Silicate Thermal Insulation Material from Rice Hull Ash

MATERIALS AND INTERFACES.

Silicate Thermal Insulation Material from Rice Hull Ash

 Uruthira Kalapathy, Andrew Proctor and John Shultz

  Department of Food Science, University of Arkansas, Fayetteville, Arkansas 72704, and Arkansas Analytical Laboratory, High-Density Electronics Center, University of Arkansas, Fayetteville, Arkansas 72701

Silicate blocks were produced from rice hull ash (RHA) derived sodium silicate, and physical properties and thermal conductivity of the blocks were investigated. A Sodium silicate solution was produced from RHA by solublizing the silica in 1 M NaOH at 100 °C, followed by filtering to remove the carbon residue. The silicate solution was then boiled at different heating rates to evaporate the water and produce highly porous thermal insulation blocks with varying densities. X-ray diffraction patterns demonstrated the amorphous nature of silicate blocks. FTIR spectra demonstrated the presence of the silicate network in the silicate. Block density, mechanical strength (the compressive force per unit area required to break the block), and thermal conductivity decreased with increasing heating rate. The block density, mechanical strength, and thermal conductivity of silicate blocks were 0.33 - 0.42 g ‚ cm - 3 , 145 - 196 N ‚ cm - 2 , and 0.103 - 0.128 W ‚ m - 1‚ K - 1 , respectively. Data suggested that RHA could be used to produce thermal insulation materials as a substitute for asbestos and organic polymer materials. Introduction The rice hull comprises 20% of the rice kernal grain and is a major coproduct of the rice industry. Rice hulls are burned commercially to produce rice hull ash (RHA) as an energy source and to minimize the hull disposal problem. RHA is marketed as high-carbon and low- carbon products with about 60 and 80% silica, respec- tively. The RHA silica is unique because it exists in a soluble amorphous form and can be readily extracted as silicate by alkali at 90 - 100 °C. 1,2 Silicate films 3 and xerogels 4 have been produced from RHA silicate extract, and a simple low-energy chemical method has been developed to produce pure silica (93%) containing 2.6% moisture from RHA at 91% extraction yield. 5 This study explores the potential of RHA as a raw material to produce a safe substitute for asbestos. Asbestos is the fibrous silicate mineral that has been used as a thermal insulator in high-temperature ap- plications. 6 Strict environmental regulations and public health threats posed by asbestos have prompted the search for new materials as asbestos substitutes. Sev- eral silicate-based materials such as glass fiber, mineral wool, and ceramic fibers have been used as asbestos substitutes. 6 However, current manufacturing processes of insulation materials from conventional crystalline silica have high energy costs because of the use of a smelting process at about 1300 °C. 1,6 Amorphous RHA silica can be used as a viable alternative raw material to produce silicates at modest temperatures. 2,7 RHA silica reacts with sodium hydroxide to form a sodium silicate solution. Solid disk or cube silica blocks can be produced from the silicate solution by lowering the pH to form silica xerogel, which is then pressed into molds and dried. 4 The density and mechanical proper- ties of xerogel blocks produced from RHA vary depend- ing on the pH at which xerogels were formed and the silica concentration of the silicate solution. Highly porous material is required for thermal insulation. Because porosity is generally inversely related to bulk density, it is necessary to produce silica gels with very low density. However, xerogels lack structural integrity and mechanical strength and have high densities rang- ing from 1.25 to 1.44 g ‚ cm -3 . Hence, silica xerogels may not be suitable for thermal insulation application. The objective of this research was to produce low- density sodium silicate blocks from amorphous RHA silica and to investigate their structure, mechanical strength, and thermal conductivity. Experimental Section RHA consisting of 87% silica and 10% carbon was obtained from Producers Rice Mill (Stuttgart, AR) and used for this study. Silicate Block Production. Silica was extracted as sodium silicate from RHA using 1 M NaOH. RHA (100 g) and 1 M NaOH (500 mL) were boiled in a covered Erlenmeyer flask for1htoproduce sodium silicate. The solution was filtered through Whatman No. 41 filter paper to remove the carbon residue. 5 The volume of this silicate solution was reduced to 250 mL by boiling. Portions (125 mL) of silicate solution were transferred into 400-mL stainless steel beakers and heated on a * Corresponding author. Current address: Department of Chemistry and Biochemistry, Claflin University, 400 Magnolia Street, Orangeburg, SC 29115. Tel: 803-535-5448. Fax: 803- 535-5776. E-mail: ukalapat@claflin.edu. † Department of Food Science, University of Arkansas. ‡ Arkansas Analytical Laboratory, High-Density Electronics Center, University of Arkansas. 46 Ind. Eng. Chem. Res. 2003, 42, 46 - 49 10.1021/ie0203227 CCC: $25.00 © 2003 American Chemical Society Published on Web 12/10/2002 hotplate (Lab-Line 1268, Pyro-Multi-Magnesstir) using a low heating rate (setting 3) to form solid disk-shaped blocks by evaporating the water. Blocks were also made using the medium (setting 5) and high (setting 8) heating rates to evaporate the water. The heating durations for low, medium, and high rate settings were about 8, 6, and 4 h, respectively. The different heating settings/durations were used to produce silicate blocks with varying densities. The blocks were removed from the stainless steel beakers and sanded (using 300 grit silicon carbide papers) to obtain disks that were 10 mm in height and 52 mm in diameter. Structural Analyses. X-ray diffraction (XRD) pat- terns were obtained by a dual goniometer X’pert XRD system (Philips Electronic Instruments, Eindhoven, The Netherlands) using a Cu target with an acceleration voltage of 45 kV and current of 40 mA. The diffraction angle (2θ, the angle between the incident and diffracted beams of X-rays) was scanned from 10 to 50°, at a rate of 2.4°/min. Diffuse reflectance Fourier transform in- frared spectra (DRIFTS) of silicates were obtained by adding 100 inteferograms at a resolution of 8.0 cm -1 using an Impact 410 Nicolet instrument (Nicolet Ana- lytical Instruments, Madison, WI). All DRIFTS were obtained using ground silicates in sample holders. For all measurements, background spectra were obtained using a blank disk. All of the DRIFTS were collected and corrected for background using Omnic 4.1 software (ThermoNicolet, Madison, WI). Block Density Measurement. Densities of silicate blocks were determined from their weights and volumes. Block weights were recorded in grams ( ( 0.0005 g) using a Mettler balance. Volumes of the blocks were calculated from the heights and diameters of the disks measured in millimeters ( ( 0.02 mm) using a stainless steel dial caliper. Mechanical Strength Determination. Mechanical strengths of blocks were determined using an Instron Universal Analyzer (model 4046, Instron Corp., Canton, MA). A cylindrical probe with a 24-cm 2 cross-head area at a loading rate of 25 mm/min was used in the compression mode to break the blocks that were pre- pared as described above. Because the silicate blocks were very brittle, the blocks were crushed into very small pieces at break point. The peak force at break point was recorded. The peak force per unit area (N ‚ cm -2 ) was calculated by dividing the peak force by the cross-head area of the probe and used as a measure of the mechanical strength of silicate blocks. Thermal Conductivity Measurement. Silicate disks were machined (Netzsch Instruments Inc., Testing Services, Bedford, MA) to obtain a uniform thickness of 6 mm. The thermal conductivity was measured at a nominal mean temperature of 50 °C by the guarded heat flowmeter method (as described in ASTM Practice F133) using a thermal conductivity analyzer (Holometrix model TCA-300, Netzsch Instruments Inc., Bedford, MA). Statistical Analyses. Block density, mechanical strength, and thermal conductivity measurements were performed in triplicate. Data were analyzed by ANOVA, and means were separated by the least significant difference test when significant F (P < 0.05) values were observed. 8 Results and Discussion Heating Rates. Silicate blocks with varying densities were made by using the low (setting 3), medium (setting 5), and high (setting 8) heating rates to evaporate the water. When silicate solutions were heated at low, medium, and high heating rates, the solution temper- ature reached 100 °C in 31, 27, and 22 min, respectively. Silica Gel Structure. XRD patterns of the silicate blocks were similar and independent of the rate of heating used to produce these blocks (Figure 1a - c). The broad XRD pattern showed the predominantly amor- phous nature of the silicate. 2,9 The small sharp peak at a diffraction angle (2θ, the angle between the incident and diffracted beams of X-rays) of 44.5° indicated the presence of some ordered structure in the RHA silicate, and RHA had a similar structure with a sharp peak at 44.5° (Figure 2). This sharp peak could be due to the presence of an ordered crystalline structure of oxides or silicates of Na, K, Ca, or Mg. The sodium oxide content of silicate blocks as calcu- lated from the amount of NaOH added to 100 g of RHA was 15%. Generally, commercial silicates have a com- position of 1:1 Na 2 O and SiO 2 . Comparatively lower Na2O in the silicate material may minimize the hygro- scopic character. Fourier transform infrared (FTIR) spectra of silicates produced by low, intermediate, and high heating rates were similar, as shown in Figure 3a - c. The FTIR spectra demonstrated that the silicate network and siloxane (Si - O - Si) bonds were dominant in the silicate structure. 4 The major absorbance band between 1000 and 1350 cm - 1 was due to the structural vibration mode and siloxane bonding in the silicate. 3 The absorbance band for Si - O - Si appears to be shifted toward higher wavenumbers compared to the transmission spectra Figure 1. XRD pattern of sodium silicate produced from RHA using (a) high, (b) intermediate, and (c) low heating rates. Figure 2. XRD pattern of RHA used for silicate production. Ind. Eng. Chem. Res., Vol. 42, No. 1, 2003 47reported in the literature. As discussed by Willey, 10 the difference in the spectral measurement mode could cause this shift due to resonance effects. The peaks between 500 and 1000 cm -1 resulted from various vibrational modes of the silicate network. The broad bands between 1500 - 2000 and 2750 - 3400 cm -1 were due to adsorbed moisture. Block Density. The densities of the silicate blocks (0.33 - 0.42 g ‚ cm - 3 ; Table 1) were very low compared to those of the silica xerogel blocks (1.25 - 1.44 g ‚ cm -3 ). 4 The low block density reflected the highly porous structure of the silicate blocks, in contrast to silica xerogels. 4 Hence, silicate blocks could be better thermal insulation materials than the xerogels. The heating rate had a significant effect on the silicate block density. As the heating rate increased, the block density decreased, probably because of the increased porosity. The me- chanical strength and thermal conductivity increased significantly with the increase in the block density. Mechanical Strength. The mechanical strength of the silicate blocks was low because of the porous structure. However, this material could be produced in any desirable size and shape or machined (sanding or slicing) to any size and shape without breaking the material. Further, mechanical strength may not be a concern for most of the thermal insulation applications. The higher density (0.42 g ‚ cm -3 ) material had a sig- nificantly higher mechanical strength compared to the lower (0.33 g ‚ cm -3 ) and intermediate (0.38 g ‚ cm -3 ) density blocks (Table 1). The difference in the mechan- ical strength of low-density and intermediate-density blocks was statistically insignificant. This could be because the mechanical strength may not be directly related to the block density and porosity because other factors including material structural characteristics and forces that govern the molecular interactions would also contribute to the mechanical strength. Hence, despite the high porosity, the low-density silicate blocks had adequate mechanical strength to retain physical struc- ture and structural integrity. To investigate the stability of the insulation material, the silicate blocks produced by low heating rate were immersed in water at room temperature or heated at 600 °C for varying time, and the mechanical strengths of these blocks were deter- mined. Table 2 shows the changes in the mechanical strength of silicate blocks after soaking in water or heating at 600 °C for 1, 6, and 24 h. The silicate blocks retained their physical structure when immersed in water, maybe because of comparatively lower Na2O (15%) content. The mechanical strength of the silicate blocks slightly decreased after soaking the silicate blocks in water. However, the differences in the mechanical strength were statistically insignificant. The silicate blocks also retained their physical structure when heated at 600 °C. The mechanical strength of the silicate blocks increased insignificantly when heated at 600 °C (Table 2). However, heating to a temperature higher than 680 °C resulted in deformation of the geometrical shape due to bending of the disk-shaped blocks. Thermal Conductivity. Increased block density resulted in increased thermal conductivity (Table 1). The thermal conductivity of a low-density silicate block (0.103 W ‚ m -1‚ K -1 ) is lower than those of most of the styrene-based insulation polymers (0.12 - 0.2 W ‚ m -1‚ K -1 ) used in low-temperature ( < 200 °C) thermal insulation. Thermal conductivities of silicates, glass, asbestos, and ceramics generally ranged from 1 to 2 W ‚ m - 1‚ K - 1 , 11 so these are used as high-temperature thermal insulators. Conclusions The data indicated that low-density silicate blocks with a mechanical strength required to maintain struc- tural integrity could be produced from RHA using a simple alkali extraction followed by boiling to evaporate the water at a controlled heating rate depending on the silicate concentration and solution volume. Thermal conductivity data showed that RHA can be used to produce silicate blocks for thermal insulation applica- tion as a safe substitute for asbestos and organic polymer based foam materials that pose public health and environmental problems. Acknowledgment We thank Producers Rice Mill (Stuttgart, AR) for providing rice hull ash. This project was funded by the USDA (NRI Competitive Grants Program, Project 9602154). Literature Cited (1) Iler, R. K. Silica gels and powders. In The Chemistry of Silica; John Wiley and Sons: New York, 1979. Figure 3. FTIR spectra of a ground sodium silicate produced by (a) high, (b) intermediate, and (c) low heating rates. Table 1. Density Mechanical Strength and Thermal Conductivity of Silicate Blocks Produced by Evaporation of Water at Three Different Heating Rates a sample density (g ‚ cm -3 ) peak force at break (N ‚ cm -2 ) conductivity (W ‚ m -1‚ K -1 ) block 1 0.33a (0.015) 145a (3.6) 0.103a (0.002) block 2 0.38 b (0.015) 161 a (3.0) 0.113 b (0.005) block 3 0.42 c (0.025) 196 b (8.6) 0.128 c (0.003) a Blocks 1-3 were produced by heating silicate solutions at a higher, intermediate, and lower heating rates, respectively. The heating times for higher, intermediate, and lower heating rates were approximately 4, 6, and 8 h, respectively. Values in the same column with different superscripts are significantly different from each other at P < 0.05. Values in parentheses are standard deviations of the corresponding measurements. Table 2. Mechanical Strength of Silicate Blocks after Soaking in Water or Heating at 600 °C for Varying Time a peak force at break (N ‚ cm -2 ) treatment time (h) water soaking heating at 600° C 1 191 a (8.4) 202 a (6.5) 6 180 a (7.5) 220 a (8.1) 24 176 a (6.3) 211 a (5.5) a Silicate blocks produced by low heating rate were used. Values in the same column with the same superscripts are not signifi- cantly different from each other at P < 0.05. Values in parentheses are standard deviations of the corresponding measurements. 48 Ind. Eng. Chem. Res., Vol. 42, No. 1, 2003(2) Kamath, S. R.; Proctor, A. Silica gel from rice hull ash: Preparation and characterization. Cereal Chem. 1998, 75, 484. (3) Kalapathy, U.; Proctor, A.; Shultz, J. Production and properties of flexible sodium silicate films from rice hull ash silica. Bioresour. Technol. 2000, 72, 99. (4) Kalapathy, U.; Proctor, A.; Shultz, J. Silica Xerogel from Rice Hull Ash: Structural, Physical and Mechanical Properties as Effected by Gelation pH, and Silica Concentration. J. Chem. Technol. Biotechnol. 2000, 75, 464. (5) Kalapathy, U.; Proctor, A.; Shultz, J. A simple method for production of pure silica from rice hull ash. Bioresour. Technol. 2000, 73, 257. (6) Virta, R. Asbestos substitutes. In Industrial Minerals and Rocks; Carr, D. D., Ed.; Society of Mining, Metallurgy, and Exploration, Inc.: Littleton, CO, 1994. (7) Sun, L.; Gong, K. Silicon-based materials from rice husks and their applications. Ind. Eng. Chem. Res. 2001, 40, 5861. (8) SAS. JMP IN, Version 3.2.1; SAS Institute Inc.: Cary, NC, 1997. (9) Proctor, A. X-ray diffraction and scanning electron micro- scope studies of rice hull ash. J. Am. Oil Chem. Soc. 1990, 67, 576. (10) Willey, R. R. Fourier transform infrared spectrometer for transmittance and diffuse reflectance measurements. Appl. Spec- trosc. 1976, 30, 593. (11) Thermal conductivity database; www.matweb.com: an online material information resource, 2002. Received for review April 29, 2002 Revised manuscript received October 25, 2002 Accepted October 28, 2002 IE0203227 Ind. Eng. Chem. Res., Vol. 42, No. 1, 2003 49