P-type (100) oriented CZ-grown silicon wafer (one side polished) having resistivity (1 - 20) Ω-cm was etched by HF-vapors in a reaction chamber (Figure 1) at STP under constant flow of HF-vapors. Vapors interact in a chamber with the sample placed on the surface of Teflon block. Out coming exhaust of vapors is dissolved in water to minimize the dangerous effects of HF-vapors. After etching samples were rinsed with de-ionized distilled water. They samples were found to emit red luminescence on exposure to UV light. Later scanning electron microscopy, thermo-gravimetric and pore size analysis were performed.

SEM, model JSM-5910 JEOL, was used to scan the sample surface. Micrographs of the samples were obtained at 20 K eV accelerating voltage and 500, 2000, 4000 and 120,000 magnifications.

Micrographs of controlled samples Figure 2(a), depicts rectangular shaped grains arranged in a regular pattern with roughly same dimensions of the boundaries. Sides of each grain appear to be either 10 or 20 µm. Top surface of every grain represents (100) plane.

Micrograph of etched samples recorded at 2000 and 4000 magnifications, Figure 2(b) and Figure 2(c) show a modified surface after etching, which is different as compared to surface of the samples shown in Figure 2(a). Rectangular grains and boundaries have totally disappeared and instead circular dots appeared on the surface. Width of the boundaries and diameter of the dots is roughly 1 µm and 2 µm respectively. These dots appeared to be lined-up in two different directions at 90˚ or 45˚ to each other indicating etched planes in either (100) or (110) directions. Etching appeared to be uniform over a wide range of the surface, Figure 2(a). Images of the etched surface recorded at 120,000 magnification gives very interesting result about the size of the sub-features on the surface of the dots. Surface of every dot appears to be covered with very small sub dots (bright spots in the figure). Roughly 16 dark and bright spots line up in 0.1 µm that yield 5 nm diameter of each sub dot, now termed as quantum dots.

The surface area and pore size of the pores etched in silicon have been measured by surface area and pore size

analyzer (Following two figures have not been discussed).

The detail of measurements is given below. The relative pressure has been measured and plotted against for unetched and etched samples. The plots of each sample are shown in Figure 3(a)Figure 3(b)). The porosity is measured from the slop of the best fit line through data points plotted in a graph as relative pressure verses for as grown and etched samples using following equations

[1] .

where, and. Assuming pores are cylindrical, the following equation is valid for the pore radius.

[20]

Pore radius calculated using above equations is roughly 6 nm.

The pore surface area, volume and diameter measured at relative pressure 2.94 × 10⁻² are listed below Surface area = 441.43 m²/g Pore value volume = 1.30 cc/g Pore diameter = 118.03 Å

Pore radius = 59 × 10⁻¹⁰ meter = 5.9 × 10⁻⁹ ≈ 6 nm.

After performing above characterization, thermo gravimetric/differential thermal analysis (TG/DTA) has been carried out to study the porosity of etched silicon. During measurements change in weight of the samples as a function of temperature was recorded by increasing the temperature of the samples gradually up to 635˚C (temperature; high enough to force hydrogen related species to disappear) and change in weight against temperature was plotted Figure 3(a), Figure 3(b). The samples and an inert reference are made to undergo identical thermal cycles. During oxidation the reaction is endothermic or exothermic (heat flows in/out relative to the inert reference) was also recorded. TG/DTA thermograms {Figure 3(a), Figure 3(b)} depict both change in mass and amount of energy which are plotted on y axis against temperature. Thermograms (temperature from −24˚C to 635˚C) for un-etched (Figure 4(a)) show maximum endothermic and exothermic heat flow at about 50˚C and 200˚C respectively. The curve also shows that heat flow is dominantly exothermic in the range 100˚C to 635˚C (positive slope) and endothermic (negative slope) in the range −24˚C to 100˚C. However loss of weight per degree centigrade during reduction and gain during oxidation is not very high, which is not more than 10⁻⁴ mg/˚C. In the above description we have considered zero line as a reference where heat flow is 0 mW. But in our opinion rate of mass loss or gain and decrease or increase rate of heat energy indicate the dominance of endothermic or exothermic reactions, which means if the slope of a part of the curve is negative reduction is dominant and incase slope is positive oxidation is dominant.

Thermogram for etched samples shown in Figure 4(b) indicates that reduction of the sample continued dominantly up to 200˚C, (negative slope from 0˚C to 100˚C and positive slope in range 100˚C - 200˚C. This reduction reaction indicates a significant amount of hydrogen related species on the surface of etched samples. The reaction for the remaining part of temperature is exothermic, which high lights dominance of oxidation of the samples over reduction. Thermogram for etched samples can be divided into four regions I, II, III and IV. In region I slope of the thermogram is negative and reduction of the samples continued dominantly up to 100˚C with mass reduction rate 4.3 × 10⁻³ mg/˚C. Reduction peaks at 100˚C and later, oxidation started to compensate reduction that converts the slope of the thermogram to positive. The rate of weight gain is 2 × 10⁻³ mg/˚C. At 200˚C reduction and oxidation balance each other (heat flow in or out is zero). In regions II, III and IV reaction is exothermic and oxidation of the samples is dominant. In regions II, III and IV mass gain rates are 3 ×10⁻³ mg/˚C, 2 × 10⁻³ mg/˚C and 4 × 10⁻³ mg/˚C respectively which point out different oxidation reactions in these regions. To check the reactivity either endothermic or exothermic temperature range can also be divided into same four ranges. In range I (0˚C - 100˚C) endothermic reactivity is dominant that goes on increasing up to 100˚C and approach to maximum at 100˚C. Afterwards in range 100˚C to 200˚C exothermic reactivity plays its role, as a result endothermic reactivity starts decreasing and at about 200˚C endothermic and exothermic reactivthe thermogram is negative in this region. In the region 100˚C to 200˚C heat energy partially starts to flow out ity balances each other. The rate of heat (energy) flow in between (0˚C - 100˚C) is 0.062 mW/K, while slope of and compensate heat flow in. Endothermic reactivity decreases as a result, slope (mW/K = 0.02) of thermogram is positive. In regions II, III, and IV reaction is exothermic and rates of heat energy flow out are 0.022, 0.014 and 0.012 mW/K respectively. This result also supports our opinion of different reaction in regions II, III, and IV and formation of different oxygen related species on the surface of etched silicon.