Environmentally Safe Method for Conditioning and Dewatering Sewage Sludge Using Iron Coagulant, Cellulose and Perlite

Abstract

A reasonable strategy for the development of sludge conditioning methods prior to dewatering appears to be the use of substances that allow the safe management of dewatered sludge. It is also justified to use mineral or organic conditioners instead of synthetic chemicals, e.g., polyelectrolytes, or to try to use other substances, e.g., waste. The properties of iron coagulant (PIX 113) combined with perlite and cellulose can be an environmentally safe method of sludge conditioning. The tests were carried out in accordance with European standards on the efficiency of mechanical dewatering of sewage sludge. The most advantageous method of sludge conditioning was the dosing of the iron coagulant PIX 113. The use of at least a coagulant dose of 0.40 g/g DS enabled the achievement of minimum sludge dewatering parameters, i.e., specific resistance of filtration (SRF) < 5.0 E12 m/kg and final hydration of filtration cake (FH) < 80%. The use of cellulose and perlite as stand-alone conditioners or in combination with PIX 113 resulted in a deterioration of the sludge dewaterability and the quality of the filtrate. It is assumed that the further development of environmentally friendly conditioning methods requires the use of easy-to-use, non-toxic and biodegradable substances. It is important to select conditioners which, in practically acceptable doses, can improve the conditioning effect or show a synergistic effect in combination with previously used conditioners.

1. Introduction

Sewage sludge is a complex mixture of solid and dissolved substances that is produced during the wastewater treatment process. The complex composition of sewage sludge includes valuable organic substances, nutrients and micro and macro elements [1,2], but also emergency contaminants such as heavy metals, polycyclic aromatic hydrocarbons, microplastics or pharmaceuticals [3,4]. Sewage sludge must be managed, but the method of management depends on many factors, such as the chemical composition of the sludge, available technologies, legal requirements, and environmental, energy, and economic aspects [5,6,7].

Generally, there are two ways of sludge management-disposal: land-based applications [3,8] or thermal treatment [9,10]. However, regardless of the method of sewage sludge management, the required condition is sludge dewatering. Sludge dewatering allows for reducing their volume and facilitates storage, transport, and disposal [11]. The dewatering process in sewage sludge technology is very demanding. This is due to the amount and binding strength of water accumulated on the surface of solid particles and the structure of sludge agglomerates [12,13].

The susceptibility of raw sludge to dewatering is usually limited by the hydrophilic organic matter content [14], colloidal particles [15], and extracellular polymeric substances (EPS) [16,17]. Conditioning is a necessity in sludge dewatering technology. Generally, the purpose of conditioning is to change the properties of the sludge using chemical, physical and biological methods in such a way that it enables effective dewatering [11,18].

For practical reasons, the key issue in the sludge dewatering process is the effectiveness and reliability of conditioning methods. Therefore, the dominant methods in sludge conditioning are the use of mineral coagulants such as aluminium and iron salts and, above all, the use of polyelectrolytes [19]. The main disadvantage of the use of polyelectrolytes is the chemicalization of sludge, which causes a potential negative impact on microorganisms and potential environmental and health problems [20,21,22]. Due to the disadvantages of using polyelectrolytes, but also to increase the efficiency of dewatering, research is carried out using many unconventional methods. Unconventional methods often combine the dosing of coagulants and polyelectrolytes with other chemical or waste reagents. Examples include the use of tannic acid [23] and coal fly ash modified by sulfuric acid [24].

A reasonable strategy for the development of conditioning methods seems to be the use of natural coagulants and structure-forming agents as an independent or combined conditioning method. The validity of this strategy is based on the advantages of natural reagents, such as being readily available, economical, easy to use, biodegradable, non-toxic, and eco-friendly [25].

A frequently used natural reagent is lignocellulosic biomass. The use of lignocellulosic materials is particularly justified due to the calorific value and energetic processing of sludge, such as incineration, gasification, pyrolysis, as feedstock or fuel [26] and carbonisation [27]. Cellulose is also treated as a structure-forming material that reduces sludge compressibility. In appropriate doses, cellulose can enhance dewatering and generate a value-added product for the land application of sludge [28,29]. Among the conditioners qualified as a filtration aid, perlite also stands out as a skeleton-forming material. The predominant use of perlite is as a preliminary layer applied to the surface of the filtration barrier [30,31]. The use of perlite is not common in sewage sludge technology. Most applications concern the use of perlite in filtration in industrial applications, especially in the food industry [32,33].

Due to the dewatering of municipal sewage sludge, the main disadvantages of cellulose and perlite are hydrophilicity [28] and water retention [30], respectively. However, as other researchers point out, despite the disadvantages, the use of cellulose and perlite in appropriate doses counteracts the compressibility of sludge and can improve dewatering efficiency [28,29,32,33]. Mineral coagulants such as iron and aluminium salts are assumed to be useful in removing water from the structure of sludge conditioned with cellulose or perlite. Iron coagulants cause destabilisation and agglomeration of solid phase particles [34]. The addition of a mineral coagulant can be one way of counteracting hydrophilicity and water retention. Unlike the use of polyelectrolytes, the only disadvantage of using iron salts is their toxicity to living organisms, which only occurs at high doses [35,36]. The use of a low dose of iron coagulant in combination with perlite and cellulose appears to be an environmentally safe method of sludge conditioning. Safe because of the presence of substances in the dewatered sludge (iron salts, cellulose, perlite), which in particular help to improve the fertilising properties of the dewatered sludge. The main aim of the research is to determine the suitability of the combination of natural reagents (cellulose and perlite) and the commercial iron coagulant PIX 113 (currently used in sewage coagulation) in the effective dewatering of sewage sludge. This simple combination of sludge conditioners has not been tested so far. Additionally, these conditioners are commonly used, and available substances and their chemical characteristics (especially cellulose and perlite) do not have a negative impact on the environment.

2. Materials and Methods

2.1. Research Substrate

Sludge samples were taken from a municipal sewage treatment plant with a capacity of 200 m³∙d⁻¹. This facility is a mechanical-biological treatment plant with activated sludge with dephosphatation, denitrification and nitrification zones. The treatment plant is made of compact, combined steel tanks. All elements of the treatment plant are located in a closed, ventilated hall. The samples collected for testing are sludge after a 25-day aerobic stabilisation process at ambient temperature. The basic physical and chemical parameters of the collected sludge are presented in

Table 1. Physicochemical properties of tested stabilised sludge.

Parameter	Unit	Value
Total solids	g/L	22.6 ± 0.6
Volatile solids	%	17.3 ± 0.7
pH	-	6.40 ± 0.21
Capillary suction time (CST)	s	634.0 ± 32.0
Specific resistance of filtration (SRT)	×1014 m/kg	2.34 ± 1.2
Filtration velocity (FV)	L of filtrate/h	0.8 ± 0.13
Final hydration of sludge cake (FH)	%	94.8 ± 1.6
Filtrate pH	-	6.40 ± 0.21
Filtrate Conductance	μS/cm	149.4 ± 47
Filtrate Turbidity	NTU	380.7 ± 43.9
Bingham (rheological model)	-	$\tau =1.1448+0.0086·\gamma$ $R^2=0.93$
Hershel Bulkley (rheological model)	-	$\tau =0.313+0.0729{\left(\gamma \right)}^{0.6771}$ $R^2=0.99$
Ostwald (rheological model)	-	$\tau =1.0·{\gamma }^{0.2499}$ $R^2=0.97$
Thixotropy (T)	Pa s−1	382.3 ± 74.9

.

2.2. Reagents

PIX 113 iron coagulant, cellulose and perlite were used to condition the sludge. Commercial product PIX 113 is an aqueous solution of Fe₂(SO₄)₃. The iron content is 11.8% ± 0.4%, the ion iron content is 0.4% ± 0.3%, and the coagulant pH ≤ 1.0. The density of the coagulant solution was 1.5 g/mL. The following doses of PIX 113 were used in the tests: 2.0, 4.0, 6.0, 10.0, 20.0 and 30.0 g/L of tested aerobic sludge. Calculated on the dry solids of sludge, the doses were 0.13, 0.27, 0.40, 0.66, 1.33 and 1.99 g/g DS, respectively.

Cellulose was used as a natural reagent for conditioning the sludge. Fibrous, unrefined pine wood cellulose obtained from a local wood processing factory was used. The second natural conditioner was expanded perlite. A commercial product called EP-100 was used. The applied doses of cellulose and perlite were 0.5, 1.0, 2.0, 4.0, 8.0, 16.0, and 32.0 g/L of sludge. The doses calculated as the weight ratio of the reagent mass to the dry solids of sludge were 0.02, 0.04, 0.09, 0.17, 0.35, 0.71, and 1.42 g/g DS.

2.3. Mixing Sludge with Reagents

The volume of the sludge sample mixed with the reagents was V = 1.0 L, and the mixing time was 40.0 s. Turbulence during the mixing of sludge with reagents was determined on the basis of the calculated Reynolds number, mixing power and the velocity gradient [37]. A 4-blade mixer was used in the mixing process. The length and height of the blade were 0.1 m and 0.015 m, respectively. The rotational speed of the mixer was $n=400 \mathrm{rpm}$. It was assumed that the dynamic viscosity coefficient of sludge, including sludge mixed with reagents, is equal to that of water at a temperature of 20 °C and is $\mathsf{\mu }=0.001 \mathrm{Pa}·\mathrm{s}=0.001 \mathrm{kg}/\mathrm{m}·\mathrm{s}$.

Reynolds number calculations:

$$Re=\frac{\mathrm{u}\times d^2\times \mathsf{\rho }}{\mathsf{\mu }}=\frac{6.7\times {0.1}^2\times 1000}{0.001}=67,000$$

(1)

Mixing power:

$$P={\mathrm{k}}_2\times d^5\times n^3·{\rho }_O=0.5\times {0.1}^5\times {6.7}^3\times 1000=1.5\mathrm{W}$$

(2)

Velocity gradient:

$$G=\sqrt{\frac{P}{V\times {\mu }_O}}=\sqrt{\frac{1.5}{0.5\times 0.001}}=38.73{\mathrm{s}}^{-1}$$

(3)

2.4. Analytical Methods

The basic technological parameters of the original sludge samples were tested according to PN-EN 12880—dry solids (DS) and water content (FH)—[38], PN-EN 12879—volatile solids (VS)—[39], PN-EN 14701-1—capillary suction time (CST)—[40], PN-EN 14701-2—specific resistance of filtration (SRT) [41].

Conductance (C) was determined by direct conductivity using the Hydromet CD-210 probe. The turbidity (T) of the filtrate after separation of the solid phase in the vacuum filtration process was determined using a Hach 2100N turbidity meter (Hach Company, Loveland, CO, USA). The rheological tests were carried out using a Brookfield RST rheometer with a CCT-40 measuring system. The tested samples were thermostated at a temperature of 20 °C. A Malvern Zetasize Lab instrument was used to determine the zeta potential of particles in the filtrate. Zeta potential was determined in samples thermostated at 20 °C.

2.5. Rheological Measurements

Rheological models are a group of equations that take into account rheological parameters such as dynamic viscosity, shear rate and shear time. These models are necessary to analytically solve problems related to the flow of non-Newtonian fluids. In order to determine the relationship between shear rate (s⁻¹) and shear stress (Pa s⁻¹), the main rheological models were used (

Table 2. Rheological models used in the research [42].

Model	Equation
Bingham	$\tau ={\tau }_0+\eta ·\gamma$
Herschel and Bulkley, (0 < n < N)	τ = τ0 + k(γ)n
Power-law (Ostwald de Vaele) Shear-thinning (pseudoplastic) n < 1. Shear-thickening (dilatant) n > 1	$\tau =k·{\gamma }^n$

$\tau$—shear stress, Pa; ${\tau }_0$ is the yield stress, Pa; $\gamma$ = shear rate, s⁻¹; n—the flow behavior index; k—the flow consistency index, Pasⁿ.

).

All experiments were conducted following two steps. The first step was to linearly increase the shear rate from 0 s⁻¹ to 300 s⁻¹ over 300 s. The second step was to linearly decrease the shear rate from 300 s⁻¹ to 0 s⁻¹ over 300 s. Thixotropy was evaluated as the area included between the two flow curves and increasing and decreasing shear rates.

2.6. Research Combinations

The tests were divided into five combinations depending on the reagents used. Single sludge conditioning, i.e., with only one reagent, and dual conditioning were carried out. Dual conditioning was the sequential dosing of two reagents, one after the other (Figure 1).

2.7. Statistical Analysis

Pearson’s R correlation coefficient was used for statistical analysis of single conditioning. The correlation coefficient was used to determine the degree of proportional (linearly dependent) relationships between the variables (CST, FV, SRT, FH, pH, C, T). Correlation coefficients in bold were significant at the p < 0.05 level.

In dual conditioning, analysis of variance (ANOVA) was used to test the significance of differences in the effects of preparation. Post hoc analysis (Tukey’s test) was used for statistically significant data p < 0.05. Values marked with the same letter (a, b, c) are not statistically significant. This means that the sludge treatment methods did not show statistically significant differences in conditioning efficiency.

STATISTICA 13.3, TIBCO Software Inc. (Palo Alto, CA, USA) was used for statistical analysis.

3. Research Results

3.1. Research Results—Combination 1

According to the coagulation theory, the iron reagent PIX 113 caused destabilisation and flocculation of solid particles, which enabled dehydration of the solid particle surface [19,34]. An increase in the reagent dose resulted in a decrease in the capillary suction time and, therefore, accelerated the separation of the sludge liquid (Figure 1). The most rational dose of PIX 113 was found to be 6.0 mL/L of sludge (0.27 g/g DS). The use of higher doses of the reagent than 6.0 mL/L did not accelerate the separation of the sludge liquid, and CST values oscillated around 50 s. The CSK values were inversely proportional to the obtained filtration velocity of the prepared sludge. These two parameters converged to indicate that the optimal coagulant dose was 6.0 mL/L (Figure 2).

The dosage of PIX 113 at 6.0–30.0 mL/L (0.40–1.99 g/g DS) caused a decrease in the specific resistance of filtration below 5.0 × 10¹² m/kg (Figure 3). This value is considered to be the upper limit of sludge filterability [41]. Depending on the type and characteristics of the sludge and the type and dose of the coagulant, it is possible to achieve substantially different SRF values. In Xu et al. studies on the dewaterability of activated sludge, SRF below 5.0 × 10¹² m/kg was obtained using PAC (polyaluminium chloride) at a dose of 5 mg/g DS or FeCl₃ at a dose of 3 mg/g DS [43]. Zhang et al. achieved a reduction in SRF of landfill sludge from 45.21 × 10¹² m/kg to 2.78 × 10¹² m/kg using FeCl₃ at a dose of 0.2 g/g DS [44]. It should be noted that the use of doses above 6.0 mL/L (0.40 g/g DS) had no significant effect on changes in filtration resistance, final hydration of sludge cake and filtration velocity. Therefore, it was found that increasing the coagulant dose above 6.0 mL/L is not justified; there is a slight improvement in dewatering effects with high coagulant consumption.

The supernatant, after the filtration process, was a highly acidified solution. An increase in PIX 113 doses resulted in a decrease in pH even to 2.3. On the other hand, an increase in the reagent dose resulted in an improvement in the quality of the filtrate, which was associated with a decrease in turbidity and conductivity (

Table 3. Parameters of the filtrate after sludge dewatering using vacuum filtration.

PIX 113 Dose, mL/L of Sludge	2.0	4.0	6.0	10.0	20.0	30.0
pH	5.39 ± 0.06	4.87 ± 0.11	4.47 ± 0.11	3.03 ± 0.14	2.44 ± 0.06	2.27 ± 0.09
Conductance, μS/cm	169.1 ± 10.5	150.2 ± 7.3	140.0 ± 9.8	103.9 ± 7.6	76.9 ± 11.1	57.3 ± 8.6
Turbidity, NTU	216 ± 11	199 ± 10	186 ± 13	151 ± 15	144 ± 9	137 ± 3

).

The highest coefficient of determination R² in rheological analysis was obtained using the Ostwald model. The Bingham and Hershel-Bulkley model was characterised by a lower accuracy of fit of the modelled flow curve to the actual measured data (

Table 4. Results of rheological tests of sludge conditioned in combination 1.

Model		Dose of PIX 113, mL/L
		2.0	4.0	6.0	10.0	20.0	30.0
Bingham	$\eta$	0.007	0.0054	-	0.0049	0.0033	0.0068
	${\tau }_0$	1.5741	3.1821	-	3.321	3.321	1.6192
	R2	0.93	0.53	<0.5	0.85	0.59	0.86
Hershel Bulkley	${\tau }_0$	0.4635	0.2564	-	0.0001	-	0.0001
	$k$	0.1613	0.3131	-	0.765	-	0.3048
	$n$	0.538	0.501	-	0.2976	-	0.4489
	R2	0.99	0.84	<0.50	0.94	<0.50	0.94
Ostwald	$k$	1.0	1.0	1.0	1.0	2.699	1.0
	$n$	0.2437	0.2512	0.2807	0.2511	0.0858	0.242
	R2	0.98	0.96	0.93	0.99	0.97	0.98
Thixotropy		382.3	498.5	674.9	704.5	767.1	439.4

). In some cases, low R² values below 0.5 were obtained. In such cases, the accuracy and reliability of the variable rheological models were considered insufficient, and the results were not presented in tables. Analysis of the parameters in the rheological models showed no significant dependence (decreasing or increasing trends) with increasing doses of PIX 113. It can be concluded that the yield stress ${\tau }_0$ values are very low, and the hydraulic transport of this type of conditioned sludge will not be problematic [45]. This is confirmed by Wei et al., who specify that the rheological properties of sludge measured in the laboratory may have a specific relationship to the flow behaviour in practice in pipeline systems [46]. Rheological measurements, although in a limited way, can even be used to obtain information on the microbiological composition of activated sludge and settleability properties [47]. It can also be concluded that the prepared sludge belongs to the shear—thinning fluids (n < 1, Oswald model), which means a decrease in viscosity with increasing shear rate and indicates liquefaction of the sludge. A similar shear thinning phenomenon was observed in studies by Santos et al., which analysed the rheological properties of digested sludge and their mixtures with waste [48]. An increase in the applied PIX dose generally resulted in an increase in the thixotropy of the sludge (Table 4). This means that the conditioned sludge becomes a structurally unstable substrate, and its properties are significantly dependent on changes in shear rate.

3.2. Research Results—Combination 2

Conditioning of the sludge with cellulose slightly increased the rate of water release from the structure of the prepared sludge. Based on the CST test and filtration velocity, the most favourable cellulose doses were 8.0 and 16.0 g/L (Figure 4). These are very high doses, close to the dry weight of the sludge (22.6 g/L). The CST and FV indicate that the dosage of cellulose significantly limits the release of water from the sludge structure, even in comparison to raw sludge. The observed phenomenon may result from the hydrophilicity of cellulose, while the possible improvement in dewatering ability may result from the three-dimensional structure of the cellulose solution forming a filter for solid particles [49,50].

The values of resistance of filtration (SRT > 70 × 10¹³ m/kg) and hydration of sludge cake (FH > 94%) did not allow cellulose to be classified as an agent supporting the dewatering process (structure-forming agent) (Figure 5). It can be suggested that the use of cellulose and other lignocellulosic reagents, due to the filtration efficiency, requires the use of high doses. Lin et al. [51] suggest using wood chips at a dose of 60% of the dry weight of the sludge. Similarly, for the effective dewatering of aerobic-stabilised sludge conditioned using wood chips, Jing et al. [52] recommend using a dose of 80% DS of sludge. It is assumed that the use of sludge conditioning methods should result in a reduction of the water content in the dewatered sludge below 80% [43,53].

The filtrate after the sludge dewatering process was highly contaminated with dissolved and solid phases, as indicated by the values of conductivity (≈100 µS/cm) and turbidity (≈500 NTU). This effect was generally independent of the dose of Regent. Similarly, the pH of the filtrate did not change and was approximately 5.5 (

Table 5. Parameters of the filtrate after dewatering sludge conditioned with cellulose.

Celulose Dose, g/L	0.5	1.0	2.0	4.0	8.0	16.0	32.0
pH,	5.5 ± 0.6	5.5 ± 0.3	5.4 ± 0.7	5.4 ± 0.2	5.5 ± 0.9	5.6 ± 0.9	5.7 ± 0.6
Conductance, μS/cm	85.3 ± 6.3	71.9 ± 9.1	91.9 ± 4.2	170.5 ± 16.8	108.8 ± 5.8	119.0 ± 7.7	93.1 ± 7.6
Turbidity, NTU	522 ± 29.7	508 ± 36.5	542 ± 29.7	471 ± 29.7	510 ± 34.5	503 ± 21.3	462 ± 9.7

). Suopajärvi et al. [54] obtained significantly better results in the dewatering of activated sludge and the quality of the supernatant using cationic nanocellulose. Importantly, the small dose of nanocellulose, 1.0 kg/t DS, resulted in an 80% reduction in impurities in the supernatant, including turbidity and chemical oxygen demand (in the current studies, the dose was 0.02–1.4 kg/kg DS).

An increase in the cellulose dose resulted in a very heterogeneous increase in thixotropy (

Table 6. Rheological characteristics of sludge conditioned in combination 2.

Model		Dose of Celulose, g/L
		0.5	1.0	2.0	4.0	8.0	16.0	32.0
Bingham	$\eta$	-	0.0071	-	0.0066	-	-	-
	${\tau }_0$	-	1.3833	-	2.3606	-	-	-
	R2	<0.50	0.94	<0.50	0.74	<0.50	<0.50	<0.50
Hershel Bulkley	${\tau }_0$	-	0.838	-		-	-	-
	$k$	-	0.0598	-	0.6481	-	-	-
	$n$	-	0.6776	-	0.3468	-	-	-
	R2	<0.50	0.99	<0.50	0.88	<0.50	<0.50	<0.50
Ostwald	$k$	2.2661	1.0	1.0	1.0	-	-	-
	$n$	0.1268	0.235	0.2936	0.2707	-	-	-
	R2	0.91	0.97	0.92	0.99	<0.50	<0.50	<0.50
Thixotropy		733.7	345.9	1239.6	434.9	1131.2	2343.2	11,234.4

). Compared to sludge prepared with PIX 113, the preparation with cellulose significantly increases the thixotropy of the sludge, and therefore, the resulting structures are very unstable. In addition, modelling shear stresses was very difficult. In most cases, the accuracy of the rheological models was not satisfactory (R² < 0.5). The irregular results of rheological analysis can be explained by the strong dependence of rheological properties on the physical and chemical parameters of sludge, including dry solids, pH, temperature and protein content in the sludge [55,56]. In the case of the conducted research, the dosed reagents caused both a significant change in pH (PIX 113) and a very significant change in dry solids (dosing of cellulose and perlite at doses up to 32 g/L, 1.42 g/g DS).

3.3. Research Results—Combination 3

Perlite conditioning was beneficial only after using the highest doses, i.e., 16.0 and 32.0 g/L (Figure 6). Comparatively, a perlite dose of 32.0 g/L allowed a reduction in CSK (235 s) and an increase in filtration speed (1.2 L/h), which corresponded to the use of the lowest doses of PIX 113 (2.0 and 4.0 mL/L).

Filtration of sludge conditioned with perlite also did not show that the tested reagent could be a good structure-forming agent (Figure 7). Only the highest dose of perlite (32.0 g/L, 1.42 g/g DS) improved the solid phase separation efficiency. Only at this dose were acceptable parameters achieved, i.e., SRF = 2.59 × 10¹² m/kg and FH = 79.8%. These values are similar to the effects of PIX 113 sludge conditioning at a dose of approximately 2.0–6.0 mL/L (0.13–0.40 g/g DS). For comparison, Xi et al. [57] showed that the use of flu ash in sludge conditioning at a dose of 10–15 g/g DS allows obtaining SRF = 1.03 × 10¹¹ m/kg and FH = 74%. However, to achieve very low hydration of the dewatered sludge of 68%, a dose of 30 g/g DS was necessary. In the research of Qi et al. [58], the filtration efficiency of digested sludge conditioned with lignite was increased six times. However, it required a solid mass ratio to sludge of 1:1.

There was no evidence that the addition of perlite had any chemical effect on the prepared sludge. The pH of the filtrate remained constant at about 5.6. Perlite conditioning, regardless of dose, increased filtrate turbidity (≈600 NTU). However, conductivity measurements of the filtrate showed that conditioning with perlite reduced the content of dissolved solids (

Table 7. Parameters of the filtrate after dewatering sludge conditioned with perlite.

Perlite Dose, g/L	0.5	1.0	2.0	4.0	8.0	16.0	32.0
pH,	5.6 ± 0.04	5.6 ± 0.07	5.6 ± 0.06	5.6 ± 0.04	5.6 ± 0.03	5.6 ± 0.01	5.6 ± 0.06
Conductance, μS/cm	142.1 ± 3.1	56.7 ± 3.6	46.1 ± 2.8	35.1 ± 2.0	30.8 ± 2.1	26.3 ± 1.9	24.0 ± 2.0
Turbidity, NTU	608 ± 33.9	586 ± 42.5	571 ± 20.6	601 ± 49.0	598 ± 20.7	556 ± 23.4	640 ± 36.2

).

The properties of perlite are a possible explanation for the changes in filtrate quality. Perlite does not react with the filtered liquid, can be crushed during mixing and works as an adsorbent [59,60].

The increase in the perlite dose was not correlated with the sludge thixotropy. The alternating increase and decrease in thixotropy may constitute the basis for determining the problems of rheological measurements, especially due to the heterogeneity of the composition and properties of sludge. The most accurate model describing the rheological properties of the tested sludge was the Ostwald model. Based on this model, it is only possible to determine the phenomenon of sediment thinning by shear (n < 1). However, these changes were not regular and independent of the increasing dose of the reagent (

Table 8. Rheological characteristics of sludge conditioned in combination 3.

Model		Dose of Perlite, g/L
		0.5	1.0	2.0	4.0	8.0	16.0	32.0
Bingham	$\eta$	0.0093	-	-	-	0.0087	0.0094	0.0153
	${\tau }_0$	1.6415	-	-	-	1.2473	2.6099	4.3343
	R2	0.94	<0.50	<0.50	<0.50	0.96	0.89	0.92
Hershel Bulkley	${\tau }_0$	0.2069	-	-	-	0.0815	-	-
	$k$	0.1886	-	-	-	0.1348	-	-
	$n$	0.558	-	-	-	0.5959	-	-
	R2	0.99	<0.50	<0.50	<0.50	0.99	<0.5	<0.50
Ostwald	$k$	1.0	6.5365	1.0	3.9677	1.0	1.0	1.0
	$n$	0.2765	0.0085	0.2654	0.0259	0.2516	0.3099	0.3965
	R2	0.98	0.91	0.95	0.94	0.96	0.99	0.99
Thixotropy		309.7	833.5	906.8	709.9	603.7	618.4	1012.3

).

Regardless of the conditioning method, the parameters characterising the sludge filtration process (CSK, FV, SRT, FH) were highly correlated, R > 0.75 (

Table 9. Correlation matrix of the examined parameters of conditioned and dewatered sewage sludge.

Parameter	FH	FV	CST	pH	C	T
Combination 1—PIX 113
SRF	0.98	−0.96	1.00	0.78	0.74	0.83
FH		−0.98	0.97	0.88	0.86	0.92
FV			−0.95	−0.91	−0.88	−0.94
CST				0.76	0.72	0.81
pH					0.99	1.00
C						0.98
Combination 2—Cellulose
SRF	0.75	0.65	0.94	−0.28	−0.94	0.67
FH		0.85	0.92	−0.79	−0.74	0.90
FV			0.78	−0.63	−0.70	0.65
CST				−0.51	−0.92	0.79
pH					0.26	−0.82
C						−0.65
Combination 3—Perlite
SRF	0.77	−0.88	0.93	0.36	0.72	−0.20
FH		−0.93	0.90	−0.12	0.45	−0.45
FV			−0.95	−0.21	−0.67	0.47
CST				0.23	0.74	−0.19
pH					0.61	−0.02
C						0.12

). In the case of sludge conditioning with PIX 113 coagulant, the filtrate parameters (pH, conductance and turbidity) were also found to be proportional to the changes in CSK, FV, SRT, and FH. Conditioning the sludge with cellulose, especially perlite, led to deterioration but also to a lack of correlation between the parameters characterising the filtration efficiency and the quality of the filtrate. The observed correlations may be the basis for the conclusion that the action of natural reagents leads to heterogeneous changes in the properties of the sludge. It can also be related to the non-uniform results of rheological analysis.

3.4. Research Results—Combination 4 and 5

Cellulose and perlite doses of 8.0 g/L and 16.0 g/L were selected for research on the dual conditioning of sludge. This choice is due to the fact that the action of cellulose and perlite can only be effective in large doses. In dual conditioning, the PIX 113 iron coagulant was used at a previously determined optimal dose of 6.0 mL/L.

Sequential dual conditioning of sludge with PIX 113 and cellulose or perlite (regardless of the order of dosing) did not result in favourable changes in CSK values. The independent use of PIX 113 coagulum at a dose of 6.0 mL/L was more effective in reducing capillary suction time (Figure 8). Similarly, the results obtained for vacuum filtration speed showed that the amount of filtrate during separation was lower than when using only PIX 113 (Figure 9).

Conditioning the sludge with PIX 113 and cellulose (or in the reverse order) did not improve the filtration ability. The obtained SRF values (>5.0 × 10¹² m/kg) do not qualify the conditioning method as justified for use. A more advantageous method of dual conditioning was the use of iron coagulant and perlite, especially at a dose of 16.0 g/L (Figure 10). Similar dependencies were observed in the hydration of the final filter cake. In dual conditioning with cellulose, FH values exceeded 80%. Replacing cellulose with perlite slightly reduced the water content in the filter cake to approximately 78% (Figure 11).

The pH of the filtrate separated during dewatering on a vacuum filter was constant and independent of the combinations used and was approximately 4.6 (

Table 10. Quality parameters of the filtrate after separation of dually conditioned sludge.

Combination	pH	Conductivity, $\mathbf{\mu }\mathbf{S}\mathbf{·}\mathbf{c}{\mathbf{m}}^{-1}$	Turbidity, NTU
PIX 113 (6.0 mL/L), (comparison level)	4.47 ± 0.11 (a)	140.1 ± 9.8 (a)	186 ±13 (a)
PIX 113 + Cellulose (6.0 mL/L + 8.0 g/L)	4.63 ± 0.33 (a)	232.1 ± 21.4 (b)	167 ± 14 (b)
PIX 113 + Cellulose (6.0 mL/L + 16.0 g/L)	4.59 ± 0.40 (a)	240.6 ± 13.8 (b)	165 ± 21 (b)
Cellulose + PIX 113 (8.0 g/L + 6.0 mL/L)	4.62 ± 0.32 (a)	205.7 ± 10.8 (b)	143 ± 11 (b)
Cellulose + PIX 113 (16.0 g/L + 6.0 mL/L)	4.60 ± 0.13 (a)	240.5 ± 16.1 (b)	148 ± 15 (b)
PIX 113 + Perlite (6.0 mL/L + 8.0 g/L)	4.66 ± 0.11 (a)	332.7 ± 32.4 (c)	144 ± 17 (b)
PIX 113 + Perlite (6.0 mL/L + 16.0 g/L)	4.38 ± 0.90 (a)	297.0 ± 21.5 (b)	143 ± 13 (b)
Perlite + PIX 113 (8.0 g/L + 6.0 mL/L)	4.43 ± 0.12 (a)	291.2 ± 24.7 (b)	145 ± 19 (b)
Perlite + PIX 113 (16.0 g/L + 6.0 mL/L)	4.58 ± 0.14 (a)	344.0 ± 23.8 (c)	146 ± 10 (b)

Values marked with the same letter (a, b, c) are not statistically significant. The sludge treatment methods did not show statistically significant differences in conditioning efficiency.

). This value corresponds to the pH of the filtrate using PIX 113 at a dose of 6.0 mL/L. The addition of PIX 113 and cellulose resulted in an increase in conductivity (average 240 μS/cm) compared to the interaction of PIX 113 alone (6.0 mL/L–140 μS/cm). Even higher conductance (>300 μS/cm) was recorded for sediment samples dually conditioned with coagulant and perlite. Filtrate turbidity was comparable for all conditioning combinations and ranged from 140 to 160 NTU. It can be mentioned that the turbidity of the filtrate after dual conditioning with perlite was approximately 40 NTU lower than for sludge conditioned individually with PIX 113 (Table 10).

Statistical analysis of the filtration parameters of sludge and filtrate showed that there were no significant differences in the effects of using cellulose or perlite. Differences in dewatering efficiency were observed only in the case of specific resistance of filtration, which resulted from the use of cellulose (group a) or perlite (group b). Moreover, only the quality of the filtrate (conductivity, turbidity) after conditioning the sludge with the PIX 113 coagulant was different from the other combinations used. This was probably related to the change in the amount of dissolved particles, colloids and fine suspensions entering the filtrate as a result of conditioning the sludge with cellulose and perlite.

As in the previous combinations, the Oswald rheological model described the shear stresses most accurately. Based on the variables of this model, a lower flow behaviour index n was noted for sludge prepared with cellulose than with perlite. Thus, cellulose-conditioned sludge was more susceptible to the shear-thinning phenomenon. Moreover, increasing the dose of cellulose or perlite increased the n values, which limited the shear thinning phenomenon. The obtained n values were not greater than 0.40. Comparatively, at similar dry solids concentrations, in the studies by Wójcik and Stachowicz [61], n = 0.4–0.7 were obtained for sludge prepared with biomass ash. In the conducted research, but also in the research of Ruiz-Hernando et al. [62] and Cao et al. [63], Oswald’s rheological model was the most accurate. On the contrary, other authors [64,65] analysing the rheological properties of wastewater sludge indicated the Herschel and Bulkley model as more appropriate.

The dually conditioned sludge had greater thixotropy than the individually conditioned PIX 113 (

Table 11. Rheological properties of sludge conditioned dually with PIX 113 and cellulose.

Model		Dose of Celulose, g/L
		6.0 mL/L + 4.0 g/L	6.0 mL/L + 8.0 g/L	4.0 g/L + 6.0 mL/L	8.0 g/L + 6.0 mL/L
Bingham	$\eta$	-	-	-	0.0061
	${\tau }_0$	-	-	-	3.4362
	R2	<0.5	<0.5	<0.5	0.83
Hershel Bulkley	${\tau }_0$	-	-	-	0.0011
	$k$	-	-	-	1.1564
	$n$	-	-	-	0.2764
	R2	<0.5	<0.5	<0.5	0.95
Ostwald	$k$	11.5798	51.7884	3.2803	1.1198
	$n$	0.0038	0.102	0.0853	0.2823
	R2	0.94	0.98	0.96	0.99
Thixotropy		2236.8	4598.2	673.9	847.9

and

Table 12. Rheological properties of sludge conditioned dually with PIX 113 and perlite.

Model		Dose of Perlite, g/L
		6.0 mL/L + 4.0 g/L	6.0 mL/L + 8.0 g/L	4.0 g/L + 6.0 mL/L	8.0 g/L + 6.0 mL/L
Bingham	$\eta$	0.0068	0.0074	0.0109	-
	${\tau }_0$	4.3328	2.9374	3.8754	-
	R2	0.67	0.90	0.68	<0.5
Hershel Bulkley	${\tau }_0$	0.0024	0.0018	3.8209	-
	$k$	1.5505	0.7508	0.0127	-
	$n$	0.2579	0.35	0.9768	-
	R2	0.82	0.99	0.68	<0.5
Ostwald	$k$	1.4733	1.0000	1.7875	1.4263
	$n$	0.2662	0.3006	0.2479	0.4019
	R2	0.99	0.99	0.95	0.97
Thixotropy		963.6	1017.7	1637.3	5321.5

). Additionally, an increased dose of cellulose or perlite (18.0 g/L) caused the sludge to be more susceptible to structure destruction by shear forces. It was also observed that higher thixotropy occurred when sludge was prepared in the PIX 113 + cellulose sequence than in the reverse order. The perlite dosing sequence in dual conditioning changed the thixotropy of the sludge in the opposite way (Table 11 and Table 12).

The analysis of thixotropy, similar to the analysis of the rheological models, did not provide a basis for identifying clear trends. However, the characteristics of sludge, the method of conditioning and the specificity of rheological tests make them one of the most difficult measurements [66].

The filtrate after dewatering the conditioned sludge was also used to determine the zeta potential. An increase in the dose of iron coagulant resulted in a rapid reduction in zeta potential. Already at a PIX 113 dose of 4.0 mL/L, the zeta potential of filtrate particles was definitely close to the isoelectric point (

Table 13. Zeta potential determined in the filtrate after the separation process of single/dual conditioned sludge.

Conditioning	Dose	Zeta Potential, mV
Raw sludge	-	−18.45	±1.26
PIX 113	(2 mL/L)	−11.45	±3.54
PIX 113	(4 mL/L)	−4.55	±1.09
PIX 113	(6 mL/L)	−3.61	±1.15
PIX 113	(10 mL/L)	−2.03	±0.92
Celulose	(8 g/L)	−23.4	±6.30
Celulose	(16 g/L)	−24.6	±4.34
Perlite	(8 g/L)	−15.4	±0.96
Perlite	(16 g/L)	−14.3	±0.71
PIX 113 + Cellulose	(6.0 mL/L + 8.0 g/L)	−3.4	±0.34
PIX 113 + Cellulose	(6.0 mL/L + 16.0 g/L)	−3.2	±0.51
Cellulose + PIX 113	(8.0 g/L + 6.0 mL/L)	−4.1	±0.87
Cellulose + PIX 113	(16.0 g/L + 6.0 mL/L)	−4.8	±1.29
PIX 113 + Perlite	(6.0 mL/L + 8.0 g/L)	−3.5	±0.64
PIX 113 + Perlite	(6.0 mL/L + 16.0 g/L)	−2.6	±0.13
Perlite + PIX 113	(8.0 g/L + 6.0 mL/L)	−3.0	±0.63
Perlite + PIX 113	(16.0 g/L + 6.0 mL/L)	−2.1	±0.20

). It is possible to associate the increase in zeta potential with the decomposition of negatively charged organic matter and the neutralisation of negative charges by positively charged iron ions [67,68]. Cellulose dosage resulted in an increase in potential, which could be related to the hydrophobicity of cellulose. Differently, dosing perlite reduced the zeta potential, although only by 4 mV compared to raw sludge. In dual conditioning, regardless of the type and dose of reagents used, the zeta potential ranged from 2.0 to 4.0 mV. The results presented indicate that in dual conditioning, the dosage and dose of the coagulant had the greatest impact on the potential reduction in zeta. Differences in zeta potential in dual conditioning combinations may result from changes in the pH of the solution and particle size changes due to the dosing of natural reagents [43,69].

4. Conclusions

• Dosing the commercial iron coagulant PIX 113 iron coagulant was the most beneficial method of sludge conditioning. Therefore, the greatest influence on the dewatering effect was the conditioner dosage, which caused destabilisation and flocculation of the sludge particles.
• Cellulose and perlite cannot be described as structure-forming agents supporting dewatering. Sludge single conditioned with cellulose or perlite excessively bound water in the structure of the sludge and ultimately limited filtration.
• The combination of reagents (PIX 113, cellulose, perlite) in dual conditioning resulted in the deterioration of sludge dewatering. Dual conditioning can only be effective if a combination of factors whose single action (single conditioning) is beneficial for the technological properties of the sludge.
• Dewatering of single/dual conditioned sludge, regardless of the doses and combinations used, generated a highly contaminated filtrate. The pH, conductivity and turbidity of the filtrate indicated that the resulting filtrate may be a problematic substrate for further purification.
• The use of rheological tests as a tool useful in determining the properties of sludge, and especially the effectiveness of conditioning methods, was limited. A high variability of the rheological parameters and no dependence on the technological parameters of the sludge were observed. The conditioned sludge were structurally unstable substrates, and their properties significantly depended on changes in shear rate. A conclusion can be drawn regarding increasing the mixing gradient or homogenising the mixture of sludge, especially with cellulose or perlite. However, it is necessary to check whether such processes affect the efficiency of dewatering and the quality of rheological analyses.
• The zeta potential was consistent with sediment dewatering effects. It seems justified to use zeta potential measurements for the initial selection of conditioners and their mixtures with sludge prior to dewatering. Minimising the zeta potential could justify the use of types and doses of conditioners of a coagulant and sorbent nature. However, it is recognised that a practical assessment of the susceptibility of sludge to dewatering can only be made on the basis of key technological parameters such as specific resistance of filtration resistance and sludge cake hydration.