Evaluation of CO₂ interactions with S-doped nanoporous carbon and its composites with a reduced GO: Effect of surface features on an apparent physical adsorption mechanism

Abstract

Adsorption of CO₂ at temperatures close to ambient and up to 1.5 MPa was measured on nanoporous polymer-derived carbon, on its composite with reduced graphene oxide (rGO), and on their air oxidized counterparts. The isotherms were fitted by the LBET class models and the adsorption mechanisms were described in terms of parameters derived from the mentioned models including the amount adsorbed, adsorption energy distribution, maximum CO₂ adsorption energy in the first layer, heterogeneity level parameter in the first adsorbed layer, cluster heights, and changes in the volume adsorbed in the first layer upon cycling experiments. The results were analyzed in conjunction with adsorbents' porosity and surface chemistry. Even though the volume of small pores is a main parameter governing the extent of the adsorption process, surface chemistry was found as important for the energetics of adsorption and for the clustering patterns the CO₂ molecules in the pore system. Regarding the quantity and quality of surface species, the results suggest that the latter might have more dominant effect on the mechanism of the adsorption process.

Introduction

Release of such gases as CH₄ or CO₂ to the atmosphere is an important issue facing contemporary society [1]. The main source of these gases is the consumption of fossil energy. To decrease the environmental effects of CO₂, its capture technologies are under development. Some of these methods are based on adsorption processes. Such adsorbents as zeolites [1], [2], [3], functionalized mesoporous silica [4], [5], carbon nanotubes [6], metal oxides [7], Metal Organic Frameworks (MOFs) [1], [8], [9], [10] or aminated graphite oxides and their composites with Cu-based MOF [11], activated carbon xerogels [12] and activated carbons [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27] are used. In many cases basic groups (as amines) [4], [5], [24], [25], or heteroatoms providing basicity, are introduced to the surfaces [18], [19], [28].

Even though the capacity of some MOFs for CO₂ adsorption is high [8], [9], [10], [11], activated carbons are still extensively investigated as CO₂ separation and storage media owing to their high porosity, inertness, and a high working adsorption capacity [13], [17], [21], [23], [29]. The latter is related to the easiness of the CO₂ recovery from their pore system. The volume of pores smaller than 1 nm was found is crucial to provide a high CO₂ adsorption capacity, especially at ambient conditions (kinetic diameter of the CO₂ molecule is 0.33 nm) [1], [13], [14], [15], [16]. Since this volume can also be linked to adsorbents' surface area, various method including soft- [15] and hard-templating [30] or KOH activation have been applied [17], [31] to increase the porosity of carbons. The CO₂ adsorption capacities of the highly nanoporous N-doped carbons synthesized by KOH activation of urea-modified petroleum coke was reported to range from 3.69 to 4.40 mmol/g and from 5.69 to 6.75 mmol/g at 298 K and 273 K, respectively, under 0.1 MPa [31]. At 0.1 MPa and 298 K 4.8 mmol/g was measured on the microporous carbons synthesized using carbon-containing metal salt [32] and 8.05 mmol/g at 0.1 MPa and 273 K on activated carbon spheres [33]. An introduction of surface basicity in the form of nitrogen functional groups [24], [25], [31] also increases CO₂ adsorption on carbons owing to acid–base interactions [19], [24], [28]. On such carbons CO₂ adsorption capacity of 3.13 mmol/g was measured on the porous carbon monolith synthesized using l-lysine [24] and 4.24 mmol/g on the N-enriched activated carbons obtained from a biomass waste by KOH activation under 0.1 MPa at 298 K [19].

Recently, it has been indicated that CO₂ adsorption on N-doped carbon is affected by hydrogen bonding interactions [19]. An enhanced adsorption of CO₂ was also reported by Xia and coworkers on sulfur doped carbon obtained within the structure of zeolite EMC-2. They reported 59 kJ/mol as a heat of adsorption at low surface coverage [20]. The high capacity on chemically activated reduced-graphene oxide/poly-thiophene composite was linked to the “oxidized S-content” and the presence of pores similar in size to CO₂ molecule [22]. Following this, we have studied the adsorption of CO₂ on S-doped polymer-derived nanoporous carbons and its composites with reduced graphene oxide [21]. The results indicated the complexity of the adsorption process and a significant role of polar sulfur-containing groups in the enhancement of CO₂ adsorption. On these materials, the high pore space utilization by CO₂ molecules was found. Specifically it was proposed that acid–base interactions of CO₂ in small pores with sulfur incorporated into aromatic rings of the pore walls and polar interactions of CO₂ with sulfoxides, sulfones and sulfonic aids, along with hydrogen bonding of CO₂ with acidic groups on the surface contribute to this effect.

Taking into account our recent findings, the objective of this paper is to further evaluate CO₂ adsorption on sulfur-containing nanoporous carbon/reduced graphene oxide composites. In the approach used we analyze the CO₂ interactions with the surface using the LBET class model [34], [35]. This methodology is based on the physical adsorption phenomenon. We have chosen this analysis route because in the vast majority of published reports on CO₂ sequestration on activated carbons only physical adsorption is addressed. By evaluation of adsorption thermodynamics from the isotherms measured at high pressure in the cycling mode the subtle and detailed changes in the interactions of CO₂ with adsorbents' surfaces are pointed out and discussed in terms of the adsorption mechanisms and the possible contributions of chemical interactions to CO₂ adsorption process. One has to remember that the results we present should not be considered as axioms; instead, they are to be approached as an expression of a certain – substantial – dose of probability that the picture of the analyzed structure is what it is, and nothing else.