc zulfide, and ruthenium. Finally, we discuss the general trends in the development of film crystallinity as function of ALD process parameters. The authors hope that this review will help newcomers to ALD to familiarize themselves with the complex world of crystalline ALD films and, at the same time, serve for the expert as a handbook-type reference source on ALD processes and film crystallinity.

Crystallinity of inorganic films grown by atomic layer deposition: Overview and general trends

The present invention provides for sequential chemical vapor deposition by employing a reactor operated at low pressure, a pump to remove excess reactants, and a line to introduce gas into the reactor through a valve. A first reactant forms a monolayer on the part to be coated, while the second reactant passes through a radical generator which partially decomposes or activates the second reactant into a gaseous radical before it impinges on the monolayer. This second reactant does not necessarily form a monolayer but is available to react with the monolayer. A pump removes the excess second reactant and reaction products completing the process cycle. The process cycle can be repeated to grow the desired thickness of film.

Sequential chemical vapor deposition

This report presents the results of steady-state and time-resolved luminescence measurements performed on suspensions of nanocrystalline ZnO particles of different sizes and at different temperatures. In all cases two emission bands are observed. One is an exciton emission band and the second an intense and broad visible emission band, shifted by approximately 1.5 eV with respect to the absorption onset. As the size of the particles increases, the intensity of the visible emission decreases while that of the exciton emission increases. As the temperature decreases, the relative intensity of the exciton emission increases. In accordance with the results presented in a previous paper, we assume that the visible emission is due to a transition of an electron from a level close to the conduction band edge to a deeply trapped hole in the bulk ( ) of the ZnO particle. The temperature dependence and size dependence of the ratio of the visible to exciton luminescence and the kinetics are explained by a model in w...

The kinetics of the radiative and nonradiative processes in nanocrystalline ZnO particles upon photoexcitation

Thin (2–10 nm) silicon nitride films have been grown by repetitive plasma nitridation of Si using a NH3 remote plasma and deposition of Si by a SiH2Cl2 thermal reaction. The deposition rate is self‐limited at nearly half‐molecular layer (ML) per one deposition cycle. The process window for the half‐ML/cycle of growth has been investigated with respect to the NH3 plasma power, SiH2Cl2 exposure time, and substrate temperature. The thickness fluctuation of the film over a 2 in. wafer is within measurement accuracy of the ellipsometer (± 1.9%) for the atomic layer controlled film while it is ± 8.5% for all the remote‐plasma chemical vapor deposition film.

Atomic layer controlled deposition of silicon nitride with self‐limiting mechanism

A reliable method to determine the threshold voltage V/sub th/ for MOSFETs with gate length down to the sub-0.1 /spl mu/m region is proposed. The method determines V/sub th/ by linear extrapolation of the transconductance g/sub m/ to zero and is therefore named "GMLE method". To understand the physical meaning of the method and to prove its reliability for different technologies 2-D simulation was applied. The results reveal that determined V/sub th/ values always meet the threshold condition, i.e., the onset of inversion layer buildup.

Physically-based threshold voltage determination for MOSFET's of all gate lengths

We report on a new roadblock which will limit the gate oxide thickness scaling of MOSFETs. It is found that statistical distribution of direct tunnel leakage current through 1.2 to 2.8 nm thick gate oxides induces significant fluctuations in the threshold voltage and transconductance when the gate oxide tunnel resistance becomes comparable to gate poly-Si resistance. By calculating the measured tunnel current based on multiple scattering theory, it is shown that the device characteristics fluctuations will be problematic when the gate oxide thickness is scaled down to less than 1 nm.

/pdf/limit-of-gate-oxide-thickness-scaling-in-mosfets-due-to-4yr2mwqu5d.pdf

Limit of gate oxide thickness scaling in MOSFETs due to apparent threshold voltage fluctuation induced by tunnel leakage current

Single crystals of cubic SiC that are free of antiphase domains were successfully grown by chemical vapor deposition on Si substrates inclined at 2° from (100) towards (011). The relationship between the generation of antiphase domains and the inclination of a surface was investigated by using spherically polished Si substrates. Inclination, except towards (011), resulted in the generation of antiphase domains. Elimination of antiphase domains was confirmed by molten KOH etching of the grown layer.

Kentaro Shibahara

Papers

Atomic layer controlled deposition of silicon nitride with self‐limiting mechanism

Physically-based threshold voltage determination for MOSFET's of all gate lengths

Limit of gate oxide thickness scaling in MOSFETs due to apparent threshold voltage fluctuation induced by tunnel leakage current

Antiphase-domain-free growth of cubic SiC on Si(100)

Atomic layer controlled deposition of silicon nitride and in situ growth observation by infrared reflection absorption spectroscopy