Xing
Li
a,
Yang
Bai
*a,
Xian
Shi
b,
Na
Su
a,
Gongzhe
Nie
a,
Rumeng
Zhang
c,
Hongbo
Nie
ad and
Liqun
Ye
*ae
aState Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, School of Oil & Natural Gas Engineering, Southwest Petroleum University, Chengdu, 610500, China. E-mail: baiyanghyq@foxmail.com
bInstitute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu, 610054, China
cKey Laboratory of Ecological Security for Water Source Region of Mid-line Project of South-to-North Water Diversion of Henan Province, College of Chemistry and Pharmaceutical Engineering, Nanyang Normal University, Nanyang, 473061, China
dCNOOC (China) Co., LTD. Tianjing Branch, Tianjing, 300452, China
eCollege of Materials and Chemical Engineering, Key Laboratory of Inorganic Nonmetallic Crystalline and Energy Conversion Materials, China Three Gorges University, Yichang, 443002, China. E-mail: lqye@ctgu.edu.cn
First published on 28th January 2021
MXenes are two-dimensional nanomaterials, which can be constructed from different elements. The rich interlayer groups, surface groups, and the flexible layer spacing of MXenes make them ideal catalysts. Among these, Ti3C2Tx has gained particular attention as a photocatalyst for photocatalytic CO2 reduction reactions (CO2RR), hydrogen evolution reactions (HER), and photocatalytic degradation reactions. The structure of Ti3C2Tx, hydrophilic surface functional groups, and the Gibbs free energy for hydrogen adsorption lead to the excellent photocatalytic HER performance of this material. Numerous surface defects on Ti3C2Tx also provide plentiful CO2 adsorption sites for CO2RR. It is the structure of two-dimensional nanomaterials and their high-speed electron transport channels that enable their excellent catalytic oxidation activity. However, at present, there are still challenges that limit their further application, the most significant of which is the material stability. In order to overcome this, the synthetic routes to prepare these photocatalysts need to be adapted.
Photocatalytic nanomaterials have an extensive number of potential applications. When their particle size is below a certain value, the Fermi level of the electronic energy levels morphs from continuous to discrete levels and the energy gap grows wider. These semiconductors are, therefore, more susceptible to photon excitation, which improves their photocatalytic activity.1
Nanomaterials can be divided into four categories, according to the dimensions of their structural scale: (1) zero-dimensional materials, e.g., groups of nanostructure clusters; (2) one-dimensional nanomaterials, e.g., fibrous nanotubes, nanowires, nanoribbons, or other related structures; (3) two-dimensional nanomaterials, e.g., layered nanomaterials, quantum wells, superlattices, and other structures; (4) three-dimensional nanomaterials, e.g., composite structures consisting of one or more zero-dimensional, one-dimensional, or two-dimensional nanomaterials. The first three are collectively known as low Vannami materials. In low Vannami materials, two-dimensional nanomaterials show significant changes in the surface, electron energy levels, state density, and other aspects compared with three-dimensional materials. This due to the fact that their thickness is greatly reduced compared to other two-dimensional materials; thus, these possess unique optical and electronic characteristics, which make them a hot topic in catalysis.2,3
MXenes are two-dimensional nanomaterials and have a general material formula of Mn+1XnTx. In this formula, M represents nitrogen or carbon, X is generally a transition metal element, and T represents the functional groups. MXenes typically consist of transition metal carbides, nitrides, or carbides that are several atomic layers thick. It was first reported in 20114 that MXene materials have comparable conductivity towards transition metal carbides due to the presence of hydroxy groups or terminal oxygen species on their surfaces. The most important feature of this range of materials is that, unlike conventional battery materials, they provide more channels for ions to move through, thus dramatically increasing their speed.
Ti3C2Tx was the first discovered MXene material and is also the most widely used MXene material in the field of photocatalysis.5–7 It was first obtained by etching the Al layer of Ti3AlC2 with hydrofluoric acid. In this paper, the application of Ti3C2Tx as a photocatalyst and approaches to improve its catalytic performance are summarized.
The preparation of MXenes can be divided into two approaches, namely, bottom-up and top-down. Presently, the top-down etching method is commonly employed. This is due to the MAX phase9,10 (commonly, M means early transition metal elements such as Ti and Nb; A represents the Al or Si layer; X represents C or N elements) as M is mainly composed of metallic bonds between the atoms, which are connected to A. The chemical properties are largely dictated by A. By using certain concentrations of hydrofluoric acid or LiF/HCl11 to etch the MAX phase of Ti3AlC2, ternary carbides within the titanium carbon layers become closer to each other. In this process, the Al layer is etched away gradually, resulting in a greater carbon–titanium interlayer spacing in the Ti3C2Tx product. In order to obtain Ti3AlC2 with a graphene-like structure consisting of only a few or single layers, mechanical or chemical intercalation dissection is required. However, when chemical intercalation is used for stripping, some organic molecules may occupy the active sites exposed on the surface, which is unfavorable for photocatalytic reactions.
Etching is a slow process, as shown in Fig. 1(a). In this process, Al layers are gradually peeled off, while the Ti–C skeleton layers are not damaged because of their strong ionic bonding.12 Free groups such as –OH and some H2O molecules enter into the framework of Ti–C and become inter-connected by hydrogen bonds, which expands the layer spacing of Ti3C2Tx. This permits ions with a large radius to enter the layer spacing,13 providing an operating space for the ion intercalation method to peel-off few layers of Ti3C2Tx. The number of –OH groups and H2O molecules within the interlayer space accounts for the large electrical capacity of Ti3C2Tx.
Ti3C2Tx obtained by direct etching with hydrofluoric acid possesses a different morphology to Ti3C2Tx obtained by etching with LiF/HCl. Furthermore, NMR spectroscopy revealed a greater number of –OH and –F functional groups on the surface of Ti3C2Tx etched by hydrofluoric acid, while LiF/HCl etching furnished a material with predominantly –O functional groups.
Various surface groups (such as –O, –F, and –OH) have supplied abundant anchored sites for the base photocatalyst to form efficient heterojunction structures, which are ideal for photocatalytic activities.26 There is also a large number of exposed metal sites on the surface, which can be used as active sites for reactions.
The surface chemical state of MXene materials has a large influence on the regulation of its physical properties. When –F on the surface is replaced by an –O group, the electrochemical performance is improved. For example, when Ti3C2Tx is treated with a KOH and CH3OOK solution, the –O groups on the surface increase, along with the electric capacity. Under an atmosphere of N2, Ar, or other inert gases, the number of –F groups on the surface of Ti3C2Tx is reduced, after which the electrical capacity is greatly increased.
Ti3C2Tx shows an excellent absorption of light between 300 nm and 500 nm.27 Recently, researchers have even found that the absorption can be broadened to the near-infrared (NIR) region. According to a further study, this may be related to its surface plasmon resonance (SPR), and the thinner the material, the stronger the SPR.28 Such a peculiarity makes Ti3C2Tx an ideal photothermal co-catalyst.
The crystal structure of Ti3C2Tx contains Ti defects, which appear to contribute significantly to the instability of this material.14,33 High angle annular dark field (HAADF)-STEM imaging is an important tool in 2D materials’ characterization and is used to unambiguously resolve the crystal structure and defect configurations.34,35 As shown in Fig. 4, widespread Ti defects were directly detected through the HAADF-STEM imaging of the single-layer Ti3C2Tx flakes. Single-layered Ti3C2Tx obtained by HF etching was observed through HAADF-STEM images in Fig. 4(a)–(c). Fig. 4(d) was obtained by calculating tens of such images, and it reveals that the relationship between HF concentration and defect formation. It was found that vacancy clusters are rarely observed after etching with 2.7 wt% HF concentration but are relatively common after etching with 7 wt% HF.33 Generally speaking, the average concentration of VTi (Ti vacancies) is positively related to that of HF.
Fig. 4 HAADF-STEM images from single-layer Ti3C2Tx MXene flakes prepared using etchants with different HF concentrations: (a) 2.7 wt% HF, (b) 5.3 wt% HF, and (c) 7 wt% HF. Single VTi vacancies are indicated by the red circles, while vacancy clusters VCTi are shown by the blue circles. (d) Scatter plot of the defect concentration from the images acquired from samples produced using different HF concentrations. The red line shows the error plot with the average and standard deviation for different HF concentrations.33 |
There are three important steps in the HER process, which are:30,40,41 (a) initial h+ + e− formation; (b) generation of H* (the intermediate adsorption state); and (c) formation of the 1/2 H2 product. The adsorption state of H* in process (b) directly affects the final hydrogen evolution efficiency and is an extremely important factor, which can be represented by the Gibbs adsorption free energy |ΔGH*|. Through simulation calculations, it was found that when all the Ti3C2Tx surface groups are –F, ΔGH* = −0.927 eV, and the adsorption is too strong. When all the surface groups are –O, |ΔGH*| is 0.003 eV, which is even better than the commonly used catalyst Pt (ΔGH* ≈ − 0.090 eV).42,43 Therefore, Ti3C2Tx is a good HER co-catalyst. Examples of Ti3C2Tx used for photolysis in recent years are summarized below (Table 1).
Name | H2 production (μmol h−1 gcatalyst−1) | AQY (%) | Activity improvement factor | Sacrificial reagent | Preparation methods | Monolayer or multilayer | Light source | Morphology | Year | Ref. |
---|---|---|---|---|---|---|---|---|---|---|
CdS/Ti3C2Tx | 14342 | 40.1% (420 nm) | 135.59 times | Lactic acid (17.6%) | One-step hydrothermal method | Ti3C2Tx NPs | 300 W Xe lamp (λ > 420 nm) | Cauliflower-structure by self-assembly of many NPs | 2017 | 30 |
2D-Layered Carbon/TiO2 | 480.8 | 1.98% (400 nm) | TEOA (10%) | Ti3C2Tx oxidation | Multilayer | 300 W Xe lamp (λ > 400 nm) | Nanosheets | 2017 | 44 | |
Ti3C2Tx/rutile TiO2 | 17.8 | 0.3% | Approximately 4 times | Methanol (25%) | Hydrothermal method | Monolayer | 200 W Hg lamp (λ > 400 nm) | 2D sheetswith TiO2 attached on the surfaces and between the sheets | 2016 | 45 |
Ti3C2/Pt/g-C3N4 | 5000 | 3.1% (420 nm) | 15 times than pristine g-C3N4 | TEOA (10%) | Hydrothermal and photodeposition method | Monolayer | 300 W Xe lamp | Nanosheets with porous nanoparticles | 2018 | 46 |
Sulfur-doped Carbon/TiO2 | 333 | 7.36% | Methanol (10%) | Ti3C2Tx oxidation | Multilayer | 300 W Xe lamp (λ > 400 nm) | Nanosheets | 2018 | 47 | |
Ti3C2Tx/TiO2 nanoflowers | 783.11 | 5.86 (350 nm) | 6 times | Methanol (20%) | Hydrothermal and calcination | Multilayer | 300 W Xe lamp | Nanoflowers | 2018 | 48 |
Zn2In2S5/Ti3C2Tx | 2596.76 | 8.96% (420 nm) | 1.97 times | 0.25 M Na2SO3/0.35 M Na2S/H2PtCl6 | Hydrothermal | Multilayer | 300 W Xe lamp (λ > 420 nm) | Flower-like microspheres | 2018 | 49 |
d-Ti3C2/TiO2/g-C3N4 | 1620 | 4.16% (420 nm) | 12.15 times than pure g-C3N4 | TEOA (10%) | Calcination | Monolayer | 300 W Xe lamp (λ > 420 nm) | 2D–2D heterostructure | 2018 | 50 |
ZnS/Ti3C2 | 502.6 | 4 times | Lactic acid (20%) | Hydrothermal | Multilayer | 300 W Xe lamp | Sphere-like structure | 2019 | 51 | |
1D CdS nanorod/2D Ti3C2 MXene nanosheet | 2407 | 35.6% (429 nm) | 6.68 times | Lactic acid (10%) | Electrostatic self-assembly | Monolayer | 300 W Xe lamp (λ > 420 nm) | 1D/2D nanosheets | 2019 | 52 |
TiO2 nanofibers/MXene Ti3C2 | 6979 | 3.8 times than TiO2 nanofibers | Methanol (10%) | Electrostatic self-assembly technique | Monolayer | 300 W Xe lamp | Nanofibers/nanosheets | 2019 | 53 | |
TiO2 nanoparticale/monolayer Ti3C2 | 2650 | 15.8% (305 nm) | 2.88 times than TiO2 nanoparticles/multilayer Ti3C2 | Methanol (25%) | Electrostatic self-assembly technique | Monolayer | 200 W Hg lamp (285–325 nm) | Nanosheets | 2019 | 54 |
MoS2/Ti3C2 | 6144.7 | 2.33 times | Methanol (30%) | Hydrothermal | Multilayer | 300 W Xe lamp (λ > 400 nm) | Spheres-like structure | 2019 | 55 | |
Ti3C2 MXene/O-doped g-C3N4 | 25124 | 17.59% (405 nm) | 1.8 times than O-doped g-C3N4 | TEOA | Electrostatic self-assembly technique | Multilayer | 300 W Xe lamp | 2D nanosheets structure | 2019 | 56 |
CdLa2S4/Ti3C2 | 11182.4 | 15.60% (420 nm) | 13.4 times | 0.25 M Na2SO3 and 0.35 M Na2S | Hydrothermal | Monolayer | 300 W Xe lamp (λ > 420 nm) | Particle-like | 2019 | 57 |
Ti3C2 MXene quantum dots/g-C3N4 | 5111.8 | 3.654% | 25.97 times | TEOA (15%) | Deposition | Ti3C2 MXene quantum dots | 300 W Xe lamp | Nanosheets | 2019 | 58 |
MoxS@TiO2@Ti3C2 | 10505.8 | 7.535% | 5.99 times than Mo2S@TiO2@Ti3C2 | TEOA | In situ growth and hydrothermal | Multilayer | 300 W Xe lamp | Nanosheets | 2019 | 59 |
Ti3C2/porous MOFs (UiO-66-NH2) | 204 | Approximately 8 times | 0.1 M Na2S and 0.1 M Na2SO3 | Hydrothermal | Monolayer | 350 W Xe lamp | 3D structure | 2019 | 60 | |
C-TiO2/g-C3N4 | 1409 | 8 times than C-TiO2 | TEOA (10%) | Calcination | Multilayer | 300 W Xe lamp (λ > 420 nm) | Smooth sheet-like structure | 2019 | 32 | |
CdS@Ti3C2@CoO | 134.46 | 1.75 times than CdS@CoO | Calcination | Monolayer | 300 W Xe lamp (λ > 420 nm) | Spheres-like structure | 2019 | 61 | ||
TiO2–Ti3C2–CoSx | 950 | 5.8 times than TiO2 | Methanol (20%) | Hydrothermal | Multilayer | 300 W Xe lamp | Smooth round block morphology | 2019 | 62 | |
Ti3C2(TiO2)@CdS/MoS2 | 8470 | 3.76 times than CdS/MoS2 | lactic acid (20%) | Hydrothermal | Multilayer | 300 W Xe lamp (λ > 420 nm) | Nanospheres | 2019 | 63 | |
Ti3C2 MXene/MoS2 nanosheets/TiO2 nanosheets | 6425.297 | 4.61% | 7.15 times than TiO2/Ti3C2 | TEOA | Ti3C2Tx oxidation | Multilayer | 300 W Xe lamp | Ti3C2 nanosheets with MoS2 nanoparticales | 2019 | 64 |
2D/3D g-C3N4/Ti3C2 (MXene) heterojunction | 116.2 | 6.64 times | TEOA (10%) | Calcination | Multilayer | 300 W Xe lamp (λ > 420 nm) | Nanosheets | 2020 | 65 | |
Au/MoS2/Ti3C2 | 12000 | Methanol (30%) | Electrostatic self-assembly technique | Multilayer | Nanosphere-like | 2020 | 66 | |||
2D/2D Ti3C2/g-C3N4 | 72.3 | 0.81% (400 nm) | 10.18 times than pure g-C3N4 | TEOA (10%) | Electrostatic selfassembly approach | Monolayer | 200 W Hg lamp | Flat irregularly shaped nanosheets of 2D/2D structures | 2019 | 67 |
MXene@Au@CdS | 17070.43 | 1.85 times than pure CdS | 0.35 mol L−1 Na2S and 0.25 mol L−1 Na2SO3 solution | Hydrothermal | Monolayer | 300 W Xe lamp (λ > 420 nm) | Nanosheets | 2020 | 68 | |
Black phosphorus quantum dots/Ti3C2@TiO2 | 684.5 | 11.35 times | TEOA (25%) | Solvent-heatmethod | Multilayer | 300 W Xe lamp (λ > 420 nm) | Nanosheets | 2020 | 69 |
Ti3C2Tx plays a significant role in HER, whether as a co-catalyst or as a part of the overall catalyst, as it greatly improves the performance of the base catalyst. As shown in Table 1, the presence of Ti3C2Tx increases the yield of H2 compared to solely the base catalyst by more than 2 times. The quantum efficiency is also significantly improved to 40.1%, whilst the maximum value of hydrogen production is 14.34 mmol g−1 h−1.
Monolayer Ti3C2Tx or quantum dot Ti3C2Tx displays better activity in HER. However, the use of monolayer Ti3C2Tx as a photocatalyst has several disadvantages: (a) the preparation of monolayer Ti3C2Tx is complex; (b) the structural stability is low and the catalyst is easily oxidized in water; (c) manipulation of the mono-layer or few-layer structures is not easy to carry out. Few-layer structures are presently prepared by electrostatic self-assembly or by in situ growth. The stability of the composite catalyst obtained by in situ growth is significantly greater than that obtained by electrostatic self-assembly.
Due to the surface hydrophilic groups,70 suitable Gibbs adsorption free energies |ΔGH*|, and excellent electron transfer efficiency, Ti3C2Tx not only plays an important role in the three-step process of HER but can participate in electron hole separation.
Xiao et al. successfully synthesized the Schottky junction of 1D CdS nanorod/2D Ti3C2 MXene nanosheet in 2019.52 As shown in Fig. 5, Xiao et al. anchored Cd2+ using the deficiency of Ti on the Ti3C2 surface and the electrostatic interaction of free Cd2+ to prepare the 1D CdS nanorods. The composite material demonstrated excellent hydrogen production performance (2407 μmol h−1 gcatalyst−1), producing 6.68 times as much H2 as pure CdS (Fig. 6).
Fig. 5 TEM images of (a–c) CdS, (d–f) exfoliated Ti3C2 MXene nanosheets, (g–i) the composite CM-20, (j) the corresponding elemental mapping results of CM-20, and (k) the oxidation process of Ti3C2Tx.52 |
Fig. 6 (a and b) Photocatalytic H2 evolution performance of different samples, (c) the recycled photocatalytic H2 evolution experiments of CM-20, (d) AQY values and the wavelength dependence of photocatalytic H2 evolution in the composite CM-20.52 |
Theoretically, the surface negative value (zeta potential value: ∼18 mV) of Ti3C2Tx is sufficient to adsorb positively charged Cd2+. Ti3C2Tx treated with DMSO forms a low-layered structure, on which Cd2+ can be anchored and one-dimensional CdS nanorods can be grown. As shown in Fig. 5, due to the constraint effect of Ti3C2Tx, the length of 1D CdS nanorods in the 1D CdS nanorods/2D Ti3C2Tx heterojunctions is smaller than that of the 1D CdS nanorods alone.
CdS equipped with Ti3C2Tx displays excellent electrochemical properties. As shown in Fig. 7, the photocurrent of 1D CdS nanorods/2D Ti3C2Tx was significantly better than that of one-dimensional CdS nanorods and the optical resistance was significantly lower than that of one-dimensional CdS nanorods. ESR tests show that the hydroxyl radical and superoxide radical signals of 1D CdS nanorods/2D Ti3C2Tx were significantly enhanced after the addition of Ti3C2Tx. In conclusion, under the same illumination conditions, 1D CdS nanorods/2D Ti3C2Tx generate more photogenic carriers. These produce oxygen-containing groups with oxidizing reductivity, which can participate in photocatalytic hydrogenation reactions.
Fig. 7 (a) Photocurrent density curves and (b) EIS Nyquist plots of CdS and CM-20.52 |
1D CdS nanorods/2D Ti3C2Tx typically exhibit better visible light response, electron hole separation efficiency, and more effective carrier transport efficiency after the formation of multi-dimensional heterojunctions. This accounts for their excellent photocatalytic hydrogen evolution capability.
Using Ti3C2Tx as the co-catalyst, Wang et al. synthesized a TiO2/Ti3C2Tx complex photocatalyst,45 which was 4 times more efficient than pure phase TiO2 in photohydrolyzing aquatic hydrogen. This is attributed to the Schottky barrier formed between TiO2 and Ti3C2Tx, which effectively improves the separation efficiency of the electron holes. As shown in Scheme 1, excited electrons can be wired to Ti3C2Tx from the conduction band of TiO2 owing to the close contact between Ti3C2Tx and TiO2; thus, negative charge is accumulated in Ti3C2Tx and a depletion layer formed at the metal–semiconductor interface, which is the Schottky barrier.45
Scheme 1 Formation of Schottky barrier at the MXene/TiO2 interface.45 |
In this work, Wang et al. treated Ti3C2Tx with DMSO to form low-layer structures. Amorphous TiO2 was formed from TiCl4 hydrolysis and then on the surface of Ti3C2Tx, amorphous TiO2 was coated. After hydrothermal treatment, anatase TiO2/Ti3C2Tx material was formed, as shown in Fig. 8. Amorphous TiO2 is micro-spherical and is coated on the surface of Ti3C2Tx, displaying a low-layered structure (Fig. 8). After water heat treatment, the whole structure forms into a brittle cake structure (Fig. 9).
Fig. 8 SEM images of (a) TiO2 (50 wt%), (d) Ti3C2Tx, (b) TiO2/Ti3C2Tx (5 wt%), and (c) TiO2/Ti3C2Tx.45 |
Fig. 9 (a) Photocatalytic hydrogen production rates and (b) recycling studies over the TiO2/Ti3C2Tx (5 wt%) sample.45 |
The TiO2/Ti3C2Tx material displays excellent photocatalytic hydrogen evolution capability with good cycling stability. The hydrogen production efficiency of TiO2/Ti3C2Tx-5% is about 4 times as high as that of pure phase TiO2, reaching 17.8 μmol h−1 gcatalyst−1. The hydrogen production efficiency of the 10% and 50% samples decreased slightly, which may be related to light energy absorption, as shown in Fig. 10(a). With the increase in Ti3C2Tx addition, the light absorption capacity of the samples in the 250–380 nm region is gradually decreased. This significant improvement in the hydrogen production efficiency is closely related to the smooth carriage of Ti3C2Tx. As shown in Fig. 10(b), after the formation of the Schottky barrier, the carrier separation efficiency is improved, thus improving its photocatalytic capacity.
Fig. 10 (a) PL spectra and (b) DRS spectra of TiO2, TiO2/Ti3C2Tx (5 wt%), TiO2/Ti3C2Tx (10 wt%), and TiO2/Ti3C2Tx (50 wt%).45 |
Thus, in conclusion, after loading with Ti3C2Tx, H2 production increased at least twice. Such an amazing promotion is mainly related with 3 aspects of Ti3C2Tx: (a) it supplies a high throughput channel as a co-catalyst for the excited electrons while the holes cannot pass the boundaries; (b) its hydrophilcity; and (c) the Gibbs free energy of Ti3C2Tx adsorption on hydrogen approaches zero infinitely.
Ti3C2Tx is also widely used in the photocatalytic CO2 reduction reaction. However, due to its own carbon source and instability, further research is needed to understand the mechanism of photocatalytic CO2 reduction of Ti3C2Tx.
In 2017, Zhang et al. summarized the CO2 reduction capacity of three MXene materials with surface groups, which terminate with –O through theoretical calculations.75 Among the three materials, Ti2CO2, V2CO2, and Ti3C2O2, Ti2CO2 showed the best photocatalytic CO2 reduction capacity. Of the two reduction paths76–78 shown in Fig. 12, the pathway of “CO2–HCOO–HCOOH” has a favorable energy barrier of about 0.53 eV.
Through DFT calculations, it was revealed that in the first step reaction of CO2 adsorption in CO2RR, the O atom of the CO2 molecule occupies an O defect position on the MXene. This mode in adsorption requires the lowest energy. The adsorption energies of the three materials were Ti3C2O2 (−0.73 eV), Ti2CO2 (−0.67 eV), and V2CO2 (−0.35 eV). Ti2CO2 has a lower adsorption energy compared to V2CO2 as the Ti atoms are more likely to lose electrons than the V atoms.
If the reaction proceeds via pathway 1 (Fig. 11), one of the oxygen atoms of the CO2 molecule is captured by the oxygen defect. This results in the breaking of the C–O bond, while CO is produced. In this step, Ti3C2O2 would lower the energy barrier of the C–O bond to about 0.86 eV. Pathway 2 (Fig. 11) has an energy barrier greater than 1 eV. In this pathway, the CO2 molecules are captured by an oxygen defect on the surface of MXene and are hydrogenated to form COOH. This is further hydrogenated and converted into the products CO and H2O. CO, which is produced, can further react to form HCOOH, HCOH, CH2OH, CH4, and other products.
Studies into the application of Ti3C2Tx in CO2RR is summarized in Table 2.
Photocatalyst | Products and yield (μmol g−1 h−1) | Activity improvement factor | Reaction conditions | Light source | Preparation method | Morphology | Monolayer or Multilayer | Year | Ref. |
---|---|---|---|---|---|---|---|---|---|
2D/2D Ti3C2 MXene/g-C3N4 nanosheet | CO (5.19) | 8.37 (CO) | 20 mg | 300 W Xe lamp (λ > 420 nm) | Calcination under N2 atmosphere | 2D/2D nanosheets | Monolayer | 2020 | 79 |
CH4 (0.044) | 2.09 (CH4) | catalyst gas–solid | |||||||
Alklinized Ti3C2/decorating g-C3N4 | CO (11.21 μmol g−1) | 5.96 (CO) | 40 mg catalyst gas–solid | 300 W Xe lamp (λ > 420 nm) | Alkali etching | 3D | Multilayer | 2019 | 80 |
CH4 (0.044 μmol g−1) | 5.6 (CH4) | ||||||||
TiO2/Ti3C2 | CO | — | 50 mg | 300 W Xe lamp | Calcination | Nanoparticles | Multilayer | 2018 | 81 |
CH4 (0.22) | catalyst liquid–solid | ||||||||
2D/2D ultrathin Ti3C2/Bi2WO6 | CO | 4.34 (CH4) | 100 mg | Xe lamp | Hydrothermal | Flat shape 2D structure | Monolayer | 2018 | 82 |
CH4 (1.78) | 6.28 (CH3OH) | catalyst liquid–solid | |||||||
CH3OH (0.44) | |||||||||
2D/2D/0D TiO2/C3N4/Ti3C2 | CO (4.39) | 1.39 (CO) (than TiO2/C3N4) | 30 mg catalyst liquid–solid | 300 W Xe lamp | Electrostatic self-assembly | 2D/2D structure | Ti3C2 quantum dots | 2020 | 83 |
CH4 (1.20) |
In 2018, Cao et al. prepared a 2D/2D heterogeneous junction of Ti3C2Tx/Bi2WO6 and the composite showed excellent photocatalytic CO2 reduction performance.82 As shown in Fig. 13, multi-layer structure Ti3C2Tx was tested with DMSO. After the formation of low-layer structure Ti3C2Tx, the oxygen-rich surface was negatively charged, which permitted Bi3+ to be adsorbed from hydrolyzed Bi(NO3)355H2O.84 After the addition of a tungsten source, a 2D/2D Ti3C2Tx/Bi2WO6 heterojunction was formed. The concurrent addition of CTAB furthermore ensures the ultrathin structure of both Bi2WO685 and Ti3C2Tx.86
Fig. 13 (a–c) Typical FESEM, AFM images, and height cutaway view of Bi2WO6, (d–f) Ti3C2 nanosheets, (g–i) TB2 (Ti3C2Tx/Bi2WO6), and (j) schematic illustration of the synthetic process.82 |
The successful preparation of heterojunctions greatly enhances the ability of Bi2WO6 to reduce CO2. The CH4 production of the sample TB2 reached 1.78 μmol h−1 g−1, while the yield of CH3OH reached 0.44 μmol h−1 g−1. The isotopic spectra of Fig. 14(b) and (c) indicates that the produced CH4 and CH3OH are formed from the photocatalytic reduction of CO2.
Fig. 14 (a) Photocatalytic activity of TB0 to TB5; (b) GC-MS spectra over TB2 after irradiation for several hours with different carbon sources; (c) GC-MS analysis of the reaction products with 12C and 13C as carbon sources.82 |
As shown in Fig. 15(a), Ti3C2Tx exhibits excellent light absorption performance between 200–800 nm. The light absorption capacity of Bi2WO6 was also significantly improved by carrying Ti3C2Tx. To be noted, as shown in Fig. 14(b), the fluorescence lifetime decreased after loading with Ti3C2Tx. This is because TC supplies a more efficient non-radiative decay pathway. In electrochemical tests, the photocurrent photoelectric impedance spectrum further revealed that the carriage of Ti3C2Tx greatly promoted the carrier strength of Bi2WO6. This further confirmed the successful construction of the Ti3C2Tx/Bi2WO6 heterojunction.
Fig. 15 (a) UV-Vis DRS of all the as-prepared samples; (b) TRPL spectra of TB0 and TB2; (c) EIS plots and (d) transient photocurrent of the prepared samples.82 |
Yang et al. prepared 2D/2D Ti3C2 MXene/g-C3N4 heterojunctions in 2020.79 As shown in Fig. 16(g), Ti3AlC2 was successfully etched to form Ti3C2, as indicated by the XRD patterns.87,88 2D g-C3N4 was found to grow on the surface of Ti3C2 under an atmosphere of N2. The formed 2D/2D Ti3C2 MXene/g-C3N4 demonstrated excellent photocatalytic CO2 reduction capability. As shown in Fig. 17, the photocatalytic performance of pure phase g-C3N4 for the production of CO and CH4 is only 0.62 μmol h−1 g−1 and 0.021 μmol h−1 g−1, respectively, in contrast to Ti3C2, wherein the production of CO and CH4 is 5.19 μmol h−1 g−1, 0.044 μmol h−1 g−1, respectively. The isotopic experiments confirm that the product is produced by the photocatalytic reduction of CO2.
Fig. 16 FESEM images of UCN (a) and 10TC (b) samples, AFM images and the corresponding height profiles of UCN (c and e), 10TC (d and f) samples, and (g) schematic illustration for the fabrication process.79 |
Fig. 17 Photocatalytic CO2 reduction performance of the as-prepared samples (a); cycling tests over the 10TC sample (b); GC-MS analysis of the products from the photoreduction of CO2 over 10TC using labelled 12CO2 and 13CO2 as the carbon sources (c and d).79 |
The tests of PL and TRPL showed that the composite rate of electron holes45,89 decreased significantly after carrying Ti3C2. As shown in Fig. 18(b), the fitted pure phase C3N4 had a lifetime of only 4.14 ns, while 10TC had a lifetime of 4.51 ns, which represents a significant increase in the lifetime of the carriers. This is closely related to the smooth carrying of Ti3C2. An excellent “storage capacitor” is produced when Ti3C2 forms a heterojunction with g-C3N4. When the electrons are transmitted to the semiconductor surface, they transfer to Ti3C2 quickly while the holes cannot. This greatly reduces the electron hole composite and improves the photocatalytic performance of the material. On the other hand, abundant defects on the Ti3C2 surface provide excellent sites for CO2 adsorption.
Fig. 18 PL spectra, EIS, and TPR plots of UCN, Ti3C2, and 10TC samples (a, c, and d); TR-PL spectra of UCN and 10TC (b).79 |
In conclusion, the application of Ti3C2Tx in CO2RR is relatively less than that of photocatalytic water splitting. This is because of its instability and its own carbon resources, which can cause interferences during the photocatalytic CO2 reduction reaction. As shown in Table 2, among limited reports, Ti3C2Tx with both single-layered structures and multi-layered structures shows an obvious production promotion. It is to be noted that there are no new products (such as C2 products, formaldehyde, and methyl ether) after loading with Ti3C2Tx compared to the base photocatalyst. This phenomenon confirms that Ti3C2Tx cannot change the energy barrier of the base photocatalyst for CO2 reduction. Thus, in general, the obvious promotion during CO2RR may be related to the two features of Ti3C2Tx: (a) abundant surface vacancies for CO2 adsorption and (b) promoting the separation of carriers.
Photocatalyst | Substrate of degradation | Removal rate (%)/rate constants (min−1) | Reaction conditions | Light source | Oxygenic species | Morphology | Monolayer or multilayer | Year | Ref. | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Ag3PO4/Ti3C2 | Methyl orange (MO) | (rate constants) 0.094 (MO) | 20 mg catalyst + 50 mL 20 mg L−1 substrate, 30 min dark adsorption | 300 W Xe lamp (λ > 420 nm) | h+ (main) | 2D Ti3C2/Ag3PO4 particles | Monolayer | 2018 | 87 | |||||||||||||||||
2,4-Dinitrophenol (2,4-DNP) | 2,4-DNP (0.005) | ˙OH | ||||||||||||||||||||||||
Tetracycline hydrochloride (TC–H) | TC–H (0.32) | |||||||||||||||||||||||||
Thiamphenicol (TPL) | TPL (0.0042) | |||||||||||||||||||||||||
Chloramphenicol (CPL) | CPL (0.025) | |||||||||||||||||||||||||
Ti3C2/SrTiO3 composites | UO22+ | (Removal rate) | 20 mg catalyst + 60 mL 50 ppm substrate, 8 hours dark adsorption | 300 W Xe lamp (λ = 320–2500 nm) | ˙OH | 2D Ti3C2/SrTiO3 particles | Multilayer | 2019 | 97 | |||||||||||||||||
77% in 180 min | ||||||||||||||||||||||||||
Ti3C2–OH/Bi2WO6 composites | Rhodamine B | (Rate constants) 0.0596 | 10 mg catalyst + 50 mL 2 × 10−5 mol L−1 substrate, 30 min dark adsorption | 300 W Xe lamp (λ = 400–2500 nm) | h+ | Porous spherical structure | Ti3C2–OH | 2019 | 98 | |||||||||||||||||
MoS2@Ti3C2 Nanohybrid | Liquid paraffin (LP) | (Rate constants) 0.0476 | A certain amount of sample + 2.0 g of deionized water and 1.0 g of LP + 10 mL dichloromethane as the sacrifice reagent, 30 min dark adsorption | 1000 W high-pressure mercury lamp | MoS2 nanosheets/Ti3C2 sheets | Multilayer | 2019 | 99 | ||||||||||||||||||
0D/2D Bi3TaO7/Ti3C2 | Methylene blue | (Rate constants) 0.032 | 50 mg catalyst + 100 mL 10 mg L−1 | 300 W Xe lamp (λ > 420 nm) | ˙OH | Bi3TaO7 nanoparticles/Ti3C2 nanosheets | Multilayer | 2020 | 100 | |||||||||||||||||
Substrate, 60 min dark adsorption | ||||||||||||||||||||||||||
2D/2D Ti3C2/Porous g-C3N4 | Phenol | (Rate constants) 0.022 | 20 mg catalyst + 50 mL 10 mg L−1 | 500 W Xe lamp (λ > 400 nm) | 2D/2D Ti3C2/PCN nanocomposite | Multilayer with ultra-sonication | 2020 | 101 | ||||||||||||||||||
Substrate, 30 min dark adsorption | ||||||||||||||||||||||||||
CdS@Ti3C2@TiO2 | Sulfachloropyridazine (SCP) | (Removal rate) SCP (about 95% in 60 min) | 50 mg catalyst + 200 mL 20 mg L−1 | Light intensity 300 mW cm−2 (λ = 400–1050 nm) | ˙O2− | CdS nanoparticles/Ti3C2@TiO2 bulk | Bulk Ti3C2@TiO2 | 2019 | 102 | |||||||||||||||||
Methylene blue (MB) | MB (about 80% in 60 min) | |||||||||||||||||||||||||
Rhodamine B (RhB) | RhB (about 99% in 60 min) | substrate, 30 min dark adsorption | ||||||||||||||||||||||||
Phenol | Phenol (about 50% in 60 min) | |||||||||||||||||||||||||
(111) TiO2-x/Ti3C2 | Mmethylene blue (MB) | (Removal rate) MB (75% in 150 min) | 10 mg catalyst + 200 mL 20 mg L−1 | 500 W Xe lamp (λ > 400 nm) | ˙OH (main) | TiO2 nanoparticles/Ti3C2 nanosheets | Multilayer | 2017 | 103 | |||||||||||||||||
(001) TiO2/Ti3C2 | Methyl orange (MO) | (Rate constants) 0.018 | 10 mg catalyst + 200 mL 20 mg L−1 substrate, 60 min dark adsorption | 300 W Xe lamp | ˙OH (main) | TiO2 square nanosheets/Ti3C2 nanosheets | Multilayer | 2016 | 104 | |||||||||||||||||
Graphene layers anchored TiO2/g-C3N4 | Rhodamine B (RhB) | (Rate constants) 0.0559 (RhB) | 10 mg catalyst + 200 mL (RhB 20 mg L−1, TC 10 mg L−1, CIP 3 mg L−1, BPA 5 mg L−1), 60 min dark adsorption | 300 W Xe lamp (λ > 400 nm) | ˙OH | 3D bulk | Bulk Ti3C2@TiO2 | 2020 | 105 | |||||||||||||||||
Tetracycline (TC) | 0.0244 (TC) | ˙O2− | ||||||||||||||||||||||||
Ciprofloxacin (CIP) | 0.0168 (CIP) | h+ | ||||||||||||||||||||||||
Bisphenol A (BPA) | 0.0194 (BPA) | |||||||||||||||||||||||||
2D/2D Ti3C2/MoS2 | Methylene orange (MO) | (Rate constants) 0.00836 | 50 mg catalyst + 50 mL 20, 30, 50 mg L−1 substrate, 30 min dark adsorption, 60 min dark adsorption | 400 W metal halide lamp | h+ | Flower-like nanosphere | Multilayer | 2020 | 106 | |||||||||||||||||
˙OH | ||||||||||||||||||||||||||
α-Fe2O3/ZnFe2O4@Ti3C2 | Rhodamine B (RhB) | (Rate constants) 0.02686 (RhB) (Removal rate) | 20 mg catalyst + 100 mL 10 mg L−1 substrate, 30 min dark adsorption | 300 W Xe lamp (λ > 400 nm) | ˙O2− | α-Fe2O3/ZnFe2O4 nanoparticles/Ti3C2 nanosheets | Multilayer | 2019 | 107 | |||||||||||||||||
Cr(VI) | Cr(VI) | ˙OH | ||||||||||||||||||||||||
Light off: about 70% in 90 min | h+ | |||||||||||||||||||||||||
Light on: about 90% in 90 min |
In 2018, Cai et al. produced a Ag3PO4/Ti3C2 composite photocatalyst, which possessed excellent photocatalytic degradation performance.87
As shown in Fig. 19, after DMSO and sonication treatment, Ti3C2 with a low-layer structure was formed. After the addition of silver nitrate, Ag+ was adsorbed due to the negative charge on the surface of Ti3C2. Ag3PO4 nanoparticles were grown in situ, forming a heterojunction between the Ag3PO4 nanoparticles and the Ti3C2 nanosheets.
Fig. 19 TEM images of (a) bulk Ti3C2, (b) single Ti3C2 sheet, (c) Ag3PO4/Ti3C2 composite. (d) EDX spectra of the Ag3PO4/Ti3C2 composite and (e) the schematic representation of single 2D Ti3C2 sheets and Ag3PO4/Ti3C2 synthesis.87 |
As shown in Fig. 20, the heterojunction of the Ag3PO4 nanoparticles/Ti3C2 nanosheets shows a photocatalytic degradation stage rate K of 0.094, 0.005, 0.32, and 0.0042 min−1 for methyl orange (MO), 2,4-dinitrophenol (2,4-DNP), tetracycline (TC–H), thiamphenicol (TPL), and chloramphenicol (CPL), respectively. According to EPR analysis, the hydroxyl radical (˙OH) plays an important role in the oxidation system, as shown in Fig. 20(f). This may be related to abundant Ti defects on the Ti3C2 surface. Ti sites exposed on the surface of Ti3C2 have strong redox reactivity, which promotes multiple electron reduction reactions (O2 → H2O2 → ˙OH).
Fig. 20 Photocatalytic degradation of various pollutants by the as-prepared catalysts. (a) CPL, (b) TPL, and (c) TC-H degradation efficiency in the presence of the as-prepared catalysts under visible light irradiation (λ > 420 nm). HPLC chromatogram of (d) CPL and (e) TPL under different degradation times using the as-prepared catalysts. (f) UV-vis absorption spectra of TC–H under different degradation times using different catalysts.87 |
As shown in Fig. 21, PL, TRPL, and the electrochemical characterization spectra indicate that the carrier separation efficiency of the material is significantly improved after carrying Ti3C2. This may be attributed to (i) the abundant surface hydrophilic functional groups of the Ti3C2 construct, which have strong interfacial contact with Ag3PO4, facilitating the separation of carriers; (ii) the strong redox reactivity of the surface Ti sites, which promote multiple electron reduction reactions to induce more ˙OH production; and (iii) a Schottky junction formed at the Ag3PO4/Ti3C2 interface enabling efficient transfer electrons to the Ti3C2 surface. This inhibits the photocossion of Ag3PO4 caused by photogeneration electrons.
Fig. 21 UV-vis diffuse reflectance spectra (a), PL spectra (b), time-resolved PL decay spectra (c), EIS Nyquist plots (d), transient photocurrent responses (e) of the as-prepared catalysts, and DMPO spin-trapping ESR spectra for DMPO–˙OH in the Ag3PO4/Ti3C2 system in the presence or absence of HA.87 |
Under high temperature conditions, Ti in the Ti3C2 skeleton layer is oxidized into TiO2, while C still exists in the form of a graphene-like layer. Therefore, under high temperature conditions, Ti3C2 can be converted into amorphous TiO2 anchored within the graphene-like layer. In 2020, Wu et al. took advantage of this material, which displayed excellent photocatalytic degradation performance.105
As shown in Fig. 22(e) and (f), high temperature treated Ti3C2 still retains its morphology and a 3D block-shaped morphology is formed after carrying g-C3N4 (Fig. 23).
Fig. 22 SEM images of C3N4 (a and b), Ti3C2 (c and d), heated Ti3C2 (e and f), and graphene layers anchored TiO2/g-C3N4 (g and h).105 |
Fig. 23 The photocatalytic degradation performance of TC (a), CIP (b), BPA (c), and RhB (d) by photocatalysts under visible light irradiation.105 |
Graphene layers anchored to TiO2/g-C3N4 show first-order kinetic constants for the degradation of rhodamine B (RhB), tetracycline (TC), ciprofloxacin (CIP), and bisphenol A (BPA) of 0.0559, 0.0244, 0.0168, and 0.0194 min−1, respectively. According to the EPR test results (Fig. 24(a)–(d)), the oxygen active species that play a role in the oxidation process mainly include ˙O2− and ˙OH. Furthermore, signals corresponding to the holes (h+) were also detected. The contribution to the degradation of these test molecules appears to be in the order of ˙O2− > h+ > ˙OH.
Fig. 24 Trapping experiment for the photocatalytic degradation of TC over GTOCN3 (a); ESR spectra of CNTOC3 for (b) DMPO–˙O2−, (c) TEMPO–h+, (d) DMPO–˙OH in the dark and under visible light irradiation, I–T curves under visible light irradiation (e) and the EIS response (f) of the samples.105 |
High-temperature treated Ti3C2 has a greatly enhanced light absorption capacity, whilst the carrier separation efficiency and transmission efficiency are also improved. The improvement of the photocurrent (Fig. 24(e) and (f)) also verified that the graphene layers anchoring TiO2 lead to the formation of a heterogeneous junction. This is due to the change in the electric field between g-C3N4.
In conclusion, as a co-catalyst, the application of Ti3C2Tx in photocatalytic degradation is mainly due to its three characteristics: (a) in a liquid–solid phase reaction, its hydrophilicity makes it easy for the adsorption or contact between the pollutants and photocatalysts; (b) high throughput electron transfer makes it easier to generate concentrated holes (h+); and (c) Ti sites exposed on the surface of Ti3C2 have strong redox reactivity, which promotes multiple electron reduction reactions, such as the reaction of activating molecular oxygen (O2→H2O2→˙OH).
The further application of Ti3C2Tx in photocatalysis depends on the development of the material itself. Methods to improve the stability of the Ti3C2Tx structure need to be explored, starting from synthetic methods. In addition, the rich groups on the surface of Ti3C2Tx and its hydrophilicity should be further explored, particularly in photocatalytic liquid phase reactions.
During photocatalytic CO2RR, although both monolayer-structured and multilayer-structured Ti3C2Tx exhibit high performance, isotope detection shows that some carbon resources come from CO2 molecular; thus, there are still some ambiguities and other possibilities. For example, the valence state of “C” in Ti3C2Tx is mostly “−4”, which makes it possible for CO2 to react with Ti3C2Tx in order to form the CO as product; this pathway involves redox reaction rather than catalysis.
This, in all, the mechanism needs to be explored more, both during the photocatalytic reaction and the oxidation of Ti3C2Tx itself.
Ti3C2Tx is the earliest material in the MXene family; thus, the improvement of its application in photocatalysis represents a great significance for the application of the whole family.
This journal is © The Royal Society of Chemistry 2021 |