Translate this page into:
Response surface methodology for the optimization of the ultrasonic-assisted rhamnolipid treatment of oily sludge
⁎Corresponding author. 295998845@qq.com (Xuefei Hu)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Abstract
This study investigates the recovery of oily sludge using ultrasound-assisted rhamnolipids and uses oil recovery yield as an evaluation index. The Box–Behnken response surface method was employed to investigate the individual and interactive effects of four different operating factors: frequency, dosage, liquid–solid ratio, and stirring speed. The model optimization results showed that the order of degree of influence of these four factors is frequency > liquid–solid ratio > dosage > stirring speed. The mathematical model predicted the highest oil recovery yield as 92.27%, which was highly accurate (in the 95% confidence interval) as from the experimental results, the highest oil recovery yield was 89.95% under optimal reaction conditions (frequency = 25.58 kHz, dosage = 150.45 mg/L, liquid–solid ratio = 4.1:1 mL/g, and stirring speed = 407 rpm). Thus, the deviation from the prediction model was only 2.32%, indicating that this method provides a theoretical basis for the treatment of oily sludge and can be implemented for practical application in Huaidong in the Xinjiang Province.
Keywords
Rhamnolipid
Ultrasound
Response surface
Oily sludge
Biosurfactant
- CMC
-
critical micelle concentration
- CV
-
coefficient of variance
- IFT
-
interfacial tension
- PHC
-
petroleum hydrocarbon
- RH
-
rhamnolipid
- SARA
-
saturates, aromatics, resins, and asphaltene
- US
-
ultrasonic
- C.I.
-
confidence interval
Abbreviations
1 Introduction
With the rapid development of technology in the petroleum industry, large amounts of oily sludge will inevitably be produced during oil and gas field development, oil storage, transportation, and processing, and the amount of oily sludge produced will continue to increase with increasing oil production (Hu et al., 2013; Xin et al., 2018). If oily sludge is not treated or poorly treated, it will cause severe environmental pollution, and valuable petroleum resources will also be wasted (Xiao et al., 2019) considering that petroleum has a high oil content and recovery value (Yu et al., 2014; Hu et al., 2016). Therefore, recycling crude oil in a more environmentally friendly, economical, and efficient manner is necessary to treat oily sludge.
Common methods by which oily sludge is treated include centrifugation, biological treatment, solidification, pyrolysis, landfill disposal, and thermal cleaning (Yu et al., 2014; Hu et al., 2015; Zhao et al., 2015; Liang et al., 2017; Gao et al., 2018). Among these methods, thermal cleaning greatly shortens the processing time and increases the processing efficiency (Duan et al., 2018). However, the key to its success is the identification of cheap, efficient, and environmentally friendly surfactants (Liang et al., 2017; Duan et al., 2018). Although surfactants are widely used and their consumption is gradually increasing, some chemical surfactants are not biodegradable. Thus, they cause secondary pollution. Compared to traditional chemical surfactants, biosurfactants have excellent surface properties and extremely low biological toxicities. Additionally, they are environmentally friendly and 100% biodegradable (Liu et al., 2018; Zhao et al., 2019). In recent years, they have become a major research focus in the field of oily sludge treatment technology (Yan et al., 2012). Particularly, rhamnolipids (RHs) are among the most widely studied biosurfactants and are associated with the most mature application technologies (Liu et al., 2018). They can reduce oil–water interfacial tension and improve the efficiency of crude oil recovery from oily sludge (Long et al., 2013).
Ultrasound supplements thermal cleaning in the oil recovery process. Some studies suggest that combining the ultrasonic and thermal cleaning methods can reduce the processing cycle and improve processing efficiency (Jin et al., 2012; Sivagami et al., 2018; He et al., 2019). In recent years, single processing models for oil recovery have been gradually abandoned in favor of a joint processing model wherein multiple methods of oil recovery are combined (Hu et al., 2015; Johnson et al., 2018). Many researchers have studied the ultrasonic-assisted chemical surfactant process (Jin et al., 2012; Gao et al., 2015; Jing et al., 2016). However, there are limited studies on ultrasound-assisted biosurfactants. Ultrasonic + biological surfactant technology may be one of the most suitable techniques for industrial application. This is because it is suitable for many types of sludge, the treatment equipment and reagents investment is low, and there are several unique automation advantages.
The response surface methodology used in the study is the Box–Behnken response surface method, which is a type of statistical experiment design. The strongest combination of each factor level is determined by the quantitative relationship between test indices and various factors. Using this method, the continuous variable surface model is established through computer operation simulation, and the factors affecting the crude oil recovery process and their interaction are evaluated. The response surface method test group is relatively small, reducing the required human and material resources. Our study focused on using the response surface method to determine the best processing environment for the use of ultrasound-assisted biosurfactant RHs, analyze the relevant mechanisms, and establish a prediction model via the optimization of relevant parameters.
In this study, the optimal ranges for different parameters were selected on the basis of single-factor tests, and the Box–Behnken response surface method was used to analyze the effects of the ultrasonic-assisted RH (US + RH) treatment method on the oil recovery process. Additionally, the effects of the ultrasonic frequency, RH dosage, liquid-to-solid ratio, and stirring speed on the oil recovery process were explored, and the properties of the residual sludge were analyzed. The findings of this study provide a theoretical foundation for resource utilization and the safe processing of oily sludge.
2 Material and methods
2.1 Materials
2.1.1 Oily sludge
The oily sludge used in this study, characterized by a high viscosity, was obtained from the Huaidong Oil Production Plant in Xinjiang Province, China. The experimental sample sludge comprised a mix of primarily ground sludge and additional tank cleaning and other sludges (e.g., polyethylene containing sludge). The site produces approximately 5 × 104 t of oily sludge annually. Currently, the company pays high yearly outsourcing fees to other companies for the disposal of oily sludge.
2.1.2 Biological surfactants
The biological surfactants, RH, trehalolipid, saponin, and sophorolipid were obtained from Beijing Soleb Biotech Co., Ltd, China. RH is a biosurfactant produced by Pseudomonas or Burkholderia and is mainly obtained by fermentation with vegetable oil as the carbon source. Trehalolipid is synthesized by Candida, red coccus, and other Actinomycetes. Saponin is mainly distributed in higher plants on land (such as Bupleurum chinense), but also exists in small quantities in sea creatures such as starfish and sea cucumbers. In the chemical structure of saponin, the glycosides have different degrees of lipophilicity, and the sugar chain has high hydrophilicity, which synthesizes saponins into a type of surfactant. Furthermore, the aqueous solution can produce persistent soap-like foam after shaking. Sophorolipids are mainly produced by yeasts, such as Candida. Sophorolipids are one of the most promising biological surfactants among glycolipids.
2.2 Oil content analysis
The oil content was measured using the Chinese method HJ637-2018 (Duan et al., 2018; Ren et al., 2020). A 5 g aliquot of oil sludge was mixed with 5 g of anhydrous sodium sulfate. After adding 10 mL of CCl4 reagent, the sample was shaken to ensure even dispersal and then subjected to ultrasonic treatment for 10 min. Next, the sample was centrifuged at 800 rpm for 10 min to separate the extract. The process was repeated three times for the same sample, resulting in three extracts from the same 5 g of oil sludge. After combining the three retrievals, the extract was dried over anhydrous sodium sulfate, filtered through an ordinary funnel, and transferred to a 50 mL volumetric flask. The extract was measured with an infrared oil measuring instrument (Oil-480 Huaxia Science and Technology, China) to determine the oil content.
2.3 Oil recovery yield
The oil recovery yield was calculated using the following equation:
2.4 Oily sludge characterization
The oil content of the oily sludge was measured according to the Chinese HJ637-2018 method (Duan et al., 2018; Ren et al., 2020) using three parallel experiments. Its water content was measured using the distillation method, and its solids content was derived using the measured oil and water contents (Duan et al., 2018; Ren et al., 2020). The oil, water, and solid contents of the oily sludge were 41.39%, 8.57%, and 50.04%, respectively. Because of its high oil (41.39%) and solid (50.04%) contents, sludge has the potential for use as an energy source. The oil phase can be recycled, and the solid phase can be synthesized into building materials and subgrade bricks. The saturates, aromatics, resins, and asphaltene (SARA) fractions of the oil portion, which are the main hydrocarbon fractions in the oily sludge, were measured in accordance with the SHT0509-2010 standard (Liu et al., 2018; Ren et al., 2020). A scanning electron microscope (QUANTA SEM, Changhai Baihe Instrument Technology Co., Ltd., China) with a magnification of 800 × was used to compare the microstructure of the oily sludge before and after the treatment (Ren et al., 2020).
2.5 Single-factor experiments
As shown in Table 1, to perform the single-factor experiments, four main factors, including frequency, RH dosage, liquid-to-solid ratio, and stirring speed, were considered. During the experiments, when each factor was examined, the other three were maintained as constant. Primarily, 10 g of the oily sludge was placed in a beaker, a certain proportion of the biosurfactant solution was added (see Table 1), and the beaker was placed in an ultrasonic waterbath (Dgci-1200 model of Degar Electronics Co., Ltd, China). The dimensions of the ultrasonic waterbath are 40 × 30 × 35 cm, and it has a transducer at the bottom which controls the ultrasonic conditions; to perform stirring, an automatic agitator (refitted in the laboratory) was used. The schematic diagram of the experimental device is shown in Fig. 1. The process was maintained at a temperature of 45 °C for 30 min. After the treatment was completed, the oil content was measured with the help of an infrared oil meter (Oil-480 Huaxia Science and Technology, China).
Factors
Levels
Frequency (kHz)
10
20
30
40
50
Dosage (mg/L)
50
100
150
200
250
Liquid–solid ratio (mL/g)
1:1
3:1
5:1
7:1
9:1
Stirring speed (rpm)
100
200
300
400
500
Schematic diagram of experimental device.
2.6 Response surface test design
Based on the single-factor experiments, a numerical experiment (the Box–Behnken experiment) was conducted using Design-Expert v11.04 software, and the response surface method was used to optimize the conditions for the US + RH treatment of oily sludge (Maran et al., 2013; Mohammad et al., 2019). The factors and levels of the Box–Behnken experiment are presented in Table 2.
Factors
Levels
−1
0
1
Frequency (kHz)
20
30
40
Dosage (mg/L)
100
150
200
Liquid–solid ratio (mL/g)
3:1
5:1
7:1
Stirring speed (rpm)
300
400
500
2.7 Oil/water interfacial tension analysis
The interfacial tension (IFT) was measured using an interfacial tension meter (Sigma 701, Biolin Company, Finland). The biosurfactant concentration was 100 mg/L (Duan et al., 2018).
2.8 Petroleum hydrocarbon fraction analysis
The fractions denoted as F2, F3, and F4 each represented a group of petroleum hydrocarbons (PHCs) in the ranges of C10–C16, C16–C34, and C34–C50, respectively. The PHC contents of the oily sludge were measured using gas chromatography coupled with flame ionization detection (GC-FID, Varian 6800 N, Varian Technology China Co., Ltd., Shanghai, China).
3 Results and discussion
3.1 Screening of biosurfactants
IFT is an important parameter for the measurement of the surface activity of biosurfactants (Elakkiya et al., 2020). As shown in Fig. 2, at the same concentration (100 mg/L), the four different biosurfactant solution systems (RH, trehalolipid, sophorolipid, and saponin) showed different IFT values. At a concentration of 100 mg/L, cleaning temperature of 45 °C, stirring speed of 300 rpm, and liquid-to-solid ratio of 5:1, the oil recovery yields of the four biosurfactants were 65.71 ± 0.61%, 39.48 ± 0.59%, 47.55 ± 0.67%, and 31.07 ± 0.48%, respectively. Additionally, a comparison of the IFT values of these surfactants and their crude oil recovery yields showed that a decrease in the IFT value resulted in an increase in the oil recovery yield (Duan et al., 2018). The oil recovery yield corresponding to the RH was significantly higher than those corresponding to the other three biosurfactants; thus, the RHs were selected for subsequent experiments.
Oil recovery yield and IFT of different biosurfactants (IFT = interfacial tension).
3.2 Comparison of methods
As shown in Fig. 3, the US + RH treatment technology was a more effective oil recovery method than the individual use of US or RH. Using the US + RH treatment method, the resulting oil recovery yield was 79 ± 1.03%. However, the individual use of US or RH meant that the resulting oil recovery yields were only 8.13% and 65.71 ± 1.21%, respectively. This is because the ultrasonic treatment of oily sludge is primarily realized via its cavitation effect (Xu et al., 2009). The cavitation effect of ultrasound can change the internal structure of the sludge, decrease the size of the solid particles in the sludge, and intensify the cleaning effect of the RH molecules (Jin et al., 2012).
Comparison of the experimental results obtained using different methods.
3.3 Single-factor experiments
3.3.1 Influence of frequency
Under an RH dosage of 150 mg/L, a liquid-to-solid ratio of 5:1, and a stirring speed of 400 rpm, the effect of the ultrasonic frequency on oil recovery was investigated. As shown in Fig. 4(a), an increase in the ultrasonic frequency resulted in an increase in the oil recovery yield; at an ultrasonic frequency of 20 kHz, the oil recovery yield was up to 82.19 ± 1.01%. However, as the ultrasonic frequency further increased, the oil recovery yield began to decrease because the degree of cavitation is closely related to the ultrasonic frequency. If the ultrasonic frequency is too high, it becomes more difficult to produce cavitation bubbles. Thus, the cavitation effect is weakened, resulting in low oil recovery (Ju et al., 2012; Li et al., 2013; Hu et al., 2016). However, under low-frequency conditions, there is an increase in the cavitation speed, which intensifies the cleaning effect of the RHs (Ju et al., 2012). Many authoritative research reports (Hu et al., 2016; Gao et al., 2018) indicate that the ultrasonic power is 20–40 kHz, and cavitation bubbles with a radius of only a few microns are easily generated. The optimal frequency range of our experiments was 10–30 kHz. The bursting of these bubbles may enhance the analytical application of adsorption molecules (Xu et al., 2009); according to previous studies (Xu et al., 2009) cavitation bubbles are not easily generated when the ultrasonic power is lower than 20 kHz. There are no reports of the best treatment being achieved at 10–20 kHz. The larger the selected range, the lower the accuracy of the response surface method. In combination with exploratory ultrasonic experiments (the ultrasonic effect of frequency 10–20 kHz was inferior to that of 20–30 kHz), we used a frequency of 20–30 kHz. However, some research results show that the cavitation bubbles produced at 30–40 kHz are the most abundant and have the best effect on the oily sludge (Xu et al., 2009; Ju et al., 2012). To determine if the optimal range is 20–30 kHz or 30–40 kHz, we conducted additional experiments at 20–40 kHz based on experimental data and literature (Xu et al., 2009; Ju et al., 2012; Gao et al., 2018; Sathiyagnanam et al., 2018).
Single-factor experiment results.
3.3.2 Influence of dosage
The highest frequency used in the experiments was 20 kHz, and the other conditions were the same as stated in Section 3.3.1. As shown in Fig. 4(b), when the RH dosage ranged as 50–150 mg/L, the oil recovery yield increased from 64.18 ± 0.79% to 85.53 ± 0.87%. At a higher dosage of 150 mg/L, the crude oil recovery yield was the highest. However, a further increase in the dosage resulted in a decrease in the oil recovery yield. This is because when the excessive surfactant dosage reaches the critical micelle concentration (CMC) value, its interaction with sludge causes the solution to produce an oil-in-water emulsion, thereby lowering the cleaning effect (Yan et al., 2012; Elakkiya et al., 2020).
3.3.3 Influence of liquid–solid ratio
As shown in Fig. 4(c), as the liquid-to-solid ratio increased from 1:1 to 5:1, the oil recovery of the combined process increased from 43.19 ± 0.62 to 86.41 ± 0.89%. This is because at a low liquid–solid ratio, the mass transfer of the system is hindered by the high solid content, which is not conducive to oil production (Tian et al., 2019). Additionally, when the liquid-to-solid ratio is greater than 5:1, the oil recovery yield begins to decrease. This is because at higher liquid-to-solid ratios, a water-in-oil emulsion is easily formed under constant stirring at a fixed temperature, causing the RH molecules to be adsorbed on the surface of the solid particles; thus, the eluted oil molecules recombine with the solid particles, resulting in a decline in oil recovery (Diao et al., 2014). Therefore, the optimal liquid–solid ratio range was found to be 3:1–7:1.
3.3.4 Influence of stirring speed
As shown in Fig. 4(d), as the stirring speed increased from 100 to 400 rpm, the oil recovery performance of the combined process increased from 67.13 ± 0.59% to 86.57 ± 0.68%. Stirring can accelerate the sediment shedding from the oily sludge and enhance the collision between particles, which is more conducive to the oil substances from the oily sludge. An extremely low stirring speed does not allow for complete contact of RH and oily sludge, thus impacting the treatment effect. The oil recovery yield also begins to decrease when the stirring speed is greater than 400 rpm. This is because part of the dispersed oil formed emulsified oil, which was easily re-adsorbed by the solid particles, thereby lowering the cleaning effect. Therefore, the optimal range of the stirring speed was found to be 300–500 rpm.
3.4 Response surface analysis
3.4.1 Box–Behnken central combination experiments and analysis of variance
The response surface design and experimental results obtained are shown in Supplementary Table S1. Each factor was fitted to the following regression equation (Eq. (2) was generated by Design-Expert v11.04 based on 29 representative experiments):
where Y represents the response variable (coded values), and X1, X2, X3, and X4 represent the frequency (kHz), dosage (mg/L), liquid–solid ratio (mg/mL), and stirring speed (min), respectively.
Supplementary Table S1 shows the response surface experimental results. The significance level of the model was p < 0.0001, indicating that the model is effective and highly significant (Table 3). The p-value corresponding to the lack of fit was 0.1048, which is >0.05, indicating that the difference in the lack of fit is insignificant, and the multiple quadratic regression model fit the actual data well. The R2 and Radj2 values were 0.9863 and 0.9725, respectively, both of which are close to 1, indicating that the model fit the regression model well, and both the R2 and Radj2 were within a reasonable range. The coefficient of variance (CV) value was 2.08, indicating that the fitted model can be reproduced. Thus, the model could be used to effectively predict the data at the real experimental points and accurately describe the relationship between the factors and response values to determine the process conditions corresponding to the highest oil recovery yield. Among the factors, the impacts of X1 and X3 were more significant, and the order of impact was X1 > X3 > X2 > X4.
Source
Sum of squares
df
Mean square
F-Value
P-Value
Model
2050.29
14
146.45
71.83
< 0.0001
X1- Frequency
178.95
1
178.95
87.77
< 0.0001
X2- Dosage
12.42
1
12.42
6.09
0.0271
X3- Liquid–solid ratio
156.46
1
156.46
76.73
< 0.0001
X4- Time
3.14
1
3.14
1.54
0.2349
X1X2
14.03
1
14.03
6.88
0.0201
X1X3
2.67
1
2.67
1.31
0.2714
X1X4
4.88
1
4.88
2.40
0.1440
X2X3
9.21
1
9.21
4.52
0.0518
X2X4
35.82
1
35.82
17.57
0.0009
X3X4
12.78
1
12.78
6.27
0.0253
X12
107.02
1
107.02
52.49
< 0.0001
X22
1575.15
1
1575.15
772.53
< 0.0001
X32
92.59
1
92.59
45.41
< 0.0001
X42
188.93
1
188.93
92.66
< 0.0001
Residual
28.55
14
2.04
Lack of Fit
25.83
10
2.58
3.81
0.1048
Pure Error
2.71
4
0.6787
Cor Total
2078.84
28
R2 = 0.9863
RAdj2 = 0.9725
CV = 2.08
Fig. 5 shows a comparison chart of the predicted and actual values (predicted values were derived from the simulation of Design-Expert V11.04). This chart shows that the predicted and actual values were close to each other. Based on the results of the analysis of variance (Table 3), it could be concluded that the multivariate quadratic model designed by this response surface fitted the data well.
Comparison of the predicted and actual oil recovery fractions (predicted values were derived from the simulation of Design-Expert v11.04).
3.4.2 Response surface interaction analysis
Fig. 6a shows the influence of the interaction between ultrasonic frequency and RH dosage on oil recovery when the liquid–solid ratio and stirring speed were 5:1 and 400 rpm, respectively. As the ultrasonic frequency increased, the oil recovery yield first increased, and subsequently decreased as the ultrasonic frequency was further increased. This may be because the number of cavitation bubbles resulting from the cavitation effect increased with increasing ultrasonic power (Xu et al., 2009; Ju et al., 2012; Sivagami et al., 2018; He et al., 2019). Thus, the cavitation effect may have increased along with the increase in oil recovery yield. However, at high frequencies, it was difficult to generate cavitation bubbles (Xu et al., 2009); we found that the optimal frequency of our experiment was 20–30 kHz instead of 30–40 kHz. Thus, there was a decline in oil recovery. When the ultrasonic power was constant, increasing the RH dosage could result in an increase the oil recovery yield via the US + RH treatment method. However, excessive RH dosages may cause the viscosity of the oil sand and RH solution mixture to increase, a phenomenon that does not favor the separation of the oil molecules and the solid particles. Thus, the oil recovery yield decreases. The slope of the curved surface indicated that the impact of the ultrasonic frequency on the oil recovery yield was greater than that of the RH dosage, which is consistent with the model predictions.
Response surface graph for the effects of the (a: biosurfactant dosage, b: liquid–solid ratio) and ultrasound frequency on oil recovery.
Fig. 6b shows the effect of the interaction between the ultrasonic frequency and the liquid–solid ratio on oil recovery yield at an RH dosage of 150 mg/L and a stirring speed of 400 rpm. With the increase in the liquid–solid ratio and ultrasonic frequency, the number of cavitation bubbles in the mixture also might have increased (Hu et al., 2016). Thus, the contact area between the cavitation bubbles and the oil sands would also have increased. However, the rupture of the cavitation bubbles produced shock waves, which caused the solid particles to burst. Thus, the oil molecules were easily separated from the solid particles, resulting in an increase in the oil recovery yield. When the ultrasonic frequency remained unchanged and the liquid–solid ratio was continuously increased, the RH molecules were adsorbed on the surface of the solid particles during the constant temperature stirring, causing the desorbed oil molecules to be re-incorporated into the oil sands, thereby decreasing the oil recovery yield.
Fig. 7a shows the effect of the interaction between the ultrasonic frequency and the stirring speed on the oil recovery yield when the RH dosage and the liquid-to-solid ratio were 150 mg/L and 5:1, respectively. As the stirring speed and ultrasonic frequency increased, the number of cavitation bubbles also would have increased. Thus, there was an increase in the efficiency of the collision between the RH molecules and the oil molecules; hence, the oil recovery yield increased. Additionally, when the optimal ultrasonic frequency value was employed, increasing the stirring speed could improve the value of the mass transfer efficiency of the RH solution, causing an increased number of oil molecules to come into full contact with RHs, thereby causing their desorption. However, when the stirring speed was too high, part of the dispersed oil formed emulsified oil, which was easily re-adsorbed by the solid particles, thereby lowering the cleaning effect. When the stirring speed was within the optimal range, increasing the frequency could result in the production of more cavitation bubbles, thereby increasing the oil recovery yield. However, continuously increasing the frequency could easily exceed the cavitation valve that does not favor the generation of cavitation bubbles and results in a decrease in the oil recovery yield. The inclination of the curved surface suggested that the effect of time on the oil recovery yield was insignificant, a conclusion that is consistent with the model predictions.
Response surface graph for the effects of the (a: ultrasound frequency, b: liquid–solid ratio) and stirring speed on oil recovery.
Fig. 7b shows the interaction between the stirring speed and the liquid–solid ratio when the frequency and the RH dosage were 30 kHz and 150 mg/L, respectively. As the stirring speed and the liquid–solid ratio increased, a larger number of surfactant molecules could contact the oily sludge particles, resulting in an increase in the oil recovery yield.
Fig. 8a shows the interaction between the RH dosage and the liquid–solid ratio when the frequency and the stirring speed were 30 kHz and 150 rpm, respectively. As the RH dosage and liquid–solid ratio both increased, more RH molecules could come into full contact with the oily sludge. Additionally, when the liquid-to-solid ratio was within the optimal range, increasing the dosage could result in an increase in the concentration of the surfactant molecules and thus, the oil recovery yield. However, when the dosage was extremely high, the viscosity of the solution and the oily sludge mixture increased. Thus, the separation of the oil molecules was hindered. When the dosage was within the optimal range, increasing the liquid–solid ratio probably caused the surfactant molecules to come into full contact with the oil sands. However, when the liquid–solid ratio was extremely large, the surfactant molecules could be adsorbed onto the surface of the solid particles, thereby reducing the oil recovery yield.
Response surface graph for the effects of (a: liquid–solid ratio, b: stirring speed) and biosurfactant dosage on oil recovery.
Fig. 8b shows the interaction between the dosage and the stirring speed when the frequency and the liquid-to-solid ratio were 30 kHz and 5:1, respectively. As stirring speed and RH dosage both increased, a larger number of RH molecules encounter the oil sand, resulting in an increase in the oil recovery yield. When the dosage was maintained within the optimal range and the stirring speed was significantly increased, the desorbed oil molecules re-formed the emulsified oil, resulting in a decrease in the oil recovery yield. In contrast, when the stirring speed was maintained within the optimal range and the RH dosage was increased significantly, the viscosity of the solution increased resulting in a decrease in the oil recovery yield.
3.4.3 Parameter optimization and experimental verification
The optimal conditions obtained using the second-order regression model were as follows: frequency = 25.58 kHz, dosage = 150.45 mg/L, liquid–solid ratio = 4.1:1, and stirring speed = 407 rpm. The response value under these conditions was 92.27%. The 95% confidence interval (C.I.) obtained by the Design-Expert v11.04 software was ∼ 87.27–93.98% for oil recovery. Three sets of parallel experiments were conducted under these optimal conditions, and an average oil recovery yield of 89.94% was attained. Furthermore, we found that the results of the three sets of experiments were within the 95% C.I. The relative error between the measured predicted value and the actual results was 2.33%, which further proves the reliability of the model in the prediction and analysis of the performance of the US + RH oily sludge treatment method.
3.5 Characteristics of the residue
3.5.1 SARA analysis
The original or fresh oily sludge and the sludge residue obtained after treatment were analyzed by determining the SARA contents, and the removal efficiencies of the different components owing to the application of the combined process (US + RH) were investigated. As shown in Table 4, compared to the original oily sludge, the saturated and aromatic hydrocarbon contents in the sludge residue decreased by 7.66% and 4.12%, respectively, while its gum and asphaltene contents increased by 6.33% and 5.42%, respectively. These findings indicate that by using the US + RH process, it is easier to remove light components than heavy components.
Saturates (%)
Aromatics (%)
Resin (%)
Asphaltene (%)
Original sludge
69.19
20.94
6.42
3.45
Sludge residue
61.53
16.85
12.75
8.87
3.5.2 Analysis of PHC
As shown in Table 5., F2, F3, and F4 fractions of the oil recovered using the combined US + RH process were 16%, 65.3%, and 12.6%, respectively. During the storage of the oily sludge, most volatile PHCs [F1 fractions (C6–C10)] vaporized, leaving a relatively heavier fraction for recovery. The recovered oil contained more F3 and F4 than the fresh oil sample, which may be suitable for heavy oil production.
F2
F3
F4
Recovered oil (%)
16.2
65.3
12.6
Fresh oil (%)
53.5
42.7
4.5
3.5.3 Scanning electron microscopy analysis
As shown in Fig. 9, before treatment, the oily sludge had a continuous structure, and the oil cement showed a continuous phase. After treatment, it had a dispersed structure, which predominantly consisted of dispersed granular solid phases. This observation shows that the US + RH process strongly affects the oily sludge and demonstrates its effectiveness as a cleaning technology for the treatment of oily sludge and its applicability in the field of resource utilization.![Scanning electron microscopy images of: (a) Original sludge sample [800×], and (b) Sludge residue [800×].](/content/184/2021/14/3/img/10.1016_j.arabjc.2020.102971-fig9.png)
Scanning electron microscopy images of: (a) Original sludge sample [800×], and (b) Sludge residue [800×].
4 Conclusions
The results of this study can be summarized as follows:
-
(1)
Based on single-factor experiments, the optimal experimental ranges for each parameter were determined as follows: frequency = 20–30 kHz, dosage = 100–200 mg/L, liquid–solid ratio = 3:1–7:1, and stirring speed = 300–500 rpm.
-
(2)
The fitted second-order regression model was statistically significant and showed a high fitting degree; thus, it could be used to predict oil recovery from oily sludge using RHs. Based on the results of orthogonal experiments, the order of influence of the different parameters was as follows: frequency > liquid–solid ratio > dosage > stirring speed.
-
(3)
The optimal process conditions predicted using the model were as follows: frequency = 25.58 kHz, dosage = 150.45 mg/L, liquid–solid ratio = 4.1:1, and stirring speed = 407 rpm. The relative error between the predicted and experimental values was 2.32%, which further demonstrates the accuracy and feasibility of the quadratic polynomial prediction model.
-
(4)
The recovered oil could be used as heavy oil. The experimental data provides a basis for future pilot projects involving oil recovery.
-
(5)
This research may provide technical support for the treatment of oily sludge or oil-contaminated soils.
Acknowledgements
The authors are grateful to Changhan Wang for providing assistance with the experiments and to Min Li for the valuable discussions.
Funding
This work was supported by the National Natural Science Foundation of China (grant number 21767025) and the Alar Science and Technology Project Fund of the First Division of Xinjiang Production and Construction Corps (grant number 2018TF02).
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- Experiment on enhanced washing of oily sludge by anionic/nonionic mixed surfactant. Chem Ind Eng Prog. 2014;30:15-17.
- [Google Scholar]
- Treatment of Daqing oily sludge by thermochemical cleaning method. Coll Surf A: Physicochem Eng Asp. 2018;554:271-278.
- [Google Scholar]
- Swift production of rhamnolipid biosurfactant, biopolymer and synthesis of biosurfactant-wrapped silver nanoparticles and its enhanced oil recovery. Saudi J Biol Sci. 2020;27:1892.
- [Google Scholar]
- Influence of ultrasonic waves on the removal of different oil components from oily sludge. Environ Technol. 2015;36:1771-1775.
- [Google Scholar]
- Ultrasonic washing for oily sludge treatment in pilot scale. Ultrasonics. 2018;90:1-4.
- [Google Scholar]
- Effect of ultrasound on oil recovery from crude oil containing sludge. Environ Technol. 2019;40:1401-1407.
- [Google Scholar]
- Recent development in the treatment of oily sludge from petroleum industry: a review. J Hazard Mater. 2013;261:470-490.
- [Google Scholar]
- A combination of solvent extraction and freeze thaw for oil recovery from petroleum refinery wastewater treatment pond sludge. J Hazard Mater. 2015;283:832-840.
- [Google Scholar]
- Oil recovery from petroleum sludge through ultrasonic assisted solvent extraction. J. Environ. Sci. Health. A Tox./Hazard. Subst Environ Eng. 2016;51:921-929.
- [Google Scholar]
- Oil recovery from oil sludge through combined ultrasound and thermochemical cleaning treatment. Ind Eng Chem Res. 2012;51:9213-9217.
- [Google Scholar]
- Petroleum sludge treatment and disposal: A review. Environ Eng Res. 2018;24:191-201.
- [Google Scholar]
- Oil recovery from refinery oily sludge via ultrasound and freeze/thaw. J Hazard Mater. 2012;203:195-203.
- [Google Scholar]
- Ultrasonic desorption of petroleum hydrocarbons from crude oil contaminated soils. Environ Lett. 2013;48:1378-1389.
- [Google Scholar]
- Solid effect in chemical cleaning treatment of oily sludge. Coll Surf A Physicochem Eng Asp. 2017;522:38-42.
- [Google Scholar]
- Oil recovery from tank bottom sludge using rhamnolipids. J Petrol Sci Eng. 2018;170:14-20.
- [Google Scholar]
- Application of rhamnolipid as a novel biodemulsifier for destabilizing waste crude oil. Bioresour Technol. 2013;131:1-5.
- [Google Scholar]
- Box-Behnken design based statistical modeling for ultrasound-assisted extraction of corn silk polysaccharide. Carbohydr Polym. 2013;92:604-611.
- [Google Scholar]
- Box-Behnken design to optimize the synthesis of new cross-linked chitosan-glyoxal/TiO2 nanocomposite: methyl orange adsorption and mechanism studies. Int J Biol Macromol. 2019;129:98-109.
- [Google Scholar]
- Treatment mechanism of sludge containing highly viscous heavy oil using biosurfactant. Coll Surf Physicochem Eng Aspects. 2020;585:124117
- [Google Scholar]
- Sathiyagnanam, A. P.,Kumar, B. R.,Saravanan, S.,Rana, D.,Sethuramasamyraja], B., 2018. Prediction of emissions and performance of a diesel engine fueled with n-octanol/diesel blends using response surface methodology. Journal of Cleaner Production 184, 423–439.
- Treatment of petroleum oil spill sludge using the combined ultrasound and Fenton oxidation process. Ultrason Sonochem. 2018;51:340-349.
- [Google Scholar]
- Ionic liquid-enhanced solvent extraction for oil recovery from oily sludge. Energ Fuel. 2019;33:3429-3438.
- [Google Scholar]
- Study on oxidation process parameters and model optimization of oil sludge. Environ Sci Technol. 2019;42:195-202.
- [Google Scholar]
- Xin, H.L., Han, L.C., Yin, Z., Jing, L.Y., Zhi, Z.Z., Shan, S.S., 2018. Application of Biosurfactant in Oil Recovery from Oily Sludge. Chem. Bio. Eng. 35, 49–54+59.
- Oil recovery from refinery oily sludge using a rhamnolipid biosurfactant-producing Pseudomonas. Bioresour Technol. 2012;116:24-28.
- [Google Scholar]
- Research on hot washing treatment process of oily sludge. Sci Technol Chem Ind. 2014;37:962-966.
- [Google Scholar]
- Recovery of crude oil from oil sludge in Shengli Oilfield by water-assisted solvent extraction method. Oilfield Chem. 2015;32:132-136.
- [Google Scholar]
- Production of rhamnolipids with different proportions of mono-rhamnolipids using crude glycerol and a comparison of their application potential for oil recovery from oily sludge. RSC Adv. 2019;9:2885-2891.
- [Google Scholar]
Appendix A
Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.arabjc.2020.102971.
Appendix A
Supplementary data
The following are the Supplementary data to this article:
