Performance of an Argentinian acid-activated bentonite in the bleaching of soybean oil

Abstract

In this work, a bentonite clay from Mendoza, Argentina was activated with H2SO4 (4 and 8 N) at 90ºC for 2 and 3.5 hours. Under these conditions several cations were removed from the octahedral sheet (Mg, Al, and Fe) and the DTA-TGA curves of the solids obtained after treatment were modified. Treatment time and acid concentration increased the degree of destruction of the bentonite structure. Activated samples were tested in order to verify their capacity to bleach soybean oil and were compared to a standard commercial bleaching clay. Treated samples are more efficient in bleaching than the standard.

clay; acid-activated bentonite; bleaching; soybean oil

Performance of an Argentinian acid-activated bentonite in the bleaching of soybean oil

E.L.Foletto^I; C.Volzone^II; L.M.Porto^I

^IDepartamento de Engenharia Química e Engenharia de Alimentos, Universidade Federal de Santa Catarina, Phone +(55) [phone], Fax +(55) [phone], P.O. Box 476, 88040-900, Florianópolis - SC, Brazil. E-mail: [e-mail], E-mail: [e-mail] ^IICentro de Tecnología de Recursos Minerales y Cerámica - CETMIC, C.C. 49, Cno Centenario y 506 (1897), M. B. Gonnet, Prov. Buenos Aires, Argentina. E-mail: [e-mail]

ABSTRACT

In this work, a bentonite clay from Mendoza, Argentina was activated with H₂SO₄ (4 and 8 N) at 90^oC for 2 and 3.5 hours. Under these conditions several cations were removed from the octahedral sheet (Mg, Al, and Fe) and the DTA-TGA curves of the solids obtained after treatment were modified. Treatment time and acid concentration increased the degree of destruction of the bentonite structure. Activated samples were tested in order to verify their capacity to bleach soybean oil and were compared to a standard commercial bleaching clay. Treated samples are more efficient in bleaching than the standard.

Keywords: clay, acid-activated bentonite, bleaching, soybean oil.

INTRODUCTION

Due to its sorptive and catalytic properties, bentonite is widely used in a variety of industrial applications. The clay is utilized as a pesticide carrier, an animal waste adsorbent, a catalyst and catalyst support, and a decolorizing agent in oil refining, and in the pharmaceutical industries. It is well known that bentonites in their natural state have limited sorbing capacity. This ability is greatly enhanced by treatment with strong acids. When bentonites are acid-activated by hot mineral acid solutions, hydrogen ions attack the aluminosilicate layers via the interlayer region (Taylor and Jenkins, 1987). This process alters the structure, chemical composition, and physical properties of the clay while increasing its adsorption capacity (Mokaya et al., 1993). In refining vegetable oils, the bleaching process is an important step, used for removal of undesired components from the oil by adsorption. This enables the production of a light-colored and stable oil, acceptable to consumers (Proctor and Palaniappan, 1989).

The behavior of bentonite when treated with acid solutions has been studied from different points of view (Volzone at al., 1986; Volzone et al., 1988; Christidis et al., 1997; Morgado, 1998; Foletto et al., 2000). The purpose of this study was to examine the structural properties of an Argentinian natural bentonite after acid treatments and to evaluate its capacity to decolorize soybean oil.

MATERIAL AND METHODS

A natural bentonite clay (from the province of Mendoza, Argentina), designated from now on as M, was used as the starting material. The natural sample particle size is < 74 µm.

Natural powder clay was treated under mechanical stirring with H₂SO₄ solution (4 and 8 N) at 90 ^oC for 2 h and 3.5 h in a three-neck glass flask under reflux conditions. The ratio of the mass of clay to the volume of acid solution was 1:10 (w/v). After activation the solids were washed until SO₄^2- free, dried at 60 ^oC, and ground to pass through a 74 µm sieve. These solids were designated as M4/2, M8/2, M4/3.5, and M8/3.5, where 4 and 8 indicate the concentration of the acid solution used and 2 and 3.5 indicate the time of treatment.

X-ray diffractograms (XRD) of samples were obtained with a Philips 3020 diffractometer equipped with a PW3710 computerized control unit, operating at 40 kV and 20 mA and using Cu-Ka (l = 1.5405 Å) radiation at a scanning speed of 1° (2q)/min from 3 to 70° (2q). The d(001) spacing to corroborate the presence of the smectite was measured in slice of the oriented specimen under different conditions: dried at room temperature, solvated with ethylene glycol, and then calcined at 550 ° C (Brindley and Brown, 1980). Elemental composition was determined by X-ray fluorescence (XRF), using a Philips PW 2400 XRF spectrometer. Infrared (IR) spectra were recorded in the 4000-350 cm^-1 region with a Perkin-Elmer 16 PC spectrophotometer, using KBr pellets prepared with a pressed-disk technique (0.3mg sample + 200 mg KBr). The Greene-Kelly test (Greene-Kelly, 1953) for differentiating between the beidellite and montmorillonite smectite present in the bentonite was used. The method is based on measurement of d(001) reflection after the Li-250° C-glycerol treatment. Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were carried out on Netzsch STA 409 analyzers at a heating rate of 10 ^oC min^-1 at an air flow rate of 35 mL min^-1 in a temperature range of 25-1000 ^oC. a -Al₂O₃ was used as reference for DTA measurements.

One hundred grams of alkali-refined soybean oil, provided by Santista-Ceval Alimentos S.A (Gaspar-SC, Brazil), was stirred and heated to 100 ^oC in a 450 mm Hg vacuum. The clay material (0.5 g) was then added to the heated oil, and the mixture was stirred mechanically for 30 min. A stream of N₂ was maintained in the oil batch throughout the experiment. The hot oil and clay mixture was filtered in vacuum at the end of the experiment just before measuring the absorbance. The bleaching efficiency of activated clays was then determined by measuring the color of the bleached oil using a UV-VIS spectrophotometer (Model WFJ 525) at 450 nm. In this study bleaching efficiency is defined by the following expression:

where A_unbleached and A_bleached are absorbances of unbleached and bleached oil, respectively, at 450 nm.

The performance of the activated samples in the bleaching of soybean oil was compared with that of a commercial bleaching material (Engelhard F118).

RESULTS AND DISCUSSION

In Figure 1 the mineralogical characteristics of the natural clay used in this study are presented.

Figure 1a shows the XRD diffractograms of the M natural bentonite. The presence of smectite (S), kaolinite (K), gypsum (G), and quartz (Q) is evident in the figure. In order to confirm the presence of the smectite phase in the original sample, we carried out XRD analyses for oriented, calcined, and glycolated samples (Figure 1b). The presence of smectite is verified thus basal distance d(001) expands to 16.92 Å when the sample is saturated with ethylene glycol and contracts to 9.79 Å when it is calcined at 500 ° C (Brindley and Brown, 1980).

The difference between the montmorillonite and beidelite present in the bentonite was found with the Greene-Kelly test (Greene-Kelly, 1953). Montmorillonite was detected by the contraction of d(001) to 8.85 Å and beidelite, by the expansion at 17.60 Å (Figure 1c). The relative amounts of the species were obtained from their corresponding peak areas. Results showed that the montmorillonite/beidelite ratio in the material studied is 18/82. Thus the bentonite is predominantly beidellitic.

Figure 1d shows the DTA curve for the natural sample. The first endothermic peak corresponds to the adsorbed water loss (at about 150 ° C) and the peaks at 530 and 660 ° C correspond to the loss of structural hydroxyl groups. This again confirms the presence of montmorillonite and beidelite in the smectite phase, since according to Grim and Kulbicki (1961), the second endothermic peak for montmorillonites appears between 600 and 700 ° C and for beidelites, between 550 and 600 ° C.

In Figure 1e we present the absorption spectrum in the infrared region for the natural sample. The absorption band at 3640 cm^-1 is attributed to stretching vibrations of the OH group while that at 3454 cm^-1, the presence of interlayer water. The amount of adsorbed water in clays is related to the deformation vibrations of the H–O–H group (1664 cm^-1). The bands at 1042 and 798 cm^-1 are attributed to Si-O stretching vibrations. The band at 770 cm^-1 corresponds to the beidellite species. The bands at 526 and 466 cm^-1 correspond to deformation vibrations of Si–O–Al and Si–O–Si, respectively (Volzone et al., 1988).

The chemical compositions of the natural and sulfuric-acid-treated samples are listed in Table 1. Acid treatment modified the smectite chemistry, as is evident from the changes in chemical composition (Table 1). Exchangeable Ca²⁺, Na⁺, and K⁺, which were also removed during the treatment, correspond to the exchange cations present in the samples. However, a considerable amount of these elements remained, in the activated samples due to the presence of insoluble impurities (feldspar) in acid solutions (Barrios et al., 1995) which might be partially dissolved. Acid activation resulted in a material with a greater silica content and a significant lower abundance of oxides in the octahedral sheet (Al₂O₃ + Fe₂O₃ + MgO).

As expected, the increase in treatment time and acid concentration resulted in a more effective attack on the structure, as observed by the decrease in the relative amount of cations belonging to the octahedral sheet and by the increase in the Si⁴⁺/(Al³⁺+Fe^2+/3++Mg²⁺) ratio.

Figure 2 shows the infrared spectra of the natural and acid-activated bentonite. In general, no important changes were observed; nevertheless, the bands at 526 cm^-1 (Si-O-Al), 466 cm^-1 (Si-O-Si), 1040 cm^-1 (Si-O), 3454 cm^-1 (adsorbed H₂O), and 3640 cm^-1 (Al-Al-OH, Mg–OH–Al) were modified. These changes are better observed by the intensity ratios of Al-Al-OH, Mg–OH–Al / H₂O and Si–O–Al / Si–O–Si, as shown in Table 2.

The IR band intensity ratios of the Al-Al-OH, Mg–OH–Al / H₂O and Si–O–Al / Si–O–Si bands decreased with an increase in acid concentration and time of treatment (Table 2). This work suggests that structural changes in the smectite structure were most pronounced for the acid treatment carried out for 3.5 h and an acid concentration of 8N.

In Figure 3 DTA/TGA curves for the natural and activated samples are presented. DTA curves (Fig. 3a) show a small change in the endothermic peak after acid treatment, corresponding to the loss of adsorbed water (at about 150 ° C), and peaks corresponding to the loss of structural hydroxyl groups (530 and 600 ° C).

The dehydroxylation weight loss (%) of the samples (Table 3) was obtained from TGA (Figure 3b) in the 450-750 ° C range. The destruction of the octahedral sheet in the acid-treated samples was calculated as the ratio of the weight loss to the mass of the original sample.

Treated samples displayed 41 to 46 % octahedral sheet destruction, suggesting that even for the most drastic conditions used (8 N and 3.5 h treatment) smectite structure is not totally destroyed. With increasing acid concentration and treatment time, a more extensive destruction of the octahedral sheet is observed.

In Table 4 the results from the bleaching runs for soybean oil using natural and activated clays as well as for a commercial product used as a reference are presented. It is obvious that the activated clays used performed better than the reference bleaching material. When compared to the natural clay an up to 12-fold increase in bleaching capacity is observed in the acid-treated samples.

In Table 4, it is obvious that bleaching efficiency increases with treatment time and acid concentration. Thus, the material treated for 3.5 h with sulfuric acid 8N showed a better bleaching performance.

Next we show some correlations we have found between bleaching tests and our previous clay characterization studies. Figures 4, 5, and ⁶ show the bleaching efficiency plotted as a function of IR intensities, Si / (Al+Mg+Fe) ratio, and octahedral sheet destruction, respectively.

In Figure 4 a rather similar behavior is observed for bleaching efficiency as a function of the Al-Al-OH, Mg–OH–Al / H₂O and Si–O–Al / Si–O–Si ratios: bleaching efficiency increases as the IR ratios decrease, indicating that the loss of structural Al due to the acid attack is beneficial to the clarifying properties of the clay.

In Figure 5 it is obvious that the decolorization efficiency of the clay is also a function of the Si / (Al+Mg+Fe) ratio. The loss of octahedral cations due to the acid treatment produces a material with higher adsorption activity.

Finally, in ^{Figure 6}, it is evident that destruction of the octahedral sheet improves bleaching activity and that lixiviation of octahedral cations improves the bleaching efficiency for the soybean clarified under the conditions described.

CONCLUDING REMARKS

Destruction of the bentonite structure as a result of the sulfuric acid treatment is strongly dependent on the acid concentration and activation time.

Structural changes observed by chemical analysis, DTA-TGA, and IR indicate that destruction of the octahedral sheet depends on the acid attack. Destruction of the octahedral sheet (between 41 and 46 %) due to the lixiviation of octahedral cations caused by the acid attack on the bentonite yields an adsorbent material that is highly efficient in the bleaching of soybean oil functions better than a standard commercial product under the same conditions.

There are correlations between the bleaching efficiency of bentonite in clarifying soybean oil and IR band intensity, and the Si / (Al+Mg+Fe) ratio as well as octahedral sheet destruction (%).

ACKNOWLEDGMENTS

The authors thank Susana Conconi and Prof. Luiz Fernando Dias Probst for the DTA/TGA and IR analyses and the Centro de Tecnologia em Cerâmica (Criciúma-SC, Brazil) for the assays used in the chemical analyses. The financial support of CAPES (scholarship provided to ELF) is also acknowledged.

Received: August 7, 2001

Accepted: December 10, 2002

• Barrios, M.S., Gonzáles, L.V.F., Rodríguez, M.A.V. and Pozas, J.M.M., Acid Activation of a Palygorskite with HCl: Development of Physico-chemical, Textural and Surface Properties, Applied Clay Science, 10, 247-258 (1995).
• Brindley, G.W. and Brown, G., Crystal Structure of Clay Minerals and Their X-ray Identification, London: Mineralogical Society (1980).
• Christidis, G.E., Scott, P.W. and Dunham, A.C., Acid Activation and Bleaching Capacity of Bentonites from the Islands of Milos and Chios, Aegean, Greece, Applied Clay Science, 12, 329-347 (1997).
• Foletto, E.L., Volzone, C., Morgado, A.F. and Porto, L.M. (2000), Análise comparativa da ativação ácida de dois materiais argilosos com diferentes composições mineralógicas, VI Jornadas Argentinas de Tratamiento de Minerales, Salta, Argentina, 43-48 (2000).
• Greene-Kelly, R., The Identification of Montmorillonoids in Clays, Journal of Soil Science, 4(2), 233-237 (1953).
• Grim, R.E. and Kulbicki, G., Montmorillonite: High Temperature Reactions and Classification, The American Mineralogist, 46, 1329-1369 (1961).
• Mokaya, R., Jones, W., Davies, M.E. and Whittle, M.E., Chlorophyll Adsorption by Alumina Pillared Acid-activated Clay, Journal of the American Oil Chemists' Society, 70, 241-244 (1993).
• Morgado, A.F., Caracterização e propriedades tecnológicas de uma argila esmectítica de Santa Catarina, Ph.D. diss., Universidade de São Paulo, São Paulo, Brazil (1998).
• Proctor, A. and Palaniappan, S., Soy Oil Lutein Adsorption by Rice Hull Ash, Journal of the American Oil Chemists' Society, 66 (11), 1618-1621 (1989).
• Taylor, D.R. and Jenkins, D.B., Acid-activated Clays, Society of Mining Engineering of the American Institute of Mechanical Engineers, Transactions, 282, 1901-1910 (1987).
• Volzone, C., López, J.M. and Pereira, E., Activación ácida de un material esmectítico, Revista Latinoamericana de Ingeniería Química y Química Aplicada, 16, 205-215 (1986).
• Volzone, C., Zalba, P.E. and Pereira, E., Activación ácida de esmectitas. II - Estudo mineralogico, Anales de la Asociación Química Argentina, 76, 57-68 (1988).

Publication Dates

• Publication in this collection
  25 June 2003
• Date of issue
  June 2003

History

• Received
  07 Aug 2001
• Accepted
  10 Dec 2002