Gate tuning of coupled electronic and structural phase transition in atomically thin Ta$_2$NiSe$_5$

Abstract

Realizing an excitonic insulator phase from narrow-gap semiconductors remains challenging, as unambiguous experimental signatures are difficult to establish. Ta$_2$NiSe$_5$ has been widely regarded as a leading candidate, yet the nature of its phase transition and insulating state remains controversial. Here, we report a systematic Raman spectroscopy study of Ta$_2$NiSe$_5$ as a function of thickness and field-effect doping, complemented by electrical transport measurements. The phase transition persists down to the monolayer limit, with the critical temperature increasing as thickness decreases. In bilayer samples, both electron and hole doping suppress the insulating state, with electron doping lowering and hole doping raising the transition temperature. Importantly, the quasi-elastic scattering, previously attributed to excitonic fluctuations, evolves monotonically across the entire doping range, inconsistent with the expected suppression of excitonic correlations by Coulomb screening. These findings rule out a dominant excitonic mechanism and instead point to a coupled electronic and structural phase transition, whose stability is tunable by carrier doping. Our doping-based approach offers a general strategy for evaluating the role of excitonic effects in candidate excitonic insulators.

Figures (5)

• Figure 1: Crystal structure and Raman characteristics of Ta$_2$NiSe$_5$.a Crystal structure of Ta$_2$NiSe$_5$ in the orthorhombic phase. The arrows in the lower part indicate atomic displacement in the monoclinic phase. b, c Temperature-dependent Raman spectra (b) and the corresponding intensity color plot (c) of bulk Ta$_2$NiSe$_5$. d Polarization-angle-dependent Raman intensity color plot of bulk Ta$_2$NiSe$_5$ at 300 K. e Fitting analysis of the low-energy Raman conductivity of bulk Ta$_2$NiSe$_5$ at 300 K. f Thickness-dependent Raman spectra of Ta$_2$NiSe$_5$ at 150 K. All data were collected in the crossed polarization ($ac$) configuration.
• Figure 2: Thickness and temperature dependent Raman conductivity.a--d Temperature-dependent Raman conductivity color plot for bulk, trilayer, bilayer, and monolayer samples. e Temperature dependence of the QES spectral weight. High-temperature data points are missing because the QES signal becomes too weak for reliable fitting. The solid lines are Gaussian fits. f--i Temperature dependence of the amplitude, frequency, full width, and $1/|q|$ of mode 2. j Thickness dependence of the temperature corresponding to the maximum or minimum values marked by arrows in e--i (filled symbols). The open squares are $T_{\mathrm{C}}$ values estimated from electrical transport measurements. Error bars in e--i are standard deviations obtained from fitting analysis, and those in j represent uncertainties in determining the characteristic temperatures. All data were collected in the crossed polarization ($ac$) configuration.
• Figure 3: Doping dependent electrical transport properties of bilayer Ta$_2$NiSe$_5$.a Schematic structure of the dual-gate bilayer Ta$_2$NiSe$_5$ device D17. Gr: graphite electrodes. b Dual-gate mapping of the resistance measured at 150 K. The dashed lines marks the charge neutrality and the arrows denote pure doping. c Doping-dependent resistance at selected temperatures. d Temperature-dependent resistance at charge neutrality and selected hole dopings. e Arrhenius plot of the resistance at typical doping levels. The solid lines are fits to the thermally activated temperature dependence over two temperature ranges. f Doping dependence of the activation gap. The shaded areas represent error bars obtained from the fitting analysis.
• Figure 4: Doping dependence of the Raman conductivity in bilayer Ta$_2$NiSe$_5$.a--d Doping-dependent Raman conductivity color plots for the bilayer sample D17 at selected temperatures. e--h Representative spectra corresponding to a--d at selected doping levels. All data were collected in the crossed polarization ($ac$) configuration.
• Figure 5: Doping and temperature dependent QES and mode 2 parameters for bilayer Ta$_2$NiSe$_5$ device D50.a, Color map showing the doping and temperature dependence of the QES spectral weight. b, Line cuts along the temperature axis at zero doping (undoped), the maximum electron doping (n-doped), and the maximum hole-doping (p-doped) in a. The solid lines are Gaussian fits to the data above 250 K, with the fitted peak center corresponding to the $T_{\mathrm{C}}$ and the full width at half maximum characterizing the temperature spread ($\Delta T$) of the QES. c Doping dependence of $T_{\mathrm{C}}$ (left axis) and $\Delta T$ (right axis). d--f, The upper panels show the doping and temperature dependent color maps of the frequency, amplitude, and $1/|q|$ for mode 2. The lower panels are the corresponding line cuts at the same doping levels as those in b. Error bars in b--f are standard deviations obtained from fitting analysis.