Dry Biosolids Reuse as Costless Biodegradable Adsorbent for Cadmium Removal from Water Systems

The recycling of untreated dry biosolids, as costless biodegradable adsorbent for the removal of cadmium from aqueous phase was characterized. The adsorption of cadmium was reported to depend on initial pH, adsorbent dose, agitation time, and initial Cd concentration. The results of the batch experiments revealed that the maximum adsorption capacity of the untreated dry biosolids was 39.22 mg g-1 under optimum operating conditions (i.e. pH: 5, adsorbent dose: 2 g l-1, contact time: 16h). Adsorption reaches equilibrium after 16h, which can be attributed to both external surface adsorption (R2 = 0.86) and intraparticle dif usion (R2 = 0.98). The Langmuir isotherm model best described cadmium adsorption (R2 = 0.99) and the pseudo-second-order kinetic model was obeyed, suggesting that the mechanism involved was chemisorption. Biodegradability would make the recovery of adsorbed Cd an environmentally friendly process. Comparing the obtained findings with the related published results, it can be concluded that treating biosolids might be an unnecessary and costly procedure for recycling biosolids as an adsorbent for cadmium.


INTRODUCTION
Pollution of water systems with cadmium (Cd) can cause several environmental problems that might negatively impact human health as well as ecological systems (Wang and Chen, 2009). Cadmium can be discharged into the environment through its mining and smelting processes as well as by other anthropogenic means, such as the application of phosphate fertilizers, land application of biosolids, and diverse industrial uses, e.g. nickel-cadmium batteries.
According to Rahimzadeh et al. (2017), longlasting exposure to cadmium through air, water, soil, and food leads to cancer and organ system toxicity. A provisional tolerable weekly intake (PTWI) from food of 7 µg Cd kg -1 body weight was identified by the WHO, which is equivalent to a daily tolerable intake level of 70 and 60 µg Cd for the average 70-kg man and 60-kg woman, respectively (WHO, 1992). The WHO has recommended a limit of 0.003 mg dm -3 as a drinking-water guideline value for cadmium based on an allocation of 10 % of the PTWI to drinking-water (WHO, 1993).
Several materials have been investigated to remove cadmium ions from water systems by adsorption, including Arundo donax leaves, Cyperus laevigatus shoots, paper solid waste, and algal tissues (Hydrodictyon reticulatum) (Ammari, 2014; Ammari et al., 2015; Ammari, 2016; Ammari et al., 2017). Adsorption was considered in these studies for being inexpensive in addition to requiring minimum skills for implementation.
Biosolids are a kind of colloidal sediment waste produced from the treatment of municipal wastewater. According to the German Corporation for International Cooperation (GIZ, 2020), there are about 31 domestic wastewater treatment plants in Jordan. Only one plant uses biosolids to generate energy. All the remaining treatment facilities do not utilize biosolids properly. Therefore, biosolids have become an ecological, environmental, and financial burden. In other parts of the world, biosolids are reused in landfilling, in forestry, in sea dumping, as soil improver, etc. In the current study, (1) the adsorption performance of untreated dry biosolids (BS) was investigated in terms of cadmium removal from aqueous phase, (2) diferent key parameters including initial pH, initial concentration, dose, and contact time were investigated to understand the adsorption of cadmium ions onto untreated dry biosolids collected from the Wastewater Treatment Plant of Wadi Shueib, and (3) adsorption kinetics and isotherm modeling were also discussed.

Preparation of the untreated dry biosolids
Dry biosolids, brought from the Wastewater Treatment Plant of Wadi Shueib, were repetitively washed using tap water followed by distilled water. The purpose was to decrease the electrical conductivity of these samples to values around 100 µS cm −1 , which were, further, dried in an oven at 105°C. The oven-dried biosolids were ground with a ceramic mortar and pestle and sieved to obtain the particles between 250 and 355 µm in size.

The Fourier transform infrared (FTIR) analyses
The FTIR analyses using thermo Nicolet nexus 670 were carried out by using the KBr pellet technique, before and after adsorption, on the BS materials using the Fourier Transform Infrared Spectroscopy to define the functional groups involved in adsorbing Cd 2+ . The spectra of FTIR of BS were determined in a 400-4000 cm -1 spectral domain and a resolution of 4 cm -1 .

Energy dispersive X-ray spectrum (EDX)
The EDX spectrum using Scanning Electron Microscope (SEM) (FEI quanta 600, Thermo Fisher) was conducted to reveal the elemental composition of the BS material. In this technique, the samples were excited by a highly focused, high-energy primary electron beam and X-rays are emitted from the samples.

Thermogravimetric analysis
The thermogravimetric analysis (TGA) and derivative thermogravimetry analysis (DTA) before and after adsorption were conducted to investigate the thermal characteristics of the BS material using a thermal analyzer (Netzsch STA 409 PG/PC) with a heating rate of 10°C /min, under nitrogen flow at a rate of 30 ml/min.

Aqueous phase and analytical procedure
Cadmium standard and aqueous solutions with concentrations ranging from 20 to 100 mg l -1 were prepared using a certified Cd reference solution (Cd metal in 2% HNO 3 , USA). All solutions were prepared in 0.01 mol l -1 NaNO 3 (UK). Diluted NaOH (Germany) or HNO 3 (China) was used to modify the pH of these solutions. All chemicals were of analytical reagent grade. The flame atomic absorption spectrophotometer (A Varian model AA-6650 flame AAS, Shimadzu, Japan) was used to measure the concentrations of Cd in solutions at 228.8 nm wavelength.

Effect of solution pH
The solutions containing 20 mg l −1 of Cd and 2 g l −1 BS were used to investigate the efect of pH at the ambient temperature of the laboratory (approx. 25±1°C). The initial pH of solutions was 2, 3, 4, 5, 6, 7, and 8. The BS-Cd solutions were agitated on an orbit shaker for a period of 24 h. This experiment was conducted in PE-bottles, previously washed with acid (Alemayehu and Lennartz, 2009) onto which no sorption of Cd was detected.

Effect of BS Dose
Diferent doses of BS (2, 3, and 4 g l -1 ) were added to the pH 5.0 modified solutions of 100 mg l −1 Cd. The BS-Cd solutions were agitated for 16 h.

Effect of Contact Time
The kinetics of the adsorption process were investigated by agitating the pH 5.0 modified solutions that contain 20 mg Cd l -1 and 2.0 g l -1 BS for 1, 5, 30, 60, 180, 360, 960, 1200, and 1440 min.

Effect of Initial Concentrations of Cd
The optimal adsorption conditions (in terms of pH and BS dose) were considered to study the efect of diferent initial concentrations of Cd. Two g l -1 BS were added to the pH 5.0 modified solutions of 20, 40, 60, 80, and 100 mg l −1 Cd.
The BS-Cd solutions were agitated for 16 h, filtered via 0.45 µm filters, and analyzed for the equilibrium concentration of Cd by flame AAS. The efficiency and capacity of adsorption were expressed using the following equations (Eqs. 1 and 2): where: Ci is the Cd initial concentration (mg l -1 ), Ce is the concentration of Cd at equilibrium (mg l -1 ), qe is the capacity of adsorption at equilibrium in mg g -1 , V is the volume of Cd solutions in l, and m is the BS weight in grams.

Batch adsorption experimental data
The Langmuir and Freundlich isotherm models; commonly available in the literature, were used to test the equilibrium data of Cd adsorption by BS. Isotherm variable parameters and determination coefficients (R 2 ) were estimated from the linearization of the above-mentioned models.

Langmuir Isotherm Model
Cadmium adsorption can be described by the Langmuir model particularly as the Cd concentration at equilibrium (C e ) becomes high (Essington, 2004). The Langmuir model is defined as (Eq. 3): where: qe is the adsorbed Cd at equilibrium [mg g -1 ], KL (l mg -1 of adsorbent) indicates the adsorption isotherm intensity (Essington, 2004) and qm is the monolayer adsorption capacity (mg g -1 ).
The Langmuir model is linearized as follows (Eq. 4): A straight line is attained by plotting C e /q e versus C e and K L and q m were computed. Slope and intercept of the straight line were used to estimate the values of q m and K L , respectively. In addition, the affinity between the adsorbate and adsorbent can be predicted using the value of R L , i.e. dimensionless separation factor (Hall et al., 1966). R L is expressed as (Eq. 5): where: KL is previously defined and C0 is the initial concentration of Cd (mg l −1 ). When RL is 0 < RL < 1, adsorption is favorable (Weber and Chakravarty, 1974).

Freundlich Isotherm Model
The adsorption data over the whole range of C e can be described by the Freundlich isotherm model [20]. This model is given as (Eq. 6): where: KF and 1/n are the Freundlich constants that describe the capacity of sorption or the number of surface sites onto which adsorption might occur (Mustafa et al., 2004) and intensity of sorption, respectively.
The linearization of this model was carried out as follows (Eq. 7): The variables of the Freundlich model were estimated by plotting log q e against log C e . The slope and intercept of the straight line were used to calculate the values of 1/n and log K F , respectively.

Kinetics of adsorption
Batch experiments were conducted to determine the equilibrium time using 100 mg l -1 Cd initial concentration, pH 5.0, and 2 g l -1 BS agitated for 1, 5, 30, 60, 180, 360, 960, 1200, and 1440 min. According to Alyuz and Veli (2009), adsorption mechanism as either mass transfer or chemical reaction, can be determined by kinetic studies. In addition, estimating the rate of adsorption is useful for the design of batch adsorption systems. Type-1 pseudo-second-order kinetic model was applied to investigate the adsorption kinetics, which is described as (Eq. 8): where: k2 (g mg -1 min -1 ) is the second-order adsorption rate constant calculated from the linear plot of t/q t against t.
Equations (9) and (10) were used to calculate the rate constant and the initial adsorption rate (mg g −1 min −1 ), respectively, and were as follows: Initial adsorption rate = k2·qe 2 (10) The Intraparticle difusion kinetic model was also applied to fit the present kinetic data, which is described as (Eq. 11): where: Kdi is the rate constant (mg/g min 1/2 ) (Weber and Morris, 1963).

Energy dispersive X-ray spectrum
The EDX analysis showed that before Cd adsorption, BS consists mainly of carbon (percent by mass = 94.65 wt.%) and traces of Si, S, Mg, Ca, and P (Fig. 1). The appearance of the main characteristic peaks of Cd in the EDX spectrum of the BS sample after Cd adsorption shown in Figure 1 reveals that Cd ions have been successfully adsorbed onto BS.

Thermogravimetric analysis
The TGA and DTG for BS before and after Cd adsorption are shown in Figure 2 (a) and (b); respectively. The thermogravimetric profiles showed that Cd 2+ adsorption caused delay in thermal degradation. In general, the mass losses occur in four phases. The first phase was observed approximately between 50°C and 175°C, in which the release of superficial absorbed water, coordinated water and possible volatile compounds  ) and after Cd adsorption were 67.7% and 60.9% before and after adsorption, respectively. The DTA curve of BS before adsorption showed a peak at around 125°C that is associated with water loss (Figure 2); whereas the position of this peak shifted to 105°C and its intensity decreased after Cd adsorption because Cd ions repelled the water molecules occupied by the BS.

Effect of pH
The ionization of functional groups increases along with the solution pH, consequently, the surface negative charges of certain adsorbents increase as well (Chen and Ting, 1995). Moreover, the solution chemistry of heavy metals can be influenced by the solution pH via hydrolysis (Wang, 2002).
Adsorption of cadmium onto BS, as afected by solution pH was examined by modifying the initial solution pH to 2, 3, 4, 5, 6, 7, and 8. Such pH values were selected because they might be found in municipal as well as industrial wastewater (Bond and Straub, 1974). The solutions were prepared using 0.01 M NaNO 3 was to simulate the actual conditions of most water systems because the ionic strength of background solutions reflects the interaction degree among ions in solutions (Sparks, 1995).
The pH-adsorption edges of Cd (II) obtained for the BS are shown in Figure 3. The results showed that the Cd removal efficiency ranged from 13.7% to 95.3% at pH 2 and 5, respectively. According to Hawari and Mulligan (2006) the heavy metal removal can be attributed to diferent causes such as (1) electrostatic attraction, (2) formation of surface complexes, and (3) precipitation. The obtained results showed that, at pH 2, the efficiency of Cd removal was lower than that at higher pH. At pH 2, the H + concentration increased and occupied almost all adsorption sites on the surface of the adsorbent, which limited the ability of Cd ions to move towards the surface due to charge repulsion. Protons also decrease the negative charges by protonation, i.e., association of the functional group with protons (Khalfa and Bagane, 2011). The removal percentage increases along with the pH (> 2), due to the decrease of H + on the surface, which results in lesser repulsion with Cd ions (Sheha and El-Zahhar, 2008). Moreover, the surface becomes progressively negatively charged, as the hydroxyl ion activity (and pH) increase, thus creating favorable conditions for specific adsorption of Cd (Zhou and Haynes, 2010).
Furthermore, the divalent Cd and Cd(OH) + are the dominant Cd species at pH less than 8 (Babic et al., 2002, Cordero et al., 2004. For the chemical composition of solutions used in our study, the free divalent Cd ranged from 97.3% of the total concentration of Cd at pH 8 to ≈98% at pH 2 as estimated by Visual MINTEQ (Data not shown). The obtained results agreed with Baes and Mesmer (1986) and Srivastava et al. (2006) who reported that the logK of Cd hydrolysis reaction is equal to -10.08 or -7.9, respectively. Therefore, when pH is less than pKa, free divalent Cd ions are dominant in solutions. Consequently, Cd speciation, as afected by pH, was not responsible for the observed "adsorption envelope". It can be concluded that the magnitude of adsorption in the current study can be attributed to the functional groups and their associated ionic state. Ait Ahsainea et al. (2017) suggested that, among others, the -OH stretch ing vibration and the stretching Fig. 2. a) TG and b) DTG thermograms for the BS before and after Cd adsorption of the N-H bond of organic compounds can be responsible for the adsorption of Cd 2+ by mesoporous treated sewage sludge. The FTIR spectra analysis of Li et al. (2018) showed that Cd (II) adsorption by live and dead biosorbents took place via the hydroxyl, amino, amide, carbonyl, and carboxyl groups. These candidate functional groups were determined by the FTIR spectra of the BS before and after adsorption (Figure 4). The peaks at 3,390, 2922 and 2852, 1,652, and 1,542 cm −1 are associated with several organic functional groups. The peak at 3,390 cm -1 is at- ; whereas, the peaks at 1,652 and 1,542 cm -1 indicated the existence of amide I (-CO-stretching vibration) and amide II (-NH bending and -CN stretching) in proteins, respectively. After adsorption, while some peaks disappeared, shifts were observed for the peaks from 3,390 and 1,652 cm −1 to 3,448 and 1,639 cm -1 , respectively, indicating that the above-mentioned functional groups might be involved in the adsorption of Cd ions. However, the increase of the band intensity observed in the region of the -OH stretching vibration when compared with the BS before adsorption could be due to the adsorbed water after adsorption. Additionally, the other bands in the region 200-2,600 cm -1 were relatively found to decrease in intensity after adsorption. According to Ait Ahsainea et al. (2017), this might be due to the electrostatic interactions between the related functional groups (-NH and -CO) and Cd ions, which is in agreement with (Mnahan, 1999a; Mnahan, 1999b; Gutierrez-Segura et al., 2012). Shift and disappearance of peaks suggest a binding between Cd ions and the BS surface. It can also be concluded that hydroxyls and amides are the most important characteristic functional groups in the BS participating in Cd 2+ adsorption.

Effect of BS Dose
The efect of BS dose on Cd adsorption was investigated and presented in Figure 5. The results showed that the removal efficiency of Cd increased along with the BS dose from 2 to 4 g l -1 (Figure 5). At the highest dose, the detected increase in Cd removal (%) is due to the increased surface area of the adsorbent and adsorption locations. Nevertheless, the reduction in the capacity of adsorption with the increase of BS dose (from 35.4 to 21.9 mg g -1 ) can be attributed to the unsaturation of binding locations via the process of adsorption (Ghodbane et al., 2008). Another reason can be due to the aggregation of particles resulting from high adsorbent concentration. Such aggregation would result in reduced the total

Effect of contact time
The binding of Cd ions on the BS as a function of contact time was investigated under optimum experimental conditions ( Figure 6). Cadmium adsorption increased along with contact time and its removal efficiency gradually reached a constant value of 81% (40.4 mg g -1 ) after 16h ( Figure 6). The obtained results agree well with earlier published data concerning the relatively long equilibrium times. Luo et al. (2006) reported that the adsorption of Cu onto activated sewage sludge was up to 90% after 0.5 hour, whereas the adsorption of Cd was up to 96% after a time of 7 hours. Rashed (2006) found that the adsorption of Pb by the stones of peach and apricot increased with contact time and reached the maximum value at 3 and 4 hours, respectively. Fonseca et al. (2006) showed that the equilibrium times for manganese and cadmium onto vermiculite were 48 and 72 hours, respectively.

Kinetics of adsorption
The adsorption kinetics provide important information about the sorption mechanism (Bulgariu and Bulgariu, 2014). The kinetics of the adsorption of Cd by BS was explored by plotting t/ q t versus t (Figure 7). One of the commonly used kinetic models is the pseudo-second-order model. This model implies that the chemical interactions among ions of Cd and surface functional groups of the BS is involved, which is the rate controlling step. The agreement between the experimental and predicted values of the model was reflected by the R 2 (determination coefficient very close to 1) (Table 1), which indicates that such kinetic model efectively describes the Cd adsorption kinetics onto BS (Figure 7 and Table 1). This model indicates that chemisorption, which comprises sharing or exchange of electrons between the BS and the adsorbate ions, is the rate limiting step (Ho and McKay, 1999). From Figure 7, the estimated q e , was calculated for the initial concentration of Cd of 100 mg l −1 as shown in Table 1 Figure 7 indicates a multilinear characteristic and the intraparticular difusion rate (K di ) and C were calculated (Table 1). It can be seen that there are two steps that control the adsorption process: (1) the first step shows mass transfer of the Cd ions from the liquid to the solid external surface known as boundary layer difusion and (2) the second step is the intraparticle difusion; the movement of Cd ions within the pores begins and the equilib rium is established. Similar results were reported by Ait Ahsainea et al. (2017) who found that the adsorption of Cd ions onto the mesoporous treated sewage sludge is a combina tion of these two phases as rate-controlling steps. Such results might explain the relatively long equilibrium time obtained in the current study (16h). It might be reasonable to conclude that the intraparticle difusion of Cd ions within the pores was evidently slow, which might delay attaining the equilibrium state.

Adsorption Isotherm Study
Specific adjustable parameters can characterize the adsorption isotherms. Such parameters describe the properties of the adsorbent surface and the affinity of the adsorbent to the adsorbed ions. As the concentration of Cd increased from 20 to 100 mg l -1 , a decrease in the Cd removal percentage by BS from 96.2% to 71.2% was observed, which is due to the saturation of adsorption sites on the BS when the Cd concentration exceeds a certain value, as explained by Semerjian (2010).
When equilibrium is attained, q e of BS increased along with concentration of Cd in the solution (from 20 to 100 mg l −1 ). These findings agree with the previously reported results of various studies (Elouear et al., 2009;Ammari, 2014;Ammari et al., 2015;Ammari, 2016;Ammari et al., 2017). With increasing initial concentrations of Cd, the observed increase in Cd ions adsorbed onto the BS can be due to the greater concentration gradient resulting from more Cd(II) available in the solution at greater concentrations of Cd. In the present study, the relation between the capacity of adsorption and the concentration of Cd at equilibrium were described and characterized by the Langmuir and Freundlich isotherm models. These isotherms are simple and able to characterize the adsorption process at the solidsolution interface. The isotherms data can be employed in designing the adsorption systems (Saygiderger et al., 2005). The constants and R 2 of the Langmuir and Freundlich isotherms are reported in Table 1. The results revealed that Table 1. Efect of initial Cd concentration (20-100 mg l -1 ) and 2 g l -1 adsorbent dose on equilibrium adsorption capacity at pH 5.0 and 0.01 M NaNO 3 , the kinetic parameters for pseudo-second-order kinetic model and isotherm adjustable parameters of Cd adsorption on BS   (Table 1). Moreover, the critical characteristic of the Langmuir model, R L , a dimensionless constant referred to as equilibrium parameter is reported in Table 1. The calculated R L values range between 0 and 1, which refer to a favorable adsorption process. A comparison of the values found in our study (39.22 mg g -1 ) with commercial activated carbon (8.21 mg g -1 ) (Azouaou et al., 2010); sewage sludge ash (7.1 mg g -1 ) (Elouear et al., 2009) ; iron oxide coated sewage sludge (14.7 mg g -1 ) (Phuengpraso et al., 2011); modified physically activated sewage sludge (31.44 mg g -1 ) (Nageeb et al., 2018); mesoporous treated sewage sludge (56.2 mg g -1 ) (Ait Ahsainea et al., 2017) revealed that the BS presented a reasonable Cd adsorption capacity from liquid phase without any kind of treatment or modification. The disposal of loaded BS with Cd ions could impose an environmental issue (Gaballah et al., 1998). Commonly, the regeneration of exhausted adsorbent materials is carried out through the desorption process, which was reported for inorganic sorbents (e.g. montmorillonite and silica gel as reported by Katsumata et al. (2003) as well as biosorbents (Abu Al-Rub et al., 2004; Bulgariu and Bulgariu, 2014). The regeneration of loaded biosolids for reuse might not be feasible or necessary because such material is increasingly generated, whereas the recovery of Cd from loaded biosolids is of environmental and industrial importance.

CONCLUSIONS
We found that dry biosolids can be recycled as costless biodegradable adsorbent for cadmium removal from polluted water systems. The removal of Cd considerably depends on the initial Cd concentration, agitation time, dose, and solution pH. The adsorption of Cd ions onto BS was characterized by the Langmuir isotherm and the pseudo-second-order model. The maximum capacity of adsorption was 39.22 mg g -1 under optimal experimental conditions. Equilibrium was attained after 16 h.
Comparing our findings with the related published results, treating biosolids both chemically or physically will increase the cost of biosolids recycling and might not be necessary. We showed that the adsorption capacity of untreated BS is greater than or, at least, comparable to that of commercial activated carbon, sewage sludge ash, iron oxide coated sewage sludge, modified physically activated sewage sludge, and mesoporous treated sewage sludge. Such findings will help reducing the sewage sludge recycling costs. In addition, biodegradability would make the recovery of adsorbed Cd an environmentally friendly process.