Particle pinholes

In comparison with sputtering, the energy of the particles which arrive at the substrate is one order of magnitude lower. This may favor the formation of pinholes,... 

One of the major problems with this type of interface, not unsurprisingly, is clogging of the pinhole. For this reason, the HPLC system has to be kept scrupulously clean with solvents being passed through narrow filters to remove any solid particles and in-line filters being incorporated to ensure that column material does not find its way into the probe. 

Common defects that are observed in this coating technique are voids, pin holes, thickness variations, and wavy surfaces. Voids and pinholes are caused by air entrapment in the film, surface contamination, or dust particles. 

Figure 10. Optical setup for LDA consisting of a laser beam being split and focused by a lens through a suspension of particles in a sample cell. Scattered light is focused through a pinhole and detected by a photomultiplier tube.

Once a satisfactory film has been obtained, the coated substrate should be transferred to a suitable storage container and moved directly to the subsequent processing step, viz., prebaking. It is important that unbaked, freshly spun films not be stored for periods of time in excess of a few hours at this stage, since many resists are particularly vulnerable to particulate contamination. Once particles contact a resist surface prior to prebaking, they are almost impossible to remove and will cause opaque spots or pinholes after exposure and development. 

As in all processing steps, cleanliness of the exposure hardware is of paramount importance. Any particle that lands on the resist prior to exposure, will shield the film underneath the particle from the exposing radiation and give rise to opaque spots in the case of positive resist, or pinholes in the case of negative resists. Particulate contamination is especially troublesome with electron beam and ion beam systems where the probability of a particle landing on a substrate is increased relative to other techniques because of the much longer exposure times involved. 

The connection between temperature and kinetic energy obtained from the kinetic-molecular theory makes it possible to calculate the average speed of a gas particle at any temperature. An important practical consequence of this relationship is Graham s law, which states that the rate of a gas s effusion, or spontaneous passage through a pinhole in a membrane, depends inversely on the square root of the gas s mass. 

This expression is identical to that calculated for w = 0 at short times, the nonexponential decay is the same as that of a free motion particle in the presence of a pinhole sink. 

The two Windscale piles were fuelled with natural uranium canned in aluminium. Coolant air was blown through the reactor and exhausted from a 120-m stack (Fig. 2.4). Filters were installed at the top of the stack, but were not very effective. Some fuel cans developed pinholes during operation, and others became damaged and lodged in the ducts behind the pile. It is estimated that about 20 kg of irradiated uranium were disseminated to atmosphere as oxide particles from these cans (Stather et al., 1986). The temperature of oxidation was 200-400°C. The particle size, measured at the top of the stack, showed a mass median diameter of 35 fim (Mossop, 1960). 

Up to this point, all of the films we have considered (Si02, Si3N4) were deposited under conditions such that they were amorphous. The only defects of interest were particles from the gas phase that might be incorporated into the growing film, or pinholes. Low-pressure CVD has reduced the incidence of particles, and thicker films can minimize the presence of pinholes. 

Figure 1. Schematic diagrams of TEB and LLS instrumentation. P, pinholes L, lenses B, polarizers C, cell Q, quarter wave plate PMT, photomultiplier tube HVG, high voltage generator MP, microprocessor TR, transient recorder CL, correlator CT, counter 6, scattering angle. For the TEB setup polarizers B-, B2 have polarization axis oriented at tt/4 with respect to the x-axis, as shown in (a). After the light beam passed through the cell with electric field in the x-direction containing a suspension of anisotropic particles and the quarter waveplate with its fast axis oriented at tt/4 with respect to the x-axis, the transmitted light beam is polarized in the direction of 71/4 + 6/2, as shown in (b). Analyzer B has polarization axis oriented at 3t/4 + a as shown in (c).

Second, the use of meshed particles versus a pressed wafer will typically lead to nonuniformity of X-ray absorption thickness. This can be directly observed by placing an X-ray sensitive camera behind the sample a sample of a powder pressed into a wafer is spectroscopically more uniform than a catalyst bed of meshed particles. Naturally the contrast becomes more extreme as the meshed particles become larger. Moreover, if the sample is spatially nonuniform then severe constraints are placed on the positional stability of the X-ray beam. Any motion of the position of the X-ray beam will then probe different thicknesses of the sample, with direct consequences on the measured S/N. From the perspective of XAFS spectroscopy, any nonuniformity of the sample thickness could directly affect the accuracy of the measurement of the amplitude of the X-ray absorption coefficient. It is the amplitude that contains information about the coordination number and site disorder. As has been discussed elsewhere (Koningsberger and Prins, 1988), these amplitude distorting effects are given the general heading of "thickness effects." In brief, a thickness effect occurs when part of the incident X-ray beam is not attenuated by the sample. In the case of meshed particles this would be in the form of pinholes in the sample. 

The Lark counter [94] was designed for measuring particles in a loose powder, recording the sizes by the use of a series of pinholes in a chart. By ruling lines on the chart corresponding to a scale of sizes, and counting the pinholes between them, the size distribution in terms of the Feret diameter could be determined at a rate of about a thousand particles per hour. 

---