Two-dimensional non-volatile valley spin valve

arXiv:2408.12028 · cond-mat.mtrl-sci, cond-mat.mes-hall · Submitted 2024-08-21 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Two-dimensional non-volatile valley spin valve".

Mira: Here's the gist of the paper. A nonvolatile valley spin valve (n-VSV) based on two-dimensional ferroelectric semiconductors is proposed to achieve a giant resistance change of up to a factor of 107 by utilizing the ferroelectric domain wall between two uniformly polarized domains in monolayer 1T’’ MoS2 Device Concept and Mechanism A spin valve represents a well-established device concept in magnetic memory technologies, whose functionality is determined by electron transmission being controlled by the relative alignment of magnetic moments of the two ferromagnetic layers.

Kai: The advent of valleytronics has conceptualized a valley spin valve (VSV) – a device that utilizes the valley degree of freedom and spin-valley locking to achieve a similar valve effect without relying on magnetism.

Mira: In this study, we propose a nonvolatile VSV (n-VSV) based on a two-dimensional (2D) ferroelectric semiconductor where the resistance of the n-VSV is controlled by the ferroelectric domain wall between the two uniformly polarized domains.

Kai: This giant VSV effect occurs due to transmission being strongly dependent on matching (mismatching) the valley-dependent spin polarizations in the two domains with the same (opposite) ferroelectric polarization orientations, when the chemical potential of 1T’’-MoS2 lies within the spin-split valleys.

Title and authors: Kai: Looking at the title and authors, we see Kai Huang, Kartik Samanta, Ding-Fu Shao, and Evgeny Tsymbal from Nebraska Center for Materials and Nanoscience. It's clear they're pulling together different expertise here.

Mira: They are bringing together solid-state physics concepts with condensed matter theory to model how this works on the material itself.

Lev: From an engineering standpoint, I always look at whether this kind of effect can actually be fabricated and measured reliably, and that's what we need to keep in mind.

Kai: Yeah, because if it's just a theoretical number without a clear path to making it work with real components like superconducting circuits or other hardware, it doesn't mean much to us in the lab.

Mira: The paper focuses on the 1T double-layer molybdenum disulfide material, which they emphasize is ferroelectric even down to a single layer <ref:2408.12028#pg1>.

Lev: That material choice is key because if we can't get that level of intrinsic ferroelectricity in a thin film, all this theory about domain walls won't translate into a working device.

Kai: Right, so the core idea is leveraging the specific properties of 1T double-layer MoS2 to achieve this non-volatile switching mechanism <ref:2408.12028#pg1>.

The paper's summary: Mira: So, the paper summarizes how they use density functional theory and quantum transport calculations to show that switching between a single domain state and two domains with opposite polarization separated by a domain wall can cause resistance changes as high as one hundred seven.

Kai: It’s this strong dependence on the matching or mismatching of spin states at the Fermi level, which is locked to the valley degree of freedom when the chemical potential is in those specific spin-split valleys.

Lev: The paper mentions that for this effect to be prominent, you need to dope it so that your Fermi level lands right inside those specific energy ranges, like the delta ECC or delta EVV.

Mira: So what this means practically is they've shown a mechanism where the domain wall itself acts as the switch, controlling electron transmission based on how well the spin states line up across that boundary.

Kai: It’s not just about changing magnetic orientation anymore; it’s about manipulating valley polarization using electric fields indirectly through the ferroelectric structure.

Lev: From a hardware perspective, that suggests a non-volatile memory element where you control the state via structural configuration rather than relying on external magnetic fields or high voltages.

The paper's improvements: Kai: The paper does touch on how they modeled the domain wall structure itself, showing two possible domain wall structures, designated as DW1 and DW2.

Mira: They used orthorhombic supercells for the polarized states and then allowed those regions to relax while keeping others fixed to find these stable structures.

Lev: I wonder about the computational cost here; modeling those complex structural relaxations across multiple configurations is a lot of work, especially when you're trying to predict how it will behave in real 2D materials <ref:2408.12028#pg1>.

Kai: Yeah, but they also confirmed that the shape and width of these domain walls stay largely stable during their calculations, which shows some structural robustness.

Mira: The polarization calculation for 1T double-layer MoS2 came out to be zero point two five picoCoulombs per meter along the z z axis, which is pretty consistent with earlier studies <ref:2408.12028#pg1>.

Lev: That consistency helps a lot because if the fundamental material properties like polarization are stable, then the switching mechanism itself should also be more predictable when we move toward fabrication.

Conclusion: Kai: So to wrap up, this paper on "Two-dimensional non-volatile valley spin valve" validates the concept of using 1T double-layer MoS2 as a core material for an nVSV <ref:2408.12028#pg1,Two-dimensional non-volatile valley spin valve>.

Mira: They've shown that exploiting the intrinsic ferroelectric properties of this material allows for non-volatile control over valley-dependent spin polarization and electron transmission without needing an external electric field.

Lev: It’s a high resistance ratio of up to one hundred seven which is what makes this result significant in terms of the potential switching capability.

Kai: It validates the idea of non-volatile valleytronics by showing a path to achieve that level of resistance control using domain wall scattering.

Mira: It opens up the possibility of controlling spin states based on electric polarization configurations within these 2D materials <ref:2408.12028#pg1>.

Lev: We'll keep watching how this translates into actual hardware, but this is a solid foundation for understanding what's possible here with valleytronics.

Kai Huang, Kartik Samanta, Ding-Fu Shao, Evgeny Y. Tsymbal

Department of Physics and Astronomy and Nebraska Center for Materials and Nanoscience, University of Nebraska · Key Laboratory of Materials Physics, Institute of Solid-State Physics, Chinese Academy of Sciences

cond-mat.mtrl-sci, cond-mat.mes-hall

Submitted: 2024-08-21

Updated: 2024-08-21

Comments: 6 pages, 4 figures

Journal ref: ACS Nano 19, 3448-3454 (2025)

DOI: 10.1021/acsnano.4c12812

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 79/100

The gist: A nonvolatile valley spin valve (n-VSV) based on two-dimensional ferroelectric semiconductors is proposed to achieve a giant resistance change of up to a factor of 107 by utilizing the ferroelectric…

Key concepts

Spin Valve
A device where electron transmission is controlled by the relative alignment of magnetic moments in two ferromagnetic layers. In this context, it's adapted to use valley degree of freedom instead of traditional magnetism.
Valley Spin Valve (VSV)
A conceptual device that uses the spin-valley locking property to achieve a switching effect without relying on magnetic materials. It exploits the difference in spin polarization between different valleys in a material.
Ferroelectric Domain Wall (DW)
A structural boundary separating two regions of 1T'' MoS2 with opposite electric polarizations. The resistance switching is controlled by how electrons interact with this specific interface, which dictates the ON/OFF ratio.
Giant Resistance Change (Factor of 107)
The extremely large ratio between the high-conductance and low-conductance states of the device. This massive effect arises from strong coupling between spin polarization and electric polarization at the domain wall, enabling nonvolatile switching.

Terminology

Summary

A nonvolatile valley spin valve (n-VSV) based on two-dimensional ferroelectric semiconductors is proposed to achieve a giant resistance change of up to a factor of 107 by utilizing the ferroelectric domain wall between two uniformly polarized domains in monolayer 1T’’ MoS2

Device Concept and Mechanism

A spin valve represents a well-established device concept in magnetic memory technologies, whose functionality is determined by electron transmission being controlled by the relative alignment of magnetic moments of the two ferromagnetic layers. The advent of valleytronics has conceptualized a valley spin valve (VSV) – a device that utilizes the valley degree of freedom and spin-valley locking to achieve a similar valve effect without relying on magnetism. p. 0

Where the paper says it

Tsymbal1,* 1 Department of Physics and Astronomy & Nebraska Center for Materials and Nanoscience, University of Nebraska, Lincoln, Nebraska 68588-0299, USA 2 Key Laboratory of Materials Physics, Institute of Solid-State Physics, HFIPS, Chinese Academy of Sciences, Hefei 230031, China Abstract: A spin valve represents a well-established device concept in magnetic memory technologies, whose functionality is determined by electron transmission being controlled by the relative alignment of magnetic moments of the two ferromagnetic layers. Recently, the advent of valleytronics has conceptualized a valley spin valve (VSV) – a device that utilizes the valley degree of freedom and spin-valley locking to achieve a similar valve effect without relying on magnetism.

In this study, we propose a nonvolatile VSV (n-VSV) based on a two-dimensional (2D) ferroelectric semiconductor where the resistance of the n-VSV is controlled by the ferroelectric domain wall between the two uniformly polarized domains. p. 0
Where the paper says it

Recently, the advent of valleytronics has conceptualized a valley spin valve (VSV) – a device that utilizes the valley degree of freedom and spin-valley locking to achieve a similar valve effect without relying on magnetism. In this study, we propose a nonvolatile VSV (n-VSV) based on a two-dimensional (2D) ferroelectric semiconductor where the resistance of the n-VSV is controlled by the ferroelectric domain wall between the two uniformly polarized domains.

This giant VSV effect occurs due to transmission being strongly dependent on matching (mismatching) the valley-dependent spin polarizations in the two domains with the same (opposite) ferroelectric polarization orientations, when the chemical potential of 1T’’-MoS2 lies within the spin-split valleys. p. 0
Where the paper says it

Focusing on the 1T’’ phase of MoS2, which is known to be ferroelectric down to a monolayer and using density functional theory (DFT) combined with the quantumtransport calculations, we demonstrate that switching between the uniformly polarized state and the state with oppositely polarized domains separated by a domain wall resultsin resistance change of as high as 107 . This giant VSV effect occurs due to transmission being strongly dependent on matching (mismatching) the valley-dependent spin polarizations in the two domains with the same (opposite) ferroelectric polarization orientations, when the chemical potential of 1T’’-MoS2 lies within the spin-split valleys.

Material Selection and Properties

Focusing on the 1T’’ phase of MoS2, which is known to be ferroelectric down to a monolayer and using density functional theory (DFT) combined with the quantumtransport calculations, we demonstrate that switching between the uniformly polarized state and the state with oppositely polarized domains separated by a domain wall results in resistance change of as high as 107. The 1T’’ phase preserves ferroelectricity down to a monolayer limit and hosts valleys at ±K points. p. 1

Where the paper says it

Recently, a ferroelectric 1T’’ phase has been confirmed experimentally to occur in MoS2 grown under appropriate conditions [28- 32]. The 1T’’ phase preserves ferroelectricity down to a monolayer limit and hosts valleys at ±𝐾𝐾 points [29,30].

The unit cell of 1T’’ MoS2 contains 4 Mo atoms and 8 S atoms and belongs to the polar space group P3m1, indicating its ferroelectric nature. p. 1
Where the paper says it

Hexagonal Brillouin zone and high symmetry points are shown at bottom left. The unit cell of 1T’’ MoS2 contains 4 Mo atoms and 8 S atoms and belongs to the polar space group 𝑃𝑃3𝑚𝑚1, indicating its ferroelectric nature.

The polarization (dipole moment per area) of the 2D material is calculated to be 0. p. 2
Where the paper says it

44 Å in previous reports. [28,43] The polarization (dipole moment per area) of the 2D material is calculated to be 0.

25 pC/m along zz-axis, aligning closely the value 0. p. 2
Where the paper says it

[28,43] The polarization (dipole moment per area) of the 2D material is calculated to be 0. 25 pC/m along 𝑧𝑧-axis, aligning closely the value 0.

24 pC/m with past studies. p. 2
Where the paper says it

25 pC/m along 𝑧𝑧-axis, aligning closely the value 0. 24 pC/m with past studies.

Modeling the Domain Wall Structure

To model this, we combine the orthorhombic supercells of P↑ and P↓ states 1T’’ MoS2 with lattice vectors of these supercells were oriented along the [210] and [010] directions. During the structural relaxation process, the regions numbered 5 through 8 and 17 through 20 (Fig. p. 2

Where the paper says it

For the calculation, three 𝑃𝑃 ↑ orthorhombic supercells and three 𝑃𝑃 ↓ orthorhombic supercells are combined. During the structural relaxation process, the regions numbered 5 through 8 and 17 through 20 (Fig.

3(b)) were kept fixed in the middle of both P↑ and P↓ domains, while other areas were allowed to relax to identify stable DW structures. p. 2
Where the paper says it

During the structural relaxation process, the regions numbered 5 through 8 and 17 through 20 (Fig. 3(b)) were kept fixed in the middle of both 𝑃𝑃 ↑ and 𝑃𝑃 ↓ domains, while other areas were allowed to relax to identify stable DW structures.

This process reveals two possible DWs, designated by DW1 and DW2. p. 2
Where the paper says it

3(b)) were kept fixed in the middle of both 𝑃𝑃 ↑ and 𝑃𝑃 ↓ domains, while other areas were allowed to relax to identify stable DW structures. This process reveals two possible DWs, designated by DW1 and DW2.

The displacement increases abruptly across the DWs. p. 2
Where the paper says it

Using the polarization displacement 𝑑𝑑̅ 𝑧𝑧 S − 𝑑𝑑̅ 𝑧𝑧 Mo shown in Fig. 3(b), we observed that the displacement increases abruptly across the DWs.

Conductance Switching Performance

The ON/OFF ratio is controlled by the valleydependent spin matching in momentum space, which is strongly influenced by the choice of transport direction. p. 3

Where the paper says it

Our results indicated qualitatively very similar behavior (Fig. We note that the ON/OFF ratio is controlled by the valleydependent spin matching in momentum space, which is

In the energy range of ∆ECC and ∆EVV, the n-VSV exhibits a giant resistive switching effect with η being as high as 107 above the bandgap and 106 below the bandgap. p. 3
Where the paper says it

4(e) shows the ON/OFF ratio, 𝜂𝜂 = (𝐺𝐺UP − 𝐺𝐺DW)/ 𝐺𝐺DW, where 𝐺𝐺UP and 𝐺𝐺DW are conductance for the UP (ON) and DW (OFF) states, respectively. It is seen that in the energy range of ∆𝐸𝐸𝐶𝐶 and ∆𝐸𝐸𝑉𝑉 , the n-VSV exhibits a giant resistive switching effect with 𝜂𝜂 being as high as ~107 above the bandgap and ~106 below the bandgap.

This huge effect reflects the significant change in conductance between UP and DW states (Fig. p. 3
Where the paper says it

It is seen that in the energy range of ∆𝐸𝐸𝐶𝐶 and ∆𝐸𝐸𝑉𝑉 , the n-VSV exhibits a giant resistive switching effect with 𝜂𝜂 being as high as ~107 above the bandgap and ~106 below the bandgap. This huge effect reflects the significant change in conductance between UP and DW states (Fig.

4(d)) due to locking between the spin and electric polarization and represents the key property of an n-VSV. p. 3
Where the paper says it

This huge effect reflects the significant change in conductance between UP and DW states (Fig. 4(d)) due to locking between the spin and electric polarization and represents the key property of an n-VSV.

The transport properties of an n-VSV are expected to depend on the DW width between the two domains with opposite polarizations. p. 3
Where the paper says it

(e) ON/OFF ratio as a function of 𝐸𝐸F. The transport properties of an n-VSV are expected to depend on the DW width between the two domains with opposite polarizations.

Critical Parameters for Effect

The Fermi energy is the critical parameter that controls the VSV effect. Doping is required to shift the Fermi level to ∆ECC or ∆EVV for the giant VSV effect to occur. p. 4

Where the paper says it

The Fermi energy is the critical parameter that controls the VSV effect. Doping is required to shift the Fermi level to ∆𝐸𝐸𝐶𝐶 or ∆𝐸𝐸𝑉𝑉 for the giant VSV effect to occur.

The ON/OFF ratio in this device is then purely due to the domain wall scattering. p. 4
Where the paper says it

In this case, the valleys at ±𝐾𝐾 overlap at the same 𝑘𝑘∥, and hence momentum-dependent spin polarizations for these valleys cancel out. The ON/OFF ratio in this device is then purely due to the domain wall scattering.

The VSV effect is expected to be negligible in a 1T’’ MoS2-based VSV if the transport is along the [010] direction. p. 4
Where the paper says it

5 strongly influenced by the choice of transport direction. For example, the VSV effect is expected to be negligible in a 1T’’ MoS2-based VSV if the transport is along the [010] direction.

Stability and Conclusion

Our calculations indicated that the shape and width of the DW remained largely unchanged, showing its structural stability. The ON/OFF ratio is controlled by the valleydependent spin matching in momentum space, which is strongly influenced by the choice of transport direction. Overall, our results demonstrate the viability of using 1T’’ MoS2 as the core material for an n-VSV. p. 4

Where the paper says it

This may be achieved by adding impurities to the bulk material or by a charge transfer from the substrate or capping layer. Overall, our results demonstrate the viability of using 1T’’ MoS2 as the core material for an n-VSV.

By exploiting the intrinsic ferroelectric properties of this 2D vdW material, we demonstrate non-volatile control over valley-dependent spin polarization and electron transmission without the need for an external electric field. p. 4
Where the paper says it

Overall, our results demonstrate the viability of using 1T’’ MoS2 as the core material for an n-VSV. By exploiting the intrinsic ferroelectric properties of this 2D vdW material, we demonstrate non-volatile control over valley-dependent spin polarization and electron transmission without the need for an external electric field.

The resulting nVSV device shows a remarkably high resistance ratio between the ON and OFF states, reaching up to 107, thus validating the concept of nonvolatile valleytronics. p. 4
Where the paper says it

By exploiting the intrinsic ferroelectric properties of this 2D vdW material, we demonstrate non-volatile control over valley-dependent spin polarization and electron transmission without the need for an external electric field. The resulting nVSV device shows a remarkably high resistance ratio between the ON and OFF states, reaching up to 107 , thus validating the concept of nonvolatile valleytronics.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Develop AI systems capable of predicting giant magnetoresistance switching effects in novel materials by incorporating density functional theory calculations combined with quantum transport calculations to model valley spin valve behavior.

  2. Create AI models that can predict the ON/OFF ratio of nonvolatile valley spin valves based on domain wall width and Fermi energy, allowing for the selection of optimal doping levels for achieving high resistance ratios up to 107.

  3. Implement machine learning algorithms trained on DFT results to screen 2D ferroic materials, specifically focusing on phases like 1T’’ MoS2, that exhibit both ferroelectricity and valleys at non-TRIM points for n-VSV realization.

  4. Design AI systems that can analyze transport direction dependence of the VSV effect, identifying cases where the effect is negligible due to valley spin polarization cancellation along specific crystal axes, such as the [010] direction.

Abstract

A spin valve represents a well-established device concept in magnetic memory technologies, whose functionality is determined by electron transmission being controlled by the relative alignment of magnetic moments of the two ferromagnetic layers. Recently, the advent of valleytronics has conceptualized a valley spin valve (VSV) - a device that utilizes the valley degree of freedom and spin-valley locking to achieve a similar valve effect without relying on magnetism. In this study, we propose a non-volatile VSV (n-VSV) based on a two-dimensional (2D) ferroelectric semiconductor where the resistance of the n-VSV is controlled by the ferroelectric domain wall between the two uniformly polarized domains. Focusing on the 1T'' phase of MoS2, which is known to be ferroelectric down to a monolayer and using density functional theory (DFT) combined with the quantum-transport calculations, we demonstrate that switching between the uniformly polarized state and the state with oppositely polarized domains separated by a domain wall results in resistance change of as high as 10 7. This giant VSV effect occurs due to transmission being strongly dependent on matching (mismatching) the valley-dependent spin polarizations in the two domains with the same (opposite) ferroelectric polarization orientations, when the chemical potential of 1T''-MoS2 lies within the spin-split valleys. Our work paves a new route for realizing high-performance nonvolatile valleytronics.

Related papers