Low Temperature Step Annealing Synthesis of the Ti 2 AlN MAX Phase to Fabricate MXene Quantum Dots

: We present the synthesis of the Ti 2 AlN MAX phase using two-step annealing at temperatures of 600 ◦ C and 1100 ◦ C, the lowest synthesis temperatures reported so far. After the successful synthesis of the Ti 2 AlN MAX phase, two-dimensional Ti 2 N MXene was prepared through wet chemical etching and further fragmented into light emitting MXene quantum dots (MQDs) with a size of 3.2 nm by hydrothermal method. Our MQDs displayed a 6.9% quantum yield at a 310 nm wavelength of excitation, suggesting promising nanophotonic applications.


Introduction
The MAX phases are a family of advanced ceramic materials comprised of hexagonal layered transition metal carbides, nitrides, and carbonitrides, which have been explored for more than a decade [1,2]. Generally, the MAX phases are denoted as M n+1 AX n , where M is an early transition metal (Sc, V, Ti, etc.), A is an element from group 13-16, and X is either nitrogen or carbon with a stoichiometry of n = 1, 2, 3 [3]. The crystal structure of the MAX phases contains a hexagonal lattice with space group P63/mmc in which adjacent M n+1 X n octahedral sites are interleaved with atomic monolayers of A atoms [4]. The MAX phases have been given considerable attention owing to their unique and interesting combination of metallic as well as ceramic properties [5]. Because of their metallic character, MAX phases exhibit high thermal and electrical conductivity along with thermal shock resistance. The ceramic property of the MAX phase material enables it to possess remarkable oxidation and corrosion resistance at high temperature operations [6]. Owing to their fascinating properties, more than 155 MAX phases have currently been synthesized [5].
As MAX phases possess both ceramic and metallic properties, their synthesis at low temperatures is difficult. Previous studies have reported the presence of significant amounts of impurities such as TiN, Ti 3 Al, TiAl, AlN, Al 2 Ti, and Al 3 Ti [2,15,21] during the synthesis of MAX phases at temperatures lower than 1400-1500 • C. Thus, most of the previous methods for synthesizing Ti 2 AlN require a high temperature, which entails much more investment and is not easily available in common laboratories. One of the main hindrances to synthesizing MAX phase materials is the use of an expensive apparatus because commonly available furnaces have temperature limitations and can only be used at temperatures of 1100-1200 • C or less. Thus, there is always a strong demand for new, easily accessible, simple, and low-cost methods for the synthesis of Ti 2 AlN MAX phases with better crystal quality and lower temperature (~1100 • C) systems compared to the methods described in previous works.
In this study, we introduce a new step annealing method to synthesize the Ti 2 AlN MAX phase at a relatively lower temperature compared to previously reported methods and soaking times. This method includes a two-step annealing process at 600 • C and 1100 • C, which is a comparatively lower temperature than previous techniques, and yields a better quality Ti 2 AlN MAX phase with completely removed unavoidable impurities such as TiN, which were mostly obtained during synthesis at low temperatures. The as-synthesized Ti 2 AlN MAX phase was etched out using wet chemical etching to prepare its corresponding 2D MXene sheets, which were further decomposed into light emitting MQDs by a hydrothermal process. The formation of the Ti 2 AlN samples, Ti 2 N Mxene sheets, and MQDs was further confirmed and comprehensively visualized and studied by X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), high-resolution transmission electron microscopy (HR-TEM), Raman spectroscopy, photoluminescence (PL) spectroscopy, X-ray photon spectroscopy (XPS) and Ultraviolet-visible (UV-vis) spectroscopy.

Synthesis of Ti 2 AlN MAX Phase via Step Annealing Method
Commercial powders of elemental titanium (Ti, −325 mesh, Sigma Aldrich, St. Louis, MO, USA), aluminum (Al, −325 mesh, Sigma Aldrich), and aluminum nitride (AlN, Sigma Aldrich) were used as the initial precursors for the preparation of the Ti 2 AlN MAX phase. All powders were used as received without further treatment.
The precursor ratio of starting powder mixtures of 2Ti + 0.8Al + AlN was used. Since Al could evaporate and dissociate during the processing, which inhibits the single-phase synthesis of Ti 2 AlN [47], an excess amount of Al was introduced to compensate for the Al loss. Ti, Al, and AlN powders were weighed on a precision scale (0.0001 g) at a molar ratio of 2:0.8:1. Powder mixtures and zirconia balls were sealed in a dry environment with a dew point of nearly −55 • C to avoid the introduction of moisture during the milling process. The powders were thoroughly ball-milled in a stainless-steel vessel by using a Fritsch Pulverisette planetary mill [11]. The powders were finely milled using a unique set of milling conditions in two steps, using a powder-to-ball ratio of 1:4. In step 1, a low rpm of 270 was used, and the powders were milled for 6 h, whereas in step 2, a high rpm of 400 was used for another 6 h of milling. These steps in the milling process enabled the thorough mixing of the precursors. The high-rpm step plays a significant role in the mechanical activation [48] of the mixed powders. Mechanical activation helps the disintegration of constituents to yield smaller sized powders while increasing the active surface area. Consequently, decreasing the time and the temperature of synthesis are desirable for Ti 2 AlN synthesis. The mixed powders were placed in a stainless-steel die and compacted into 13 mm diameter pellets by applying a uniaxial pressure of 4 MPa.
For this particular step annealing synthesis method, the pellets were placed in an alumina crucible and placed in a double-tube furnace, where a constant flow of argon gas was maintained throughout the thermal annealing process. This method includes two different steps, involving different heating rates and temperatures. First, a heating rate of 5 • C/min was used to reach 600 • C with a soaking time of 1 h. Second, a relatively high heating rate of 20 • C/min was used to reach 1100 • C from 600 • C, and the sample was soaked for 3 h at that temperature. After completion of the synthesis process, the pellets were naturally cooled to room temperature and ground with an agate mortar pestle to obtain Ti 2 AlN MAX phase powder.

Synthesis of MXene and MQDs
Ti 2 N MXene was prepared by chemical etching of the Al layers from the Ti 2 AlN MAX phase by following a previously reported method [10]. During the chemical etching process, the etchant was prepared by dissolving 6 g KF in 5M HCl then 1 g of the MAX phase was added to 20 mL of the prepared etchant solution and kept undisturbed for 3 h. Subsequently, the etching solution with the MAX phase was bath-sonicated for 1 h at 40 • C to accelerate intercalation, which further assisted the exfoliation process. Subsequently, the solution was centrifuged and washed with DI water until the pH became neutral. Finally, the MXene sheets were filtered and dried overnight at 40 • C. The yield of MXene formation was found to be approximately 70%, which is comparable to the previously reported yield [10].
MQDs were synthesized using a hydrothermal method followed by centrifugation [35]. The obtained Ti 2 N MXene after Al etching from Ti 2 AlN MAX was dispersed in DI water, and the pH of the solution was adjusted to 10 by the addition of ammonium hydroxide (NH 4 OH) solution. This solution was poured into a Teflon container that was placed in an autoclave. The autoclave was then placed in an oven at 100 • C for 6 h and cooled to room temperature. After cooling, the solution was centrifuged (10,000 rpm) for 20 min, and the MQDs were collected as dispersed in the supernatant. The supernatant was carefully separated and stored in a refrigerator for frequent use for further characterization. The XRD patterns of the Ti 2 AlN MAX phase and corresponding Ti 2 N MXene were recorded using a Bruker AXS diffractometer (D2 PHASER A26-X1-A2B0B2A Karlsruhe, Germany) at room temperature. Raman spectra were recorded using a WITec confocal microscope system (Alpha-300 SR, WITec GmbH, Ulm, Germany) equipped with a 100× objective lens (N.A. = 0.9) with an excitation wavelength of 532 nm. UV-Vis absorption spectra of the Ti 2 N MQDs were recorded in deionized water (Optizen UV-Vis spectrometer). PL and PLE spectra were measured using a Cary Eclipse spectrometer. The quantum yield of the Rhodamine 6G (R6G) solution was measured using a Quantaurus-QY, Hamamatsu Photonics spectrometer. FE-SEM (JSM-7600F, JEOL Corp., Tokyo, Japan) images of the Ti 2 AlN MAX phase and Ti 2 N MXene were captured at field voltages ranging from 5 kV to 10 kV. XPS of MQDs was recorded using X-ray photon spectrometer (ESCALAB250). The HR-TEM (JEM-3010, JEOL Corp.) images of the MQDs were obtained at 300 kV.

Results and Discussion
The schematic diagram in Figure 1 illustrates a brief process of the new step annealing method for synthesizing the (211) Ti 2 AlN MAX phase beginning from the precursors. This is comprehensively explained in the experimental section. Generally, MAX phase ceramics are synthesized at high temperatures [49], and it is very difficult to obtain relative purity at lower temperatures. As illustrated in the schematic diagram, the 3D ceramic MAX phase of Ti 2 AlN was prepared using a unique step annealing method in an argon environment after pelletizing the precursors. In this particular method, a temperature soaking step at 600 • C for 1 h was maintained in step 1 because the melting point of Al is relatively low compared to other precursors, that is, 660 • C, which causes inevitable weight loss at higher temperatures [47]. The desorption of Al occurs at temperatures greater than 600 • C because of its high vapor pressure [50]. The absence of Al at high temperatures can change the composition of the reaction system. To overcome this phenomenon, the precursor ratio was initially adjusted and excess of Al was added to the composition. The increased formation of TiN phase impurities occurs in the temperature range of 900 • C to 1000 • C [50]. Therefore, from step 1 to step 2 of annealing, a higher step rate of 20 • min −1 was used to inhibit TiN formation. In step 2, the temperature soaking step was maintained at 1100 • C for 3 h to complete the MAX phase synthesis. All the diffraction peaks of the as-synthesized max phase can be indexed to the Ti 2 AlN reflections from JCPDS card no. 00-055-0434, which indicates the high purity of the as-synthesized Ti 2 AlN MAX phase. This is consistent with previously reported XRD results [10,51]. The c-lattice parameter (c-LP) of the two-step annealed Ti 2 AlN presented in Table 1 was calculated to be 13.59 Å using "unit cell" software from the XRD patterns explained in Figure 2a, and it closely matches with the reported Ti 2 AlN c-LP value [21]. This is strong evidence of high purity of the as-synthesized MAX phase. For comparison, Ti 2 AlN was also synthesized by a typical annealing process at 1100 • C without any step; its XRD pattern (red profile) is also shown in Figure 2a, which contains a prominent TiN peak at 42.6 • and other impurities such as Ti 2 Al 5 at 43.3 • along with many unspecified peaks. This comparative study of the XRD patterns clearly shows the improvement of the step annealing method over the normal annealing method. The FE-SEM image shown in Figure 2b depicts the layered compact structure, which is a typical signature of the MAX phases [4], confirming the successful synthesis of the 3D MAX phase using the unique step annealing method. Because the peak positions in the Raman spectra are correlated to the vibrational modes distinctive to a specific material, Raman spectroscopy is a powerful tool for identifying and characterizing the fingerprint of materials. The vibrational modes for the 211 MAX phases are theorized to be 24, but only four modes are Raman active and can be observed [52]. Of these four modes, ω 3 is infrared and Raman active, whereas the ω 4 , ω 1 , and ω 2 modes are only Raman active [52]. For the Ti 2 AlN MAX phase, the ω 3 and ω 4 modes are associated with the parallel and perpendicular vibrations of the Ti-Ti planes, whereas the ω 1 and ω 2 vibrational modes arise from the shear vibrations of the Ti-Al planes [10,51,53]. The representative Raman spectra of step annealed Ti 2 AlN are shown in Figure 2c, where all four Raman active modes are clearly observed at the peak positions of 149 cm −1 (ω 1 ), 229 cm −1 (ω 2 ), 234 cm −1 (ω 3 ), and 365 cm −1 (ω 4 ), providing strong evidence for the formation of the Ti 2 AlN MAX phase [10,51,54].  The step annealed Ti 2 AlN MAX phase was further etched by wet chemical etching to obtain multilayered Ti 2 N MXene sheets. In this particular method, the Al layer was removed by acid-salt treatment of the Ti 2 AlN MAX phase in a KF-HCl etchant solution [10,26,45,55]. The effective formation and phase composition of multilayered Ti 2 N MXene sheets were analyzed using XRD, FE-SEM, and Raman spectroscopy respectively, and the corresponding results are demonstrated in Figure 3. The prominent peaks observed in the XRD pattern were consistent with previously reported results for Ti 2 N MXene, and they were indexed according to the JCPDS card no. 00-017-0386 of Ti 2 N, as shown in Figure 3a. Comparative analysis of the XRD patterns of the Ti 2 AlN MAX phase and Ti 2 N MXene shows that the sharp peak assigned to the (0 0 2) peak in the Ti 2 AlN MAX phase completely disappeared, and a broad curve was observed at that position. This observation may have resulted from the removal of the Al layer between two consecutive Ti 2 N MXene layers [26]. Similarly, the intense peak assigned to the (1 0 3) plane observed in the Ti 2 AlN MAX phase was completely absent in the XRD pattern of Ti 2 N MXene, which is significant evidence of the effective etching of the Al layer from the Ti 2 AlN MAX phase [10,26]. A new peak at approximately 54.4 • marked with "*" in Figure 3a is observed, which is assigned to the XRD signal of the potassium fluoride (KF) residue [56], which was used along with HCl in the etchant. This may have remained in the exfoliated Ti 2 N MXene layers during the washing process. The formation of Ti 2 N MXene was further confirmed by FE-SEM. The FE-SEM image in Figure 3b shows the typical morphological characteristics of MXenes [55,57,58]. The etched Ti 2 N MXene showed visible delamination and increased space between the layers owing to the removal of the Al layer. In some areas, stacked layers appear instead of the delaminated and opened layers. This may be attributed to the intercalation of K + or water ions between the Ti 2 N layers during the etching process. The presence of only two vibration modes at 222 cm −1 (ω 3 ) and 355 cm −1 (ω 4 ) confirms the formation of Ti 2 N MXene sheets after etching the Al layer from the Ti 2 AlN MAX phase observed in the Raman spectrum shown in Figure 3c. Comparative analysis of the Raman spectra of the Ti 2 AlN MAX phase and the as-synthesized Ti 2 N MXene sheets shows that the Ti-Al vibration mode (ω 1 ) is completely absent in the case of the Ti 2 N MXene sheets, indicating the complete removal of the Al layer and formation of the Ti 2 N MXene structure. Simultaneously, a clear red shift in the peak positions of ω 3 and ω 4 was observed for the Ti 2 N MXene sheets compared to its parent MAX phase, clearly showing better etching of the Al layer along with exfoliation [10]. The exfoliated Ti 2 N MXene sheets were further decomposed into 0D Ti 2 N MQDs using a simple hydrothermal method similar to the synthesis of QDs of other 2D materials, such as graphene and TMDs [30,59,60]. The effective synthesis of Ti 2 N MQDs is based on two crucial steps: first, the Al layer is etched from the 3D Ti 2 AlN MAX phase to obtain 2D Ti 2 N MXene, and second, the MXene nanosheets are broken down into ultra-small Ti 2 N MQDs, which determines the quality of the MQDs [26]. Since MXene is metastable it may yield its oxidation products in alkaline medium at high temperature reaction system. However, we emphasize that the hydrothermal reaction process at 100 • C using NH 4 OH solution was adopted for the synthesis of MQDs where surface-terminated -NH groups from the NH 4 OH solution maintain the crystal structure of pristine MXene, preventing the formation of TiO 2 or TiO 2 quantum dots [35]. The morphology and size of the Ti 2 N MQDs were analyzed by HR-TEM and are presented in Figure 4a,b, respectively. The low-and high-magnification HR-TEM images in Figure 4a,b reveal the formation of uniform and ultra-small Ti 2 N MQDs, respectively. The fringes observed in the single MQD in the HR-TEM image presented in Figure 4b are highly parallel with separation distance of 0.26 nm that corresponds to (1 0 1) lattice space of the Ti 2 N [26] and reveal good crystallization of the Ti 2 N MQDs. The fast Fourier transform (FFT) pattern of the MQDs is shown in the inset of Figure 4b. The average lateral diameter of the Ti 2 N MQDs was determined to be 3.2 ± 0.2 nm from the bell-shaped size distribution curve, as shown in Figure 4c, which matches the size of the single MQD shown in Figure 4b. The element distribution was measured by XPS and the content of nitrogen in the MQDs was approximately 17.6%, which is comparable with the previously reported value of Ti 2 N MXene [10]. To study and understand the optical properties of the as-synthesized Ti 2 N MQDs, absorption, PLE, and PL spectra were acquired from aqueous Ti 2 N MQDs, and the results are presented in Figure 5. As-synthesized Ti 2 N MQDs exhibited the PL emission with the peak wavelength around at 420 nm upon the excitation wavelength between 390 nm to 230 nm (Figure 5c). The demonstration of PL behavior indicated the successful fabrication of light emitting Ti 2 N MQDs. The light emitting properties were observable due to quantum confinement effect which caused the bandgap expansion of semi-metal MXene that were previously observed from the fabrication of other MQDs and 2D material-derived QDs [26,33,35,46,60,61]. PLE spectrum is the plot of PL intensity at a fixed detection wavelength vs. the excitation wavelength. Because PLE intensity corresponds to the probability of optical transition, PLE spectra resemble the absorption spectra but with higher signal-tonoise ratio [62]. In our case, we performed PLE measurements to understand the correlation between PLE with PL and UV-vis absorption spectra. The measurement was done by fixing the detection wavelength of 410 ± 10 nm with varying excitation wavelengths ranging 225-400 nm. As shown in Figure 5a (black curve), the PLE spectrum displayed the two distinct peaks at 233 nm and 310 nm, which were consistent with the two absorption edges of the absorption spectrum (blue curve) [63]. The PLQY of 3 µM concentration of R6G solution in methanol has been found to be~100% under visible excitation and can be used as the reference material to estimate the PLQY of other emissive materials [64]. The PLQY of MQDs was estimated using the following equation [65], which is presented in Figure 5b.
Here, Φ MQDs is the QY of the MQDs to be measured, Φ Ref is the QY of the reference, I PL MQDs represents the PL intensity of the MQDs, I PL Re f represents the PL intensity of the reference, A MQDs is the absorption of the MQDs, and A Ref is the absorption of the reference.
The maximum PLQY was estimated to be~6.9% with 310 nm excitation, which is consistent with the PLE and PL results presented in Figure 5a. Furthermore, excitation wavelength (within UV, ranging from 230 nm to 390 nm) dependent PL spectra were obtained, which are broad in shape owing to the inhomogeneous size distribution of MQDs in aqueous solution [61,66], as shown in Figure 5c. Red shifts in the PL peak positions along with a decrease in intensity were observed for the excitation wavelengths from 310 nm to 390 nm, which may be due to the various sizes of the MQDs dispersed in the aqueous medium. Under the higher wavelength (lower energy) excitation, the small MQDs with a large band gap will be less excited, resulting in a lesser intensity due to the loss of emission at short wavelengths along with red-shifted peak positions and vice-versa. Remarkably, we observed a high intensity of the emission under excitation at~230 nm and 310 nm, which may be due to most of the MQDs being excited under this illumination. This is quite consistent with our absorption, PLE, and PLQY results shown in Figure 5a,b.

Conclusions
The Ti 2 AlN MAX phase was prepared by two-steps annealing method at a significantly lower temperature than that used in previous methods. Furthermore, 2D MXene and MQDs were obtained that displayed efficient PL upon UV light absorption. The facile method of light emitting MQDs reported here will help expand the use of MXene and MQDs in nanophotonic applications.

Conflicts of Interest:
The authors declare no conflict of interest.