2.1. Adsorbent Preparation
The commercial GFH adsorbent used was Ferrosorp GW
® (HeGo Biotec GmbH, Giessen, Germany) with sizes in the range of 0.5–2 mm. This kind of adsorbent mainly consists of Fe(OH)
3 and is obtained from an industrial low-cost product [
12].
Two methods were used for the size reduction of the GFH solid.
The first method involved a two-step milling process, in which the first step consisted of dry milling, using a hammer mill (MF 10 basic microfine grinder drive, IKA-Werke GmbH, Staufen Germany) at 6000 rpm. Then, 250 g of the GFH sample was passed through this mill three times and sieved at 100 µm. In the second milling step, 6 g of this pre-milled sample was placed in a 250 mL carbon steel vial, alongside 200 g of S110 spherical high-carbon steel shots (Pometon S.p. A, Italy) and 150 mL of DI water. The material was milled at 250 rpm for 5 h using a Planetary ball mill (Fritsch GmbH, P-5, Mahlen und Messen, Germany). After milling, the slurry was separated from the steel grinding balls using a 75 µm sieve. The retained material was washed several times with DI water over the recovered slurry until reaching a final volume of 500 mL. The slurry was dried overnight in an oven at 100 °C (J.P. Selecta, Digiheat, Abrera, Spain) and the dry solid was disaggregated using an agate mortar. This milled solid is referred to as OF-M in the present work.
The second method used for the size reduction was disaggregation by ultrasonic waves. For this purpose, around 5 g of solid GFH was mixed with DI water in a 250–500 mL volumetric flask and sonicated for 5 min at 20 kHz in a lab ultrasonic cleaner (ATU Ultrasonidos, ATM40-2L-CD, Paterna, Spain). The suspension was centrifuged at 4000 rpm for 5 min to remove supernatant water. The solid was dried and disaggregated like in the milling method (see above). The dry particles obtained from this ultrasonic-based protocol are referred to as OF-U.
2.2. Characterization of Adsorbents
The content of total iron and calcium in the GFH sample was determined by sieving the dried solid below 40 µm and sampling 50 mg for acid digestion using HCl 30% Suprapure® (Supelco, Germany). After filtering the digested solid with a 0.45 µm filter, the total iron and calcium contents were obtained using an AAS instrument (Analytik Jena GmbH, contrAA 800, Jena, Germany).
The general morphology and individual particle size were studied by scanning electron microscopy (SEM) (Zeiss, Gemini ultra plus, Jena, Germany) equipped with energy-dispersive X-ray spectroscopy (EDS) (Oxford Instruments, X-MAX 50 mm2, Abingdon, UK). Sample preparation involved two steps: (I) deposition and room evaporation over aluminium pins, and (II) sample metallization with Au-Pd 30 s at 18 mA (Quorum, Emitech SC7620 mini sputter, Laughton, UK).
The suspensions’ granulometry was analysed using laser diffraction spectrometry for the OF-M samples (Beckman Coulter, LS 13320, Brea, CA, USA) and with a Malvern Panalytical Mastersizer 3000 (Malvern Panalytical Ltd., UK) for the OF-U samples in DI water.
The nitrogen Brunauer–Emmett–Teller (BET) specific surface area (SSA) was measured (Micromeritics ASAP 2020, Aachen, Germany). Degassing was carried out for several hours at a maximum temperature of 100 °C. Sample preparation for this analysis involved drying 1 g of the slurry. It was performed at 60 °C in a biological incubator (Raypa, Incuterm Digit, Terrassa, Spain) for 72 h. Thereafter, a dried cake was obtained, which was crushed to a fine powder using a manual agate mortar. The above-mentioned drying procedure was performed in a negative pressure lab cabinet to avoid exposure to nanopowder.
The zeta potential of 1 g/L suspensions of OF-M and OF-U was measured using 0.01 M analytical grade NaCl (Panreac, Spain) at equilibrium pH and dynamic light scattering detection (Brookhaven Instruments, NanoBrook Zeta Pals; Holtsville, NY, USA).
The pH of the point of zero charge (PZC) of GFH was evaluated via the immersion technique [
33] using 25 g/L suspensions with an initial pH adjusted in the range of 3–12 with analytical grade hydrochloric acid (HCl) or sodium hydroxide (NaOH) (Panreac, Spain), as measured with a pH-meter (Crison, GLP 22, Alella, Spain).
2.3. Batch Adsorption Procedure
Phosphate test solutions were prepared by weighing K2HPO4 analytical grade (Scharlau, Spain). The dissolutions and dilutions were performed using deionized (DI) water (Merck-Millipore, Elix 70, Darmstadt, Germany). The pH was adjusted to 8.0 using HCl. This initial value of pH was chosen as a compromise between the buffer capacity of phosphate and the observed equilibrium pH. In this way, the theoretical predominant speciation of phosphate in all the synthetic samples was HPO42−. Phosphate equilibrium adsorption with GFH was also tested in two types of treated wastewater. The first wastewater (MR-1) was generated during the cleaning operation with phosphate products in a pig slaughterhouse and the treatment of water applied consisted of coarse and fine screening, homogenization, flotation, and nitrification-denitrification biological treatment. The second wastewater (MR-2) was obtained from a truck cover manufacturing textile company in which phosphate was associated with manufacturing process and cleaning operations. Treatment for MR-2 included physicochemical treatment, secondary biological treatment, and tertiary treatment with activated carbon.
Equilibrium and kinetic batch tests were performed by placing 50 mL of the phosphate dissolutions in Falcon tubes with different weighted amounts (Sartorius AG, Practum 513-S, Göttingen, Germany) of each adsorbent at room temperature. Then, stirring at 8 rpm (Heidolph GmbH, REAX 20, Schwabach, Germany) was performed under dark conditions. After stirring the mixtures, the suspensions were centrifuged (J.P. Selecta, Centronic-BLT, Abrera, Spain) at 4000 rpm for 5 min. The supernatant solution was filtered with a regenerated cellulose (RC) 0.2 µm filter and analyzed using ion chromatography (Dionex, ICS 2100, Sunny Valley, OR, USA) with a 4 × 250 mm Ion Pac AS19 Column. The analysis method used 1 mL/minute of the generated KOH mobile phase with a gradient of 10–35 mM. The quantitation limit of the method was 0.5 mg PO43−/L. All masses of phosphate in the present work are expressed as mg PO43−.
The adsorption of phosphate onto the materials was calculated from the following mass balance equation:
where q is the amount of phosphate adsorbed onto the solid at time t (mg/g); C
o and C are the phosphate concentrations at time zero and at time t (mg/L), respectively; V is the volume of dissolution in the batch experiment (L); and m is the mass of the solid adsorbent in the experiment (g).
The removal of phosphate, η (mass %), is given in Equation (2) and can also be expressed in terms of batch experiment conditions using Equation (1):
Equilibrium batch tests were performed using replicate experiments of phosphate dissolutions (10, 20, 33, 50, 90, and 150 mg/L) with different weighed amounts of the three adsorbents (1, 2, and 3 g/L) for 120 h. The wastewater samples were filtered with a 0.45 µm nylon filter before the equilibrium experiment, which was performed with 2 and 3 g/L of GFH for 120 h and analyzed by ion chromatography, as mentioned before. Chloride, nitrate, and sulfate concentrations were also determined to check potential competition of these frequent anions with phosphates.
Freundlich (Equation (3)) and Langmuir (Equation (4)) isotherms were used to fit experimental equilibrium data C
e and q
e (obtained from the corresponding equilibrium using Equation (1)). These models have been originally applied in the literature [
34,
35].
where K
F (mg
1−1/n·(L)
1/n/g) and n are the constants of the Freundlich model that can be obtained by fitting log C
e versusvs log q
e. In the Langmuir isotherm, q
max is the maximum adsorption capacity (mg/g) and b is the binding constant (L/mg). Both parameters can be obtained by rearranging Equation (4):
The theoretical calculation of phosphate removal was performed from the equilibrium concentration of GFH isotherm using Equations (1), (2),) and (4), coded into a Microsoft Excel™ spreadsheet.
The kinetic behaviour of the three adsorbents was determined through duplicate kinetic batch tests with 2 g/L of the materials and 30 mg/L phosphate, whichthat is the minimum expected concentration in the wastewater to be analyzed. Test tubes were extracted after 1, 2, 3, 4, 5, 6, 22, 24, 27, 48, 52, 72, 75, 96, 144, and 168 h of contact time.
Pseudo first- (Equation (6)) and pseudo second-order (Equation (7)) models were employed to identify batch kinetics:
where k
1 is the first-order kinetic constant (h
−1) and k
2 is the second-order kinetic constant (g/mgh).
The corresponding linear forms of pseudo first- and pseudo second-order are as follows, respectively:
Pseudo second-order kinetics model (Equation (7)) could be linearized up to five different linear forms [
36]. Regression using the linear form given in Equation (9) is the most used for phosphate kinetic studies [
7,
8,
9,
10,
15]. This linear form could predict q
e very well, but several references [
36,
37,
38] indicate that this linearization could be inappropriate to obtain a good value of k
2 and the initial adsorption rate value (k
2·qe
2). In order to investigate this question, non-linear models using Equation (7) were applied [
39]. The methodology used for non-linear regression consisted of the direct fitting of n data pairs (t
i, q
i) to Equation (7) using the values of q
e and k
2 that minimize the sum of squared errors, SS
err.
The Solver function in Microsoft Excel™ software with the GRG Nonlinear solving method was used to minimize SSerr as the objective function to obtain the fitted qe and k2. The qe and k2 results obtained in the linear model were used as a starting point for iteration.
2.4. Statistical Analysis
Two statistical methods were used to compare the fitting slopes parameters (m) used in linear models (Equations (5) and (9)) in order to investigate the effects of size reduction on qmax and qe.
The first method was to obtain the 95% confidence interval (CI) for m, CI
m, which is given by the following:
where t
(0.025,n−2) is the two-tailed Student’s
t-test value for a significance level of α = 0.05 and n − 2 degrees of freedom (n is the number of fitting points), and s
e the standard error for the slope m. The extremes of this interval are the lower and the upper confidence levels (LCL and UCL, respectively). The comparison of these CI
m allows detecting the absence of overlap between the adsorbents, and thus confirming the difference between the slope fitting parameters (m).
The second method used was the comparison of the slope parameters of two experiments incorporating a dichotomous (dummy) independent variable, D, [
40,
41,
42,
43]. This D variable allows the combination of two sets of experimental data that correspond to the two categories (e.g., comparison of kinetics from GFH and OF-U adsorbents using Expression (9)), extending the linear model to a multilinear model in the following form:
where Y
i and X
i are the linearized data values of the two sets of data (e.g., t
i/q
i and t
i for GFH and OF-U kinetics). The variable D
i takes the value of zero for the data points of the set of reference (e.g., GFH adsorbent) and the value of 1 for the other set of data points (e.g., OF-U).
Testing the statistical significance of b3 parameter in this model, it is possible to know if the slopes of the two sets are equal. In the case where the CI of b3 contains the zero, the hypothesis that the slopes of the two sets are equal cannot be rejected.
In both methods, the fitting calculations and statistical values were obtained using the LINEST function in Microsoft Excel™ software.