Monday, October 16, 2006

Ion Implantation Process in Semiconductor Manufacturing

For performing ion implantation, atoms or molecules are ionized, accelerated in an electric field and implanted into the target material. A wide variety of combinations of target material and implanted ions are possible. The dose of the implanted ions can vary between 1011 and 1018 cm-2. Usually, the acceleration energy lies between several keV and several hundred keV, but with special equipment it is possible to work with energies up to several MeV. The range of the implanted ions in the substrate depends on the mass of the implanted ions, their energy, the mass of the substrate atoms, crystal structure and the direction of incidence. As an example, the mean range of 100 keV phosphorus ions in silicon is about 150 nm.
Main advantages of ion implantation (in comparison to diffusion) for the doping of semiconductors are:
Short process times, good homogeneity and reproducibility of the profiles
Exact control of the amount of implanted ions by integrating the current. This is of particular importance for low concentrations, e.g. for adjusting the threshold voltage of MOS transistors
Relatively low temperatures during the process
Different materials can be used for masking, e.g. oxide, nitride, metals, and resist
Implantation through thin layers (e.g. SiO2 or Si3N4) is possible
Low penetration depth of the implanted ions. This allows modification of thin areas near the surface with high concentration gradients
Sequences of implantation steps (with different energies and doses) allow optimization of the dopant profiles
There are also some disadvantages, such as:
Damage of the substrate is caused by the implanted ions
The change of material properties is restricted to the substrate domains close to the surface
Additional effects during or after implantation (such as channeling or diffusion) make it difficult to achieve very shallow profiles and to theoretically predict the exact profile shapes.

Diffusion Process in Semiconductor Manufacturing

Diffusion of impurity atoms in silicon during processing is important for the electrical characteristics of silicon devices. Various ways of introducing dopants into silicon by diffusion are used and have been studied with the goals of controlling dopant distribution, total dopant concentration, uniformity, and reproducibility.
Diffusion is used to form base, emitter, and collector regions in bipolar device processing, to form source, drain and channel regions, and to dope polysilicon in MOS processing. Dopant atoms that span a wide range of concentrations can be introduced into silicon in many ways. The most commonly used methods are
Diffusion from a chemical source in a vapor form at high temperatures,
Diffusion from a doped-oxide source, and
Diffusion and annealing from an ion-implanted layer (see ion implantation).
The electrical activation of ion-implanted species is carried out by annealing. This causes a redistribution of the impurity atoms which should be as low as possible. In order to optimize the electrical behavior of the device, it is important to know how the impurities redistribute during the anneal. The development of appropriate models and simulation programs to predict the diffusion is one major topic in semiconductor technology research.

Etching Process in Semiconductor Manufacturing

Etching processes are used to partly remove material in order to create patterns to obtain the desired device or interconnect geometry. Particles in the etching component (a liquid or gas) remove material by attacking the open surface.
The material may be isotropically removed, known as chemical or "wet" etching, or be etched in a plasma reactor ("dry"-etching). In the case of a dry-etching process, the total etch rate consists of an ion-assisted rate and a purely chemical etch rate due to etching by neutral radicals, which may still have a directional component. The total etch rate can depend on shadowing within the reactor and by the structure on the substrate itself, the angle-dependent flux distribution of particles from the reactor volume, the angle of incidence of the particles relative to the surface normal direction, reflection/re-emission of etching particles, and surface diffusion effects. Reactive ion etching (RIE) provides high anisotropy which is achieved by etching that is enhanced (via different mechanisms) by ions impinging onto the surface. The main advantage of RIE is enhanced directionality which becomes increasingly important as device sizes decrease substantially and etching must proceed in vertical direction without affecting adjacent features.

Lithography in Semiconductor Manufacturing

Lithography is the process of transferring geometric shapes on a mask to the surface of a silicon wafer. These shapes make up the parts of the circuit, such as gate electrodes, contact windows, metal interconnections, an so on. The final integrated circuit (IC) is made by sequentially transferring the features from each mask level by level, to the surface of the silicon wafer. For example, between each successive image transfer an ion implant, drive-in, oxidation, or metallization operation may take place.
In the IC lithographic process, a photosensitive polymer film is applied to the silicon wafer, dried, and then exposed with the proper geometrical patterns through a photomask to ultraviolet (UV) or other radiation. After exposure, the wafer is brought into contact (e.g. by dipping or spraying) with a solution that develops the images in the photosensitive material. Depending on the type of polymer used, either exposed (in the case of positive resists) or non-exposed (for negative resists) areas of the film are removed in the developing process. After development the resist acts as a mask, for instance for etching patterns into underlying layers.
Resists are made that are sensitive to UV light, electron beams, or ion beams. At present, optical lithography is the prevailing method. In the figure below, an example for the application of resist on the wafer is shown.

Deposition Process in Semiconductor Manufacturing

For the deposition of conducting and insulating materials in semiconductor processing, a wide range of different techniques is used. Among them, physical vapor deposition (PVD) and chemical vapor deposition (CVD) are the most important ones.
Materials in semiconductor manufacturing that can be deposited by CVD are for example: Silicon dioxide, silicon nitride, polysilicon, tungsten, copper, titanium nitride, aluminum.
Examples for CVD processes
Materials in semiconductor manufacturing that can be deposited using PVD are for example: aluminum, titanium, titanium nitride, tantalum, copper.
Sputter reactor
Usually the conformality (perfect conformality means, that for all positions at the device surface the same layer thickness is achieved) of CVD layers is much better than the conformality of PVD layers. This is demonstrated by the 2 movies: the first one shows a 3D simulation of a sputter deposition of aluminum into a contact hole, the second one shows a 3D simulation of tungsten CVD into the same contact hole. The CVD process yields better layer conformality than the PVD process. The 3D simulations shown in the movies have been carried out using the topography simulator SC-TOP from the software house SIGMA-C. Within SC-TOP, a 3D module for layer deposition developed at FhG-IIS-B is used.

Oxidation Process in Semiconductor Manufacturing

Today's ULSI technology is to a large extent based on the excellent properties of thermally grown silicon dioxide layers. SiO2 is used as gate dielectric in MOS devices, as implantation or doping mask, and for device isolation purposes.
Most of the silicon oxidation processes today are carried out at atmospheric pressure in oxidation furnaces at temperatures between 800 and 1,200°C. The appropriate oxidant gases are dry oxygen, wet oxygen (water steam). For wet oxidation, the water steam can be extracted by influx of gases (oxygen, argon, or nitrogen) into a vessel filled with water. Another possibility is to initiate a reaction of oxygen and hydrogen at the inlet of the reactor tube. The wafers are placed vertically or horizontally in quartz boats and the temperature of the hot zone is controlled with an accuracy of 1°C, in order to grow oxide layers in a reproducible manner.
It has been proven by a number of experiments that thermal oxidation of silicon proceeds by diffusion of the oxidant through the growing oxide. The oxidation reaction itself takes place at the oxide-silicon interface. During oxidation, the oxide-silicon interface moves into the silicon material as the silicon is oxidized. Taking into account the densities of silicon and of SiO2, it can be shown that about 44% of the oxide layer grows into the silicon substrate. The remaining 56% grows on top of the silicon, resulting in a non-planar surface if oxidation is local.
The lateral and vertical isolation of device structures in ULSI determines to a large extent the performance of the technology with respect to packaging density and parasitic effects. Classical local oxidation (LOCOS) has been widely used to laterally isolate the active regions of an integrated circuit. The oxide is grown at those parts of the wafer surface where a Si3N4 masking layer has its openings. Before starting the oxidation there is already a thin oxide layer in between the nitride layer and the silicon substrate. Because of diffusion of the oxidant into the region below the nitride, also the oxide beneath the nitride is growing leading to the characteristic bird's beak shape after completion of the process. This situation is illustrated in the Figure 1, which shows the result of an oxidation in wet oxygen at 950°C for 10 hours(from: C. Claeys, J. Vanhellemont, G. Declerck, J. Van Landuyt, R. Van Overstraeten, S. Amelinckx, VLSI Science and Technology/1984, K.E. Bean, G. Rozgoni, Eds., The Electrochemical Society, Pennington, 1984, p. 272.).

Wafer Fabrication

The first step in wafer production is the fabrication of a silicon single crystal. A small crystal, called the seed crystal, is rotated and slowly withdrawn from the molten hyper-pure silicon to give a cylindrical crystal. The silicon crystal is sliced to obtain thin disks, known as wafers on which chips are made, hundreds at a time, during the following processing. After slicing, the wafers are finely ground, mirror-smooth polished and cleaned.
The most common semiconductor material used today is silicon. In addition to silicon, numerous other materials have been studied, and gallium arsenide (GaAs) is used instead of silicon in some applications at present (e.g. high-frequency devices).
All processes require extremely clean rooms with filtered air to have a low concentration of particles of dust or other contaminants so that a high chip yield can be attained.

How Semiconductors Work?

Semiconductors have had a monumental impact on our society. You find semiconductors at the heart of microprocessor chips as well as transistors. Anything that's computerized or uses radio waves depends on semiconductors.
Today, most semiconductor chips and transistors are created with silicon. You may have heard expressions like "Silicon Valley" and the "silicon economy," and that's why -- silicon is the heart of any electronic device. A diode is the simplest possible semiconductor device, and is therefore an excellent beginning point if you want to understand how semiconductors work. In this article, you'll learn what a semiconductor is, how doping works and how a diode can be created using semiconductors.
Silicon is a very common element -- for example, it is the main element in sand and quartz. If you look "silicon" up in the
periodic table, you will find that it sits next to aluminum, below carbon and above germanium.
Carbon, silicon and germanium (germanium, like silicon, is also a semiconductor) have a unique property in their electron structure -- each has four electrons in its outer orbital. This allows them to form nice crystals. The four electrons form perfect covalent bonds with four neighboring
atoms, creating a lattice. In carbon, we know the crystalline form as diamond. In silicon, the crystalline form is a silvery, metallic-looking substance.
Metals tend to be good conductors of electricity because they usually have "free electrons" that can move easily between atoms, and electricity involves the flow of electrons. While silicon crystals look metallic, they are not, in fact, metals. All of the outer electrons in a silicon crystal are involved in perfect covalent bonds, so they can't move around. A pure silicon crystal is nearly an insulator -- very little electricity will flow through it.
You can change the behavior of silicon and turn it into a conductor by doping it. In doping, you mix a small amount of an impurity into the silicon crystal.
There are two types of impurities:
N-type - In N-type doping,
phosphorus or arsenic is added to the silicon in small quantities. Phosphorus and arsenic each have five outer electrons, so they're out of place when they get into the silicon lattice. The fifth electron has nothing to bond to, so it's free to move around. It takes only a very small quantity of the impurity to create enough free electrons to allow an electric current to flow through the silicon. N-type silicon is a good conductor. Electrons have a negative charge, hence the name N-type.
P-type - In P-type doping,
boron or gallium is the dopant. Boron and gallium each have only three outer electrons. When mixed into the silicon lattice, they form "holes" in the lattice where a silicon electron has nothing to bond to. The absence of an electron creates the effect of a positive charge, hence the name P-type. Holes can conduct current. A hole happily accepts an electron from a neighbor, moving the hole over a space. P-type silicon is a good conductor. A minute amount of either N-type or P-type doping turns a silicon crystal from a good insulator into a viable (but not great) conductor -- hence the name "semiconductor."
N-type and P-type silicon are not that amazing by themselves; but when you put them together, you get some very interesting behavior at the junction.
A diode is the simplest possible semiconductor device. A diode allows current to flow in one direction but not the other. You may have seen turnstiles at a stadium or a subway station that let people go through in only one direction. A diode is a one-way turnstile for electrons.
When you put N-type and P-type silicon together as shown in this diagram, you get a very interesting phenomenon that gives a diode its unique properties.
Even though N-type silicon by itself is a conductor, and P-type silicon by itself is also a conductor, the combination shown in the diagram does not conduct any electricity. The negative electrons in the N-type silicon get attracted to the positive terminal of the battery. The positive holes in the P-type silicon get attracted to the negative terminal of the battery. No current flows across the junction because the holes and the electrons are each moving in the wrong direction.
If you flip the battery around, the diode conducts electricity just fine. The free electrons in the N-type silicon are repelled by the negative terminal of the battery. The holes in the P-type silicon are repelled by the positive terminal. At the junction between the N-type and P-type silicon, holes and free electrons meet. The electrons fill the holes. Those holes and free electrons cease to exist, and new holes and electrons spring up to take their place. The effect is that current flows through the junction.
A device that blocks current in one direction while letting current flow in another direction is called a diode. Diodes can be used in a number of ways. For example, a device that uses batteries often contains a diode that protects the device if you insert the batteries backward. The diode simply blocks any current from leaving the battery if it is reversed -- this protects the sensitive electronics in the device.
When reverse-biased, an ideal diode would block all current. A real diode lets perhaps 10 microamps through -- not a lot, but still not perfect. And if you apply enough reverse voltage (V), the junction breaks down and lets current through. Usually, the breakdown voltage is a lot more voltage than the circuit will ever see, so it is irrelevant.
When forward-biased, there is a small amount of voltage necessary to get the diode going. In silicon, this voltage is about 0.7 volts. This voltage is needed to start the hole-electron combination process at the junction.
A transistor is created by using three layers rather than the two layers used in a diode. You can create either an NPN or a PNP sandwich. A transistor can act as a switch or an amplifier.
A transistor looks like two diodes back-to-back. You'd imagine that no current could flow through a transistor because back-to-back diodes would block current both ways. And this is true. However, when you apply a small current to the center layer of the sandwich, a much larger current can flow through the sandwich as a whole. This gives a transistor its switching behavior. A small current can turn a larger current on and off.
A silicon chip is a piece of silicon that can hold thousands of transistors. With transistors acting as switches, you can create
Boolean gates, and with Boolean gates you can create microprocessor chips.
The natural progression from silicon to doped silicon to transistors to chips is what has made microprocessors and other electronic devices so inexpensive and ubiquitous in today's society. The fundamental principles are surprisingly simple. The miracle is the constant refinement of those principles to the point where, today, tens of millions of transistors can be inexpensively formed onto a single chip.