Observation of an Incommensurate Charge Density Wave in Monolayer ${\mathrm{TiSe}}_2/\mathrm{CuSe}/\mathrm{Cu}(111)$ Heterostructure

${\mathrm{TiSe}}_2$ is a layered material exhibiting a commensurate ($2\times 2\times 2$) charge density wave (CDW) with a transition temperature of $\sim 200\mathrm{K}$. Recently, incommensurate CDW in bulk ${\mathrm{TiSe}}_2$ draws great interest due to its close relationship with the emergence of superconductivity. Here, we report an incommensurate superstructure in monolayer ${\mathrm{TiSe}}_2/\mathrm{CuSe}/\mathrm{Cu}(111)$ heterostructure. Characterizations by low-energy electron diffraction and scanning tunneling microscopy show that the main wave vector of the superstructure is $\sim 0.41a^{*}$ or $\sim 0.59a^{*}$ (here $a^{*}$ is in-plane reciprocal lattice constant of ${\mathrm{TiSe}}_2$). After ruling out the possibility of moiré superlattices, according to the correlation of the wave vectors of the superstructure and the large indirect band gap below the Fermi level, we propose that the incommensurate superstructure is associated with an incommensurate charge density wave (I-CDW). It is noteworthy that the I-CDW is robust with a transition temperature over 600 K, much higher than that of commensurate CDW in pristine ${\mathrm{TiSe}}_2$. Based on our data and analysis, we present that interface effect may play a key role in the formation of the I-CDW state.

Figure 1

(a) Schematic illustrations of the growth process and the structure of monolayer ${\mathrm{TiSe}}_2/\mathrm{CuSe}/\mathrm{Cu}(111)$. (b) Large-scale STM image of monolayer ${\mathrm{TiSe}}_2$ grown on $\mathrm{CuSe}/\mathrm{Cu}(111)$ ($V_S=-2\mathrm{V}$, $I_S=50\mathrm{pA}$, $T=80\mathrm{K}$). (c) High-resolution STM image of monolayer $M\text{−}{\mathrm{TiSe}}_2$ in the red square in (b) ($V_S=-50\mathrm{mV}$, $I_S=1\mathrm{nA}$, $T=80\mathrm{K}$). (d) XPS of the Ti $2p$ core levels. (e) XPS of the Se $3d$ core levels. The blue and red curves are fitting curves corresponding to the Se $3d$ core levels from ${\mathrm{TiSe}}_2$ and CuSe, respectively. Discrete points represent the raw data and the black line is the fitting line.

Figure 2

(a),(d), and (g) High-resolution STM images of monolayer-${\mathrm{TiSe}}_2/\mathrm{CuSe}/\mathrm{Cu}(111)$ at different temperatures, showing the superstructure. The scanning parameters are $V_S=-0.05\mathrm{V}$, $I_S=0.1\mathrm{nA}$ in (a), and $V_S=-1\mathrm{V}$, $I_S=1\mathrm{nA}$ in (d) and (g), respectively. (b),(e), and (h) Fast Fourier transform (FFT) of STM images in (a),(d), and (g), respectively. The circles indicate the positions of the superstructure. (c),(f), and (i) Representative line cuts from the central point to ${\mathrm{TiSe}}_2$ point in (b),(e), and (h), respectively. (j) LEED pattern of the sample at 300 K, showing diffraction spots from Cu(111) substrate (white circles), ${\mathrm{TiSe}}_2$ (green circle), CuSe (red circle), two superstructure peaks 2 and 3 (yellow and cyan circles). The inset is the same LEED pattern with the sample rotated, so that the central area close to the reflected spot of incident electron beam (brown circle) is visible, showing the absence of peak 1 at this temperature. (k) LEED pattern of the sample at 600 K, showing one superstructure peak 2. (l) Schematic drawing of variations of the superstructure peaks with temperature.

Figure 3

(a) and (d) Constant-energy map at $E_F$, obtained by ARPES, at 300 and 5.4 K, respectively. The Brillouin zones of ${\mathrm{TiSe}}_2$ and CuSe are indicated by the red and blue hexagons. (b) and (e) ARPES intensity plots along $M^{\prime }-\mathrm{\Gamma }-{\mathrm{\Gamma }}^{\prime }$ at 300 and 5.4 K, respectively. The red and blue curves are energy distribution curves (EDCs) at $M$ and ${\mathrm{\Gamma }}^{\prime }$. (c) and (f) Detailed electronic band around $\mathrm{\Gamma }$ (along $\mathrm{\Gamma }\text{−}M$ direction) at 300 and 5.4 K, respectively.