Upgrading Anaerobic Sludge Digestion by Using an Oil Refinery By-Product

Abstract

Carbon-based conductive additives have been studied for their positive effects on anaerobic digestion (AD) using synthetic substrates, but their importance in wastewater sludge digestion has not been sufficiently explored. This research investigated and compared the effects of two conductive materials (graphene and petroleum coke) with and without trace metal supplementation. The results indicated that supplementing reactors with graphene and petroleum coke could significantly improve biogas production. The supplementation of 1 g/L petroleum coke and 2 g/L graphene, without trace metal addition, led to an increase in the biogas production by 19.10 ± 1.04% and 16.97 ± 5.00%, respectively. Thus, it can be concluded that petroleum coke, which is an oil refinery by-product, can be used to enhance biogas production in a similar way to other carbon-based conductive materials that are currently available on the market. Moreover, using petroleum coke and graphene, the average chemical oxygen demand (COD) removal was 42.84 ± 1.23% and 42.80 ± 0.45%, respectively, without the addition of trace metals. On the other hand, supplementation of the reactors with trace elements resulted in a COD removal of 34.65 ± 0.43% and 34.05 ± 0.45% using petroleum coke and graphene, respectively.

1. Introduction

In anaerobic digestion, organic matter is decomposed by various microorganisms under anaerobic conditions. Biogas and nitrogen-rich organic residues are the resulting products of AD. Due to its ability to reduce chemical oxygen demand (COD) and biological oxygen demand (BOD) from waste streams and produce renewable energy, this technology has successfully been implemented in the treatment of agricultural wastes, food wastes, and wastewater sludge [1,2].

Due to the global energy crisis, which poses a significant challenge to socioeconomic affairs and the sustainability of the environment, it is imperative to find an alternative method of energy production [3]. The potential of AD as a bioenergy source has drawn considerable attention in recent years, especially due to its potential to produce a valuable biogas that primarily consists of methane CH₄ [1,4]. However, a number of challenges have been associated with the use of AD technology (low efficiency and stability, metal accumulation, low organic loading rate, etc.), and it is worth investigating these to find solutions [3].

In order to solve these problems and enhance methane generation, certain conductive materials (CMs) have been employed [5] including granular activated carbon (GAC) [6], graphene [7], biochar [8], carbon fiber [9], carbon cloth [10], magnetite [11], etc. The improvements in the AD process obtained by supplementing CMs in the digester are believed to be due to the direct interspecies electron transfer (DIET) between certain species of microorganisms [12,13,14,15,16,17].

Graphene has received increased interest for biotechnological applications, such as electrode materials in microbial fuel cells, because of its excellent electrical conductivity [18]. In some circumstances, graphene is known to have antimicrobial characteristics [19]. It has a high electrical conductivity, a high mechanical strength, and a small surface area. These properties suggest that it may have the potential to significantly improve DIET and AD efficiency [20]. Lin et al. (2017) [20] studied charcoal and graphene for their effect on anaerobic digestion. They attributed the higher AD efficiency achieved by the addition of graphene to its smaller size compared to charcoal, which resulted in a larger specific surface area and improved interactions with microbes.

Crude oil refineries produce petroleum coke as a waste product in the coking process. Excepting the inhalation of coke particles smaller than 10 μm, this is a stable substance and causes no significant environmental toxicity concerns [21]. Metallurgical industries (iron and steel industries) use petroleum coke as a carbon source in their processes to reduce iron ore to iron [22]. It is also used as a fuel source in cement kilns and as a graphite electrodes [23]. Furthermore, studies have indicated that petroleum coke has the ability to absorb heavy metals from wastewater, as well as naphthenic acids [24,25,26]. Additionally, petroleum coke and its ash have been shown in recent studies to be suitable for soil remedies [27].

Furthermore, the microorganisms in the digester require certain trace elements to grow and survive. Depending on their concentration, trace elements can act as inhibitors, stimulants or even toxicants in digestates. These materials include nickel (Ni), cobalt (Co), molybdenum (Mo), iron (Fe), selenium (Se), tungsten (W), zinc (Zn), copper (Cu), and manganese (Mn) [28,29,30,31]. Insufficient amounts of these essential elements may negatively affect the function and activity of key enzymes. Hence, trace element supplementation may also be an effective way of enhancing AD performance [31,32,33]. The effect of trace metal supplementation on the anaerobic digestion of food waste was studied by Akturk and Demirer (2020) [34]. They found that trace metals supplementation enhanced the production of biogas and methane yields.

Several studies have demonstated that carbon-based conductive materials stimulate the anaerobic digestion process in methanogenic digesters by stimulating DIET. Despite their positive effects on anaerobic digestion (AD) using synthetic substrates, these additives have not been adequately studied for their importance in wastewater sludge digestion. Therefore, in this study, the effect of two carbonaceous conductive materials, graphene and petroleum coke, on the AD of wastewater sludge was investigated. Our interest in exploring whether petroleum coke has a positive effect on the anaerobic digestion of wastewater sludge was further motivated by the fact that, to the best of our knowledge, this electrically conductive material [35,36,37] has not been applied to the anaerobic digestion of sludge. Therefore, the main purposes of this study are to: (1) investigate the effect of petroleum coke on biogas production from wastewater sludge, (2) compare the capability of graphene and petroleum coke on the AD process, and (3) observe the influence of trace elements on AD performance.

2. Materials and Methods

2.1. Substrate

Raw sludge was obtained from a municipal wastewater treatment plant located in Mount Pleasant, Michigan, and used as a substrate for this study. This sample had a total and volatile solids (TS and VS) concentration of 26,100 ± 300 and 17,500 ± 400 mg/L, respectively. This sludge contained a total chemical oxygen demand (COD) concentration of 33,940 ± 315 mg/L.

Table 1. Characterization of substrate and inoculum.

Parameter	Substrate	Inoculum
COD (mg/L)	33,940 ± 315	41,660 ± 1082
TS (mg/L)	26,067 ± 306	44,267 ± 416
VS (mg/L)	17,467 ± 416	33,800 ± 200
Total Nitrogen (TN) (mg/L)	2073 ± 77	3737 ± 107
Total Phosphorus (TP) (mg/L)	1285 ± 18	1134 ± 37

presents a summary of the characteristics of this sample.

2.2. Inoculum

The inoculum was obtained from the anaerobic digester at Michigan State University in East Lansing, Michigan. Following the collection, the sample was left undisturbed at room temperature to allow for starvation and settling of solids. Furthermore, one-third of the supernatant was discarded, while the remainder was used as an inoculum for the AD process. The total chemical oxygen demand (COD) of this sample was 41,660 ± 1082 mg/L, with TS and VS contents of 44,267 ± 416 and 33,800 ± 200 mg/L, respectively. The characteristics of this sample are summarized in Table 1.

2.3. Conductive Material

Petroleum coke was acquired from the University of Tulsa, Oklahoma. The samples were obtained from a pilot plant operating at a temperature of 900 °F and 70 Psig at the University of Tulsa Delayed Coker Project (TUDCP). Samples were characterized by Loring Laboratories LTD in Alberta, Canada. The results of the analysis are shown in

Table 2. Characterization of the petroleum-coke used in this study.

Analysis	Content (%)
Carbon	87.5
Hydrogen	3.4
Nitrogen	1.4
H2O	0.50
Volatile Matter	10.02
Sulphur	6.1
Ash	0.03

[26]. Moreover, the graphene flake powder (0.3–1 nm) used in this experiment was purchased from the AZ Laboratories.

2.4. Analytical Methods

The standard methods were adopted to measure both the TS (2540 B) and the VS (2540E) [37]. pH values were measured using a digital pH meter (Apera PH700). To measure COD (HACH/EPA method 8000), TN (HACH/EPA method 10072), and TP (HACH/EPA method 8190), Hach kit vials were used (Hach, Loveland, CO, USA). The methane content of biogas was determined using an Agilent 8860 gas chromatograph (GC) equipped with an Agilent 7697A headspace autosampler, a flame ionization detector (FID) and a column (Agilent GS-GasPro GC Column, 15 mm length, 0.32 mm diameter). A water displacement was utilized to measure the amount of biogas produced in each reactor [38,39].

2.5. Experimental Setups

Batch reactors with total and effective volumes of 250 and 150 mL, respectively, were used to examine the effect of conductive materials on the biogas production potential of wastewater sludge with and without supplementation of trace elements.

The first experimental setup was run to determine and compare the effect of petroleum-coke and graphene on anaerobic digestion without adding any trace metals. For this purpose, 0, 0.5, 1.0, and 2.0 g/L of these two conductive materials were used. The substrate to inoculum ratio was maintained at 3.0 with COD concentration of 38,549 ± 1047 mg/L in all reactors (T1-T6), and deionized water was added to adjust the total volume to 150 mL. All the reactors were purged with nitrogen gas, sealed with rubber stoppers, and maintained at 34.5 ± 0.5 °C with constant mixing at 120 rpm during incubation. The initial pH value in the reactors was 7.85 ± 0.04.

A second experimental setup was run to determine the effect of trace metals on the biogas production of wastewater sludge using conductive materials in the presence of a source of alkalinity in the reactors. All reactors (NT1-NT6) contained S/I ratio of 3.0, COD concentration of 19,393 ± 480 mg/L, 50 mL trace metal solution, and 6000 mg/L NaHCO₃ as an alkalinity source. The trace metal solution contained 10 mg/L FeCl₂.4H₂O, 2 mg/L ZnCl₂, 0.5 mg/L NiCl₂.6H₂O, 0.2 mg/L NaWO₄.2H₂O, 0.05 mg/L CoCl₂.6H₂O, 0.05 mg/L CuCl₂.2H₂O, 0.05 mg/L Na₂MoO₄.2H₂O. The initial pH value in the reactors in the second setup was 8.00 ± 0.01. The details of both experimental setups are provided in

Table 3. The experimental setups.

First Setup: Without Trace Metal Supplementation
Reactor	Substrate (mL)	Inoculum (mL)	Trace Metals	Alkalinity	Graphene (g/L)	Petroleum Coke (g/L)
Blank	116	0	−	−	0	0
Control	116	20	−	−	0	0
T1	116	20	−	−	0	0.5
T2	116	20	−	−	0	1
T3	116	20	−	−	0	2
T4	116	20	−	−	0.5	0
T5	116	20	−	−	1	0
T6	116	20	−	−	2	0
Second Setup: With Trace Metal Supplemetation
Reactor	Substrate (mL)	Inoculum (mL)	Trace Metals	Alkalinity	Graphene (g/L)	Petroleum coke (g/L)
Blank	58	0	−	+	0	0
Control	58	10	−	+	0	0
Control 2	58	10	+	+	0	0
NT1	58	10	+	+	0	0.5
NT2	58	10	+	+	0	1
NT3	58	10	+	+	0	2
NT4	58	10	+	+	0.5	0
NT5	58	10	+	+	1	0
NT6	58	10	+	+	2	0

.

In the reactors, a substrate to inoculum ratio of 3.0 was maintained, since low substrate concentrations may result in lower biomass activity, while high substrate concentrations can inhibit the anaerobic digestion of thickened sludge due to the presence of volatile fatty acids and lower pH levels [40].

2.6. Statistical Analysis

The experiments were all carried out in duplicate, and results were expressed as a mean ± standard deviation.

Statistical significance was evaluated through a single sample t-test. The following formula can be used to determine a t-value:

$$t=\left(\overline{x}-\mu \right)/\left(s/\surd n\right)$$

where $\overline{x}$ is the mean of the sample, μ is the theoretical value, s is the standard deviation of the sample, and n is the sample size [41].

The resulting t-value can be compared with the critical t-values for the appropriate degrees of freedom (number of cases minus 1) and significance level found in the t-distribution table provided by Wheeler et al. (2010) [42]. If the calculated t-value is less than the critical t-value, it can be concluded that the values of the sample are acceptable. However, if the calculated t-value is greater than the critical t-value, then the values in the sample are not acceptable and should be rejected.

For this study, a p-value ≤ 0.05 was defined as statistically significant [43,44].

3. Results and Discussion

3.1. Effect of Petroleum Coke and Graphene on Biogas Production

Biogas was measured for 36 and 49 consecutive days for the setups without and with trace metal supplementation, respectively. The results indicated that supplementing the reactors with graphene and petroleum coke, without the addition of trace elements, promoted the production of biogas from wastewater sludge. As shown in Figure 1a, cumulative biogas production in the first experimental setup increased by 14.05 ± 5%, 19.10 ± 1.04%, and 7.45 ± 0.09% by adding 0.5, 1, and 2 g/L petroleum coke in the digester, respectively. Moreover, supplementing reactors with 0.5, 1, and 2 g/L graphene in this setup increased the cumulative biogas production by 16.52 ± 5%, 7.03 ± 4.41%, and 16.97 ± 5%, respectively (Figure 1b). The results demonstrated that 1 g/L petroleum coke and 2 g/L graphene were the optimum concentrations for biogas production when no trace metals were added. Moreover, a lower amount of petroleum coke (1 g/L) can lead to increased biogas production in comparison to the use of graphene under the same conditions. According to previous studies, the increase in biogas production using graphene is due to the direct interspecies electron transfer (DIET) induced by this material [21,45,46]. Considering the fact that carbon-based conductive materials can enhance DIET and, therefore, biogas production [46], an increase in the yield of biogas using petroleum coke can be attributed to the improved DIET stimulated by this carbonaceous conductive material. Lin et al. (2017) [21] investigated the effect of graphene on the biogas production of activated sludge (mainly containing cellulose). According to their findings, 1.0 g/L of graphene could increase biogas production by 25.0% through the promotion of DIET.

Figure 2 illustrates the normalized biogas production when reactors were supplemented with trace metals. The experiments showed that using 0.5 g/L petroleum coke resulted in a 6.89 ± 5% decrease in biogas production, while by using 1, and 2 g/L petroleum coke, the cumulative biogas yield increased by 0.94 ± 5% and 14.22 ± 4.30%, respectively, when compared to the control reactors (Figure 2a). On the other hand, the addition of graphene (Figure 2b) resulted in a slight reduction in the production of biogas. This negligible decrease could be due to the experimental errors. In this study, values within the range of ±5% are considered statistically significant; therefore, the 0.94% increase in biogas production can be ignored. Consequently, petroleum coke is capable of increasing the production of biogas from anaerobic digestion, even at lower concentrations, in comparison to other conductive additives. The effect of different CMs on biogas production has been investigated in various studies and has been reported to increase biogas production by 10–20% [13,47,48].

GC headspace was used to determine the biogas content of each setup when petroleum coke and graphene were used at their optimum concentrations. According to the results, the methane content of the produced biogas using 1 g/L petroleum coke and 2 g/L graphene was 67.31 ± 3.30% and 61.76 ± 1.78% in the setup with no trace metal addition, and 64.95 ± 2.22% and 57.60 ± 1.52% using 2g/L petroleum coke and 2 g/L graphene, respectively, when trace metal solutions were used. Accordingly, given that the AD process produces biogas with a methane content ranging between 55 and 75% of its total volume [49], the results of the two setups indicate that the cumulative methane yield was within this range.

To predict and compare methane production with the experimental results, a modified Gompertz model was used [50].

P = P_max.exp{−exp[(R_max.e.(λ − t)/P_max) + 1]}

This equation can be expressed as follows: P (mL) is the cumulative amount of CH₄ produced at time t, P_max (mL) is the maximum amount of CH₄ produced at the end of incubation, t (h) is the time, R_max (mL/h) is the maximum production rate of CH₄, λ(h) is the lag phase, and e is equal to 2.71828.

In this research, the Gompertz model was applied to reactors that contained the optimum amount of CMs. Considering the results of the study provided in

Table 4. Modified Gompertz model parameters.

Without Trace Metal Supplementation
	Rmax (mL/h)	Pmax (mL)	λ (h)	R2
Petroleum coke (1 g/L)	3.61	354.7	172.8	0.987
Graphene (2 g/L)	3.64	348.3	172.8	0.986
With Trace Metal Supplementation
	Rmax (mL/h)	Pmax (mL)	λ (h)	R2
Petroleum coke (2 g/L)	1.025	121.7	46.08	0.989
Graphene (2 g/L)	1.42	107.2	46.08	0.919

, cumulative methane production followed the modified Gompertz model. It is also worth noting that the lag phase was decreased by 73.33% in the reactors supplemented with trace metals.

3.2. COD Removal

The initial COD concentrations in the reactors without and with trace elements supplementation were kept constant at 38,549 ± 1047 and 19,393 ± 480 mg/L, respectively. Figure 3 illustrates the total COD removal in both experimental setups. Using petroleum coke and graphene without the addition of trace elements (Figure 3a), the average COD removal was 42.84 ± 1.23% and 42.80 ± 0.45%, respectively, while adding the same amounts of the aforementioned conductive materials in the presence of trace elements (Figure 3b) indicated a COD removal of 34.65 ± 0.43% and 34.05 ± 0.45%, respectively. It is important to emphasize that the COD removal efficiency in the reactors without trace metal addition was quite favorable, considering the fact that the typical range of COD removal efficiency in AD of sludge is 40–50% [51].

The relatively lower performance observed in the second setup could be due to the addition of trace elements. It has been demonstrated that a trace element concentration of less than a certain amount can limit the growth of methanogen cultures in terms of their density, which negatively affects the process of anaerobic digestion [52].

3.3. pH Variation and Organic Matter Removal

The initial and final pH of each reactor was measured in both experimental setups. The average values of the initial and final pH without trace elements’ supplementation were 7.85 ± 0.04 and 7.09 ± 0.06, respectively, while these values for the setup supplemented with trace metals were 8.00 ± 0.01 and 7.47 ± 0.05, respectively. Despite a decrease in pH in the reactors, the pH was still within a range that methanogens could tolerate.

The VS concentration is an indicator of an organic fraction within the substrate, as it represents a portion of the total solids [53]. It indicates the potential of biogas yield from the initial material [54]. In this study, the average VS removal of 35.17 ± 0.01% (without trace metal supplementation) and 33.11 ± 0.02% (with trace elements addition) were achieved. In comparison to control reactors, volatile solids removal, when there was no trace elements supplementation, increased by 7.71 ± 0.01%, and with the addition of trace metals, an increase of 5.78 ± 0.02% was observed. A study conducted by Dastyar et al. (2021) [55] used powdered activated carbon as a conductive additive in anaerobic digestion and reported a VS removal of 34%.

4. Conclusions

Carbonaceous conductive materials’ supplementation was found to improve methane yield from organic matter. This research explored the effect of graphene and petroleum coke on the anaerobic digestion of wastewater sludge with and without trace metal supplementation. Based on the findings of this research, without the addition of trace elements, the maximum biogas yield was achieved using 2 g/L graphene and 1 g/L petroleum coke. By supplementing the reactors with trace metals, the highest biogas production was observed at a 2 g/L concentration of both graphene and petroleum coke. Using petroleum coke and graphene, the average COD removal was 42.84 ± 1.23% and 42.80 ± 0.45%, respectively, while adding the same amounts of the aforementioned conductive materials in the presence of trace elements in the reactors resulted in a COD removal of 34.65 ± 0.43% and 34.05 ± 0.45%, respectively. In this study, the average VS removal of 35.17 ± 0.01% and 33.11 ± 0.02% was achieved without and with the addition of trace metals to the reactors, respectively. Furthermore, the modified Gompertz model indicated a 73.33% decrease in the lag phase of reactors supplemented with trace metals. A notable finding of this study is that petroleum coke is capable of enhancing the production of biogas from anaerobic digestion in the same way as other conductive materials. It should be noted that one of the main advantages of using petroleum coke over other conductive materials is that it is an undesirable by-product, which means that using this material will create a symbiotic relationship between industries. This may contribute to the circular economy by optimizing the use of natural resources and reusing waste products. Various microbial analyses should be performed in future studies to demonstrate the benefit of adding petroleum coke to the reactors in terms of direct interspecies electron transfer.