Introduction

The extensive emission of greenhouse gases, specifically methane (CH4) and carbon dioxide (CO2), is a primary cause of global warming1,2. Catalytic dry reforming of methane (DRM) offers an environmentally friendly and viable route for large-scale greenhouse gases utilization (\({{{\mbox{CO}}}}_{2}+{{{\mbox{CH}}}}_{4}\to 2{{\mbox{CO}}}+2{{{\mbox{H}}}}_{2}\), \(\Delta {{{\mbox{H}}}}_{298{{\mbox{K}}}}=+ 247{{\mbox{kJ}}}{{{\mbox{mol}}}}^{-1}\)), converting two gases into valuable chemical feedstock3,4. Unfortunately, DRM is an endothermic reaction, requiring high temperatures (700−1000 oC) to overcome unfavorable dynamics and thermodynamics5. At present, thermocatalysis driven by external heating was limited by high temperature and high pressure, resulting in high energy consumption. Photothermal catalysis occurred at a relatively low temperature, but the catalytic efficiency possessed major room for improvement6. Furthermore, due to the high-temperature reaction, coke deposition and concomitant catalyst deactivation during DRM are inevitable problems7. Consequently, it is imperative to explore novel catalytic modalities that combine mild reaction conditions with high catalytic performance.

The thriving exploration of catalysts for DRM has been witnessed in the past decades. Especially, the Group VIII metals, whether precious metals (Rh, Ru, Pt)3,8,9,10 or non-precious metals (Ni, Co, Fe)11,12,13, have been proven to possess catalytic activity in DRM reactions. Transition metal carbides, such as molybdenum carbide (Mo2C) and tungsten carbide (WC)14,15, exhibited remarkable catalytic activity in methane aromatization reaction16,17. However, under thermal catalytic reaction condition, Mo2C possesses limited methane cracking ability in DRM reaction18, which leads to over-oxidation and eventually the deactivation of the catalyst in carbon-deficient environments19,20,21. Hence, improving the methane cracking capacity is of great significance to the Mo2C catalyst.

Pulsed laser is a type of high-energy density light produced by the process of stimulated emission and light amplification22. Owing to its narrow pulse width and high-energy density, the interaction between pulsed laser and materials can give rise to some fascinating effects23,24. The local thermal effect produced by pulsed laser can trigger photothermochemical reactions25,26,27. More importantly, high-energy electrons generated by an infocus pulsed laser can cause the breakdown of gas molecules, resulting the formation of highly active plasma from the gas reactants28,29,30. Studies have confirmed that plasma enabled a thermodynamic-limited reaction to occur with a fast reaction rate at low temperatures31,32. It is worth noting that, unlike pulsed laser, the continuous wave (CW) laser usually produced a high temperature instead of plasma due to the efficient photothermal conversion via non-radiative relaxation33. Consequently, it is rational to speculate that pulsed laser can be applied in DRM as both heat and plasma sources, which represent an efficient supplementary to the deficiency of Mo2C.

Here, we reported that by using the pulsed laser to drive the DRM reaction at relatively mild reaction condition. Record-high activities of H2 (14300.8 mmol·h−1·g−1) and CO (14949.9 mmol·h−1·g−1) were reached over a simple Mo2C catalyst by this pulsed laser-driven DRM reaction. The thermal and plasma effect on Mo2C catalyst induced by a 16 W pulsed laser were identified to be critical for the pulsed laser-driven DRM reaction. More importantly, the pulsed laser-induced CH plasma (*CH) avoided the step-by-step dehydrogenation of CH4, which boosted methane cracking on the surface of the catalyst, established a surface carbon-oxygen equilibrium and protected Mo2C from over-oxidation, which is essential for the high activity and stability of the catalyst. This finding holds significant implications for expanding the research concepts within photothermal catalytic systems.

Results

Characterization of Mo2C/BaSO4 tablet

Mo2C nanosheets were synthesized with a rough surface, as depicted in Fig. 1a. The crystal phase of Mo2C was confirmed by X-ray diffraction analysis (Fig. 1b), which exhibited distinct peaks of β-Mo2C. The HRTEM image (Fig. 1c) showed lattice spacings of 0.228 nm (assigned to (101) planes) and 0.237 nm (assigned to (002) planes) for β-Mo2C, confirming the successful synthesis of β-Mo2C. Before the laser-catalytic DRM, the Mo2C/BaSO4 tablet was pressed by Mo2C and BaSO4 powders (Fig. 1d). The catalytically inert BaSO4 was selected as the substrate and didn’t affect the catalytic activity of Mo2C (Supplementary Fig. 1, 2). The top Mo2C layer exhibited a thickness of ~36 μm, as evident from the cross-sectional SEM image in Fig. 1e. The corresponding EDS mapping of Mo, C, Ba, and O elements in Fig. 1f also confirmed successful construction of razor-thin Mo2C layer on BaSO4 substrate.

Fig. 1: Characterization of Mo2C/BaSO4 tablet and laser irradiation effects.
figure 1

a SEM image, b XRD pattern, and c HRTEM image of Mo2C powder. d Photograph of tablets pressed by Mo2C and BaSO4 powders. e SEM image and f corresponding EDS element mapping of the cross-section of Mo2C/BaSO4 tablet. g Schematic diagram of laser-catalytic DRM reaction via Mo2C/BaSO4 tablet. h Schematic diagram of the Mo2C/BaSO4 tablet illuminated by pulsed laser and i SEM images of the circular spot of Mo2C/BaSO4 tablet irradiated by pulsed laser (the laser irradiates a specific area in a line scanning mode, taking a circular area with a diameter of 0.94 mm as an example).

The experimental equipment of laser-catalytic DRM is shown in Fig. 1g and Supplementary Fig. 3a. During the laser-catalytic DRM, the obtained Mo2C/BaSO4 tablet was placed at the bottom of the quartz reactor. A 16 W fiber optic laser with a laser wavelength of 1064 nm and a pulse duration of 100 ns was used to irradiate the Mo2C/BaSO4 tablet. As depicted in Fig. 1h, laser irradiation in a small area (demonstrated by a circular area of 0.94 mm diameter) in a line sweep was performed. The pulsed laser was concentrated on a tiny point on the surface of Mo2C/BaSO4 tablet. SEM image in Fig. 1i exhibited that the irradiated tiny area formed a regular overheating morphology due to the localized high temperature generated by the laser. As a consequence, it was reasonably speculated that the effective mass of the catalyst in the laser-catalytic DRM process was only in the region treated by the laser spot.

Laser-catalytic DRM performance

To gain more insight, the laser-catalytic DRM performances of Mo2C/BaSO4 tablets with different tablet areas of 12.57, 7.07, 3.14, and 0.79 mm2 were discussed (Fig. 2a and b, Supplementary Fig. 4 and Fig. 5, Supplementary Table 1-3). The product rate of H2 (6.864 ~ 7.330 mmol h−1) and CO (7.176 ~ 7.611 mmol h−1) did not change significantly when the tablet area was not less than 3.14 mm2. Thus, it is indicated that laser-catalysis occurs in a tiny area, which is consistent with the SEM image of the laser-irradiated area (Fig. 1i). However, when the tablet area was reduced to 0.79 mm2, the thin Mo2C layer was easily stripped by the laser resulting in a decrease in the catalytic reaction rate after 20 mins irradiation (Fig. 2c). Consequently, it is deduced that an overly small catalyst area is not beneficial for maintaining catalyst stability, and a tablet area of 3.14 mm2 appears to be optimal. Additionally, we discussed the laser-catalytic DRM performances of Mo2C/BaSO4 tablets with different Mo2C thicknesses (36, 54, 110, 220, and 315 μm). Remarkably, the product rates of H2 (6.998 ~ 6.276 mmol h−1) and CO (7.176 ~ 6.201 mmol h−1) did not exhibit significant changes as the Mo2C thickness increased. This further underscores that the actual catalytic dose interacting with the laser remains very small, regardless of whether it operates at the area or depth level.

Fig. 2: Performance evaluation of laser-catalytic DRM.
figure 2

a, b Laser-catalytic DRM performances and c catalytic stability of Mo2C/BaSO4 tablets with different areas (laser-catalysis in closed system, laser output power: 16 W, infocus mode, CO2:CH4:Ar = 47.5%:47.5%:5%). Error bars represent standard deviation. d DRM performance in laser-catalysis, thermocatalysis (temperature: 900 °C), and Xenon lamp-driven photothermal catalysis (optical power density: 3 W cm−2). Error bars represent standard deviation. e Catalytic stability under CO2 (CO2:Ar = 95%:5%), CH4 (CH4:Ar = 95%:5%), and CO2 + CH4 (CO2:CH4:Ar = 47.5%:47.5%:5%) atmosphere. f Laser-catalytic DRM performances of Mo2C/BaSO4 tablet in different proportions of CO2 and CH4. Error bars represent standard deviation. g The comparison of product yield for laser-catalytic DRM in this work and previously reported results.

To elucidate the advantages of laser-catalytic DRM, the product yields of thermocatalytic DRM, as well as photothermal catalytic DRM driven by Xenon lamp, were investigated (Fig. 2d). Specifically, the optimal reactive activity of thermocatalytic DRM at 900 °C with yields of H2 (29.4 mmol h−1 g−1) and CO (97.4 mmol h−1 g−1) was achieved (Supplementary Fig. 7), which was in good accordance with previous studies18,34. The performance of Xenon-lamp-driven photothermal catalytic DRM was also unsatisfying, only 0.3 mmol h−1 g−1 (yield of H2) and 0.8 mmol h−1 g−1 (yield of CO) were detected (Fig. 2d), which was attributed to the low photothermal temperature (maximum 361 °C at 3 W cm−2 in Supplementary Fig. 8) and poor methane activation capability of low-energy photons. The yields of the continuous wave (CW) laser-driven DRM were substantially lower in comparison to those of the pulsed laser, yielding only 5099.8 mmol h−1 g−1 (for H2) and 8000.6 mmol h−1 g1 (for CO). Remarkably, the ultrahigh yields of H2 of 14300.8 mmol h−1 g−1 and CO of 14949.9 mmol h−1 g−1 via pulsed laser-catalytic DRM without the assistance of external heating source were obtained (Fig. 2d and Supplementary Fig. 9), which were about 486 times (for H2) and 153 times (for CO) higher than those of thermocatalysis and about 56975 times (for H2) and 18143 times (for CO) higher than those of Xenon lamp-driven photothermal catalysis. Even without comparing the mass of the catalysts, the production rate (6.864 mmol h−1 for H2 and 7.176 mmol h−1 for CO) of the laser-catalysis system with less catalyst (0.48 mg Mo2C) was still better than that (1.471 mmol h−1 for H2 and 4.870 mmol h−1 for CO) of the thermocatalysis system (50 mg Mo2C), suggesting the superior intrinsic catalytic activity of laser-driven DRM (Supplementary Fig. 10). Under the same small amount (0.48 mg), the production rate of H2 (0.009 mmol h−1) and CO (0.037 mmol h−1) and yields of H2 (17.7 mmol h−1 g−1) and CO (77.2 mmol h−1 g−1) in thermocatalysis were also much lower than those of laser catalysis (yields of H2 of 14300.8 mmol h−1 g−1 and CO of 14949.9 mmol h−1 g−1) (Supplementary Fig. 11). This confirmed that thermocatalysis could not achieve high catalytic activity under a small amount of catalysts, only laser catalysis can achieve high DRM activity under such a small amount of catalysts, which embodied the advantage of laser catalysis.

To illuminate the pulsed laser-catalytic DRM reaction of the Mo2C/BaSO4 tablet, the products under different reaction atmospheres were detected at the same pulsed laser irradiation condition. As shown in Fig. 2d, under CO2 atmosphere, only CO with a low yield of 9707.9 mmol h−1 g−1 was detected. Under CH4 atmosphere, only H2 was detected, and with the high H2 yield of 12161.3 mmol h−1 g−1, which verified the strong methane cracking capacity of pulsed laser with Mo2C as laser absorber and catalyst. Whether in single CO2 or CH4 atmosphere, the catalytic reactions were unstable, with significant attenuation observed, from 9.860 to 2.608 mmol h−1 for H2 and from 3.288 to 0.550 mmol h−1 for CO. In contrast, under CO2/CH4 (1:1) atmosphere, the laser-catalytic DRM with the Mo2C/BaSO4 catalyst exhibited relatively satisfactory stability for 500 mins (Fig. 2e), even more than 50 hours (Supplementary Fig. 12). The XRD patterns (Supplementary Fig. 13) and HRTEM images (Supplementary Fig. 14) revealed that Mo2C phase remained intact under CO2: CH4 feed ratio of 1:1 atmosphere. Although isotopic labelling experiments confirmed that the carbon in Mo2C may participate in the carbon cycle of DRM reaction (Supplementary Fig. 15), the stable existence of the final Mo2C phase and the long-term catalytic stability indicated the establishment of the C-O equilibrium reaction.

In addition, the DRM yields of Mo2C/BaSO4 tablet in varying CO2/CH4 ratios from 2:1 to 1:2 also confirmed the strong methane cracking capacity of pulsed laser. Increasing the CH4 amount during the laser-catalytic DRM resulted in heightened H2 production, as shown in Fig. 2f. In contrast, even if the proportion of CH4 in the reactants was increased, the product of thermocatalytic DRM remained predominantly CO (Supplementary Fig. 16–18), demonstrating the weak methane cracking capacity of Mo2C. Raman and XPS results of Mo2C after laser-catalytic and thermocatalytic DRM also confirmed the above conclusion. The C consumption (Supplementary Fig. 19a) and pronounced oxidation of Mo2C (Supplementary Fig. 19b), arising from its weak methane cracking capacity during thermocatalytic DRM, led to poor catalytic stability. In previously reported results, loading methane activation sites on Mo2C is a common catalyst design strategy, such as creating Metal-MoxC dual-site catalysts (such as Ni-Mo2C, Co-Mo2C, etc.)35,36,37,38. Herein, with the aid of the pulsed laser, the cracking of CH4 was significantly boosted and C-O equilibrium on Mo2C surface was established, yielding excellent DRM activity and stability with pure Mo2C as the catalyst. The laser-catalytic DRM reaction of Mo2C/BaSO4 tablet possessed the high activity of 14300.8 mmol h−1 g−1 (yield of H2), 14949.9 mmol h−1 g−1 (yield of CO) and stability (No significant decay over 50 h), which were superior to recently reported DRM results, as summarized in Fig. 2g and Supplementary Table 43,10,39,40,41,42,43,44,45,46,47. The mass activity of the laser-catalytic DRM without external heating using Mo2C as a catalyst was the highest value up to now in the fields of thermocatalysis and photothermal catalysis.

Laser-induced localized high temperature

The UV-Vis-NIR absorption spectra depicted in Fig. 3a for both Mo2C and BaSO4 revealed that Mo2C possessed an ideal capacity for laser-induced heat generation due to its effective absorption at 1064 nm wavelength. Due to the negligible absorption ability to laser, BaSO4 could only be heated up to 60 °C under laser irradiation (Supplementary Fig. 20), which caused negligible DRM performance. The radial temperature distribution on the Mo2C/BaSO4 tablet was illustrated in Fig. 3b during laser irradiation, revealing a gradual decline from the central focal point outward. The laser’s focused temperature reached 772 °C, which thermodynamically sufficed to propel the DRM reaction. The local high temperature generated by laser irradiation on Mo2C/BaSO4 tablet is one of the prerequisites for DRM reaction. The incident laser was absorbed within the skin depth of the Mo2C surface, instantaneously transforming into heat within sub-nanosecond intervals. Of course, the local temperature of the laser is currently difficult to measure accurately, which is a problem for the industry. The current temperature test is the measurement of the average temperature within a certain region, the actual local temperature value may be different from the measured temperature, but we are unified test conditions and test equipment, to ensure that the temperature trend is accurate.

Fig. 3: Localized high temperature and corresponding DRM performance under different laser irradiation modes.
figure 3

a UV-Vis-NIR absorption spectra of Mo2C and BaSO4. b Temperature distribution of laser irradiation on Mo2C/BaSO4 tablet. c The temperatures and d the laser-catalytic DRM performances of the Mo2C/BaSO4 tablet by different pulsed laser powers (4 W, 8 W, 12 W, 16 W, 20 W). Error bars represent standard deviation. e The temperatures and f the laser-catalytic DRM performances of the Mo2C/BaSO4 tablet by pulse laser with different defocusing amounts. Error bars represent standard deviation.

The effect of laser power on both temperature and product yields in laser-catalytic DRM was explored, as presented in Fig. 3c, d. With the pulsed laser power ascending from 4 W to 8 W, 12 W, and 16 W, the temperature on the Mo2C/BaSO4 tablet escalated from 542 °C to 550 °C, 619 °C, and 772 °C, respectively. The according laser-catalytic DRM reaction performance of H2 and CO yields was improved from 0/45.2 mmol h−1 g−1 to 132.0/755.6 mmol h−1 g−1, 645.5/2748.7 mmol h−1 g−1 and 14300.8/14949.9 mmol h−1 g−1. In addition, no noticeable difference in temperature was detected under 16 W (772 °C) and 20 W (784 °C). However, the yield of H2 was obviously improved from that of 16 W (14300.8 mmol h−1 g−1) to that of 20 W (19736.0 mmol h−1 g−1), implying the thermal effect was not the sole driver of the enhanced performance of laser-catalytic DRM. In particular, the H2/CO molar ratio was significantly increased with increased laser power, suggesting that pulsed laser boosted the CH4 cracking capacity. Furthermore, the effect of different defocusing amounts (Supplementary Fig. 31 and Supplementary Table 5) on laser-catalytic DRM performance also confirmed the above speculation. While the temperatures remained relatively consistent in both under focus (defocusing amount = 20, 15, 10, 5 mm) and in focus modes (defocusing amount = 0 mm), ranging from 746 ~ 793 °C (Fig. 3e and Supplementary Table 6), the laser-catalytic DRM performances exhibited significant discrepancies, as shown in Fig. 3f. In the in focus mode, the H2/CO yields of 14300.8/14949.9 mmol h−1 g−1 were significantly higher than those in the under focus modes (3212.8/4953.9, 4681.0/7654.8, 6813.9/10094.0 and 6160.8/9389.7 mmol h−1 g−1). These results confirmed that the pulsed laser in the in focus mode caused the heightened DRM.

Laser-induced plasma of CH4 and CO2

The free electrons can be accelerated by bremsstrahlung absorbing energy of pulsed laser, which increases the electron density like a cascade. Until they have enough energy to collide and ionize the surrounding gas to generate gas plasma, which can be carried out even at low pulsed laser power48,49. It is plausible that this laser-induced plasma effect could potentially influence the DRM reaction. Figure 4a presents the excitation spectra of plasmas generated during the DRM process under three conditions: Infocus-Pulsed Laser, Infocus-CW Laser, and Underfocus-Pulsed Laser. With an in focus pulsed laser applied to the Mo2C/BaSO4 tablet, distinct peaks in the range of 350 to 604 nm corresponding to CO2 and CH4 plasma were detected30. Conversely, no plasma generation on BaSO4 under pulsed laser in focus mode confirmed the weak interaction between laser and BaSO4 (Supplementary Fig. 21). In contrast, neither CO2 nor CH4 plasma was observed under CW laser infocus mode (Fig. 4c) or pulsed laser under focus mode (Fig. 4d). Only the envelope peak of thermal radiation (600 ~ 800 nm) was detected.

Fig. 4: Demonstration of laser-induced plasma.
figure 4

a Excitation spectra of laser-catalytic DRM under different catalytic conditions. Time-dependent dynamic spectra generated by b pulsed laser (Infocus), c CW laser (Infocus), and d pulsed laser (Underfocus). e Schematic diagram of the pulsed laser induced plasma on Mo2C. simulated spatial density distributions of f electrically-charged particles including electron, g *CH, and h *CO of pulsed laser induced plasma. i Schematic diagram of high-speed camera equipped with band-pass filter to take plasma optical pictures. j The plasma images captured by the high-speed camera equipped with band-pass filters (central wavelength: 430 nm and 450 nm). k Performance of laser-catalytic DRM under different catalytic conditions. Error bars represent standard deviation.

The time-dependent dynamic spectra generated under pulsed laser in infocus mode were showed in Fig. 4b. The plasma of CH4 and CO2 were swiftly generated within 50 ms, implying the splitting bremsstrahlung process by pulsed laser. In addition, the plasma intensities were increased with the pulsed laser irradiation time, and peak positions remained relatively stable, confirming the steady generation of plasma by the pulsed laser. Notably, the intensity of plasma increased as the output power of pulsed laser escalated from 4 W to 20 W in infocus mode (Supplementary Fig. 22). In contrast, no spectra of CH4 and CO2 plasma were detected under the CW laser in infocus mode (Fig. 4c) and the pulsed laser in underfocus mode (Fig. 4d). Only the continuous spectrum emitted by atoms at thermodynamically high temperature was observable, which was stable with time. Under the action of CW laser, excited electrons continuously collide with the lattice and emit phonons, converting laser energy into thermal effects and further diffusing into the crystal through heat transfer. While the input energy of the pulsed laser is discontinuous, the collision of excited electrons with the lattice is not sufficient and more electrons can escape from the lattice to become the initial electrons of the cascade reaction, which is necessary to induce avalanche ionization to produce plasma50,51,52. Therefore, the infocus pulsed laser is a pivotal requirement for plasma generation.

A two-dimensional axial symmetry physical model of plasma at the gas-solid interface produced by pulsed laser was established in Fig. 4e. The simulated spatial density distributions of electrons, *CH, and *CO produced by pulsed laser were shown in Fig. 4f, g, and h. It could be inferred that the generation process of laser induced plasma was as following: Firstly, the hot carriers were generated through interaction between laser and Mo2C. Secondly, hot carriers on Mo2C absorbed the focused laser energy by a non-resonant process of inverse harsh radiation, which caused avalanche ionization53, and further cracked CH4 and CO2 molecules to the *CH plasma and *CO plasma, respectively.

According to the excitation spectra of laser-induced plasma of CH4 and CO2, the corresponding plasma images were also captured by a high-speed CMOS (Complementary Metal Oxide Semiconductor) camera equipped with band-pass filters centered at 430 nm and 450 nm (Fig. 4i), which recorded the time-resolved shadowgraphs of plasma expansion (*CH at 431.4 nm, *CO at 451.1 nm), respectively. As shown in Fig. 4j, for the Mo2C, the plasma plume was captured under pulsed laser in the infocus mode, but no capture occurred under the underfocus mode of pulsed laser or the infocus mode of the CW laser.

As shown in Fig. 4k, the H2/CO yields of laser-catalytic DRM in CW laser infocus mode and pulsed laser underfocus mode were 5099.8/8000.6 mmol h−1 g−1 and 6160.8/9389.7 mmol h−1 g−1, respectively, at similar thermal effect temperatures (Fig. 3e and Supplementary Fig. 23). In contrast, the synergy of laser-induced plasma and laser-induced thermal effects achieved in pulsed laser infocus mode not only doubled the yields but also promoted H2 production.

The above results confirmed that the plasmonization effect induced by the focused pulsed laser was the dominant factor in accelerating the DRM catalytic activity. The synergy between laser-induced plasma and laser-induced thermal effects on Mo2C contributes to its exceptionally high catalytic activity for DRM. Pulsed laser produces a high-temperature thermal region on the surface of Mo2C, which is thermodynamically sufficient to drive the DRM reaction. Furthermore, the interaction between pulsed laser and Mo2C generates high-energy electrons, which in turn induce the plasmaization of CH4 and CO2 at the gas-solid interface to enhance the DRM activity. In contrast, the CW laser solely triggers the thermal effect on Mo2C due to electron–phonon relaxation, resulting in poorer DRM performance and stability (Supplementary Fig. 24) under the same temperature condition.

Mo2C was an excellent catalyst for CO2 activation during DRM, but its capacity for CH4 activation is relatively weak21,35. The schematic mechanism of DRM is shown in Supplementary Fig. 25. The production of C* requires a four-step process of stepwise dehydrogenation of CH4, which is the rate-limiting step on Mo2C. In conventional thermocatalytic DRM, the reaction potential for generating C* is higher than that for O* formation, creating a C*-deficient environment that hampers the structural stability of Mo2C. This scenario renders Mo2C prone to oxidation, ultimately transforming into MoO2 (as illustrated in Supplementary Fig. 26b), which is consistent with previous work20,54. The pulsed laser-induced plasma breaks the limiting step of dehydrogenation of CH4, resulting in enhanced catalytic activity (Fig. 2d) and improved stability (Supplementary Fig. 26a) for DRM using a Mo2C catalyst.

Laser-catalytic DRM in flow system

A flow-type catalytic system is the predominant method for evaluating DRM performance55. Consequently, to validate the practical significance of laser catalysis, we established a flow-type laser-catalytic DRM system. (Fig. 5a, b and Supplementary Fig. 27a). Despite laser catalysis being localized to a specific point, the rapid movement of the pulsed laser, with a speed of 1000 mm s−1, surpasses the cross-section velocity (0.67 mm s−1) of the CH4/CO2 gases (Fig. 5c). This design allows the moving pulsed laser to act as a steady laser line, traversing the reactor chamber vertically along the gas flow direction. The laser line’s rapid oscillation within the reactor maintained the methane conversion of DRM reaction at a stable value of 50.5 %. It is worth noting that the pure Mo2C with poor intrinsic catalytic activity as a catalyst showed relatively poor DRM catalytic performance in the thermocatalytic system18,32. Compared to the thermocatalytic system (15.7 %, H2/CO ≈ 0.46, Supplementary Fig. 27b), not only a higher conversion rate (50.5 %) but also a higher H2/CO ratio (H2/CO ≈ 0.86) was implemented in laser-catalytic DRM (Fig. 5d and e). This further affirms the laser’s potential to enhance methane cracking, thereby significantly improving DRM’s activity and stability. The structural stability of Mo2C after laser-catalysis in flow system was confirmed by XRD pattern (Supplementary Fig. 28), XPS (Supplementary Fig. 29) and HRTEM results (Supplementary Fig. 30), which also demonstrated the above conclusion. Hence, the application of laser-catalytic DRM holds promising potential.

Fig. 5: Performance and energy consumption evaluation of laser-catalytic DRM in flow system.
figure 5

a The laser-catalytic DRM device in flow system. b, c Schematic diagram of the interaction among gas molecules, Mo2C and pulsed laser. Comparison of the d CH4 conversion and e selectivity of laser-catalytic DRM (16 W) and thermocatalytic DRM (800 W, 800 °C) via the same quantity of Mo2C, Gas velocity: 60 mL min−1. f Total energy efficiency and the cost efficiency of different catalytic systems.

Energy efficiency was also one of the pivotal indicators of different DRM catalytic systems, which were compared in Fig. 5f. Obviously, laser-catalytic DRM demonstrated superior energy efficiency (0.98 mmol kJ−1) and more cost-effective electricity conversion (46.8 mmol kW−1 h−1) compared with recently reported experimental conditions of thermocatalysis18 and photocatalysis10, signifying its practical application potential. Moreover, from an industrial perspective, industrial-grade medium-power nanosecond lasers are priced at less than $10,000. It is highly automated and easy to regulate with a stable laser output performance and a long working life, generally up to 100,000 hours. Meanwhile, the reactor used for laser-catalysis is simple in structure and does not need to be subjected to high temperature and pressure during the reaction process. The surface temperature of the reactor during the laser-catalytic DRM reaction was close to room temperature and very mild due to the thermal effect of the laser was localized (Supplementary Fig. 3b and Supplementary Fig. 32). Furthermore, the catalyst for laser-catalytic DRM is the single Mo2C, easily and inexpensively scalable for large-scale production. Hence, the future application prospects for laser catalysis as a novel catalytic system are highly promising.

Discussion

A pulsed laser-catalytic DRM reaction was demonstrated without external heating, using simple Mo2C both as the laser carrier and catalyst. The 16 W pulsed laser induced both thermal and plasma effects on Mo2C simultaneously in the infocus mode. This innovative approach enhanced the activation capacity of CH4 through laser-induced plasmaization, effectively breaking the rate-limiting step of DRM. Consequently, the laser-catalytic DRM exhibited an exceptionally high catalytic activity, yielding H2 (14300.8 mmol h−1 g−1) and CO (14949.9 mmol h−1 g−1), respectively. Simultaneously, an equilibrium reaction between CO2 and CH4 was facilitated by the enhanced laser-induced cracking of CH4, creating a balanced environment that prevented Mo2C from over-oxidation. This laser-catalytic approach enabled efficient DRM without relying on an external heating source, a breakthrough with significant implications for advancing the research landscape of photothermal catalytic systems.

Methods

Synthesis of Mo2C

Firstly, MoO2 nanosheets (NSs) as precursors were prepared by a chemical vapor reduction process in a long quartz tube. Briefly, the phase transition reaction from commercial MoO3 to MoO2 NSs occurred in the existence of Ar−H2 (10% H2) mixture (200 mL min−1) at 900 °C for 2 h. After the reaction cooled down naturally, MoO2 NSs were collected at the tail end of the chemical vapor deposition (CVD) system. Due to the equilibrium between the sublimation of MoO3 and the reduction of gaseous MoO3 by Ar-H2, the collected MoO2 NSs exhibit uniformly dispersed hexagonal nanosheet structures of a few microns in size with smooth surfaces and edges. Secondly, the Mo2C NSs were then prepared by carbonization of MoO2 NSs in a tube furnace. Typically, the obtained MoO2 NSs were placed in a ceramic crucible in the tube furnace and heated up to 1000 °C under Ar (50 mL min−1) atmosphere. Once the 1000 °C reached, the Ar gas was shut off and CH4 gas (50 mL min−1) was introduced into the tube for 30 mins. Then, the tube furnace was cooled to room temperature naturally under Ar atmosphere. The powders were collected and subsequently washed with ethanol to obtain dark gray Mo2C NSs.

Characterization

Phase compositions of the as-made materials were measured by D8 Advance (Germany Bruker) X-ray diffractometer (XRD) with Cu Kα radiation (λ = 0.15406 nm). Morphologies and element mapping of the materials were identified by a field emission scanning electron microscope (SEM, Zeiss, Gemini 300; EDS, Oxford, X-MaxN 50) and a transmission electron microscope (TEM, a JEM-2100F Field Emission Electron Microscope, JPN) at an acceleration voltage of 200 kV. X-ray photoelectron spectroscopic (XPS) measurement was performed using a PHI X-tool instrument (Ulvac-Phi). UV-Vis-NIR absorption spectra of samples were recorded by a UV-Vis-NIR spectrophotometer (UH4150, Hitachi High-Technologies Corporation). The production rates of hydrogen and carbon monoxide were measured by a gas chromatograph (GC, Agilent 7890B). High-speed video camera (FuHuang AgileDevice Co., Ltd, X213) was used to verify the plasma plume induced by laser. Time-resolved transmittance spectra were conducted on a home-made device with a portable spectrometer (Aurora 4000, GE-UV-NIR, Changchun New Industries Optoelectronics Tech. Co., Ltd) with step size of 1 nm and dwell time of 50 ms. Infrared thermal images were captured by an infrared thermal imager (Magnity, MAG32HT).

Pulsed laser-catalytic dry reforming of methane in closed system

Before the laser-catalytic DRM, the synthesized Mo2C NSs (20 mg) and BaSO4 (0.5 g) were pressed into a catalyst tablet (diameter 13 mm) by a tablet press at a pressure of 100 bar, the catalyst tablet was composed of a thin layer of Mo2C on the surface of BaSO4 substrate. The Mo2C/BaSO4 tablets with different areas were prepared by laser cutting, including 0.79, 3.14, 7.07, and 12.57 mm2, as shown in Supplementary Fig. 4. Then, by placing the Mo2C/BaSO4 tablet on the bottom of the quartz reactor (Supplementary Fig. 3a), the mixed gas of CO2, CH4 and Ar at a ratio of 47.5%:47.5%:5% was poured into the reactor for 15 mins to remove air in the reactor, then closed the reactor vent. During the laser-catalytic process, a fiber optic laser system (LSF20D, Hgtech laser) with a laser wavelength of 1064 nm and a pulse duration of 100 ns was used to irradiate the catalyst tablet. A computer was connected to the laser system and the software (named EzCad2) was used to set the experimental parameters and map the catalytic reaction zone. A linear scanning mode with a repetition rate of 20 kHz, maximum power of 20 W, scanning spacing of 0.01 mm and scanning speed of 500 mm s−1 was used to perform the laser-catalysis. In order to investigate the effects of laser power and underfocus/infocus on the DRM performance, different powers (4, 8, 12, 16, and 20 W) and different distance of underfocus (defocusing amount = 0, 5, 10, 15, and 20 mm) were utilized during laser-catalysis. Without special explanation, the laser catalytic DRM performances were all obtained in 16 W laser power and infocus laser condition by pulsed laser. The post-reaction gases were analyzed using a GC to obtain the relative amounts of CO, H2, CO2, and CH4.

CW laser-catalytic DRM was used as a control system to confirm the contribution of the pulsed laser-induced plasma effect. CW laser-catalysis induced a high temperature (789 °C) similar to that of the pulse laser-catalysis through 1064 nm CW laser. The catalysts (Mo2C/BaSO4 tablet), quartz reactor and reaction gas (CO2:CH4:Ar = 47.5%:47.5%:5%) used were consistent with the pulse laser-catalysis system.

Pulsed laser-catalytic dry reforming of methane in flow system

The laser-catalytic DRM of Mo2C NSs in a flow system was carried out in a quartz reactor at room temperature and atmospheric pressure. The DRM activity was evaluated under reactive gas flow (CO2:CH4:Ar = 16.7%:16.7%:66.6%). Gas hourly sp/ace velocity (GHSV) with 120 mg Mo2C catalyst was controlled at 30 L·gcat−1 · h−1. A linear scanning mode with a repetition rate of 20 kHz, single pulse energy of 0.8 mJ, scanning spacing of 0.05 mm and scanning speed of 1000 mm s−1 was used to perform the laser-catalytic DRM.

The thermocatalytic DRM of Mo2C NSs in a flow system was carried out in a fixed bed quartz reactor at atmospheric pressure. The activity was evaluated at 800 °C with the same catalyst amount and total flow rate of the feed gas (CO2:CH4:Ar = 16.7%:16.7%:66.6%, GHSV = 30 L gcat–1 h–1) as those of laser-catalysis.