Brazil is the most important Latin-American rice producer and the State of
Rio Grande do Sul (RS) is the countrys main producer (IBGE,
2011). Rice husks in our country have begun to be regarded as an agro-industrial
residue of importance and as a potential raw material for thermal and chemical
processes, as well as a source of soluble silica.
However, attempts to use rice husks for industrial purposes are still facing
technical difficulties, mostly due to their poor protein content, low degradability
and high mineral percentage (Zhang et al., 2010).
For these reasons, if a suitable application could be found for residual rice
husks, it could provide a challenging business opportunity, while also bringing
about benefits to the environment and public health (Martins
et al., 2007; Zakaria et al., 2010).
At the moment, most of the rice husks remain unused which causes environmental
problems, for example, emission of CO2 and long lasting smoke from
open sky combustion and methane, when they are left to degrade on the ground
(Umeda and Kondoh, 2010).
For every 4 tons of rice, 1 ton of rice husks is produced. After combustion
for power generation, 15-30% of this results in RHA (Mehdinia
et al., 2011) comprising around 95% silica (Kermani
et al., 2006). Thus, it is estimated that the availability of silica
is around 380,000 ton year-1 in the state of RS. Previous studies
have shown that it is possible to obtain silica from rice husks and Rice Husk
Ash (RHA) with similar yields compared to other processes (Zaky
et al., 2008; Zhang et al., 2010;
Lima et al., 2011). A simplified method of sol-gel
extraction of silica at room temperature was also reported in the literature
(Lima et al., 2011).
Zaky et al. (2008) have prepared silica nanoparticles
from RHA by employing statistical design to optimise the parameters affecting
the dissolution of silica, such as stoichiometry, time and temperature as well
as achieving efficiency of 99% on the extraction. Zhang
et al. (2010) produced amorphous silica from RHA treated with HCl
solution, obtaining surface area of 287.86 m2 g-1, mean
particle size of 50 mm and 99.87% purity. Other works have described rice husks
and RHA as excellent sources for high quality silica (Lima
et al., 2011; Kalapathy et al., 1999;
Witoon et al., 2008). However, these silicas
have only been proposed as substitutes for commercial sources, such as nanosilicas,
ceramic and cement additives.
On the other hand, silica also plays an important role in almost all living
organisms. It is absorbed by the roots of plants and deposited on the outer
walls of epidermal cells as a silica gel, where it acts as a physical barrier
against pathogenic fungi and attacks by insects. Moreover, it reduces the loss
of water through transpiration, stiffens the cell walls and keeps the leaves
more erect (Nakata et al., 2008; Abdalla,
The rice plant responds most positively to the application of Si and the use
of soluble silica as a fertiliser is largely employed (Ramos
et al., 2008). The benefits for the culture of rice include increased
growth and production, positive interactions with fertilisers, as well as higher
resistance against diseases, plagues, drought and salinity (biotic and abiotic
stresses) (Voleti et al., 2008). According to
Rodrigues et al. (2003), an increase in the Si
content of rice plants by the application of Ca silicate to the soil, as wolastonite,
explained the significant reduction of the occurrence of sheath blight on the
The Response Surface Methodology (RSM) is a statistic-mathematical tool based
on factorial planning (Mohagir et al., 2009)
which allows to achieve more precise results with less experiments, obtaining
higher product yield, lower response time and material consumption (Charyulu
and Gnanamani, 2010). It is not recommended for a large number of variables
(Gummadi and Kumar, 2006). Besides this, the RSM has
been much used for the optimization of the process parameters and to raise the
yield of the desired products (Jaiswal et al., 2011).
Thus, the purpose of this study was essentially to evaluate the potential use of micronized RHA for the production of xerogel silicas and their possible use as fertilisers in rice cultivation. Tests were also carried out with non-micronised RHA, where similar results were obtained (not examined in this study).
MATERIALS AND METHODS
This study was developed from March 2009 to March 2010 at the Chemistry Department of the Federal University of Santa Maria with the collaboration of colleagues from the Institute of Agrarian Sciences of the Federal University of Uberlândia, Brazil.
Preparation of the RHA: The micronised RHA used in this investigation
was supplied by a local market developer for RHA from the thermoelectric combustion
of rice shells (Fig. 1).
|| Extraction of silica from rice husks ash
After being sent to the laboratory, the RHA was classified in a sieving system
(Bertel, São Paulo, Brazil) using a 2.4 mm mesh sieve. The RHA samples
were dried in a kiln (Ideal, São Paulo, Brazil) for 24 h at 110 °C;
after that, they were left to cool in a desiccator and were later stored in hermetically-sealed
glass flasks for subsequent use. Following this, X-ray fluorescence spectrometric
analysis was conducted to determine the silica content of the micronised RHA.
Silica production: The silicas were prepared with the aid of NaOH and
KOH extracting solutions with the respective bicarbonates being used as catalysts,
following the process (with some modifications) described by Kalapathy
et al. (2002). Micronised RHA (20 g) was placed in a 250 mL round-bottom
flask and mixed with the corresponding alkaline solution and catalyst, in concentrations
predicted by the Response Surface Methodology (RSM) for sodium-based silica,
using as independent variables the pH (1, 3, 5, 7 and 9), base concentration
(2.0, 3.5, 5.0, 6.5 and 8.0 mol L-1), catalyst concentration (0.6,
1.3, 2.0, 2.7 and 3.4 mol L-1) and time (1, 2, 3, 4 and 5 h). The
yield of the silica extraction was regarded as the dependent variable.
The mixture was then kept under reflux for 3 h in accordance with the DOE (design of experiment). Following this, the solution was passed through a paper filter with 2.0 μm porosity (Whatman, Springfield Mill, UK) and the filtered material was acidified to the pH predicted by the RSM with a solution of 5.5 N H2SO4 which formed a slightly pink silica precipitate. It was filtered again in the same way as mentioned before, now adding 20 mL of 3% H2O2.
The discoloured silica xerogel was dried in a kiln for 24 h; after that, it was washed with distilled water to remove the excess acid (pH~6) and put back in the kiln for a further drying period of 24 h. In the final stage, the silica xerogel was cooled in a desiccator, ground in porcelain mortar and classified at 0.24 mm mesh.
Characterization of silica xerogels: The X-ray diffractograms of the silica xerogels and the RHA were obtained with the aid of a Shimadzu Diffractometer Model XD 7A (Shimadzu, Tokyo, Japan), using CuKα radiation, acceleration voltage of 30 kV, current of 30 mA and scan angle of 2-72°.
The specific surface areas were calculated through the Brunauer, Emmet and
Teller (BET) method (Brunauer et al., 1938), in
a volumetric apparatus fitted with a turbomolecular pump (Edward, Milwaukee,
OR, USA). The analysis of the pore distribution was carried out by laser diffraction
technique with a Cilas 1064 L particle size analyzer (Madison, WI, USA). A Philips
X-ray spectrometer PW 2400 (Philips, The Netherlands) with a 3 kW tube and rhodium
target was used for fluorescence measurements.
Experiments involving the addition of silica xerogels to different types of soil: These experiments were conducted at the Institute of Agrarian Sciences, Federal University of Uberlandia, with the aim to confirm the benefits of the use of the silica xerogels as fertilisers. The experiments were carried out in a greenhouse using soil samples classified as dystrophic Red Latosol with a moderate A horizon and a medium texture (77% sand).
The silica xerogels were weighed and mixed with the soil together with basic fertilisers which consisted of 0.3 g kg-1 urea, 0.5 g kg-1 Triple Superphosphate (Shanghai Shilei Ltd., Shanghai, China; 54% P2O5; small% S; Ca>10%), 0.35 g kg-1 KCl and 0.1 g kg-1 formulated micronutrients (with 9% Zn, 1.8% B, 2% Mn; 8% Cu; 0.1% Mo; 3% Fe). Half of the N and K was supplied in the sowing season and the other half 20 days later. The soil was placed in 8 kg plastic containers and remained in incubation for 20 days until the rice-sowing season.
The reactivity of the silicon sources was measured by adding to the soil increasing amounts of a standard silicate-Recmix Siligran-(12% Si; Fertion, Rio Branco do Sul, PR, Brazil) in granulated form and in powder; sodium xerogel silica and potassium xerogel silica (0, 100, 200 and 400 kg Si ha-1) in random blocks with four repetitions. Ouro Minas rice was sown and cultivated in an average cycle, lasting for a period of 94 days from sprouting to blooming. It displayed the following characteristics: Fine long grain rice (needle-shaped); a yield of between 55 and 65% of whole grains; resistant to lodging and higher resistance to blast.
Statistical analysis: A Design of Experiments (DOE) was employed to
reduce the number of experiments and to describe the effects on the variables
investigated (Bas and Boyaci, 2007). Thus, the chosen
model used was RSM with four independent variables at five levels for the extraction
of silica from the RHA with lye. The results were submitted to an Analysis of
Variance (ANOVA) to determinate the variables or interactions between them,
that would be more significant. Checking the most suitable model for the designs
was carried out with the aid of Statistics 6 Software.
RESULTS AND DISCUSSION
Characterization of silica xerogels: The diffractrograms of the silica
xerogels can be observed in Fig. 2. The diffractrograms of
the sodium-based and potassium-based silica samples show peaks in the range
of 12.5° which confirmed the presence of amorphous regions. In the case
of the potassium-based silica, a peak of 29° was confirmed with regard to
K6Si3O9 (trisilicate of hexapotassium) or tridymite,
a crystalline form of silica (Real et al., 1996;
Leu and Chou, 2011).
The specific surface areas and the mesopore volumes of the prepared silicas are provided in Table 1.
As can be observed, the high values for the surface area of the sodium and
potassium-based silicas correspond to the values found in the literature (Liou,
2004, Zhang et al., 2010). Figure
3 shows the nitrogen adsorption-desorption isotherms. The lower part of
the loop is traced out on adsorption and the upper portion on desorption.
||Specific surface area, mesopore volume and diameter and mean
particle diameter of silicas and micronized rice husks ash
|*Liou (2004), **Zhang
et al. (2010)
||Diffractograms of the sodium and potassium xerogel silicas
|| Isotherms of adsorption/desorption of nitrogen
The isotherm loop of the sodium silica resembles type IV of the Brunauers
classification but without a significant occurrence of mesopores. It can be
also seen for the potassium silica at low values of P/Po an isotherm profile
similar to type III but the adsorption increases markedly at P/Po from 0.6 to
0.9 which characterizes a superimposed multi-layer. It was confirmed that the
silicas obtained had mean particle diameters of 28.41 and 67.93 μm, respectively
which are higher values than those recorded in the literature (Della
et al., 2002; Liou, 2004; Zhang
et al., 2010). However, it should be stressed that in the last case,
the investigation was initiated using coarse RHA.
Table 2 shows the results of the X-ray fluorescence analysis (FRX) of the micronised RHA sample and the prepared silica xerogels. The silicas produced showed contents of 75.5 and 77.2% of SiO2. These values can be regarded as quite satisfactory for industrial usage, considering that the raw material was coarse RHA which had not undergone any previous purification treatment. The literature records values for silicas of chemically treated raw materials with a SiO2 content of up to 95%.
The optimization model for the parameters of silica production: As mentioned before, CCDs were employed in the experiments for sodium-based silica to evaluate the influence of the variables and determine the optimal quantitative relations.
||Oxide composition of the rice husks ash and xerogel silicas
obtained by X-ray fluorescence analysis
|nd: Not determined
Effect of the silicas on rice cultivation: The application of increasing amounts of silica sources as an additive ensures a greater availability of Si in the soil, where it can be absorbed into the rice root system and accumulated in the leaf tissue.
As can be seen at Fig. 4, the levels of Si in the soil increased with the added amounts of both silicas in the quantity of 400 kg ha-1 of Si, higher than the control definition. The added amounts had an effect on the accumulation of Si in the area surrounding the plant; the best results were obtained with quantities of 200 and 400 kg ha-1 of Si, higher than the control definition.
A significant amount of Si accumulated in the surrounding area when standard Recmix was added together with the sodium and potassium-based silicas. In terms of the production of dry material, the best amounts were 100 and 200 kg Si ha-1, higher than the control definition. In terms of grain production, the sodium and potassium-based silicas were better than the control definition (Fig. 4), which shows that increasing the amount of silica added to the soil leads to an increase in production. This effect suggests that grain production could be even higher if the amounts of Si added were higher than 400 kg ha-1.
The increase in production was due to the effect of Si in controlling the severity
of blast disease on the leaves (Nakata et al., 2008).
The epidermal cells became thicker and there was a greater degree of lignification
and/or silicification with a physical protective barrier being formed, enabling
it to withstand attacks by fungi and insects. When the silicon sources were
used, improved results were obtained with regard to the source of Siligran (Recmix)
which were caused by the greater solubility of the sources of sodium-based and
|| Available silicon in soil (mg dm-3) as a function
of the dosage (kg ha-1)
The chemical conversion of micronised RHA with the purpose to produce xerogel silicas was shown to be a possible way of making use of this inconvenient agricultural residue by generating products of some added value which can feed back into the business chain. The xerogel silicas, moreover, showed high values for the specific surface area and favourable mean particle size for these purposes. The tests of the xerogel silicas for the cultivation of rice indicated that these sources of silicon provided increments in the soil Si content and as a consequence, increasing the grain yield and dry matter production in the rice plant leaves. Therefore, the costs of agrochemicals for the control of plagues and diseases in the cultivation of rice can be reduced, by using a low-cost, renewable and most abundant product.
Our thanks to CNPq (National Council for Scientific and Technological Development) for providing financial assistance and scholarships.