Lli

■ » 1 > ■ 1 * 1 ■ ' ■ 1 1 ' ■ ' ■ 1 ■

450 500 550 600 650 700 WAVELENGTH (nm)

Figure 6.17. Optical absorption spectrum of 20- and 80-nm gold nanoparticles embedded in glass. (Adapted from F. Gonella et al., in Handbook of Nanostwctured Materials and Nanotechnology, H. S. Nalwa, ad., Academic Pre», San Diego, 2000, Vol. 4, Chapter 2, p. 85.)

the optical absorption shifts to shorter wavelengths when the nanoparticle size decreases from 80 to 20 nm. The spectrum is due to plasma absorption in the metal nanoparticles. At very high frequencies the conduction electrons in a metal behave like a plasma, that is, like an electrically neutral ionized gas in which the negative charges are the mobile electron, and the positive charges reside on the stationary background atoms. Provided the clusters are smaller than the wavelength of the incident visible light, and are well dispersed so that they can be considered non-interacting, the electromagnetic wave of the light beam causes an oscillation of the electron plasma that results in absorption of die light. A theory developed by Mie may be used to calculate the absorption coefficient versus the wavelength of the light. The absorption coefficient a of small spherical metal particles embedded in a nonabsoibing medium is given by nnNsVn04/l , where N, is the number of spheres of volume V, e j and e2 are the real and imaginary parts of die dielectric constant of the spheres, n0 is the refractive index of the insulating glass, and A is die wavelength of the incident light

Another technologically important property of metallic glass composites is that they display nonlinear optical effects, which means that their refractive indices depend on the intensity of the incident light. The glasses have an enhanced third-order susceptibility that results in an intensity dependent refractive index n given by n = n0 + n2I (6.9)

where / is the intensity of the light beam. Nonlinear optical effects have potential application as optical switches, which would be a major component of photon-based computers. When metal particles are less than lOnm in size, confinement effects become important, ¡aid these alter the optical absorption properties. Quantum confinement is discussed in Chapter 9.

The earliest methods for making composite metal glasses involve mixing metal particles m molten glasses. However, it is difficult to control the properties of the glasses, such as the aggregation of the particles. More controllable processes have been developed such as ion implantation. Essentially, the glasses are subjected to an ion beam consisting of atoms of the métal to be implanted, having energies in the range from lOkeV to lOMeV. Ion «change is also used to put metal particles into glasses. Figure 6.18 shows an experimental setup for an ion exchange process designed to put silver particles m glasses. Monovalent surface atoms such as sodium present near the surface of all glasses are replaced with other ions such as silver. The glass substrate is placed in a molten salt bath that contains die electrodes, and a voltage is applied across the electrodes with the polarity shown in Fig. 6.18. The sodium ion diffuses in the glass toward die negative electrode, and the silver diffuses from the silver electrolyte solution into the surface of the glass.

V, battery

positive electrode substrate negative electrode cM

salt bath

Figure 6.18. Electric field assisted ion exchange apparatus for doping glasses (substrate) with metals such as Ag+ ions. [Adapted from G. De Marchi et a)., J. Non-Cryst Solids 196, 79 (1996).}

6.1.8. Porous Silicon

When a wafer of silicon is subjected to electrochemical etching, the silicon wafer develops pores. Figure 6.19 is a scanning electron microscope picture of the (100) surface of etched silicon showing pores (dark regions) of micrometer dimensions.

Figure 6.19. Scanning electron microscope (SEM) picture of the surface of n-doped etched silicon. The micrometer-sized pores appear as dark regions. [With permission from C. Levy-Clement et al., J. Electrochem. Soc. 141, 958 (1994).]

This silicon is called porous silicon (PoSi). By controlling die processing conditions, pores of nanometer dimensions can be made. Research interest in porous silicon was intensified in l990 when it was discovered that it was fluorescent at room temperature. Luminescence refers to the absorption of energy by matter, and its re-emission as visible or near-visible light. If the emission occurs within 10-8s of die excitation, then the process is called fluorescence, and if there is a delay in the emission it is called phosphorescence. Nonporous silicon has a weak fluorescence between 0.96 and 1.20 eV in die region of the band gap, which is 1.125 eV at 300 K. This fluorescence is due to band gap transitions in the silicon. However, as shown in Fig. 6.20, porous silicon exhibits a strong photon-induced luminescence well above 1.4 eV at room temperature. The peak wavelength of the emission depends on the length of time the wafer is subjected to etching. This observation generated much excitement because of the potential of incorporating photoactive silicon using current silicon technology, leading to new display devices or optoelectronic coupled elements. Silicon is die element most widely used to make transistors, which are the on/off switching elements in computers.

Figure 6.21 illustrates one method of etching silicon. Silicon is deposited on a metal such as aluminum, which forms the bottom of a container made of polyethylene or Teflon, which will not react with the hydrogen fluoride (HF) etching solution. A voltage is applied between the platinum electrode and the Si wafer such that the Si is the positive electrode. The parameters that influence the nature of the pores are the concentration of HF in the electrolyte or etching solution, the

Wavelength (/tm)

Figure 6.20. Photoluminescence spectra of porous silicon for two different etching times at room temperature. Note the change in scale for the two curves. [Adapted from L. T. Camham, Appt. Phys. Lett. 57, 1046 (1990).]

Wavelength (/tm)

Figure 6.20. Photoluminescence spectra of porous silicon for two different etching times at room temperature. Note the change in scale for the two curves. [Adapted from L. T. Camham, Appt. Phys. Lett. 57, 1046 (1990).]

Pt Electrode

Backside si Wafer

Electrical Connection

Si Wafer

Figure 6.21. A cell for etching a silicon wafer in a hydrogen fluoride (HF) solution in order to introduce pores. (With permission from D. F. Thomas et al., in Handbook of Nanostructured Materials and Nanolechnoiogy, H. S. Nalwa, ed.. Academic Press, San Diego, 2000, Vol. 4, Chapter 3, p. 173.)

Pt Electrode

\ Etching Solution,

Backside si Wafer

Electrical Connection

Si Wafer

Figure 6.21. A cell for etching a silicon wafer in a hydrogen fluoride (HF) solution in order to introduce pores. (With permission from D. F. Thomas et al., in Handbook of Nanostructured Materials and Nanolechnoiogy, H. S. Nalwa, ed.. Academic Press, San Diego, 2000, Vol. 4, Chapter 3, p. 173.)

magnitude of current flowing through the electrolyte, the presence of a surfactant (surface-active agent), and whether the silicon is negatively (n) or positively (p) doped.

The Si atoms of a silicon crystal have four valence elections, and are bonded to four nearest-neighbor Si atoms. If an atom of silicon is replaced by a phosphorus atom, which has five valence electrons, four of the electrons will participate in the bonding with the four neighboring silicon atoms. This will leave an extra electron available to carry current, and thereby contribute to the conduction process. This puts an energy level in the gap just below the bottom of the conduction band. Silicon doped in this way is called an n-type semiconductor. If an atom of aluminum, which has three valence electrons, is doped into the silicon lattice, there is a missing electron referred to as a hole in one of the bonds of die neighboring silicon atoms. This hole can also carry current and contribute to increasing die conductivity. Silicon doped in this manner is called a p-type semiconductor. It turns out that the size of the pores produced in the silicon is determined by whether silicon is n- or p-type. When p-type silicon is etched, a very fine network of pores having dimensions less than lOnm is produced.

A number of explanations have been offered to explain the origin of the fluorescence of porous silicon, such as the presence of oxides on the surface of the pores that emit molecular fluorescence, surface defect states, quantum wires, quantum dots and the resulting quantum confinement, and surface states on quantum dots. Porous silicon also displays electroluminscence, whereby the luminescence is induced by the application of a small voltage across electrodes mounted on the silicon, and cathodoluminescence from bombarding electrons.

0 0

Post a comment