Calcium Phosphate Flocs and the Clarification of Sugar Cane Juice

Jan 9, 2015 - The conventional method for juice clarification in Australian sugar factories is termed simple defecation. In the defecation ...... We t...
39 downloads 41 Views 2MB Size
Article pubs.acs.org/JAFC

Calcium Phosphate Flocs and the Clarification of Sugar Cane Juice from Whole of Crop Harvesting Caroline C. D. Thai,* Lalehvash Moghaddam, and William O. S. Doherty Centre for Tropical Crops and Biocommodities, Queensland University of Technology, Brisbane, QLD 4000, Australia S Supporting Information *

ABSTRACT: Sugar cane biomass is one of the most viable feedstocks for the production of renewable fuels and chemicals. Therefore, processing the whole of crop (WC) (i.e., stalk and trash, instead of stalk only) will increase the amount of available biomass for this purpose. However, effective clarification of juice expressed from WC for raw sugar manufacture is a major challenge because of the amounts and types of non-sucrose impurities (e.g., polysaccharides, inorganics, proteins, etc.) present. Calcium phosphate flocs are important during sugar cane juice clarification because they are responsible for the removal of impurities. Therefore, to gain a better understanding of the role of calcium phosphate flocs during the juice clarification process, the effects of impurities on the physicochemical properties of calcium phosphate flocs were examined using small-angle laser light scattering technique, attenuated total reflectance Fourier transformed infrared spectroscopy, and X-ray powder diffraction. Results on synthetic sugar juice solutions showed that the presence of SiO2 and Na+ ions affected floc size and floc structure. Starch and phosphate ions did not affect the floc structure; however, the former reduced the floc size, whereas the latter increased the floc size. The study revealed that high levels of Na+ ions would negatively affect the clarification process the most, as they would reduce the amount of suspended particles trapped by the flocs. A complementary study on prepared WC juice using cold and cold/intermediate liming techniques was conducted. The study demonstrated that, in comparison to the one-stage (i.e., conventional) clarification process, a two-stage clarification process using cold liming removed more polysaccharides (≤19%), proteins (≤82%), phosphorus (≤53%), and SiO2 (≤23%) in WC juice but increased Ca2+ (≤136%) and sulfur (≤200%). KEYWORDS: whole of cane harvesting, trash, tops, stalk, sugar cane juice, clarification, calcium phosphate



increases the proportion of sucrose lost to final molasses.4,5 Recent work by the authors6,7 on juice expressed from green cane containing only half of the initial trash present (GE) confirmed the issues highlighted by the previous authors4,5 but noted that satisfactory clarification (i.e., separation of solid from liquid) performance can be achieved with juices expressed from green sugar cane containing 75% of the fiber is in the range of 350 mg/kg P2O5 was added compared to the addition of 300 mg/kg P2O5 (Table 2). Supplying phosphoric acid to the sugar solution provided

RESULTS AND DISCUSSION

Effect of Sodium on Calcium Phosphate Flocs. The scattering exponent and average size of the flocs formed in the presence of Na+ ions are shown in Table 2. A linear decrease in the scattering exponent (from 2.50 to 2.25) with increasing Na+ 1576

DOI: 10.1021/jf502229f J. Agric. Food Chem. 2015, 63, 1573−1581

Article

Journal of Agricultural and Food Chemistry

Figure 3. X-ray powder diffraction patterns of calcium phosphate flocs: (a) 5 mg/kg of NaCl added to synthetic juice; (b) synthetic juice; (c) 200 mg/kg starch added to synthetic juice.

representation of the variation of the interatomic distances between P−O−P structural units, which is related to the arrangement of the phosphorus molecules.14 The addition of phosphate broadens the XRD halo relative to the control, indicating that the calcium phosphate phase is far less ordered (smaller particles, i.e., more amorphous) than that of the control. The increased amorphous nature of the calcium phosphate particles will increase the amount of impurities removed because of an increase in surface area.18 The crystalline peaks associated with starch at 2θ values of 15° and 17°, which is typical of the A-type pattern,25 are highest when an increasing amount of starch was added prior to lime neutralization and lowest when the Na+ level was increased prior to neutralization. Starch is electronegative, so cations attract starch hydroxyl groups and destabilize starch granules, whereas anions tend to repel starch hydroxyl groups and stabilize starch granules.27 Chiotelli27 and co-workers postulated that Na+ ions tend to penetrate starch granules and replace hydrogen ions, which reduces the water phase. When the electrostatic driving force of the anions in solution exceeds the repulsive force associated with the Na+ ions, the anions enter the starch granule and rupture the hydrogen bond between starch molecules. This results in gelatinization, and so the amount of starch trapped (or rather removed) by the calcium phosphate flocs is reduced. The presence of starch in sugar cane juice processed in the factory can create boiling and viscosity problems, filterability impairment, and poor molasses exhaustion.28 As a consequence, high levels of Na+ ions in juice have two negative effects on clarification: (i) increasing the amount of starch molecules present in the CJ and (ii) loosening the floc structure, thereby reducing the amount of suspended particles not trapped by the flocs.12 Attenuated Total Reflectance Fourier Transform Infrared spectroscopy. The ATR-FTIR spectra of calcium phosphate precipitate formed in the control synthetic juice and

additional phosphate ions, which interacted with calcium ions from lime saccharate and resulted in an increase in the growth of calcium phosphate particles. The results show that there is a wide P2O5 range over which there is stabilization of floc structure. It is unclear whether the stabilization is due to the changes in surface charge of the calcium phosphate, via changes in proportion of the phases that constitute the floc.9 In a typical clarification process in a sugar factory, juice phosphate concentration >300 mg/kg as P2O5 does not influence clarification performance significantly.9 Effects of Starch on Calcium Phosphate Flocs. As shown in Table 2, increasing the amount of starch did not change the floc structure, but slightly affected the floc size. At the lowest dosage of starch (100 mg/kg), the floc size of the formed flocs was 390 μm. Increasing starch concentration to 500 mg/kg dropped the size to approximately 270 μm. The reduction in floc size may be associated with starch molecules adhering to the surface of microflocs inhibiting growth, without affecting packing arrangement.25 Starch molecules when present to a significant extent in juice do affect sugar crystallization, so the same may occur with calcium phosphate.25 X-ray Powder Diffraction Results. XRD was used to identify the phases of calcium phosphate precipitates formed in the presence of Na+ ions and starch, as these two components were shown to most strongly influence the scattering exponent and size of flocs. The XRD patterns of selected precipitates are shown in Figure 3. Calcite (2θ, 23.06°, 29.46°, 35.98°, 39.46°, 43.18°, 47.54°, 48.56°, and 57.40°) is the main crystal phase in all of the samples. The prominent halo in the 2θ region between 26° and 36° is characteristic of an amorphous compound,26 which is presumably calcium phosphate, because energy dispersive spectroscopy gave a mass ratio of Ca to P similar to that of tricalcium phosphate. As was reported for SiO2, the variation in the size of the halo may be a 1577

DOI: 10.1021/jf502229f J. Agric. Food Chem. 2015, 63, 1573−1581

Article

Journal of Agricultural and Food Chemistry

of phosphate, NaCl, and SiO2 reduces the amount of starch trapped by the calcium phosphate flocs. The peak detected at 1643 cm−1 in the spectra (Figure 4ii) is correlated to the H−O−H symmetric bending vibration.31 The bands observed at 1457, 1416, and 876 cm−1 are derived from two antisymmetric stretching and bending vibrations of CO32−, respectively. The presence of these peaks is due to the formation of CaCO3.32 The partially resolved shoulder at 1336 cm−1 detected from the control and synthetic juice, in which starch was added, is assigned to the C−H bending of CH2 groups in starch.30 This peak was not detected in the spectra obtained from the analysis of other samples. The other peaks present as shoulders at 1143, 1074, and 934 cm−1, characteristic of starch, were detected in the control (Figure 4ii(a) and calcium phosphate sample (Figure 4ii(b)). These peaks are associated with C−O, C−C stretching, C−O−H bending, and C−O−C vibrations in starch molecular structure, respectively.29 Each spectrum shown in Figure 4ii reveals an intense peak between 1200 and 896 cm−1, which is related to P−O stretching derived from the phosphate group.32 The position of this peak shifted in the spectra of the samples, which may be related to the salt added during flocculation and the change in crystallinity of the calcium phosphate floc. In summary, the results obtained using ATR-FTIR correlate with the XRD data, in terms of the relative proportions of starch in the various calcium phosphate samples. The calcium phosphate flocs are not effective in trapping starch molecules. However, increasing the concentration of phosphate increases the floc size and presumably the amount of flocs formed to improve clarification performance.6 Thai et al.6 investigated the clarification performance of sugar cane juice derived from burnt cane (BE) and green cane with half of the trash removed (GE) and concluded that the latter can be clarified with acceptable juice turbidity when up to 100 mg/kg of P2O5 is added. Clarification of Sugar Cane Juice Derived from Whole Sugar Cane Crop. The compositions of sugar cane juices obtained from the factory (SJ) and those prepared from factory juice and hand-cut trash material (prepared WC) are shown in Table 3. The inclusion of juice derived from trash contributed significantly to the level of soluble proteins, polysaccharides, reducing sugars, calcium, phosphate, magnesium, and silicon ions. The polysaccharide content of juice derived from WC was higher than that of SJ by ∼36% (Table 3). Starch, polysaccharides, and proteins are polar and thus would influence the coagulation and flocculation processes of juice particles and their floccing ability.18 Furthermore, if a higher proportion of these impurities is carried through the various processing stages, they would increase the viscosity of molasses (sugar syrup), reduce the rate of sucrose crystallization, and produce colored compounds.33 The results of the soluble inorganic ions (Table 3) show that Ca2+ and phosphate contents in the WC juices were higher than those in the SJ by ∼30% and ∼52%, respectively. As a result, Ca2+ and phosphate ions are expected to increase the coagulation of WC juice particles and hence improve clarification compared to juice expressed from SJ. However, it is important to note that the proportion of Ca2+ ions in WC juice that takes part in particle aggregation is not known.24 Clarification Performance. As shown in Table 3, the onestage (i.e., conventional) clarification processes using cold and cold/intermediate liming resulted in relatively high CJ turbidities of 15.7 and 29.0 NTU, respectively. For good

in the presence of starch, SiO2, NaCl, and phosphate are shown in Figure 4. The broad band observed at 3325 cm−1 is

Figure 4. ATR-FTIR spectra ((i) 3800−2600 cm−1 and (ii) 1650−650 cm−1) of calcium phosphate precipitate formed during juice flocculation of synthetic juice: (a) control; (b) 200 mg/kg starch; (c) 25 mg/kg SiO2; (d) 5 mg/kg NaCl; (e) 250 mg/kg phosphate.

associated with the O−H group stretch vibration of calcium phosphate.29 The presence of adsorbed water may also be associated with this peak. Figure 4I shows that the addition of SiO2 significantly reduced the intensity of this broad peak. This peak was also reduced, albeit to a lesser extent, for the calcium phosphate precipitates derived from the synthetic juice in the presence of NaCl and phosphate ions. It is therefore likely that these added compounds affect the properties of calcium phosphate at differing extents. The band at 2932 cm−1 is related to the C−H stretching vibration of methyl groups, which is related to the presence of starch and/or dextran.30 From Figure 4i, the intensity of this peak is highest in the synthetic juice obtained in the presence of starch (Figure 4i(b) followed by the control (Figure 4i(a) in comparison to the intensities of the other samples. This confirms that the addition 1578

DOI: 10.1021/jf502229f J. Agric. Food Chem. 2015, 63, 1573−1581

Article

Journal of Agricultural and Food Chemistry Table 3. Clarification Results of Prepared WC Juicea at Tully Mill description/liming technique clarification juice pH additives pH after liming alkali vol (mL) floc size (visual) settling rate, cm/min final mud level (%) turbidity (NTU) polysaccharides (mg/kg)b proteins (mg/kg)b Ca (mg/kg)b Mg (mg/kg)b P (mg/kg)b Si (mg/kg)b S (mg/kg)b a

SJ

prepared WC

cold liming

cold/ intermediate

not treated 5.8

not treated 5.5

normal 5.5

normal 5.6

7.1 9.8 small 30

47300

64300

30 15.7 41700

2460 1010 924 924 773 840

15400 1310 1560 1400 533 1150

6220 2120 1240 387 314 1170

cold liming

cold/ intermediate

cold liming

cold/intermediate

6.2/7.8 3.8/12.0 small 10

two-stage 5.2 H2SO4, pH 3.5 7.1 13.9 large flocs floated

two-stage 5.3 H2SO4, pH 3.5 6.2/7.8 11.1/14.2 large flocs floated

two-stage 5.3 H2SO4, pH 3.5 10 mg/kg LT525 7.1 13.4 large flocs floated

two-stage 5.3 H2SO4, pH 3.5 10 mg/kg LT525 6.2/7.8 5.0/13.2 large flocs floated

15.0 29.0 52600

5.2 44000

6.7 46700

4.8 42900

8.1 44700

7080 3380 1220 460 338 1220

3000 5000 1270 246 262 3490

5290 5850 1300 220 276 3660

2800 4960 1240 240 256 3490

5170 5570 1220 214 260 3510

Composite juice samples. bUnits are on dry solids.

pH level of 7.1 (instead of 7.8) reduced the proportion of the protein and polysaccharide contents in CJ. The reason for this is not known. It was observed that there was no benefit in using the coagulant during the first stage of the two-stage process. Amino acids formed by protein denaturation in sugar cane juice react with reducing sugars (e.g., glucose and fructose) via the Maillard reaction to form undesirable high molecular weight dark colored compounds that become incorporated into the sucrose crystals.37 Thus, the two-stage process, which reduced proteins (and polysaccharides) may thus affect the proportions of colored bodies formed during sugar manufacturing. Previous studies by Lindeman and O’Shea38,39 revealed a strong association between colored compounds and polysaccharides in sucrose crystals. In comparison to the one-stage clarification process, the twostage clarification process resulted in CJ with higher reducing sugar content by ≤28% (≥2.0% on dry solids) and sucrose levels were lower by ≤4.5% (∼93% on dry solids) (please refer to the Supporting Information for glucose, fructose, and sucrose concentrations in WC juices). The results showed that sucrose degradation occurred, due to the use of acid to drop the juice pH to 3.5 during the pretreatment step. Table 3 shows that lime consumption was notably higher using the cold/intermediate liming technique than cold liming because additional lime saccharate was required for neutralization. There were no changes in the proportion of Mg2+ ion concentrations irrespective of liming technique or clarification process. In general, cold liming resulted in lower inorganic ion levels than the cold/intermediate liming technique. For both cold and cold/intermediate liming techniques, there was an improved removal of phosphate and silicon using the two-stage clarification process compared to the one-stage clarification process by ≤53% and ≤23%, respectively, whereas, Ca2+ and sulfur levels were considerably higher by ≤136% and ≤200%, respectively. The higher the sulfur content in juice treated using the dual clarification process is due to the addition of H2SO4 during pretreatment. The high levels of these inorganic components, particularly Ca2+, phosphate, silicon, and sulfur, will cause severe scaling in

clarification, this value is usually