Lei Ma ,Hongxia Lv ,Haonan Yu ,Lingtong Kong ,Rongyue Zhang ,Xiaoyan Guo ,Haibo Jin,*,Guangxiang He,*,Xiaoyan Liu

1 Beijing Key Laboratory of Fuels Cleaning and Advanced Catalytic Emission Reduction Technology,College of Chemical Engineering,Beijing Institute of Petrochemical Technology,Beijing 102617,China

2 Water Resource Office of Jining City,Shandong Province,Jining 272019,China

Keywords:Oil-absorption resin Response surface method Oil removal rate Waste water Absorption Polymerization

ABSTRACT A series of high oil-absorption resins with low cross-linking degree were synthesized by suspension polymerization using stearyl methacrylate(SMA),2-Ethylhexyl methacrylate (EHMA),and styrene(St) as monomers.Response surface methodology(RSM)with central composite design(CCD)was also applied to determine the optimal parameters that are mainly known to affect their synthesis.Thus,the effects of the monomer mass ratio(EHMA:SMA),the rigid monomer(St) dosage,the porous agent (acetone)dosage,and their pairwise interaction on the resin's oil-absorption capacity were analyzed,highlighting PSES-R2 as the resin with the optimum performance.The pure oil-absorption rates of PSES-R2 for gasoline,diesel,and kerosene were 11.19 g·g−1,16.25 g·g−1,and 14.84 g·g−1,respectively,while the oil removal rates from oily wastewater were 98.82%,65.11%,and 99.63%,respectively.

1.Introduction

With the rapid development of industrial society,the demand for oil resources and petroleum products is also growing.As a result,the amount of oily pollutants discharged into water environments due to uncontrolled emissions from factories and/or unexpected accidents,such as oil spill,is increasing.Although several measures have been taken to control this pollution,including the selection of oil skimmers for the recovery of thick oil layers,the incineration of oil layers in local seas within a controlled range to reduce oil accumulation,and the selection of oil boom to prevent oil spread,there is still a small amount of petroleum products that cannot be dissolved,posing a potential threat to the environment,as they can give a bad smell to fish,impair the quality of water sources,and inhibit the growth of aquatic organisms.

Among the available methods,absorption is the most effective for removing water-soluble oil products from contaminated water environments.High oil-absorption resins are synthetic polymer materials,which generally exists in the form of a single component homopolymer or a multicomponent copolymer.It has been proved that under optimized polymerization conditions,this type of resin can efficiently absorb oil products.Thus,several studies have focused on improving the existing polymerization process by exploring the effect of various parameters such as polymerization monomers,cross-linking agents,dispersants,and initiators[1].Besides,the internal structure and oilabsorption performance of acrylic resins have been improved during the polymerization process[2].

It should also be noted that the pairwise interaction among the single factors may also significantly affect the polymerization process.Hence,in order to determine the appropriate parameters for the synthesis of high oil-absorption resins,the response surface methodology(RSM) has been applied in previous studies [2].In general,RSM is used to evaluate the functional relationships between a response variable and a set of design variables,which can be further used to determine the optimal combination and levels of the response variables[3].Thus,for experiments affected by multiple factors,RSM is an ideal optimization method.In addition,a multiple quadratic regression equation is used to fit the functional relationship between the factors and response values,yielding the optimal process parameters.Random examples of the RSM application include the determination of the optimal preparation conditions of a novel adsorbent for lead ions[4],the optimization of the photodegradation conditions of azo dyes on TiO2catalyst[5],and the optimization of the deposition conditions for the synthesis of a Cu2ZnSnS4thin film.

Ji et al.synthesized a copolymer resin using butyl methacrylate(BMA)and methyl methacrylate(MMA)for the first time by suspended emulsion polymerization.The absorption capacity of the resin for methylbenzene was about 17 g·g−1,which was better than that achieved by the traditional suspension polymerization or the emulsion polymerization process[6].Sun et al.employed three separate monomers,i.e.,long-chain alkyl lauryl methacrylate(LMA),hexadecyl methacrylate(HMA),and methyl stearate(SMA),to prepare three types of high oil-absorption resins with large pore volume by a high internal phase emulsion polymerization process.Their study indicated that the porous structure of resin considerably favored the contact between the oil and the resin,thus improving the oil-absorption capacity of the synthesized resins.Specifically,the absorption rates of the materials obtained under the optimized process were 24 g·g−1,32 g·g−1,and 34 g·g−1for LMA,HMA,and SMA,respectively [7].Furthermore,Wang et al.grafted the lipophilic monomer BMA on the surface of a plant fiber using the initiator benzoyl peroxide (BPO),resulting in a resin with significantly higher absorption rates for various organic solvents(33–66 g·g−1)[8].

In our previous work,we used RSM to optimize the preparation conditions of a high oil-absorption resin (PSES-R1).In particular,the amounts of the initiator,cross-linker,and dispersant were optimized[2],resulting in gasoline,diesel,and kerosene absorption rates of 10.14 g·g−1,10.73 g·g−1,and 11.69 g·g−1,respectively.Moreover,the removal rates of the oily water samples were 98.66%,63.24%,and 99.74%,respectively.However,some factors,critical for the synthesis of high oil-absorption resins,such as the amount of the porous agent[9],were not considered in that study.Therefore,in this work,the effect of the monomer mass ratio of 2-Ethylhexyl methacrylate(EHMA)and stearyl methacrylate (SMA),the amount of styrene (St) as the rigid monomer,and the amount of acetone as the porous agent(acetone)on the oil absorption was evaluated and discussed in detail using RSM.

2.Materials and Methods

2.1.Materials

SMA(Shanghai Dibo Chemical Technology Co.Ltd.,Shanghai,China),EHMA(Shanghai Maclean Biochemical Technology Co.,Ltd.,Shanghai,China),and St(Shanghai Maclean Biochemical Technology Co.,Shanghai,China)were washed three times with NaOH and neutralized with deionized water.Then,vacuum distillation was applied only for St.All the samples were finally dried and stored for subsequent experiments.The initiator,2,2′-azobisisobutyronitrile(AIBN,Shanghai Maclean Biochemical Technology Co.,Shanghai,China),was recrystallized from ethanol (Beijing Chemical Plant,Beijing,China).Divinylbenzene (DVB,Saan Chemical Technology (Shanghai) Co.,Shanghai,China),which was used as the cross-linking agent,was washed three times with 10%alkali,neutralized with deionized water,and distilled under negative pressure.Finally,it was purified by column chromatography.Polyvinyl alcohol(PVA,Tianjin Guangfu Fine Chemical Research Institute,Tianjin,China)and acetone(Beijing Chemical Plant,Beijing,China)were directly used in the experiments without special treatment.All reagents,except DVB(industrial reagent grade),were of analytical grade.

2.2.Reaction mechanism and synthesis procedure

The highly oil-absorption resin was synthesized using the suspension polymerization process.The first step of the reaction mechanism was the chain initiation process,where the initiator(AIBN),dissolved in the monomer,decomposed to generate primary free radicals,which reacted with the monomers EHMA and SMA to form the monomer free radicals by an addition reaction.The chain growth process followed,where the EHMA and SMA free radicals cleaved the π bonds of the DVB ethylenic molecules,forming new free radical species.As the chain structure developed,heat was also released in the system.After some time,the free radicals reacted with each other under the effect of the cross-linking agent and gradually weaved a network structure.At this step,the porous agent(acetone),which did not participate in the reaction,was embedded in the network structure,followed by the chain termination step,where the unstable free radicals of the reaction system interacted and terminated the process.In this experiment,the chain termination method was induced by the increasing temperature of the reaction system.

Fig.1.Experimental setup used for the synthesis of the high oil-absorption resin.

The experimental setup used in this study is shown in Fig.1.First,100 ml deionized water and 3 g PVA were added to a four-necked flask and stirred at low speed for 12 h.Then,the temperature of the water bath was raised to 85°C to dissolve the suspended PVA.Afterward,the system was cooled to<40°C.The prepared mixture,which consisted of one of the three monomers each time,the initiator,and the cross-linking agent,was added to the flask through a constant pressure funnel at 1 drip·s−1upon stirring at 500 r·min−1.After the dripping process was completed,the temperature was kept at 40 °C for 30 min,and then increased to 70°C for 2 h,ra ised to 85°C for 2 h,kept at 85°C for three hours,finally to 90°C for 1 h.The generated products were then filtered,washed 3–5 times with 50°C deionized water,and dried in an oven at 80°C for 12 h.After drying,the high oil-absorption resin was obtained and used for the corresponding experiments and characterizations.

Furthermore,the central composite design(CCD)model was used to explore the second-order multi-factor interaction,allowing the acquisition of the representative experimental results with fewer experiments.

2.3.Oil-absorption tests

2.3.1.Pure oil

The absorption capacity of the prepared resin for pure oil was evaluated according to the ASTM F726-12 guidelines[10].The oil-absorption(OA)rate was calculated as follows:

where m1(g)is the weight of the resin sample before the OA test,m2(g)is the weight of the resin sample and the filter paper bag,and m3(g)is the weight of the resin and the filter paper bag after 24 h absorption.Based on this method,the gasoline(GAR),diesel(DAR),and kerosene(KAR)(absorption rates were also measured.

2.3.2.Oily water samples

A 100 ml oily water sample was divided into two 50 ml samples(water A and water B).Water A was added to a conical flask containing0.1 g of the resin and stirred for 24 h.Filtration followed and the resin sample with absorbed oil was obtained,while the remaining water A solution was extracted with n-hexane to calculate the oil concentration(c2).Instead,water B was directly extracted with n-hexane to estimate the initial oil concentration(c1).The oil removal efficiency η was calculated using the following formula:2.3.3.Oil retention test

The resins obtained after the oil absorption were centrifuged at 3000 r·min−1for 5 min and weighted.The oil retention was calculated by the following formula:

where m3and m4(g)represent the weight of the resin before and after the centrifugal process,respectively.

2.4.Experimental design

Using the five levels of the three factors,the actual values could be converted to the externally encoded values.The corresponding principles are shown in Table 1.Moreover,based on the experimental schemeoutlined in Table 2,the resin was synthesized and subjected to the OA test.The absorption results obtained for the three different oils are also listed in Table 2.

Table 1Correspondence table of actual and coded values

Table 2Experimental CCD table

Table 3Variance analysis table

3.Results and Discussion

3.1.Statistical analysis and model optimization

The variance analysis was performed by statistically analyzing the experimental data of Tables 2,3,while the significance test of the model simulated by the experimental results was conducted before analyzing the effect of each factor.The F values of the model for the three types of OA rate were<3,indicating that the noise variable did not significantly affect the model simulation process and that the model's significance was excellent.However,the lack-of-fit values in the three models were relatively high,suggesting that the model design was easily affected by the pure error.Therefore,we concluded that the model was less significant in the presence of all factors (Table S1-S3 in Supporting information).

Hence,the model was streamlined based on the p>F-value and the variables with small influence and great error were omitted,resulting in the best fitting result.According to Table 3,the interaction between the monomer ratio and the porous agent dosage on the absorption rates for the three oil types was not obvious.Thus,the interaction term was not included in the fitted mathematical model.

Fig.2.Residual normal and random distribution maps of the mathematical model.a,b,and c represent the mathematical models of gasoline,diesel,and kerosene,respectively,while the subscripts 1 and 2 represent the residual normal and random distribution maps,respectively.

The significance of the model was further verified by the residual normal and random distribution maps(Fig.2).It is clear from the staggered normal distribution map that most of the residual points were on the fitted straight line,while some individual points deviated within the allowable error range.Moreover,the analysis of the random distribution of residuals showed that the residual values of the mathematical models of the three oils were evenly distributed on both sides of the horizontal line and that the numbers were roughly equal.Therefore,it was inferred that the significance of the mathematical model of fit was excellent.In addition,the fitting equations could be expressed as follows:

3.2.Effect of single factors on the oil-absorption capacity of resin

3.2.1.Effect of the monomer ratio on resin oil absorption

According to Fig.3,the effect of the monomer ratio on the absorption of the three oils on the resin was limited.In particular,as the m(EHMA)∶m(SMA)ratio increased,the gasoline absorption rate at first increased until a peak at a ratio of 0.5 was reached and then decreased,indicating that the optimal absorption effect of gasoline could be achieved with respect to the m(EHMA)∶m(SMA)ratio.In contrast,the absorption rate of diesel slowly decreased as the EHMA dosage increased.Furthermore,the monomer ratio only slightly affected the absorption of kerosene[11].

Fig.3.Effect of the monomer ratio and the St and acetone content on the oil-absorption rate.

3.2.2.Effect of the rigid monomer(St)amount on resin oil absorption

Given that St significantly affected the resin structure during synthesis,a more pronounced effect was expected on the OA rate with increasing St content.Indeed,as shown in Fig.3,increasing the St amount increased the gasoline absorption in the given value range,whereas the absorption rate of diesel firstly increased and then decreased slowly with increasing St dosage.An increasing tendency was also observed for the kerosene absorption rate at relatively low St content.However,when the St amount was more than 2.5 g,the rate remained almost stable[10].

3.2.3.Effect of the porous agent(acetone)amount on resin oil absorption

The amount of acetone,which was used as the porous agent in the prepared resin,significantly affected the OA capacity of the resin for gasoline and diesel (Fig.3).More specifically,the amount of acetone slightly affected the absorption of gasoline at low dosages conditions,but when the dosage exceeded 3.5 g,the absorption of gasoline was considerably improved.In contrast,the absorption rate of diesel firstly increased and then decreased with increasing acetone dosage,but the optimum acetone amount of 3.5 g could be acquired within the given value range.Nevertheless,the effect of the porous agent on the kerosene absorption was very limited[12],as the rate increased slightly with increasing acetone dosage and stabilized rapidly.

3.3.Effect of multi-factor interactions on the resin oil-absorption capacity

3.3.1.Effect of pairwise interaction of A,B,and C on gasoline absorption efficiency

Based on the two-dimensional(2D)contour map(Fig.4),there was a significant pairwise interaction between the three factors A,B,and C that affected the absorption efficiency of the resin for gasoline.Moreover,an analysis of the three-dimensional(3D)response surface map indicated a significant synergy effect between factors A and B,while the optimal applicable value point was determined in the given value range [3].However,the response surfaces resulting from the synergy between factors B and C and factors A and C were concave,indicating their antagonistic relationship.In general,the higher the bending degree of the response surface,the more effective the interaction between two factors.Consequently,the highest interaction was observed between the monomer ratio and the acetone content (factors A and B).The lowest value of the fitted model could also be determined by the concave response surface,thus providing a reference range for the subsequent optimization process.

Fig.4.2D contour maps(left)and 3D response surface diagrams(right)of the synergistic effect of factors A,B,and C on the gasoline oil-absorption rate.

3.3.2.Effect of pairwise interaction of A,B,and C on diesel absorption efficiency

The curves in the 2D contour maps of factors A and C and factors B and C in Fig.5 indicated their synergistic effect on the diesel absorption rate and confirmed the use of an effective range of values.In addition,the pairwise interaction of the three factors during the absorption of diesel oil was obvious from the 3D response surface diagram,where the three factors were mutually reinforced.The interaction between factors B and C was the most representative,while the best dosage could be obtained by the statistical analysis of the three investigated parameters.

3.3.3.Effect of pairwise interaction of A,B,and C on kerosene absorption efficiency

In the case of kerosene,the synergistic interaction between factors A,B,and C considerably varied compared to that of gasoline and diesel (Fig.6).The curvature of the contour lines obtained from the interaction between factors A and C and factors B and C was very low and the response surface maps were also almost plane,indicating that the interactions were relatively weak during the kerosene absorption.However,the bending and upward convex surface observed in the response surface diagram of A and B(Fig.6) indicated a more significant interaction between these factors.

Based on the above experimental results and the factor effect analysis,the optimal combination of the three factors for the preparation of a highly OA resin was identified by simulating and optimizing the mathematical model and consisted of a monomer mass ratio(EHMA:SMA)of 1:5(total mass 10.0 g),a St dosage of 3.93 g,and an acetone dosage of 5.0 g.The resin obtained according to the optimal formula was indicated as PSES-R2.

Fig.5.2D contour maps(left)and 3D response surface diagrams(right)of the synergistic effect of factors A,B,and C on the diesel oil-absorption rate.

3.4.Characterization of PSES-R2 and oil-absorption performance test

3.4.1.Characterization of PSES-R2

Fourier transform infrared (FTIR) spectroscopy was used to characterize the optimized resin PSES-R2.In particular,the peak at 2920 cm−1in the IR spectrum of Fig.7(a)was assigned to the stretching vibration of the C—H bond of the methyl group.Moreover,the characteristic absorption peaks of the benzene ring(698 cm−1)[12]and the ester group(1720 cm−1)could be clearly observed,while no absorption peak of a double bond connected to the fatty chain was detected,indicating that the two monomers were well combined provided that the ester group was not destroyed.Based on these data,the skeleton structure of the resin was suggested,as illustrated in Fig.8.

Fig.6.2D contour maps(left)and 3D response surface diagrams(right)of the synergistic effect of factors A,B,and C on the kerosene oil-absorption rate.

Furthermore,as shown in Fig.7(b),the thermogravimetric analysis(TGA)curve of PSES-R2was stable at temperatures lower than 231°C.However,when the temperature was higher than the critical temperature,the mass of the resin started to decrease,indicating its decomposition by heat.The mass decrease continued until the resin was completely decomposed at 434°C.In addition,based on the derivative thermogravimetric(DTG)curve,the decomposition rate of the resin first increased and then decreased,while the highest decomposition rate was achieved at 363°C.Therefore,according to the information obtained from the two curves,the resin not only optimized the OA performance,but was also thermally stable below 231°C,which makes it a suitable material for conventional oil spill accidents or the dissolution of oil products under ambient temperature conditions.

The morphology of samples Nos.1,4,and 19,and the optimized resin PSES-R2in the experimental group were monitored by scanning electron microscopy (SEM) (Fig.9).The overall morphology of the four resin samples was observed at a lower resolution ratio(200 μm),indicating that samples No.1 and 19 had a spherical shape.Instead,sample 4 had a poorer OA effect and based on its appearance(Fig.9),its structural toughness was poor and easily destroyable,while internal collapse may also occur.By magnifying the SEM pictures by 10,000 times(1–3 μm),a large number of small spherical structures could be observed on the surface of PSES-R2.Moreover,the wrinkles on its surface were clear,which would also increase the contact area between the material and the oil,thus improving the OA effect of the resin.In addition,the samples No.1 and 19,which had a better OA effect than No.4,were full of gullies,while No.4 displayed a smooth surface and smaller pore structures,probably due to its poor absorption capacity.

Fig.7.(a)FTIR spectrum and(b)thermogravimetric and weight loss rate curves of PSES-R2.

Fig.8.Suggested skeleton structure of PSES-R2.

The contact angles of samples No.1 and 4 and the optimized resin PSES-R2were also measured.As shown in Fig.10,all the water contact angles were greater than 90°,suggesting that the three materials were all hydrophobic.However,small differences were observed among the estimated values.Namely,the water contact angle of PSES-R2was the highest(118.55°),suggesting that the optimized resin has excellent hydrophobic properties in contrast to the previously synthesized samples and the application potential in the treatment of oily wastewater.

The specific surface area of the optimal resin was measured using the Brunauer–Emmett–Teller(BET)method with N2adsorption and desorption at −195.8°C(Fig.11).The N2adsorption–desorption curves of PSES-R2were typical type-IV isotherms[13],while one distinct hysteresis loop was observed at a relatively higher pressure than that expected for the sample,indicating the existence of a mesoporous structure[14,15].Moreover,the pore size distribution of PSES-R2was determined using the Barrett–Joyner–Halenda(BJH)method,assuming a cylindrical pore model.As depicted in the inset of Fig.11,the resin had a two-stage pore structure and exhibited multi-peak distributions with pore sizes(diameter)centered at 1.0,1.3,3.5,4.0,and 5.5 nm.The significant distribution of the pore size mainly in the region of 1.0–5.5 nm indicated that the resin at this stage was a microporous or mesoporous material.

3.4.2.Oil-absorption effect of PSES-R2

The PSES-R2resin absorbed the oil molecules through its pore structure and the lipophilic long-chain alkyl groups[16].During adsorption,the resin started to expand due to its swelling properties,and the pore structure decreased.As a result,the absorbed oil molecules were stored into the resin.This microscopic adsorption process(Fig.12)could also be mathematically presented after determining the optimal synthetic conditions,using absorption kinetics and isotherms[17].

Fig.9.SEM images of samples Nos.1,4,and 19,and PSES-R2 under low and high resolution.

Fig.10.Contact water angles of samples Nos.1,4,and PSES-R2.

Fig.11.N2 adsorption–desorption isotherms and the corresponding pore size distribution curve(inset)of PSES-R2.

The efficiency of PSES-R2for the absorption of the three oil types was also examined(Table 4).Compared to the results of RSM(Table 2),both the diesel and kerosene absorption on PSES-R2have been apparently improved.More specifically,the diesel absorption rate on PSES-R2increased to 16.25 g·g−1,which was about a 33%increase than that on sample No.1.Similarly,the kerosene absorption rate(14.84 g·g−1)on PSES-R2increased by about 25% compared to that on sample No.1.The increase in the OA rates further demonstrated the excellent performance of the response design.The change in the absorption rate of thethree oil types over time shown in Fig.12 indicated that the saturated absorption time of gasoline,diesel,and kerosene was 8,10,and 9 h,respectively.Moreover,as the saturated oil-absorption amount increased,the saturated OA time also increased.

Table 4Results of the oil-absorption test on PSES-R2

The values of the oil retention rate on PSES-R2(Table 4)indicated the significant improvement of the retention rate of diesel and kerosene through the optimization process[1].Especially for diesel,a retention rate of 98.34%.was achieved.Therefore,the resin structure not only affected the OA process,but also played an important role in the retention of the oils after their absorption.

Furthermore,as observed in Table 4,the performance of PSES-R2in removing oils from water samples was excellent for gasoline(98.66%)and kerosene(99.74%)water samples,whereas its efficiency in removing diesel from water samples was low(63.24%).This significant difference could be attributed to the weak selectivity of the resin for diesel in water,which did not allow the complete filtration and storage of the water surrounded diesel molecules[18].

4.Conclusions

Fig.12.Adsorption curves of PSES-R1 and PSES-R2 for gasoline,diesel,and kerosene oils.

In this study,in order to improve the OA rate of a corresponding resin,the effect of the monomer mass ratio,the rigid monomer amount,and the porous agent amount on its OA performance was investigated.Through RSM,the influence of each factor and their synergistic interactions on the oil absorption was evaluated.The statistical analysis and the simplification of the mathematical model revealed an optimized formula,which was used for the synthesis of the optimal resin PSES-R2and involved a monomer mass ratio(EHMA:SMA)of 1:5,a St content of 39.3 wt%,and an acetone content of 50 wt%.The absorption rates of pure diesel and kerosene oils on PSES-R2were significantly improved by 33%and 25%compared to those of sample No.1.The improvement of the resin's OA efficiency also demonstrated the excellent performance of the response design.Furthermore,two samples(No.1 and 4)and the optimized resin PSES-R2were characterized by SEM,TGA,and FTIR,while their water contact angle was also estimated.The results indicated that PSES-R2had a more regular spherical morphology and higher hydrophobicity than the other two samples.However,the oil–water selectivity and the thermodynamic properties of the sample were slightly weaker than those of PSES-R1.As a result,RSM is one effective method for the fabrication of oil-absorption resin.It would be possible to apply the optimized sample(PSES-R2)into the treatment of oily water.

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.

Acknowledgements

This work was funded by Project of Construction of Innovative Teams and Teacher Career Development for Universities and Colleges under Beijing Municipality (IDHT20180508),Scientific Research Common Program of Beijing Municipal Commission of Education (KM202010017006),the Talents Project of Beijing Organization Department (2018000020124G091),the Strategic Priority Research Program of the Chinese Academy of Sciences(XDA21021101) and the National Key Research and Development Program of China(2019YFA0705803).

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2020.07.032.