- Email: [email protected]
Decolorization of sugar syrups using commercial and sugar beet pulp based activated carbons
Available online at www.sciencedirect.com
Bioresource Technology 99 (2008) 3528–3533
Decolorization of sugar syrups using commercial and sugar beet pulp based activated carbons H.L. Mudoga, H. Yucel, N.S. Kincal
*
Department of Chemical Engineering, Middle East Technical University, Ankara 06531, Turkey Received 3 November 2006; received in revised form 10 July 2007; accepted 28 July 2007 Available online 7 September 2007
Abstract Sugar syrup decolorization was studied using two commercial and eight beet pulp based activated carbons. In an attempt to relate decolorizing performances to other characteristics, surface areas, pore volumes, bulk densities and ash contents of the carbons in the powdered form; pH and electrical conductivities of their suspensions and their color adsorption properties from iodine and molasses solution were determined. The color removal capabilities of all carbons were measured at 1/100 (w/w) dosage, and isotherms were determined on better samples. The two commercial activated carbons showed different decolorization efficiencies; which could be related to their physical and chemical properties. The decolorization efficiency of beet pulp carbon prepared at 750 °C and activated for 5 h using CO2 was much better than the others and close to the better one of the commercial activated carbons used. It is evident that beet pulp is an inexpensive potential precursor for activated carbons for use in sugar refining. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Sugar syrup; Decolorization; Activated carbon; Beet pulp
1. Introduction Activated carbon adsorbents having a large internal surface area and relatively nonpolar surface are primarily being used to remove organic compounds from both liquid and gaseous streams (Smisek and Cerny, 1970). One significant application makes use of its decolorization properties in several areas including beet and cane sugar refining. The presence of colored compounds in the sugar syrups results from reactions occurring during the production. The chemical structure of some of these coloring materials is quite complex and difficult to determine in many cases. The most significant colored substances that develop during sugar processing can be classified in three general groups: (a) melanins (b) melanoidins and (c) caramels (Kearsley and Dziedzic, 1995). Conventional methods to prevent formation of these compounds include treatment with sulfur
*
Corresponding author. Tel.: +90 312 2102617; fax: +90 312 2102600. E-mail address: [email protected] (N.S. Kincal).
0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.07.058
dioxide or hydrogen peroxide, however, although these treatment steps retard the formation of some of these compounds, they cannot eliminate them altogether. In order to produce high quality white sugar or when very good clarity is required in the syrup products such as those used in colorless soda drinks, a treatment of juice with activated carbon or ion exchange resins is usually required (van der Poel et al., 1998). Activated carbons can be produced from various carbon containing species by physical and chemical activation methods. Most common feed stocks for commercial production may be given as coal, peat and wood (Smisek and Cerny, 1970). Other commercial and potential feed stocks include renewable agricultural by-products, namely nut shells (hazelnut, walnut, pecan, etc.), fruit kernels/seeds/ stones (apricot, apple, etc.), hulls (soybean, rice, etc.), pulps ¨ zer (paper, apple, beet) and olive stone (Balcı et al., 1994; O et al., 1998; Bac¸aoui et al., 2001; Yun et al., 2001; Aygu¨n et al., 2003). Ahmedna and co-workers (Ahmedna et al., 1997b, 2000) extensively studied the preparation, properties and
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
decolorization applications of activated carbons produced from agricultural by-products (sugar cane bagasse, rice hulls, rice straw, pecan shells using different binders (beet and sugar cane molasses, corn syrup and coal tar). They found that the type of by-product, binder used and the activation method determine the physical and chemical properties and thus decolorization properties of granular activated carbons. It has been demonstrated previously that activated carbon can be produced from beet sugar pulp which is a by¨ zer et al., 1998 and product of beet sugar processing (O 2002). Exploration of the sugar decolorization properties of carbons made from sugar beet pulp was undertaken to find another important outlet for this beet sugar plant waste primarily being used as cattle food. The aim of the present work was to compare the decolorization capabilities of beet pulp activated carbons produced locally in the laboratory scale with some commercial sugar decolorizing activated carbons and to look for possible correlations between the physical properties of activated carbons and their sugar syrup decolorization capabilities; with the hope of shedding some light on potential studies of preparing activated carbons for decolorization.
2. Methods 2.1. Syrup The syrup (thick juice) was obtained from the Ankara Sugar Factory. A series of grab samples of the syrup were collected at the point just before the sugar-boiling step in the factory. The sampling was done every 6 h during a period of 48 h. The samples were stored in the refrigerator in a plastic container. The properties of the syrup, determined according to Anon (1994) are given in Table 1. The refractive index values of the syrup (determined using Abbe BS 60/70 Refractometer) were 1.4450 at its natural pH of 7.7; 1.3630 at pH 4 and 1.4048 at pH 9. The indicator value, I.V., defined as the ratio of the refractive indices at pH value of 9 to that at pH 7 (Kearsley and Dziedzic, 1995) can be estimated on the basis of these values to be 1.04, indicating the major colorants in the syrup to be mel-
3529
Table 1 Properties of the syrup Color, ICUMSA units Brix, % Polarity, % Purity, % Ash, % Absorbance pH
3149 66.83 61.73 92.37 4.01 2.4376 7.7
anoidins (I.V. 1.0–1.2) or caramels (I.V. 1.0–1.5) Godshall (1997). 2.2. Commercial activated carbons Two reference commercial carbons, selected on basis of the manufacturer’s recommendations, were obtained from Waterlink Sutcliffe Carbons, England. The carbons used were DCL320 (wood base, activated with phosphoric acid) and DCL200 (coal base, steam activated). 2.3. Beet pulp activated carbons Eight different beet pulp activated carbon samples pre¨ zer et al. (1998) and O ¨ zer and C pared by O ¸ am (2002) were ¨ zer et al., 1998) by used. Two samples were prepared (O direct carbonization and were activated at 750 °C under a stream of carbon dioxide for 2 h or 5 h (750C2 and ¨ zer and C 750C5). Six samples were prepared (O ¸ am, 2002) by first impregnating with phosphoric acid overnight, followed by activating under a stream of nitrogen at 300 °C or 500 °C for 1.5–5 h (300P1.5, 300P3, 300P5, 500P1.5, 500P3, 500P5). In the coding of the samples, the first three digits indicate the activation temperature; the following letter is C for the directly carbonized – CO2 activated samples and P for the phosphoric acid impregnated – N2 activated samples; and the number following the letter indicates the time of activation in hours. 2.4. Methods of characterization of the activated carbons The commercial and beet pulp based activated carbons are characterized with the methods below and the results are given in Table 2.
Table 2 Some properties of the activated carbons Activated carbon
pH of susp.
Electrical cond. of susp, lS
Ash, mg g1
Pore size, nm
Pore volume, cm3 g1
Density, g cm3
DCL 320 DCL 200 750C5 750C2 500P5 500P3 500P1.5 300P5 300P3 300P1.5
4.5 6.7 6 10.2 2.8 2.8 2.8 2.8 2.8 2.8
285 17.1 24.8 972,000 3200 2600 2410 1820 2320 1870
36 168 183 348 210 183 193 197 192 202
3.6 2.6 2.2 3.8 3.3 2.7 2.6 3.5 3.4 4.3
1.30 0.54 0.65 0.25 0.55 0.27 0.22 0.28 0.28 0.20
0.28 0.35 0.28 0.25 0.67 0.60 0.58 0.62 0.71 0.61
3530
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
2.4.1. Surface area, pore volume and pore size The BET surface area of the samples were determined by nitrogen adsorption at 77 K using Micrometrics ASAP 2000 automatic surface area analyzer after the samples were dried under vacuum at 100 °C for 3 h to remove moisture. The surface area was determined using the N2 adsorption data at 77 K in the relative pressure range of 0.05–0.25, and taking the cross-sectional area of the nitrogen molecule to be 0.162 nm2. The total pore volume was based on the N2 amount adsorbed at relative pressure greater than 0.98, and taking the liquid molar volume of N2 to be 34.65 cm3/g at 77 K. The average pore diameter, D, was calculated from the equation based on uniform cylindrical pores as given below
D¼
ð4Þ ðPore volumeÞ BET area
2.4.2. Bulk density Powdered activated carbon samples were placed in graduated cylinders, tapped several times until constant volume (10 mL) was obtained, and measuring the weight, the bulk density was calculated as the ratio of weight to volume. 2.4.3. Moisture and ash contents Moisture content was determined by the measurement of dry weight after bringing the sample (about 1 g) to constant weight in an air circulation oven at 115 °C; and ash content was determined based on the weight of the residue obtained after 1.5 h in a furnace at 950 °C under air circulation. 2.4.4. pH Suspensions of the samples (1% w/w) in distilled water were heated to 90 °C, subjected to continuous stirring for 20 min, cooled to room temperature and the pH of the solution was determined using a Leeds Northrup 7415 pH-meter (Ahmedna et al., 1997b). 2.4.5. Electrical conductivity Suspensions of the samples (1% w/w) in distilled water were subjected to continuous stirring for 20 min, and the electrical conductivity of the solution was measured (Ahmedna et al., 1997b) at 25.6 °C using a Jenway 4020 conductivity-meter. 2.4.6. Iodine test, molasses test and syrup decolorization experiments These experiments were all similar in the sense that an iodine solution, diluted molasses or sugar syrup was contacted with the active carbon; the color of this sample was compared to that of a blank prepared and treated the same way without the activated carbon (Pendyal et al., 1999; Ahmedna et al., 1997a, 1997b). For the iodine test, 100 mL of stock solution (2.7 g I2 and 4.1 g KI in 1 L of solution) was added to 0.5 g activated carbon in 10 mL of 5% HCl solution, 5 min contact time was allowed, the sample was filtered through Whatman No. 110 filter paper and
the percentage color removal was based on the volumes of 0.1 M sodium thiosulphate required (using starch as the indicator) for the sample and the blank filtrates. For the molasses test, the test solution was prepared by dissolving 10 g of sugar beet molasses and 15 g of Na2HPO4 in 500 mL of water and sufficient H3PO4 to make pH 6.5, diluting to 1 L and filtering through a thin layer of filter aid (diatomaceous earth). Fifty milliliter of this solution was added to 0.5 g of activated carbon; the mixture was well stirred, and was placed on a heating plate along with the blank and brought to boil. The percentage color removal was based on the absorbance readings taken 420 nm using a Hitachi U-3200 Spectrophotometer after the sample and the blank were filtered through Whatman No. 4 filter paper. For the syrup decolorization experiments, syrup samples were adjusted to pH 7.0 using 0.1 M HCl and Thymol Blue indicator, two 100 g of samples were brought to 80 °C in a water bath with shaking at 160 rpm, 0.2–1.0 g of activated carbon was added to one of the samples, and 20 min was allowed, which had been verified to be sufficient for the attainment of equilibrium. Then, the samples were clarified in two steps. The first filtration was through Whatman No. 110 filter paper, followed by addition of 1% diatomaceous earth and heating to 70 °C for reducing the viscosity. The second filtration was through Nalgene Filtering assembly with a 0.45 lm membrane for removal of very fine carbon particles generated by attrition. The absorbance was determined at 420 nm on the clarified sample. The color was expressed in ICUMSA units (IU) defined according to equation (Anon, 1994); IU ¼ 1000 A=ðb cÞ; where A = absorbance at 420 nm of the test sample b = length (cm) of the adsorbing path c = concentration (g sugar/mL) of the test sample The IU is therefore a measure of the weight ratio of a hypothetical coloring substance to the sugar present. All activated carbon samples were studied for their sugar syrup decolorization properties at 1.0 g per 100 g syrup, and the effect of level of carbon dosage was studied for samples showing better decolorization capability. The color removal at different carbon dosages were expressed as adsorption isotherms. The amount of adsorbate held per g of activated carbon was expressed as the difference between initial and residual levels of color in IU divided by the carbon dosage in g/100 g syrup. Similar approaches were used by Agudo et al. (2002) and C ¸ elebi and Kincal (2007) in their isotherm expressions. 3. Results and discussion The syrup decolorization performances of the activated carbons are given in Fig. 1 along with their iodine and
3531
100 90
Syrup
80
Iodine
70
Molasses
60 50 40 30 20 10
P1 .5 30 0
30 0P 3
0P 5 30
0P 1. 5 50
50 0P 3
0P 5 50
C 2 75 0
75 0C 5
D C
D C
L2 00
0
L3 20
Syrup, Iodine or Molasses Decolorization, %
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
Fig. 1. Performances of commercial and prepared activated carbons in the decolorization of sugar syrup, molasses and iodine solutions.
molasses decolorization performances. The beet pulp carbon 750C5 is almost as good as the commercially used DCL320 and better than DCL200. Comparing the decolorization of syrup with those of the iodine solution and diluted molasses with all samples of activated carbons, iodine solution decolorization performance seems to be a better indicator of syrup decolorization performance. The BET surface areas of the activated carbons are given in Fig. 2. Comparison of the decolorization performances of the carbons in Fig. 1 with their BET surface areas in Fig. 2 shows that the three best decolorizing carbons, which can remove more than 60% of the color, all have BET surface areas above 800 m2 g1. However, the
sample 750C2, the BET surface area of which is a little less than 400 m2 g1 exhibits almost no decolorization potential; while the samples 500P5 and 500P3, both of which exhibiting around 45% decolorization have substantially different BET surface areas. It appears that the iodine decolorization is the best indicator of the syrup decolorization performance, followed by molasses decolorization, and the BET surface area is a poorer indicator than these two properties. The pH values of 1% suspensions of the activated carbons are given in Table 2. The suspensions of the three best decolorizing carbons, which can remove more than 60% of the color (Fig. 1) have pH values in the range of 4.5 and
Fig. 2. BET surface areas of the commercial and prepared activated carbons.
3532
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
6.7; while the other suspensions are highly alkaline or highly acidic. Moderate pH values seem to favor the decolorization performance. The electrical conductivity values of 1% suspensions of activated carbons were measured as an indicator of their soluble ash contents, and are given in Table 2. The extremely poor decolorization performance of 750C2 seems to be best explained by the extremely high (3–4 orders of magnitude higher) electrical conductivity of its suspension. The comparatively lower BET surface areas, and the acidic pH values and high electrical conductivities of the suspensions of phosphoric acid activated prepared carbons, which may be the reason of their poorer performance, may be due to traces of the acid occluded in the pores in spite of thorough washing right after preparation. The ash content, average pore size, pore volume and densities of the activated carbons are given in Table 2. These properties do not seem to indicate any direction in the syrup decolorization performance. For example, the ash content does not seem to be a good indicator; although the poorest performer has considerably high ash content compared to the others, the beet pulp carbons activated at 500 °C and 300 °C, having ash contents similar to that of 750C5 are much poorer performers. Similarly, the best performing carbons have higher pore volumes, but another carbon with similar pore volume (500P5) has quite poorer performance. The ability of an activated carbon sample to adsorb organic substances is essentially related to its textural and surface properties. The hydrophobic and apolar nature of the adsorbent is essential for a carbon to remove the organic substances responsible for color. Indeed, the samples with best performance are those with nearly neutral pH and low conductivity, both characteristics indicating a more nonpolar surface. Although the number of samples is too small to conclude it from this set of data alone, decolorizing ability is proportional to surface area, as should be expected, for the materials with more apolar surfaces. Beet pulp being abundantly available as a raw material, the effect of sorbent loading (0.2–1.0 g carbon/100 g syrup) was studied for the best performing one of the beet pulp
(IUo-IUeq)/(g C/100 g syrup)
8000 y = 2.1625x R2 = 0.9704
y = 3.365x R2 = 0.9979
6000
DCL320 750C5
4000
DCL200
2000
y = 1.5443x R2 = 0.9836
0 0
500
1000
1500
2000
2500
3000
Residual color, IU
Fig. 3. Sorption isotherms for the better decolorizing carbons.
activated carbons, namely 750C5, and the commercial carbons DCL 320 and DCL 200. The results are shown in Fig. 3 as adsorption isotherms. The isotherms were found to be in the linear range as indicated by the very high regression coefficients for straight lines through the origin. The studied range of coloring substance loadings appears to be low enough for Henry’s law to be valid. Studies at higher activated carbon dosages would be necessary to obtain the whole isotherm. 4. Conclusion The activated carbon obtained from sugar beet pulp by direct carbonization for 5 h at 750 °C under a stream of carbon dioxide has a decolorization performance that can compete with commercially used activated carbons. Among the studied physical characteristics of the activated carbons, iodine decolorization performance appear to be the best indicator of sugar syrup decolorization performance, followed by the molasses decolorization performances, electrical conductivities of their suspensions and their BET surface areas. Suspension pH values between 4.4 and 6.7 and electrical conductivities less than 300 lS are indicated; while the syrup decolorization performance does not seem to be related to the pore volume or size, density or ash contents. Acknowledgements ¨ zer of The authors would like to thank Dr. Ahmet O Fırat University for kindly providing the beet pulp based activated carbons. References Agudo, J.A.G., Cubero, M.T.G., Benito, G.G., Miranda, M.P., 2002. Removal of coloured compounds from sugar solutions by adsorption onto anionic resins: equilibrium and kinetic study. Sep. Purif. Technol. 29, 199–205. Ahmedna, M., Clarke, S.J., Rao, R.M., Marshall, W.E., Johns, M.M., 1997a. Use of filtration and buffers in raw sugar color measurements. J. Sci. Food Agric. 75, 109–116. Ahmedna, M., Johns, M.M., Clarke, S.J., Marshall, W.E., Rao, R.M., 1997b. Potential of agricultural by-product-based activated carbons for use in raw sugar de-colorization. J. Sci. Food Agric. 75, 117–124. Ahmedna, M., Marshall, W.E., Rao, R.M., 2000. Surface properties of granular activated carbons from agricultural by-products and their effects on raw sugar de-colorization. Bioresource Technol. 71, 103–112. Anon, 1994. ICUMSA Colour Determination Method, Method 2. International Commission for Uniform Methods of Sugar Analysis. Aygu¨n, A., Yenisoy-Karakasß, S., Duman, I., 2003. Production of granular activated carbon from fruit stones and nutshells and evaluation of their physical, chemical and adsorption properties. Micropor. Mesopor. Mater. 66, 89–195. Bac¸aoui, A., Yacoubi, A., Dahbi, A., Bennouna, C., Phan Tan Luu, R., Maldonado-Hodar, F.J., Rivera-Utrilla, J., Moreno-Castilla, C., 2001. Optimization of conditions for the preparation of activated carbons from olive-waste cakes. Carbon 39, 425–432. Balcı, S., Dogu, T., Yucel, H., 1994. Characterization of activated carbon produced from almond shell and hazelnut shell. J. Chem. Technol. Biotechnol. 60, 419–426.
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533 _ Kincal, N.S., 2007. Color formation in wheat starch based C ¸ elebi, I., glucose syrups and use of commercially available and laboratoryprepared agricultural waste-based activated carbons for decolorization. Sep. Sci. Technol. 42, 1761–1773. Godshall, M.A., 1997. Symposium on Sugar Color SIT, Part Two – Color Analysis SIT, 46, 193–211. Kearsley, M.W., Dziedzic, S.Z., 1995. Handbook of Starch Hydrolysis Products and Their Derivatives. Chapman Hill, London. ¨ zer, A., Tanyıldızı, M.S., Tu¨men, F., 1998. Study of cadmium O adsorption from aqueous solution on activated carbon from sugar beet pulp. Environ. Technol. 19, 1119–1125. ¨ zer, A., C O ¸ am, G., 2002. The characterization of activated carbons from sugar beet pulp impregnated with phosphoric acid. J. Sci. Eng. 14 (1), 191–197 (in Turkish).
3533
Pendyal, B., Jones, M.M., Marshall, W.E., Ahmedna, M., Rao, R.M., 1999. Removal of sugar colorants by granular activated carbons made from binders and agricultural by-products. Bioresource Technol. 69, 45–51. Smisek, M., Cerny, S., 1970. Active Carbon Manufacture, Properties and Applications. Elsevier, New York. van der Poel, P.W., Schiweck, H., Schwartz, T., 1998. Sugar Technology: Beet and Cane Sugar Manufacture. Verlag Dr. Albert Bartens KG, Berlin, p.516. Yun, C.H., Park, Y.H., Park, C.R., 2001. Effects of pre-carbonization on porosity development of activated carbons from rice straw. Carbon 39, 559–567.
Bioresource Technology 99 (2008) 3528–3533
Decolorization of sugar syrups using commercial and sugar beet pulp based activated carbons H.L. Mudoga, H. Yucel, N.S. Kincal
*
Department of Chemical Engineering, Middle East Technical University, Ankara 06531, Turkey Received 3 November 2006; received in revised form 10 July 2007; accepted 28 July 2007 Available online 7 September 2007
Abstract Sugar syrup decolorization was studied using two commercial and eight beet pulp based activated carbons. In an attempt to relate decolorizing performances to other characteristics, surface areas, pore volumes, bulk densities and ash contents of the carbons in the powdered form; pH and electrical conductivities of their suspensions and their color adsorption properties from iodine and molasses solution were determined. The color removal capabilities of all carbons were measured at 1/100 (w/w) dosage, and isotherms were determined on better samples. The two commercial activated carbons showed different decolorization efficiencies; which could be related to their physical and chemical properties. The decolorization efficiency of beet pulp carbon prepared at 750 °C and activated for 5 h using CO2 was much better than the others and close to the better one of the commercial activated carbons used. It is evident that beet pulp is an inexpensive potential precursor for activated carbons for use in sugar refining. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Sugar syrup; Decolorization; Activated carbon; Beet pulp
1. Introduction Activated carbon adsorbents having a large internal surface area and relatively nonpolar surface are primarily being used to remove organic compounds from both liquid and gaseous streams (Smisek and Cerny, 1970). One significant application makes use of its decolorization properties in several areas including beet and cane sugar refining. The presence of colored compounds in the sugar syrups results from reactions occurring during the production. The chemical structure of some of these coloring materials is quite complex and difficult to determine in many cases. The most significant colored substances that develop during sugar processing can be classified in three general groups: (a) melanins (b) melanoidins and (c) caramels (Kearsley and Dziedzic, 1995). Conventional methods to prevent formation of these compounds include treatment with sulfur
*
Corresponding author. Tel.: +90 312 2102617; fax: +90 312 2102600. E-mail address: [email protected] (N.S. Kincal).
0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.07.058
dioxide or hydrogen peroxide, however, although these treatment steps retard the formation of some of these compounds, they cannot eliminate them altogether. In order to produce high quality white sugar or when very good clarity is required in the syrup products such as those used in colorless soda drinks, a treatment of juice with activated carbon or ion exchange resins is usually required (van der Poel et al., 1998). Activated carbons can be produced from various carbon containing species by physical and chemical activation methods. Most common feed stocks for commercial production may be given as coal, peat and wood (Smisek and Cerny, 1970). Other commercial and potential feed stocks include renewable agricultural by-products, namely nut shells (hazelnut, walnut, pecan, etc.), fruit kernels/seeds/ stones (apricot, apple, etc.), hulls (soybean, rice, etc.), pulps ¨ zer (paper, apple, beet) and olive stone (Balcı et al., 1994; O et al., 1998; Bac¸aoui et al., 2001; Yun et al., 2001; Aygu¨n et al., 2003). Ahmedna and co-workers (Ahmedna et al., 1997b, 2000) extensively studied the preparation, properties and
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
decolorization applications of activated carbons produced from agricultural by-products (sugar cane bagasse, rice hulls, rice straw, pecan shells using different binders (beet and sugar cane molasses, corn syrup and coal tar). They found that the type of by-product, binder used and the activation method determine the physical and chemical properties and thus decolorization properties of granular activated carbons. It has been demonstrated previously that activated carbon can be produced from beet sugar pulp which is a by¨ zer et al., 1998 and product of beet sugar processing (O 2002). Exploration of the sugar decolorization properties of carbons made from sugar beet pulp was undertaken to find another important outlet for this beet sugar plant waste primarily being used as cattle food. The aim of the present work was to compare the decolorization capabilities of beet pulp activated carbons produced locally in the laboratory scale with some commercial sugar decolorizing activated carbons and to look for possible correlations between the physical properties of activated carbons and their sugar syrup decolorization capabilities; with the hope of shedding some light on potential studies of preparing activated carbons for decolorization.
2. Methods 2.1. Syrup The syrup (thick juice) was obtained from the Ankara Sugar Factory. A series of grab samples of the syrup were collected at the point just before the sugar-boiling step in the factory. The sampling was done every 6 h during a period of 48 h. The samples were stored in the refrigerator in a plastic container. The properties of the syrup, determined according to Anon (1994) are given in Table 1. The refractive index values of the syrup (determined using Abbe BS 60/70 Refractometer) were 1.4450 at its natural pH of 7.7; 1.3630 at pH 4 and 1.4048 at pH 9. The indicator value, I.V., defined as the ratio of the refractive indices at pH value of 9 to that at pH 7 (Kearsley and Dziedzic, 1995) can be estimated on the basis of these values to be 1.04, indicating the major colorants in the syrup to be mel-
3529
Table 1 Properties of the syrup Color, ICUMSA units Brix, % Polarity, % Purity, % Ash, % Absorbance pH
3149 66.83 61.73 92.37 4.01 2.4376 7.7
anoidins (I.V. 1.0–1.2) or caramels (I.V. 1.0–1.5) Godshall (1997). 2.2. Commercial activated carbons Two reference commercial carbons, selected on basis of the manufacturer’s recommendations, were obtained from Waterlink Sutcliffe Carbons, England. The carbons used were DCL320 (wood base, activated with phosphoric acid) and DCL200 (coal base, steam activated). 2.3. Beet pulp activated carbons Eight different beet pulp activated carbon samples pre¨ zer et al. (1998) and O ¨ zer and C pared by O ¸ am (2002) were ¨ zer et al., 1998) by used. Two samples were prepared (O direct carbonization and were activated at 750 °C under a stream of carbon dioxide for 2 h or 5 h (750C2 and ¨ zer and C 750C5). Six samples were prepared (O ¸ am, 2002) by first impregnating with phosphoric acid overnight, followed by activating under a stream of nitrogen at 300 °C or 500 °C for 1.5–5 h (300P1.5, 300P3, 300P5, 500P1.5, 500P3, 500P5). In the coding of the samples, the first three digits indicate the activation temperature; the following letter is C for the directly carbonized – CO2 activated samples and P for the phosphoric acid impregnated – N2 activated samples; and the number following the letter indicates the time of activation in hours. 2.4. Methods of characterization of the activated carbons The commercial and beet pulp based activated carbons are characterized with the methods below and the results are given in Table 2.
Table 2 Some properties of the activated carbons Activated carbon
pH of susp.
Electrical cond. of susp, lS
Ash, mg g1
Pore size, nm
Pore volume, cm3 g1
Density, g cm3
DCL 320 DCL 200 750C5 750C2 500P5 500P3 500P1.5 300P5 300P3 300P1.5
4.5 6.7 6 10.2 2.8 2.8 2.8 2.8 2.8 2.8
285 17.1 24.8 972,000 3200 2600 2410 1820 2320 1870
36 168 183 348 210 183 193 197 192 202
3.6 2.6 2.2 3.8 3.3 2.7 2.6 3.5 3.4 4.3
1.30 0.54 0.65 0.25 0.55 0.27 0.22 0.28 0.28 0.20
0.28 0.35 0.28 0.25 0.67 0.60 0.58 0.62 0.71 0.61
3530
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
2.4.1. Surface area, pore volume and pore size The BET surface area of the samples were determined by nitrogen adsorption at 77 K using Micrometrics ASAP 2000 automatic surface area analyzer after the samples were dried under vacuum at 100 °C for 3 h to remove moisture. The surface area was determined using the N2 adsorption data at 77 K in the relative pressure range of 0.05–0.25, and taking the cross-sectional area of the nitrogen molecule to be 0.162 nm2. The total pore volume was based on the N2 amount adsorbed at relative pressure greater than 0.98, and taking the liquid molar volume of N2 to be 34.65 cm3/g at 77 K. The average pore diameter, D, was calculated from the equation based on uniform cylindrical pores as given below
D¼
ð4Þ ðPore volumeÞ BET area
2.4.2. Bulk density Powdered activated carbon samples were placed in graduated cylinders, tapped several times until constant volume (10 mL) was obtained, and measuring the weight, the bulk density was calculated as the ratio of weight to volume. 2.4.3. Moisture and ash contents Moisture content was determined by the measurement of dry weight after bringing the sample (about 1 g) to constant weight in an air circulation oven at 115 °C; and ash content was determined based on the weight of the residue obtained after 1.5 h in a furnace at 950 °C under air circulation. 2.4.4. pH Suspensions of the samples (1% w/w) in distilled water were heated to 90 °C, subjected to continuous stirring for 20 min, cooled to room temperature and the pH of the solution was determined using a Leeds Northrup 7415 pH-meter (Ahmedna et al., 1997b). 2.4.5. Electrical conductivity Suspensions of the samples (1% w/w) in distilled water were subjected to continuous stirring for 20 min, and the electrical conductivity of the solution was measured (Ahmedna et al., 1997b) at 25.6 °C using a Jenway 4020 conductivity-meter. 2.4.6. Iodine test, molasses test and syrup decolorization experiments These experiments were all similar in the sense that an iodine solution, diluted molasses or sugar syrup was contacted with the active carbon; the color of this sample was compared to that of a blank prepared and treated the same way without the activated carbon (Pendyal et al., 1999; Ahmedna et al., 1997a, 1997b). For the iodine test, 100 mL of stock solution (2.7 g I2 and 4.1 g KI in 1 L of solution) was added to 0.5 g activated carbon in 10 mL of 5% HCl solution, 5 min contact time was allowed, the sample was filtered through Whatman No. 110 filter paper and
the percentage color removal was based on the volumes of 0.1 M sodium thiosulphate required (using starch as the indicator) for the sample and the blank filtrates. For the molasses test, the test solution was prepared by dissolving 10 g of sugar beet molasses and 15 g of Na2HPO4 in 500 mL of water and sufficient H3PO4 to make pH 6.5, diluting to 1 L and filtering through a thin layer of filter aid (diatomaceous earth). Fifty milliliter of this solution was added to 0.5 g of activated carbon; the mixture was well stirred, and was placed on a heating plate along with the blank and brought to boil. The percentage color removal was based on the absorbance readings taken 420 nm using a Hitachi U-3200 Spectrophotometer after the sample and the blank were filtered through Whatman No. 4 filter paper. For the syrup decolorization experiments, syrup samples were adjusted to pH 7.0 using 0.1 M HCl and Thymol Blue indicator, two 100 g of samples were brought to 80 °C in a water bath with shaking at 160 rpm, 0.2–1.0 g of activated carbon was added to one of the samples, and 20 min was allowed, which had been verified to be sufficient for the attainment of equilibrium. Then, the samples were clarified in two steps. The first filtration was through Whatman No. 110 filter paper, followed by addition of 1% diatomaceous earth and heating to 70 °C for reducing the viscosity. The second filtration was through Nalgene Filtering assembly with a 0.45 lm membrane for removal of very fine carbon particles generated by attrition. The absorbance was determined at 420 nm on the clarified sample. The color was expressed in ICUMSA units (IU) defined according to equation (Anon, 1994); IU ¼ 1000 A=ðb cÞ; where A = absorbance at 420 nm of the test sample b = length (cm) of the adsorbing path c = concentration (g sugar/mL) of the test sample The IU is therefore a measure of the weight ratio of a hypothetical coloring substance to the sugar present. All activated carbon samples were studied for their sugar syrup decolorization properties at 1.0 g per 100 g syrup, and the effect of level of carbon dosage was studied for samples showing better decolorization capability. The color removal at different carbon dosages were expressed as adsorption isotherms. The amount of adsorbate held per g of activated carbon was expressed as the difference between initial and residual levels of color in IU divided by the carbon dosage in g/100 g syrup. Similar approaches were used by Agudo et al. (2002) and C ¸ elebi and Kincal (2007) in their isotherm expressions. 3. Results and discussion The syrup decolorization performances of the activated carbons are given in Fig. 1 along with their iodine and
3531
100 90
Syrup
80
Iodine
70
Molasses
60 50 40 30 20 10
P1 .5 30 0
30 0P 3
0P 5 30
0P 1. 5 50
50 0P 3
0P 5 50
C 2 75 0
75 0C 5
D C
D C
L2 00
0
L3 20
Syrup, Iodine or Molasses Decolorization, %
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
Fig. 1. Performances of commercial and prepared activated carbons in the decolorization of sugar syrup, molasses and iodine solutions.
molasses decolorization performances. The beet pulp carbon 750C5 is almost as good as the commercially used DCL320 and better than DCL200. Comparing the decolorization of syrup with those of the iodine solution and diluted molasses with all samples of activated carbons, iodine solution decolorization performance seems to be a better indicator of syrup decolorization performance. The BET surface areas of the activated carbons are given in Fig. 2. Comparison of the decolorization performances of the carbons in Fig. 1 with their BET surface areas in Fig. 2 shows that the three best decolorizing carbons, which can remove more than 60% of the color, all have BET surface areas above 800 m2 g1. However, the
sample 750C2, the BET surface area of which is a little less than 400 m2 g1 exhibits almost no decolorization potential; while the samples 500P5 and 500P3, both of which exhibiting around 45% decolorization have substantially different BET surface areas. It appears that the iodine decolorization is the best indicator of the syrup decolorization performance, followed by molasses decolorization, and the BET surface area is a poorer indicator than these two properties. The pH values of 1% suspensions of the activated carbons are given in Table 2. The suspensions of the three best decolorizing carbons, which can remove more than 60% of the color (Fig. 1) have pH values in the range of 4.5 and
Fig. 2. BET surface areas of the commercial and prepared activated carbons.
3532
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533
6.7; while the other suspensions are highly alkaline or highly acidic. Moderate pH values seem to favor the decolorization performance. The electrical conductivity values of 1% suspensions of activated carbons were measured as an indicator of their soluble ash contents, and are given in Table 2. The extremely poor decolorization performance of 750C2 seems to be best explained by the extremely high (3–4 orders of magnitude higher) electrical conductivity of its suspension. The comparatively lower BET surface areas, and the acidic pH values and high electrical conductivities of the suspensions of phosphoric acid activated prepared carbons, which may be the reason of their poorer performance, may be due to traces of the acid occluded in the pores in spite of thorough washing right after preparation. The ash content, average pore size, pore volume and densities of the activated carbons are given in Table 2. These properties do not seem to indicate any direction in the syrup decolorization performance. For example, the ash content does not seem to be a good indicator; although the poorest performer has considerably high ash content compared to the others, the beet pulp carbons activated at 500 °C and 300 °C, having ash contents similar to that of 750C5 are much poorer performers. Similarly, the best performing carbons have higher pore volumes, but another carbon with similar pore volume (500P5) has quite poorer performance. The ability of an activated carbon sample to adsorb organic substances is essentially related to its textural and surface properties. The hydrophobic and apolar nature of the adsorbent is essential for a carbon to remove the organic substances responsible for color. Indeed, the samples with best performance are those with nearly neutral pH and low conductivity, both characteristics indicating a more nonpolar surface. Although the number of samples is too small to conclude it from this set of data alone, decolorizing ability is proportional to surface area, as should be expected, for the materials with more apolar surfaces. Beet pulp being abundantly available as a raw material, the effect of sorbent loading (0.2–1.0 g carbon/100 g syrup) was studied for the best performing one of the beet pulp
(IUo-IUeq)/(g C/100 g syrup)
8000 y = 2.1625x R2 = 0.9704
y = 3.365x R2 = 0.9979
6000
DCL320 750C5
4000
DCL200
2000
y = 1.5443x R2 = 0.9836
0 0
500
1000
1500
2000
2500
3000
Residual color, IU
Fig. 3. Sorption isotherms for the better decolorizing carbons.
activated carbons, namely 750C5, and the commercial carbons DCL 320 and DCL 200. The results are shown in Fig. 3 as adsorption isotherms. The isotherms were found to be in the linear range as indicated by the very high regression coefficients for straight lines through the origin. The studied range of coloring substance loadings appears to be low enough for Henry’s law to be valid. Studies at higher activated carbon dosages would be necessary to obtain the whole isotherm. 4. Conclusion The activated carbon obtained from sugar beet pulp by direct carbonization for 5 h at 750 °C under a stream of carbon dioxide has a decolorization performance that can compete with commercially used activated carbons. Among the studied physical characteristics of the activated carbons, iodine decolorization performance appear to be the best indicator of sugar syrup decolorization performance, followed by the molasses decolorization performances, electrical conductivities of their suspensions and their BET surface areas. Suspension pH values between 4.4 and 6.7 and electrical conductivities less than 300 lS are indicated; while the syrup decolorization performance does not seem to be related to the pore volume or size, density or ash contents. Acknowledgements ¨ zer of The authors would like to thank Dr. Ahmet O Fırat University for kindly providing the beet pulp based activated carbons. References Agudo, J.A.G., Cubero, M.T.G., Benito, G.G., Miranda, M.P., 2002. Removal of coloured compounds from sugar solutions by adsorption onto anionic resins: equilibrium and kinetic study. Sep. Purif. Technol. 29, 199–205. Ahmedna, M., Clarke, S.J., Rao, R.M., Marshall, W.E., Johns, M.M., 1997a. Use of filtration and buffers in raw sugar color measurements. J. Sci. Food Agric. 75, 109–116. Ahmedna, M., Johns, M.M., Clarke, S.J., Marshall, W.E., Rao, R.M., 1997b. Potential of agricultural by-product-based activated carbons for use in raw sugar de-colorization. J. Sci. Food Agric. 75, 117–124. Ahmedna, M., Marshall, W.E., Rao, R.M., 2000. Surface properties of granular activated carbons from agricultural by-products and their effects on raw sugar de-colorization. Bioresource Technol. 71, 103–112. Anon, 1994. ICUMSA Colour Determination Method, Method 2. International Commission for Uniform Methods of Sugar Analysis. Aygu¨n, A., Yenisoy-Karakasß, S., Duman, I., 2003. Production of granular activated carbon from fruit stones and nutshells and evaluation of their physical, chemical and adsorption properties. Micropor. Mesopor. Mater. 66, 89–195. Bac¸aoui, A., Yacoubi, A., Dahbi, A., Bennouna, C., Phan Tan Luu, R., Maldonado-Hodar, F.J., Rivera-Utrilla, J., Moreno-Castilla, C., 2001. Optimization of conditions for the preparation of activated carbons from olive-waste cakes. Carbon 39, 425–432. Balcı, S., Dogu, T., Yucel, H., 1994. Characterization of activated carbon produced from almond shell and hazelnut shell. J. Chem. Technol. Biotechnol. 60, 419–426.
H.L. Mudoga et al. / Bioresource Technology 99 (2008) 3528–3533 _ Kincal, N.S., 2007. Color formation in wheat starch based C ¸ elebi, I., glucose syrups and use of commercially available and laboratoryprepared agricultural waste-based activated carbons for decolorization. Sep. Sci. Technol. 42, 1761–1773. Godshall, M.A., 1997. Symposium on Sugar Color SIT, Part Two – Color Analysis SIT, 46, 193–211. Kearsley, M.W., Dziedzic, S.Z., 1995. Handbook of Starch Hydrolysis Products and Their Derivatives. Chapman Hill, London. ¨ zer, A., Tanyıldızı, M.S., Tu¨men, F., 1998. Study of cadmium O adsorption from aqueous solution on activated carbon from sugar beet pulp. Environ. Technol. 19, 1119–1125. ¨ zer, A., C O ¸ am, G., 2002. The characterization of activated carbons from sugar beet pulp impregnated with phosphoric acid. J. Sci. Eng. 14 (1), 191–197 (in Turkish).
3533
Pendyal, B., Jones, M.M., Marshall, W.E., Ahmedna, M., Rao, R.M., 1999. Removal of sugar colorants by granular activated carbons made from binders and agricultural by-products. Bioresource Technol. 69, 45–51. Smisek, M., Cerny, S., 1970. Active Carbon Manufacture, Properties and Applications. Elsevier, New York. van der Poel, P.W., Schiweck, H., Schwartz, T., 1998. Sugar Technology: Beet and Cane Sugar Manufacture. Verlag Dr. Albert Bartens KG, Berlin, p.516. Yun, C.H., Park, Y.H., Park, C.R., 2001. Effects of pre-carbonization on porosity development of activated carbons from rice straw. Carbon 39, 559–567.