Efficient CO2 adsorption using chitosan, graphene oxide, and zinc oxide composite

This study was deeply focused on developing a novel CTS/GO/ZnO composite as an efficient adsorbent for CO2 adsorption process. To do so, design of experiment (DOE) was done based on RSM-BBD technique and according to the DOE runs, various CTS/GO/ZnO samples were synthesized with different GO loading (in the range of 0 wt% to 20 wt%) and different ZnO nanoparticle’s loading (in the range of 0 wt% to 20 wt%). A volumetric adsorption setup was used to investigate the effect of temperature (in the range of 25–65 °C) and pressure (in the range of 1–9 bar) on the obtained samples CO2 uptake capability. A quadratic model was developed based on the RSM-BBD method to predict the CO2 adsorption capacity of the composite sample within design space. In addition, CO2 adsorption process optimization was conducted and the optimum values of the GO, ZnO, temperature, and pressure were obtained around 23.8 wt%, 18.2 wt%, 30.1 °C, and 8.6 bar, respectively, with the highest CO2 uptake capacity of 470.43 mg/g. Moreover, isotherm and kinetic modeling of the CO2 uptake process were conducted and the Freundlich model (R2 = 0.99) and fractional order model (R2 = 0.99) were obtained as the most appropriate isotherm and kinetic models, respectively. Also, thermodynamic analysis of the adsorption was done and the ∆H°, ∆S°, and ∆G° values were obtained around − 19.121 kJ/mol, − 0.032 kJ/mol K, and − 9.608 kJ/mol, respectively, indicating exothermic, spontaneously, and physically adsorption of the CO2 molecules on the CTS/GO/ZnO composite’s surface. Finally, a renewability study was conducted and a minor loss in the CO2 adsorption efficiency of about 4.35% was obtained after ten cycles, demonstrating the resulting adsorbent has good performance and robustness for industrial CO2 capture purposes.


List of symbols
First-order kinetic model rate constant (min −1 ) k F Freundlich model constant (mmol.g−1 .bar−1/n ) k 2  Second-order model rate constant (g.mmol −1 .min−1 ) k n Fractional-order rate constant [(mmol/g) (1−n) min −m ] m Mass of adsorbed gas(mg) M w Molecular weight (g/mol) P Pressure (bar) P e Equilibrium pressure (bar) q Adsorption capacity (mg/g) q e Equilibrium adsorption capacity (mmol/g) q m Maximum adsorption capacity (mmol/g) q t Adsorption capacity at time t (mmol/g) R Gas constant (8.314J. K −1 .mol−1 ) R 2 Correlation coefficient with the lowest amount of ascorbic acid as a reducing agent exhibited the highest specific surface area, porosity, and level of functionalization achieving a capacity of 2.1 mmol/g at 25 °C and 1 bar pressure.Hsan et al. 35 presented a sustainable and environmentally friendly approach for developing chitosan (CTS) grafted GO composite as a solid sorbent for the CO 2 gas capture process.The obtained composite exhibits an adsorption capacity of approximately 0.257 mmol/g at 1 bar for CO 2 gas, which is significantly higher than that of pure CTS.
Recent research showed that adding metal oxide nanoparticles to adsorbents considerably increased their ability to adsorb CO 2 molecules [36][37][38] .Zhou et al. 39 investigated the impact of Li doping on the structural properties of a MOF material namely HKUST-1 and explored the CO 2 uptake capability of the resulting material.The morphological characteristics of HKUST-1 remained intact, and the introduction of Li helped rectify certain defects which led to an optimized CO 2 uptake capacity of 4.3 mmol/g.Krap et al. 40 synthesized MFM-300(Ga 2 ), a Ga-based MOF, using a solvothermal reaction of Ga(NO 3 ) 3 and H 4 L. Method of Fe-doping improved gas uptake capabilities, especially for CO 2 adsorption, with a 49% enhancement in CO 2 uptake (2.86 mmol/g at 273 K at 1 bar).MOF-74(Ni, Co) was synthesized, characterized, modified with Pd-loaded AC, and evaluated for CO 2 adsorption capacity and CO 2 /N 2 separation efficiency by Adhikari and Lin 41 Modified MOFs showed enhanced adsorption capacity of 12.24 and 11.42 mmol/g at 298 K and 32 bar and improved bond distances.Another study by Al-Mamoori et al. 42 presented novel metal oxide-doped CaO adsorbents with high capture capacity, fast kinetics, and long-term stability.Fe and Ga-doping improved adsorption performance, achieving high adsorption capacities (13.7 and 14.2 mmol/g, respectively) and reversible performance.The materials exhibit about a 5% loss in adsorption capacity after ten cycles and exhibit favorable CO 2 uptake.
Although, chitosan as an amine-rich biopolymer is appropriate for CO 2 adsorption purposes but the low surface area of this type of porous polymer leads to a low CO 2 uptake capability which creates some limitations for large-scale CO 2 capture applications.As a promising technique to tackle this issue, dispersion of a porous material such as GO or a metal oxide such as ZnO nanoparticles on the surface of the chitosan support causes improving adsorbent surface's heterogeneity and gaining adsorbent's surface area leading to increasing the CO 2 uptake capability of the chitosan support 24 .Therefore, this study was conducted aim of developing a novel chitosan/graphene oxide/zinc oxide composite, denoted as CTS/GO/ZnO composite for CO 2 adsorption purposes.To do so, different composite samples with various compositions were prepared to investigate the effect of GO and ZnO nanoparticles loading weight percent on the composite's CO 2 adsorption capacity.By reviewing the literature, it was concluded that many researchers only considered the physiochemical properties of the CTS/GO samples, while developing a predictive model is vital for industrial process design applications.Therefore, the CO 2 adsorption process modeling was applied using the RSM-BBD approach aim of provide a predictive model to correlate the dependency of the CTS/GO/ZnO adsorbent's adsorption capacity to the operational condition such as temperature and pressure and synthesis parameters including GO and ZnO loading weight percent.Moreover, the best composition of the CTS/GO/ZnO composite was determined through optimizing the CO 2 adsorption process, and the obtained adsorbent was characterized using FTIR, XRD, SEM, and BET analysis.Finally, isotherm modeling, kinetic modeling, and thermodynamic assessment of the adsorption process were done to investigate the CO 2 adsorption mechanism regarding the interactions between CO 2 molecules and the adsorbent's surface.

Graphene oxide (GO) synthesis
To synthesize the GO nanoparticles from graphite powder, the modified Hummer's method 43 was conducted similar to the work conducted by Kumar S.A.K et al. 44 .In a typical procedure, the graphite powder (1.5 g) was charged to a round bottom flask containing the mixture of concentrated H 2 SO 4 /H 3 PO 4 solution with a ratio of 9:1 (180:20 ml).Next, KMnO 4 powder (9 g) was introduced to the flask gradually and the mixture temperature was kept at 40 °C, after the complete addition of KMnO 4 powder the temperature was increased up to 60 °C and the flask contents were mixed for 12 h.Then, the flask contents were cooled down and the mixture was entered into ice (200 ml) containing H 2 O 2 30% (2 ml), which resulted in the solution color changing from brown to yellow.Finally, the mixture was filtrated and the obtained solid was purified three times with 250 ml of distilled water, followed by 250 ml of HCL solution (30%) and pure methanol (250 ml).After the purification stage, the GO sample was obtained by centrifugation of the resulting powder at 5000 rpm for 40 min followed by drying inside a vacuum oven at 60 °C for 11 h.The general route for the synthesis of GO from graphite powder is shown in Fig. 1.

Synthesis of ZnO nanoparticles
Similar to the method reported by Deb A. et al. 45 , the Zinc Oxide (ZnO) nanoparticles were synthesized through precipitation of zinc ions (Zn 2+ ) in an alkali solution.In a typical method, the ZnSO 4 .7H 2 O (0.72 g) was dissolved in 120 ml of deionized water (DI) to prepare 120 ml of Zn 2+ solution (0.025 M).Also, NaOH solution (0.5 M) was prepared by dissolving NaOH (2.0 g) in 120 ml of DI.Next, the NaOH solution was gradually added to the Zn 2+ solution to precipitate zinc ions, then 20 ml CTAB solution (100 mg/l) was added to the mixture to prevent www.nature.com/scientificreports/ZnOH crystal growth.Finally, the obtained solid was filtrated and purified with distilled water and methanol several times and the purified sample was calcined at 420 • C for 3 h which yielded the ZnO nanoparticles.

Synthesis of CTS/GO/ZnO composite
The CTS/GO/ZnO composite samples with various weight percent of the GO and ZnO nanoparticles were synthesized similarly to the method reported by Zhang, et al. 46 .For example, to prepare a composite sample with the 20 wt% of GO and 20 wt% of ZnO nanoparticles, the chitosan powder (0.75 g) was dissolved in 30 ml acetic acid solution (1.5% w/v) and the GO nanoparticles (0.25 g) was added to the solution.The solution was stirred at room temperature for 3 h followed by adding the ZnO nanoparticle (0.25 g), and the flask contents were mixed for 30 min.Afterward, the flask contents were poured into a Teflon-lined autoclave and the autoclave was placed inside an oven at the temperature of 120 °C for 12 h.Finally, the autoclave's contents were cooled down to room temperature and the resulting composite was purified using deionized water and methanol followed by drying in an oven at 95 °C for 10 h.The general synthesis procedure of the CTS/GO/ZnO composite is illustrated in Fig. 2. Similar to the mentioned procedure, the CTS/GO/ZnO-10 wt% sample was prepared by using the chitosan powder (0.80 g), GO nanoparticles (0.10 g), and ZnO nanoparticle (0.10 g), also for the case of CTS/GO/ ZnO-30 wt%, the chitosan powder (0.60 g), GO nanoparticles (0.45 g), and ZnO nanoparticle (0.45 g) were used.

Adsorbent characterization
X-ray diffraction analysis (XRD, STOE STADIMP, Germany) was used to determine the crystalline structure, and phases of the samples under Cu Ka radiation, 40 kV voltage.Also, X'Pert HighScore Plus software was utilized for phase detection.To characterize the samples' textural properties and pore characteristics, nitrogen adsorption/desorption experiment was performed by utilizing an ASAP 2020 M analyzer.Fourier Transfer Infrared (FTIR) analysis was conducted using a PerkinElmer FTIR spectrometer to detect the samples' functional groups.

Adsorption setup
The CO 2 adsorption capacity of the obtained solid adsorbents was measured by utilizing a volumetric setup which was illustrated in Fig. 3.As can be seen in Fig. 3, the pure CO 2 gas is warmed up by utilizing an electrical heater followed by entering into a mixing chamber, and inside the chamber, both the pressure and the temperature of the CO 2 gas equalize.Next, the gas is entered into the adsorption vessel and the gas will be contacted to the bulk powder of the adsorbent.The adsorption vessel consists of a stainless steel chamber with a total volume of 254 cm 3 , inner diameter of 6 cm, and height of 9 cm, which is sealed by an appropriate cap to eliminate the leak of the gas to the environment.The adsorption vessel's temperature is adjusted at the set point temperature using a water jacket placed around the adsorption vessel.As soon as the CO 2 gas is entered into the adsorption vessel, the CO 2 uptake process will be started and the vessel's features including the pressure and temperature and their corresponding adsorption time are collected.The CO 2 adsorption capacity can be calculated by considering the CO 2 gas's mass differences between the initial and final adsorption time using Eqs.(1) and (2) 47 .Which m i and m f refer to the mass of the CO 2 gas at the initial and final adsorption time.The subscripts i and f are related to the initial and final condition.Also, the terms including P is pressure, V is adsorption vessel volume, R is the global gas constant, T is temperature, M w is CO 2 molecular weight, W is adsorbent's weight, and Z is gas compressibility factor.The term B is the second virial coefficient which can be calculated by using an empirical correlation for both polar and non polar gaseous system introduced by Tsonopoulos at 1974 48,49 .The mentioned correlation and its parameters are presented in Eqs. ( 3), ( 4) and ( 5) General procedure of synthesis of the CTS/GO/ZnO composite.

Response surface methodology
Recently, in the various fields of engineering, the design of experiment (DOE) has been used as an applicable tool that can correlate independent variables to the response regarding considering the variables' interactions with each other.DOE by utilizing response surface methodology (RSM) can be applicable to investigate and model the situations where a modeling's response is influenced by some independent variables within a defined design space.Investigation of the variables' interaction concerning decreasing the number of experiments is considered a main advantage of the RSM technique.In this method, coefficients of a quadratic model represented in Eq. ( 6), are determined by fitting the experimental values on the mentioned equation, also the obtained model's significance is evaluated through analysis of the variance (ANOVA) using some statistical criteria such as model's P-value, F-value, and correlation coefficient (R 2 ) 3,15 .The R 2 value and absolute average relative error (AARE%) can be calculated using Eqs.( 7) and ( 8), respectively 50 .
where the term β 0 refer to the model's constant, β i indicates linear effect, β ii refers to the quadratic effect, β ij is the interaction term, and n is equal to the variable's count.
The RSM technique' Box Behnken Design (BBD) is regarded as a successful strategy for correlating independent factors to response.In this method, the range of variables is separated into three levels: − 1, 0, and + 1, which are related to the interval's lower bound, middle point, and upper bound, respectively 51 .In this study, the DOE was applied in the Design Expert software (version 11) by utilizing RSM-BBD approach to correlate the effect of independent variables including graphene oxide (GO) loading (wt.%),Zinc Oxide (ZnO) loading (wt.%), adsorption process's temperature (T), and pressure (P) on the CO 2 adsorption capacity of the GO-ZnO-chitosan composite adsorbent.The mentioned variables' intervals are reported in Table 1.

Analysis of variance (ANOVA)
The ANOVA results of the CO 2 uptake process using CTS/GO/ZnO composite and the impact of effective factors such as GO loading weight percent, ZnO loading weight percent, adsorption process temperature, and pressure are presented in Table 2. To determine the accuracy of the obtained RSM-based model within the design space, two statistical criteria including P-value and F-value are presented which refer to the significance of the obtained model or the model's terms when the P-value is less than 0.05 and F-value is more than 1.Therefore, by considering the P-value of the model (less than 0.0001) in Table 2, it can be concluded the obtained model has enough precision within design space, also the P-values of the model's terms show that the terms including Table 1.Independent variables levels in RSM-BBD modeling.GO loading (wt%), temperature, and pressure are particularly significant and more effective than the term ZnO loading (wt%), in the obtained model.The model's F-value (132.65) is significantly higher than 1, which highlights the model's importance and the low likelihood that it is the result of noise (probability is less than 0.01%) 52 .The generated model is trustworthy because the R 2 = 0.992 is higher than 0.8 and the differences between the predicted R 2 = 0.954 and the adjusted R 2 = 0.985 is less than 0.2.So, the developed predictive model can be applicable for designing industrial processes because the adequate precision value for signal-to-noise evaluation is 43.21, which is more than 4.

CO 2 uptake model based on the RSM-BBD approach
To correlate the effect of the mentioned parameters including GO loading weight percent, ZnO loading weight percent, adsorption process temperature, and pressure on the CO 2 uptake capability of the CTS/GO/ZnO composite, A quadratic model, obtained from the RSM-BBD method, is presented in Eq. (9).Also, comparison between the CO 2 adsorption quantities, predicted using RSM-based model and the experimental values is illustrated in Fig. 4. According to this figure, the provided semi-empirical model's superior precision can be proved by considering the sufficient centralization of the predicted and the experimental values of the CO 2 adsorption capacity over the diagonal line.

Effect of input parameters on the CO 2 uptake capability of CTS/GO/ZnO composite
To investigate the effect of input factors on the CTS/GO/ZnO composite's CO 2 adsorption capability, perturbation plot is prepared and shown in Fig. 5.In this figure, the A, B, C, and D curves correspond to the factors namely GO loading weight percent, ZnO loading weight percent, temperature, and pressure, respectively.According to the Fig. 5a, an increase in CO 2 uptake capability of the CTS/GO/ZnO composite can be observed by gaining the GO and ZnO nanoparticles loading on the chitosan bio adsorbent up to 22.5 wt% and 18.3 wt%, respectively, while increasing the loading more than mentioned values causes a gradual reduction in CO 2 adsorption capacity of the solid sorbent.The CTS/GO/ZnO composite's CO 2 adsorption improvement can be attributed to the increasing surface area of the composite sample in comparison with pristine samples such as chitosan bio adsorbent, GO, and ZnO nanoparticles.Also, the oxygen-rich nature of the GO and ZnO samples and also the presence of various electrons withdrawing functional groups such as -NH 2 , -COOH, -NO 2 , and -OH in the GO sample's structure, make these nanoparticles a promising candidate for CO 2 uptake applications 53 .The mentioned functional group's dispersion on the surface of the chitosan sample causes increasing heterogeneity of the composite sample, resulting in gaining the CO 2 uptake capability of the CTS/GO/ZnO composite through improving dipolequadropole interaction between the adsorbent's surface and CO 2 molecules 24 .However, increasing the GO and ( 9) www.nature.com/scientificreports/ZnO nanoparticles loading value more than the critical quantity (GO = 22.5 wt% and ZnO = 18.3 wt%), causes a decrease in the CO 2 capture capability of the composite sample.It can be related to the pore filling and decreasing adsorbent's pore width and pore blocking effect in the excess loading values of the GO and ZnO nanoparticles 54 .
The CO 2 adsorption capacity in CTS/GO/ZnO is influenced by surface area and pore volume.A larger surface area allows more CO 2 interaction and adsorption, with materials like GO and ZnO having high surface areas.
Pore volume provides space for CO 2 trapping, increasing the adsorption capacity.The composite nature of CTS/ GO/ZnO leads to synergistic effects, where their properties complement each other for higher CO 2 adsorption.
As shown in the Fig. 5a, increasing the adsorption process's pressure causes a sharp enhancement in the CO 2 uptake capability of the CTS/GO/ZnO composite.It can be attributed to the improved mass transfer of CO 2 molecules into the previously inaccessible cavities as well as the decrease in the desorption of the captured CO 2 molecules 55 .Considering a decreasing behavior of the CO 2 adsorption by increasing the process's temperature, it can be concluded the CO 2 uptake reduction may be related to the physical adsorption of the CO 2 molecules on the surface of the composite sample, also an increase in temperature increases the movement of absorbed CO 2 molecules, which leads to improved desorption of CO 2 molecules at higher temperatures 24 .
To better show the effect of the mentioned factors and their interactions on the CO 2 adsorption capacity of the CTS/GO/ZnO sample, three-dimensional plots are provided and shown in Fig. 5c to Fig. 5d.Increasing the presence of GO and ZnO in Fig. 5b is shown to boost CO 2 adsorption.However, it's worth noting that an excessive amount of these materials can diminish the overall CO 2 adsorption capacity.This happens because, past a certain point, the surface sites on these materials get filled up or blocked.When there's an excess of GO and ZnO, it can impede these materials' ability to efficiently capture and retain CO 2 molecules.Instead of enhancing adsorption, an overabundance of material can create obstacles that restrict the interaction between CO 2 and the available adsorption sites, resulting in a decrease in the overall adsorption capacity.Therefore, maintaining an optimal balance in the loading levels of these materials is crucial for achieving the best CO 2 adsorption performance.As shown in Fig. 5b, an increase in pressure at the optimum loading of GO in composite can increase CO 2 adsorption.This occurs because higher pressure provides a driving force for CO 2 molecules to be more readily attracted to the adsorption sites on the GO surface.With increased pressure, there are more CO 2 molecules in the gas phase, and they are more likely to interact with and be adsorbed by the available sites on the GO material.As a result, the overall CO 2 adsorption capacity is enhanced under these conditions, making it an effective strategy for improving the adsorption process.
Increasing in pressure and decreasing in temperature can increase CO 2 adsorption capacity.This phenomenon occurs because higher pressure and lower temperature conditions favor the formation of stronger interactions between CO 2 molecules and the adsorption sites on the material's surface.At higher pressure, there are more CO 2 molecules available in the gas phase, increasing the chances of these molecules coming into contact with and being adsorbed by the adsorption sites.Additionally, lower temperatures reduce the kinetic energy of CO 2 molecules, making them more likely to be captured and held by the adsorption sites, as their movement is slowed down.These combined effects result in a higher adsorption capacity for CO 2 , making it a more efficient process as shown in Fig. 5d.According to these figures, the highest CO 2 adsorption capacity of the CTS/GO/ZnO composite can be obtained at the maximum pressure, minimum process's temperature, and the middle ranges of GO and ZnO nanoparticles loading weight percent.CO 2 capture process optimization Industrial use of the CTS/GO/ZnO composite requires fine-tuning the operational parameters to achieve the highest CO 2 uptake capacity.The CO 2 adsorption optimization performed here was accomplished with the help of the optimization module of the Design Expert software.To determine the maximum uptake capacity of the composite, the effective parameters including GO loading, ZnO loading, temperature, and pressure were set as 'in range' and the adsorption capacity was set to 'maximize' followed by running optimization.Table 3 reports the optimum condition with the highest desirability.By conducting the CO 2 capture experiment under the obtained conditions, the RSM-BBD approach's optimized CO 2 adsorption capability was experimentally verified.As a result, the experimental CO 2 uptake capacities of 479.08, 476.33, 475.08, and 479.19 mg/g and an insignificant AARE of about 1.42% was obtained after four replications of the adsorption test under optimum conditions.Table 3.The result of CO 2 uptake process optimization using RSM-BBD.

Adsorbents characterization
The FTIR spectra of the samples namely chitosan, GO, and CTS/GO/ZnO (GO = 23.8wt%, ZnO = 18.2 wt%) are displayed in Fig. 6a.In the chitosan FTIR spectra, a band allocated at 3348 cm -1 is correspond to the O-H and N-H stretching vibration of hydroxyl group and primary amine group (NH 2 ), also the characteristic peaks allocated at 2880, 1660, and 1089 cm -1 are attributed to the C-H stretching vibration of CH and CH 2 , N-H bending vibration of the NH 2 , and C-O stretching vibration of the hydroxyl moiety, respectively.Moreover, two peaks at 1428 1382 cm -1 refer to the deformation vibration of the N-H bond in the NH 2 44,56 .The FTIR spectrum of the GO sample reveals peaks at 3424, 1729, 1619, 1218, and 1039 cm -1 which corresponds to the O-H stretching vibration of carboxylic acid or hydroxyl functionalities, C=O stretching vibration of carboxylic acid, C=C bond's vibration of unoxidized graphite moiety, C-O stretching vibration, and C-O-C stretching vibration of the epoxy group 57 .Considering the FTIR spectra of the CTS/GO/ZnO sample and GO sample, it can be observed the intensity of the C=O bond (1723 cm -1 ) is reduced, which refers to the chemical bonding between carboxylic acid and amine functionalities and formation of the amide group.A strong peak at 3436 cm -1 is attributed to the stretching vibration of O-H, N-H, and hydrogen bonding of polysaccharide molecules, also the peak at 1564 cm -1 is ascribed to the N-H vibration of the amide group.Moreover, a band at 494 cm -1 is related to the Zn-O bond of ZnO nanoparticles of the CTS/GO/ZnO composite 46,57 .The results of XRD analysis of the GO, www.nature.com/scientificreports/chitosan, and CTS/GO/ZnO composite (GO = 23.8wt% and ZnO = 18.2 wt%) are illustrated in Fig. 6b.According to the GO's XRD pattern a sharp diffraction peak at 2θ = 10.9 • is related to the crystalline nature of the GO 35 .
The XRD pattern of the chitosan sample exhibited two diffracted peaks near the 2θ = 9.9 • and 2θ = 20.8°,which are attributed to the crystalline structure of commercial chitosan sample (JCDPS no.39-1894) 58 .Considering the XRD pattern of the CTS/GO/ZnO composite, it can be observed that some characteristic peaks have appeared at the 2θ quantities equal to 31.6°, 34.7°, 36.1° and 47.5°.These peaks are attributed to the ZnO nanoparticle's crystal plains of 100, 001, 101, and 102, respectively (JCDPS no.36-1451) 46 .Moreover, the intensity of the peaks located at 2θ = 10.4° and 2θ = 20.5 • in the composite sample's XRD pattern are weaker than the peaks of the GO and chitosan pristine samples, therefore it can be concluded the crystallinity of the composite sample is reduced and the final structure turned into amorphous structure 44 .
The charactrization properties of the mentioned samples were investigated by conducting a Nitrogen adsorption-desorption experiment at 77.3 K.The isotherm plots of the N 2 adsorption-desorption are illustrated in the Fig. 6c and the morphological characteristics of the samples are presented in Table 4. Based on the findings of the Fig. 6c, all samples exhibit IV type isotherm (based on IUPAC isotherm classification) and H2 hysteresis loop.Considering isotherm plots, a minor quantity of the N 2 was adsorbed by the GO, ZnO, and chitosan sample at a relative pressure of less than 0.2, therefore these samples are rarely constructed by microspores.Meanwhile, the composite sample indicated more quantity of N 2 adsorption at the mentioned relative pressure range, so it can be concluded the micropore volume of the composite sample is much more than other samples 35,59 .Considering the hysteresis loop presence in the samples isotherm plots at the relative pressure between 0.2 and 0.8, it can be noticed that all samples especially GO, and the CTS/GO/ZnO samples formed by meso-pores, also the hysteresis loop at the relative pressure higher than 0.8 indicate the existence of macro-pores and intraparticle cavities 49 .As a result, the surface area of a composite of GO, ZnO, and CTS is enhanced due to their synergistic effects.GO, being a two-dimensional carbon material, provides a large surface area for adsorption and increased porosity, while ZnO contributes to surface roughness and provides active sites for chemical interactions.Chitosan acts as a binder, promoting the dispersion of GO and ZnO, further increasing the available surface area.This unique combination maximizes the accessible surface for CO 2 adsorption.SEM images of the chitosan, GO, and CTS/ GO/ZnO composite are shown in Fig. 7. Based on the SEM images, the surface porosity of the composite sample is more than the GO and chitosan samples, also the composite sample is formed from cavities with a narrower pore size compared with GO sample.The area circled as red color in the Fig. 7c indicate ZnO nanoparticles were successfully dispersed and loaded on the surface of the composite sample.In addition, sections marked in the Fig. 7c with arrows indicate the porosity for CO 2 adsorption.

Adsorption isotherms
To effectively design adsorption systems, isotherm modeling may provide relevant data on the adsorption process mechanism.Such models explain the nature of the interactions and reactions that occur between the solid sorbent's surface and adsorbate's molecules during gas capture processes.The current study used isotherm models such as Langmuir, Freundlich, Dubbin-Radushkevich (D-R), and Temkin as two-parameter models to explain the adsorption process.The aforementioned models are highlighted in Eqs.(10), (11), ( 12), ( 13), respectively 60 .
where q e and P e refer to equilibrium CO 2 uptake capacity (mg.g -1 ) and equilibrium pressure (bar).Other terms such as q l is maximum uptake capacity (mg.g -1 ) based on the Langmuir model, K l (bar -1 ) is Langmuir model's constant, K F (mg.g -1 .bar 1/n ) and n F are Freundlich model's constants.In the Temkin model, the term A T (L.mol -1 ) is the model's constant and B is the virial equation of the state's first coefficient ( Langmuir: q e = q l K l P e 1 + K l P e , the terms (mol 2 .J -1 ) and ω (J.mol -1 ) in the D-R model refer to the model's constant and Polanyi potential, respectively 60 .
To perform isotherm modeling, a new composite sample was prepared according to the optimum GO and ZnO nanoparticles loading weight percent (GO = 23.8wt% and ZnO = 18.2 wt%), followed by conducting CO 2  www.nature.com/scientificreports/adsorption tests at the different pressures between 1 to 9 bar and the temperature of 298 K, 308 K, 318 K, and 328 K.
Next, the mentioned models were fitted on the equilibrium adsorption data and the model's terms were obtained which are reported in Table 5, also the fitted curves of the isotherm models at 298 K are displayed in Fig. 8. Based on the Table 5 findings, a descending order of the Langmuir model's constant ( q l ) by increasing temperature, proves the exothermic nature of CO 2 uptake process.In addition, by considering the K F value of the Freundlich model that refers to the tendency of the adsorbate molecule to be adsorbed by solid sorbent, it can be observed increasing temperature caused a gradual decrease in the K F value, so the CO 2 adsorption by CTS/GO/ZnO composite occurred via physisorption mechanism dominantly 61 .Also, a favorable CO 2 capture process is reflected by the Freundlich model's parameter (n F ), which lies between 1 and 2. Furthermore, the D-R model's constant ( ω ) stands for the free energy of adsorption, a physisorption mechanism is indicated by the ω values below 8 kj/mol, while chemisorption is indicated by the ω values between 8 and 16 kj/mol.Thus, the ω values below 8 kj/mol confirm the previously established physical adsorption of CO 2 molecules 62 .As a result of comparing different isotherm models, the Freundlich model was chosen as the most appropriate model with the highest R 2 value.Therefore, it can be concluded that the surface of CTS/GO/ZnO composite is heterogeneous, and multi-layer CO 2 adsorption is performed on the adsorbent's surface.A high Freundlich constant shows that the composite has a high adsorption capacity, while a low exponent means a more linear adsorption isotherm 49 .

Kinetic modeling
Generally, investigation of an adsorption process kinetic can provide some useful insights about the rate of adsorption which significantly affects adsorption efficiency.Typically, kinetic models are employed to determine the gas phase's retention time and adsorption rate, which are practical for industrial process design purposes such as fixed bed columns.Therefore, to investigate the CO 2 adsorption kinetic, common kinetic models such as first order, second order, Elovich, Rate Controlling, and Fractional order models were applied.The aforementioned kinetic models are introduced in Eqs. ( 14), ( 15), ( 16), ( 17) and (18), respectively.The experimental CO 2 adsorption data were collected by utilizing the newly synthesized CTS/GO/ZnO composite (GO = 23.8wt% and ZnO = 18.2 wt%) at the pressure of 5 bar and the temperature 298 K, 308 K, 318 K, and 328 K.The results of kinetic modeling and the model's parameters are summarized in Table 6, also the curves of the kinetic models were plotted and illustrated in Fig. 9. (14)  First order : q t = q e 1 − e k 1 t , (15) Second order : q t = (q 2 e k 2 t)/(1 + q e k 2 t), (16) Elovich: q t = βln(αβ) + βln(t), (17)  Rate controlling : q t = k c t 0.5 , (18) Fractional order : q t = q e − (n − 1) .   where the terms including k 1 , k 2 , k c , and k n are the constant of their corresponding kinetic models 63 .The firstorder model assumes that the rate of adsorption is proportional to the difference between the saturation concentration and the quantity of the adsorbed constituent.The growing influence of chemical adsorption on the adsorption process is reflected in a decline in the R 2 quantity of the first-order model, as reported in Table 6 64 .
The rate controlling model suggests that the rate of the adsorption is affected by intraparticle diffusion.Increasing the R 2 value of this model at higher temperatures suggests that diffusion is the rate-controlling mechanism 65 .Table 6 shows that the fractional order kinetic model successfully correlates the CO 2 adsorption capacity to the adsorption time, as measured by the correlation coefficient (R 2 ).A more accurate representation of the adsorption process that departs from integer order kinetics is provided by the latter model.It takes into account factors like surface heterogeneity, multilayer adsorption, and interactions between adsorbate molecules and adsorbent surface, all of which add to the adsorption process' complexity 66 .

Adsorption thermodynamic
Thermodynamic analysis of the CO 2 uptake process was applied to determine parameters such as enthalpy changes (ΔH), Gibbs free energy changes (ΔG), and entropy changes (ΔS).Using Eq. ( 19), the distribution factor (K d ) was calculated at a constant pressure of 5 bar and the temperatures of 298 K, 308 K, 318 K, and 328 K. Van't Hoff plot, shown in Fig. 10, was prepared through plotting the obtained K d values versus their correspond reverse temperature ( 1 T ) .As presented in Eq. ( 20), the van't Hoff plot's slope and intercept refer to the enthalpy change ( H 0 ) and entropy change ( S 0 ), respectively.The G 0 of the process can be measured using Eq. ( 21).The thermodynamic parameters of the CO 2 adsorption process are reported in Table 7.
where P is the adsorption vessel's pressure difference during the process, V is the vessel's volume, W is defined as the adsorbent mass, and R refers to the gas constant (8.314J.mol -1 .K -1 ) 67 .According to the table, the H 0 value (− 19.21 kJ/mol) reflects the physical adsorption of the CO 2 molecules on the CTS/GO/ZnO composite regarding the heat releases quantity less than 20 kJ/mol for physisorption mechanism and heat releases more than 40 kJ/mol for chemisorption mechanism 68 .In addition, significant information about the randomized or organized interaction between the gas phase and the solid sorbent can be derived from the adsorption S 0 value.A positive S 0 value indicates that the adsorption is taken place more random, whereas negative S 0 value proves less randomized adsorption.Considering the negative value of S 0 (− 0.032 kJ/mol K), it can be inferred that ( 19)  the gas-solid interface is less random.The negative values of the G 0 at different temperatures highlight the favorability of the CO 2 adsorption process and also the processes occurred spontaneously 69 .

CTS/GO/ZnO composite renewability
As the most important economic aspect for the development of new solid adsorbents, the recycling and reusability of the adsorbent should be investigated to further evaluate the feasibility of CO 2 adsorption in an industrial scale.To investigate the stability and renewability of the resulting composite, ten cycles of the CO 2 adsorption process were conducted at a pressure of 5 bar and temperature of 45 °C.In each cycle, the spent adsorbent was recycled at the temperature of 90 °C in a vacuum oven for 2 h.The composite's efficiency in CO 2 adsorption after ten cycles is plotted and illustrated in Fig. 11.According to this figure, minor losses of about 4.35% in the adsorbent efficiency for capturing CO 2 can be observed after ten cycles.Therefore, the obtained composite can be used as a promising candidate for CO 2 capture applications.

Comparison between the current work and similar research
In this section, the CO 2 uptake performance of the optimum CTS/GO/ZnO composite is compared with similar studies in this area.The results of the research on CO 2 adsorption by utilizing GO, chitosan, and composite adsorbent and their corresponding operational conditions are reported in Table 8.According to this table, the synthesized CTS/GO/ZnO composite exhibits high CO 2 adsorption capacity in comparison with other adsorbents, therefore it can be used as a promising adsorbent with the most efficiency for industrial scale CO 2 capture applications.

Conclusion
In this study, some types of the CTS/GO/ZnO composite with different loading weight percent of the GO and ZnO nanoparticles were prepared, and the CO 2 adsorption experiments were conducted using resulting solid adsorbents.RSM-BBD technique was applied to investigate the effect of some parameters including GO loading,  ZnO loading, temperature, and pressure on the CO 2 adsorption capability of the adsorbents.Perturbation plots and three-dimensional response surfaces, prepared based on the RSM method, indicate that increasing the GO and ZnO loading weight percent up to 22.5 wt% and 18.3 wt% causes an improvement in the CO 2 adsorption capacity of the composite adsorbent, meanwhile increasing the GO and ZnO loading more than mentioned quantity caused decreasing in the CO 2 adsorption capacity due to pore blocking effect of the excess GO and ZnO nanoparticles.Optimization based on the RSM-BBD approach was conducted and the optimum structure of the CTS/GO/ZnO composite was obtained with the GO loading equal to 23.8 wt%, and ZnO loading equal to 18.2 wt%.The optimum composite characterization was performed using BET, XRD, FTIR, and SEM analysis.The BET analysis of the optimum CTS/GO/ZnO composite indicated that the surface area of the composite sample (SA BET = 78.75m 2 /g) is much more than pristine samples such as GO (SA BET = 36.53m 2 /g), chitosan (SA BET = 7.76 m 2 /g), and ZnO (SA BET = 17.91 m 2 /g).Additionally, isotherm modeling was performed and the results indicated the CTS/GO/ZnO composite's surface is heterogeneous and CO 2 adsorption occurred as a multi-layer adsorption process.Moreover, the fractional order kinetic model's most accurate to fitting the experimental CO 2 adsorption data exhibited the reaction order can't be an integer number and the adsorption kinetic was influenced by some factors such as heterogeneity and adsorbate-adsorbent interaction.Finally, thermodynamic parameters evaluation such as ΔH°, ΔG°, and ΔS° indicated the exothermic and dominant physisorption mechanism of the CO 2 uptake on the CTS/GO/ZnO surface.A regeneration study of the CTS/GO/ZnO composite resulted in high stability and robustness of the adsorbent after 10 cycles of adsorption-desorption regarding the minor losses in the CO 2 adsorption efficiency (around 4.36%).

Figure 1 .
Figure 1.The general procedure of GO Synthesis from graphite powder.

Figure 5 .
Figure 5. CO 2 adsorption capacity dependency to the (a) all factors, (b) GO and ZnO loading weight percent at T = 45 °C and P = 5 bar, (c) GO loading weight percent and pressure at = 45 °C and ZnO = 20 wt%, and (d) temperature and pressure when GO = 20 wt% and ZnO = 20 wt%.

Figure 8 .
Figure 8.The fitted plots of CO 2 adsorption isotherm at the temperature of 25 °C.

Figure 9 .
Figure 9. Kinetic models fitting on the experimental data of the CO 2 adsorption at the pressure of 5 bar and temperature of (a) 298 K, (b) 308 K, (c) 318 K, and (d) 328 K.

Figure 10 .
Figure 10.Van's Hoff plots of the CO 2 adsorption process at the pressure of 5 bar.

Table 4 .
Characteristics of all samples based on N 2 adsorption-desorption analysis.a BET Surface area, b Micropore volume, c Total pore volume, d Average pore diameter (measured using BET surface area and 4V/A equation).

Table 5 .
Parameters of the isotherm models at the temperature of 298 K, 308 K, 318 K and 328 K.

Table 6 .
The parameters of the kinetic models at the pressure of 5 bar.

Table 7 .
Thermodynamic parameters of the CO 2 adsorption process.

Table 8 .
Comparison of the resulting CTS/GO/ZnO composite with similar adsorbents in capturing CO 2 .