Jiahao Cui,Shejiang Liu,Hua Xue,Xianqin Wang,Ziquan Hao,Rui Liu,Wei Shang,Dan Zhao,Hui Ding,3,

1 School of Environmental Science and Engineering,Tianjin University,Tianjin 300072,China

2 School of Chemical Engineering and Technology,Tianjin University,Tianjin 300072,China

3 Tianjin Key Laboratory of Indoor Air Environmental Quality Control,School of Environmental Science and Engineering,Tianjin University,Tianjin 300072,China

Keywords:Catalytic ozonation Normal temperature Activated carbon fiber Ethyl acetate Noble metal catalysts

ABSTRACT Catalytic treatments of VOCs at normal temperature can greatly reduce the cost and temperature of processing,and improve the safety factor in line with the requirements of green chemistry.Activated carbon fiber(ACF)was pretreated with 10%H2SO4 by single factor optimization to increase specific surface area and pore volume obviously.The catalytic ozonation performance of ACF loaded with Au,Ag,Pt and Pd noble metals on ethyl acetate was investigated and Pd/ACF was selected as the optimal catalyst which had certain stability.Pd is uniformly distributed on the surface of ACF,and Palladium mainly exists in the form of Pd0 with a amount of Pd+2.The specific surface area of the catalysts gradually decreases as the loading increases.The activation energy of ethyl acetate calculated by Arrhenius equation is 113 kJ﹒mol-1.With 1% Pd loading and the concentration ratio of ozone to ethyl acetate is 3:1,catalytic ozonation performance is maximized and the conversion rate of ethyl acetate reached to 60% in 30–50 °C at 15,000–30,000 h-1.

1.Introduction

VOCs are a general definition for organic compounds with a saturated vapor pressure of more than 133.3 Pa at normal temperature and a boiling point of 50–260 °C at atmospheric pressure[1,2].VOCs mainly come from decoration materials,adhesives,building materials,chemicals,petrochemicals,printing industries[3–7].Ethyl acetate is a representative pollutant in the automobile manufacturing,pharmaceutical,electronics manufacturing,coatings and inks manufacturing and plastics manufacturing industries,and it is also the main pollutant of composite film dry compound process in the packaging and printing industry [8,9].For the environment,VOCs can form ozone through photochemical reactions,producing ozone pollution.They can also be converted into secondary organic aerosol and further form PM2.5[10–14].For humans,VOCs can harm the human body and affect the health of human beings.Not only do they have stimulating effects on organs such as human vision,smell and breathing,but also damage the internal organs such as heart,lung and liver.VOCs can even cause acute and chronic poisoning,which can also cause cancer and mutagenicity [15–19].

Commonly used treatment technologies for VOCs at home and abroad are divided into physically based recycling technologies and chemically based elimination technologies.Recycling technologies such as adsorption,condensation,and membrane separation cannot remove VOCs effectively,and the elimination technologies have various shortcomings.For example,combustion technology has high combustion temperature and high energy consumption,which causes secondary pollution to the environment.Plasma technology has high requirements for power supply and reactor,and is easy to produce harmful by-products such as NOxand CO.The catalytic efficiency of photocatalytic degradation technology is not high,and it is only suitable for low concentration organic waste gas [20–23].Since the degradation reaction of ethyl acetate is exothermic,according to the second law of thermodynamics,the Gibbs free energy is negative at normal temperature,and the reaction can proceed spontaneously.Without the help of the outside conditions,a suitable catalyst can be prepared to reduce the energy barrier of the reaction and accelerate the degradation of VOCs.In this way,catalytic treatment of VOCs at normal temperature can greatly reduce the temperature of processing and energy consumption and equipment loss,improving the safety factor.The operation process is much simpler than the traditional process,in line with the requirements of green chemistry.

Activated carbon fiber (ACF) which is made from fiber is called the king of new materials and is widely used in water purification,air purification,aviation,military,nuclear industry,food and other industries due to its high specific surface area,fast adsorption speed,low density and light mass [24].In this work,we found the appropriate type of catalyst for degrading ethyl acetate from the ACF with different active metals loaded,and studied the effects of preparation conditions on the catalytic performance,such as acid treatment,loading and concentration of sodium borohydride.The reaction temperature,space velocity,the concentration ratio of ozone to ethyl acetate were systematically investigated and optimized.The morphology,structure and valence state of the catalyst were characterized by Brunauer–Emmett–Teller (BET),X-ray diffraction (XRD),scanning electron microscopy (SEM),transmission electron microscopy (TEM),Energy dispersive spectroscopy(EDS),X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR).

2.Materials and Methods

2.1.Materials

H2PtCl6﹒6H2O (AR),AuCl3﹒HCl﹒4H2O,PdSO4﹒2H2O (Pd,51.2–53.9%)were obtained from Chemart Chemical Ltd.(Tianjin,China).AgNO3(AR),C4H8O2(AR),C2H5O(AR)were purchased from Yuanli Chemical Ltd.(Tianjin,China).NaBH4was obtained from Saan Chemical Technology Ltd.(Tianjin,China).N2(99.99%) were purchased from Tianjin Liufang Industrial Gas Distribution Company(Tianjin,China).All of the chemicals were used as purchased without further purification.Solvents used in this study were strictly redistilled.

2.2.Synthesis of catalyst

First,a 2–3 mm thick activated carbon fiber(ACF)was cut into a 2 × 15 cm rectangle.Then,5%,10%,and 15% H2SO4solutions containing ACF were placed in a 60°C water bath for 2 h.The ACF was then washed with distilled water until neutral and naturally dried.A certain amount of modified ACF was placed in the HAuCl4,AgNO3,H2PtCl6,and PdSO4immersion liquids in the manner of all addition.Then,the impregnated ACF was allowed to stand at room temperature for 24 h after the mixture was gently shaken until the immersion liquid was completely absorbed.Add 0.5 mol﹒L-1sodium borohydride solution and stir vigorously until hydrogen was no longer produced in the solution.The aqueous phase containing the black suspension was poured off and the ACF was washed several times with a certain amount of deionized water.Then the washed catalyst was baked in a vacuum oven at 80 °C for 4 h to obtain a finished catalyst.

2.3.Characterization

The morphology of ACF was characterized by SEM using Hitachi S4800 and TEM using JEOL 2100F.The XRD pattern was recorded on a Bruker D8 Advance powder X-ray diffractometer using the CuKα radiation (λ=0.15418 nm).The FTIR of the sample was collected using a Bruker Thermo Nicolet 6700 IR spectrometer(4 cm-1) in KBr media.XPS was performed with a Thermo Fisher 250XI electron spectrometer with AlKα radiation,and the binding energies were referenced to the C 1s peak at 284.8 eV.The surface area and pore volume of catalysts were calculated from the BET method based on the nitrogen adsorption–desorption isotherm measured at 200 °C on a TRISTAR II3020.

2.4.Activity measurements

A continuous flow of nitrogen gas having a flow rate of 30 ml﹒min-1through a mass flow controller was introduced into an ethyl acetate bubbling bottle placed in an ice water bath to produce a certain concentration of ethyl acetate.The other nitrogen was mixed as a balance gas with ethyl acetate.Ozone was mixed with ethyl acetate and nitrogen in a mixing flask,and then introduced into a reactor to carry out a reaction of catalyzing ethyl acetate at normal temperature.The catalyst loading was 0.2 g,and the ethyl acetate content in the injection gas was 350 mg﹒m-3.The ethyl acetate concentration of the inlet and exhaust ports was measured online by gas chromatography(GC-6980A,China)to calculate the degradation efficiency of ethyl acetate Fig.1.

The ethyl acetate conversion (QE) was calculated using Eq.(1):

Fig.1.Catalyst performance evaluation device diagram.1.Nitrogen steel cylinder,2.Hydrogen generator,3.Air generator,4.Mass flowmeter,5.Check valve,6.Rotameter,7.Ozone generator,8.Purging bottle,9.Icewater bath,10.Mixing bottle,11.Water bath pot,12.U-shaped tube,13.Catalyst,14.Gas chromatograph,15.Tee valve,16.chromatographic workstation,17.Ozone detector.

where Coutand Cin(mg﹒m-3) are the ethyl acetate concentration of the inlet and exhaust ports measured online by gas chromatography,respectively.

2.5.Kinetics

The concentrations of ethyl acetate in the tail gas at 20°C,30°C,40°C,50°C were measured by changing the inlet concentration of ethyl acetate.Thereby the conversion rate QEand reaction rate rEof ethyl acetate at different temperatures were obtained.The reaction order n and the reaction rate constant k in the series dynamics model under different temperatures can be obtained by taking ln(-rE) as the ordinate and ln(-Ci) as the abscissa.ln k was plotted on the ordinate and 1/RT was plotted on the abscissa according to the deformation formula of the Arrhenius formula.

The activation energy Eaand the pre-exponential factor A of the reaction can be determined from the slope and the intercept,respectively.

The reaction rate in a fixed bed reactor under ideal conditions is calculated according to Eq.(2) [27]:

where -rEis the reaction rate of ethyl acetate in the reactor(mol﹒cm-3﹒s-1);VRis the volume of the catalyst bed (cm3);FEis the molar flow rate of ethyl acetate in the reactor (mol﹒s-1);QEis the conversion of ethyl acetate.

The power law model refers to the reaction of the main reaction rate proportional to the n-th order equation of the reactant content in the system.Reaction rate constant is calculated according to Eq.(3)

The concentration of O3is much larger than that of ethyl acetate in the experiment,so Eq.(3) can be simplified to Eq.(4)

k is the reaction rate constant (mol﹒cm-3)1-m-n﹒s-1;k’ is the approximate reaction rate constant (mol﹒cm-3)1-m-n﹒s-1;-riis the reaction rate (mol﹒cm-3)1-n﹒s-1;and Ciare the concentration of O3(mol﹒cm-3) and the concentration of ethyl acetate(mol﹒cm-3);m and n are the reaction order.

Reaction activation energy is calculated according to Eq.(5)which is deformation of the Arrhenius equation.

3.Results and Discussion

3.1.Effect of acid treatment

ACF was treated by 5%,10%,15% H2SO4solution in 60 °C water bath for 2 hours,respectively.ACF treated with 10% and 5% H2SO4had better adsorption performance for ethyl acetate than ACF treated without acid treatment and with 15% H2SO4(Fig.2(c)).The specific surface area,pore volume,and micropore volume of ACF with low concentration acid treatment were increased until 10%H2SO4solution (Table 1),indicating that acid treatment with low concentration acid can produce defects and wash away the ash on the ACF surface.Acid treatment with higher concentration acid can reduce micropore specific surface area of ACF and affect adsorption performance.

Fig.2.Adsorption performance(a)of ACF with different acid treatment.Degradation performance(b)of 1 wt%Pd/ACF with and without acid treatment.Reaction conditions:the initial concentration of ethyl ester=350 mg﹒m-3,the concentration of ozone=1000 mg﹒m-3,the space velocity=30,000 h-1 and the reaction temperature=20 °C.

Table 1 Specific surface area and average pore diameter of different catalyst

Fig.3.SEM images of Pd/ACF(a),Ag/ACF(b),Au/ACF(c),Pt/ACF(d).Degradation performance(e)and XRD patterns(f)of different catalysts.Reaction conditions:the initial concentration of ethyl ester=350 mg﹒m-3,the concentration of ozone=1000 mg﹒m-3,the space velocity=30000 h-1 and the reaction temperature=30 °C.

In order to further confirm the catalytic effect of acid-treated ACF loaded with precious metals can be improved,the degradation effect of 1wt% Pd/ACF on ethyl acetate was determined.The adsorption of ethyl acetate on the support ACF was saturated at 30 min,after which the conversion of ethyl acetate decreased,indicating that it entered the catalytic oxidation stage.The degradation effect of the acid-treated catalyst on ethyl acetate is better than that of catalyst without acid treatment (Fig.2(d)).

3.2.Effects of different active components on ACF

As we can see,Pd and Au were uniformly dispersed on the ACF without obvious agglomeration (Fig.3(a),(b)).There was light agglomeration on Ag/ACF or Pt/ACF(Fig.3(c),(d)).The degradation performance of ACF with different active metals can rank in order 1 wt%Pd/ACF>1 wt%Ag/ACF>1 wt%Pt/ACF>1 wt%Au/ACF>ACF(Fig.3(e)),indicating that Pd/ACF is a catalyst with high catalytic activity and the catalytic performance is related to the degree of dispersion of the active ingredient.Different active metals did not change the XRD spectrum skeleton (Fig.3(f)).The two peaks at 23° and 44.3° corresponded to the (0 0 2) and (1 0 0) crystal faces of the graphite structure,respectively.The Au/ACF had a peak at 38.2°,indicating that the particle size of Au loaded on the ACF was larger,which was in accord with the result obtained from the SEM image.Pd had no obvious diffraction peak,indicating that Pd was uniformly dispersed on the activated carbon fiber.

3.3.Effects of different loadings of Pd on ACF

The catalytic performance gradually increased as the loading increased.With 1wt% Pd loading,the catalytic performance was maximized,and afterwards the catalytic performance was significantly reduced with increasing loading (Fig.4(c)).TEM images shows that Pd was uniformly dispersed on the ACF with 1 wt%loadings (Fig.4(a),(d)).

The XRD spectrums of Pd/ACF with different loadings had not changed and no sharp peaks were found (Fig.4(d)),indicating Pd was only dispersed uniformly on the surface of the ACF combing with Fig.5.There were strong peaks in ACF around 2θ=23° and 44° which were mainly caused by the microcrystalline carbon formed by the viscose activation of the viscose nonwoven fabric,indicating that the structure of viscose-based activated carbon fiber was composed of graphite-like microcrystalline structure[25],in which the broad peak of ACF shows that the order of crystallites is disordered [26].There was also no change in the structure of the amorphous carbon,indicating that the preparation conditions such as reduction of sodium borohydride had no significant effect on the structure of the activated carbon fiber.

The main component of the catalyst was C element,which accounts for 99.84% of the total (Figs.5,6(b) and Table 2).Other elements such as N,O,S,Si,P,Pd were relatively small.The proportion of P elements in ACF was higher than that of other elements.This was mainly due to the phosphoric acid-containing activator used in the activation of the viscose-based carbon fibers.

The trends of the N2adsorption isotherms of the catalysts was the same,and the isotherms belonged to the I type and the H4type hysteresis loop (Fig.6(a)).At a relatively low relative pressure (P/P0<0.2),nitrogen was adsorbed on the monolayer of the catalyst surface because nitrogen had a weaker effect on multi-layer adsorption on the solid surface.There was an N2hysteresis loop in the region where the relative pressure was high(P/P0>0.4).This was mainly due to the presence of mesopores in the sample,which existed multiple layers of adsorption and capillary condensation at higher partial pressure.

Fig.4.TEM images of ACF (a) and 1wt% Pd/ACF (b).Degradation performance (c) and XRD patterns (d) of Pd/ACF with different loadings.Reaction conditions:the initial concentration of ethyl ester=350 mg﹒m-3,the concentration of ozone=1000 mg﹒m-3,the space velocity=30,000 h-1 and the reaction temperature=30°C,concentration of sodium borohydride=0.5 mol﹒L-1.

Fig.5.EDS elemental mapping analysis of 1 wt% Pd/ACF.

The specific surface area of the ACF ranged from 760 m2﹒g-1to 920 m2﹒g-1(Table 3).As the Pd loading increased,the specific surface area of the catalyst was continuously decreasing because the loaded Pd filled the pores of the ACF.The microporous specific surface area of the five catalysts accounted for about 70% of the total specific surface area,demonstrating the existence of a large number of micropores and small mesopores and macropores in the ACF.The load of Pd showed less effect on the pore size of ACF.

Table 2 Analysis of elements content of active carbon fiber surface

Table 3 Textural characterization and average pore diameter of different load rate catalysts

According to the XPS spectra,the peaks with binding energies around 336.0 eV and 341.1 eV correspond to Pd0(3d5/2) and Pd0(3d3/2),and the peaks with binding energies around 338.0 eV and 343.3 eV correspond to Pd2+(3d5/2) and Pd2+(3d3/2) (Fig.6(c)).The Pd supported on the ACF mainly existed in the form of Pd0with an amount of 58.8% and Pd2+with an amount of 41.2%.

Fig.6.N2 adsorption isotherms at 77.4 K (a).XPS spectra (b),Pd3d XPS spectra (c) and FTIR spectra (d) of Pd/ACF catalysts.

The absorption peak at 3753 cm-1was due to stretching of free O—H groups on the surface (Fig.6(d)).The broad and strong absorption peak at 3435 cm-1corresponded to the Hydrogenbonded OH groups [30].The peaks at 2921 cm-1are due to the C—H symmetric and asymmetric stretches of residual methylene groups on the surface [28–30].It has been reported that bands at 1100–1450 cm-1could be related to C—O stretching.Peaks at 1550–2050 cm-1corresponded to C=O stretching of carboxylic acid and lactone [31,32].The absorption peaks at 1388 cm-1was due to carboxylic salt [30].The absorption peak at 1728 cm-1and 1123 cm-1was due to the vibration of the aldehyde group and the carboxyl group,respectively.A strong absorption peak at 1630 cm-1was due to water in the sample.The absorption peaks at 595 cm-1were due to the vibration of.The band at 670 cm-1shows the presence of physisorbed carbon dioxide[28].It is worth noting that the bands intensity at 1123,1388,1728 and 2921 cm-1were gradually diminished after repeated use of the catalyst which proves that carboxyl group,methyl group,aldehyde group and methylene group were reduced during the reaction.There was no band intensity at 3753 cm-1coming out on the curve of the prepared catalyst,indicating that free hydroxyl groups were not adsorbed on the surface of the prepared catalyst.After repeated uses of the catalyst,the vibration peak of the catalyst at 3753 cm-1was significantly enhanced,indicating that hydroxyl groups were present on the surface of the catalyst after catalytic ozonation.Therefore,it can be considered that indicating that parts of the hydroxyl radicals produced by the reaction between ozone and water was adsorbed on the surface of the catalyst,and the other parts were used to degrade ethyl acetate at normal temperature.

Fig.7.The effects of NaBH4 concentrations (a),temperatures (b),space velocity (c) and inlet concentrations (d) on catalytic performance of catalysts.

3.4.Effects of the concentration of sodium borohydride

When Pd was not reduced by sodium borohydride,the degradation rate of the catalyst to ethyl acetate was very low,only about 10% (Fig.7(a)).With the increase of sodium borohydride concentration,the degradation rate of ethyl acetate gradually increased.When the concentration of sodium borohydride was 0.5 mol﹒L-1,the conversion rate reached the maximum.Therefore,the reduction concentration of sodium borohydride is 0.5 mol﹒L-1.

Fig.8.Reaction rate constant k for ethyl acetate ozone catalytic oxidation based on power-rate law (a) and activation energy (b).

Fig.9.The gas chromatogram spectra of components in outlet gas (a).Effect of repeated use of catalyst on catalytic performance (b).Reaction conditions:the initial concentration of ethyl ester=350 mg﹒m-3,the concentration of ozone=1000 mg﹒m-3 and the reaction temperature=30 °C and the space velocity=30,000 h-1.

The 1 wt%Pd/ACF catalyst mainly existed adsorption at the first 20 min,and almost completely failed after 40 min at 0 °C (Fig.7(b)).After the experiment,it was found that the catalyst contained water,which was mainly caused by the condensation of water in the mixed gas.It can completely affect its adsorption and generated the worst effect.The effects of the adsorption phase and the catalytic phase between 30 °C and 50 °C are relatively close,and the conversion rate is about 55%.At the reaction temperature of 70 °C,the degradation of ethyl acetate by 1wt% Pd/ACF catalyst decreased significantly with 20% degradation rate after 2 h.This is mainly because the ozone is self-decomposed at higher temperature.When the reaction temperature is 90 °C,the decomposition rate of ozone is higher,so the catalytic effect is worse.

The catalytic effect of the catalyst was only 20%when the space velocity is 40,000 h-1while the catalytic effect was maintained at about 60% when the space velocity is less than 30,000 h-1(Fig.7(c)).When the space velocity was too large,the amount of the ethyl acetate which passes through the catalyst per unit time became large,and the residence time of ethyl acetate on the catalyst was short.The suitable space velocity of the catalyst is less than 30,000 h-1.

Under the same ozone concentration,the smaller the inlet concentration,the slower the adsorption saturation of ethyl acetate on ACF,the more ozone was exposed per unit time,and the more ethyl acetate was degraded.The catalytic activity was maximum at a ratio of initial concentration to ozone concentration of 1:3(Fig.7(d)).

3.5.Activation energy of ethyl acetate

The fitting analysis obtained an adjusted R2degree of fit values of 0.993,0.987,0.994 and 0.997 for the four selected temperature points(20,30,40 and 50°C),respectively.The results indicate that the power law model was well fitted.The reaction rate constant of ethyl acetate increases with the reaction temperature,which is in line with Arrhenius theory.The values of m increases with the reaction temperature,indicating that the higher the temperature at room temperature,the more sensitive the reaction rate of ethyl acetate to the change of concentration.The different conversions of ethyl acetate at 20 °C,30 °C,40 °C and 50 °C were measured,respectively.Then the power factor model and Arrhenius formula were used to obtain the pre-exponential factor A=1 × 108s-1and the reaction activation energy was 113 kJ﹒mol-1(Fig.8(a),(b)).

The degradation products of ethyl acetate on Pd/ACF catalyst at normal temperature were detected by gas chromatography and the carbon balance was calculated (Fig.9(a)).It can be found from Fig.9(a)that in addition to the main product CO2,a certain amount of CO and trace amounts of CH4were also detected.It was detected that the CO2concentration was 320.76 mg﹒m-3,the CO concentration was 27.40 mg﹒m-3,the CH4concentration was 0.35 mg﹒m-3,and the ethyl acetate concentration was 139.17 mg﹒m-3.Since the ethyl acetate concentration before the reaction was 350 mg﹒m-3,the total carbon in the gas phase can be detected as 91.87%.The degradation rate of ethyl acetate in the first three times of the catalyst was 60% (Fig.9(b)).And the degradation rate was reduced to 50% after continuing to recycle twice,indicating that the catalyst had certain stability.

4.Conclusions

The specific surface area and pore volume of the ACF with acid treatment was larger than not.Acid treatment and loading of Pd do not significantly affect the average pore size.The specific surface areas of the catalysts gradually decrease as the loading increases.There are a large number of micropores in the ACF,with a small amount of mesopores.The Pd/ACF prepared has certain stability.Pd is uniformly distributed on the surface of ACF without obvious agglomeration,and Palladium mainly exists in the form of Pd0with a amount of Pd2+.The main component of Pd/ACF was C element,which accounted for 99.84%of the total.There were few other elements such as N,O,S,Si,P,Pd.

The catalytic performance is related to the degree of dispersion of the active ingredients.Catalytic ozonation performance is maximized and the conversion of ethyl acetate is 60%when the loading is 1wt%,the normal temperature is 30–50°C,the feed space velocity is 15,000–30,000 h-1and the concentration ratio of ozone to ethyl acetate is 3:1.The activation energy of ethyl acetate by Arrhenius equation is 113 kJ﹒mol-1.

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 supported by the National Key R&D Program of the Ministry of Science and Technology,China (Grant No.2018YFC0705304);and the Key Scientific and Technological Support Projects,Tianjin City,China (Grant No.19YFZCSF01090).