Crystalline Defects in Silicon

Like anything else in this world, crystals inherently possess imperfections, or what we often refer to as 'crystalline defects'. The presence of most of these crystalline defects is undesirable in silicon wafers, although certain types of 'defects' are essential in semiconductor manufacturing. Engineers in the semiconductor industry must be aware of, if not knowledgeable on, the various types of silicon crystal defects, since these defects can affect various aspects of semiconductor manufacturing - from production yields to product reliability.

Crystalline defects may be classified into four categories according to their geometry. These categories are: 1)  zero-dimensional or 'point' defects; 2) one-dimensional or 'line' defects; 3) two-dimensional or 'area' defects; and 4) three-dimensional or 'volume' defects. Table 1 presents the commonly-encountered defects under each of these categories.

Table 1. Examples of Crystalline Defects

There are many forms of crystal point defects. A defect wherein a silicon atom is missing from one of these sites is known as a 'vacancy' defect.  If an atom is located in a non-lattice site within the crystal, then it is said to be an 'interstitial' defect.  If the interstitial defect involves a silicon atom at an interstitial site within a silicon crystal, then it is referred to as a 'self-interstitial' defect.  Vacancies and self-interstitial defects are classified as intrinsic point defects.

If an atom leaves its site in the lattice (thereby creating a vacancy) and then moves to the surface of the crystal, then it becomes a 'Schottky' defect.  On the other hand, an atom that vacates its position in the lattice and transfers to an interstitial position in the crystal is known as a 'Frenkel' defect. The formation of a Frenkel defect therefore produces two defects within the lattice - a vacancy and the interstitial defect, while the formation of a Schottky defect leaves only one defect within the lattice, i.e., a vacancy. Aside from the formation of Schottky and Frenkel defects, there's a third mechanism by which an intrinsic point defect may be formed, i.e., the movement of a surface atom into an interstitial site.

Extrinsic point defects, which are point defects involving foreign atoms, are even more critical than intrinsic point defects. When a non-silicon atom moves into a lattice site normally occupied by a silicon atom, then it becomes a 'substitutional impurity.'   If a non-silicon atom occupies a non-lattice site, then it is referred to as an 'interstitial impurity.' Foreign atoms involved in the formation of extrinsic defects usually come from dopants, oxygen, carbon, and metals.

The presence of point defects is important in the kinetics of diffusion and oxidation.  The rate at which diffusion of dopants occurs is dependent on the concentration of vacancies. This is also true for oxidation of silicon.

Crystal line defects are also known as 'dislocations', which can be classified as one of the following: 1) edge dislocation; 2) screw dislocation; or 3) mixed dislocation, which contains both edge and screw dislocation components. 

An edge dislocation may be described as an extra plane of atoms squeezed into a part of the crystal lattice, resulting in that part of the lattice containing extra atoms and the rest of the lattice containing the correct number of atoms.  The part with extra atoms would therefore be under compressive stresses, while the part with the correct number of atoms would be under tensile stresses. The dislocation line of an edge dislocation is the line connecting all the atoms at the end of the extra plane.

Figure 1.  An edge dislocation; note the insertion

of atoms in the upper part of the lattice

If the dislocation is such that a step or ramp is formed by the displacement of atoms in a plane in the crystal, then it is referred to as a 'screw dislocation.'  The screw basically forms the boundary between the slipped and unslipped atoms in the crystal. Thus, if one were to trace the periphery of a crystal with a screw dislocation, the end point would be displaced from the starting point by one lattice space. The dislocation line of a screw dislocation is the axis of the screw.

Figure 2.  A screw dislocation; note the screw-like

'slip' of atoms in the upper part of the lattice

If the dislocation consists of an extra plane of atoms (or a missing plane of atoms) lying entirely within the crystal, then the dislocation is known as a 'dislocation loop.'  The dislocation line of a dislocation loop forms a closed curve that is usually circular in shape, since this shape results in the lowest dislocation energy.

Dislocations are generally undesirable in silicon wafers because they serve as sinks for metallic impurities as well as disrupt diffusion profiles.  However, the ability of dislocations to sink impurities may be engineered into a wafer fabrication advantage. i.e., it may be used in the removal of impurities from the wafer, a technique known as 'gettering.'

Area defects in crystals consist of stacking faults, grain boundaries, and twin boundaries. A 'stacking fault' pertains to a disturbance in the regularity of the stacking of planes of atoms in a crystal lattice.  This usually occurs when a plane is inserted into or removed from the lattice. The insertion of an extra plane in the stacking is known as an 'extrinsic' stacking fault, while the removal of a plane is referred to as an 'intrinsic' stacking fault.

Stacking faults can become electrically active when decorated by impurity atoms. Electrically active stacking faults can cause device degradation, examples of which are higher reverse bias currents in p-n junctions and storage time reduction in MOS circuits.

Figure 1.  Photo of a Stacking Fault

Image Source: http://lmass.uah.edu

- J. A. Gavira-Gallardo, J. D. Ng and M.A. George      

A 'twin' is an area defect wherein a mirror image of the regular lattice is formed during the growth of the silicon ingot, usually caused by a perturbation.  The 'twin boundary' is the mirror plane of the twin formation.

A  'grain boundary' refers to the transition or interface between crystals whose atomic arrangements are different in orientation with respect to each other.

Volume defects in a crystal are also known as 'bulk' defects, which include voids and precipitates of extrinsic and intrinsic point defects.

Every impurity introduced into a crystal has a certain level of solubility, which defines the concentration of that impurity that the solid solution of the host crystal can accommodate.  Impurity solubility usually decreases with decreasing temperature.

If an impurity is introduced into a crystal at the maximum concentration allowed by its solubility at a high temperature, the crystal will become supersaturated with that impurity once it is cooled down.  A crystal under such supersaturated conditions seeks and achieves equilibrium by precipitating the excess impurity atoms into another phase of different composition or structure.  

Precipitates are considered undesirable because they have been known to act as sites for the generation of dislocations.  Dislocations arise as a means of relieving stress generated by the strain exerted by precipitates on the lattice.  Precipitates induced during silicon wafer processing come from oxygen, metallic impurities, and dopants like boron.

See Also:  Crystal Defect Effects;  Incoming Wafers;  Epitaxy;  Polysilicon;  

Ion Implant;  Gettering;  Crystal Growing

HOME

Copyright © 2004-Present www.EESemi.com. All Rights Reserved.