Introduction

Methanol is an important raw material for chemical production, and the direct conversion of methane into methanol has commercial value. Currently, the industrial process to synthesize methanol involves the use of syngas1, which is usually prepared by methane reforming2,3, but the cost of this process is high4. The direct one-step oxidation of methane to methanol is more advantageous than the syngas route. Noble metals (e.g., Au, Pd, and Rh) have been reported to exhibit excellent catalytic properties for the conversion of methane5,6,7,8,9,10. Agarwal et al.5 oxidized methane to methanol using Pd–Au colloidal nanoparticles in the presence of H2O2 and O2. Xiao et al.7 designed a “molecular fence” catalyst synthesized by fixation of AuPd alloy nanoparticles within aluminosilicate zeolite crystals followed by modification of the external zeolite surface with organosilanes. Methane was converted into methanol with high efficiency, with a production rate of 91.6 mmol gAuPd−1 h−1. In addition, some non-noble metals have also been reported to show potential for this reaction. Inspired by the biocatalysis of methane monooxygenase (MMOS)11,12, Grundner et al.13 prepared a Cu-MOR catalyst by ion exchange, with active sites similar to those present in MMOS. The conversion of methane to methanol was realized by multiple reaction steps (activating active sites under an O2 atmosphere, methane reaction, and methanol hydrolysis). Except for intermittent reactions, Narsimhan et al.14 reported the simultaneous passage of methane, oxygen, and water into the reactor for the reaction. Although the reported conversion was not high, the oxidation of methane to methanol under continuous conditions was realized for the first time.

Besides the complexity of the operation process, a high reaction temperature is employed due to the intrinsic characteristics of methane, which is difficult to activate15,16,17, and the oxidation of the product methanol is easier than that of methane, typically leading to the continuous oxidation of methanol to CO2 under the reaction conditions. Hence, unsatisfactory results are obtained18,19. NTP is a form of plasma with the characteristics of low temperature and high-energy electrons20. Therefore, the introduction of NTP into the reaction system can convert some difficult-to-activate molecules into active groups21 and considerably accelerate the reaction speed. Nevertheless, the inhibition of the over-oxidation of methane is still an urgent issue.

In this study, a new route to synthesize methanol from methane using NTP is designed, which reasonably and completely exploits the catalytic characteristics of plasma. At room temperature under atmospheric pressure, the formation of methanol is realized by directly mixing CH4 and a weak oxidant H2O without cumbersome operational steps. The reaction between CH4 and H2O to afford methanol is performed in a dielectric barrier discharge (DBD) reactor (Fig. 1). The reaction environment involves three parts: the plasma phase, the liquid phase, and the plasma-liquid interface. Each of these parts corresponds to the formation of methyl radicals (CH3·), formation of hydroxyl radicals (·OH), and production of methanol. Methane in the plasma phase is activated by inelastic collision with high-energy electrons, becomes an excited state (CH4*), and is decomposed to methyl radicals (CH3·). These activated molecules and radicals contribute predominantly to methanol production. Liquid-phase H2O provides the required ·OH for the reaction. By constructing a TiO2 surface heterojunction in the liquid phase and using the photons and electrons22 produced in the plasma, the formation of hydroxyl radicals is significantly promoted. In addition, the liquid-phase environment is conducive to the instantaneous transfer of methanol produced at the liquid-plasma interface, inhibiting the decomposition and transformation of methanol.

Fig. 1: Schematic of the plasma system.
figure 1

The reaction between CH4 and H2O to afford methanol is performed in a dielectric barrier discharge (DBD) reactor.

Full size image

Results and discussion

Catalytic performance

Methanol can be produced in all experiments by this route, and it is crucial that CO2 is not produced. The presence of a small amount of methanol is detected under the plasma-only mode (Fig. 2). On the one hand, the CH3· produced by CH4 activation can directly react with H2O to form methanol; on the other hand, the plasma decomposes H2O molecules when they reach the phase interface, and the gas phase comprises a small amount of water vapor, which is also converted into ·OH by high-energy electrons. Clearly, the addition of the TiO2 catalyst in the DBD plasma enhances the production rate of methanol to 29.7 mmol gcat−1 h−1 (Supplementary Fig. 1). Compared with the absence of a catalyst, a 3-fold increase in the methanol production rate is achieved. The selectivity of gas phase and liquid phase products are shown in Table S1, it can be seen that 4.5–12.7% of methane is converted to methanol.

Fig. 2: Alcohol production rate.
figure 2

Effects of different catalysts on the reaction (CH4 flow rate of 5 mL min−1, discharge power of 30 W, ca. 5 mg catalyst, error bars obtained from repeated three sets of experiments on the same catalyst).

Full size image

X-ray diffraction (XRD) was employed to characterize the chemical structures of a series of TiO2-t samples. Peaks at 25.3°, 37.8°, 48.0°, and 62.7° correspond to the characteristic diffraction peaks of anatase, while those at 27.4°, 36.1°, 41.2°, and 54.3° correspond to those of rutile. The XRD spectra reveal that the catalyst contains both anatase and rutile and that the rutile is the main crystal form (Supplementary Fig. 2). High-resolution transmission electron microscopy (HRTEM) reveals that the TiO2-700 shows crystal plane spacings at ~2.44 and 3.52 Å, corresponding to the (101) crystal plane of the rutile phase and the (101) crystal plane of the anatase phase (Fig. 3f), respectively. This is consistent with the XRD analysis. This catalyst is advantageous as the hybrid catalyst consisting of anatase and rutile can form a heterojunction. Hence, electrons and holes can be effectively separated and lead to better catalyst performance. The surface phase composition of TiO2 is the key factor that affects the catalytic performance21. Visible Raman and UV Raman spectra (Fig. 3a, b) reveal the presence of anatase and rutile phase junctions on the catalyst surface, and the characteristic peak value of anatase in the TiO2-700 sample decreases, indicating that the surface phase changes. The samples show similar band gaps in the UV–Vis spectra (Fig. 3c), but the light absorption intensity of TiO2 gradually increases with increasing treatment temperature, corresponding to the maximum at 700 °C. The X-ray photoelectron spectroscopy valence band (XPS-VB) spectra (Fig. 3d) reveal that the valence band energy of each sample after treatment gradually increases and that the peak value is attained at 700 °C. Figure 3e shows a schematic of the band-gap structure. The VB energy value of 2.43 eV for unprocessed raw powder TiO2 changes to 2.69 eV for TiO2 treated at 700 °C, indicative of the improved oxidation performance of the holes formed by the treated catalyst. The catalytic performance from experiments is consistent with the characterization results. With the increase in the treatment temperature, the production rate of methanol increases to the peak value and then decreases, and TiO2-700 exhibits excellent performance.

Fig. 3: Spectra and HRTEM image of the samples.
figure 3

a Visible Raman spectra, b UV Raman spectra, c UV–Vis spectra, d XPS spectra showing the valence band (VB) levels, e schematic of the band-gap structure, and f HRTEM image of the TiO2-700 sample.

Full size image

Possible reaction pathways

Figure 4 shows the possible reaction pathways for the formation of methanol and ethanol. Isotope tracer experiments were performed with D2O and analyzed by gas chromatography-mass spectrometry (GC-MS) (Fig. 5). The GC-MS spectra reveal that most of the produced methanol is CH3OD and that few molecules are CH3OH. The results reveal that the methanol produced by the reaction mainly originates from the combination of methyl radicals (CH3·) and hydroxyl radicals (·OH) (Eq. 10). Moreover, few CH3· radicals combine with ·O to form CH3O· and then combine with H· to form CH3OH (Eqs. 7-9). Experimental results show that in addition to CH3OH, C2H5OH is a main by-product. Therefore, increasing the CH3OH production rate and making efficient use of CH3· and ·OH are the key issues for improving methanol production rate and selectivity.

Fig. 4: Possible reaction pathways for the reaction between CH4 and H2O with DBD.
figure 4

The methanol produced by the reaction mainly originates from the combination of methyl radicals (CH3·) and hydroxyl radicals (·OH).

Full size image
Fig. 5: GC-MS spectra of methanol formed using H2O and D2O.
figure 5

The GC-MS spectra reveal that most of the produced methanol is CH3OD and that few molecules are CH3OH.

Full size image

Effect of adding inert gas on methanol production rate

The addition of an inert gas (e.g., Ar, He) to the reaction feed to improve the reaction performance was explored from two aspects. On the one hand, in DBD plasma, CH3· radicals constitute the main free radicals produced from CH423, but other free radicals are also present (e.g., CH2· and CH·) (Eqs. 1–3). At this time, due to the dilution of the original radical composition by Ar or He, the self-coupling of methyl radicals is reduced (Eq. 4). Hence, the produced methyl radicals are utilized for methanol synthesis to a greater extent, and the selectivity of methanol is subsequently considerably improved. On the other hand, the added Ar or He gas can produce metastable atoms with a longer average life span, thereby leading to Penning ionization. Ar or He activated by electrons transfers its energy to CH4 for the activation of an increased number of CH4 molecules24 (Eqs. 15, 16). The formation of CH3· from CH4 in a plasma is caused by the energetic electron-initiated dissociation of CH4 (bond dissociation energy: 4.52 eV). Furthermore, ·OH can be formed by the electron impact dissociation of H2O (bond dissociation energy: 5.19 eV). Therefore, the average electron energy in a DBD plasma is a key factor affecting the reaction. The electrons in the high-energy tail in the Maxwellian distribution are responsible for the dissociation of CH4 and H2O. Introducing He or Ar can increase the average electron energy in the reactor, which leads to higher CH4 and H2O conversions. The addition of Ar or He considerably increased the production rate of methanol. With the addition of Ar, the production rate reached 58.1 mmol gcat−1 h−1 (Fig. 6a). Compared to previous studies, this system exhibits an excellent methanol production rate, and most of the previous studies used O2 or H2O2 as the oxidant (Table 1). With the addition of He, the production rate reaches 56.7 mmol gcat−1 h−1. Although the production rate increase with He is slightly lower than that with Ar, the selectivity of methanol clearly increases from 65 to 93% in the liquid products under CH4/He plasma. After the addition of Ar or He, the plasma discharge current waveform was monitored (Supplementary Fig. 3). The addition of Ar increases the electron density25, which in turn increases the collision probability between electrons and CH4 in the reaction. As a result, an increased number of CH4 molecules are activated and transformed (Supplementary Fig. 4). The addition of He leads to an increase in the average electron energy in the plasma region but reduces the electron density of the system26. Hence, the CH4 conversion increase is less than that with the addition of Ar.

Fig. 6: Alcohol formation rate and methanol selectivity.
figure 6

a Effect of the addition of He or Ar on the reaction at total flow rate of 15 mL min−1. b Effect of the He/CH4 feed ratio on the reaction at CH4 flow rate 5 mL min−1 (discharge power of 30 W, ca. 5 mg TiO2-700 catalyst, error bars obtained from repeated three sets of experiments on the same catalyst).

Full size image
Table 1 Comparison of methanol production rates obtained in this study and in previous studies.
Full size table

Effect of the flow ratio of He/CH4 on the methanol production rate

The addition of He improves methanol production rate as well as methanol selectivity. Next, the effects of the addition of different amounts of He on the methanol production rate and selectivity were investigated (Fig. 6b). The results reveal that the selectivity of methanol increases with the increase in He addition. At a He/CH4 ratio of 6:1, the selectivity of methanol reaches 96%. Simultaneously, with the increase in the He/CH4 feed ratio, the methanol production rate initially increases and reaches a peak at a He/CH4 feed ratio of 2:1. A further increase in the He/CH4 ratio leads to a decrease in methanol production rate due to a possible decrease in the gas–liquid contact time by the excess airspeed.

In conclusion, a new route for the direct conversion of methane to methanol in one step at room temperature under atmospheric pressure was designed and demonstrated with simple operation. The entire reaction was further tailored by considering two aspects: CH3· production and ·OH production. A methanol production rate of 56.7 mmol gcat−1 h−1 and a selectivity of 93% in the liquid products were obtained, and CO2 formation was not observed, successfully inhibiting the over-oxidation of methane. This designed reaction system may also be applied for the selective oxidation of other stable chemicals such as benzene to phenol.

Methods

Catalyst preparation

Purchased TiO2 powders were ground in an agate mortar, then placed in a crucible and transferred to a muffle furnace. The TiO2 samples were heat treated in an air atmosphere at 500 °C, 600 °C, 700 °C, or 800 °C for 4 h to obtain a series of catalysts, which were recorded as TiO2-t, where t is the calcination temperature.

Catalyst characterization

Powder X-ray diffraction (XRD) patterns were measured using a Rigaku Smart Lab diffractometer, with a nickel-filtered Cu Kα X-ray source, at a scanning rate of 0.02° over a 2θ range of 5° to 80°. Transmission electron microscopy (TEM) tests were performed using a Tecnai G2 20 S-twin instrument (FEI Co.) with an accelerating voltage of 200 kV. TEM samples were ultrasonicated in ethanol, dropped onto carbon-coated copper mesh, and dried under ambient conditions. High-resolution transmission electron microscopy (HRTEM) measurements were obtained to determine the lattice spacing of the TiO2 for a comprehensive understanding of the crystal structure. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha), with an Al Kα excitation source (hν = 1486.6 eV), was used to determine core level binding energies of surface species. Sample charging was corrected by referencing all measurements to the C (1s) peak at 284.8 eV. Raman spectra were recorded with a Horiba Scientific LabRAM HR Evolution spectrograph using 532 and 325 nm laser lines. UV–Vis absorption spectra were obtained in the range of 200–800 nm using a UV-2600 spectrophotometer. The methanol product in the isotope-labeled reaction was analyzed by a gas chromatograph-mass spectrometer (GCMS) using an Agilent GCMS 7890B-5977B equipped with an HP-5ms chromatographic column.

Experimental

Supplementary Fig. 5 shows a process flow diagram of the reaction system. 5 mg catalyst was ultrasonically dispersed in 3 mL H2O and placed in a dielectric barrier discharge (DBD) reactor. The details of the plasma reaction system are shown in Supplementary Fig. 6. The methane feed into the reactor was controlled by a mass flow meter at a constant rate of 5 mL min−1. The plasma was generated using a high-voltage power supply, and the reaction was allowed to proceed for 20 min. The plasma power (30 W) is the input power and is calculated by multiplying the input voltage (50 V) and the input current (0.6 A). In addition, a Lissajous figure was measured with 30 W input power using an oscilloscope. Supplementary Fig. 7 shows the Lissajous figure at an input power of 30 W. According to Supplementary Fig. 7, the actual output power corresponding to the input power of 30 W can be calculated to be 22.4 W. The applied voltage peak to peak (Vpk-pk) was approximately 27 kV (Supplementary Fig. 8). The reaction solution was filtered and then analyzed by gas chromatography to determine the product composition. Methanol and ethanol were analyzed by a gas chromatograph (GC) equipped with an FID detector and a Porapak Q column. The standard calibration curves of the peak areas and methanol and ethanol concentrations are shown in Supplementary Figs. 9 and 10. Methane was analyzed using a GC equipped with an FID detector and a 30 m PLOT-Q capillary column. CO2 was analyzed using a TCD detector and a TDX-01 molecular sieve column.

Calculation method of discharge power

We calculated the average power by finding the area under the curve for the V–Q Lissajous plot and multiplying it by the frequency.

$$P=\oint v\left(t\right)\times q\left(t\right)\times {{{{{{\rm{d}}}}}}t}=\frac{f}{2\pi }\times S$$

where P is the average power in W, v(t) is the voltage measured by the oscilloscope, q(t) is the charge measured by the oscilloscope in C, f is the frequency in kHz, and S is the area under the curve for one cycle.