betaFeedback

WorksPhysics and Astronomy

Charge density wave induced nodal lines in LaTe3_3

Abstract

LaTe3_3 is a noncentrosymmetric (NC) material with time reversal (TR) symmetry in which the charge density wave (CDW) is hosted by the Te bilayers. Here, we show that LaTe3_3 hosts a Kramers nodal line (KNL), a twofold degenerate nodal line that connects the TR invariant momenta in NC achiral systems, using angle resolved photoemission spectroscopy (ARPES), density functional theory (DFT), effective band structure (EBS) calculated by band unfolding, and symmetry arguments. DFT incorporating spin-orbit coupling (SOC) reveals that the KNL – protected by the TR and lattice symmetries – imposes gapless crossings between the bilayer-split CDW-induced shadow bands and the main bands. In excellent agreement with the EBS, ARPES data corroborate the presence of the KNL and show that the crossings traverse the Fermi level. Furthermore, spinless nodal lines - entirely gapped out by the SOC - are formed by the linear crossings of the shadow and main bands with a high Fermi velocity.

I Introduction

Recent years have witnessed a rapid development in the understanding of the physics of cooperative charge density wave (CDW) electronic state [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]. In particular, interplay of the CDW electronic state with the non-trivial topological phases provides an interesting platform for discovery of novel quasiparticles such as, axionic insulator [3, 13], quantum spin-hall insulator [14], fractional Chern insulator states [15] and manipulation of topologically protected states [16, 17]. CDW can drive topological phase transitions by modifying the symmetry of the lattice, such as breaking the inversion symmetry [18]. Interesting topological phases are frequently found in noncentrosymmetric materials, such as, nodal chain fermions [19], Dirac and Weyl fermions [20, 21, 22], hourglass fermions protected by glide reflection [23], Kramers Weyl semimetal (KWS) [24] and recently predicted Kramers nodal line (KNL) metal [25]. KNLs differ from the Weyl nodal lines because they join two time reversal invariant momenta (TRIM) points and should appear in all achiral noncentrosymmetric time reversal symmetry (TRS) preserving systems [25]. For the subclass of nonsymmorphic symmetry, KNLs emerge from the Γ\Gamma points. Unlike the previously known nodal lines manifested by band inversion[26], the KNLs are robust under spin-orbit coupling (SOC) unless the protecting lattice symmetries such as TRS, mirror, or roto-inversion symmetries are broken. KNL fermions have been predicted to exhibit novel physical properties such as quantized optical conductivity [25]. However, in this emerging field, to the best of our knowledge, the experimental evidence of KNL is limited to the work by Shang et al. [27] who reported that transition metal ruthenium silicides belong to this class and exhibit unconventional superconductivity based on muon spin spectroscopy and density functional theory (DFT).

In recent years, multiple fascinating findings in LaTe3\mathrm{LaTe}_{3} [1, 4, 7, 28, 29, 30] – a member of the RTe3\mathrm{RTe}_{3} (R represents a rare earth element) family with highest CDW transition temperature of 670 K [31, 32] – have rekindled the scientific interest of the community in this noncentrosymmetric TRS preserving material. The detection of an axial Higgs boson mode in LaTe3\mathrm{LaTe}_{3} from Raman spectroscopy has been related to unconventional CDW excitation [1]. Photoinduced CDW state with topological defects in this material has been discovered from ultrafast electron diffraction and related studies [7, 28]. Transport studies have revealed that LaTe3\mathrm{LaTe}_{3} possesses an unusually high non-saturated longitudinal magnetoresistance [30], which is similar to that of nodal line material [33]. In addition, LaTe3\mathrm{LaTe}_{3} possesses very high carrier mobility [30], which, in conjunction with its high transition temperature, makes it a promising contender for next-generation electronics. Theoretical calculation of the electronic susceptibility as well as experiments have shown that qq dependent electron phonon coupling plays an important role in stabilizing the CDW state in RTe3\mathrm{RTe}_{3} [34, 35]. While there are no DFT investigations, angle resolved photoemission spectroscopy (ARPES) measurements on LaTe3\mathrm{LaTe}_{3} are also scarce in literature [7, 36]. Using time resolved ARPES, Zong et al. [7] examined light-induced melting of the CDW state of LaTe3\mathrm{LaTe}_{3}. A previous research by Brouet et al. [36] demonstrated that the CDW-induced shadow bands hybridize with the main bands existing in the non-CDW state creating a CDW gap along a particular direction of the Brillouin zone (BZ).

In light of the fact that LaTe3\mathrm{LaTe}_{3} is a noncentrosymmetric achiral material with TRS intact, topological phases brought on by inversion symmetry breaking in conjunction with other symmetries may be anticipated in the CDW state. Here, from an in-depth study of the band structure of LaTe3\mathrm{LaTe}_{3} in the CDW state by combining ARPES and ab−ab-initio DFT using a realistic experiment-based structure, we establish the existence of a KNL in a CDW material for the first time. It originates from the interaction of the shadow band and the main band, and is hosted by the TRS and the lattice symmetries. Furthermore, spinless nodal lines that are entirely gapped out by SOC are also identified.

Results

Modulated structure of LaTe3 in the CDW state

Figure 1a shows a 7-fold modulated (1×1×7)(1 \times1 \times7) supercell structure of LaTe3LaTe_3. It is derived from the experimental structure obtained from single crystal x-ray diffraction (XRD) at 100 K [37] (see Methods). Its space group is C2cmC2cm (SG #40), which is same as that reported for the average structure [37]. LaTe3LaTe_3 is made up of two main structural units: La-Te1 corrugated slabs and Te2-Te3 bilayer. The latter, highlighted by blue double-sided arrows, are weakly coupled by van der Waals interaction. The Te bilayer of the 7-fold modulated structure hosts the CDW with qCDW=27c∗=0.286c∗q_{\mathrm{CDW}} = \frac{2}{7}c^{*} = 0.286c^{*}, where c∗c^{*} is the reciprocal lattice vector along kzk_z in the non-CDW state. This value is close to the experimental values determined from ARPES (subsection II B) and reported in previous literature [37, 38]. Moreover, the amplitude of the CDW modulation is nearly similar to that from XRD [37], see Supplementary Figs. S1a,b.

Structure, Brillouin zone and the Fermi surface of LaTe3 in the CDW state

Figure 1. Structure, Brillouin zone and the Fermi surface of LaTe3\mathrm{LaTe}_3 in the CDW state. (a) The 7-fold (1×1×7)(1\times1\times7) modulated primitive unit cell of LaTe3\mathrm{LaTe}_3 comprising of 56 atoms with qCDW=27c∗q_{\mathrm{CDW}}=\frac{2}{7}c^* viewed perpendicular (left) and parallel (right) to the c′\mathbf{c}' direction. The lattice constants are a′=b′=13.256a'=b'=13.256 Å, c′=30.778c'=30.778 Å with α′=β′=90∘\alpha'=\beta'=90^\circ, and γ′=160.99∘\gamma'=160.99^\circ. The cleavage plane (light blue) occurs in between the two weakly interacting Te layers. (b) High resolution scanning tunneling microscopy topography image obtained with bias voltage of 0.2 V and a tunneling current of 0.4 nA. (c) Low energy electron diffraction pattern measured with 57 eV primary beam energy in inverted gray scale. (d) The CDW Brillouin zone (BZ) (black) is shown within that of the non-CDW state (light blue). The high symmetry points [39] are indicated in these respective colors. Γ\Gamma, YY and XX points coincide for both. Γ2X2\Gamma_2X_2 (dashed line) is in the 2nd BZ along kzk_z. (e) The Fermi surface (FS) in the CDW state measured by ARPES, where the length of the white dashed arrows (#1, #2) that join the shadow branches with the main branches of the FS represents qCDWq_{\mathrm{CDW}}.

The CDW modulation has been directly observed in the LaTe3LaTe_3 crystal studied here from a high resolution scanning tunneling microscopy (STM) topography image. In Figure 1b, the modulated Te net formed by connecting the Te atoms (orange circles) is shown by white lines. Low energy electron diffraction (LEED) pattern also shows CDW related satellite spots (encircled) besides the (1×1)(1 \times1) spots (Figure 1c). Satellite spots are also observed in the Fourier transform of the STM image, see Supplementary Note 1 and Fig. S2.

The experiment based 7 fold modulated structure discussed above has been used for the DFT calculations. The corresponding BZ is shown in Figure 1d inscribed within the non-CDW BZ. Since the BZ is related to the ordering of the lattice constants of the conventional cell, its comparison with the primitive unit cell of the 7-fold structure is shown in Supplementary Figs. S3a,b. In our notation, the horizontal plane in the BZ is represented by kxk_x-kzk_z and qCDWq_{\mathrm{CDW}} is oriented along kzk_z. The CDW BZ, containing features such as the primitive reciprocal lattice vectors and all the pertinent high symmetry points and directions as well as their coordinates, is depicted in Supplementary Fig. S3c.

Crossing of the bilayer-split shadow and main bands from ARPES

An E(kz)E(k_z) ARPES intensity plot in a generic direction parallel to ΓZ\Gamma Z at kx=0.68k_x = 0.68 Å−1^{-1} i.e., near the X point (the length of ΓX\Gamma X being 0.7370.737 Å−1^{-1}) is shown in Figure 2a (see Methods for the

Crossings between the bilayer-split shadow and the main bands

Figure 2. Crossings between the bilayer-split shadow and the main bands. E(kz)E(k_z) ARPES intensity plots measured at 100 K for kx=k_x= (a) 0.68, (b) 0.63, (c) 0.59, and (d) 0.56 Å−1^{-1}. The crossing regions are highlighted by green dashed ovals. (e) A zoomed view of the near EFE_F region of panel a around kz=0.21k_z=0.21 Å−1^{-1}, the white dashed lines are obtained by curve fitting (see Methods). (f) Curvature plot of a part of panel c shown in an expanded scale: the crossings indicated by LL, BB, RR and TT are shown by blue, green, yellow and red circles, respectively. The TT and BB crossings occur at the X2X_2 point. The dashed red (black) line representing the main (shadow) band is obtained by curve fitting as in panel e. (g) Cut AA at kz=0.216k_z=0.216 of an EE-kzk_z-kxk_x ARPES intensity plot shows the dispersion of the RR crossing and the mumu and sdsd bands within the yellow rectangle. A shadow band is observed in the cut BB. (h-k) The effective band structure (EBS) along kzk_z obtained by band unfolding at similar kxk_x values as in panels a-d, respectively. The band crossing regions are highlighted by green dashed ovals.

experimental details). The direction of the measurement is shown by a red line denoted by “a” in Figure 2g. It shows two main bands (inner and outer) centered around Γ\Gamma that cross EFE_{\mathrm{F}} at kz=±0.15k_z = \pm0.15 and ±0.21\pm0.21 Å−1^{-1}, respectively. The outer main band disperses down to binding energy (EE) of about 1.25 eV, while the inner band has a nearly flat bottom at ∼\sim0.8 eV. It is interesting to note that a relatively weaker band centered around kz=±0.41k_z = \pm0.41 Å−1^{-1} is a replica of the main band shifted by qCDW(=0.28c∗=0.41q_{\mathrm{CDW}} (= 0.28c^{*} = 0.41 Å−1^{-1}), as shown by two horizontal white dashed arrows. This replica band crosses EFE_{\mathrm{F}} at kz=0.21k_z = 0.21 and 0.610.61 Å−1^{-1} and disperses down to E∼1.25E \sim1.25 eV. No shift in its position along EE compared to the main band is observed, which indicates that it is related to the initial state CDW superlattice [40]. It has been referred to in literature as the shadow band [36, 40, 41]. The signature of the shadow band is also evident in the Fermi surface (FS) shown in Figure 1e, where shadow FS branches appear at a separation of qCDWq_{\mathrm{CDW}} from the main branches, as shown by white dashed arrows in the metallic region around the XX point parallel to the ΓZ\Gamma Z direction. In the Supplementary Note 2, a discussion about the FS and qCDWq_{\mathrm{CDW}} obtained from the average separation of the shadow and main branches are provided. qCDW=0.28±0.005c∗q_{\mathrm{CDW}} = 0.28 \pm0.005c^{*} determined in this way is close to 27c∗\frac{2}{7}c^{*} and is in excellent agreement with the values obtained from STM and LEED, see Supplementary Note 1.

In Figure 2a, The shadow and the main bands resemble an inverted “V” and meet each other close to EFE_{\mathrm{F}} at kz=±0.21k_z = \pm0.21 Å−1^{-1} (highlighted by a green dashed oval). This region is shown in an expanded scale in Figure 2e, where the white dashed lines suggest a possible crossing of nearly linear bands at EFE_{\mathrm{F}}. The Fermi velocities of these bands, determined using the expression (1ℏdEdkz)\left(\frac{1}{\hbar}\frac{dE}{dk_z}\right), turn out to be 1.2±0.05×1061.2 \pm0.05 \times10^{6} and 1±0.05×1061 \pm0.05 \times10^{6} m/s for the shadow and main bands, respectively. These values are comparable to graphene (1×1061 \times10^{6} m/s [42]), indicating large mobility of LaTe3_3 in agreement with recent report from Hall conductivity measurements [30].

To further investigate these bands and their possible crossing, ARPES was performed over a range of kxk_x out of which three representative plots (along the red lines parallel to “a” up to “d” in Figure 2g) are shown in Figure 2b-Figure 2d. Interestingly, as kxk_x decreases, both the main and the shadow bands spread out in kzk_z and the crossings – highlighted by green dashed ovals – shift below EFE_{\mathrm{F}} to larger EE. However, note that their kzk_z position remains unchanged around X2≈±12qCDWX_2 \approx\pm\frac{1}{2}q_{\mathrm{CDW}}. In Figure 2b-Figure 2d, splittings in both the shadow and the mains bands are observed, which are related to the coupling between the two adjacent weakly coupled Te layers. This

is the bilayer splitting that has been reported in other RTe3_3 members [36, 43, 44], bilayer graphene [45], and cuprate superconductors [46]. Here, we find that the splitting increases with EE, e.g., it varies from 0.016 to 0.03 Å−1^{-1} as EE increases from 0.1 to 0.5 eV (Figure 2c). Evidence of varying bilayer splitting is also observed in the FS (see Supplementary Note 2).

Both the bilayer-split shadow bands - denoted by susu (sdsd) for smaller (larger) EE, see Figure 2f - are shifted by qCDWq_{\mathrm{CDW}} from the corresponding main bands (denoted by mumu and mdmd). This is shown by the white dashed horizontal arrows in Figure 2a,Figure 2b. This leads to formation of four crossings between the main and the shadow bands, as shown in Figure 2f. These are also visible in a constant energy kzk_z-kxk_x isosurface through the band crossing e.g. at E=0.5E=0.5 eV and a stack of momentum distribution curves (MDCs), see Supplementary Figs. S8a,b.

The crossings are denoted by left (LL, blue circle, crossing of mumu and sdsd i.e., mu⊗sdmu \otimes sd), right (RR, orange circle, md⊗sumd \otimes su), top (TT, red circle, mu⊗sumu\otimes su), and bottom (BB, green circle, md⊗sdmd \otimes sd) (Figure 2f). While Figure 2a-Figure 2d show the band dispersion at discrete kxk_x values, we show a continuous dispersion of the crossings, for example the RR in cut AA of Figure 2g. Here, the E(kx)E(k_x) at kz=0.216k_z=0.216 Å−1^{-1} shows the loci of the RR crossing within the yellow rectangle.

ARPES with different photon energies shows negligible kyk_y dependence of the crossings. For example at kx=0.58k_x=0.58 Å−1^{-1}, Supplementary Figs. S9a-d show that their position (highlighted by green dashed ovals) remains almost unchanged with kyk_y. This is summarized in Fig. S9e through a kyk_y-kzk_z map at E=0.39E=0.39 eV, where the two crossings at kz=±0.2k_z=\pm0.2 Å−1^{-1} (marked by yellow arrows) show almost no change with kyk_y, indicating the quasi-2D nature of LaTe3_3.

Effective band structure from DFT compared to ARPES

Multiple crossings indicated by ARPES is a surprising result since hybridization gap is generally expected at the band touching points between the Bloch states connected by qCDWq_{\mathrm{CDW}} [44, 47]. To understand this, we have performed DFT calculations using the modulated structure discussed in subsection II A.

The E(kz)E(k_z) bands calculated at different kxk_x as in the experiment are shown in an extended zone scheme spread over the non-CDW BZ in Supplementary fig:S10. Many bands are observed in the CDW state due to band folding. The complexity of these bands hides the influence of the CDW on the electronic bands and impedes their interpretation and direct comparison with ARPES. So, an effective band structure (EBS) has been calculated by unfolding the bands in the non-CDW BZ [48, 49, 50]. In Figure 2h-Figure 2k, EBS shows the distribution of states as a function of their energy and momenta with appropriate spectral weight, where broadening similar to the experiment has been applied. Variations in the spectral weight is evident resulting in dissimilar EBS in the different CDW BZs. This complexity in a CDW system is discussed further in subsection II F.

The importance of the EBS calculation is that despite the CDW effect being small with the modulation amplitude only ∼\sim4.5% of the average Te2-Te3 distances, the EBS reveals the shadow bands that are shifted from the main bands by qCDWq_{\mathrm{CDW}} (blue dashed arrows) in excellent agreement with ARPES (compare Figure 2h-Figure 2k with Figure 2a-Figure 2d). In contrast, the bands in the non-CDW state, where obviously the CDW amplitude is zero, shows only the main bands (Supplementary Fig. S11a). An analysis of the orbital character establishes that the shadow bands (as well as the main bands) are of predominantly in-plane Te pxp_x-pzp_z character, see the Supplementary Figs. S12a-f. The out-of-plane pyp_y character becomes slightly significant only for E>0.6E > 0.6 eV, and the contributions from the La-Te1 block is negligible. However, transfer of electrons from La to the Te net determines the band filling in the Te2-Te3 layer and this has been calculated using the Bader charge analysis (see Supplementary Note 3 including Refs. 51–53).

From Figure 2h, we find that the outer branch of the shadow band disperses towards EFE_F and crosses the main band around kz=±0.21k_z = \pm0.21 Å−1^{-1} (highlighted by a dashed green oval). A bunch of bands that disperse weakly between 1.15 to 1.6 eV are observed in both theory and experiment at similar EE. A bilayer splitting of 0.17-0.2 eV (red double arrow in Figure 2h) is observed in the EBS with LL and RR crossings close to EFE_F, while TT (BB) appears above (below) EFE_F. However, these are not visible separately in Figure 2a, indicating that possibly at this kxk_x the experimental bilayer splitting is smaller compared to the DFT value. The inner branch of the shadow band that disperses to about 0.6 eV is more prominent in the EBS, but also has its counterpart in the curvature plot of ARPES (see Supplementary Fig. S11b). In Figure 2i-Figure 2k, for progressively smaller kxk_x values, EBS portray all the four crossings at similar EE-kzk_z observed in ARPES (compare Figure 2i-Figure 2k with Figure 2b-Figure 2d). The agreement is also good at larger EE: the main bands around 0.8 eV become flatter and move to lower EE.

Important to note is that from ARPES the crossings appear to be gapless (Figure 2a-Figure 2f) within the experimental and lifetime broadening. Although this is supported by the EBS (Figure 2h-Figure 2k), it should be noted that the calculations have been performed with a kk step size (δkz\delta k_z) of 6×10−3 A˚−16 \times10^{-3}\,\mathrm{\AA}^{-1}, and with broadening comparable to ARPES, which might conceal the potential presence of minigaps. In the subsequent subsections II D and II E, we show the DFT bands calculated with smaller δkz\delta k_z to probe these crossings further.

Spinless nodal lines formed by the LL and RR crossings

The E(kz)E(k_z) DFT bands in the crossing region with small δkz\delta k_z (=5×10−4 A˚−1= 5 \times10^{-4}\,\mathrm{\AA}^{-1}) in absence of SOC indeed indicate that the LL and RR crossings are gapless (Figure 3a). This is reconfirmed in Figure 3b,Figure 3c by the bands calculated with even smaller δkz\delta k_z (=1×10−5 A˚−1= 1 \times10^{-5}\,\mathrm{\AA}^{-1}). It is interesting to note that these crossings occur at generic points of the BZ in the kxk_x-kzk_z plane and the bands involved are linear (Figure 3a-Figure 3c). ARPES intensity plots in Figure 2c,Figure 2f at this kxk_x also demonstrate the linearity of the bands around LL and RR in excellent agreement with DFT. The velocities calculated from their slopes are similar to that obtained at EFE_F from Figure 2e. Although these bands originate from the in-plane pp orbitals with small difference in contributions from the pxp_x and pzp_z orbitals as shown in Supplementary Figs. S12a-f, mdmd and sdsd bands belong to M1M_1 irreducible band representation (irrep), while mumu and susu belong to M2M_2 irrep. Thus both LL and RR crossings are formed by bands belonging to different irreps (Figure 3a-Figure 3c).

Band crossings and the Kramers nodal line

Figure 3. Band crossings and the Kramers nodal line. (a) E(kz)E(k_z) band dispersion from DFT at kx=0.59k_x=0.59 Å−1^{-1} without (w/o) SOC. The irreducible representations (irreps) are shown. The vertical dashed line represents the kzk_z point on Γ2X2\Gamma_2X_2 i.e., the Σ\Sigma line. Zoomed colored rectangles of panel a show the bands around (b) LL (mu⊗sdmu \otimes sd), (c) RR (md⊗sumd \otimes su), (d) TT (mu⊗sumu \otimes su), and (e) BB (md⊗sdmd \otimes sd). Comparison of the E(kx)E(k_x) ARPES intensity plot with DFT for the following crossings: (f) RR at kz=0.229k_z=0.229 Å−1^{-1} and (g) LL at kz=0.196k_z=0.196 Å−1^{-1}. The positions of these crossings obtained from DFT (orange and blue filled circles for RR and LL, respectively) are superimposed. (h) A schematic representation of the gapless LL and RR crossings in the EE-kzk_z-kxk_x space (red dashed lines) and their projection on the kxk_x-kzk_z plane showing the spinless nodal lines (thick red lines on both sides of the Σ\Sigma line). (i-m) Same as panels a-e except that the calculations are performed with SOC. (n) E(kx)E(k_x) ARPES intensity plot at kz=0.204k_z=0.204 Å−1^{-1} (dashed curves here and in panels f and g serve as guide to the eye) compared with the positions of the crossings enforced by the KNL obtained from DFT for TT (red, light red circles) and BB (black, gray circles). (o) A schematic representation of the four crossings (green dashed lines) related to the upper and lower branches of TT and BB. The Kramers nodal line appears along the Σ\Sigma line and is denoted by a green thick line on the kxk_x-kzk_z plane. The gapped out spinless nodal lines at LL and RR are also shown. The energy dispersion along kxk_x is not shown here as well as in panel h.

The crossings disperse with kxk_x and at larger kx=0.685 A˚−1k_x = 0.685\,\mathrm{\AA}^{-1} compared to 0.59 A˚−10.59\,\mathrm{\AA}^{-1}, LL and RR crossings traverse the EFE_F (Supplementary Fig. S13a-c). In fact, calculations for a series of kxk_x values establish that the gapless linear crossings occur over an extended range of the EE-kk space (Supplementary Fig. S14). The loci of each crossing form a continuous curve in the momentum space that has been referred to as a nodal line. A direct comparison of the (EE, kxk_x) cuts from ARPES (Figure 3f,Figure 3g for RR and LL, respectively) show that the positions of both the nodal lines are in excellent agreement with DFT. Both disperse between (EE, kxk_x) = (00 eV, ∼0.7 A˚−1\sim0.7\,\mathrm{\AA}^{-1}) to (∼0.6\sim0.6 eV, ∼0.5 A˚−1\sim0.5\,\mathrm{\AA}^{-1}) with kzk_z at 0.1950.195 and 0.22 A˚−10.22\,\mathrm{\AA}^{-1} for LL and RR, respectively. Thus, the crossings appear within an energy window of E∼0.6E \sim0.6 eV to the EFE_F. Their projections in the kxk_x-kzk_z plane form a pair of approximately parallel nodal lines that are 0.2 A˚−10.2\,\mathrm{\AA}^{-1} in length and appear at a separation of kz∼0.02 A˚−1k_z \sim0.02\,\mathrm{\AA}^{-1} in a general direction on this plane (red solid lines in Figure 3h). These appear parallel to but on either sides of Γ2X2\Gamma_2X_2 that occurs at kz=0.204 A˚−1k_z = 0.204\,\mathrm{\AA}^{-1}. These nodal lines are formed by two fold crossings of

nondegenerate bands in the absence of spin and so are referred to spinless nodal lines.

With inclusion of SOC, it is intriguing to find that both the LL and RR crossings (and hence the nodal lines) are entirely gapped out by minigap of 14-17 meV into upper and lower branches, as shown in Figure 3i-Figure 3k and Supplementary Figs. S13g,h. A schematic is shown in Figure 3o. The gap is formed by hybridization of bands belonging to same one dimensional double group irrep involving both M‾3\overline{M}_3 and M‾4\overline{M}_4.

Note that noncentrosymmetry should lift the SU(2) spin degeneracy. This is visible through the spin splitting of both the upper and lower branches in Figure 3j,Figure 3k. These spin-split bands belong to different irreps M‾3\overline{M}_3 and M‾4\overline{M}_4 and do not cross each other. The splittings for both LL and RR are kk-dependent, it is maximum at the extrema (∼\sim3 meV for lower and 1 meV for the upper branch in Figure 3j,Figure 3k) and decreases away from it.

Evidence of Kramers nodal line from the TT and BB crossings

In the case of TT and BB, the shadow and main bands belong to the same irrep, leading to hybridization-related minigaps without SOC and the creation of upper and lower branches (Figure 3a,Figure 3d,Figure 3e). The inclusion of SOC provides a fascinating outcome: the spin-split bands in both the upper and lower branches of TT and BB exhibit gapless crossings (Figure 3l,Figure 3m), which contrasts the bands at LL and RR (Figure 3j,Figure 3k). Each branch of TT and BB crosses at the same kz=0.204k_z = 0.204 Å−1^{-1} (dashed lines in Figure 3i,Figure 3l,Figure 3m) resulting in four crossings. This value of kzk_z is special, since it falls on the Γ2X2\Gamma_2X_2 line of the CDW BZ (i.e., ΓX\Gamma X or the Σ\Sigma line) (Figure 1d). At kx=0.59k_x = 0.59 Å−1^{-1}, the four crossings appear at different EE: 0.256 (0.267) eV for the upper (lower) branch of TT, and 0.448 (0.469) eV for upper (lower) branches of BB.

These gapless crossings are observed over a range of kxk_x, as in the case of LL and RR (subsection III D). For example, at kx=0.685k_x = 0.685 Å−1^{-1}, the crossings disperse to smaller EE and in this case both the branches of TT (BB) are above (below) the EFE_F (Supplementary Fig. S13f-h). Significantly, the crossings always appear at same kzk_z i.e., along the Σ\Sigma line, suggesting that the crossings may be enforced by the lattice symmetries along this direction.

The band irreps with SOC shown in Figure 3l,Figure 3m are as follows: the crossing belongs to a double valued irrep that is two dimensional (Σ‾5\overline{\Sigma}_5), while away from it, the spin-split bands have one dimensional (M‾3\overline{M}_3 or M‾4\overline{M}_4) representation. Significantly, the Σ\Sigma line that emerges from the Γ\Gamma point has the little group that is isomorphic to C2vC_{2v} point group and has symmetries such as the two fold rotation about the kxk_x-axis denoted by {2100∣0,0,0}\{2_{100}|0,0,0\}, glide reflection perpendicular to the yy axis in the kxk_x-kzk_z plane followed by a translation of 12c\frac{1}{2}c given by {m010∣0,0,12}\{m_{010}|0,0,\frac{1}{2}\} and an off-centered mirror perpendicular to the kzk_z axis {m001∣0,0,12}\{m_{001}|0,0,\frac{1}{2}\}. Γ\Gamma is a TRIM point, where according to the Kramers theorem, each band is at least doubly degenerate. The little group of Σ\Sigma is related to that of Γ\Gamma by the compatibility relations. We find that Γ\Gamma and Σ\Sigma are represented by two dimensional double-valued irreps: Γ‾5\overline{\Gamma}_5 and Σ‾5\overline{\Sigma}_5, respectively. These representations are similar, as shown in Supplementary Table S2 [54]. The Σ\Sigma line passes through the XX point with co-ordinates (0.257,0.257,0)(0.257, 0.257,0) and meets the TRIM point Y2Y_2 (0.5,0.5,0)(0.5, 0.5, 0) in the next BZ (see Supplementary Fig. S15). Y2Y_2 also belongs to double-valued two dimensional irrep Y‾25\overline{Y}_{25} that is same as Γ\Gamma and Σ\Sigma (Supplementary Table S2). The condition that at the TRIM points the representations should be time reversal invariant is satisfied since both Γ‾5\overline{\Gamma}_5 and Y‾25\overline{Y}_{25} are pseudo-real [54]. Thus, Γ‾5\overline{\Gamma}_5-Σ‾5\overline{\Sigma}_5-Y‾25\overline{Y}_{25} is able to support two fold degeneracy of the bands along ΓXY2\Gamma XY_2 i.e., the Σ\Sigma line. This is a Kramers nodal line (KNL) that occurs along the Σ\Sigma line in the mirror-invariant kk-plane in presence of TRS, and the additional rotational symmetry constrains the KNL along a high symmetry direction [25]. Xie et al. [25] have proved the existence of KNL along the C2C_2 rotational axis ({2100∣0,0,0}\{2_{100}|0,0,0\}), which in our case is along kxk_x that lies in the kxk_x-kyk_y mirror plane ({m001∣0,0,12}\{m_{001}|0,0,\frac{1}{2}\}). In Figure 3o, the projection of the loci of the crossings of the four pairs of bands in the EE-kzk_z-kxk_x space on the kxk_x-kzk_z plane shows the KNL (thick green line) that enforces the crossings.

The bands in a plane that cuts the KNL perpendicularly – as is the case in Figure 3i – have been described in the literature as two-dimensional massless Dirac Hamiltonian with the Berry curvature concentrated at the crossing [25]. These authors demonstrated that the Berry phase around a KNL is quantized as mπm\pi mod 2π2\pi. In case of quadratic and cubic dispersion of these bands, the crossing has been dubbed as a higher-order Dirac point [25]. In the present case, both the upper and lower branches of TT and BB exhibit quadratic dispersion very close to the crossings, as is evident in Figure 3l,Figure 3m. Consequently, the gapless crossings in LaTe3\mathrm{LaTe}_3 that are associated with the KNL are higher-order Dirac points.

ARPES and DFT along the KNL and other directions

In Figure 4a, the E(kx)E(k_x) bands calculated along the KNL (i.e., ΓX\Gamma X) show two pairs of degenerate bands related to the TT and BB crossings, highlighted by green shading in the insets ii, iiii. The ARPES intensity plot in Figure 3n represents these bands that are also identified by the degenerate crossings of bands in the perpendicular E(kz)E(k_z) direction (Figure 3l,Figure 3m) enforced by the KNL. The existence of the KNL between 0.5 to 0.7 Å−1^{-1} along kxk_x is affirmed by ARPES through an excellent agreement of the positions of the crossings obtained from DFT (filled circles) that are overlaid on the experimental data in Figure 3n. The agreement is also evident in the Supplementary Fig. S16, where Figure 3i is superposed on Figure 2f. However, the energy separations between the upper and lower branches of both TT and BB are too small to be resolved by ARPES. Similarly, the quadratic nature visible in a tiny kzk_z range shown in Figure 3l,Figure 3m is not distinguished, see the regions enclosed by the red and black rectangles in Supplementary Fig. S16. Nonetheless, ARPES is consistent with DFT, and both show that the crossings disperse from E∼0.6E\sim0.6 eV at kx∼0.55k_x\sim0.55 Å−1^{-1} and traverse the EFE_F at kx=k_x= 0.65-0.7 Å−1^{-1}.

DFT and ARPES along and perpendicular to the KNL

Figure 4. DFT and ARPES along and perpendicular to the KNL. (a) The band structure of LaTe3\mathrm{LaTe}_{3} along XΓZ′Z\mathrm{X}\Gamma Z^{\prime}Z in the CDW state. On the left, zoomed regions from the blue rectangles show the two fold degeneracy (dashed-green and black) of the bands along the KNL. On the right, zoomed regions from the red rectangles show the spin-splitting along the ΓZ\Gamma Z direction in contrast to ΓX\Gamma X. (b) ARPES intensity plot measured using photon energy of 24.4 eV and (c) the EBS obtained from DFT towards ΓX\Gamma X and ΓZ\Gamma Z.

We find that every band along ΓX\Gamma X is degenerate due to the double degeneracy enforced by the KNL (see for example the insets ii-vv of Figure 4a) and belong to two dimensional Σ‾5\overline{\Sigma}_5 irrep. Besides the TT and BB related bands discussed above, there are other bands that cross EFE_F in Figure 4a. These bands – numbered as 11, 22 and 33 and highlighted by orange shading – have not been observed in the ARPES along Γ2X2\Gamma_2X_2 (kz=0.204k_z=0.204 Å−1^{-1}) due to their low spectral weight at this kzk_z. However, these three bands are clearly visible in the ARPES intensity plot along ΓX\Gamma X (kz=0k_z=0 Å−1^{-1}) in Figure 4b and disperse across the EFE_F in splendid agreement with the EBS along the same direction in Figure 4c. At larger EE other bands (numbered as 44-1212) in the EBS along ΓX\Gamma X are in very good agreement with ARPES (Figure 4b). On the other hand, the bands around the TT and BB crossings both along and perpendicular to the KNL have negligible spectral weight around ΓX\Gamma X and so are observed neither in EBS nor ARPES in Figure 4b,Figure 4c. Similarly, in Figure 3a,Figure 3i parabolic bands around E=0.3E=0.3 eV and E=0.5E=0.5-0.60.6 eV are not visible in the EBS or ARPES. The is caused by the variation of the spectral weight mentioned in subsection II C. On the other hand, the band structure shows the crossings formed by band folding in every BZ, see Supplementary Fig. S10, where the crossing region is highlighted by red ovals. This shows the importance of performing ARPES over multiple CDW BZs in the direction of qCDWq_{\mathrm{CDW}} to decipher the influence of the CDW on the electronic band structure.

Perpendicular to the KNL along ΓZ\Gamma Z (Λ\Lambda), the bands are represented by one dimensional irreps (Supplementary Fig. S15 and Supplementary Table S3). Since the little group along this direction has lesser symmetry than C2vC_{2v}, degeneracy is not enforced. This is shown by the spin-splitting along ΓZ\Gamma Z in the zoomed insets iv,viv,v of Figure 4a. This is also true for the bands calculated along various other high symmetry directions (Supplementary Fig. S17), where the insets show degeneracy along ΓX\Gamma X and splitting along other directions. In the ΓZ\Gamma Z direction no bands are found to cross the EFE_{F}, and this is corroborated by ARPES as well as EBS. Here, a hybridization related gap is observed with bands 13,1413,14 being the highest occupied

ones (red double arrow in Figure 4b,Figure 4c). This gap has been referred to in previous literature as the CDW gap in LaTe3_3 [36] and other RTe3_3 systems [40, 43, 44, 55, 56]. A discussion about the variation of the CDW gap of LaTe3_3 with kxk_x and kyk_y is provided in the Supplementary Note 4.

Discussions

In this work, we show the existence of the recently predicted [25] KNL in the CDW state of LaTe3_3, along the Σ\Sigma direction with C2vC_{2v} point group (Figure 3o) based on the results of ARPES, ab-initio DFT, EBS obtained by band unfolding and symmetry arguments. The origin of the KNL is associated with the time reversal symmetry (LaTe3_3 being non-magnetic in contrast to the other RTe3_3 materials) as well as the symmetries of the noncentrosymmetric and nonsymmorphic lattice (C2C_2 rotation along kxk_x and mirror in the kxk_x-kzk_z plane).

The crossings of the CDW-induced shadow and the main bands enforced by the KNL occur from E∼0.6E\sim0.6 eV, and disperse in EE to traverse the EFE_{\mathrm{F}} as kxk_x increases (Figure 3n). The dispersion in EE is related to the dispersion of the main band (the crossing of the main band with EFE_{\mathrm{F}} moves to larger kzk_z as kxk_x decreases, see Figure 2a-Figure 2d, Figure 2h-Figure 2k) coupled with the constraint that the shadow band is always separated from it by qCDWq_{\mathrm{CDW}}. Other spin degenerate bands also cross the EFE_{\mathrm{F}} along the KNL. We characterize LaTe3_3 as a KNL metal in the CDW state based on multiple bands crossing the EFE_{\mathrm{F}} both along and perpendicular to the KNL.

We have furthermore identified spinless nodal lines in a general direction on the kxk_x-kzk_z plane in the limit of negligible SOC (Figure 3h). These nondegenerate bands with distinct irreps involved in the crossings are linear and traverse the EFE_{\mathrm{F}} with high Fermi velocity. The nodal lines are entirely gapped with SOC. These characteristics resemble those of noncentrosymmetric topological nodal line semi-metals such as pinictides e.g., CaAgAs [57], in which the nodal line is entirely gapped by SOC resulting in a topological insulator [57, 58]. Whether the SOC-induced gap in LaTe3_3 has a topological character is an open question that would require further investigation.

The appearance of these two distinct types of nodal lines can be attributed to the bilayer splitting, that splits both the shadow and the main bands to bands belonging to different irreps. Thus, the foundation of our investigation is the unambiguous observation of the bilayer-split shadow bands by ARPES with which the theory is in impressive agreement.

This occurred due to the 7 fold modulated structure with qCDW=27c∗=0.286c∗q_{\mathrm{CDW}}=\frac{2}{7}c^*=0.286c^*, employing which DFT was done for the first time with qCDWq_{\mathrm{CDW}} and the amplitude of the CDW modulation close to the experimental value. However, the experimental values differ depending on the method: for example, electron diffraction showed qCDW=0.28±0.01c∗q_{\mathrm{CDW}}=0.28\pm0.01c^* [38], whereas XRD reported qCDW=0.2757(4)c∗q_{\mathrm{CDW}}=0.2757(4)c^* [37] in the temperature range of 90-100 K. Here, from ARPES we find qCDWq_{\mathrm{CDW}} to be 0.28±0.005c∗0.28\pm0.005c^*. Considering these values, it is possible that there might be a minor deviation of experimental qCDWq_{\mathrm{CDW}} from 27c∗\frac{2}{7}c^*; hence, is often considered as “incommensurate” [37, 38]. In order to answer the question whether the KNL would exist if qCDWq_{\mathrm{CDW}} deviates, we find that it is possible to construct a 29-fold (1×1×29) structure with qCDW=829c∗=0.2759c∗q_{\mathrm{CDW}}=\frac{8}{29}c^*=0.2759c^* from the experimental XRD structure [59]. Note that it has the same space group (SG #40) as the 7-fold structure. Its qCDWq_{\mathrm{CDW}} equals the experimental XRD value, which is most precise among the distinct methods, rounded to the third decimal place. Moreover, the required displacement of the 232 atoms in the primitive unit cell from the atomic positions from XRD [37] is zero (0.0000) for virtually all the atoms, with the exception of only a couple of Te atoms where it is 0.0001 Å. Since the symmetries of SGSG #40 as well as the TRS are retained in the 29-fold structure, it is expected that the KNL will remain intact.

In conclusion, the discovery of CDW-induced KNL in LaTe3_3 – a non-centrosymmetric, quasi-2D, TRS-preserving material – is the cornerstone of our work, which we feel will stimulate further investigation into the fascinating realm of CDW materials.

Methods

Experimental: Single crystals of LaTe3_3 with residual resistivity ratio (RRR, ρ(300 K)/ρ(2 K)\rho(300\ \mathrm{K})/\rho(2\ \mathrm{K})) of ∼\sim270 were grown by tellurium flux technique, as discussed in our previous work [30].

The ARPES measurements presented here were performed at the SGM3 beamline at the ASTRID2 synchrotron facility [60]. ARPES data at SGM3 beamline were collected with an energy resolution of 18 meV at hν=24.4h\nu=24.4 eV. The angular resolution was 0.2° (0.008 Å−1^{-1}). The measurements were performed at 100 K with different photon energies ranging from 13 eV to 140 eV. All the measurements were done on freshly cleaved surfaces at a chamber base pressure better than 2×10−102\times10^{-10} mbar. The ARPES intensity plots measured as a function of photon energy (hνh\nu) have been converted to kyk_y assuming the free electron final states [61]. The raw data (EkinE_{\mathrm{kin}} vs hνh\nu) have been converted to EE vs kyk_y by utilizing the expression ky=1ℏ2m(Ekincos⁡2θ+V0)k_y=\frac{1}{\hbar}\sqrt{2m(E_{\mathrm{kin}}\cos^2\theta+V_0)}, where EkinE_{\mathrm{kin}} is the kinetic energy of the photoelectrons, θ\theta is the emission angle and the inner potential V0V_0 has been assumed to be 7.5 eV based on the best matching with the ARPES data. The curvature plots have been obtained as in our earlier work [62, 63] with respect to both EE and kk axes following the method proposed by Zhang et al.[64]. This method improves the visibility of the weaker bands in the ARPES intensity plots and is better than the second derivative approach. The data analysis has been performed using IGOR pro (version 9). The dashed curves in Figure 2e,Figure 2f are obtained by a curve fitting the maxima of the MDCs. The maxima are identified by fitting the MDCs with Lorentzian functions. The least-square error method has been used for the curve fitting.

The STM measurements were carried out at a base pressure of 2×10−112 \times10^{-11} mbar using a variable temperature STM from Omicron Nanotechnology GmbH, while LEED and preliminary ARPES measurements were carried out in a workstation from Prevac sp. z o.o. STM was performed in the constant current mode using a tungsten tip that was cleaned by sputtering and voltage pulse method. The tip was biased and sample was kept at the ground potential. LEED is performed using a four grid rear view optics from OCI Vacuum Microengineering. Both STM and LEED were performed in the CDW state at room temperature.

Density functional theory: The DFT calculations have been performed for the 7-fold modulated structure with C2cmC2cm space group (SGSG #40). This structure has been derived from the experimental atomic positions at 100 K reported in Ref. 37 using the PSEUDO program with average displacement of the atoms being ≤0.01\le0.01 Å [59]. VESTA software has been used for Crystal structure visualization [65].

We have employed the DFT-based Vienna Ab-initio Simulation Package (VASP) [66, 67] within the framework of the projector augmented wave method (PAW) [66, 67] to investigate the electronic structure of LaTe3_3. The exchange-correlation functional is treated under the generalized gradient approximation (GGA) given by Perdew, Burke, and Ernzerhof [68]. We have considered 11 valence electrons of the La atom (5s25p65d16s25s^2 5p^6 5d^1 6s^2) and 6 valence electrons of Te atom (5s25p45s^2 5p^4) in the PAW pseudopotential. The energy cut-off is set to 500 eV for the expansion of the planewaves. The convergence criterion for energy in the self-consistent-field cycle and total force tolerance on each atom are taken to be 10−610^{-6} eV and 0.02 eV/Å, respectively. The SOC is employed by a second-variation method as implemented in the VASP code [67]. To calculate the EBS, we have unfolded the band structure of the CDW state into the primitive BZ of the non-CDW state, using the PyProcar python code [50]. All the DFT bands (and consequently the EBS) are rigidly shifted to larger EE by 0.1 eV with respect to the EFE_F for comparison with the ARPES data.

Furthermore, to reconfirm the band crossings and to identify their irreps, DFT calculations have been performed using the all electron WIEN2k programme package [69] and Quantum Espresso software package [70]. Supplementary Fig. S18 shows that the bands are in agreement between the different methods. WIEN2k was performed with energy cut-off of 16 Ry, where the RMTKmax⁡R_{\mathrm{MT}}K_{\max} value is taken to be 9.5. Further, we have used 10 for the maximum value of angular momentum for the (l,m) expansion of wave function or density. Convergence criteria for energy and charge have been taken to be 10−510^{-5} Ry and 0.001 e−e^{-}, respectively. Quantum Espresso calculations have been carried out using fully relativistic PAW pseudopotentials for La and Te atoms. A planewave cutoff of 80 Ry and a 6×6×16 \times6 \times1 k-grid were taken along with the energy accuracy of 10−810^{-8} Ry.

Acknowledgment

S.S., P.S. and S.R.B. gratefully acknowledge the financial support from Department of Science and Technology, Government of India within the framework of the DST-Synchrotron-Neutron Project to perform experiments at ASTRID2 synchrotron facility. A part of this work was supported by VILLUM FONDEN via the Centre of Excellence for Dirac Materials (Grant No. 11744). The Computer division of Raja Ramanna Centre for Advanced Technology is thanked for installing the DFT codes and providing support throughout.

Author contributions

S.S., P.S., D.C., and S.R.B. conducted the ARPES measurements with assistance and support from M.B. and P.H. LEED was carried out by S.S. and P.S., while STM was performed by V.K.S. J.B. and R.D. did the DFT calculations under the supervision of A.C. The explanation of the results was provided by S.S., T.D., A.C., and S.R.B. The single crystals of LaTe3_3 were grown by A.P., S.R. and P.M, the latter introduced us to this system. S.S. analyzed the experimental data with initial help from D.C., performed the post-analysis of the DFT results with some inputs from J.B., and prepared the figures. The project was planned and led by S.R.B. who wrote the paper with significant contributions from S.S., P.H., and A.C.

Competing interests

The authors declare no competing interests.

References

  1. [1]Wang, Y. et al., Axial higgs mode detected by quantum pathway interference in RTe₃, Nature 606, 896 (2022).DOI
  2. [2]Jiang, Y.-X. et al., Unconventional chiral charge order in kagome superconductor KV₃Sb₅, Nature Materials 20, 1353 (2021).DOI
  3. [3]Shi, W. et al., A charge-density-wave topological semimetal, Nature Physics 17, 381 (2021).
  4. [4]Rettig, L. et al., Persistent order due to transiently enhanced nesting in an electronically excited charge density wave, Nature Communications 7, 10459 (2016).DOI
  5. [5]Luo, H. et al., Electronic nature of charge density wave and electron-phonon coupling in kagome superconductor KV₃Sb₅, Nature Communications 13, 273 (2022).DOI
  6. [6]Yu, F. H. et al., Unusual competition of superconductivity and charge-density-wave state in a compressed topological kagome metal, Nature Communications 12, 3645 (2021).DOI
  7. [7]Zong, A. et al., Evidence for topological defects in a photoinduced phase transition, Nature Physics 15, 27 (2018).
  8. [8]Li, H. et al., Observation of unconventional charge density wave without acoustic phonon anomaly in kagome superconductors AV₃Sb₅ (A = Rb, Cs), Phys. Rev. X 11, 031050 (2021).DOI
  9. [9]Song, Z. et al., Observation of an incommensurate charge density wave in monolayer TiSe₂/CuSe/Cu(111) heterostructure, Phys. Rev. Lett. 128, 026401 (2022).DOI
  10. [10]Lv, B. Q. et al., Unconventional hysteretic transition in a charge density wave, Phys. Rev. Lett. 128, 036401 (2022).
  11. [11]Pan, H., Xie, M., Wu, F. and Das Sarma, S., Topological phases in ab-stacked MoTe₂/WSe₂: Z₂ topological insulators, chern insulators, and topological charge density waves, Phys. Rev. Lett. 129, 056804 (2022).DOI
  12. [12]D’Souza, S. W. et al., Coexistence of charge-density wave and ferromagnetism in Ni₂MnGa, Phys. Rev. B 85, 085123 (2012).DOI
  13. [13]Gooth, J. et al., Axionic charge-density wave in the weyl semimetal (TaSe₄)₂I, Nature 575, 315 (2019).DOI
  14. [14]Qian, X., Liu, J., Fu, L. and Li, J., Quantum spin hall effect in two-dimensional transition metal dichalcogenides, Science 346, 1344 (2014).
  15. [15]Polshyn, H. et al., Topological charge density waves at half-integer filling of a moiré superlattice, Nature Physics 18, 42 (2021).DOI
  16. [16]Mitsuishi, N. et al., Switching of band inversion and topological surface states by charge density wave, Nature Communications 11, 2466 (2020).DOI
  17. [17]Lei, S. et al., Band engineering of dirac semimetals using charge density waves, Advanced Materials 33, 2101591 (2021).DOI
  18. [18]Hsu, M.-C. et al., Topological theory of inversion-breaking charge-density-wave monolayer 1T-TiSe₂, New Journal of Physics 23, 093025 (2021).DOI
  19. [19]Bzdušek, T., Wu, Q., Rüegg, A., Sigrist, M. and Soluyanov, A. A., Nodal-chain metals, Nature 538, 75 (2016).
  20. [20]Xia, Y., Cai, X. and Li, G., Multitype dirac fermions protected by orthogonal glide symmetries in a noncentrosymmetric system, Phys. Rev. B 102, 041201 (2020).DOI
  21. [21]Gao, H. et al., Dirac-weyl semimetal: Coexistence of dirac and weyl fermions in polar hexagonal ABC crystals, Phys. Rev. Lett. 121, 106404 (2018).DOI
  22. [22]Oh, Y.-T., Min, H.-G. and Kim, Y., Dual topological nodal line and nonsymmorphic dirac semimetal in three dimensions, Phys. Rev. B 99, 201110 (2019).DOI
  23. [23]Leonhardt, A. et al., Symmetry-enforced topological band crossings in orthorhombic crystals: classification and materials discovery, Phys. Rev. Materials 5, 124202 (2021).DOI
  24. [24]Chang, G. et al., Topological quantum properties of chiral crystals, Nature Materials 17, 978 (2018).
  25. [25]Xie, Y.-M. et al., Kramers nodal line metals, Nature Communications 12, 3064 (2021).
  26. [26]Yu, R., Weng, H., Fang, Z., Dai, X. and Hu, X., Topological node-line semimetal and dirac semimetal state in antiperovskite Cu₃PdN, Phys. Rev. Lett. 115, 036807 (2015).
  27. [27]Shang, T. et al., Unconventional superconductivity in topological Kramers nodal-line semimetals, Science Advances 8, eabq6589 (2022).DOI
  28. [28]Kogar, A. et al., Light-induced charge density wave in LaTe₃, Nature Physics 16, 159 (2019).DOI
  29. [29]Zong, A. et al., Role of equilibrium fluctuations in light-induced order, Phys. Rev. Lett. 127, 227401 (2021).DOI
  30. [30]Pariari, A. et al., Interplay between charge density wave order and magnetic field in the nonmagnetic rare-earth tritelluride LaTe₃, Phys. Rev. B 104, 155147 (2021).DOI
  31. [31]Hu, B. F., Cheng, B., Yuan, R. H., Dong, T. and Wang, N. L., Coexistence and competition of multiple charge-density-wave orders in rare-earth tritellurides, Phys. Rev. B 90, 085105 (2014).DOI
  32. [32]Yumigeta, K. et al., Advances in rare-earth tritelluride quantum materials: Structure, properties, and synthesis, Advanced Science 8, 2004762 (2021).DOI
  33. [33]Singha, R., Pariari, A. K., Satpati, B. and Mandal, P., Large nonsaturating magnetoresistance and signature of nondegenerate dirac nodes in ZrSiS, Proceedings of the National Academy of Sciences 114, 2468 (2017).DOI
  34. [34]Johannes, M. D. and Mazin, I. I., Fermi surface nesting and the origin of charge density waves in metals, Phys. Rev. B 77, 165135 (2008).
  35. [35]Eiter, H.-M. et al., Alternative route to charge density wave formation in multiband systems, Proceedings of the National Academy of Sciences 110, 64 (2012).DOI
  36. [36]Brouet, V. et al., Angle-resolved photoemission study of the evolution of band structure and charge density wave properties in RTe₃ (R=Y, La, Ce, Sm, Gd, Tb, and Dy), Phys. Rev. B 77, 235104 (2008).DOI
  37. [37]Malliakas, C. D. and Kanatzidis, M. G., Divergence in the behavior of the charge density wave in RETe₃ (RE = rare-earth element) with temperature and RE element, Journal of the American Chemical Society 128, 12612 (2006).
  38. [38]DiMasi, E., Aronson, M. C., Mansfield, J. F., Foran, B. and Lee, S., Chemical pressure and charge-density waves in rare-earth tellurides, Phys. Rev. B 52, 14516 (1995).
  39. [39]Setyawan, W. and Curtarolo, S., High-throughput electronic band structure calculations: Challenges and tools, Computational Materials Science 49, 299 (2010).DOI
  40. [40]Komoda, H. et al., High-resolution angle-resolved photoemission study of incommensurate charge-density-wave compound CeTe₃, Phys. Rev. B 70, 195101 (2004).DOI
  41. [41]Mans, A. et al., Experimental proof of a structural origin for the shadow fermi surface of Bi2Sr2CaCu2O8, Physical Review Letters 96, 10.1103/physrevlett.96.107007 (2006).
  42. [42]Novoselov, K. S. et al., Two-dimensional gas of massless dirac fermions in graphene, Nature 438, 197 (2005).
  43. [43]Gweon, G.-H. et al., Direct observation of complete fermi surface, imperfect nesting, and gap anisotropy in the high-temperature incommensurate charge-density-wave compound SmTe₃, Phys. Rev. Lett. 81, 886 (1998).DOI
  44. [44]Brouet, V. et al., Fermi surface reconstruction in the cdw state of CeTe₃ observed by photoemission, Phys. Rev. Lett. 93, 126405 (2004).
  45. [45]Ohta, T., Bostwick, A., Seyller, T., Horn, K. and Rotenberg, E., Controlling the electronic structure of bilayer graphene, Science 313, 951 (2006).
  46. [46]Feng, D. L. et al., Bilayer splitting in the electronic structure of heavily overdoped Bi₂Sr₂CaCu₂O₈+δ, Phys. Rev. Lett. 86, 5550 (2001).
  47. [47]Yang, B.-J. and Kee, H.-Y., Searching for topological density-wave insulators in multiorbital square-lattice systems, Phys. Rev. B 82, 195126 (2010).DOI
  48. [48]Ku, W., Berlijn, T. and Lee, C.-C., Unfolding first-principles band structures, Phys. Rev. Lett. 104, 216401 (2010).
  49. [49]Popescu, V. and Zunger, A., Effective band structure of random alloys, Phys. Rev. Lett. 104, 236403 (2010).
  50. [50]Herath, U. et al., PyProcar: A python library for electronic structure pre/post-processing, Computer Physics Communications 251, 107080 (2020).DOI
  51. [54]Elcoro, L. et al., Double crystallographic groups and their representations on the bilbao crystallographic server, Journal of Applied Crystallography 50, 1457 (2017).
  52. [55]Lee, E. et al., The 7 × 1 fermi surface reconstruction in a two-dimensional f -electron charge density wave system: PrTe₃, Scientific Reports 6, 30318 (2016).
  53. [56]Chikina, A. et al., Charge density wave-generated fermi surfaces in NdTe₃, arXiv preprint , arXiv:2209.04226v1 (2022).DOI
  54. [57]Yamakage, A., Yamakawa, Y., Tanaka, Y. and Okamoto, Y., Line-node dirac semimetal and topological insulating phase in noncentrosymmetric pnictides CaAgX(X= P, As), Journal of the Physical Society of Japan 85, 013708 (2016).DOI
  55. [58]Lv, B. Q., Qian, T. and Ding, H., Experimental perspective on three-dimensional topological semimetals, Rev. Mod. Phys. 93, 025002 (2021).DOI
  56. [59]Capillas, C. et al., A new computer tool at the bilbao crystallographic server to detect and characterize pseudosymmetry, Zeitschrift für Kristallographie 226, 186 (2011).DOI
  57. [60]Hoffmann, S., Søndergaard, C., Schultz, C., Li, Z. and Hofmann, P., An undulator-based spherical grating monochromator beamline for angle-resolved photoemission spectroscopy, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 523, 441 (2004).DOI
  58. [61]Ngankeu, A. S. et al., Quasi-one-dimensional metallic band dispersion in the commensurate charge density wave of 1T−TaS₂, Phys. Rev. B 96, 195147 (2017).DOI
  59. [62]Singha, R. et al., Signatures of topological surface state and unconventional magnetotransport properties in elemental ruthenium, Submitted (2022).DOI
  60. [63]Sadhukhan, P. et al., Electronic structure of Au-Sn compounds grown on Au(111), Phys. Rev. B 100, 235404 (2019).DOI
  61. [64]Zhang, P. et al., A precise method for visualizing dispersive features in image plots, Review of Scientific Instruments 82, 043712 (2011).DOI
  62. [65]Momma, K. and Izumi, F., Vesta 3 for three-dimensional visualization of crystal, volumetric and morphology data, Journal of Applied Crystallography 44, 1272 (2011).
  63. [66]Kresse, G. and Furthmüller, J., Efficient iterative schemes for ab−initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54, 11169 (1996).
  64. [67]Kresse, G. and Joubert, D., From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B 59, 1758 (1999).DOI
  65. [68]Perdew, J. P., Burke, K. and Ernzerhof, M., Generalized gradient approximation made simple, Phys. Rev. Lett. 77, 3865 (1996).
  66. [69]Blaha, P. et al., WIEN2k : An augmented plane wave plus local orbitals program for calculating crystal properties, Techn. Universitat 60 (2001).
  67. [70]Baroni, S., de Gironcoli, S., Dal Corso, A. and Giannozzi, P., Phonons and related crystal properties from density-functional perturbation theory, Rev. Mod. Phys. 73, 515 (2001).

Paper details

vlnc:W_xsjjxgy7
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents