0% found this document useful (0 votes)
16 views8 pages

Ultrafast Charge Ordering by Self-Amplified Exciton-Phonon Dynamics in TiSe2

Uploaded by

20110190057
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
16 views8 pages

Ultrafast Charge Ordering by Self-Amplified Exciton-Phonon Dynamics in TiSe2

Uploaded by

20110190057
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

ARTICLE

https://doi.org/10.1038/s41467-019-13672-7 OPEN

Ultrafast charge ordering by self-amplified


exciton–phonon dynamics in TiSe2
Chao Lian 1, Sheng-Jie Zhang 1, Shi-Qi Hu 1, Meng-Xue Guan 1 & Sheng Meng 1,2,3*

The origin of charge density waves (CDWs) in TiSe2 has long been debated, mainly due to
the difficulties in identifying the timescales of the excitonic pairing and electron–phonon
coupling (EPC). Without a time-resolved and microscopic mechanism, one has to assume
simultaneous appearance of CDW and periodic lattice distortions (PLD). Here, we accom-
plish a complete separation of ultrafast exciton and PLD dynamics and unravel their interplay
in our real-time time-dependent density functional theory simulations. We find that laser
pulses knock off the exciton order and induce a homogeneous bonding–antibonding transition
in the initial 20 fs, then the weakened electronic order triggers ionic movements antiparallel
to the original PLD. The EPC comes into play after the initial 20 fs, and the two processes
mutually amplify each other leading to a complete inversion of CDW ordering. The self-
amplified dynamics reproduces the evolution of band structures in agreement with photo-
emission experiments. Hence we resolve the key processes in the initial dynamics of CDWs
that help elucidate the underlying mechanism.

1 Beijing National Laboratory for Condensed Matter Physics and Institute of Physics, Chinese Academy of Sciences, Beijing 100190, P. R. China. 2 School of

Physical Sciences, University of Chinese Academy of Sciences, Beijing 100190, P. R. China. 3 Songshan Lake Materials Laboratory, Dongguan, Guangdong
523808, P. R. China. *email: [email protected]

NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications 1


ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7

C
harge density wave (CDW) in 1T-TiSe2 has been one of movements exactly antiparallel to the original PLD, but cannot
the persistent eye-drawing topics over decades. It is not solely drives the observed inversion in CDW/PLD. Instead,
only an excellent playground to study the interplay assisted by EPC, a self-amplification mechanism between electron
between CDW and superconductivity1–23, but also an evidenced dynamics and lattice distortion emerges after the initial excita-
excitonic insulator24–28. It was heavily debated which mechanism tion, reproducing well experimental features observed in tr-
—the electron–phonon coupling (EPC)29–40 or the excitonic ARPES measurements. We propose that both exciton pairing and
pairing40–61—is the major driving force for the formation of EPC contribute to the CDW formation, albeit in a different
CDW in TiSe2 . These two mechanisms disagree on the role of timescale (see Fig. 1): CDW is predominantly initiated by exciton
periodic lattice distortions (PLD) in CDW: PLD is essential in binding (<20 fs) and subsequently enhanced by EPC (>20 fs).
forming the CDW according to the EPC mechanism, while it is These insights hint for a complete microscopic understanding on
only a passive consequence of the CDW if the excitonic pairing the nature of charge ordering in quantum materials.
dominates.
Isolating the PLD from the CDW can solve the debate, but it is
not achievable in the ground state. Inspiringly, the ultrafast mea- Results
surements can distinguish PLD and CDW based on their different Ground state properties. We calculate the atomic geometries of
time scales62. However, conclusions from previous ultrafast mea- the normal phase and the CDW 2 ´ 2 ´ 2 phase. The bulk TiSe2 is
surements are highly controversial. Utilizing time-resolved (tr) and used with the interlayer separation of 6.69 Å. The optimized PLD
angle-resolved photoemission spectroscopy (ARPES) measure- displacements {di } are shown in Fig. 2a, where the displacements
ments, Rohwer et al. observed very fast (<30 fs) collapses of the δ Ti ¼ 0:091 Å, δ Se ¼ 0:030 Åand δ Ti =δSe ¼ 3:03 : 1. These results
long-range order63, which was interpreted by Mathias et al. as the excellently reproduce the experimental measurements
signals of the mutually amplified carrier multiplication and gap δ Ti ¼ 0:085 ± 0:014 Åand δ Ti =δSe  3 : 182. To directly compare
quenching64. However, this mechanism completely neglects the with ARPES measurements, we calculate the effective band struc-
possible involvements of ionic dynamics and the effects of geometry ture (EBS) along K-M-Γ-K by unfolding energy bands from the
relaxation. Möhr-Vorobeva et al. observed the nonthermal melting 2 ´ 2 BZ to the 1 ´ 1 BZ, as shown in Fig. 2. The calculated EBS
within 250 fs in the ultrafast X-ray measurements and suggested agrees well with the experimental spectra. In the normal state, we
that excitonic pairing generates the CDW65. Hellmann et al. also reproduce the electron and hole pockets at M and Γ points,
supported the excitonic mechanism66. They observed that the 100 fs respectively. The CDW opens an indirect gap of 0.18 eV and creates
change in the tr-ARPES signals is comparable to the buildup time of folded valence bands (VB) at M and folded conduction bands (CB)
the electron-hole screening in the exciton formation. On the other at Γ. We note that the PBE+U approach with the Hubbard
hand, Porer et al. indicated that excitonic pairing was not the sole U Ti ¼ 3:5 eV incorrectly predicts that the CDW state is unstable83.
driving force of CDW67. They separately studied the electronic and Thus, we use U Ti ¼ 0 eV for all the dynamic simulations.
structural orders via monitoring the characteristic peaks of the two Based on the consistency between DFT results and experi-
in the transient energy loss spectra. They found that the PLD can mental measurements, we briefly discuss the accuracy of DFT in
persist with the quenched excitonic order. Despite extensive efforts describing the excitonic paring and EPC in TiSe2 . It is well known
in the past, the conclusions in all these previous studies were mainly that the semilocal exchange-correlation (XC) functionals (e.g.,
derived from indirect mappings between the spectra and the CDW/ PBE) poorly describes the long-range Coulomb screening
PLD orders. Notwithstanding the cutting-edge techniques used in 2π=jk0  kj  1 in vertical excitations q ¼ k 0  k ¼ 084. Com-
these studies, the time-resolved spectra can only provide the aver- putationally expensive corrections such as Bethe-Salpeter equation
aged response of the material without atomic resolution. With (BSE)85,86 can considerably improve the accuracy. The experi-
recent significant progresses in real-time (rt) time dependent (TD) mentally observed superlinear feature is absent in our simulations
density functional theory (DFT) algorithms and computing (Supplementary Fig. 2). However, the long-range attractions
power68–81, it become now possible to perform ultrafast quantum produce spatially uniform forces on the ions. Based on the concept
dynamics simulations fully from first principles, to provide a unified of the excitonic insulator24–26, the PLD stability is only affected by
atomic-level picture of ultrafast CDW dynamics. the inter-valley excitons formed by an attractive interaction VðwÞ
Here we take advantages of recently developed rt-TDDFT between the electron pocket at the M point and the hole pocket at
methods for ab initio simulations of ultrafast dynamics in com- the Γ point, as shown in Fig. 2d. Here, w ¼ ± bi =2 and bi
plex materials such as TiSe2 . We demonstrate that laser pulses (i ¼ 1; 2) being the reciprocal lattice vector along the ith direction.
knock down the CDW order without disturbing PLD through Therefore, VðwÞ is a short-range interaction with a characteristic
inducing a homogeneous bonding–antibonding electronic length scale 1=jwj ¼ a, where a ¼ jai j and ai is the lattice vector.
transition. The reduced CDW order then triggers the ionic Note VðwÞ is different from typical long-range interactions in

Exciton binding Electron–phonon coupling Phonon scattering


0 fs 20 fs 100 fs

Ti Lattice Band gap


distortion decrease
Se

Carrier
relaxation

Charge order melting Photoinduced forces Self-amplified process

Fig. 1 Schematic of atomic processes in photoexcited 1T-TiSe2 . The laser pulse melts charge order within 20 fs, producing the forces that trigger the ionic
movements. The self-amplified dynamics is assisted by electron–phonon couplings after initial excitation.

2 NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications


NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7 ARTICLE

a c
b1

δSe
2a2

δTi

Γ/Γ K K/K
2a1 w
M
b M/Γ
2×2
ZSe b2
1×1

d e
0.5
Ti 3d

0.0
Attraction Folding
V (W)
Energy (eV)

Se 4p
–0.5

–1.0

–1.5
K M Γ K K M Γ K

Fig. 2 Atomic structure and electronic properties of 1T-TiSe2 . a Top view and b side view of the atomic structure of 1T-TiSe2 . The silver, yellow and
orange spheres denote the Ti atoms, the Se atoms on the bottom layer and the Se atoms on the top layer, respectively. The blue and the red arrows denote
the PLD displacements {di } of the Ti and Se atoms, respectively. c Brillouin zones (BZ) of TiSe2 . The solid red lines and blue dash lines denote the BZ of the
2 ´ 2 and 1 ´ 1 cell, respectively. {Γ, M, K} and {  K}
Γ, M,  denote the special k points in 1 ´ 1 and 2 ´ 2 BZ, respectively. Effective band structures (EBS) of the
d normal state with U ¼ 0 eV (dots) and U ¼ 3:5 eV (dashed lines) and e CDW state. The plots in shaded areas in (d) and (e) are ARPES spectra
reproduced from ref. 63. The Fermi level is shifted to the experimental value.

Wannier excitons. The semilocal XC already includes exciton PLD. We analyze the differential charge density ρtot ðr; tÞ ¼
binding between the electron and hole pockets at M and Γ, ρchg ðr; tÞ  ρatom ðr; tÞ as a function of time, where ρchg is the charge
respectively, albeit slightly underestimating the screening effect. density and ρatom is the superposition of the atomic charge densities.
To quantitatively demonstrate the validity of the semilocal Thus, ρtot ðr; tÞ features the spacial distribution of the bonding (+)
functional in describing the intervalley exciton with momentum
and antibonding (−) densities. We divide ρtot ðr; tÞ into two parts
q ¼ w, we compare the linear-response TDDFT results obtained
ρ1 ´ 1 ðrÞ and ρCDW ðrÞ, where ρ1 ´ 1 ðrÞ ¼ ½ρtot ðr þ ai Þ þ ρtot ðrÞ=2,
from adiabatic PBE (APBE) and BSE kernels. At q ¼ w, the
and ρCDW ðrÞ ¼ ρtot ðrÞ  ρ1 ´ 1 ðrÞ. Obviously, ρ1 ´ 1 ðr þ ai Þ ¼
APBE kernel yields similar absorption spectra with those from
BSE kernel (Supplementary Fig. 3). Since the BSE kernel is a well- ρ1 ´ 1 ðrÞ characterizes the original 1 ´ 1 order, while ρCDW ðrÞ char-
accepted accurate description of excitons85,86, this indicates acterizes the strength of the CDW order. A laser-induced charge
that the electron-hole exchange effect has been well described difference is characterized by Δρi ðrÞ ¼ ρi ðr; t f Þ  ρi ðr; t 0 Þ (i ¼ tot,
in the semilocal XC. Therefore, due to the unique band 1 ´ 1, CDW). Figure 3 shows the two-dimensional contour of the
structures of TiSe2 , the semilocal XC yields acceptable excitonic charge density ρi ðx; y; tÞ (i ¼ tot, 1 ´ 1, CDW), which is ρi ðr; tÞ
interactions. averaged over z direction. Comparing Fig. 3a–c, we find that
Furthermore, the density functional perturbation theory and ρ1 ´ 1 ðx; y; t 0 Þ is the main ingredient of ρtot ðx; y; t 0 Þ, even with the
molecular dynamics calculations reproduce well experimental presence of PLD, while ρCDW ðx; y; t 0 Þ is localized around the Ti
phonon spectra87,88 and thermal conductivity83,89, confirming positions. Starting from the CDW ground state with the fixed PLD,
that DFT can accurately describe the EPC in TiSe2 . Thus, the the CDW order ρCDW ðx; y; tÞ decreases after the laser illumination:
semilocal XC and electron-ion dynamics simulations are suitable (i) The laser induces electron transfer from the bonding area to the
for tracking photoexcitation physics of TiSe2 . antibonding area. The induced charge Δρtot ðx; y; tÞ (Fig. 3g) is
opposite to ground state charge ρtot ðx; y; t 0 Þ (Fig. 3a). (ii) The
majority of the induced charge has the 1 ´ 1 periodicity, i.e.,
Dynamics of charge orders. To identify the dominant mechanism, Δρ1 ´ 1 ðx; y; tÞ (Fig. 3h) dominates Δρtot ðx; y; tÞ (Fig. 3g). (iii) The
we try to supply direct proofs by separating the CDW dynamics from most important feature is that the induced CDW charge density
the PLD in the time domain. Instead of directly creating CDW at the ΔρCDW ðx; yÞ (Fig. 3i) is opposite to the original ρCDW ðx; y; t 0 Þ
normal state, we aim at decreasing the CDW order with the fixed (Fig. 3c), indicating a 20% decrease in the excitonic order.

NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications 3


ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7

ρtot = ρ1 × 1 + ρCDW
0.2
a b c j 0.00

0.1 –0.056
ρ (t0) –0.02
–0.200 –0.195
0.0 |tot|
=

d e f |1 × 1| –0.04

Charge (a.u.)
|CDW|
ρ (tf)

–0.1
|Δtot| 0.06
–0.2 |Δ1 × 1|
|ΔCDW|

0.1 0.04
g h i
0.0 0.02
Δρ

k 0.00
–0.1
0 5 10 15
Time (fs)

Fig. 3 Dynamics of charge order. Two-dimensional contour plots for a ρtot ðx; y; t0 Þ, b ρ1 ´ 1 ðx; y; t0 Þ, c ρCDW ðx; y; t0 Þ, d ρtot ðx; y; tf Þ, e ρ1 ´ 1 ðx; y; tf Þ, f ρCDW ðx; y; tf Þ,
g Δρtot ðx; yÞ, h Δρ1 ´ 1 ðx; yÞ, i ΔρCDW ðx; yÞ, where t0 ¼ 0 and tf ¼ 20 fs. j The bonding charge and k anti-bonding charge as a function of time. The initial values in
(j) are shifted −0.2, −0.195, −0.056 for ρtot , ρ1 ´ 1 , ρCDW , respectively.

R
We analyze
R the integrated charges Qi ðtÞ ¼ jρi ðr; tÞjdr and dynamical PES of the PLD, as shown in Fig. 4c. The dynamical
Ci ðtÞ ¼ jρi ðr; tÞ  ρi ðr; t 0 Þjdr (i ¼ tot, 1 ´ 1, CDW) for quanti- potential energy decreases due to the in-plane PLD movements,
tative comparisons. The former characterizes the strength of together with the energy oscillations attributed to the out-of-
bonding, while the latter denotes the weakening of bonding states, plane A1g mode (Fig. 4g). This indicates that the laser induced
i.e., the strength of anti-bonding. As shown in Fig. 3j, the percentage PLD movements stabilize the system. Within the simulation time,
of the decrease ½Qi ðt f Þ  Qi ðt 0 Þ=Qi ðt 0 Þ is 23.6%, 23.8% and 19.7% di ðtÞ has been completely inverted di ðtÞ ¼ di ðt 0 Þ at around
for i ¼ tot, 1 ´ 1, CDW, respectively. Meanwhile, C i ðtÞ increases, 300 fs, lowering the dynamical PES and creating a new quasi-
with a ratio Ctot ðt f Þ : C1 ´ 1 ðt f Þ : CCDW ðt f Þ ¼ 1 : 0:98 : 0:20, slightly equilibrium state. We note that the inversion of PLD produces a
different from the ratio of the initial bonding charges degenerate CDW state which is symmetry-equivalent to the
Qtot ðt 0 Þ : Q1 ´ 1 ðt 0 Þ : QCDW ðt 0 Þ ¼ 1 : 0:98 : 0:27. Thus, laser-induced original state.
bonding–antibonding transfer is nearly homogeneous, lowering The above data imply a scenario of ultrafast PLD switch, which
both the 1 ´ 1 and CDW order proportionally. Since the decrease in includes three consecutive steps: (i) In the first 20 fs, the laser pulses
jQ1 ´ 1;bond j affects all chemical bonds homogeneously, as an overall pump the electrons from the bonding state to the anti-bonding state,
effect, the decrease in jQtot;bond j lowers the stability of CDW. which leads to the change of the PES. The calculated double-well
PES yields a vibration frequency of 4.3 THz, which is consistent with
the oscillation period of the amplitude mode of about 220–240 fs
Photoexcited lattice dynamics. The laser-induced CDW observed in dynamic simulations (Fig. 4d). This frequency is
instability would further trigger dynamic changes in PLD, which comparable to the experimental value of 3.4 THz observed for the
are simulated from first principles. Among the atoms in the layer A1g CDW amplitude mode90,91. The difference between theory and
of 2 ´ 2 cell, only one Ti atom and one Se atom are nontrivial
symmetry-inequivalent atoms, as shown in Fig. 4a. We present experiment is attributed to possible mixing of other phonon modes
the time-dependent displacement di ðtÞ(i ¼ Ti; Se) by the pro- and inadequate accuracy of semilocal functionals to treat low-energy
jection of di ðtÞ on the PLD direction δi ðtÞ ¼ di ðtÞ  di ðt 0 Þ=jdi ðt 0 Þj phonons. The softening of the double-well potential is evidenced by
and the angle θðtÞ between di ðtÞ and di ðt 0 Þ (Fig. 4b). We find that the fact that the CDW mode oscillations disappear when the laser
laser triggers a set of movements Δdi ðtÞ ¼ di ðtÞ  di ðt 0 Þ which fluence increases from I 0 =4 to I 0 . (ii) The lattice transforms into a
are exactly antiparallel to di ðt 0 Þ, noticing that δi decreases with all structure with the opposite PLD along the non-equilibrium TD-PES
during 20–300 fs. (iii) The photoexcited system might relax into the
cos θ  ± 1 as shown in Fig. 4d–e. Besides the in-plane move-
ground state after the recombination of photocarriers. This process
ments, the overall bonding–antibonding charge transition intro-
is beyond our simulations and requires multichannel and large-scale
duces an out-of-plane breathing mode between the Se layer and
modeling to properly account for the decoherence92–94 and
the Ti layer z Se ðtÞ (Fig. 4f). This is the established out-of-plane
dissipation effects95,96.
A1g mode with the periodicity of 175 fs (5.7 THz) when I ¼ I 0 =4,
We complete the story by briefly discussing the long-time
which is slightly smaller than the experimental value 5.9 THz90,91. behaviors. In the experiments, the ionic movements in different
Higher laser fluence I ¼ I 0 further decreases the frequency to 5.1 unit cells have different phases due to the finite thickness of the
THz, due to the weakened Ti–Se bonds by photocarriers. material, the inhomogeneous spatial distribution of the laser spot,
We analyze the dynamical potential energy surface (PES) as well as thermal fluctuations. Via phonon–phonon scattering,
related to the PLD dynamics. To focus on the in-plane PLD the equilibrium temperature gradually forms at the timescale of
movements other than the out-of-plane A1g mode, we plot the picoseconds. The thermally driven phase transitions may occur
minus of the kinetic energies of these two movements Ekin and through an increasing population of phonons and changes in the
Ekout , respectively. Since the total energy Etot ðtÞ ¼ Ep ðtÞ þ atomic potential via anharmonic phonon interactions. Thus, the
Ekin ðtÞ þ Ekout ðtÞ is conserved, where Ep ðtÞ is the potential nonthermal CDW dynamics observed here is separated from
energy, we use Ekin ðtÞ as a function of δðtÞ to characterize the thermal transitions at different time scales.

4 NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications


NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7 ARTICLE

a c (1:4%), the PLD dynamics is completely quenched (Fig. 4, dotted


3.0 350 fs
lines). It strongly implies that the increase in Ek ðtÞ is nonlinear,
Ti
Se 2.5 300 fs with a lower rate at the beginning. The fragile initial movement of
ii
δðtÞ will be completely quenched with a small dissipation rate. Since

Energy (meV atom–1)


250 fs
2.0 the PLD dynamics is sensitively dependent on its own trajectory, we
200 fs infer that there exists a self-amplified electron–phonon mechanism
b PLD direction
1.5 in the PLD dynamics initiated by exciton binding.
150 fs
i Accordingly, we demonstrate the impact of ionic movements
1.0
θ
100 fs on electronic structure through analyzing TD-EBS along Γ-M in
D

Fig. 5a–f. The laser pulse first shifts the EBS downwards and
is
pl
ac

0.5
enlarges the bandgap; then VBs are raised by 0.18 eV while CBs
em

iii 50 fs
en
t

δ 0.0 barely change during 20-80 fs, in excellent agreement with tr-
–0.15 –0.10 –0.05 0.00 0.05 0.10 0.15 ARPES data. It leads to the vanishing band gap and forms hole
 (Å)
and electron pockets, boosting the relaxation of hole carriers
d e towards the maximum of VBs (VBM). The carrier relaxation
0.1
1.0 lowers the total energy of the system. In comparison, in the
quenched case where ions barely move (Fig. 5h–l), the dynamic
In-plane  (Å)

0.5
electronic evolution—the gap closing and carrier relaxation—are
cos

0.0 Inversion 0.0


completely absent, implying that the measured changes in EBS
–0.5 are highly correlated with ionic movements. We thus identify a
Ti self-amplified process from these observations: the decrease in the
–1.0
Se
–0.1
PLD ! the decrease in the bandgap ! the decrease in the energy
f 0.3 g 0.02 of carriers ! the further decrease in PLD. We expect such a
In-plane ×20
dynamic interplay between electronic order and PLD, ignored in
Energy (eV atom–1)

Out-of-Plane
Out-of-plane δz (Å)

0.2
0.00 previous studies, would sustain in a variety of CDW materials
such as NbSe2 , TaS2 , LaTe3 , etc.89,97. The results highlight the
–0.02
entangled dynamics between different degree of freedoms in
0.1
–0.04 quantum materials.

0.0 –0.06
0 100 200 300 0 100 200 300 Discussion
Time (fs) Time (fs)
To complete our story, we note the following facts towards
Fig. 4 Photoexcited lattice dynamics. a All the symmetrically inequivalent building a unified picture for experimental observations.
atoms with nontrivial PLD. b The illustration of δ and θ. c The potential (i) The effective lattice temperature is relatively low during the
energy surfaces (PES). The black and colored lines denote the ground state whole laser-induced dynamics. The maximum kinetic energy of
PES and the non-equilibrium TD-PES, E½δðtÞ, respectively. The d δðtÞ, e ions is 0.06 eV atom1 in a temporary period (corresponding to a
cos θðtÞ, and f δzðtÞ as a function of time. g The minus of the kinetic transient effective temperature of 450 K). Although the low kinetic
energies of out-of-plane movements Ekin ðtÞ and in-plane movements energy of ions may not be directly connected to nonthermal
Ekout ðtÞ as a function of time. The solid, dash and dotted lines denote the processes, we believe the processes discussed above are primarily
I ¼ I0 , I ¼ I0 =4, and the quenched case (I ¼ I0 but with 1.4% energy nonthermal since they are short-lived (<300 fs) and the excited-
dissipation), respectively. state carriers are clearly present (Fig. 5). In this case the structural
soft-mode potential is relaxed or even driven to a single well-
potential by excitation of the electronic degree of freedom. This is
Dynamic interplay. The highly directional PLD movements are also consistent with the experimental observations65 and previous
clearly not the reason but the consequence of the reduced studies97–99.
excitonic order. In contrast to the complete PLD inversion for (ii) The CDW order is disturbed instantly by the laser pulse
the I ¼ I 0 case, di ðtÞ only oscillates around the original value within the timescale of 10 fs, since it is originated from homo-
when laser intensity decreases to I 0 =4. The formation/decay of geneous electronic transitions. No buildup time in electron-hole
the CDW/PLD is electronically initialized, which supports the screening is needed. It shows that, the 100 fs shift in tr-ARPES
excitonic mechanism. This intensity dependence is also signals66 is not caused by the buildup of electron-hole screening,
experimentally observed65,67: the A1g CDW mode is softened as but by the EPC-assisted PLD dynamics di ðtÞ. Apparently, the
phonon response within 100 fs is contradictory to the well-
the laser fluence increases, which serves as a precursor of the accepted timescale of phonon dynamics τ ph >1 ps. However, we
PLD melting. Meanwhile, there are extra factors in the fol- note that the photo-induced PLD dynamics are coherently driven
lowing dynamics. Assuming that the excitonic pairing is the not by the phonon–phonon scattering, but directly by the elec-
only driving force, an inversion of electronic order tronic transitions. Thus, the response shares the same timescale as
ρCDW ðr; t f Þ ¼ ρCDW ðr; t 0 Þ is required to reverse the PLD the electron–phonon scattering of 100 fs.
di ðtÞ ¼ di ðt 0 Þ. Considering ρCDW only decreases by about (iii) With the laser intensity used in our simulations, the laser
20%, the PLD should be only perturbed. pulses induce the PLD dynamics but barely change the intrinsic
We design another numerical experiment by introducing a phonon modes (e.g., out-of-plane A1g ). This explains the sus-
thermostat and removing a small fraction of kinetic energy at the tainable characteristic peaks of PLD with the melted exciton order
rate of 0.01 eV atom1 ps1 . We note that the kinetic energy Ek ðtÞ as previously observed in ref. 67.
increases by 0.66 eV atom1 ps1 within 60 fs, as shown in Fig. 4g. (iv) There are open discussions in literature on whether exci-
Thus the thermostat, affecting only 1.4% of the kinetic energy tonic interaction or the Jahn-Teller distortion is dominant for the
change, is expected to slightly delay the dynamics of PLD δðtÞ. formation of CDW in TiSe2 . Our simulations seem to indicate the
However, despite that the dissipation in the kinetic energy is small two effects could coexist in a short period perturbed by

NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications 5


ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7

0 fs 20 fs 40 fs 60 fs 80 fs Γ M
1 0.4 1.0
a b c d e f g
0.2

0 0.0
0.8
–0.2

–1 –0.4

–0.6 0.6
Energy (eV)

–2 –0.8
1 0.4
h i j k l m n
0.2 0.4

0 0.0

–0.2
0.2
–1 –0.4

–0.6

–2 –0.8 0.0
Γ M Γ M Γ M Γ M Γ M 0 20 40 60 0 20 40 60
Time (fs)

Fig. 5 Time dependent band structure. Snapshots of TD-EBS for (a–e) the PLD dynamics, and (h–l) the quenched case. EBS at Γ and M point as a function
of time for (f–g) PLD dynamics and (m-n) the quenched case. The color bar denotes the carrier population. The squares in (g) mark the experimental tr-
ARPES data, reproduced from ref. 66.

photoexcitation. From Figs. 3 and 4 it is shown that while there is similar to that used in experiment, reproduces the measured number of excited
no significant lattice distortion (thus no changes in Jahn-Teller carriers in TiSe2 (Supplementary Fig. 2).
interaction), the electronic order is suppressed by 20% in the first
20 fs, beaconing excitonic interactions. On the other hand, the Data availability
following self-amplifying process suggests dynamic Jahn-Teller The data that support the findings of this study are available from the corresponding
author upon reasonable request.
lattice distortion is crucial for the subsequent decay/formation of
CDW state.
We demonstrate an entangled exciton-phonon mechanism for Code availability
The code that was used to simulate the findings of this study is available from the
charge ordering in 1T-TiSe2 by ab initio ultrafast rt-TDDFT corresponding authors upon reasonable request.
simulations of photoexcitation dynamics. We show that laser
pulses knock down the CDW order via inducing a homogeneous
bonding–antibonding electronic transition. The weakened CDW Received: 3 February 2019; Accepted: 14 November 2019;
order then triggers ionic movements exactly anti-parallel to the
original PLD, but cannot solely drive the observed PLD inversion.
Instead, assisted by the EPC, a self-amplification mechanism
drives the PLD dynamics initiated by laser excitation, in excellent
agreement with the tr-ARPES measurements. We propose that References
both the excitonic pairing and the EPC contribute to the CDW/ 1. Morosan, E. et al. Superconductivity in Cux TiSex . Nat. Phys. 2, 544–550
(2006).
PLD formation, but in a different timescale: the excitonic pairing 2. Li, G. et al. Semimetal-to-semimetal charge density wave transition in 1T-
initializes the formation of CDW, while the EPC promotes the TiSe2 . Phys. Rev. Lett. 99, 027404 (2007).
following dynamics through a self-amplification process. 3. Kusmartseva, A. F., Sipos, B., Berger, H., Forró, L. & Tutiš, E. Pressure induced
superconductivity in pristine 1T-TiSe2 . Phys. Rev. Lett. 103, 236401 (2009).
4. Yao, Q. et al. Charge transfer effects in naturally occurring van der waals
Methods heterostructures (PbSe)1:16 (TiSe1:16 )1:16 (m = 1, 2). Phys. Rev. Lett. 120,
Numerical calculations. Following our previous efforts in developing time
106401 (2018).
dependent ab initio package (TDAP)100–102, we implemented TDDFT algorithm in
5. Wei, M. J. et al. Manipulating charge density wave order in monolayer 1T-
the plane-wave code Quantum Espresso103. The details are described in Sup-
TiSe2 by strain and charge doping: a first-principles investigation. Phys. Rev. B
plementary Note I. We used the projector augmented-waves method (PAW)104
96, 165404 (2017).
and the Perdew-Burke-Ernzerhof (PBE) exchange-correlation (XC) functional105
in both DFT and TDDFT calculations. Pseudopotentials were generated using 6. Kogar, A. et al. Observation of a charge density wave incommensuration near
pslibrary106. The plane-wave energy cutoff was set to 55 Ry. Brillouin zone the superconducting dome in Cux TiSex . Phys. Rev. Lett. 118, 027002 (2017).
(BZ) was sampled using Monkhorst-Pack scheme107 with a 6 ´ 6 ´ 3 k-point mesh. 7. Yan, S. et al. Influence of domain walls in the incommensurate charge density
Band unfolding techniques were utilized to generate the EBS108,109 with a modified wave state of Cu intercalated 1T-TiSe2 . Phys. Rev. Lett. 118, 106405 (2017).
version of BandUP code110–112. Onsite Coulomb repulsion U ¼ 3:5 eV was added 8. Medvecká, Z. et al. Observation of a transverse Meissner effect in Cux TiSex
to the Ti atom to reproduce the experimental band structure, while we used U ¼ 0 single crystals. Phys. Rev. B 93, 100501 (2016).
in the dynamic TDDFT calculations and structural optimization. The electron 9. Das, T. & Dolui, K. Superconducting dome in MoS2 and TiSe2 generated by
timestep δt is 4 ´ 104 a.u. = 0:1935 attosecond and the ion timestep Δt is 0:194 fs. quasiparticle-phonon coupling. Phys. Rev. B 91, 094510 (2015).
Laser pulses with a wavelength λ ¼ 800 nm, width σ ¼ 12 fs and fluence 10. Luna, K., Wu, P. M., Chen, J. S., Morosan, E. & Beasley, M. R. Point-contact
I 0 ¼ 2:1 mJ cm2 centered at t ¼ 10 fs is utilized. The linear-response TDDFT tunneling spectroscopy measurement of Cux TiSex : disorder-enhanced
calculations were carried out with the YAMBO package113. This setup, which is Coulomb effects. Phys. Rev. B 91, 094509 (2015).

6 NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications


NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7 ARTICLE

11. Ganesh, R., Baskaran, G., van den Brink, J. & Efremov, D. V. Theoretical 44. Stoffel, N., Lévy, F., Bertoni, C. & Margaritondo, G. Direct evidence for d-
prediction of a time-reversal broken chiral superconducting phase driven by band involvement in the TiSe2 phase transition. Solid State Commun. 41, 53
electronic correlations in a single TiSe2 layer. Phys. Rev. Lett. 113, 177001 (1982).
(2014). 45. Monney, G., Monney, C., Hildebrand, B., Aebi, P. & Beck, H. Impact of
12. Joe, Y. I. et al. Emergence of charge density wave domain walls above the electron-hole correlations on the 1T-TiSe2 electronic structure. Phys. Rev. Lett.
superconducting dome in 1T-TiSe2 . Nat. Phys. 10, 421 (2014). 114, 086402 (2015).
13. Kačmarčík, J. et al. Heat capacity of single-crystal Cux TiSex superconductors. 46. Cazzaniga, M. et al. Ab initiomany-body effects in TiSe2 : a possible excitonic
Phys. Rev. B 88, 020507 (2013). insulator scenario from GW band-shape renormalization. Phys. Rev. B 85,
14. Husaníková, P. et al. Magnetization properties and vortex phase diagram of 195111 (2012).
Cux TiSex single crystals. Phys. Rev. B 88, 174501 (2013). 47. Monney, C. et al. Temperature-dependent photoemission on 1T-TiSe2 :
15. Iavarone, M. et al. Evolution of the charge density wave state in Cux TiSex . Interpretation within the exciton condensate phase model. Phys. Rev. B 81,
Phys. Rev. B 85, 155103 (2012). 155104 (2010).
16. Jeong, J., Jeong, J., Noh, H.-J., Kim, S. B. & Kim, H.-D. Electronic structure 48. Anderson, O., Manzke, R. & Skibowski, M. Three-dimensional and relativistic
study of Cu-doped 1T-TiSe2 by angle-resolved photoemission spectroscopy. effects in layered 1T-TiSe2 . Phys. Rev. Lett. 55, 2188 (1985).
Physica C 470, S648 (2010). 49. Peng, J.-P. et al. Molecular beam epitaxy growth and scanning tunneling
17. Zaberchik, M. et al. Possible evidence of a two-gap structure for the Cux TiSex microscopy study of TiSe2 ultrathin films. Phys. Rev. B 91, 121113 (2015).
superconductor. Phys. Rev. B 81, 220505 (2010). 50. Koley, S., Laad, M. S., Vidhyadhiraja, N. S. & Taraphder, A. Preformed
18. Giang, N. et al. Superconductivity at 2.3 K in the misfit compound(PbSe)1:16 excitons, orbital selectivity, and charge density wave order in 1T-TiSe2 . Phys.
(TiSe1:16 )1:16 . Phys. Rev. B 82, 024503 (2010). Rev. B 90, 115146 (2014).
19. Morosan, E. et al. Multiple electronic transitions and superconductivity in Pdx 51. Novello, A. et al. Stripe and short range order in the charge density wave of
TiSex . Phys. Rev. B 81, 094524 (2010). 1T-Cux TiSex . Phys. Rev. Lett. 118, 017002 (2017).
20. Hillier, A. D. et al. Probing the superconducting ground state near the charge 52. Monney, C., Monney, G., Aebi, P. & Beck, H. Electron-hole fluctuation phase
density wave phase transition in Cu0:06 TiSe0:06 . Phys. Rev. B 81, 092507 in 1T-TiSe2 . Phys. Rev. B 85, 235150 (2012).
(2010). 53. Monney, C. et al. Probing the exciton condensate phase in 1T-TiSe2 with
21. Barath, H. et al. Quantum and classical mode softening near the charge- photoemission. New J. Phys. 12, 125019 (2010).
density-wave-superconductor transition of Cux TiSex . Phys. Rev. Lett. 100, 54. Monney, C., Battaglia, C., Cercellier, H., Aebi, P. & Beck, H. Exciton
106402 (2008). condensation driving the periodic lattice distortion of 1T-TiSe2 . Phys. Rev.
22. Li, S. Y., Wu, G., Chen, X. H. & Taillefer, L. Single-gaps-wave Lett. 106, 106404 (2011).
superconductivity near the charge-density-wave quantum critical point in Cux 55. Monney, C. et al. Spontaneous exciton condensation in 1T-TiSe2 : BCS-like
TiSex . Phys. Rev. Lett. 99, 107001 (2007). approach. Phys. Rev. B 79, 045116 (2009).
23. Cui, X. Y. et al. Direct evidence of band modification and suppression of 56. Sugawara, K. et al. Unconventional charge-density-wave transition in
superstructure in TiSe2 upon Fe intercalation: an angle-resolved monolayer 1T-TiSe2 . ACS Nano 10, 1341 (2015).
photoemission study. Phys. Rev. B 73, 085111 (2006). 57. Hildebrand, B. et al. Short-range phase coherence and origin of the 1T-TiSe2
24. Kohn, W. Excitonic phases. Phys. Rev. Lett. 19, 439 (1967). charge density wave. Phys. Rev. B 93, 125140 (2016).
25. Jérome, D., Rice, T. M. & Kohn, W. Excitonic insulator. Phys. Rev. 158, 462 58. Watanabe, H., Seki, K. & Yunoki, S. Charge-density wave induced by
(1967). combined electron-electron and electron-phonon interactions in 1T-TiSe2 : a
26. Halperin, B. I. & Rice, T. M. Possible anomalies at a semimetal-semiconductor variational Monte Carlo study. Phys. Rev. B 91, 205135 (2015).
transistion. Rev. Mod. Phys. 40, 755 (1968). 59. Pillo, T. et al. Photoemission of bands above the Fermi level: the excitonic
27. Rossnagel, K., Kipp, L. & Skibowski, M. Charge-density-wave phase transition insulator phase transition in 1T-TiSe2 . Phys. Rev. B 61, 16213 (2000).
in 1T-TiSe2 : excitonic insulator versus band-type Jahn-Teller mechanism. 60. May, M. M., Brabetz, C., Janowitz, C. & Manzke, R. Charge-density-wave
Phys. Rev. B 65, 235101 (2002). phase of 1T-TiSe2 : the influence of conduction band population. Phys. Rev.
28. Cercellier, H. et al. Evidence for an excitonic insulator phase in 1T-TiSe2 . Lett. 107, 176405 (2011).
Phys. Rev. Lett. 99, 146403 (2007). 61. Chen, C., Singh, B., Lin, H. & Pereira, V. M. Reproduction of the charge
29. Hughes, H. P. Structural distortion in TiSe2 and related materials-a possible density wave phase diagram in 1T-TiSe2 exposes its excitonic character. Phys.
Jahn-Teller effect? J. Phys. C Solid State 10, L319 (1977). Rev. Lett. 121, 226602 (2018).
30. Wakabayashi, N., Smith, H., Woo, K. & Brown, F. Phonons and charge 62. Sundaram, S. K. & Mazur, E. Inducing and probing non-thermal transitions in
density waves in 1T-TiSe2 . Solid State Commun. 28, 923 (1978). semiconductors using femtosecond laser pulses. Nat. Mater. 1, 217 (2002).
31. Gaby, J. H., De Long, B., Brown, F., Kirby, R. & Lévy, F. Origin of the 63. Rohwer, T. et al. Collapse of long-range charge order tracked by time-resolved
structural transition in TiSe2 . Solid State Commun. 39, 1167 (1981). photoemission at high momenta. Nature 471, 490 (2011).
32. Motizuki, K., Suzuki, N., Yoshida, Y. & Takaoka, Y. Role of electron-lattice 64. Mathias, S. et al. Self-amplified photo-induced gap quenching in a correlated
interaction in lattice dynamics and lattice instability of 1T-TiSe2 . Solid State electron material. Nat. Commun. 7, 12902 (2016).
Commun. 40, 995 (1981). 65. Möhr-Vorobeva, E. et al. Nonthermal melting of a charge density wave in
33. Lopez-Castillo, J. M. et al. Phonon-drag effect in TiSe2x S2x mixed TiSe2 . Phys. Rev. Lett. 107, 036403 (2011).
compounds. Phys. Rev. B 36, 4249 (1987). 66. Hellmann, S. et al. Time-domain classification of charge-density-wave
34. Holt, M., Zschack, P., Hong, H., Chou, M. Y. & Chiang, T.-C. X-ray studies of insulators. Nat. Commun. 3, 1069 (2012).
phonon softening in TiSe2 . Phys. Rev. Lett. 86, 3799 (2001). 67. Porer, M. et al. Non-thermal separation of electronic and structural orders in a
35. Bussmann-Holder, A. & Bishop, A. R. Suppression of charge-density persisting charge density wave. Nat. Mater. 13, 857 (2014).
formation in TiSe2 by Cu doping. Phys. Rev. B 79, 024302 (2009). 68. Runge, E. & Gross, E. K. U. Density-functional theory for time-dependent
36. van Wezel, J., Nahai-Williamson, P. & Saxena, S. S. Exciton-phonon-driven systems. Phys. Rev. Lett. 52, 997 (1984).
charge density wave in TiSe2 . Phys. Rev. B 81, 165109 (2010). 69. Bertsch, G. F., Iwata, J.-I., Rubio, A. & Yabana, K. Real-space, real-time
37. Rossnagel, K. Suppression and emergence of charge-density waves at the method for the dielectric function. Phys. Rev. B 62, 7998 (2000).
surfaces of layered 1T-TiSe2 and 1T-TaS2 by in situRb deposition. New J. Phys. 70. Yabana, K., Nakatsukasa, T., Iwata, J.-I. & Bertsch, G. F. Real-time, real-space
12, 125018 (2010). implementation of the linear response time-dependent density-functional
38. Calandra, M. & Mauri, F. Charge-density wave and superconducting dome in theory. Phys. Status Solidi B 243, 1121 (2006).
TiSe2 from electron-phonon interaction. Phys. Rev. Lett. 106, 196406 (2011). 71. Otobe, T. et al. First-principles electron dynamics simulation for optical
39. Zhu, Z., Cheng, Y. & Schwingenschlögl, U. Origin of the charge density wave breakdown of dielectrics under an intense laser field. Phys. Rev. B 77, 165104
in 1T-TiSe2 . Phys. Rev. B 85, 245133 (2012). (2008).
40. Zenker, B., Fehske, H., Beck, H., Monney, C. & Bishop, A. R. Chiral charge 72. Otobe, T., Yabana, K. & Iwata, J.-I. First-principles calculation of the electron
order in 1T-TiSe2 : importance of lattice degrees of freedom. Phys. Rev. B 88, dynamics in crystalline SiO2. J. Phys. Condens. Matter 21, 064224 (2009).
075138 (2013). 73. Otobe, T., Shinohara, Y., Sato, S. A. & Yabana, K. Femtosecond time-resolved
41. Kidd, T. E., Miller, T., Chou, M. Y. & Chiang, T.-C. Electron-hole coupling dynamical Franz-Keldysh effect. Phys. Rev. B 93, 045124 (2016).
and the charge density wave transition in TiSe2 . Phys. Rev. Lett. 88, 226402 74. Yabana, K., Sugiyama, T., Shinohara, Y., Otobe, T. & Bertsch, G. F. Time-
(2002). dependent density functional theory for strong electromagnetic fields in
42. Monney, C., Monney, G., Aebi, P. & Beck, H. Electron-hole instability in 1T- crystalline solids. Phys. Rev. B 85, 045134 (2012).
TiSe2 . New J. Phys. 14, 075026 (2012). 75. Shinohara, Y. et al. First-principles description for coherent phonon
43. van Wezel, J., Nahai-Williamson, P. & Saxena, S. S. An alternative generation in diamond. J. Phys. Condens. Matter 22, 384212 (2010).
interpretation of recent ARPES measurements on TiSe2 . EPL-Europhys. Lett. 76. Shinohara, Y. et al. Coherent phonon generation in time-dependent density
89, 47004 (2010). functional theory. Phys. Rev. B 82, 155110 (2010).

NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications 7


ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-019-13672-7

77. Shinohara, Y. et al. Nonadiabatic generation of coherent phonons. J. Chem. 104. Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B 50, 17953
Phys. 137, 22A527 (2012). (1994).
78. Sato, S. A. et al. Time-dependent density functional theory of high-intensity 105. Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation
short-pulse laser irradiation on insulators. Phys. Rev. B 92, 205413 (2015). made simple. Phys. Rev. Lett. 77, 3865 (1996).
79. Sato, S. A., Taniguchi, Y., Shinohara, Y. & Yabana, K. Nonlinear electronic 106. Corso, A. D. Pseudopotentials periodic table: from H to Pu. Computational
excitations in crystalline solids using meta-generalized gradient Materials Science 95, 337 (2014).
approximation and hybrid functional in time-dependent density functional 107. Monkhorst, H. J. & Pack, J. D. Special points for Brillouin-zone integrations.
theory. J. Chem. Phys. 143, 224116 (2015). Phys. Rev. B 13, 5188 (1976).
80. Ren, J., Vukmirović, N. & Wang, L.-W. Nonadiabatic molecular dynamics 108. Ku, W., Berlijn, T. & Lee, C.-C. Unfolding first-principles band structures.
simulation for carrier transport in a pentathiophene butyric acid monolayer. Phys. Rev. Lett. 104, 216401 (2010).
Phys. Rev. B 87, 205117 (2013). 109. Popescu, V. & Zunger, A. Extracting E versus k effective band structure from
81. Wang, Z., Li, S.-S. & Wang, L.-W. Efficient real-time time-dependent density supercell calculations on alloys and impurities. Phys. Rev. B 85, 085201
functional theory method and its application to a collision of an ion with a 2D (2012).
material. Phys. Rev. Lett. 114, 063004 (2015). 110. Medeiros, P. V. C., Stafström, S. & Björk, J. Effects of extrinsic and intrinsic
82. Salvo, F. J. D., Moncton, D. E. & Waszczak, J. V. Electronic properties and perturbations on the electronic structure of graphene: retaining an effective
superlattice formation in the semimetal TiSe2 . Phys. Rev. B 14, 4321 (1976). primitive cell band structure by band unfolding. Phys. Rev. B 89, 041407
83. Bianco, R., Calandra, M. & Mauri, F. Electronic and vibrational properties of (2014).
TiSe2 in the charge-density-wave phase from first principles. Phys. Rev. B 92, 111. Medeiros, P. V. C., Tsirkin, S. S., Stafström, S. & Björk, J. Unfolding spinor
094107 (2015). wave functions and expectation values of general operators: Introducing the
84. Sharma, S., Dewhurst, J. K., Sanna, A. & Gross, E. K. U. Bootstrap unfolding-density operator. Phys. Rev. B 91, 041116 (2015).
approximation for the exchange-correlation kernel of time-dependent density- 112. Lian, C. & Meng, S. Dirac cone pairs in silicene induced by interface Si-Ag
functional theory. Phys. Rev. Lett. 107, 186401 (2011). hybridization: a first-principles effective band study. Phys. Rev. B 95, 245409
85. Sottile, F., Olevano, V. & Reining, L. Parameter-free calculation of response (2017).
functions in time-dependent density-functional theory. Phys. Rev. Lett. 91, 113. Marini, A., Hogan, C., Grüning, M. & Varsano, D. yambo: an ab initio tool for
056402 (2003). excited state calculations. Comput. Phys. Commun. 180, 1392–1403
86. Marini, A., Del Sole, R. & Rubio, A. Bound excitons in time-dependent (2009).
density-functional theory: optical and energy-loss spectra. Phys. Rev. Lett. 91,
256402 (2003).
87. Weber, F. et al. Electron-phonon coupling and the soft phonon mode in TiSe2 . Acknowledgements
Phys. Rev. Lett. 107, 266401 (2011). We acknowledge partial financial supports from MOST (grants 2016YFA0300902 and
88. Singh, B., Hsu, C.-H., Tsai, W.-F., Pereira, V. M. & Lin, H. Stable charge 2015CB921001), NSFC (grants 91850120, 11774396, 11934003), and CAS
density wave phase in a 1T-TiSe2 monolayer. Phys. Rev. B 95, 245136 (2017). (XDB07030100).
89. Hellgren, M. et al. Critical role of the exchange interaction for the electronic
structure and charge-density-wave formation in TiSe2 . Phys. Rev. Lett. 119, Author contributions
176401 (2017). S.M. conceived and directed the research. C.L. performed the calculations and analyzed
90. Snow, C. S., Karpus, J. F., Cooper, S. L., Kidd, T. E. & Chiang, T.-C. Quantum the data. C.L., S.-J.Z., S.-Q.H., M.-X.G. and S.M. participated in the discussion. C.L. and
melting of the charge-density-wave state in 1T-TiSe2 . Phys. Rev. Lett. 91, S.M. wrote the paper with contributions from all the other authors.
136402 (2003).
91. Wang, H. et al. Large-area atomic layers of the charge-density-wave conductor
TiSe2. Adv. Mater. 30, 1704382 (2018). Competing interests
92. Floss, I. et al. Ab initio multiscale simulation of high-order harmonic The authors declare no competing interests.
generation in solids. Phys. Rev. A 97, 011401 (2018).
93. Yamada, S., Noda, M., Nobusada, K. & Yabana, K. Time-dependent density
functional theory for interaction of ultrashort light pulse with thin materials.
Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/s41467-
Phys. Rev. B 98, 245147 (2018).
019-13672-7.
94. Floss, I., Lemell, C., Yabana, K. & Burgdörfer, J. Incorporating decoherence
into solid-state time-dependent density functional theory. Phys. Rev. B 99,
Correspondence and requests for materials should be addressed to S.M.
224301 (2019).
95. Tully, J. C. Molecular dynamics with electronic transitions. J. Chem. Phys. 93,
Peer review information Nature Communications thanks the anonymous reviewers for
1061 (1990).
their contribution to the peer review of this work. Peer reviewer reports are available.
96. Gossel, G. H., Agostini, F. & Maitra, N. T. Coupled-trajectory mixed
quantum-classical algorithm: a deconstruction. J. Chem. Theory. Comput. 14,
Reprints and permission information is available at http://www.nature.com/reprints
4513 (2018).
97. Zhang, J. et al. Photoexcitation induced quantum dynamics of charge density
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
wave and emergence of a collective mode in 1T-TaS2. Nano Lett. 19,
published maps and institutional affiliations.
6027–6034 (2019).
98. Lian, C., Zhang, S. B. & Meng, S. Ultrafast carrier relaxation and its Pauli drag
in photo-enhanced melting of solids. https://arxiv.org/abs/1901.00609 (2019).
99. Lian, C., Zhang, S. B. & Meng, S. Ab initio evidence for nonthermal Open Access This article is licensed under a Creative Commons
characteristics in ultrafast laser melting. Phys. Rev. B 94, 184310 (2016). Attribution 4.0 International License, which permits use, sharing,
100. Meng, S. & Kaxiras, E. Real-time, local basis-set implementation of time- adaptation, distribution and reproduction in any medium or format, as long as you give
dependent density functional theory for excited state dynamics simulations. J. appropriate credit to the original author(s) and the source, provide a link to the Creative
Chem. Phys. 129, 054110 (2008). Commons license, and indicate if changes were made. The images or other third party
101. Lian, C., Hu, S.-Q., Guan, M.-X. & Meng, S. Momentum-resolved TDDFT material in this article are included in the article’s Creative Commons license, unless
algorithm in atomic basis for real time tracking of electronic excitation. J. indicated otherwise in a credit line to the material. If material is not included in the
Chem. Phys. 149, 154104 (2018). article’s Creative Commons license and your intended use is not permitted by statutory
102. Lian, C., Guan, M., Hu, S., Zhang, J. & Meng, S. Photoexcitation in solids: regulation or exceeds the permitted use, you will need to obtain permission directly from
first-principles quantum simulations by real-time TDDFT. Advanced Theory the copyright holder. To view a copy of this license, visit http://creativecommons.org/
and Simulations 1, 1800055 (2018). licenses/by/4.0/.
103. Giannozzi, P. et al. QUANTUM ESPRESSO: a modular and open-source
software project for quantum simulations of materials. J. Phys. Condens.
Matter 21, 395502 (2009). © The Author(s) 2020

8 NATURE COMMUNICATIONS | (2020)11:43 | https://doi.org/10.1038/s41467-019-13672-7 | www.nature.com/naturecommunications

You might also like