Is it possible to create magnetic semiconductors that function at room temperature?

  Is it possible to create magnetic semiconductors that function at room temperature?

As electronic devices continue to shrink to just a few nanometers, enabling the integration of billions of devices in computers, power consumption has surged beyond control, exceeding thermal limits and leading to failure. To address this issue, researchers are exploring an alternative to conventional electronics by harnessing the property of “spin” rather than “charge.” This emerging field, known as spintronics, compared to its counterpart electronics, holds the promise of more energy-efficient quantum computing and data storage, and beyond. Recent successes with van der Waals (vdW)–layered diluted magnetic chalcogenide semiconductors (dMCSs) are promising and have sparked a vibrant research community. This field is still in its infancy on the path to realizing the ultimate spintronics. But it is highly likely that the remaining scientific challenges can be overcome within the next decade or so.

In spintronics, the predominant approach thus far has involved current-injection devices, which can lead to high power consumption. However, a more sustainable solution may lie in the manipulation of magnetic order through electrical gating in magnetic semiconductors. This method allows for the voltage-based control of spins with minimal charge modulation, enabling the engineering of magnetic order below the Curie temperature (_T_C). Above _T_C, the switching effect diminishes, highlighting the potential for sustainable power reduction.

Scientists have attempted to realize _T_C above room temperature in various semiconductors but encountered considerable challenges. This difficulty arises because the available pool of materials capable of achieving _T_C values near room temperature while maintaining gating tunability is limited. As a potential solution to this issue, researchers have turned to magnetic semiconductors, in which magnetism is introduced through the incorporation of magnetic dopants such as Mn, Co, Ni, and others. For instance, Mn doping has been applied to semiconductors such as GaAs, as well as certain nitrides and oxides, leading to the creation of what is known as diluted magnetic semiconductors.

Unfortunately, achieving both _T_C at or above room temperature together with gate tunability has remained elusive in the realm of magnetic semiconductors. This enduring challenge has long been recognized as one of the unsolved questions in the scientific community, encapsulated in the query: Is it possible to create magnetic semiconductors that function at room temperature? This question has been prominently featured among the 125 unsolved questions in the field of science, as noted in 2005.

The foundation of diluted magnetic semiconductors relies on several key prerequisites, including a _T_C above room temperature, gate tunability, absence of dopant segregation in semiconductors, and the establishment of long-range magnetic order. Although the isolated demonstrations have showcased one or a few of these essential ingredients, the goal of realizing all these components within bulk semiconductors remains unattained thus far.

Recently, vdW-layered magnetic semiconductors have emerged as promising candidates in the pursuit of meeting the aforementioned requirements for spintronics. A multitude of vdW-layered ferromagnetic semiconductors have been explored, including notable examples such as CrI3, Cr2Si2Te6, and others. However, despite these promising developments, the challenge of achieving room-temperature _T_C remains unmet. Although researchers have made substantial progress in identifying vdW ferromagnetic semiconductors with reasonably elevated _T_C values, their choices are still constrained by the limited selection available in the materials library.

Inspired by the lessons learned from diluted magnetic semiconductors in bulk materials and fueled by the emergence of vdW-layered dichalcogenides, researchers have embarked on a transformative journey. This journey involves the introduction of magnetic dopant into a two-dimensional (2D) vdW-layered chalcogenide semiconductor, giving rise to a new category known as diluted magnetic chalcogenide semiconductors.

One distinctive feature of 2D dMCSs is their pronounced spin-orbit coupling (SOC), a consequence of the incorporation of heavy-mass atoms found in transition metals such as Mo and W. This is further interlinked with magnetocrystalline anisotropy—the difference in magnetic moment depending on crystallographic direction—which is inherently implemented in vdW-layered materials and hence is key to strengthening long-range magnetic order. Such distinctive combinations of strong SOC, magnetocrystalline anisotropy, and strong Coulomb interaction stemming from ultrathin 2D vdW-layered dMCSs hold the potential to enhance long-range magnetic order and pave the way for achieving room-temperature _T_C. These advancements in dMCSs represent an exciting frontier in the realm of spintronics, offering a multitude of opportunities for transformative breakthroughs. Understanding of the combined complexity phenomena could be resolved theoretically and experimentally by engineering heavy-mass dopants having different SOC strengths. Room-temperature _T_C and long-range magnetic order can be realized by introducing magnetic proximity at heterostructures where interfacial interaction is inherently strong at vdW-layered materials.

For these reasons, vdW-layered dMCSs have undergone thorough investigation. Various systems involving vdW-layered semiconductors combined with magnetic dopants, such as WSe2, MoS2, or GaTe, with elements like V, Cr, Fe, Co, or Ni, have been explored. Notably, consensus seems to have emerged in the pursuit of manifesting _T_C near or above room temperature, particularly in materials such as V-doped WSe2, Fe-doped MoS2, or Cr-doped GaTe, although gate tunability has yet to be demonstrated.

The interesting scientific question is why vdW dMCSs have shown success in achieving _T_C values near room temperature, in contrast to their bulk diluted magnetic semiconductor counterparts. The answer may lie in the interplay of SOC, magnetocrystalline anisotropy, exchange interaction, and Coulomb interaction, leading to long-range magnetic order and ultimately facilitating the attainment of high _T_C values and gate tunability. In this sense, it is fortunate that such key features can be incorporated within the framework of vdW-layered dMCSs. However, the underlying mechanisms remain somewhat hypothetical at this stage. One way to clarify this issue is to use strong SOC dMCS such as WSe2 on proximitized yttrium-iron garnet (YIG), a magnetic insulator with a _T_C above room temperature. Such versatility is a remarkable feature of the abundant presence of vdW-layered semiconductor materials, offering a promising avenue for further investigation.

Despite noteworthy achievements in the realm of vdW-layered dMCSs, challenges still persist in accurately measuring _T_C. Many measurement techniques struggle to provide clear evidence of _T_C, primarily owing to the limited sensitivity in detecting magnetic order. This limitation is exacerbated by the inherently low doping concentration of magnetic dopants, typically a few percent or even smaller. Such low concentrations are necessitated by the propensity for dopants to segregate at higher doping levels, a phenomenon that hinders the formation of an appreciable net magnetic moment.

An illustrative example of these challenges is found in the case of V dopants introduced into WSe2. These dopants are distributed randomly throughout the entire WSe2 lattice, achieving concentrations as high as 10% without inducing segregation, a notable contrast to the behavior of counterparts such as Mn-doped GaAs, which exhibits segregation at lower doping concentrations. However, despite the favorable distribution, the magnetic moment in V-doped WSe2 remains modest, measuring only 1 Bohr magneton, which is smaller than that of typical magnetic dopants. This limitation is somewhat mitigated by the material’s ability to exhibit a _T_C near room temperature.

To address these challenges and enhance magnetic properties, researchers are exploring alternative doping strategies. One solution could be introduction of co-doping with different dopants such as V atoms to maintain a high _T_C and Gd atoms to confer a strong magnetic moment. Another solution is to take advantage of, for example, metal defects or chalcogen vacancies, edges, or grain boundary defects in colloidal metal chalcogenides. The synergistic combination of these dopants and defects aims to overcome the limitations of weak magnetic moments while preserving elevated _T_C values.

The next pivotal challenge revolves around the ability of materials with room-temperature _T_C to exhibit gate tunability. The literature offers limited data demonstrating the modulation of magnetic order through gating. Some notable attempts have been made, such as the modulation of magnetic phase shifts or magnetic domains using a Cr/Au tip in magnetic force microscopy (MFM). However, such gating effects are often localized by the MFM tip, influencing local magnetic order rather than the entire sample. Although magnetoresistance hysteresis is well documented and observed with external magnetic fields, more rigorous measurement techniques are required. For instance, it is essential to directly measure the magnetic transition curve or magnetic moment under a specific gate bias. Alternatively, the modulation of source-drain current (or resistance) should be examined for a given magnetic field, akin to the magnetoresistance hysteresis. Demonstrating gate tunability for materials with _T_C values near room temperature is paramount. One noticeable gap in the current research landscape is the absence of magnetic switching in field-effect transistor devices. To advance the field of vdW-layered dMCSs toward the realization of ultimate spintronics, leading researchers should prioritize demonstrating gate tunability in these devices, as it represents a critical step forward.

Another key criterion for achieving gate tunability lies in the closeness of the dopant energy levels to the Fermi level. This similarity allows for effective modulation of magnetic order. For instance, the energy level of V atoms is conveniently situated near the Fermi level in V (hole)–doped WSe2, as observed in prior research. However, the energy levels of other magnetic dopants, such as Fe, Co, and Ni, in various semiconductors remain unanswered. Theoretical understanding of the energy levels of these dopants is essential to ensure further applications before demonstrating gate tunability by experiment.

When the doping concentration is exceptionally low—as little as 0.1% in specific cases—a distinctive phenomenon known as random telegraph noise emerges in the magneto-tunneling junction devices. This noise gives rise to considerable resistance variance, often reaching levels as high as 80% due to interlayer coupling between layers. These fluctuations stem from the dynamic behavior of spins in multilayered V-doped WSe2, exhibiting similarities to the phenomenon of giant magnetoresistance. In two-terminal devices, the distinctive characteristics of high-resistance and low-resistance signals, representing the two states of spin alignment (parallel and antiparallel), can be switched by changing the voltage polarity. This intriguing behavior holds promise and may serve as a catalyst for future advancements in spintronics.

One remaining challenge is determining the optimal doping concentration to achieve _T_C values near room temperature. An excessively low doping concentration results in a weak magnetic moment or magnetic noise, often leading to the prevalence of random telegraph noise. Conversely, an excessively high doping concentration can trigger the aggregation of magnetic dopants and may lead to degenerate semiconductors, hindering gate tunability.

One viable solution is to use a moderate doping concentration, typically ranging from 0.2 to 2%. This range ensures the presence of high magnetic moments, prevents dopant aggregation, and facilitates gate tunability. However, the precise doping concentration may vary slightly from one material to another. Under such conditions, random telegraph noise may persist even at room temperature, manifesting as two distinct high- and low-resistance signals that correspond to parallel and antiparallel spin states, and magnetic order vanishes at high temperatures above _T_C. This phenomenon arises from the interplay between interlayer and intralayer coupling, warranting further investigation.

One last challenge on the horizon is the availability of single-crystal vdW dMCSs on a large scale for seamless integration. Bulk transition-metal dichalcogenides can be synthesized through chemical vapor transport or flux methods, but they typically yield samples limited to a few millimeters in size, hindering facile exfoliation and transfer to larger substrates. The overarching question is whether wafer-scale films can be synthesized or single-crystal multilayer films grown directly on the desired substrate. Notably, researchers have achieved single-crystal transition-metal dichalcogenide monolayers at a wafer scale using chemical vapor deposition with an atomic sawtooth substrate. The next frontier is to push these boundaries further, aiming for the growth of single-crystal multilayer films at a wafer scale or the direct synthesis of single-crystal films on designated substrates. This could be achieved by once again incorporating an atomic sawtooth substrate into prepatterned substrates for ultimate integration.