High crystallinity of 2D α-RuCl₃

April 26, 2023 - α-RuCl₃ is considered particularly interesting for fundamental studies of magnetic properties in low-dimensional structures. A team of researchers from the Universities of Heidelberg, Ulm and Kassel as well as the Max Planck Institute for Solid State Research and the Bundeswehr University in Germany, Trinity college in Ireland and the National Institute of Technology Durgapur (India) have demonstrated sonication-assisted liquid exfoliation with confirmed structural integrity.

Layered ruthenium chloride (α-RuCl₃) is a candidate for quantum spin liquid states due to the triangular arrangement of magnetic ions and strong relativistic spin-orbit coupling. The spin-1/2 model with bond-dependent spin interactions features a quantum spin liquid (QSL) ground state, which can be considered as a topological state with the potential for coherent transport of quantum information and fractional excitations [1]. The latter can be understood as an exotic phenomenon that splits an electron into well-defined quasiparticles. In the case of a Kitaev 2D honeycomb lattice[2], this supports the formation of gauge fluxes and Majorana fermions [3]. While fractional excitation has so far only been described in theory, the prospect of experimentally realizing such interactions in α-RuCl₃ (or any other material) could be considered a breakthrough in quantum computation technology [4].

However, apart from such exotic interactions, non-Kitaev interactions in RuCl₃ also lead to extensive antiferromagnetic ordering at low temperatures [5]. Previously, the ordering in single crystals was found to proceed via a single sharp phase transition at a Neel temperature of T_N = 7 K, as indicated by magnetic susceptibility and specific heat measurements [5], or via a sequence of two phase transitions occurring at T_N1 ≈ 14 K and T_N2 ≈ 8 K, respectively [6-9]. It has been argued that the number, position, and shape of the magnetic transitions change as a function of the stacking order and stacking errors of the hexagonal layers of Ru³⁺ ions in single-crystalline samples. However, in polycrystalline samples, single antiferromagnetic phase transitions are reported at T_N = 11 K [10], T_N = 13 K [11], and T_N = 15.6 K [12], respectively.

While most studies on the optical and magnetic properties of α-RuCl₃ have focused on the bulk material, little is known about the properties of the material after exfoliation. This is an important aspect since delamination of the crystal is associated with a reduction in stacking disorder, which is thought to be responsible for an enhanced appearance of antiferromagnetic ordering [5, 9]. Although it has been recently shown that α-RuCl₃ can be exfoliated by chemical [9] and electrochemical exfoliation [13], in both cases the material loses its original electronic and magnetic properties due to electron doping. To avoid this, there is an alternative strategy for liquid phase exfoliation (LPE) based on mechanical delamination by the forces exerted on the starting material due to cavitation bubble collapse [14] and colloidal stabilization by suitable surfactants or solvents [15]. This method has been applied to over 15 classes of layered materials that have been successfully exfoliated into 2D nanostructures, including transition metal dichalcogenides (TMDs) [19], transition metal hexathiohypodiphosphates (M P S226 ) [21, 22], III-VI and IV-VI semiconductors [23], pickntogens [24], silicates [25], oxides [26], hydroxides [27], and other minerals [28], as well as synthetic structures such as organic polymers [29], covalent organic 2D frameworks [30], and others. Usually, no phase transition or heavy doping is observed, so this could be a suitable approach to study the layer number dependent properties of α-RuCl₃.

The researchers applied additive-free delamination of α-RuCl₃ by sonication-assisted liquid-phase exfoliation (LPE) under inert conditions. To investigate the effect of layer number on the optical and magnetic response, samples were selected for size using previously established centrifugation-based techniques [31]. The lateral size and thickness distribution of the α-RuCl₃ nanosheets in the fractions was quantified by statistical atomic force microscopy (AFM). The structural integrity of the nanosheets was confirmed by a combination of transmission electron microscopy (TEM), Raman spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). In addition, the magnetic susceptibility of α-RuCl₃ was investigated for three samples with different nanosheet size distributions in dispersion and for drop-cast samples, respectively. In addition, absorbance and absorbance measurements were performed on nanosheet dispersions to reveal systematic changes in the photospectroscopic response. From this, parameters for the material concentration and the average number of layers can be derived. Finally, time- and temperature-dependent absorption measurements were used to investigate and quantify the environmental stability for different nanosheet sizes.

Nanosheets of α-RuCl₃
Ground α-RuCl₃ single crystals were exfoliated under inert gas conditions in dried, distilled and degassed N-cyclohexyl-2-pyrrolidone (CHP) using bath sonication. The crystallinity of the starting material was confirmed by Raman spectroscopy and powder XRD (see Figure 1). CHP was chosen as the exfoliation medium because pyrrolidone-based solvents are known to be suitable stabilizing solvents for the LPE of various layered crystals and can also promote surface passivation, suppressing material decomposition [21]. After exfoliation, the obtained parent nanomaterial dispersion was size-selected by liquid cascade centrifugation (LCC) [31] to divide the initial polydisperse nanosheet mixture of different sizes and thicknesses into fractions with well-defined size distribution. For this purpose, successive centrifugation steps were performed, starting with a step at low centrifugal acceleration (i.e., 100 g) leading to sedimentation of large and thick nanosheets.

The nanosheet length is defined as the longest dimension of a sheet, while the width is the dimension perpendicular to the length. The height of a sheet is extracted from the average height of both profiles (see [19,32] for more details). For accurate determination of the number of layers from the measured apparent thickness, additional contributions from impurities, solvent residues, chemical potential, and measurement parameters must be considered. Therefore, the measured apparent height in liquid-exfoliated nanosheets is usually overestimated, but is proportional to the expected crystallographic thickness [21, 33, 34, 35]. This problem can be solved by a step height analysis on incompletely exfoliated nanosheets to convert the apparent thickness to the number of layers [21, 33, 36]. In addition, the new research showed that thinner nanosheets tend to be smaller and thicker nanosheets tend to be larger. This is typical of LPE nanosheets and is due to the fact that both delamination and tearing of the nanosheets occur [38].

To confirm the crystallinity and structural integrity of the nanosheets, a combination of TEM, Raman spectroscopy, XPS and XRD measurements were performed on the exfoliated nanomaterial. The results of the microscopic and spectroscopic characterization are summarized in Figure 2. For the TEM investigation, a non-size-selected dispersion, with only unpeeled material at 100 g removed as sediment, was dropped onto Quantifoil TEM grids in N2 atmosphere. This sample was selected to avoid exposure to the environment as much as possible, since centrifugation (in sealed vials) had to be performed outside the glovebox. CHP residues were removed in the microscope's pre-vacuum (~2 × 10⁵ mbar) at 150 °C for 40 min before the high-resolution low-dose TEM measurements (see Figures 3(A)-(D) for more details). Figure 2(A) shows a TEM overview image containing a relatively large nanosheet several μm long. The nanosheets shown in the image have sharp edges, which agrees well with the morphology from our previous material characterization measurements. Electron diffraction (SAED) was measured at the region indicated by the dashed circle and is shown as inset in Figure 2(A). The diffraction pattern shows sharp and bright spots confirming the crystallinity of the exfoliated nanomaterial. Figures 2(B) and (C) show HRTEM images of exfoliated α-RuCl₃ nanosheets revealing the atomic lattice. The HRTEM results are in good agreement with a simulated trigonal lattice structure (see SI, Figures 3(E)-(H) for direct comparison of the measurement with the expected patterns for various simulated RuCl₃ polymorphs) and with previous literature reports [56]. The inset in Figure 2(B) shows a fast Fourier-transformed reciprocal spatial pattern of the studied region, further emphasizing the crystalline character of the sample. In addition to SAED, energy dispersive X-ray (EDX) mapping was performed on the freestanding nanosheets using scanning transmission electron microscopy (STEM). The magnified high angular dark field (HAADF) STEM image and elemental distribution maps for ruthenium (red), chlorine (green), and oxygen (white) are shown in Figure 2(D). The measurements confirm the stoichiometry of the material and suggest only minor oxidation. This finding is further supported by core loss electron energy loss spectroscopy (see SI, Figures 4(A)-(C)). The average EDX spectrum of the sample shown in Figure 2(D) and EDX measurements on additional nanosheets are shown in the SI (Figures 6(D)-(F)). Based on the EDX measurements, the material stoichiometry is proposed as RuCl O2.60.02, RuCl O2.70.3, and RuCl O3.10.01 for the different regions studied. Small and thin nanosheets exhibit high sensitivity to electron beams even at 80 kV, as indicated by significant changes in the nanosheet lattice upon irradiation, see Figures 4(G)-(H)).

Researchers had previously shown [39], for exfoliated α-RuCl₃ that the two commonly observed phases trigonal (point group P31 12, [3]) and monoclinic (point group C2/m, [37]) can be distinguished from each other using high-resolution STEM images along the c* direction of the crystals, based on the different stacking order of these two out-of-plane phases.[39] Therefore, HRTEM images of few-layers (6) of the trigonal (point group P31 12, Figure 5F), monoclinic (point group C1 2/m, Figure 5G) and, for completeness, the α-RuCl₃ (point group P63 /mcm, Figure 5H, [16]) are simulated along the c* direction. The simulation parameters were adjusted to the given aberration conditions of the microscope (C₃ ≈-7.5 mm, C₅ ≈2.7 mm) before the experiment, resulting in bright atomic contrast conditions for the chosen defocus value (Δf=-C₁≈-4 nm).[17] In the new study, comparison of the simulated and experimental HRTEM images shows that the best agreement with the trigonal structure (P31 12, Figure 5F) can be obtained for these few-layer samples due to the different contrast of the atomic columns. However, atomic vacancies and stacking defects can be seen in the experimental image, which are indicated by a decrease in intensity in the atomic columns and lattice distortions. These defects are mainly caused by the damage from the electron beam irradiation during the acquisition of the image. Furthermore, it was already reported [39] that differences were observed between bulk and "quasi-2D" exfoliation layers.

EELS and EDX analyses were performed to analyze the nanosheet composition (Figure 4A-F). The core loss edges of the Cl(L) and Cl(K) lines as well as the Ru(M) series are resolved (Figure 4A-B), while contributions from oxygen are barely detectable (Figure 4C). This finding is further supported by EDX measurements on different nanosheets. For this purpose, EDX elemental maps and spectra were recorded on different nanosheet samples. Figure 4D shows the average EDX spectrum of the nanosheet shown in Figure 2D. Additional samples can be seen in Figure 4E-F. Although slight point-to-point deviations from the exact stoichiometry are observed, the overall results confirm the stoichiometry of RuCl₃ and indicate low oxidation of the material. However, we note that HRTEM measurements on ultrathin nanosheets were not possible because significant radiation damage (80 kV) was observed during image acquisition. An example is shown in Figure 4G, where an overview image of a nanosheet is shown after high-resolution scans were acquired in the highlighted area. Altered areas on the surface of the nanosheet can be clearly seen (see Figure 4 G). For clarity, the images at the beginning and at the end of the image acquisition are shown in comparison (Figure 4 H-I). The comparison shows clear changes in nanosheet morphology in the real and reciprocal space patterns (from FFT, see Figure 4H-I). Such structural changes make imaging of ultrathin structures (i.e., below about 6 layers, which was estimated using the log-ratio method) [18] impossible at atomic resolution.

In addition, to evaluate the structural integrity of the nanomaterials after exfoliation and size selection, especially for small/thin nanosheets that cannot be reliably studied in the TEM, measurements were performed on ensembles of nanosheets. For this purpose, samples of different nanosheet sizes were drop-cast in an inert atmosphere onto a gold-coated aluminum support for Raman spectroscopy and XPS measurements (Figures 2(E) and (F)) and onto a quartz substrate for powder XRD measurements (Figure 2(G)). Raman spectra of three different sizes representative of large, medium, and small nanosheets (i.e., 0.1-0.4 kg, 6-12 kg, and 22-50 kg) are shown in Figure 2(E) along with an average spectrum measured on a single crystal. The spectra are shown horizontally offset for clarity. While all modes of the bulk material can also be observed in the exfoliated nanosheets, a significant broadening of the modes is observed for the exfoliated material, which can be attributed to finite size effects and the random orientation of the droplet material. We note that a similar effect has also been observed for micromechanically cleaved nanosheets [20]. Since no additional modes appear in the spectra of the nanosheets, no clear evidence of material decomposition after LPE and LCC can be detected by Raman measurements. However, the species formed during decomposition may not be Raman active or resonant with the excitation wavelength.

For this reason, additional XPS measurements were performed. The spectrum of the ruthenium 3d core level is shown in Figure 2(F) for the sample with the smallest fraction of nanosheets (22-50 kg). In addition to the expected signals from RuCl₃, additional contributions from carbon, elemental ruthenium, ruthenium oxide, hydroxide, and oxyhydroxide (RuO₂, Ru(OH)₃, and RuO(OH)₂) are identified. The spectra indicate that only about 50% of the original RuCl₃ remains on the surface of the sample. Fitting the data to the response of each component reveals that ˜10% of the signal is from elemental ruthenium, ~21% is from RuO₂, and 16% is from Ru(OH)₃.

XRD powder measurements were performed for further characterization (Figure 2(G)). Diffractograms of the samples with large, medium, and small nanosheets show patterns that differ from those of the α-RuCl₃ mass, which is shown in gray for comparison. To understand these differences, the reflections are additionally compared to the recently reported patterns of randomly restacked RuCl₃ nanosheets (dark gray trace) [9]. The most intense Bragg reflections of the exfoliated nanomaterial are found at angles similar to those of the restacked material. In addition, a signal RuCl₃ broadening is observed that can be attributed to a structural distortion of the restacked material after drop casting, similar to the effects observed in the Raman spectra. A more detailed analysis shows that as the material size decreases, an increasing number of unassignable reflections are observed, which could indicate that crystalline degradation products are formed and are more evident in the samples with higher surface-to-volume ratios. However, an alternative interpretation of the data could be that these are due to effects of preferential orientation, which are less pronounced in the smaller plates, so that out-of-plane reflections become more dominant compared to larger plates, where mainly reflections with (001) character are observed due to the reflection geometry. This would imply that the other reflections at higher angles (2Θ) are attributable to other polymorphs or decomposition products of RuCl₃. However, simulations for different crystal lattices and stacking orientations would be required to confirm this, especially because the reflections cannot be assigned to the expected species from XPS measurements [9, 41]. Overall, the bulk characterization clearly shows that exfoliated α-RuCl₃ is susceptible to degradation, which will be discussed in more detail below. Since the material is exposed to ambient conditions in most of the measurements, no clear conclusions can be drawn about the degree of oxidation. However, the results of the TEM measurements, where exposure to oxygen and water was minimized, are encouraging and show very little oxidation of the nanomaterial in its original form. Unfortunately, the smallest/thinnest nanosheets are very sensitive to radiation damage, both in TEM and XPS, so their composition cannot be accurately determined. However, as we will show below, the absorption spectra show the same spectral profile for all size-selected fractions, indicating a similar initial composition, even for the fraction of smallest/thinnest nanosheets. In the following, this will be analyzed in more detail and also used to track the degradation as a function of time and temperature after the nanosheets are exposed to the ambient atmosphere, leading to significant changes in the spectral profile. Overall, the results emphasize that special care should be taken when preparing liquid-phase nanomaterials, as the reactivity of the nanosheets is significantly increased compared to their solid counterparts.

Measuring the magnetic properties
As mentioned above, α-RuCl₃ is considered to be of particular interest for fundamental studies of magnetic properties in low-dimensional structures and a possible realization of a two-dimensional QSL in its ground state [9, 42, 43]. To date, most of the work has been performed on micromechanically exfoliated (ME) nanosheets [20, 39, 44], with few examples of exfoliation in liquids [9] and no report of pristine, undoped nanosheets fabricated by sonication-assisted LPE. To this end, measurements were performed in a superconducting quantum interference device on three different sizes, representative of large, medium, and small (i.e., 0.1-0.4 kg, 6-12 kg, and 22-50 kg) α-RuCl₃ nanosheets, in liquid environments and on drop-cast nanosheets deposited in an inert atmosphere. The results of the magnetic measurements are shown in Figure 5. The magnetic response at low temperatures of the different sizes of CHP-dispersed nanosheets is shown for two separate batches of exfoliated material. A measurement on the starting material is also included for clarity. Due to background uncertainties resulting from non-negligible contributions from the solvent, the data have been scaled to the susceptibility of the crystal at 50 K for better comparability. The bulk material shows a uniform Curie-Weiss-like increase in magnetization upon cooling from room temperature, followed by a broad anomaly indicative of the development of long-range antiferromagnetic order at about 10 K. The bulk material is found to have a large magnetic field. This anomaly is suppressed in the large (0.1-0.4 kg) nanosheet dispersion, and instead a sharp transition is visible at T = 2 K (Figure 5(A), inset). No such transition is observed for the smaller nanosheet sample (22-50 kg) within the accessible temperature range. However, a similar transition can be observed for one of the two batches of intermediate size nanosheets (6-12 kg) shown in Figure 5(A). We note that the magnetic response at low temperatures for the dispersion of the small and medium size nanosheets is different between the respective batches, which we attribute to a combination of material decomposition, solvent effects, and partial aggregation of the dispersed nanosheets upon approaching low temperatures.

The high-temperature magnetism for the different nanosheet pellets is proportional to the bulk material, as shown by the susceptibility curves in Figure 5(B). The susceptibilities of the platelets deviate from those of the bulk material below 75 K, indicating stronger short-range antiferromagnetic correlations in the bulk material than in the platelets. Both the medium and large nanosheet films show similar behavior, with a broad plateau-like hump at about 4 K and a sharp drop at the lowest accessible temperatures (Figure 5(B), inset 1). The broad hump transitions to a well-resolved peak when a low magnetic field is applied (B = 0.01 T, see Figure 5(B), Inset 2). This implies the development of long-range antiferromagnetic order at T_N = 3 K for the medium and large nanosheets, while the magnetic order is either completely suppressed or shifted to temperatures below the accessible range when B = 1 T is applied. Moreover, the size reduction stabilizes the antiferromagnetic phase, as evidenced by a sharp peak in susceptibility at B = 1 T at 2.2 K for the sample with the smallest nanosheets (Figure 5(B), inset 1). Note that due to the limited amount of material, measurements at B = 0.01 T for the smallest nanosheets were not successful. Despite the fact that the long-range magnetic ordering appears to be more pronounced in the small platelets compared to the fractions with larger nanosheets, the broad hump is not observed, indicating less pronounced short-range antiferromagnetic correlations in this temperature range. The general trend, then, is that the antiferromagnetic correlations at high temperatures are suppressed by the reduction in size, while at the same time a sharp, more 3D antiferromagnetic phase transition appears to be stabilized. It could be argued that the low-temperature antiferromagnetic phase that appears in the solid nanosheet samples is similar in nature to the phase indicated by the sharp decrease in susceptibility of the samples dispersed in solution.

The research demonstrated additive-free liquid exfoliation of α-RuCl₃ into nanosheets of different sizes and thicknesses in an inert atmosphere. The nanomaterial was separated into fractions with different size and thickness distributions following established cascade centrifugation protocols.

"We confirmed the structural integrity and crystallinity of the individual nanosheets by TEM measurements and further substantiated by a combination of Raman spectroscopy, XPS and XRD measurements averaged over a large number of nanosheets." said the authors. "For the first time, susceptibility measurements were performed on α-RuCl₃ nanosheets in a liquid medium and compared with randomly restacked nanosheets in drop-cast samples."

No QSL state could be confirmed for any of the nanosheet sizes. In contrast, high-temperature antiferromagnetic correlations are suppressed upon downsizing, while a sharp, 3D-like antiferromagnetic phase transition appears to be stabilized. Moreover, the photospectroscopic response of α-RuCl₃ nanosheets for different size fractions allowed determine spectroscopic metrics for the average number of material layers and concentration in dispersion. The researchers showed that an unusually small change in the extinction and absorption intensity ratio is observed with the lateral size of the material. Although this is qualitatively similar to previously reported material systems [25, 66, 68], it is not considered a reliable metric due to the magnitude of the observed changes. They believe the results presented are important for future applications that exploit the magnetic properties of nanosheets. The low activation energy for material decomposition emphasizes that exfoliation, storage, and processing parameters can lead to significant differences in material properties due to partial oxidation of the nanomaterial and can now be further studied.

Resource: Synnatschke, K., Jonak, M., Storm, A., Laha, S., Köster, J., Petry, J., Ott, S., Szydlowska, Duesberg, G. S., Kaiser, U., Klingeler, R., Lotsch, B., & Backes, C. (2023). Sonication-assisted liquid exfoliation and size-dependent properties of magnetic two-dimensional α-RuCl₃. Journal of Physics D: Applied Physics, 56, 274001. https://doi.org/10.1088/1361-6463/accc3e