Journal of Energy Chemistry 24(2015)127–137

Synthesis of layered double hydroxides/graphene oxidenanocomposite as a novel high-temperature CO2 adsorbent

Junya Wanga, Xueyi Meia, Liang Huanga, Qianwen Zhenga, Yaqian Qiaoa,Ketao Zangb, Shengcheng Maob, Ruoyan Yanga, Zhang Zhanga,Yanshan Gaoa, Zhanhu Guoc, Zhanggen Huangd, Qiang Wanga,d∗

a. College of Environmental Science and Engineering, Beijing Forestry University, Beijing 100083, China;b. Institute of Microstructure and Property of Advanced Materials, Beijing University of Technology, Beijing 100124, China;

c. Integrated Composites Laboratory, Dan F Smith Department of Chemical Engineering, Lamar University, Beaumont, TX 77710, USA;d. State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, Shanxi, China

[ Manuscript received October 17, 2014; revised November 23, 2014 ]

AbstractIn this contribution, a novel high-temperature CO2 adsorbent consisting of Mg-Al layered double hydroxide (LDH) and graphene oxide (GO)nanosheets was prepared and evaluated. The nanocomposite-type adsorbent was synthesized based on the electrostatically driven self-assemblybetween positively charged Mg-Al LDH single sheet and negatively charged GO monolayer. The characteristics of this novel adsorbent wereinvestigated using XRD, FE-SEM, HRTEM, FT-IR, BET and TGA. The results showed that both the CO2 adsorption capacity and the multi-cycle stability of LDH were increased with the addition of GO owing to the enhanced particle dispersion and stabilization. In particular, theabsolute CO2 capture capacity of LDH was increased by more than twice by adding 6.54 wt% GO as support. GO appeared to be especiallyeffective for supporting LDH sheets. Moreover, the CO2 capture capacity of the adsorbent could be further increased by doping with 15 wt%K2CO3. This work demonstrated a new approach for the preparation of LDH-based hybrid-type adsorbents for CO2 capture.

Key wordsCO2 capture; global warming; graphene oxide; hybrid materials; recycling

1. Introduction

Recently, there is a growing consensus that global cli-mate change is occurring. As measured by Scripps Institute ofOceanography, CO2 concentration was ca. 315 ppm in March1958, which was increased to 391 ppm in January 2011, andto 398 ppm in January 2014 [1]. Now, it is widely acceptedthat the strong greenhouse gas effect significantly contributesto the global warming [2], which causes environmental prob-lems like continuous rise of water-level in sea and the increas-ing number of ocean storms and floods, etc. [3]. Reducing theemission of CO2 becomes the worldwide concerned problem,for which many techniques have been investigated, includ-ing substituting nuclear power for fossil fuels, increasing theefficiency of fossil plants, and capturing CO2 prior to emis-sion into the environment. All of these techniques have theattractive feature of limiting the amount of CO2 emitted into

the atmosphere, but each has economic, technical, or societallimitations [4].

Sorption enhanced water gas shift (SEWGS) is wellknown as a promising pre-combustion CO2 capture technol-ogy, which is a combination of WGS reaction and CO2 sorp-tion, as shown in Equation (1). Due to the existence of asolid CO2 adsorbent, the produced CO2 can be in situ cap-tured. And in the meantime, the conversion of CO and theproduction of H2 can be increased as well [5−8].

CO(g)+H2O(g)+adsorbent(s)↔adsorbent-CO2(s)+H2(g)(1)

The key to ensure the success of this process is choosinga good CO2 capturing material. Bearing in mind the limita-tions of the existing materials, there is an intense search fornew adsorbents that exhibit good CO2 adsorption capacity forSEWGS process. Layered double hydroxides (LDHs) are an

∗ Corresponding author. Tel: +86-13699130626; E-mail: qiang.wang.ox@gmail.com; qiangwang@bjfu.edu.cnThis work was supported by the Fundamental Research Funds for the Central Universities (BLYJ201509), the Fundamental Research Funds for the Central

Universities (TD-JC-2013-3), the Program for New Century Excellent Talents in University (NCET-12-0787), Beijing Nova Programme (Z131109000413013),the National Natural Science Foundation of China (51308045), and the Foundation of State Key Laboratory of Coal Conversion (Grant No. J14-15-309), Instituteof Coal Chemistry, Chinese Academy of Sciences.

Copyright©2015, Science Press and Dalian Institute of Chemical Physics. All rights reserved.doi: 10.1016/S2095-4956(15)60293-5

128 Junya Wang et al./ Journal of Energy Chemistry Vol. 24 No. 2 2015

interesting class of inorganic compounds, and in particu-lar, their derivatives produced upon-calcination have desiredproperties as CO2 adsorbents in pre-combustion capture ap-plications [8−13].

For LDHs-derived CO2 sorbents, starting from the nat-ural mineral (Mg-Al-CO3), many works have been done.For instance, the effects of divalent cations [14], trivalentcations [15], charge compensating anions [13,16,17], Mg/Alratio [18], synthesis method [19], the presence of SO2 andH2O [9,20], particle size [21,22], alkali metal (K, Cs) doping[10,16,22−27], supported LDHs [28], and operational pres-sure [29] on CO2 adsorption have all been reported. Exceptfor changing the composition of LDHs, another effective wayto improve the CO2 capture capacities is to control the particlesize of the LDHs. For the adsorbents, decreasing the particlesize or improving the specific surface area then improves thesurface adsorption activity site, which then improves the CO2capture capacities. Meis et al. [28] have studied the relation-ship between platelet sizes of LDHs and their CO2 sorptioncapacities. The results showed that the LDHs with a plateletsize between 30 nm and 2 µm have an invariant adsorption ca-pacity (∼0.1 mmol/g). However, the supported LDH sampleswith a platelet size of ∼20 nm had much higher capacities.They proposed that the CO2 sorption capacities of LDHs aredetermined by the amounts of low-coordination oxygen sitesin Mg(Al)Ox nanoparticles. Besides, Wang et al. [12] alsofound that the nano-sized spherical Mg3Al1-CO3 LDHs withan average particle size of ca. 20 nm showed an increasedCO2 capture capacity. Recently, Garcia-Gallastegui et al. [30]synthesized a LDH/graphene oxide (GO) hybrid material byco-precipitation method for CO2 capture, and they confirmedthat both the CO2 sorption capacity and recyclability were in-creased when LDHs were supported onto GO, which could beattributed to the enhanced particle dispersion.

It is believed that reducing the particle size of LDH canincrease its CO2 adsorption capacity. However, due to thelayered feature of LDHs, it is difficult to further reduce theparticle size lower than 20 nm. In this contribution, we reportthat the delamination of LDHs into single layer may be themost effective solution to this problem [31]. The delaminatednanosheets have an exceedingly high two-dimensionality witha molecular thickness, higher specific surface area, more de-gree of freedom, and more crystal orientation, etc. [31]. How-ever, because of the strong surface charge density of LDHsmonolayer [32], it is easy to restack and it is very difficult toseparate the delaminated LDHs monolayer from solution. Inorder to tackle the restacking and separation issue, negativelycharged graphene oxide single sheets are introduced to the ex-foliated LDHs dispersion solution [33]. The electrostatic in-teraction between GO and LDHs creats a layered assemblyof two sheets. The resulting nanocomposite contains exfoli-ated LDHs and GO nanosheets and its performance for CO2capture is reported for the first time. In principle, such as-sembly is beneficial for maximizing the dispersion of LDHnanosheets on the surface of GO, which can consequently in-crease the utilization rate of LDH as active CO2 adsorptionmaterials.

2. Experimental

2.1. Synthesis of samples

2.1.1. Synthesis of Mg-Al-NO3 LDHs

A salt solution (100 mL) containing a mixture of0.075 mol Mg(NO3)2·6H2O and 0.025 mol Al(NO3)3·9H2Owas added dropwise to a basic solution (100 mL) contain-ing 0.05 mol Na2NO3. The pH value of the mixture solutionwas kept constant at 10 by the addition of a NaOH solution(3.4 mol/L). The resulting mixture was hydrothermally treatedat 120 ◦C overnight. After hydrothermal aging, the samplewas filtered and washed with deionized water until pH = 7,then dried at 100 ◦C in an oven. The obtained sample wasMg-Al-NO3 LDHs.

2.1.2. Synthesis of graphite oxide

Graphite oxide was prepared from the natural graphitepower according to a modified Hummers method [34]. Thegraphitic oxide was prepared by stirring some powdered flakegraphite and sodium nitrate into the concentrated H2SO4.Then KMnO4 was added gradually with stirring while keep-ing the temperature of the mixture below 20 ◦C with ice-bath.The mixture was then stirred at 35 ◦C for 30 min, followed bythe addition of some distilled water. After another 30 min stir-ring, some distilled water was then added to terminate the re-action. Subsequently, 30 wt% H2O2 was added and the colorof the mixture changed to bright yellow. The mixture wascentrifuged and washed with 10 wt% HCl solution to removethe residual metal ions. The precipitate was then washed withMilli-Q water and centrifuged repeatedly until the solution be-came neutral, then dried at 100 ◦C in an oven. The obtainedsample was graphite oxide.

2.1.3. Delamination of LDHs

For the delamination of LDHs, 0.5 g Mg-Al-NO3 LDHswere put into 50 mL formamide, followed by magnetic stir-ring till no sediment was observed upon standing. The con-centration of delaminated Mg-Al-NO3 LDHs dispersion was10 g/L, which was marked Mg-Al-NO3 LDH-NS.

2.1.4. Delamination of graphite oxide

To delaminate graphite oxide, the product was treatedwith an ultrasonic probe at 700 W for 1 h, followed by cen-trifuging at 8000 rpm for 10 min. The delaminated graphiteoxide was obtained in the supernatant. The precipitate wasdelaminated repeatedly. The concentration of delaminatedgraphite oxide dispersion, which was 1 g/L, was determinedby drying 100 mL supernatant at 60 ◦C for two days, followedby measuring the mass using electronic balance.

Journal of Energy Chemistry Vol. 24 No. 2 2015 129

2.1.5. Synthesis of Mg-AL-NO3-NS and graphene oxidenanocomposite (Mg-Al-NO3 LDH-NS/GO)

The nanocomposite containing exfoliated Mg-Al-NO3LDH and GO nanosheets (Mg-Al-NO3 LDH-NS/GO) wasproduced by simply mixing the dispersions of each compo-nent together. Six different LDH/GO nanocomposites wereprepared by varying LDHs/GO volume ratios of 1/0.3, 1/0.5,1/0.7, 1/1, 1/5 and 1/10. The precipitation was centrifuged un-der 8000 rpm for 10 min, followed by washing with Milli-Qwater and anhydrous ethanol repeatedly to remove formamideand water completely. After drying at 100 ◦C for 24 h, theprecipitation was obtained.

2.2. Characterization of samples

Powder XRD analyses were conducted on a ShimadzuXRD-7000 X-ray diffractometer with Cu Kα radiation.Diffraction patterns were recorded within the range of2θ = 5o–65o with a step size of 0.02o. The morphologiesof samples were characterized by field emission scanningelectron microscope (FE-SEM, SU-8020). High resolutiontransmission electron microscopy (HR-TEM) images wereobtained on a JEOL 2010, operating at 200 kV. Sampleswere prepared by dispersing the sample in isopropanol using0.01 mg LDH/GO per mL of solvent, and allowing a drop todry onto a holey carbon copper grid (300 mesh, Agar Scien-tific). Fourier transform infrared spectrometer (FT-IR) exper-iments were performed using a FTS 3000 MX FT-IR spec-trophotometer. The Langmuir specific surface areas (SSA)were measured from N2 adsorption and desorption isothermsat 77 K collected from an ASAP 2020 physisorption analyzer(Micromeritics). Before each measurement, fresh and cal-cined samples were first degassed at 110 ◦C and 220 ◦C re-spectively, for overnight. The zeta potentials of delaminatedLDHs and GO were measured using zeta potential analyzer.(ZPA, Nanosizer Nano ZS, Malvern Instruments).

2.3. Evaluation of CO2 capture capacity

Thermogravimetric adsorptions of CO2 on the sampleswere measured using a Q50 TGA analyzer. Samples werefirstly calcined at 400 ◦C for 5 h in N2 (60 mL/min) beforeperforming adsorption. To avoid the error caused by mem-ory effect, the test was carried out immediately after the firstcalcination. And the samples were further calcined in situ at400 ◦C for 1 h in N2 (60 mL/min) before adsorption. CO2adsorption experiments were carried out at 1 atm with a con-stant flow of CO2 (20 mL/min). The regeneration and stabil-ity of the adsorbents were assessed by adsorption/desorptioncycling tests, in which the adsorption step was carried out at200 ◦C for 30 min with a constant flow of CO2 (20 mL/min),and the desorption was performed at 400 ◦C for 30 min with aconstant flow of N2 (60 mL/min).

3. Results and discussion

3.1. Delamination and characterization of LDH and GOnanosheets

Highly crystalline Mg-Al-NO3 LDHs were synthesizedusing coprecipitation method, followed by hydrothermal ag-ing at 120 ◦C for overnight. Powder XRD data confirmedthat the synthesized Mg-Al-NO3 LDHs was pure phase [35].The Mg-Al-NO3 LDHs material had a hydrotalcite-type struc-ture, as shown in Figure 1(a). However, after the Mg-Al-NO3LDH powder was put in formamide, and stirred for severalhours, a clear and transparent colloidal dispersion was ob-tained. The inset of Figure 1(b) shows that a clear Tyndalllight scattering was observed, which indicated the presenceof exfoliated nanosheets of the layered solid well dispersedin the formamide [36]. But for the XRD pattern of delami-nated LDH (Figure 1b), the characteristic diffractions of LDHstructure disappeared, with only a halo at 2θ = 20o–30o came

Figure 1. XRD patterns of Mg-AL-NO3 LDH (a) and delaminated LDH (b) and inset shows its Tyndall effect

130 Junya Wang et al./ Journal of Energy Chemistry Vol. 24 No. 2 2015

from formamide [32]. The absence of sharp basal peaksclearly suggested that the host sheets were not in parallel toinduce interference of the X-rays [31]. Therefore, it couldbe identified that the LDH had been delaminated into indi-vidual nanosheets. In order to further prove that LDH hasbeen exfoliated, TEM analysis was performed. Very thin LDHnanosheets with translucent plate-like morphology could beclearly seen in Figure 2(a). However, the original LDH par-ticles displayed a flower-like morphology with some degreeof amorphous conglomeration in FE-SEM, as shown in Fig-ure 2(b).

GO was produced by the exfoliation of graphite oxide,while the graphite oxide was produced by the oxidative treat-ment of pristine graphite via modified Hummers method. Be-cause of the hydrophilic nature of graphite oxide, it could beeasily delaminated in aqueous media [37]. As a result, after

a suitable ultrasonic treatment, such exfoliation can producestable dispersions of very thin graphene oxide sheets in water[38,39]. From the TEM image in the inset of Figure 3(a), itcan be seen that GO nanosheets dispersed in water were thinnanoplatelets, which exhibited a typical wrinkled morphologyof graphene oxide [40]. Figure 3 shows the XRD pattern of“dry” GO and pristine graphite. The inter-sheet distance forGO varies with the amount of absorbed water, with valuesranging from 0.61 to 0.63 nm reported for “dry” GO samples,and complete drying of GO is probably impossible [41]. Thecharacteristic diffraction peak of exfoliated GO at 11.8o (002)was observed with interlay space of 0.75 nm, which was con-sistent with previous reports [30,42,43]. The large interlayerdistance was attributed to the formation of carbonyl, carboxyland hydroxyl groups [43].

Figure 2. (a) HR-TEM image of delaminated Mg-Al-NO3 LDH, (b) FE-SEM of Mg-Al-NO3 LDH

Figure 3. (a) XRD patterns of “dry” GO, and inset shows its TEM image; (b) XRD pattern of pristine graphite

The zeta potentials of positively exfoliated Mg-Al-NO3LDH nanosheet colloidal dispersion and negatively chargedGO dispersion were measured (Figure 4a and 4b), whichwere +11.1 mV and –34.6 mV, respectively. This data sug-

gested that both of these two dispersions were stable and well-dispersed. Due to the opposite charge of these two disper-sions, Mg-Al-NO3 LDH-NS/GO nanocomposite was synthe-sized by electrostatic self-assembly of Mg-Al-NO3 LDH and

Journal of Energy Chemistry Vol. 24 No. 2 2015 131

GO nanosheets. Figure 4(c) indicates that sediment was im-mediately formed once the Mg-Al-NO3 LDH-NS dispersionwas added to the GO dispersion. The nanocomposite was eas-ily obtained by separating with centrifugation. The negativelycharged GO was complementary to the positive charge of thedelaminated LDH sheets which was likely to contribute to thestabilization of the growing nanocomposite. Chemical inter-

actions between alkaline earth metal ions and GO have beenobserved elsewhere [44], which may be particularly relevantto heterogeneous nucleation effects. In any case, the abilityof GO to stabilize the adjacent delaminated LDH sheets mustdepend on the relative charge density of the two single layeredmaterials [30].

Figure 4. (a) Zeta potential of Mg-Al-NO3 LDH-NS in formamide, (b) Zeta potential of GO in water, (c) Digital photographs of an aqueous dispersion of GO(left), an aqueous dispersion of Mg-Al-NO3 LDH-NS (middle), and a mixture of Mg-Al-NO3 LDH-NS and GO (right)

3.2. Preparation and characterization of LDH-NS/GOnanocomposites

The obtained flocculent Mg-Al-NO3 LDH-NS/GOnanocomposites were then thoroughly investigated usingXRD, FE-SEM, HR-TEM and FT-IR. Figure 5 indicates thatthe XRD patterns of Mg-Al-NO3 LDH-NS/GO samples weresimilar to that of LDH sample. The Mg-Al-NO3 LDH-NS/GOnanocomposites displayed the characteristic reflections corre-sponding to 2D hydrotalcite materials, which can be indexedaccordingly. Also it is because that GO reflection at 11.8o

is indexed to the (002) plane [43]. The basal reflection ofLDH corresponding to (003) appeared at around the same 2θof 11.3o, overlapping with GO characteristic peak. More-over, a series of well-developed (003) reflections were ob-served in low 2θ region for all the present Mg-Al-NO3 LDH-NS/GO nanocomposites, indicating the formation of a layer-by-layer ordered structure [45]. However, in high angle re-gion, the (110) peak of the LDH phase was obviously observ-able for all the present nanocomposites, clearly demonstratingthe maintenance of the LDH nanoplates upon hybridizationwith graphene nanosheets [46,47].

The nanocomposite structure, crystal shape and localcrystal structure of the as-prepared Mg-Al-NO3 LDH-NS/GOnanocomposite was examined using TEM and SAED. As pre-sented in Figure 6(a), it clearly showed a flower-like morphol-ogy that was composed of curved nanosheets. The formationof Mg-Al-NO3 LDH-NS/GO nanocomposite was evidencedby the SAED analysis, as shown in the inset of Figure 6(a).A typical electron diffraction pattern of the hexagonalLDH phase with two weak diffuse rings corresponding to the

Figure 5. XRD patterns of Mg-Al-NO3 LDH-NS/GO nanocomposite withdifferent volume ratios. (1) LDH, (2) 1 : 0.3, (3) 1 : 0.5, (4) 1 : 0.7, (5) 1 : 1,(6) 1 : 5, (7) 1 : 10

graphene lattice was observed [46]. Figure 6(b) showsthe HR-TEM image, from which the Mg-Al-NO3 LDHnanoplates could be clearly seen. Similar HR-TEM imageswith Zn-Cr-LDH/GO nanocomposites have been reported byGunjakar et al. [46].

132 Junya Wang et al./ Journal of Energy Chemistry Vol. 24 No. 2 2015

Figure 6. (a) Plane view of TEM image and SAED pattern of Mg-Al-NO3 LDH-NS/GO nanocomposite, (b) Plane view of HRTEM images of Mg-Al-NO3LDH-NS/GO nanocomposite

The morphology of the obtained Mg-Al-NO3 LDH-NS/GO nanocomposite was also monitored by FE-SEM,as illustrated in Figure 7. Based on the electrostaticallydriven self-assembly between one-atom thick GO sheets andMg-Al-NO3 LDH-NS, the porous Mg-Al-NO3 LDH-NS/GOnanocomposite was formed by cross-linked nanoflakes with ahouse-of-cards-type stacking structure of the sheet-like crys-tallites, strongly suggesting the formation of a mesoporousstructure [45−48]. This observation indicates the advantageof the self-assembly of 2D nanostructures in forming thehighly porous structure [45].

Figure 7. FE-SEM image of Mg-Al-NO3 LDH-NS/GO nanocomposite

In addition, the chemical bonding nature of GOnanosheets in the as-prepared Mg-Al-NO3 LDH-NS/GOnanocomposites was further probed using FT-IR spectroscopy.The results are shown in Figure 8. The characteristic peaksof LDH were observed for all the samples synthesized. The–OH stretch in the brucite-like layer appeared at around 3460–3480 cm−1, the vibration of angular deformation of H2Omolecules was seen at 1640 cm−1, and the Al–O and Mg–Ovibrations were found at 584–655 cm−1. Moreover, the peaks

Figure 8. FT-IR spectra of Mg-Al-NO3 LDH (1) and Mg-Al-NO3 LDH-NS/GO nanocomposites with different volume ratios of (2) 1 : 0.3, (3) 1 : 0.5,(4) 1 : 0.7, (5) 1 : 1, (6) 1 : 5 and (7) 1 : 10

at 1382 cm−1 and 839 cm−1 for Mg-Al-NO3 LDH were as-signed to the vibrations of nitrate. There was an obvious de-crease in the intensity of the 1640 cm−1 stretching vibrationpeak in Mg-Al-NO3 LDH-NS/GO compared with that in Mg-Al-NO3 LDH because the solid LDH sheets were separatedfrom nitrate in the process of exfoliation, and the negativelycharged GO instead of nitrate hybridized with the positivelycharged LDH sheets via electrostatic interactions [49]. Andall of the present Mg-Al-NO3 LDH-NS/GO composites dis-played strong IR bands corresponding to the carbon-oxygenbonds in the wavenumber region of 900–1800 cm−1 [50,51].These bands were assigned as stretching vibrations of carbon-ate [47], confirming the incorporation of carbonate ions into

Journal of Energy Chemistry Vol. 24 No. 2 2015 133

the interlayer space of Mg-Al-NO3 LDH component upon theself-assembly process [45].

3.3. CO2 capture performance of LDH-NS/GO nanocompos-ites

This type of Mg-Al-NO3 LDH-NS/GO nanocomposite isbelieved to be very promising in many applications. For in-stance, Gunjakar et al. [45] prepared similar Zn-Cr-LDH/GOnanocomposites as highly efficient photocatalysts. Herein, weare interested in using it for CO2 capture which has neverbeen investigated. LDHs have been widely identified as themost suitable CO2 adsorbents in the temperature range of200−400 ◦C for SEWGS process. Mg-Al-NO3 LDH and Mg-Al-NO3 LDH-NS/GO nanocomposites were firstly calcinedat 400 ◦C for 5 h before each CO2 adsorption test. Then thethermogravimetric sorption of CO2 on Mg-Al-NO3 LDH andMg-Al-NO3 LDH-NS/GO nanocomposites were measured at200 ◦C using a Q50 TGA analyzer. The CO2 adsorption ca-

pacities of Mg-Al-NO3 LDH and Mg-Al-NO3 LDH-NS/GOnanocomposites are list in Table 1. From which we can seethat Mg-Al-NO3 LDH-NS/GO nanocomposites had better ad-sorption capacities than that without GO. Moreover, Mg-Al-NO3 LDH-NS/GO nanocomposite with 6.54 wt% GO showedthe maximum adsorption capacity, which is more than twicelarger than the pure Mg-Al-NO3 LDH. The enhancement inadsorption capacity in the presence of GO can be attributed tothe LDH single sheets and their dispersion and stabilizationon the support [49]. With 6.54 wt% GO loading, the geomet-ric and electrostatic compatibility between LDH single sheetsand GO single sheets appeared to favor the heterogeneous nu-cleation, dispersion, and stabilization [30].

Figure 9 shows the Langmuir SSA and the pore size dis-tributions of both fresh and calcined samples. The Lang-muir SSA of fresh Mg-Al-NO3 LDH-NS/GO nanocompos-ite (76.6 m2/g) was smaller than fresh Mg-Al-NO3 LDH(132.7 m2/g). However, after calcination, the Langmuir SSAof Mg-Al-NO3 LDH-NS/GO nanocomposite was increased

Figure 9. BET isotherms and corresponding pore size distributions of fresh Mg-AL-NO3 LDH (a), fresh Mg-Al-NO3 LDH-NS/GO nanocomposites(V : V = 1 : 0.7) (b), calcined Mg-AL-NO3 LDH (c) and calcined Mg-Al-NO3 LDH-NS/GO nanocomposites (V : V = 1 : 0.7) (d)

134 Junya Wang et al./ Journal of Energy Chemistry Vol. 24 No. 2 2015

Table 1. Chemical composition and CO2 capture capacity of synthesized Mg-Al-NO3 LDH-NS/GO nanocomposites

Mg-Al-NO3 LDH-NS/GO (V/V ) GO content (wt%) CO2 capture capacity (mmol/g) CO2 capture capacity per mass of LDH (mmol/g)LDH 0 0.24 0.241 : 10 50.0 0.16 0.311 : 5 33.3 0.30 0.451 : 1 9.09 0.38 0.42

1 : 0.7 6.54 0.47 0.511 : 0.5 4.76 0.42 0.441 : 0.3 2.91 0.32 0.32

0 100 0 −

Figure 10. XRD patterns of LDH, GO, and LDH-NS/GO after being calcinedat 400 ◦C

from 76.6 m2/g to 371.4 m2/g, which was much larger thanthe calcined LDH. Moreover, both the fresh and calcined com-posites exhibited Brunauer-Deming-Deming-Teller (BDDT)-type I and IV shape of isotherms with an H3-type hystere-sis loop in the IUPAC classification, which can be regardedas an evidence for the presence of the open slit-shaped cap-illaries with very wide bodies and narrow short necks. Thistype of isotherm is frequently observed for the mesoporousstacking structure of sheet-like 2D crystallites [45]. The poresize distribution of the nanocomposite was calculated usingthe Barrett-Joyner-Halenda (BJH) method. As shown in Fig-ure 9, both fresh and calcined composites possessed a rela-tively narrow distribution of pores with an average diameter of36 to 42 nm, different from LDHs, confirming the formationof regular mesoporous materials. The formation of mesoporeswas attributable to the house-of-cards-type stacked structureof nanocomposite crystallites [45], which agreed well with itsmorphology observed in SEM and TEM analysis. It also wellexplained why the Mg-Al-NO3 LDH-NS/GO nanocompositehad good CO2 adsorption capacities.

It is well known that fresh LDH itself does not possessany CO2 capture capacity. Upon thermal treatment, LDHsgradually lose their interlayer water, and then dehydroxylateand decarbonate to a large extent, leading to the formationof a mixed metal oxide with a poorly defined 3D network[52]. The resultant partly-amorphous solid possesses a highsurface area and high surface basicity, which is suitable forCO2 adsorption [15]. The XRD patterns of calcined LDH,

GO, and LDH-NS/GO are shown in Figure 10. After be-ing calcined, LDH-NS/GO transformed into amorphous-likepericlase MgO, which is similar to pure LDH. This indicatesthat the layered structure was destroyed, leading to higherspecific surface area. Pure GO transferred into graphite af-ter being calcined at 400 ◦C, however graphite was not seenin the calcined LHD-NS/GO nanocomposite, suggesting thatGO was highly dispersed in the sample as supporting sub-strate. Therefore, we investigated the effect of pre-calcinationtemperature on the CO2 capture capacity of 6.54 wt% Mg-Al-NO3 LDH-NS/GO nanocomposite, as shown in Figure 11(a).At 300 ◦C, the CO2 capture capacity was very low, onlyca. 0.24 mmol/g. At 400 ◦C, the Mg-Al-NO3 LDH-NS/GOnanocomposite reached its maximum CO2 capture capacity

Figure 11. (a) Effect of pretreatment temperature on CO2 adsorption capac-ity of Mg-Al-NO3 LDH-NS/GO nanocomposites (CO2 capture capacity wastested at 200 ◦C), (b) Effect of adsorption temperature on CO2 adsorptioncapacity of Mg-Al-NO3 LDH-NS/GO nanocomposites

Journal of Energy Chemistry Vol. 24 No. 2 2015 135

of 0.47 mmol/g. When the calcination temperature was500 ◦C, the CO2 capture capacity started to decrease to as lowas ca. 0.37 mmol/g. The trend was the same as the LDH-CO3reported by Gao et al. [53].

To integrate CO2 adsorbents into the SEWGS pro-cess, CO2 adsorption temperatures should lie in the tem-perature window of the WGS reaction, which is normally200−400 ◦C. Therefore, the CO2 adsorption capacity per-formance at different temperatures is of great interest. Be-cause the pre-calcination temperature has a great effect onthe CO2 capture, as the results of above, we investigated theeffect of adsorption temperature ex situ calcination at 400 ◦Cfor 5 h, immediately followed by in situ calcination at 400 ◦Cfor 1 h. Figure 11(b) shows the CO2 capture capacity as afunction of adsorption temperature. The results implied themaximum CO2 capture took place at 60 ◦C (1.0 mmol/g) andwith the increase of adsorption temperature the adsorptioncapacity decreased. It is indicated that LDH/GO nanocom-posite showed good CO2 capture capacity in a wide temper-ature range. Thus, our data demonstrated that the Mg-Al-NO3 LDH-NS/GO nanocomposite is a promising CO2 adsor-bent, not only for the SEWGS process, but also for the post-combustion flue gases.

In addition to CO2 capture capacity, the continuous

Figure 12. (a) CO2 adsorption/desorption cycling test for Mg-Al-NO3 LDH-NS/GO nanocomposite, (b) calculated CO2 capture capacity of Mg-Al-NO3LDH-NS/GO nanocomposite in adsorption/desorption cycling tests

adsorption/desorption cycling stabilities of novel Mg-Al-NO3LDH-NS/GO nanocomposite (V : V = 1 : 0.7) was also eval-uated. Figure 12(a) shows the CO2 adsorption/desorption cy-cling test in a typical temperature swing adsorption (TSA)process. The adsorption was performed at 200 ◦C for 30 minwith pure CO2, and the desorption was performed at 400 ◦Cfor 30 min with pure N2. The adsorption capacity graduallydropped for the first seven cycles and became stable from theeighth cycle. From Figure 12(b), it can be seen that the ad-sorption capacity of the nanocomposite was still much higherthan that of pure Mg-Al-NO3 LDH after 22 cycles. This re-sult emphasized the potential of Mg-Al-NO3 LDH-NS/GOnanocomposite containing low amounts of GO for CO2 cap-ture processes.

3.4. Promoting ef fect of K2CO3 doping

It is well known that the CO2 sorption capacity of LDHscan be enhanced by doping with K2CO3. K2CO3 presumablyincreases the basicity of sorbents, which is favorable for thesorption of acidic CO2 [10]. Therefore, in this work, K2CO3promoted Mg-Al-NO3 LDH-NS/GO nanocomposite has alsobeen investigated. A series of samples with various K2CO3loadings of 5, 10, 15, 18 and 20 wt% were prepared usingincipient wetness impregnation (IWI) method. Before dopingwith K2CO3, the sample was firstly calcined at 400 ◦C for 5 h.Next, the K2CO3 solution was dropped into the calcined sam-ple with constantly milling. The mixture was dried at 60 ◦Cin an oven, and re-calcined at 400 ◦C for 5 h before CO2 cap-ture tests. The results indicated that the CO2 capture capacitywas increased with the increase of K2CO3 loading from 5 to15 wt%. The highest CO2 capture capacity of ca. 0.6 mmol/gwas obtained with 15 wt% K2CO3 (Figure 13). While theCO2 capture capacity started to drop with a further increaseof K2CO3 loading to 20 wt%, which might due to the poreblockage by excessive K2CO3.

Figure 13. CO2 capture capacity of Mg-Al LDH & GO nanocomposites withdifferent K2CO3 loadings

4. Conclusions

The synthesis of Mg-Al-NO3 LDH-NS/GO nanocompos-

136 Junya Wang et al./ Journal of Energy Chemistry Vol. 24 No. 2 2015

ite via electrostatic self-assembly from exfoliated LDH andGO nanosheet dispersions as a novel high-temperature CO2adsorbent was reported for the first time. Mg-Al-NO3 was de-laminated into single nanosheets in formamide and the GOwas obtained by the exfoliation of graphite oxide in waterthrough ultra-sonication. The obtained Mg-Al-NO3 LDH-NS/GO nanocomposites were thoroughly characterized usingXRD, SEM, HR-TEM, FT-IR and TGA. The result indicatedthat GO with high surface area can support the LDH singlesheet, improving the dispersion of LDH. Then a series ofMg-Al-NO3 LDH-NS/GO nanocomposites with different GOloadings from 2.9 to 50.0 wt% were prepared for CO2 capture.The Mg-Al-NO3 LDH-NS/GO nanocomposite with 6.54 wt%GO showed the maximum adsorption capacity, which is morethan twice larger than the pure LDH. The enhancement in ad-sorption capacity in the presence of GO can be attributed tothe LDH single sheets and their dispersion and stabilizationon the support. The optimal calcination temperature for thenanocomposite was determined to be 400 ◦C, which is simi-lar to the pristine LDH. CO2 capture tests indicated that Mg-Al-NO3 LDH-NS/GO nanocomposite had good CO2 capturecapacity in a wide temperature range, which suggested that itis a promising CO2 adsorbent not only for the SEWGS pro-cess but also for the post-combustion flue gases. The Mg-Al-NO3 LDH-NS/GO nanocomposite also showed good CO2adsorption/desorption cycling tests during 22 cycles. The CO2capture capacity can be further increased by doping 15 wt%K2CO3. In this contribution, we demonstrated a promis-ing approach for the preparation of highly dispersed LDHs-based CO2 adsorbents, through the exfoliation of LDHs andsubsequent electrostatic self-assembly with other negativelycharged nanosized layered materials, which is expected to in-crease its CO2 capture capacity and mechanical property.

AcknowledgementsThis work was supported by the Fundamental Research Funds

for the Central Universities (BLYJ201509), the Fundamental Re-search Funds for the Central Universities (TD-JC-2013-3), the Pro-gram for New Century Excellent Talents in University (NCET-12-0787), Beijing Nova Programme (Z131109000413013), the NationalNatural Science Foundation of China (51308045), and the Founda-tion of State Key Laboratory of Coal Conversion (Grant No. J14-15-309), Institute of Coal Chemistry, Chinese Academy of Sciences.

References

[1] http://srippscO2. ucsd. edu/data/atmospheric CO2. html[2] Pearson P N, Palmer M R. Nature, 2000, 406: 695[3] Liu J G, Chen M D, Ji Y, Zhang H. J Fuel Chem Technol (Ran-

liao Huaxue Xuebao), 2009, 37: 740[4] Burchell T D, Judkins R R. Energy Conv Manag, 1996, 37: 947[5] Balasubramanian B, Lopez Ortiz A, Kaytakoglu S, Harrison D

P. Chem Eng Sci, 1999, 54: 3543[6] Rusten H K, Ochoa-Fernández E, Lindborg H, Chen D, Jakob-

sen H A. Ind Eng Chem Res, 2007, 46: 8729[7] Lopez Ortiz A, Harrison D P. Ind Eng Chem Res, 2001, 40: 5102[8] Wang Q, O’Hare D. Chem Rev, 2012, 112: 4124

[9] Ram Reddy M K, Xu Z P, Lu G Q, Diniz da Costa J C. Ind EngChem Res, 2006, 45: 7504

[10] Wang Q, Tay H H, Chen L W, Liu Y, Chang J, Zhong Z Y, LuoJ Z, Borgna A. J Nanoeng Nanomanuf, 2011, 1: 298

[11] Evans D G, Xue D A. Chem Commun, 2006, (5): 485[12] Wang Q, Gao Y S, Luo J Z, Zhong Z Y, Borgna A, Guo Z H,

O’Hare D. RSC Adv, 2013, 3: 3414[13] Wang Q, Tay H H, Zhong Z Y, Luo J Z, Borgna A. Energy Env-

iron Sci, 2012, 5: 7526[14] Lwin Y, Abdullah F. J Therm Anal Calorim, 2009, 97: 885[15] Wang Q, Tay H H, Ng D J W, Chen L W, Liu Y, Chang J, Zhong

Z Y, Luo J Z, Borgna A. ChemSusChem, 2010, 3: 965[16] Wang Q, Wu Z H, Tay H H, Chen L W, Liu Y, Chang J, Zhong

Z Y, Luo J Z, Borgna A. Catal Today, 2011, 164: 198[17] Hutson N D, Attwood B C. Adsorption, 2008, 14: 781[18] Sharma U, Tyagi B, Jasra R V. Ind Eng Chem Res, 2008, 47:

9588[19] Zhang Z, Wang J Y, Huang L, Gao Y S, Umar A, Huang Z G,

Wang Q. Sci Adv Mater, 2014, 6: 1154[20] Ram Reddy M K, Xu Z P, Lu G Q, Diniz da Costa J C. Ind Eng

Chem Res, 2008, 47: 2630[21] Dadwhal M, Kim T W, Sahimi M, Tsotsis T T. Ind Eng Chem

Res, 2008, 47: 6150[22] Wang Q, Tay H H, Guo Z H, Chen L W, Liu Y, Chang J, Zhong

Z Y, Luo J Z, Borgna A. Appl Clay Sci, 2012, 55: 18[23] Oliveira E L G, Grande C A, Rodrigues A E. Sep Purif Technol,

2008, 62: 137[24] Ebner A D, Reynolds S P, Ritter J A. Ind Eng Chem Res, 2006,

45: 6387[25] Lee K B, Verdooren A, Caram H S, Sircar S. J Colloid Interface

Sci, 2007, 308: 30[26] Walspurger S, Boels L, Cobden P D, Elzinga G D, Haije W G,

van den Brink R W. ChemSusChem, 2008, 1: 643[27] Wu Y J, Li P, Yu J G, Cunha A F, Rodrigues A E. Chem Eng

Technol, 2013, 36: 567[28] Meis N N A H, Bitter J H, de Jong K P. Ind Eng Chem Res,

2010, 49: 1229[29] Reynolds S P, Ebner A D, Ritter J A. Ind Eng Chem Res, 2006,

45: 4278[30] Garcia-Gallastegui A, Iruretagoyena D, Gouvea V, Mokhtar M,

Asiri A M, Basahel S N, Al-Thabaiti S A, Alyoubi A O, Chad-wick D, Shaffer M S P. Chem Mater, 2012, 24: 4531

[31] Hou W Y, Kang L P, Sun R G, Liu Z H. Colloid Surf A, 2008,312: 92

[32] Kang H L, Huang G L, Ma S L, Bai Y X, Ma H, Li Y L, YangX J. J Phys Chem C, 2009, 113: 9157

[33] Abellán G, Latorre-Sánchez M, Fornés V, Ribera A, Garcı́a H.Chem Commun, 2012, 48: 11416

[34] Hummers W S, Offeman R E. J Am Chem Soc, 1958, 80: 1339[35] Wang J Y, Huang L, Gao Y S, Yang R Y, Zhang Z H, Guo Z,

Wang Q. Chem Commun, 2014, 50: 10130[36] Naik V V, Ramesh T N, Vasudevan S. J Phys Chem Lett, 2011,

2: 1193[37] Wang Y, Zhang D, Bao Q, Wu J J, Wan Y. J Mater Chem, 2012,

22: 23106[38] Stankovich S, Piner R D, Chen X Q, Wu N Q, Nguyen S T,

Ruoff R S. J Mater Chem, 2006, 16: 155[39] Stankovich S, Piner R D, Nguyen S T, Ruoff R S. Carbon, 2006,

44: 3342[40] Fu X L, Chen L X, Li J H. Analyst, 2012, 137: 3653

Journal of Energy Chemistry Vol. 24 No. 2 2015 137

[41] Stankovich S, Dikin D A, Piner R D, Kohlhaas K A, Klein-hammes A, Jia Y, Wu Y, Nguyen S T, Ruoff R S. Carbon, 2007,45: 1558

[42] Jeong H K, Lee Y P, Lahaye R J W E, Park M H, An K H, KimI J, Yang C W, Park C Y, Ruoff R S, Lee Y H. J Am Chem Soc,2008, 130: 1362

[43] Lv Z Z, Yang X, Wang E K. Nanoscale, 2013, 5: 663[44] Park S, Lee K S, Bozoklu G, Cai W, Nguyen S T, Ruoff R S.

ACS Nano, 2008, 2: 572[45] Gunjakar J L, Kim I Y, Lee J M, Lee N S, Hwang S J. Energy

Environ Sci, 2013, 6: 1008[46] Gunjakar J L, Kim T W, Kim H N, Kim I Y, Hwang S J. J Am

Chem Soc, 2011, 133: 14998[47] Li L, Ma R Z, Ebina Y, Fukuda K, Takada K, Sasaki T. J Am

Chem Soc, 2007, 129: 8000[48] Kim T W, Hur S G, Hwang S J, Choy J H. Chem Commun, 2006,

(2): 220

[49] Huang Z J, Wu P X, Gong B N, Fang Y P, Zhu N W. J MaterChem, 2014, 2: 5534

[50] Kim I Y, Lee J M, Kim T W, Kim H N, Kim H I, Choi W, HwangS J. Small, 2012, 8: 1038

[51] Lee Y R, Kim I Y, Kim T W, Lee J M, Hwang S J. Chem-Eur J,2012, 18: 2263

[52] Yang W S, Kim Y, Liu P K T, Sahimi M, Tsotsis T T. Chem EngSci, 2002, 57: 2945

[53] Gao Y S, Zhang Z, Wu J W, Yi X F, Zheng A M, Umar A,O’Hare D, Wang Q. J Mater Chem, 2013, 1: 12782

Synthesis of layered double hydroxides/graphene oxide nanocomposite as a novel high-temperature CO2 adsorbent1. Introduction2. Experimental2.1. Synthesis of samples2.1.1. Synthesis of Mg-Al-NO3 LDHs2.1.2. Synthesis of graphite oxide2.1.3. Delamination of LDHs2.1.4. Delamination of graphite oxide2.1.5. Synthesis of Mg-AL-NO3-NS and graphene oxide nanocomposite (Mg-Al-NO3 LDH-NS/GO)

2.2. Characterization of samples2.3. Evaluation of CO2 capture capacity

3. Results and discussion3.1. Delamination and characterization of LDH and GO nanosheets3.2. Preparation and characterization of LDH-NS/GO nanocomposites3.3. CO2 capture performance of LDH-NS/GO nanocomposites3.4. Promoting ef fect of K2CO3 doping

4. ConclusionsAcknowledgementsReferences