Since ferroelectricity was first observed in 2011, HfO2-based ferroelectrics have garnered significant attention, owing to their compatibility with complementary metal–oxide–semiconductors. Moreover, their thickness scalability facilitates the miniaturization of integrated circuit systems. The ultrafast polarization switching speed in the range of sub-nanoseconds helps in the fabrication of fast-operation devices. The origins of ferroelectricity in HfO2-based ferroelectrics differ from those of conventional perovskite ferroelectrics, with more complex behaviors associated with polarization switching. In this Perspective, recent investigations on the complex behaviors pertaining to polarization switching, including wake-up, split-up, fatigue, negative capacitance, accumulative switching, and some of their relations are discussed. Furthermore, the polarization switching dynamics have also been studied. Finally, the potential applications and investigations of HfO2-based ferroelectrics are discussed.

Ferroelectrics exhibit spontaneous polarization that can be switched by external electric fields. The switchable polarization endows ferroelectrics with significant potential in numerous device applications, such as nonvolatile memories, ferroelectric field effect transistors, and microelectromechanical devices.1 Among various ferroelectrics, perovskite ferroelectrics have been extensively studied for decades, owing to their outstanding electromechanical properties.2 However, the applications of conventional perovskite ferroelectrics in nanoscale devices are limited by their thickness limit and poor compatibility with complementary metal–oxide–semiconductors (CMOS), which consequently hamper the miniaturization of the devices.3 

Since fluorite HfO2-based materials were first discovered to exhibit ferroelectric properties in 2011,4 they have been recommended as one of the promising ferroelectric materials due to their compatibility with CMOS and the high thickness scalability. These properties counter the drawbacks of the perovskite ferroelectrics. The difference in structure compared to perovskite ferroelectrics leads to a distinctive origin of ferroelectricity in fluorite HfO2-based ferroelectrics.5 The different origin of ferroelectricity and the structural complexity result in the complex polarization switching behaviors of HfO2-based ferroelectrics. Accordingly, the elucidation of the polarization switching behaviors in HfO2-based ferroelectrics has become an urgent requirement.6–13 

The polarization switching behavior of HfO2-based ferroelectrics is more complex than that of conventional perovskite ferroelectrics. For instance, although the wake-up effect was reported in both perovskites14 and HfO2-based ferroelectrics,15 the mechanisms of the effect in HfO2-based ferroelectrics can be more complicated because of the complex structural evolution.16,17 While HfO2-based ferroelectrics exhibit the electrical fatigue effect18,19 similar to perovskite ferroelectrics, the underlying mechanisms can prove to be complicated, like those of the wake-up effect. However, the reversible split-up effect has not been detected in perovskite ferroelectrics.20,21 This increases the potential applications of HfO2-based ferroelectrics. Apart from these, other complex phenomena induced by polarization switching also exist in HfO2-based ferroelectrics, such as negative capacitance (NC),22,23 accumulative switching,24,25 and resistive switching.26 Furthermore, the switching dynamics have been explored based on various models18,22,27–29 such as the Kolmogorov–Avrami–Ishibashi (KAI),8 nucleation-limited switching (NLS),30 and inhomogeneous field mechanism (IFM) models.30–33 

In this perspective, we discuss recent studies undertaken on the switching behavior of HfO2-based ferroelectrics. In Sec. II, the origin of ferroelectricity in the fluorite structure will be discussed. Following this, the complex polarization switching behaviors and switching dynamics will be discussed in Secs. III and IV, respectively. Finally, we will conclude and discuss the perspectives of the switching behavior in HfO2-based ferroelectrics.

The structural origin of spontaneous polarization in ferroelectrics is a non-centrosymmetric phase. Several possible structures were reported for HfO2-based materials, including monoclinic (M), tetragonal (T), orthorhombic (O), and cubic (C), among which only the O-phase (Pca21, No. 29) is non-centrosymmetric.36,37 Besides that, polar rhombohedral (R) phase was also found in epitaxially strained Hf0.5Zr0.5O2 thin films.38 For pure HfO2 with an Si substrate, non-ferroelectric M structure is the most stable phase, as is evident from the grazing incidence x-ray diffraction (GI-XRD) result in Fig. 1(a).34 It was found that the polar O-phase can be stabilized via Zr doping. Other than Zr, various dopants were found to be capable of stabilizing the polar O-phase, including Si,4 Y,39 Al,40 and Gd,40 to name a few. However, the coexistence of O, M, and T phases was also observed. Using a top electrode TiN as a mechanical confinement, the polar O-phase can be more stabilized as shown in Fig. 1(b). Apart from doping and mechanical confinement, thickness also plays an important role in the stabilization of the polar O-phase. For un-doped HfO2 with a top electrode and Hf0.5Zr0.5O2 grown on TiN/Si, ferroelectricity disappeared for thicknesses exceeding 20–30 nm.41,42 However, recent studies have found that the polar O-phase can be stabilized in 930 nm-thick 7% Y:HfO2 films. Up to 70% of the O-phase was observed via XRD and the well-shaped polarization–electric field (P–E) hysteresis loops; moreover, the ferroelectricity was insensitive to the thickness.43 Different material processing methods can also give a significant impact on it. Li et al. grew epitaxial Zr:HfO2 and Si:HfO2 thin films by pulsed laser deposition (PLD), showing a pure polar Pca21 O phase with preferred orientations.44,45 The switching behavior was confirmed by piezoresponse force microscopy (PFM). Besides, other factors including surface energy effect, thermal expansion mismatch, and island coalescence can be crucial for the stabilization of polar O-phase.46 Recently, another possible origin of ferroelectricity in HfO2 thin films was also suggested,47 in which ferroelectricity was induced by electrochemical coupling from the special configuration of oxygen vacancies in HfO2 thin films.

FIG. 1.

(a) GI-XRD patterns of HfO2, Hf05Zr0.5O2, and ZrO2, with strong reflections labeled, (b) in situ GI-XRD patterns of TiN-capped and un-capped Hf0.5Zr0.5O2 samples. (a) and (b) Reproduced with permission from Müller et al., Nano Lett. 12(8), 4318–4323 (2012). Copyright 2012 American Chemical Society. Schematic of polarization switching of (c) perovskite PZT and (d) O-phase HfO2. Reproduced with permission from Clima et al., Appl. Phys. Lett. 104(9), 092906 (2014). Copyright 2014 AIP Publishing LLC. (e) P–E hysteresis loops of 10 nm-thick Si:HfO2 from 100 K to 350 K. Reproduced with permission from Schultheiß et al., Acta Mater. 157, 355–363 (2018). Copyright 2018 Acta Materialia Inc.

FIG. 1.

(a) GI-XRD patterns of HfO2, Hf05Zr0.5O2, and ZrO2, with strong reflections labeled, (b) in situ GI-XRD patterns of TiN-capped and un-capped Hf0.5Zr0.5O2 samples. (a) and (b) Reproduced with permission from Müller et al., Nano Lett. 12(8), 4318–4323 (2012). Copyright 2012 American Chemical Society. Schematic of polarization switching of (c) perovskite PZT and (d) O-phase HfO2. Reproduced with permission from Clima et al., Appl. Phys. Lett. 104(9), 092906 (2014). Copyright 2014 AIP Publishing LLC. (e) P–E hysteresis loops of 10 nm-thick Si:HfO2 from 100 K to 350 K. Reproduced with permission from Schultheiß et al., Acta Mater. 157, 355–363 (2018). Copyright 2018 Acta Materialia Inc.

Close modal

In conventional perovskite ferroelectrics such as Pb(Zr,Ti)O3 (PZT), polarization switching is induced by the movement of Zr4+/Ti4+ ions under external electric fields, as shown in Fig. 1(c). Meanwhile, the polarization of HfO2-based ferroelectrics switches via the displacement of oxygen atoms along the c-axis away from the equilibrium position, as shown in Fig. 1(d).5,36 The different modes of oxygen atoms movements (parallel or anti-parallel with respect to each other) lead to different phonon vibration modes. Lee et al. reported that the mixture of different vibration modes result in flat phonon bands in HfO2, leading to robust switchable electric dipoles with the scale down to the length of half a cell (about 3 Å). As a result, 2D slices of polarized regions are formed between thin non-polar regions with no cost of formation energy and no interactions.48,49 Although exhibiting different switching mechanisms, HfO2-based ferroelectrics show well-characterized ferroelectric properties. It was reported that 10 nm-thick Si:HfO2 exhibited well-shaped P–E hysteresis loops in a wide temperature range as shown in Fig. 1(e).35 

HfO2-based ferroelectrics possess complex polarization switching behaviors under external electric bias, including wake-up, split-up, NC, and accumulative switching. It is crucial to study these phenomena to elucidate their switching behaviors.

Wake-up, also called de-pinning, de-aging, or self-rejuvenation, is a phenomenon that a pinched P–E hysteresis loop of a pristine ferroelectric HfO2-based sample opens up, accompanied by an increase of remnant polarization (Pr), after certain treatment of electric field cycling. The wake-up has been widely detected and investigated in polycrystalline HfO2-based ferroelectrics.50–52 A typical wake-up process is shown in Fig. 2(a) and the inset of Fig. 2(e), in which the de-pinning of the P–E hysteresis loop and an increase of Pr can be seen.

FIG. 2.

(a) PE hysteresis loops of 10 nm-thick Si:HfO2 thin films with 5.6 mol. % Si during the electric field cycling, TEM images (b) before and (c) after the wake-up process of 10 nm-thick Si:HfO2 thin films with 5.6 mol. % Si; insets display the corresponding FFT images. (a)–(c) Reproduced with permission from Martin et al., Adv. Mater. 26(48), 8198–8202 (2014). Copyright 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. PD distribution of Sr:HfO2 capacitors in the (d) pristine [(e) wake-up] state, the corresponding main hysteresis loop shown as solid curve in the inset [dashed curves corresponds to wake-up [(d) pristine] states]. Reproduced with permission from Schenk et al., ACS Appl. Mater. Interfaces 7(36), 20224–20233 (2015). Copyright 2015 American Chemical Society.

FIG. 2.

(a) PE hysteresis loops of 10 nm-thick Si:HfO2 thin films with 5.6 mol. % Si during the electric field cycling, TEM images (b) before and (c) after the wake-up process of 10 nm-thick Si:HfO2 thin films with 5.6 mol. % Si; insets display the corresponding FFT images. (a)–(c) Reproduced with permission from Martin et al., Adv. Mater. 26(48), 8198–8202 (2014). Copyright 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. PD distribution of Sr:HfO2 capacitors in the (d) pristine [(e) wake-up] state, the corresponding main hysteresis loop shown as solid curve in the inset [dashed curves corresponds to wake-up [(d) pristine] states]. Reproduced with permission from Schenk et al., ACS Appl. Mater. Interfaces 7(36), 20224–20233 (2015). Copyright 2015 American Chemical Society.

Close modal

Researchers found that structural changes accompany the variation of electrical properties during electric field cycling. A pinched and unclosed P–E hysteresis loop shown in Fig. 2(a) was observed for the pristine Si:HfO2 films. After several cycles, a well-shaped loop without pinching was detected, indicating the wake-up effect. Phase compositions before and after wake-up was also examined by transmission electron microscopy (TEM), as shown in Figs. 2(b) and 2(c). A transition from M-phase to O/C-phase was observed based on the fast Fourier transform (FFT) result. Similar results were also found for Sr:HfO250 and Gd:HfO216 capacitors, in which transitions from M-phase to O-phase after wake-up were observed.

However, the structural evolution during the wake-up process was not distinctively characterized by XRD, which is common for structural characterizations. Fields et al. observed the wake-up effect in Zr:HfO2 capacitors;53 however, no distinctive change was observed in the XRD patterns during cycling, except for a slight decrease of the full width at half-maximum of the T/O superimposed peak. The decrease was attributed to the phase exchange between T- and O-phases. A similar result was reported by Hoffmann et al. in Gd:HfO2.54 The reason why the structure evolution was not distinctively observed by XRD may be that the structure evolutions occurred locally in regions with size much smaller than the diameter of the x-ray light source in conventional XRD.16 The XRD results may also contain regions in which wake-up did not happen.

As a universal technique that is sensitive to hysteretic properties,55 the first-order reversal curve (FORC) was applied to analyze the switching properties of Sr:HfO2 in pristine and wake-up states.20 The calculated distributions of Preisach density (PD) with respect to the electric field (E) and the reversal electric field (Er) are presented in Figs. 2(d) and 2(e) with corresponding main hysteresis loops (solid curve) shown in the insets (with the dashed curve representing the other states). The PD distribution in the ErE coordinate system can be represented in the bias field (Ebias)–coercive field (EC) coordinate system via the relation Ebias = (E + Er)/2 and EC = (E − Er)/2, namely, a 45° rotation of the ErE coordinate system. A double-center distribution with strong internal bias fields was detected in the pristine state, as shown in Fig. 2(d), a collective effect of which resulted in a pinched P–E hysteresis loop. As a result of the cycling, the two peaks gradually merged into one single peak with the centers of the bias fields almost decreasing to zero, as shown in Fig. 2(e). This led to the de-pinching of the P–E hysteresis loop and weakening of the imprint (horizontal shift of the P–E hysteresis loop). The result is in accordance with other reports on the wake-up treatment.27,50,56 The magnitude of PD represents the intensity of the response of the polarization switching. Higher values of the PD magnitude represent a larger number of switched polarization dipoles and/or that the switching is more complete. Therefore, the increased PD magnitude in the wake-up state, as exhibited in Fig. 2(e), was therefore revealed as a collective effect of the increase of the Pr value.

The mechanisms behind domain evolutions during the wake-up process were investigated via PFM. The domains of La:HfO2 capacitors in the pristine state were only partially switched by voltages of ±3 V [Figs. 3(a) and 3(b)], while more domains were switched in the wake-up state [Figs. 3(c) and 3(d)]. This suggests that the domain walls were pinned by defects, such as oxygen vacancies or trapped electrons in the pristine state. Meanwhile, the domain walls were de-pinned by the redistribution of defects during cycling, allowing the switching of more domains. In addition, a phase transition from the M-phase to the O-phase after the wake-up process was possible. The de-pinning was not symmetric, and some domains with downward polarization were still pinned, as is evident from Fig. 3(c). Meanwhile, nearly all of the upward domains were switched downward by a voltage of +3 V [Fig. 3(d)]. Such effects of defect accumulation around domain walls were widely studied in perovskite ferroelectrics.57–59 For HfO2-based ferroelectrics, there are indirect experimental results on the pinning and de-pinning effects.60 However, to the best of our knowledge, there is no direct result on the interaction between defects and domain walls in HfO2-based ferroelectrics. Two of the possible reasons may be that the minimum switchable ferroelectric dipole unit can be half a length of a single cell (3 Å), sandwiched between two non-polar domain walls with the same size (3 Å)48 requiring measurements with atomic resolution to detect such effects directly, and multiple structure phases coexist in the material. Therefore, the interactions between defects and domain walls can be hard to be distinguished directly.

FIG. 3.

PFM phase images of 10 nm-thick La:HfO2 capacitors in pristine [(a) and (b)] and wake-up states [(c) and (d)] after −3 V [(a) and (c)] and +3 V poling [(b) and (d)]. (a)–(d) Reproduced with permission from Buragohain et al., Appl. Phys. Lett. 112(22), 222901 (2018). Copyright 2018 AIP Publishing LLC. (e) Schematic of the three types of domains, and overlapped PFM phase images during the wake-up process of Zr:HfO2 capacitors, corresponding to (f) 1.5 and 1 cycle, (g) 14.5 and 15 cycles, and (h) 999.5 and 1000 cycles. (e)–(h) Reproduced with permission from Chouprik et al., ACS Appl. Electron. Mater. 1(3), 275–287 (2019). Copyright 2019 American Chemical Society.

FIG. 3.

PFM phase images of 10 nm-thick La:HfO2 capacitors in pristine [(a) and (b)] and wake-up states [(c) and (d)] after −3 V [(a) and (c)] and +3 V poling [(b) and (d)]. (a)–(d) Reproduced with permission from Buragohain et al., Appl. Phys. Lett. 112(22), 222901 (2018). Copyright 2018 AIP Publishing LLC. (e) Schematic of the three types of domains, and overlapped PFM phase images during the wake-up process of Zr:HfO2 capacitors, corresponding to (f) 1.5 and 1 cycle, (g) 14.5 and 15 cycles, and (h) 999.5 and 1000 cycles. (e)–(h) Reproduced with permission from Chouprik et al., ACS Appl. Electron. Mater. 1(3), 275–287 (2019). Copyright 2019 American Chemical Society.

Close modal

Chouprik et al. studied the domain evolution of (111)-textured Zr:HfO2 capacitors in detail28 and found that three types of domains existed at the initial stages of the wake-up process: (1) normal domains with polarization dipoles aligned along the external field; (2) un-switched static domains; and (3) anomalous domains with polarization switched opposite to the external field, as shown in Figs. 3(f)3(h). The anomalous domains were found to occupy 70% of the capacitor area and were found to decrease with increasing Pr during the cycling process. It was reported that the PFM signals may exhibit non-piezoelectric effects, including electrostatic effect, electrochemical strain, electrostriction effect, flexoelectricity effect, and joule heating.61 It can be speculated that the anomalous domains may arise from the non-piezoelectric effects. Besides, the wake-up effect may also be associated with the polycrystalline nature of the materials, because it has rarely been detected for single-phase epitaxial HfO2-based ferroelectrics.62,63 However, the wake-up effect was observed for epitaxial Zr:HfO2 films thinner than 4.6 nm, which may result from the effects of electrodes.64 

Owing to the increasing need to fabricate more powerful devices, enhancing the ferroelectricity of materials has become an urgent concern. Being an effective method to control polarization switching, control of the wake-up process has drawn significant attention. Bouaziz et al. successfully controlled the wake-up effect in Zr:HfO2 capacitors via different deposition methods.65 It was also found that the wake-up behaviors varied significantly for different dopants.66 Other properties strongly associated with polarization switching, such as negative capacitance, also exhibit strong correlations with the wake-up behavior; this will be discussed later.

Split-up is a complex phenomenon of HfO2-based ferroelectrics and has not been reported in conventional ferroelectrics with perovskite structures. The phenomenon was first reported by Schenk et al.21 The split-up effect is the splitting of the already merged transient current peak into two peaks when an already woken-up sample is subjected to electric field cycles. This is a sub-loop of the wake-up process, as shown in Fig. 4(a). In addition to that, more complex double-split-up phenomena were observed when the single-split-up sample is cycled, and subjected to the sub-loop of the single-split-up process, as shown in Fig. 4(e). Shown in Figs. 4(b), 4(d), and 4(f) are the distributions of the PD of Sr:HfO2 during the split-up process.20 Split-up occurs on HfO2-based ferroelectrics in the wake-up state after an external voltage lower than the voltage for the wake-up is applied, as shown in Fig. 4(a). The split-up was attributed to the local imprint effect.20 The sub-loop cycling injected charges into the material, imprinting the hysterons (switching units in FORC based on the Preisach model) within the E and Er ranges of the sub-loop cycling. The material exhibited double-peak PD distribution, as shown in Fig. 4(b) (single-split-up) and its behavior reverted to its pre-wake-up condition. A subsequent series of voltages [Fig. 4(c)] remerged the double peaks to form a single peak, as shown in Fig. 4(d). Therefore, the transition between wake-up and single-split-up states is reversible. Furthermore, if a series of voltages [lower than the voltages for the single-split-up, as shown in Fig. 4(e)] is applied to an already single-split-up sample, the double-peak PD distribution would split into four peaks, as shown in Fig. 4(f). The mechanism of the complex split-up behavior was attributed to the local imprint effect induced by injected charges from the electrodes.20 

FIG. 4.

Complex cycling effects of Sr:HfO2 capacitors: Schematics of the processes of single-split-up, re-merging, and double-split-up [(a), (c), and (e), respectively], and corresponding distribution of Preisach density and the MHP shown as a solid curve and other states shown as dashed curves in the insets [(b), (d), and (f)]. Reproduced with permission from Schenk et al., ACS Appl. Mater. Interfaces 7(36), 20224–20233 (2015). Copyright 2015 American Chemical Society.

FIG. 4.

Complex cycling effects of Sr:HfO2 capacitors: Schematics of the processes of single-split-up, re-merging, and double-split-up [(a), (c), and (e), respectively], and corresponding distribution of Preisach density and the MHP shown as a solid curve and other states shown as dashed curves in the insets [(b), (d), and (f)]. Reproduced with permission from Schenk et al., ACS Appl. Mater. Interfaces 7(36), 20224–20233 (2015). Copyright 2015 American Chemical Society.

Close modal

Recently, it was reported by Mayergoyz and Korman that if a physical system with hysteretic behavior could be fabricated as a device, it would be possible to produce a brand new memory device by utilizing the staircase-like interface in the Preisach plane.67 Combined with the complex and reversible split-up behavior of HfO2-based ferroelectrics exhibiting multi-center PD distribution in the Preisach plane, the fabrication of a new and powerful memory device would be feasible.

Electrical fatigue, which is an effect of decreasing Pr with cycling, is one of the most detrimental effects in the application of memory device because it can severely shorten the lifetime. Despite numerous advantages, HfO2-based ferroelectrics still exhibit inferior endurance against electric field cycling, owing to the higher coercive field. The best cycling endurance ever reported is in La:Hf0.5Zr0.5O2 thin films, reaching 4 × 10 10 cycles;68 however, some BiFeO3-based ferroelectrics were reported to be fatigue-free.69,70 Therefore, elucidating the fatigue mechanisms to further improve the cycling endurance of HfO2-based ferroelectrics is highly desirable.

The fatigue mechanisms of HfO2-based ferroelectrics have been widely investigated. Numerous researchers suggested that the generation of new defects, such as oxygen vacancies during cycling, were responsible for the fatigue effect.50,68,71,72 Besides, the charge injection could be another culprit for the fatigue effect.50,73 Moreover, it was suggested that domain wall pinning during electric field cycling could contribute to the fatigue effect. The above-mentioned fatigue mechanisms have also been widely investigated in conventional ferroelectrics with perovskite structures; however, the true mechanism is not still clear.74,75 Lou et al. proposed a model called local phase decomposition and claimed it to be generic.76 The model can indeed account for the various fatigue phenomena in perovskite ferroelectrics, but so far, there is no direct evidence of local phase decomposition in HfO2-based ferroelectrics. However, this may be probable with further studies. Recently, the recovery of the cycling endurance failure of HfO2-based FeFET via annealing was reported.77 Recovery of ferroelectric properties by heating is a crucial indicator of local phase decomposition for perovskite ferroelectrics, owing to the recovery of the pyrochlore phase to the perovskite phase.76 It can be expected that there is a heat-induced polar phase in the fatigued HfO2-based ferroelectrics after annealing.

For the application of memory devices, the speed of data reading and writing is strongly associated with the speed of polarization switching of the memory cell. As discussed above, the domain wall velocity of HfO2-based ferroelectrics is two orders slower than that of PZT capacitors,27 suggesting that the polarization switching should be significantly slower. It has been reported that the switching time of PZT-based ferroelectrics, which was confirmed experimentally, can be as fast as 220 ps.79 Researchers also have endeavored to investigate the switching speed of HfO2-based ferroelectrics.28,78,80–82 Most of the researchers achieved the switching time in the range of several nanoseconds, while Lyu et al. managed a switching time of 360 ps for a 15 nm-thick Zr:HfO2 device with a crossbar array structure, as shown in Fig. 5.83 A clear transient current peak can be seen in Fig. 5(a). The corresponding evolution of polarization exhibited a stable polarization state after switching. Moreover, the switching time is influenced by the area of the capacitor. The sample with an area of 8.4 μm2 displayed the strongest current peak and shortest switching time. Considering that the switching speeds of Zr:HfO2 and PZT are of the same order, and the domain wall movements, which are extrinsic contributions, are two orders slower for Zr:HfO2 than PZT, it can be inferred that the intrinsic polarization switching speed may be faster for HfO2-based ferroelectrics than perovskite ferroelectrics. Note that the domain wall movements are significantly slower. Moreover, it is evident that non-ferroelectric phases (e.g., M, T, or non-polar O phases) are distributed randomly and can even be mixed with the ferroelectric phase;84 furthermore, domain growth may lead to larger domain size than the grain size27 for HfO2-based ferroelectrics. It is reasonable to expect some special interactions between the domain walls and the non-ferroelectric phase during polarization switching.

FIG. 5.

(a) Transient current and (b) corresponding polarization of Zr:HfO2 capacitors with different areas with respect to time. Reproduced with permission from Lyu et al., Technical Digest—International Electron Devices Meeting, IEDM, December 2019 (IEEE, 2019), pp. 342–345. Copyright 2019 IEEE.

FIG. 5.

(a) Transient current and (b) corresponding polarization of Zr:HfO2 capacitors with different areas with respect to time. Reproduced with permission from Lyu et al., Technical Digest—International Electron Devices Meeting, IEDM, December 2019 (IEEE, 2019), pp. 342–345. Copyright 2019 IEEE.

Close modal

Studies were carried out to further accelerate the polarization switching. Polarization switching was found to be faster for antiferroelectric Zr:HfO2 (4.5 ns) than for ferroelectric Zr:HfO2 (5.4 ns) with the same thickness.28 Choi et al. concluded that polarization switching of un-doped HfO2 capacitors was slower in capacitors with larger thickness.85 Yoo et al. achieved polarization that was three times faster in Si:HfO2 capacitors by controlling the grain size to lower the coercive field.80 

There exist other complex switching behaviors for HfO2-based ferroelectrics, including NC and accumulative polarization switching. Accumulative polarization switching was first discovered in HfO2-based FeFET.25 As shown in Fig. 6(a), complete polarization switching occurred at certain number of pulses of a voltage train with a short pulse duration (1 μs) and at certain amplitude (V = 2.2 V). An all-or-none behavior was also confirmed by the threshold voltage, as is evident from Fig. 6(b). The authors attributed the accumulative polarization switching behavior to the successive nucleation and growth of opposite nanodomains, which finally led to complete polarization switching [stages 1–3 in Figs. 6(a) and 6(b)]. With the aid of careful adjustments, the authors found it possible to emulate neuronal behavior, just like biological neurons.24 

FIG. 6.

Accumulative switching and negative capacitance. (a) Drain current vs number of pulses of HfO2-based FeFET. (b) Threshold voltage vs number of pulses of HfO2-based FeFET. Circled numbers correspond to different stages of successive nucleation and the growth of opposite nanodomains. (a) and (b) Reproduced with permission from Mulaosmanovic et al., ACS Appl. Mater. Interfaces 10(28), 23997–24002 (2018). Copyright 2018 American Chemical Society. (c) Voltage transients of Gd:HfO2 capacitor in series with a resistor. Reproduced with permission from Hoffmann et al., Adv. Funct. Mater. 26(47), 8643–8649 (2016). Copyright 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (d) Voltage drops in the negative capacitance range vs Landau fitting coefficient during the wake-up process of the Si:HfO2 capacitor. Reproduced with permission from Park et al., Curr. Appl. Phys. 19(3), 347–350 (2019). Copyright 2019 Published by Elsevier B.V. on behalf of Korean Physical Society.

FIG. 6.

Accumulative switching and negative capacitance. (a) Drain current vs number of pulses of HfO2-based FeFET. (b) Threshold voltage vs number of pulses of HfO2-based FeFET. Circled numbers correspond to different stages of successive nucleation and the growth of opposite nanodomains. (a) and (b) Reproduced with permission from Mulaosmanovic et al., ACS Appl. Mater. Interfaces 10(28), 23997–24002 (2018). Copyright 2018 American Chemical Society. (c) Voltage transients of Gd:HfO2 capacitor in series with a resistor. Reproduced with permission from Hoffmann et al., Adv. Funct. Mater. 26(47), 8643–8649 (2016). Copyright 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (d) Voltage drops in the negative capacitance range vs Landau fitting coefficient during the wake-up process of the Si:HfO2 capacitor. Reproduced with permission from Park et al., Curr. Appl. Phys. 19(3), 347–350 (2019). Copyright 2019 Published by Elsevier B.V. on behalf of Korean Physical Society.

Close modal

NC, related to the negative dielectric loss,87 was directly observed during the polarization switching of polycrystalline HfO2 ferroelectrics with a significant potential to overcome the Boltzmann tyranny and subsequently reduce the power consumption of the device.23,88 For the NC effect, the ferroelectric is depolarized during polarization switching, exhibiting negative capacitance (defined as the derivation of the charge of the ferroelectric over the applied voltage) as shown in Fig. 6(c).23 It was recently reported that the NC effect is strongly associated with the wake-up effect, as shown in Fig. 6(d).86 The voltage range in which the NC effect exists was tuned by electric field cycling, indicating an alternative way to modulate the NC effect.86 

As discussed above, various crucial ferroelectric properties of HfO2-based ferroelectrics are based on the polarization switching. It is of great significance to understand the polarization switching dynamics of HfO2-based ferroelectrics. Considering the complex multi-phase characteristics of HfO2-based ferroelectrics, one can intuitively draw a conclusion that the switching properties may vary spatially. Chouprik et al. observed the non-uniform distribution of built-in fields in Zr:HfO2 capacitors.89 The non-uniform switching performance was also confirmed by Hyun et al. by assuming an inhomogeneous distribution of the applied electric fields.19 These studies indicated non-uniform polarization switching in different domains of HfO2-based ferroelectrics. However, researchers have found it feasible to apply models with and without considering the spatial variation of switching properties.

KAI model is a classical theory of polarization switching dynamics.8,10,90,91 In the KAI model, it was presumed that the applied electric field was distributed uniformly in the material, and polarization switching is determined by two mechanisms: (1) nucleation of new domains and (2) growth of new domains by domain wall movements. Mechanism (1) prevailed under small electric fields, while mechanism (2) dominated the switching process under large electric fields. The KAI model can appreciably characterize the polarization switching dynamics of ferroelectric single crystals and some thin films.8,91–93 However, in most non-ideal cases, such as polycrystalline materials, in which nucleation and domain wall movements occur simultaneously,94 the KAI model failed in characterizing the polarization switching dynamics.9,31–33 The non-ideal nature of materials favors models that take the statistical distribution of properties into account. Considering the fluctuation of local environments, Tagantsev et al.30 proposed nucleation-limited switching (NLS), presuming that the switching time of a material follows the statistical distribution p(t) = 1 − exp[−(t/t0)n], where p is the fraction of the switched volume by time t, and t0 and n are the parameters depending on the nucleus of the switched domains. With the NLS model, the polarization switching dynamics of ferroelectric polymers,95 organic ferroelectrics,96 and nanoscale capacitors9 were appreciably characterized. Although it is powerful, the NLS model does not consider the physical origin of the spatial variation of the switching time. Zhukov et al.31 proposed the inhomogeneous field mechanism (IFM) model, attributing the spatial variation of the switching time to a localized distribution of the effective external electric field. Subsequently, IFM has been widely applied for polycrystalline ferroelectrics,32 polymers,97 and thin films.98 

For HfO2-based ferroelectrics, multiple coexisting phases, and complex polarization switching behaviors, various models have been applied to characterize the polarization switching dynamics. Some researchers applied the KAI model for the characterization of polarization switching dynamics of HfO2-based ferroelectrics,85,99 while many researchers applied the NLS model18,27,35,100,101 and IFM model19,98 considering the fluctuation of the local environment for such multi-phase materials. Other researchers also used models based on the time-dependent Ginzburg–Landau approach22 and Monte Carlo simulations.102 All the models exhibited appreciable fitting results, providing researchers with multiple alternatives to investigate the polarization switching dynamics in HfO2-based ferroelectrics. When considering complex polarization switching processes, such as wake-up, which is strongly related to the evolution of polarization switching as discussed above, polarization switching dynamics can be more complex. Lee et al.84 and Hyun et al.19 investigated the wake-up effect in terms of switching dynamics for Si:HfO2 and Zr:HfO2 capacitors by using NLS and IFM models, respectively, as shown in Fig. 7. Lee et al.84 applied the NLS model103 based on the statistics of nucleation without considering the grain coalescence [Eq. (1)],
Δ P ( t ) 2 P S = [ 1 exp { ( t t 0 ) 2 } ] F ( log t 0 ) d ( log t 0 ) ,
(1)
where
F ( log t 0 ) = A π [ w ( log t 0 log t 1 ) 2 + w 2 ] .
(2)
FIG. 7.

Switching dynamics of Si-doped (left) and Zr-doped (right) HfO2 capacitors: normalized switched polarization of (a) pristine and (b) wake-up states; (c) distribution function of the switching times of different applied voltages with solid curves representing the pristine state and dashed curves representing the wake-up state. (a)–(c) Reproduced with permission from Lee et al., ACS Appl. Mater. Interfaces 11(3), 3142–3149 (2019). Copyright 2019 American Chemical Society. (d) Switched polarization with different numbers of electric field cycling; (e) fitting result using Eq. (3); and (f) fitted transient current of different applied voltages. (d)–(f) Reproduced with permission from Hyun et al., ACS Appl. Mater. Interfaces 10(41), 35374–35384 (2018). Copyright 2018 American Chemical Society.

FIG. 7.

Switching dynamics of Si-doped (left) and Zr-doped (right) HfO2 capacitors: normalized switched polarization of (a) pristine and (b) wake-up states; (c) distribution function of the switching times of different applied voltages with solid curves representing the pristine state and dashed curves representing the wake-up state. (a)–(c) Reproduced with permission from Lee et al., ACS Appl. Mater. Interfaces 11(3), 3142–3149 (2019). Copyright 2019 American Chemical Society. (d) Switched polarization with different numbers of electric field cycling; (e) fitting result using Eq. (3); and (f) fitted transient current of different applied voltages. (d)–(f) Reproduced with permission from Hyun et al., ACS Appl. Mater. Interfaces 10(41), 35374–35384 (2018). Copyright 2018 American Chemical Society.

Close modal

The fitting results were shown as solid curves in Figs. 7(a) and 7(b). The Lorentzian distributions of switching times under different biases in pristine (solid curve) and wake-up (dashed curve) states were shown in Fig. 7(c). The fitting results of Figs. 7(a) and 7(b) were in good accordance with the NLS model. After wake-up, the distributions of switching times were sharper and the values of t1 were more retarded, indicating a more uniform distribution of switching times and the annihilation of disorder (such as defects)-induced internal fields.

Despite the feasibility of the NLS model, Hyun et al.19 found that the NLS model failed to characterize the polarization switching dynamics of Zr:HfO2. Instead the IFM model was applied. The results shown in Fig. 7(d) were fitted with Eq. (3),
Δ P ( V , t ) / V Δ P ( V , t ) / V | max = 1 ξ 2 exp [ 1 1 ξ 2 γ ( 1 ξ ) 2 ξ 2 ] ,
(3)
where ξ = V / V d m and γ = 2 / ( 1 + 8 σ 2 1 ). After wake-up, Pr increased, as shown in Fig. 7(d), and the value of the fitting parameter σ decreased from 0.53 after 103 cycles to 0.32 after 105 cycles. This suggests a more uniform distribution of switching properties. It should be noted that both of the results are in accordance with the wake-up behavior of ferroelectrics with perovskite structures104 analyzed by FORC, in which the fatigue mechanism was attributed to the local phase decomposition.104 The similarity renders the possibility of occurrence of mechanisms similar to local phase decomposition in HfO2-based ferroelectrics worth investigating.

It should be noted that switching dynamics have also been studied by synchrotron XRD under different electrical biases as in the perovskite materials.59 However, it can be difficult for HfO2-based ferroelectrics due to the issues we discussed in Sec. III A. But potential application exists by using micro-spot XRD, which can be applied to study the microstructures of materials.

Possessing the significant advantages of compatibility with CMOS and thickness scalability, HfO2-based ferroelectrics exhibit considerable potential for applications in various devices. The structural origin of ferroelectricity (displacements of oxygen atoms) in HfO2-based ferroelectrics differs from perovskite ferroelectrics. This consequently results in the complex polarization switching behaviors in HfO2-based ferroelectrics. Electric-field-cycling-induced wake-up and split-up effects were observed, and the different split-up states were switchable between each other. Cycling endurance can be up to 1010 cycles. Negative capacitance in HfO2-based ferroelectrics was strongly associated with the wake-up process. Furthermore, the accumulative switching allows the fabrication of devices mimicking biological neurons. HfO2-based ferroelectrics also exhibit ultrafast switching speed, enhancing the operation speeds of the devices. Multiple theoretical models with and without considering the inhomogeneous distribution of polarization switching times were suggested and found to be suitable for the study of the polarization switching dynamics.

The state transition among different split-up states can be clearly seen in a Preisach plane. A Preisach plane of a specific material itself has potential in novel memory devices. Combined with the switchable properties of different split-up states of HfO2-based ferroelectrics, the corresponding Preisach plane exhibits great potential in the fabrication of novel high-density memory devices. Moreover, the relations between wake-up and negative capacitance have been investigated, but not widely. The relations hint at the numerous possibilities of tunable negative capacitance and accumulative switching behaviors, which can be useful in fabricating multi-functional devices. For the structure evolutions during wake-up and split-up, the microscopic mechanisms of which have not been clearly revealed, there is a potential to be distinctively characterized by micro-spot XRD due to the mature application of XRD in the electric cycling in perovskite samples. According to some experimental evidences, the mechanism of electric fatigue in HfO2-based ferroelectrics may also result from local phase decomposition. It is worth investigating whether local phase decomposition occurs in HfO2-based ferroelectrics, along with the elucidation of the phase to which it is decomposed. For polarization switching dynamics, homogeneous and inhomogeneous distributions of polarization switching times have been assumed. Apart from the differences induced by the preparation methods and the different dopants and compositions, it can be beneficial to consider the ultrafast switching nature down to nanoscale and take into account the structural complexity.

C. Wang and H. Qiao contributed equally to this article.

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (Nos. 2020R1F1A1072355 and 2019R1I1A1A01063888). This research was also supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. 2019R1A6A1A03033215).

The authors declare no competing financial interests.

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

1.
K. M.
Rabe
,
M.
Dawber
,
C.
Lichtensteiger
,
C. H.
Ahn
, and
J.-M.
Triscone
,
Physics of Ferroelectrics: A Modern Perspective
(
Springer
,
Berlin
,
2007
), pp.
1
30
.
2.
F.
Li
,
S.
Zhang
,
T.
Yang
,
Z.
Xu
,
N.
Zhang
,
G.
Liu
,
J.
Wang
,
J.
Wang
,
Z.
Cheng
,
Z.-G.
Ye
,
J.
Luo
,
T. R.
Shrout
, and
L.-Q.
Chen
,
Nat. Commun.
7
(
1
),
13807
(
2016
).
3.
M. H.
Park
,
Y. H.
Lee
,
H. J.
Kim
,
Y. J.
Kim
,
T.
Moon
,
K. D.
Kim
,
J.
Müller
,
A.
Kersch
,
U.
Schroeder
,
T.
Mikolajick
, and
C. S.
Hwang
,
Adv. Mater.
27
(
11
),
1811
1831
(
2015
).
4.
T. S.
Böscke
,
J.
Müller
,
D.
Bräuhaus
,
U.
Schröder
, and
U.
Böttger
,
Appl. Phys. Lett.
99
(
10
),
102903
(
2011
).
5.
S.
Clima
,
D. J.
Wouters
,
C.
Adelmann
,
T.
Schenk
,
U.
Schroeder
,
M.
Jurczak
, and
G.
Pourtois
,
Appl. Phys. Lett.
104
(
9
),
092906
(
2014
).
6.
J.
Schultheiß
,
L.
Liu
,
H.
Kungl
,
M.
Weber
,
L.
Kodumudi Venkataraman
,
S.
Checchia
,
D.
Damjanovic
,
J. E.
Daniels
, and
J.
Koruza
,
Acta Mater.
157
,
355
363
(
2018
).
7.
S. C.
Hwang
,
J. E.
Huber
,
R. M.
McMeeking
, and
N. A.
Fleck
,
J. Appl. Phys.
84
(
3
),
1530
1540
(
1998
).
8.
E.
Fatuzzo
and
W. J.
Merz
,
Phys. Rev.
116
(
1
),
61
68
(
1959
).
9.
Y.
Kim
,
H.
Han
,
W.
Lee
,
S.
Baik
,
D.
Hesse
, and
M.
Alexe
,
Nano Lett.
10
(
4
),
1266
1270
(
2010
).
10.
Y.
Kim
,
A.
Kumar
,
O.
Ovchinnikov
,
S.
Jesse
,
H.
Han
,
D.
Pantel
,
I.
Vrejoiu
,
W.
Lee
,
D.
Hesse
,
M.
Alexe
, and
S. V.
Kalinin
,
ACS Nano
6
(
1
),
491
500
(
2012
).
11.
E.
Parsonnet
,
Y.-L.
Huang
,
T.
Gosavi
,
A.
Qualls
,
D.
Nikonov
,
C.-C.
Lin
,
I.
Young
,
J.
Bokor
,
L. W.
Martin
, and
R.
Ramesh
,
Phys. Rev. Lett.
125
(
6
),
067601
(
2020
).
12.
J.
Jiang
,
Z. L.
Bai
,
Z. H.
Chen
,
L.
He
,
D. W.
Zhang
,
Q. H.
Zhang
,
J. A.
Shi
,
M. H.
Park
,
J. F.
Scott
,
C. S.
Hwang
, and
A. Q.
Jiang
,
Nat. Mater.
17
(
1
),
49
56
(
2018
).
13.
D.
Lee
,
B. C.
Jeon
,
S. H.
Baek
,
S. M.
Yang
,
Y. J.
Shin
,
T. H.
Kim
,
Y. S.
Kim
,
J.-G.
Yoon
,
C. B.
Eom
, and
T. W.
Noh
,
Adv. Mater.
24
(
48
),
6490
6495
(
2012
).
14.
J.
Glaum
,
Y. A.
Genenko
,
H.
Kungl
,
L.
Ana Schmitt
, and
T.
Granzow
,
J. Appl. Phys.
112
(
3
),
034103
(
2012
).
15.
D.
Zhou
,
J.
Xu
,
Q.
Li
,
Y.
Guan
,
F.
Cao
,
X.
Dong
,
J.
Müller
,
T.
Schenk
, and
U.
Schröder
,
Appl. Phys. Lett.
103
(
19
),
192904
(
2013
).
16.
E. D.
Grimley
,
T.
Schenk
,
X.
Sang
,
M.
Pešić
,
U.
Schroeder
,
T.
Mikolajick
, and
J. M.
LeBeau
,
Adv. Electron. Mater.
2
(
9
),
1600173
(
2016
).
17.
D.
Martin
,
J.
Müller
,
T.
Schenk
,
T. M.
Arruda
,
A.
Kumar
,
E.
Strelcov
,
E.
Yurchuk
,
S.
Müller
,
D.
Pohl
,
U.
Schröder
,
S. V.
Kalinin
, and
T.
Mikolajick
,
Adv. Mater.
26
(
48
),
8198
8202
(
2014
).
18.
N.
Gong
,
X.
Sun
,
H.
Jiang
,
K. S.
Chang-Liao
,
Q.
Xia
, and
T. P.
Ma
,
Appl. Phys. Lett.
112
(
26
),
262903
(
2018
).
19.
S. D.
Hyun
,
H. W.
Park
,
Y. J.
Kim
,
M. H.
Park
,
Y. H.
Lee
,
H. J.
Kim
,
Y. J.
Kwon
,
T.
Moon
,
K. D.
Kim
,
Y. B.
Lee
,
B. S.
Kim
, and
C. S.
Hwang
,
ACS Appl. Mater. Interfaces
10
(
41
),
35374
35384
(
2018
).
20.
T.
Schenk
,
M.
Hoffmann
,
J.
Ocker
,
M.
Pešić
,
T.
Mikolajick
, and
U.
Schroeder
,
ACS Appl. Mater. Interfaces
7
(
36
),
20224
20233
(
2015
).
21.
T.
Schenk
,
U.
Schroeder
,
M.
Pešić
,
M.
Popovici
,
Y. V.
Pershin
, and
T.
Mikolajick
,
ACS Appl. Mater. Interfaces
6
(
22
),
19744
19751
(
2014
).
22.
M.
Hoffmann
,
A. I.
Khan
,
C.
Serrao
,
Z.
Lu
,
S.
Salahuddin
,
M.
Pešić
,
S.
Slesazeck
,
U.
Schroeder
, and
T.
Mikolajick
,
J. Appl. Phys.
123
(
18
),
184101
(
2018
).
23.
M.
Hoffmann
,
M.
Pešić
,
K.
Chatterjee
,
A. I.
Khan
,
S.
Salahuddin
,
S.
Slesazeck
,
U.
Schroeder
, and
T.
Mikolajick
,
Adv. Funct. Mater.
26
(
47
),
8643
8649
(
2016
).
24.
H.
Mulaosmanovic
,
E.
Chicca
,
M.
Bertele
,
T.
Mikolajick
, and
S.
Slesazeck
,
Nanoscale
10
(
46
),
21755
21763
(
2018
).
25.
H.
Mulaosmanovic
,
T.
Mikolajick
, and
S.
Slesazeck
,
ACS Appl. Mater. Interfaces
10
(
28
),
23997
24002
(
2018
).
26.
R.
Jiang
,
X.
Du
, and
Z.
Han
,
J. Semicond.
37
(
8
),
084006
(
2016
).
27.
P.
Buragohain
,
C.
Richter
,
T.
Schenk
,
H.
Lu
,
T.
Mikolajick
,
U.
Schroeder
, and
A.
Gruverman
,
Appl. Phys. Lett.
112
(
22
),
222901
(
2018
).
28.
A.
Chouprik
,
M.
Spiridonov
,
S.
Zarubin
,
R.
Kirtaev
,
V.
Mikheev
,
Y.
Lebedinskii
,
S.
Zakharchenko
, and
D.
Negrov
,
ACS Appl. Electron. Mater.
1
(
3
),
275
287
(
2019
).
29.
A. K.
Saha
,
K.
Ni
,
S.
Dutta
,
S.
Datta
, and
S.
Gupta
,
Appl. Phys. Lett.
114
(
20
),
202903
(
2019
).
30.
A. K.
Tagantsev
,
I.
Stolichnov
,
N.
Setter
,
J. S.
Cross
, and
M.
Tsukada
,
Phys. Rev. B
66
(
21
),
214109
(
2002
).
31.
S.
Zhukov
,
S.
Fedosov
,
J.
Glaum
,
T.
Granzow
,
Y. A.
Genenko
, and
H.
von Seggern
,
J. Appl. Phys.
108
(
1
),
014105
(
2010
).
32.
S.
Zhukov
,
J.
Glaum
,
H.
Kungl
,
E.
Sapper
,
R.
Dittmer
,
Y. A.
Genenko
, and
H.
von Seggern
,
J. Appl. Phys.
120
(
6
),
064103
(
2016
).
33.
S.
Zhukov
,
Y. A.
Genenko
, and
H.
von Seggern
,
J. Appl. Phys.
108
(
1
),
014106
(
2010
).
34.
J.
Müller
,
T. S.
Böscke
,
U.
Schröder
,
S.
Mueller
,
D.
Bräuhaus
,
U.
Böttger
,
L.
Frey
, and
T.
Mikolajick
,
Nano Lett.
12
(
8
),
4318
4323
(
2012
).
35.
D.
Zhou
,
Y.
Guan
,
M. M.
Vopson
,
J.
Xu
,
H.
Liang
,
F.
Cao
,
X.
Dong
,
J.
Mueller
,
T.
Schenk
, and
U.
Schroeder
,
Acta Mater.
99
,
240
246
(
2015
).
36.
R.
Materlik
,
C.
Kunneth
, and
A.
Kersch
,
J. Appl. Phys.
117
(
13
),
134109
(
2015
).
37.
S.
Kim
,
J.
Ha
, and
J.-B.
Kim
,
Eur. Phys. J. Plus
131
(
11
),
382
(
2016
).
38.
Y.
Wei
,
P.
Nukala
,
M.
Salverda
,
S.
Matzen
,
H. J.
Zhao
,
J.
Momand
,
A. S.
Everhardt
,
G.
Agnus
,
G. R.
Blake
,
P.
Lecoeur
,
B. J.
Kooi
,
J.
Íñiguez
,
B.
Dkhil
, and
B.
Noheda
,
Nat. Mater.
17
(
12
),
1095
1100
(
2018
).
39.
T.
Shimizu
,
K.
Katayama
,
T.
Kiguchi
,
A.
Akama
,
T. J.
Konno
,
O.
Sakata
, and
H.
Funakubo
,
Sci. Rep.
6
(
1
),
32931
(
2016
).
40.
M. H.
Park
,
T.
Schenk
,
C. M.
Fancher
,
E. D.
Grimley
,
C.
Zhou
,
C.
Richter
,
J. M.
LeBeau
,
J. L.
Jones
,
T.
Mikolajick
, and
U.
Schroeder
,
J. Mater. Chem. C
5
(
19
),
4677
4690
(
2017
).
41.
P.
Polakowski
and
J.
Müller
,
Appl. Phys. Lett.
106
(
23
),
232905
(
2015
).
42.
M. H.
Park
,
H. J.
Kim
,
Y. J.
Kim
,
W.
Lee
,
T.
Moon
, and
C. S.
Hwang
,
Appl. Phys. Lett.
102
(
24
),
242905
(
2013
).
43.
T.
Mimura
,
T.
Shimizu
, and
H.
Funakubo
,
Appl. Phys. Lett.
115
(
3
),
032901
(
2019
).
44.
T.
Li
,
N.
Zhang
,
Z.
Sun
,
C.
Xie
,
M.
Ye
,
S.
Mazumdar
,
L.
Shu
,
Y.
Wang
,
D.
Wang
,
L.
Chen
,
S.
Ke
, and
H.
Huang
,
J. Mater. Chem. C
6
(
34
),
9224
9231
(
2018
).
45.
T.
Li
,
M.
Ye
,
Z.
Sun
,
N.
Zhang
,
W.
Zhang
,
S.
Inguva
,
C.
Xie
,
L.
Chen
,
Y.
Wang
,
S.
Ke
, and
H.
Huang
,
ACS Appl. Mater. Interfaces
11
(
4
),
4139
4144
(
2019
).
46.
Z.
Fan
,
J.
Chen
, and
J.
Wang
,
J. Adv. Dielectr.
06
(
2
),
1630003
(
2016
).
47.
M. D.
Glinchuk
,
A. N.
Morozovska
,
A.
Lukowiak
,
W.
Stręk
,
M. V.
Silibin
,
D. V.
Karpinsky
,
Y.
Kim
, and
S. V.
Kalinin
,
J. Alloys Compd.
830
,
153628
(
2020
).
48.
H.-J.
Lee
,
M.
Lee
,
K.
Lee
,
J.
Jo
,
H.
Yang
,
Y.
Kim
,
S. C.
Chae
,
U.
Waghmare
, and
J. H.
Lee
,
Science
369
(
6509
),
1343
(
2020
).
49.
B.
Noheda
and
J.
Íñiguez
,
Science
369
(
6509
),
1300
(
2020
).
50.
M.
Pešić
,
F. P. G.
Fengler
,
L.
Larcher
,
A.
Padovani
,
T.
Schenk
,
E. D.
Grimley
,
X.
Sang
,
J. M.
LeBeau
,
S.
Slesazeck
,
U.
Schroeder
, and
T.
Mikolajick
,
Adv. Funct. Mater.
26
(
25
),
4601
4612
(
2016
).
51.
H. J.
Kim
,
M. H.
Park
,
Y. J.
Kim
,
Y. H.
Lee
,
T.
Moon
,
K. D.
Kim
,
S. D.
Hyun
, and
C. S.
Hwang
,
Nanoscale
8
(
3
),
1383
1389
(
2016
).
52.
M.
Ghatge
,
G.
Walters
,
T.
Nishida
, and
R.
Tabrizian
,
Nat. Electron.
2
(
11
),
506
512
(
2019
).
53.
S. S.
Fields
,
S. W.
Smith
,
P. J.
Ryan
,
S. T.
Jaszewski
,
I. A.
Brummel
,
A.
Salanova
,
G.
Esteves
,
S. L.
Wolfley
,
M. D.
Henry
,
P. S.
Davids
, and
J. F.
Ihlefeld
,
ACS Appl. Mater. Interfaces
12
(
23
),
26577
26585
(
2020
).
54.
M.
Hoffmann
,
U.
Schroeder
,
T.
Schenk
,
T.
Shimizu
,
H.
Funakubo
,
O.
Sakata
,
D.
Pohl
,
M.
Drescher
,
C.
Adelmann
,
R.
Materlik
,
A.
Kersch
, and
T.
Mikolajick
,
J. Appl. Phys.
118
(
7
),
072006
(
2015
).
55.
I.
Urbanavičiūtė
,
T. D.
Cornelissen
,
X.
Meng
,
R. P.
Sijbesma
, and
M.
Kemerink
,
Nat. Commun.
9
(
1
),
4409
(
2018
).
56.
P. D.
Lomenzo
,
Q.
Takmeel
,
C.
Zhou
,
C. M.
Fancher
,
E.
Lambers
,
N. G.
Rudawski
,
J. L.
Jones
,
S.
Moghaddam
, and
T.
Nishida
,
J. Appl. Phys.
117
(
13
),
134105
(
2015
).
57.
Y.
Noguchi
,
I.
Miwa
,
Y.
Goshima
, and
M.
Miyayama
,
Jpn. J. Appl. Phys.
39
(
Part 2, No. 12B
),
L1259
L1262
(
2000
).
58.
L.
Liu
,
K.
Xu
,
Q.
Li
,
J.
Daniels
,
H.
Zhou
,
J.
Li
,
J.
Zhu
,
J.
Seidel
, and
J.-F.
Li
, “
Giant domain wall conductivity in self-assembled BiFeO3 nanocrystals
,”
Adv. Funct. Mater.
(published online) (
2020
).
59.
L.
Liu
,
T.
Rojac
,
J.
Kimpton
,
J.
Walker
,
M.
Makarovic
,
J.-F.
Li
, and
J.
Daniels
,
Appl. Phys. Lett.
116
(
12
),
122901
(
2020
).
60.
S.
Starschich
,
S.
Menzel
, and
U.
Böttger
,
Appl. Phys. Lett.
108
(
3
),
032903
(
2016
).
61.
D.
Seol
,
B.
Kim
, and
Y.
Kim
,
Curr. Appl. Phys.
17
(
5
),
661
674
(
2017
).
62.
J.
Lyu
,
I.
Fina
,
R.
Solanas
,
J.
Fontcuberta
, and
F.
Sánchez
,
Appl. Phys. Lett.
113
(
8
),
082902
(
2018
).
63.
J.
Lyu
,
I.
Fina
,
J.
Fontcuberta
, and
F.
Sánchez
,
ACS Appl. Mater. Interfaces
11
(
6
),
6224
6229
(
2019
).
64.
J.
Lyu
,
T.
Song
,
I.
Fina
, and
F.
Sánchez
,
Nanoscale
12
(
20
),
11280
11287
(
2020
).
65.
J.
Bouaziz
,
P. R.
Romeo
,
N.
Baboux
, and
B.
Vilquin
,
ACS Appl. Electron. Mater.
1
(
9
),
1740
1745
(
2019
).
66.
U.
Schroeder
,
E.
Yurchuk
,
J.
Müller
,
D.
Martin
,
T.
Schenk
,
P.
Polakowski
,
C.
Adelmann
,
M. I.
Popovici
,
S. V.
Kalinin
, and
T.
Mikolajick
,
Jpn. J. Appl. Phys.
53
,
08LE02
(
2014
).
67.
I. D.
Mayergoyz
and
C. E.
Korman
,
Math. Model. Nat. Phenom.
15
,
20
(
2020
).
68.
A. G.
Chernikova
,
M. G.
Kozodaev
,
D. V.
Negrov
,
E. V.
Korostylev
,
M. H.
Park
,
U.
Schroeder
,
C. S.
Hwang
, and
A. M.
Markeev
,
ACS Appl. Mater. Interfaces
10
(
3
),
2701
2708
(
2018
).
69.
C.
Yang
,
Y.
Han
,
J.
Qian
,
P.
Lv
,
X.
Lin
,
S.
Huang
, and
Z.
Cheng
,
ACS Appl. Mater. Interfaces
11
(
13
),
12647
12655
(
2019
).
70.
C.
Yang
,
P.
Lv
,
J.
Qian
,
Y.
Han
,
J.
Ouyang
,
X.
Lin
,
S.
Huang
, and
Z.
Cheng
,
Adv. Energy Mater.
9
(
18
),
1803949
(
2019
).
71.
S.
Starschich
,
S.
Menzel
, and
U.
Böttger
,
J. Appl. Phys.
121
(
15
),
154102
(
2017
).
72.
C.
Liu
,
F.
Liu
,
Q.
Luo
,
P.
Huang
,
X. X.
Xu
,
H. B.
Lv
,
Y. D.
Zhao
,
X. Y.
Liu
, and
J. F.
Kang
, in
Technical Digest—International Electron Devices Meeting, IEDM, December 2018
(IEEE, 2019), pp. 16.14.11–16.14.14.
73.
X.
Liu
,
D.
Zhou
,
Y.
Guan
,
S.
Li
,
F.
Cao
, and
X.
Dong
,
Acta Mater.
154
,
190
198
(
2018
).
74.
A. K.
Tagantsev
,
I.
Stolichnov
,
E. L.
Colla
, and
N.
Setter
,
J. Appl. Phys.
90
(
3
),
1387
1402
(
2001
).
75.
X. J.
Lou
,
J. Appl. Phys.
105
(
2
),
024101
(
2009
).
76.
X. J.
Lou
,
M.
Zhang
,
S. A. T.
Redfern
, and
J. F.
Scott
,
Phys. Rev. Lett.
97
(
17
),
177601
(
2006
).
77.
H.
Mulaosmanovic
,
E. T.
Breyer
,
T.
Mikolajick
, and
S.
Slesazeck
,
IEEE Electron Device Lett.
40
(
2
),
216
219
(
2019
).
78.
X.
Lyu
,
M.
Si
,
P. R.
Shrestha
,
K. P.
Cheung
, and
P. D.
Ye
, in Technical Digest—International Electron Devices Meeting, IEDM, December 2019 (IEEE, 2019), pp. 342–345.
79.
J.
Li
,
B.
Nagaraj
,
H.
Liang
,
W.
Cao
,
C. H.
Lee
, and
R.
Ramesh
,
Appl. Phys. Lett.
84
(
7
),
1174
1176
(
2004
).
80.
H. K.
Yoo
,
J. S.
Kim
,
Z.
Zhu
,
Y. S.
Choi
,
A.
Yoon
,
M. R.
MacDonald
,
X.
Lei
,
T. Y.
Lee
,
D.
Lee
,
S. C.
Chae
,
J.
Park
,
D.
Hemker
,
J. G.
Langan
,
Y.
Nishi
, and
S. J.
Hong
, in
2017 IEEE International Electron Devices Meeting (IEDM)
(IEEE, 2017), Vol. 5, pp. 19.16.11–19.16.14.
81.
W.
Chung
,
M.
Si
,
P. R.
Shrestha
,
J. P.
Campbell
,
K. P.
Cheung
, and
P. D.
Ye
, in
Digest of Technical Papers—Symposium on VLSI Technology
, June 2018 (IEEE, 2018), pp. 89–90.
82.
S.
Boyn
,
A.
Chanthbouala
,
S.
Girod
,
C.
Carrétéro
,
A.
Barthélémy
,
M.
Bibes
,
J.
Grollier
,
S.
Fusil
, and
V.
Garcia
,
Appl. Phys. Lett.
113
(
23
),
232902
(
2018
).
83.
X.
Lyu
,
M.
Si
,
P. R.
Shrestha
,
J. P.
Campbell
,
K. P.
Cheung
, and
P. D.
Ye
, in
2020 IEEE Silicon Nanoelectronics Workshop (SNW)
(IEEE, 2020), Vol. 2(3), pp. 7–8.
84.
T. Y.
Lee
,
K.
Lee
,
H. H.
Lim
,
M. S.
Song
,
S. M.
Yang
,
H. K.
Yoo
,
D. I.
Suh
,
Z.
Zhu
,
A.
Yoon
,
M. R.
Macdonald
,
X.
Lei
,
H. Y.
Jeong
,
D.
Lee
,
K.
Park
,
J.
Park
, and
S. C.
Chae
,
ACS Appl. Mater. Interfaces
11
(
3
),
3142
3149
(
2019
).
85.
S.-N.
Choi
,
S.-E.
Moon
, and
S.-M.
Yoon
,
Ceram. Int.
45
(
17
),
22642
22648
(
2019
).
86.
S.
Park
,
M. C.
Chun
,
S.
Park
,
G.
Park
,
M.
Jung
,
Y.
Noh
,
S.-E.
Ahn
, and
B. S.
Kang
,
Curr. Appl. Phys.
19
(
3
),
347
350
(
2019
).
87.
L.
Liu
,
T.
Rojac
,
D.
Damjanovic
,
M.
Di Michiel
, and
J.
Daniels
,
Nat. Commun.
9
(
1
),
4928
(
2018
).
88.
M.
Hoffmann
,
F. P. G.
Fengler
,
M.
Herzig
,
T.
Mittmann
,
B.
Max
,
U.
Schroeder
,
R.
Negrea
,
P.
Lucian
,
S.
Slesazeck
, and
T.
Mikolajick
,
Nature
565
(
7740
),
464
467
(
2019
).
89.
A.
Chouprik
,
S.
Zakharchenko
,
M.
Spiridonov
,
S.
Zarubin
,
A.
Chernikova
,
R.
Kirtaev
,
P.
Buragohain
,
A.
Gruverman
,
A.
Zenkevich
, and
D.
Negrov
,
ACS Appl. Mater. Interfaces
10
(
10
),
8818
8826
(
2018
).
90.
Y.
Ishibashi
,
Ferroelectrics
98
(
1
),
193
205
(
1989
).
91.
M.
Avrami
,
J. Chem. Phys.
8
(
2
),
212
224
(
1940
).
92.
D. C.
Lupascu
,
S.
Fedosov
,
C.
Verdier
,
J.
Rödel
, and
H.
von Seggern
,
J. Appl. Phys.
95
(
3
),
1386
1390
(
2004
).
93.
C.
Verdier
,
D. C.
Lupascu
,
H.
von Seggern
, and
J.
Rödel
,
Appl. Phys. Lett.
85
(
15
),
3211
3213
(
2004
).
94.
S. V.
Kalinin
,
A. N.
Morozovska
,
L. Q.
Chen
, and
B. J.
Rodriguez
,
Rep. Prog. Phys.
73
(
5
),
056502
(
2010
).
95.
D.
Mao
,
I.
Mejia
,
H.
Stiegler
,
B. E.
Gnade
, and
M. A.
Quevedo-Lopez
,
J. Appl. Phys.
108
(
9
),
094102
(
2010
).
96.
A. V.
Gorbunov
,
T.
Putzeys
,
I.
Urbanavičiūtė
,
R. A. J.
Janssen
,
M.
Wübbenhorst
,
R. P.
Sijbesma
, and
M.
Kemerink
,
Phys. Chem. Chem. Phys.
18
(
34
),
23663
23672
(
2016
).
97.
J.
Schütrumpf
,
S.
Zhukov
,
Y. A.
Genenko
, and
H.
von Seggern
,
J. Phys. D Appl. Phys.
45
(
16
),
165301
(
2012
).
98.
Y.
Li
,
J.
Li
,
R.
Liang
,
R.
Zhao
,
B.
Xiong
,
H.
Liu
,
H.
Tian
,
Y.
Yang
, and
T.-L.
Ren
,
Appl. Phys. Lett.
114
(
14
),
142902
(
2019
).
99.
S.-J.
Yoon
,
S.-Y.
Na
,
S.-E.
Moon
, and
S.-M.
Yoon
,
J. Vac. Sci. Technol. B
37
(
5
),
050601
(
2019
).
100.
S.
Mueller
,
S. R.
Summerfelt
,
J.
Müller
,
U.
Schroeder
, and
T.
Mikolajick
,
IEEE Electron Device Lett.
33
(
9
),
1300
1302
(
2012
).
101.
C.
Alessandri
,
P.
Pandey
,
A.
Abusleme
, and
A.
Seabaugh
,
IEEE Electron Device Lett.
39
(
11
),
1780
1783
(
2018
).
102.
C.
Alessandri
,
P.
Pandey
,
A.
Abusleme
, and
A.
Seabaugh
,
IEEE Trans. Electron Devices
66
(
8
),
3527
3534
(
2019
).
103.
J. Y.
Jo
,
H. S.
Han
,
J.-G.
Yoon
,
T. K.
Song
,
S.-H.
Kim
, and
T. W.
Noh
,
Phys. Rev. Lett.
99
(
26
),
267602
(
2007
).
104.
S. R.
Lee
,
L.
Baasandorj
,
J. W.
Chang
,
I. W.
Hwang
,
J. R.
Kim
,
J.-G.
Kim
,
K.-T.
Ko
,
S. B.
Shim
,
M. W.
Choi
,
M.
You
,
C.-H.
Yang
,
J.
Kim
, and
J.
Song
,
Nano Lett.
19
(
4
),
2243
2250
(
2019
).