A novel CO2 utilization technology for the synergistic co-production of multi-walled carbon nanotubes and syngas

Dry reforming of methane (DRM) is a well-known process in which CH4 and CO2 catalytically react to produce syngas. Solid carbon is a well-known byproduct of the DRM but is undesirable as it leads to catalyst deactivation. However, converting CO2 and CH4 into solid carbon serves as a promising carbon capture and sequestration technique that has been demonstrated in this study by two patented processes. In the first process, known as CARGEN technology (CARbon GENerator), a novel concept of two reactors in series is developed that separately convert the greenhouse gases (GHGs) into multi-walled carbon nanotubes (MWCNTs) and syngas. CARGEN enables at least a 50% reduction in energy requirement with at least 65% CO2 conversion compared to the DRM process. The second process presents an alternative pathway for the regeneration/reactivation of the spent DRM/CARGEN catalyst using CO2. Provided herein is the first report on an experimental demonstration of a 'switching' technology in which CO2 is utilized in both the operation and the regeneration cycles and thus, finally contributing to the overall goal of CO2 fixation. The following studies support all the results in this work: physisorption, chemisorption, XRD, XPS, SEM, TEM, TGA, ICP, and Raman analysis.

www.nature.com/scientificreports/ its concept are provided in supplementary Section 1. The second process, on the other hand, utilizes CO 2 during the catalyst regeneration cycle for the removal of surface carbon. A combination of CARGEN and the new CO 2 regeneration process forms a switching technology that utilizes CO 2 during both operations as well as the regeneration cycle.

Results and discussion
The thermodynamic equilibrium assessment of the CARGEN reactor shows that solid carbon is a favorable product in a temperature range of 420 °C to 600 °C as shown in Fig. 2a. The feed gas composition was CH 4 /CO 2 / O 2 = 1/0.6/0.1, that closely resembles flue/landfill/biogas composition 14 for ultimate FTS application 15 . However, this feed ratio may be changed depending upon available feed gas at site to get different results, and an optimization can also be done to achieve a desired quantity of carbon, or achieve a specific syngas ratio. Figure 2b shows the experimental conversion profile of the CARGEN reactor at a temperature of 550 °C, which closely matches with the theoretical results reported in Fig. 2a-ii. It should be noted that all the experimental studies in this work were conducted using commercially available 20% Ni/Al 2 O 3 catalyst, as it is the most widely used reforming catalyst. However, other materials like Fe, Co, Mo, Ru, Pt, Rh etc. supported with SiO 2 , TiO 2 etc. could be used, but will result in higher cost and sintering challenges. It can be seen in Fig. 2 that experimental results closely match thermodynamic results within a margin of 10-15%. It is important to note that the experimental conversions are lower compared to the estimated thermodynamic conversion due to kinetics and mass transfer limitations that arise during the catalytic reaction 16 . A closer examination of the actual spent catalyst from this experimental run using SEM and TEM (Fig. 2c) reveals the formation of MWCNTs. Raman analysis of a 30-min CH 4 pyrolysis experiment on 20 wt% Ni/Al 2 O 3 at 700, 600, 500, and 400 °C showed that the MWCNT formation regime is gradually diminished with a decrease in temperature, and only nano-carbon remains at 400 °C. These results demonstrate that MWCNT begins forming after between 400 and 500 °C. Figure 3 illustrates the corresponding ex-situ micro-Raman spectra, which indicates that the quality of the MWCNT gets better (D/G ratio decreases while G'/G ratio increases) with an increase in temperature. It should be noted that the operational window of CARGEN is in the temperature range of 400-600 °C, which is the most suitable condition for MWCNT formation per Raman results. High-value products coming from GHGs have attracted significant media attention [17][18][19][20] as well as interest from global energy corporations due to its encouraging preliminary economic data 21 . The commercial value of high-quality MWCNTs can be in the range of 500-10,000 USD/kilogram 22 . Although a detailed techno-economic assessment is underway, the preliminary life cycle assessment (LCA) study revealed that both the CO 2 footprint and the operational cost of the CARGEN process are approx. 40% that of conventional ATR processes 21 .
A thermo-gravimetric analysis (TGA) experiment coupled with material balance on residual gas analyzer (RGA) data enabled direct assessment of the carbon formation rate and feed conversions as a function of TOS. These results are provided in Fig. 4 and detailed calculation steps of material balance in supplementary Sect. 3. Around 20 mg carbon formation was observed in 138 min TOS at 550 °C on 20 mg of commercial 20 wt% Ni/γ-Al 2 O 3 catalyst entailing a remarkable carbon formation rate of 0.00722 mg MWCNT /mg Cat. /min. It should be reiterated that the target for CARGEN reactor (the first reactor in CARGEN technology) is only to produce solid carbon. The second reactor on the other hand, which is a combined reformer offers a great flexibility in producing a desirable syngas ratio as discussed in detail in previous publications 12, 23 .  www.nature.com/scientificreports/ Subsequent to the production of milli-grams of carbon in TGA, a larger batch of solid carbon was produced in a chemical vapor deposition (CVD) setup to produce multi-grams of carbon. One of the sample produced from this experiment was at ~ 80 wt% carbon purity (and remaining catalyst weight). In order to deduce the quality of the as-produced samples, a TGA air oxidation experiment was then conducted. The protocol for this test involved air combustion of carbon at 50 mL/min in the TGA on ca. 10 mg of the sample under a temperature ramp of 10 °C/min from room temperature to 400 °C, and at 5 °C/min from 400 °C to 800 °C. Figure 5a,b shows that only < 1% of the sample was oxidized between 200 and 500 °C, which is the literature reported range of amorphous carbon [24][25][26] . As the combustion occured in the range of 480 to 700 °C temperature indicates that crystalline form of carbon was primarily present, of which MWCNTs are predominant category [24][25][26] . Consolidating the TGA information to the SEM and TEM micrographs infers that most of the carbon belongs to MWCNT category with diameters in the range of 50 to 100 nm and length in the range of 10 to 30 µm as shown in Fig. 5. Moreover, the HR-TEM images at 10 nm scale bar shows that a tip-growth mechanism of CNT production would have taken place during the CARGEN process.
One of the primary challenges in any industrial process that involves carbon formation like DRM, or catalytic cracker units is catalyst deactivation. Due to this, the entire plant has to be taken into maintenance, and the bulk of carbon has to be removed by either scraping or by using other carbon removal techniques like etching, sonication, etc. However, even after the removal of bulk carbon, the catalyst nano sites are still covered with surface carbon 27 , which needs to be removed to reactivate the catalyst. In this paper, a novel approach of a singlestep catalyst regeneration procedure, applicable for any process that suffers from catalyst deactivation via coke formation 17 is presented. The said process utilizes CO 2 as a soft oxidant 28 for the removal of surface carbon that deactivated the catalyst. This presents a great improvement over conventional procedure, as the conventional procedure requires two steps-first, oxidation with O 2 for carbon removal as CO 2 , followed by the reduction of nickel oxide with H 2 to produce active nickel. The new procedure is precisely similar to temperature-programmed oxidation (TPO) using O 2 but utilizes CO 2 in its place. The reaction is proposed to happen via reverse-boudouard route as follows: This reaction was first tested using thermodynamic equilibrium assessment at temperatures in the range of 650 to 800 °C, followed by laboratory proof of concept study in a flow-through reactor. Under a temperature ramp from 650 to 800 °C, when the CO 2 gas was passed through the spent DRM catalyst bed, it reacted with surface carbon and produced CO (which could be used as a precursor for several industrial chemicals). A subsequent DRM immediately after the regeneration showed activity without the need for a second reduction step by H 2 . Figure 6 shows the results of this test for three cycles that ensures repeatability. A combination of the operation and regeneration cycle as in Fig. 6 demonstrates an interesting switching process wherein uninterrupted CO 2 utilization happens during both the cycles.
In order to evaluate the CO 2 oxidation capability, pre-reduced 20 wt% Ni/Al 2 O 3 catalyst was heated continuously from room temperature to 700 °C in CH 4 , then CH 4 supply was cut off and one CO 2 pulse was injected for 5 s and for 40 s respectively in two independent experiments. Figure 7 shows the SEM and Raman results of this study. It was observed that the quality of the CO 2 -treated MWCNTs becomes worse (D/G ratio increases while G'/G ratio decreases). It seems that the more defective carbon structures are preferentially removed from the Ni surface while the robust ones remain on the surface with more defects.
Results of XPS analysis of a coked commercial catalyst showed that the surface carbon atom percentage was about 75%. On the other hand, the surface carbon coverage of post-CO 2 TPO treated catalyst was about 5%, while for post O 2 TPO was 4.5%. This shows the equivalency of both the regeneration techniques in removing surface carbon. It should be emphasized that even though pure CO 2 is required for CO 2 TPO, it has a greater importance for DRM since its use will enable overall CO 2 fixation.
(1) www.nature.com/scientificreports/ In terms of sustainability benefits, it was observed that, the new CO 2 TPO process is capable of converting at least 3 kg of CO 2 per kg of surface carbon removed, while O 2 TPO leads to about 0.5 kg of CO 2 emission per kg of surface carbon. Detailed calculation steps are provided in Sect. 5 of the Supplementary Information. It should be emphasized that the use of CO 2 regeneration technique also saves the active catalyst (Ni) from undergoing cycles of oxidation (forming inactive NiO) and reduction (forming Ni) as in the case of conventional regeneration technique.

Conclusion
In conclusion, the reported work experimentally demonstrates the formation of MWCNTs in the novel CAR-GEN process. Additionally, a unique single step procedure to regenerate coked spent DRM/ CARGEN catalyst using CO 2 as an activation media is also reported. A combination of the two processes (DRM/CARGEN + CO 2 regeneration) provides a novel switching technique that utilizes CO 2 during both operations as well regeneration cycles, proving to be a strong candidate for the commercialization of the DRM/CARGEN process that enables overall CO 2 fixation.