Introduction

Carbon decolorization still extensively applied in the field of food production such as sugar refining to remove color precursors from different types of juice (Bourzutsckky 2005; Chou 2010). The reactions that take place during the production are the main reason for such colored compounds in the sugar manufacturing. Furthermore, it is not possible to determine the chemical structure of the relevant coloring compounds and difficult to control in many cases. Melanin, melanoidins and caramel types are the most proper colored components formed during sugar carbonation process (Singh et al. 2011). There are twenty-one phenolic compounds in the cane sugar plantation according to Farber and Carpenter, who also reported, four colored types in refined sugar, and nearly ten in raw sugar (Farber and Carpenter 1971). Moreover, the red-colored quinones, indole and melanin derivatives, result from the oxidation of phenolic compounds. Therefore, we can conclude that the polyphenols are responsible for color formation in sugar processing (Gross and Coombs 1976). With increasing energy costs related to the sugar production and environmental impacts, it is highly important to re-estimate the old decolorization technologies with respect to their virtues. These techniques include affination, carbonation or phosphatation that are classified as a primary decolorization stage, and secondary expensive decolorization technique-based ion-exchange resins (Singh et al. 2011). Moreover, treatment with hydrogen peroxide or sulfur dioxide have some toxic effect overdose; therefore, to produce high juice clarity and good-quality white sugar products, a new cost-reduction material used for treatment of juice like powdered activated carbon is developed (van der Poel et al. 1998).

Existence of large specific surface area (up to 4000 m2/g) makes powdered activated carbon (PAC) a superior food decolorant and efficient adsorbent for the removal of contaminants from wastewater streams (Gupta et al. 2009; Gupta and Rastogi 2009; Rao et al. 2010; Lu et al. 2011).

Powdered activated carbon waste (PACW) as a secondary product in some sugar factories attributed to the hydrothermal activation of the main granulated active carbon (GAC) after burning process in the kiln. By such pretreatment, a wet PACW biomass is dehydrated and used again. As requirements for hygiene and safety in sugar decolorization process, the decolorizing material should possess large surface area, good porosity and pH sensitivity for sugar colorants such as flavonoids and phenolic structures (Singh et al. 2012; Liu et al. 2012). Moreover, during processing, especially after polishing filtration, the sugar mud juice (SMJ) was obtained which is a very complicated sludge (high color, bad odor, high viscosity, high brix, etc.), and this is our noble target for decreasing the color as much as possible. There are physical and chemical treatments for dealing with PAC, but the thermal and chemical activation methods are the most predominant techniques for regeneration of exhausted PAC (Gupta et al. 2009; Lu et al. 2011; Guo et al. 2011; Lim and Okada 2005; Simaratanamongkol and Thiravetyan 2010). Thus, the temperature was raised up to 600 °C in the thermal activation process (Martin and Ng 1984). Color removal from SMJ considered new tools and new idea to decrease color of non-sugar compounds such as betaine and phenolic substances. In this respect, a practical and economic way to add value to the waste carbon in sugar factory is deeply required.

In the present paper, we intend to find out an alternative, effective cost-reduction process to get rid of the unacceptable sugar defects, such as color and turbidity, from highly concentrated sugar mud tank (black mud liquor). In addition, the use of PACW is determined and it is compared with reference commercial carbon via adsorption technique for the SMJ process decolorization, and the surface parameters of the used PACW are measured according to IUPAC system.

Materials and methods

Materials

The sugar mud juice (SMJ) dark liquid (Fig. 1) was a generous gift from the United Sugar Company, Saudi Arabia, Jeddah port. Series of samples of the SMJ were collected at the point after polish filters empty slurry. The samples were freshly taken and stored in the refrigerator prior to use. The physical properties of the SMJ determined according to International Commission for Uniform Methods of Sugar Analysis (ICUMSA) (Lopez et al. 2003) are given in Table 1. The Carbo UA (CUA), DCL 320 (reference carbon), was purchased from Sigma. PACW was collected from cyclone outlet ball valve (Fig. 2). Filter aid, commercial strong amorphous silica FW60, was a gift from Spain. All other reagents were used as received.

Fig. 1
figure 1

Photograph of SMJ dark color liquid

Full size image
Table 1 Physicochemical properties of SMJ sample
Full size table
Fig. 2
figure 2

Cyclone outlet ball valve sample collector (a) and the huge PACW (b, c)

Full size image

Methods

Decolorization of SMJ by continuous process

To detect the analytical methods towards SMJ decolor performances, initially batch-technique adsorption experiments assisted. The SMJ solution was filtered through 0.45-µm filter paper to remove any heavy suspended particles; for the turbid solution pH was adjusted to 7.0 using 3-(N-morpholino) propane sulfonic acid (MOPS) buffer. A known mass (0.5 g) of CUA, PACW and FW60 adsorbents were added to three solutions in three separate closed Pyrex glass bottles. The different bottles were kept in a thermo-stated water bath at constant temperature and regular constant shaking until equilibrium was attained. Then the solutions were filtered again and at wavelength 420 nm the absorbance was determined using a UV–visible spectrophotometer. The absorbance readings were converted into color in ICUMSA units (IU) using the equation:

$${\text{IU}} = \frac{{A_{\text{bs}} \times 1000}}{b \times c}$$
(1)

where Abs is the equal absorbance at 420 nm of the test sample, b is the cell length (cm) and c is the sugar concentration (g sugar/ml). The color removal (CR%) and decolorization capacity (DC) are expressed as follows:

$${\text{CR}}\% = \frac{{{\text{IU}}_{\text{i}} - {\text{IU}}_{\text{f}} }}{{{\text{IU}}_{\text{i}} }} \times 100$$
(2)
$${\text{DC}} = \frac{{{\text{IU}}_{\text{i}} - {\text{IU}}_{\text{f}} }}{{{\text{IU}}_{\text{i}} - m}} \times 100$$
(3)

where IUi, IUf and m are the ICUMSA color of the initial, final (treated) solution and weight of dry decolorant (g), respectively.

Measurements of iodine and molasses number

A stock solution from 2.7 g I2 and 4.1 g KI in 1L of solution was prepared, 50 ml from this solution was added to 0.5 g PACW plus CUA in 10 mL of 5% HCl solution. After stirring for 15 min, the sample was filtered and the CR% was recorded for the sample and the blank filtrates based on the volumes of 0.1 M sodium thiosulphate required (using starch as the indicator) (Anon 1994). For the molasses number (MN) test, 10 g of dark sugar beet molasses and 15 g of Na2HPO4 were dissolved in a bottle of 500 mL water and sufficient amount of H3PO4 was added to adjust the pH around 6.5; the above mixture was diluted to 1 L and filtered using 5 gm of light filter aid (silica type). In another bottle, 50 mL of this solution was added to 0.5 g of each decolorant with vigorous stirring until the solution was brought to boil. The CR% for the sample and the blank was determined spectrophotometrically.

Characterization and test methods

FT-IR spectrometer (Nicolet, NEXUS-670) was used to detect the functional groups of the PACW (before and after decolorization), between 4000 and 500 cm−1 using KBr pellets. The SBET and related parameters based on isotherms of adsorption–desorption isotherms was measured at 77 K using (Nova 3200 USA) Micrometrics gas adsorption analyzer. The surface texture was studied by SEM (S-3400 N II, Hitachi, Japan) at two different magnifications; the sample was coated with gold prior to evaluation. Absorbance was measured using a double-beam UV spectrophotometer (Shimadzu UV-1208 model).

Results and discussion

Characterization tools of CUA, PACW and FW60

SEM morphology

Figure 3 illustrates the surface morphology and textures of decolorants at 20 and 100µ magnification. Figure 3a shows that CUA as reference is very ordered, regular and has multi-porous-like layers; on the other hand, in Fig. 3b, PACW shows resemblance to the reference CUA. The surface texture resembles amorphous-like sheets and multi-branch layers due to high white magnesium oxide content (Hassler 1963). In case of FW60 (Fig. 3c), ultimately different structure can be viewed due to silica origin marine source and the surface texture of FW60 has deep pores and rings, and the crystalline pattern has no meaning. All these textural features help in sugar color adsorption due to the possibility of disaccharide molecules to take up these spaces (Hao et al. 2014).

Fig. 3
figure 3

Micrograph of CUA reference carbon (a), PACW (b), and FW60 (c) at magnification 100 and 20 µm

Full size image

BET characterization

According to IUPAC, there are three pore size categories: macropores (> 50 nm), mesopores (2–50 nm) and micropores (< 2 nm) (Cardoso et al. 2012). Figure 4a depicts the isotherms in the form of type II isotherm function (Ewecharoen et al. 2009). In addition, the adsorption increases sharply in the initial low relative pressure indicating the existence of a large amount of small pores on PACW structure. There is a small inflection in the region p/p0 > 0.3, and at a high relative pressure, p/p0 > 0.9, rises rapidly. These types of isotherms were characterized for adsorption onto open surfaces with multilayer formation (Lowell et al. 2004). Figure 4b calculates the pore size distribution using BJH method. The plot shows the PACW has internal micropore radius Dv(r) of 3.452 nm and total pore volume with radius smaller than 224.9 Å, harmonious with the highest specific surface area 613.887 m2/g compared with the BET of other carbon decolorants as shown in Fig. 5.

Fig. 4
figure 4

Nitrogen adsorption isotherm of PACW at 77 K (a), and pore size distribution of PACW by BJH method desorption dV(r) (b)

Full size image
Fig. 5
figure 5

Surface area parameters comparison between PACW and other commercial adsorbents

Full size image

FT-IR illustration

Figure 6 shows the FT-IR of PACW before and after color adsorption behavior. The peak at 2856 cm−1 featured for symmetric or asymmetric stretching –CH in aliphatic series. By perusing literatures (Rao et al. 2010; Lu et al. 2011; Singh et al. 2012; Hao et al. 2014; Cardoso et al. 2012; Ewecharoen et al. 2009; Lowell et al. 2004; da Cunha Gonçalves et al. 2016), all spectra show broad absorption band around 3250–3700 cm−1, which belongs to the (–OH) stretching vibration in carboxyl, phenols, alcohols, and in the absorbed water from PACW. Si–O–Si network vibrations were identified at 1041.3 and 876.4 cm−1. At 1636 cm−1 strong peak related to the stretching vibration of C–O carboxyl or carbonyl groups. The aromatic ring that contains the skeletal units of C–C vibration appeared at 1455 cm−1 (Danish et al. 2014; Shoueir et al. 2016). Apparently, after decolorization, several bands were shifted, which confirmed the presence of oxides in PACW.

Fig. 6
figure 6

FTIR spectra of PACW before and after adsorption of SMJ

Full size image

Clarification of SMJ according to IN and MN tests

IN largely determines the porosity in decolorants of active carbons. For high specific surface area carbons (> 1000 m2 g−1), the IN is a relative indicator of BET surface area measurements, whilst for low specific surface area IN will devolve away. Figure 7 shows the IN and MN comparison with the PACW and CUA (referenced carbon). As mentioned before, the IN correlated to the surface area which explains the differences in IN between adsorbents indicate higher surface area and decreased pore volume to attain micropore configuration (da Cunha Gonçalves et al. 2016). In addition, figure illustrates that referenced CUA has a very high MN, whereas PACW showed relatively lower MN color removal and the presence of micropore radius and total pore volume with radius smaller than 224.9 Å are responsible for this differences. The high specific surface area 613.887 m2/g helping access and facile penetration of any colored organic compounds into the interior surface of PACW (Danish et al. 2014). Comparing the decolorization unit of mud syrup with those of IN and MN with two samples concluded that the presence of small and large pores in the configuration structure led to excellent pore size distribution and adsorption of poly-dispersed colorants. The macropore network causes fast diffusion of color structure than the smaller pores, where they are adsorbed (Shoueir et al. 2016).

Fig. 7
figure 7

Data showing the effect of IN and MN for both PACW and CUA

Full size image

Clarification of SMJ by adding FW60 to PACW

It was prone to use only PACW decolorants but also FW60 was added physically in the second stage to enhance CR%. Table 2 shows the mother mud juice after filtration and still contains a large amount of impurities, such as colloids, suspended solids, reducing sugar and others. In this respect, adding FW60 adds value to PACW (Table 2) unless the color enhanced and there was no detected difference in ash of SMJ. Results of SMJ quality after addition of different doses from FW60 to PACW are presented in Fig. 8. The treated SMJ with higher doses of FW60 had slightly increased CR% until certain dose (0.5gFW60/0.5gPACW/50 mlSMJ) at the optimal clearance CR% was viewed. FW60 was used in sugar manufacture to underlay the filter for more filtration process. Adsorption of colorants and organic molecules like protein on FW60/PACW reduced SMJ color, turbidity and this is due to the greater adsorption surface (Pelekani and Snoeyink 2000).

Table 2 Decolorization performance criteria of the different adsorbents for SMJ solution in terms of CR%, turbidity and ash content (0.50 g sample; 50 mL solution; 25oC)
Full size table
Fig. 8
figure 8

Effect of FW60 different doses added to PACW

Full size image

Determination of CR% under static motion

The static adsorption of color was used to determine the performances of the decolorant materials. Typically, 50 ml from SMJ with pH 7.0 ± 0.1 using MOPS buffer was placed in the beaker flasks with the soaking of 0.5 g of CUA, PACW, FW60, for 2 h with vigorous shaking at 30° C. The initial and final absorbencies of decolorants are calculated according to equations presented Sect. 2.2.1. Table 2 tabulates the results, which indicates that the CR% of the PACW was up to 81.03%. The results give reflection on both quality assurance and control that give PACW the capability to remove both color and turbidity. Moreover, the turbidity is unacceptable at all for white sugar product and an important criterion for deciding the quality of a food product. Also, the turbidity was always tiny found but improved in the final decolorized solution (Jahed et al. 2014).

Effect of time on clarification

The experimental data behavior for the adsorption of turbid mud juice is embraced in this study as a function of time. Figure 9 shows increasing time gives adsorbent the ability to adsorb organic substances; PACW amongst kith is efficient with CR% reaching 90.4% at 30 min equilibrium. During thermal activation, the high specific area increased which gives the hydrophobicity features for PACW. Indeed, the PACW superior food decolorant due to neutral pH, low conductivity, low turbidity and high surface area which is directly proportional to specific surface area (Gupta and Rastogi 2009; Frank et al. 1989; Hao et al. 2014).

Fig. 9
figure 9

Decolorization of SMJ as a function of time

Full size image

Environmental impacts

PACW is considered a waste and reuse it again decouple their importancy. The removal efficiency of PACW due to thermal activation in the kiln that carbonization and activation are carried out simultaneously, so in this case we can consider PACW manufacturing is a cheap and eco-friendly carbon source. Although, many of other commercial materials are not environmentally green. As example, animal origin carbon source is becoming of concern due to release of segments with pathogenic effect (El-Hendawy et al. 2008). These treatments are wearisome, high cost, not safe for the environment and time consuming. Ion exchange is an expensive process due to the high costs and severity of regeneration after treatment and the maximum color dealing with no more than 1500 IU (Guimaraes et al. 1996).

Decolorization mechanism

Most of the postulated decolorization mechanism is based on the nature of surface texture functionality and the surface charge affinity. Thus, our supposal is based on the following items:

(1) If the structure of phenolic substances shows partially negatively charged carboxylate ion, it can interact with positively charged surface of PACW by electrostatic attraction forces:

$$p{\text{ -HOC}}_{ 6} {\text{H}}_{ 4} {\text{ -COO}}^{ - } + {\text{ PACW}}^{ + } \longrightarrow p{\text{ -HOC6H4-COO}}^{ - } \ldots {\text{PACW}}^{ + } .$$

The spotted line represents the electrostatic attraction, (2) the partially negative-charged carboxylate ion can provide H.b sites with the PACW+ surfaces which enhances the adsorption capability of phenolic acids on the surface (Rao et al. 2010; Drago et al. 1964; Mattson et al. 1969).

$${\text{PACW}}^{ + } + {\text{HO-C6H4-COO}}^{ - } \longrightarrow {\text{PACW}}^{ + } \ldots {\text{HO-C6H4-COO}}^{ - }$$

where the dotted bond represents H.b. Under normal liming conditions, polyphenols are not sufficiently ionized, and at neutral pH 7.0, the polyphenols existing in SMJ solution will be also in neutral form, which preferred being adsorbed onto the PACW surfaces. The phenolic adsorption may be due to the interaction between the ring of phenols and the aromatic species of the ash (Salame and Bandosz 2003).

Comparison of PACW with other decolorizing agents

The maximum CR% using raw sugar sample (970 IU) as an adsorbate solution was used as model for comparison between PACW, CUA, dowex 2, Accrual and Indion 830-s. The result of this study is reported in Table 3, and it is clear that PACW has high CR% fitted with standard decolorizing structures. The performance of any sugar decolorant systematically depends on its CR%; however, CR% alone is not sufficient due to the mother load color not included. To determine the performance of the CR%, process technology should be conjugated with color feeding. Table 4 illustrates that load color index in the case of present process is very high (3700 IU) as compared to the standard procedures. The adsorption capacities of the adsorbent are comparatively higher than the other different carbon or ion exchanger-based decolorants. Our results reveal the potential of these dynamics to be an effective waste adsorbent for removing organic color from aqua system.

Table 3 Decolorization of soluble raw sugar solution by different standard adsorbent materials
Full size table
Table 4 Comparison of the present process with the standard procedures
Full size table

Conclusions

Powdered activated carbon waste (PACW) has tremendous activity for the decolorization of sugar mud juice in the sugar refining. The high specific surface area of PACW enhances adsorption and color removal. PACW was found to be comparable to those of the adsorbents actually used for sugar remediation. In addition, the surface morphology of PACW much similar to CUA reference carbon, the exactly CR% was 81.20% which able to remove most of phenolic compounds and the mechanism was explained on the basis that the nature of surface texture functionality and the surface charge affinity.