Enhancing electric-field control of ferromagnetism through nanoscale engineering of high-Tc MnxGe1−x nanomesh


Voltage control of magnetism in ferromagnetic semiconductor has emerged as an appealing solution to significantly reduce the power dissipation and variability beyond current CMOS technology. However, it has been proven to be very challenging to achieve a candidate with high Curie temperature (Tc), controllable ferromagnetism and easy integration with current Si technology. Here we report the effective electric-field control of both ferromagnetism and magnetoresistance in unique MnxGe1−x nanomeshes fabricated by nanosphere lithography, in which a Tc above 400 K is demonstrated as a result of size/quantum confinement. Furthermore, by adjusting Mn doping concentration, extremely giant magnetoresistance is realized from 8,000% at 30 K to 75% at 300 K at 4 T, which arises from a geometrically enhanced magnetoresistance effect of the unique mesh structure. Our results may provide a paradigm for fundamentally understanding the high Tc in ferromagnetic semiconductor nanostructure and realizing electric-field control of magnetoresistance for future spintronic applications.


For low power dissipation and low variability in the continuous scaling of complementary metal-oxide-semiconductor (CMOS), the search for novel materials and devices has been highlighted in the International Technology Roadmap of Semiconductors1,2. Spintronics, which utilizes the spin of electrons as another d.f. for information processing, offers a promising pathway to meet this challenge3. Recently, the use of spin-polarized current to manipulate magnetism2,4 has drawn great attentions for low-power spintronics devices. Further significant reduction of switching power can be envisaged by using an electric field rather than current. With carrier-mediated ferromagnetism, ferromagnetic semiconductors (FMS)5,6 can open up great opportunities in realizing voltage controlled non-volatile and low-power spintronic devices, such as magnetoresistance (MR)-type memories7, magnetic-field sensors8 and spinFETs9,10. Among various FMS materials, MnxGe1−x has attracted significant attentions due to its compatibility with the mature Si technology and its potentially high Curie temperature (Tc)11,12. In practice, however, the realization of high-quality and high-Tc MnxGe1−x thin films with electric-field-controlled ferromagnetism has been proven to be difficult due to the extremely low Mn solubility in Ge. Intermetallic compounds (such as Mn5Ge3 and Mn11Ge8) and Mn-rich nanophases8 are usually formed in a random and uncontrollable manner12,13,14. Although great efforts have been devoted to avoid the formation of these compounds, completely eliminating them remains elusive. Therefore, the quest for spintronic devices to work at ambient temperature calls for a new material candidate or structure, in which a high Tc can be attained with controllable ferromagnetism and easy integration with current CMOS platform.

For seeking such new candidates, the pioneer work on one-dimensional (1D) MnxGe1−x nanowires15,16 grown by chemical vapour deposition has shown a Tc over 300 K, however, the electric-field control of ferromagnetism was never demonstrated16,17. Later, our molecular beam epitaxy (MBE)-grown MnxGe1−x quantum dots (QDs)11 were also found to exhibit a Tc higher than room temperature, and more significantly we demonstrated electric-field controlled ferromagnetism. The underlying mechanism of the high Tc was attributed to the strong quantum confinement effect, which significantly enhanced the exchange coupling between the confined holes and localized Mn ions11,17,18,19. Those inspiring results point to a new direction of using nanostructures to realize high-Tc MnxGe1−x FMS. However, MnxGe1−x QDs suffer from high growth temperature, inhomogeneity from self-assembly growth and inherent difficulty for transistor. And above all, it is difficult to process them for practical devices. Likewise, for 1D nanowires, the fabrication process usually relies on low-throughput e-beam lithography20. To overcome the above obstacles, new nanostructures must be produced in a large-scale uniform fashion with easy device fabrication and convenient integration with current Si technology. Nanomesh21 appears to be an attractive candidate since it can be easily fabricated in large scale with high throughput and good uniformity.

Commonly, two strategies could be employed to realize nanomesh structures21,22. One is post-synthesis top-down etching the two-dimensional film into nanomesh structures23, and the other is to form a nanopattern first followed by the growth of nanomesh structures24. In comparison, the former method still suffers the aforementioned precipitates problem, and unavoidably produces rough and disordered edges during the etching process, which may also give rise to additional carriers scattering and reduce carrier mobility22. Therefore, the latter one is preferred for the high-quality nanomesh fabrication. Among nanofabrication methods, nanosphere lithography is a good choice due to its simplicity, low cost and high throughput in the fabrication of large-area and highly-ordered patterns25. Furthermore, the pattern size can be easily controlled by choosing specific nanosphere diameters and by adjusting the etching time. All these advantages make it extremely suitable for nanomesh fabrication in large scale with a controllable size.

Herein, wafer-scale MnxGe1−x nanomeshes with controlled Mn-doping concentrations are successfully grown by MBE at low temperature on a pre-patterned Ge substrate fabricated by nanosphere lithography. The nanomesh width can be well controlled by oxygen-plasma etching of the nanospheres. Comprehensive transmission electron microscopy (TEM) characterizations of the nanomeshes clearly show the coherent growth of the MnxGe1−x nanomesh on Ge substrate, without observable intermetallic precipitates. The Tc of the MnxGe1−x nanomesh is found to be above 400 K measured by a superconducting quantum interference device (SQUID). Furthermore, the magnetotransport measurement demonstrates the MR can be controlled from positive and negative values by applying a gate bias. With an appropriate Mn doping concentration, a giant MR as high as 8,000% is obtained at low temperature. We attribute the giant MR to a geometrical enhancement of the unique nanomesh structure. The successful growth and engineering of the high-Tc MnxGe1−x nanomesh with a tunable giant and electric-field controlled MR have the potential to exploit the Ge-based spintronics devices, such as spin field-effect transistor (spinFET)9, magnetic random-access memory26,27 and magnetic-field sensor8, among others.


Pattern design

Figure 1a–c shows the process-flow diagrams on how to fabricate periodic pattern on Ge substrate. After self-assembled patterning nanospheres, a typical scanning electron microscopy (SEM) image (see Fig. 1d) clearly shows the close-packed single layer of nanospheres with a diameter of 220 nm. By controlling the O2 plasma-etching rate, the nanospheres diameter is successfully shrunk to 160 nm with a 60 nm gap between them as shown in Fig. 1e. The nanosphere pattern was then transferred onto the SiO2 layer by dry etching firstly, followed with short-time wet etching. Due to the anisotropic dry etching, SiO2 nanopillars were obtained, between which the Ge substrate was exposed for the nanomesh growth. The periodic SiO2 nanopillars are clearly shown in the SEM image (Fig. 1f).

Figure 1: Process flow diagrams for the pattern fabrication.

(a) Self-assembly growth of closely-packed single layer of nanospheres on the Ge substrate. (b) O2-plasma etching of the nanospheres to reach the desired size. (c) SiO2 nanopillars formed by dry etching, masked by the nanospheres. (df) The corresponding SEM images of every steps. Scale bar, 100 nm.

Structural characterization of the MnxGe1−x nanomesh

The SiO2 mask was subsequently removed after the MBE growth by selective etching and only the MnxGe1−x nanomesh was left on the Ge substrate. Figure 2a is a schematic illustration of the MnxGe1−x nanomesh, in which the yellow-colored nanomesh sits on top of the blue-colored Ge (111) substrate. The typical morphology of the sample was captured by SEM as shown in Fig. 2b. A periodic nanomesh structure has a nanomesh width of 60 nm and a hole diameter of 160 nm; a magnified SEM image is also shown in the inset. To further characterize the microstructure and composition of the formed nanomesh, the cross-sectional TEM was used and the results are shown in Fig. 2c–g, for which a focused ion beam was employed to cut the sample along the diameter of the hole. In this process, a Cr/Au layer was deposited to protect the sample from damage of the ion beam. Figure 2c is a low-resolution cross-sectional TEM image of the MnxGe1−x nanomesh, which clearly shows that the nanomesh is grown on the Ge substrate as defined by the SiO2 pattern. The zoom-in image (Fig. 2e) shows that the MnxGe1−x nanomesh has a height of 25 nm and a width of 60 nm, consistent with the SEM result. The high-resolution TEM (HRTEM) image (Fig. 2f) clearly demonstrates the coherent growth of the MnxGe1−x nanomesh on Ge substrate. It does not reveal observable intermetallic compounds. The Fourier-transform image as shown in Fig. 2g gives only one set of periodic pattern, proving a perfectly epitaxial growth. This image can be indexed to the [011] zone axis of the Ge diamond lattice and the epitaxial growth direction is along [111]. To understand the nanomesh composition, energy-dispersive spectroscopy (EDS) was used and the result is given in Fig. 2d. The Mn doping concentration can be estimated to be 3%. The additional peaks, Cr and Si, come from the protection layer and the SiO2 pattern, respectively. The EDS mapping and line scan analysis can further confirm the uniform distribution of Mn in MnxGe1−x nanomesh, shown in Supplementary Fig. 1. Combined with the HRTEM analysis, the grown MnxGe1−x nanomesh is believed to be single crystalline FMS11.

Figure 2: Structural properties of MnxGe1−x nanomeshes.

(a) Schematic illustration of the formed MnxGe1−x nanomeshes. (b) Typical SEM image of the MnxGe1−x nanomeshes, with a nanomesh width of 60 nm and a nanohole diameter of 160 nm. Scale bar, 200 nm. The inset is the magnified SEM image. Scale bar, 100 nm. (c) Typical cross-sectional TEM image of the MnxGe1−x nanomeshes, in which the nanomeshes sit well on the Ge substrate and are spaced by the SiO2 mask. Scale bar, 50 nm. (d) EDS spectrum, showing a Mn composition of 3%. (e) Magnified cross-sectional TEM image of the MnxGe1−x nanomesh. Scale bar, 20 nm. (f) HRTEM image of the interface between the MnxGe1−x nanomesh and the Ge substrate, clearly showing that the MnxGe1−x nanomesh is a single crystal. Scale bar, 2 nm. (g) Fourier-transform image of (f).

Magnetic properties of the MnxGe1−x nanomesh

To well understand the magnetic property of the MnxGe1−x nanomesh, comparative samples with different Mn doping concentrations (3% and 5%) are measured by SQUID. Figure 3a shows the temperature-dependent hysteresis loops of the 3% Mn-doped nanomesh (Mn0.03Ge0.97), when an external field is applied parallel to the sample surface. The S-shaped hysteresis loops indicate the ferromagnetism above 350 K8. Figure 3b is the magnified hysteresis loop obtained at 10 K, clearly showing a small coercivity of 100 Oe. The saturation magnetization (Ms) per Mn is estimated to be 0.87 μB at 10 K. In addition, Arrott plots were also used to evaluate the Tc, as shown in Fig. 3d. We observe that even at 350 K, the intercept, namely reciprocal of the susceptibility, does not vanish, indicating that the Tc has not been reached yet. The extrapolated dashed line further indicates the Tc is beyond 350 K, which agrees well with the hysteresis loops. Figure 3e shows the temperature-dependent Ms ranging from 10 to 400 K; it clearly shows a weak temperature dependence and a large magnetization remained at 400 K. All of the data support that the Tc is beyond 400 K. The temperature-dependent coercivity is shown in Fig. 3f, in which the coercivity decreases from 100 to 35 Oe with increasing temperature. The small coercivity in the entire investigated temperature range indicates that the soft ferromagnetism of our sample may come from the diluted nature of Mn8. The magnetic property of the 5% Mn-doped MnxGe1−x nanomesh (Mn0.05Ge0.95) is shown in Fig. 3g–i. Figure 3g shows a similar temperature dependence of hysteresis loops as that in the Mn0.03Ge0.97 nanomesh. Furthermore, a high Tc (above 400 K) can also be seen from both the S-shaped hysteresis loop (Fig. 3h) and Arrott plot (Fig. 3i). However, a careful examination indicates that the saturated magnetic moment per Mn in the Mn0.05Ge0.95 nanomesh is a little smaller than that in the Mn0.03Ge0.97 nanomesh. The underlying mechanism may arise from the increased interstitial Mn density, as the Mn doping concentration increases28. It has been well documented that the Mn interstitials are ferromagnetically inactive29. Meanwhile, Mn interstitials acting as double donors29 can compensate a significant fraction of free holes, which can further weaken the hole-mediated ferromagnetism6,30.

Figure 3: Magnetic properties of MnxGe1−x nanomeshes.

(a) Magnetic hysteresis loops of the Mn0.03Ge0.97 nanomeshes measured at different temperatures from 10–350 K. (b,c) The magnified hysteresis loop obtained at 10 and 350 K, respectively. (d) Arrott’s plots showing that the Tc is above 350 K. (e) The temperature-dependent saturation moment. (f) Temperature-dependent coercivity decreasing from 100 Oe at 10 K to 35 Oe at 400 K. (g) Temperature-dependent MH curves of the Mn0.05Ge0.95 nanomeshes. (h) The magnified hysteresis loop obtained at 400 K. (i) Arrott’s plots showing that its Tc is above 400 K.

To date, several papers have reported a similar high Tc phenomenon8,14,15,16,31 in MnxGe1−x systems, which can be classified into two types of structures: Mn-rich MnxGe1−x nanocolumn8 and Mn-diluted MnxGe1−x nanostructure14,15,16,31. The Mn-rich MnxGe1−x nanocolumn exhibited a Tc over 400 K8, but no electric-field control of ferromagnetism was reported. Recently, the nanocolumn has been proven to be new phase of MnGe2 (ref. 32). The Mn-diluted MnxGe1−x nanostructures14,15,16,31, including nanowires and nanodots, usually displayed a Tc over 300 K. In particular, Mn0.05Ge0.95 QDs showed an electric-field controlled ferromagnetism. To explore the origin of the high Tc in our sample, the zero-field cooled (ZFC) and field cooled (FC) magnetization measurement were performed and the superimposed ZFC and FC curves (Supplementary Fig. 2) indicate there are no any nanophase precipitates29. Together with the TEM evidence, we believe the origin of the high Tc in our sample comes from the nanosize and quantum confinement effect of our MnxGe1−x nanomesh. As the sample is scaled down to nanoscale, the quantum confinement effect in nanostructures can significantly boost the exchange coupling between the confined holes and the localized Mn11, leading to a higher Tc compared with bulk materials1,11,17. Additionally, when it comes to the nanostructure, the ability to accommodate the strain without formation of the lattice defects can be dramatically enhanced due to its large surface33,34. In the MnxGe1−x thin film, the lattice strain associated with substitutional Mn can affect and limit the growth of high-quality FMS35; the MnxGe1−x lattice swells up to a critical strain above which Mn atoms segregate into Mn-rich nanocrystals35. In the MnxGe1−x nanomesh structure, however, the strain induced by Mn doping could be relieved due to the large surface area of the nanomesh and thus more Mn atoms can be accommodated into the substitutional position, giving rise to a higher Tc than thin films29,36,37. To understand the size confinement on the growth quality, we give a systematic ZFC and FC measurement on different size nanomesh structures, shown in Supplementary Fig. 2, and the result clearly proves the size confinement effect can suppress the compound formation and improve the Tc (detailedly discussed in Supplementary Note 1). Finally, we should point out that previously reported Mn-diluted nanostructures14,15,16,31, including the chemical vapour deposition-grown nanowires and self-assembly grown QDs, were usually formed at a high growth temperature. In comparison, our MnxGe1−x nanomesh, which was grown at low temperature, can fully utilize the advantages of the non-equilibrium MBE technology. More Mn can be incorporated into the Ge lattice in our nanomesh structure in a non-equilibrium manner, and thus a much higher Tc can be expected. Overall, the unique use of MnxGe1−x nanomeshes facilitates the formation of the high-Tc ferromagnetic semiconductor.

Magnetotransport measurement of the MnxGe1−x nanomesh

Furthermore, the magnetotransport property of the Mn0.03Ge0.97 nanomesh was revealed by physical property measurement system (PPMS) and the results are shown in Fig. 4. Figure 4a shows the temperature-dependent resistivity of the Mn0.03Ge0.97 nanomesh under the magnetic field of 0 T (blue square) and 4 T (red circle), respectively, in which a huge difference can be observed. A clear metal-to-insulator transition without applying magnetic field (0 T) can be observed with a low-temperature (T<30 K) activation region and a high-temperature (T>30 K) saturation region. From the Arrhenius relation38, the activation energy of our Mn0.03Ge0.97 nanomesh is estimated to be 11 meV, which, however, is lower than the substitutional Mn acceptor energy level (160 meV)11. A similar low activation energy (30 meV) of Mn in Ge has also been reported in a previous study39. The underlying mechanism comes from the high-doping level and the presence of exchange interactions, inducing the broadening and possible splitting of the impurity band39. Above 30 K, the RT curve could be well fitted by a power-law relation (Tα) with α≈1.6 close to the value 1.5 predicted for hole scattering by phonons in Ge8. The fitting was plotted in red over the blue data in the RT curve. As shown in Fig. 4b, the ordinary Hall coefficient at 300 K is positive, further proving that the majority carriers are holes provided by the substitutional Mn acceptors.

Figure 4: Temperature-dependent magnetotransport measurement of MnxGe1−x nanomeshes.

(a) Temperature-dependent resistivity of the Mn0.03Ge0.97 nanomeshes measured without an applied magnetic field (blue square) and with a magnetic field of 4 T (red circle), clearly showing a metal–insulator transition and a large resistivity deviation between them. The red solid line over the blue data is the fitting curve of a power-law relation to T1.6, indicating the carrier scattering by phonons. (b) The Hall resistance at 300 K. (c,d) Temperature-dependent MR measured at low-temperature and high-temperature regions, respectively. (e) The schematic illustration of scattering mechanism. (f) Temperature-dependent resistivity of the Mn0.05Ge0.95 nanomeshes, showing a much larger resistivity compared with that in the Mn0.03Ge0.97 nanomeshes. (g,h) Temperature-dependent MR measured at low-temperature and high-temperature regions, respectively, showing a MR transition from a negative to positive value. (i) The schematic illustration of two competitive scattering mechanisms from magnetic polarons in Mn0.05Ge0.95 nanomeshes.

An intriguing phenomenon is the observation of a giant positive MR in the Mn0.03Ge0.97 nanomesh as shown in Fig. 4c,d, when an out-of-plane magnetic field was applied. At a magnetic field of 4 T, the MR as high as 2,000% at 10 K is observed and still 75% at 300 K; however, the maximum of 8,000% occurs at 30 K. Traditionally, the positive MR comes from the orbital MR, which is proportional to (μB)2, where μ is the carrier mobility13. However, the estimated orbital MR is too small to account for our giant MR. A similar MR effect has been reported previously in a MnxGe1−x film, which was due to the formation of Mn-rich nanocolumns8,14. In our Mn0.03Ge0.97 nanomesh, the presence of such Mn-rich nanocolumns has been excluded from our extensive TEM analysis as discussed above. To reveal the reason behind the giant MR, an unpatterned Ge wafer was simultaneously loaded into the chamber for thin film growth as a reference, along with the Mn0.03Ge0.97 nanomesh sample. However, only 1% positive MR can be detected in the thin film (shown in Supplementary Fig. 3). Such distinct MR behaviours between the Mn0.03Ge0.97 nanomesh and thin film (see Supplementary Note 2) give the conclusion that the giant MR can only be attributed to the geometrically enhanced MR26,40 owing to the unique mesh structure. To understand it, the nanomesh structures can be considered as a highly conductive Mn0.03Ge0.97 percolation network with periodic nanoholes. Without applying a magnetic field, the current flows through the Mn0.03Ge0.97 nanomesh, and the current direction is parallel to the local electric field. As magnetic field is applied, the current is deflected due to the Lorentz force41; the current and the local electric field are no longer collinear. The angle between them is determined by the Hall angle θ=arctan(μHH), where the μH is the Hall mobility. For a sufficiently large magnetic field, the current is obviously deflected from the highly conductive nanomesh to the insulated nanoholes, resulting in a high resistance. The transition from the extremely low resistance at zero magnetic field to the extremely high resistance at a large magnetic field gives rise to the giant MR for the Mn0.03Ge0.97 nanomesh, as illustrated in Fig. 4e. Thus, it may be concluded that the lower the initial resistance of the nanomesh is, the larger the MR became at a given magnetic field. It can be verified from the deviations of the RT curves between 4 and 0 T, as already shown in Fig. 4a. The largest deviation happens at the lowest resistivity (at 30 K and 0 T), which agrees well with our proposed model.

For comparison, the magnetotransport property of the Mn0.05Ge0.95 nanomesh was also measured by PPMS, and the RT curve is shown in Fig. 4f. A similar metal-to-insulator transition could be observed, however, the resistivity was found to be more than one order of magnitude larger than that of the Mn0.03Ge0.97 nanomesh. The increased resistivity may come from the increased scattering centres from the higher Mn doping concentration42. Figure 4g,h shows the temperature-dependent MR of the Mn0.05Ge0.95 nanomesh sample. Intriguingly, the geometrically enhanced giant MR becomes less pronounced in the Mn0.05Ge0.95 nanomesh sample. This may be due to the dramatically increased resistivity and the decreased Hall angle from the lower mobility compared with that in the Mn0.03Ge0.97 nanomesh. In addition, the Mn0.05Ge0.95 nanomesh shows a negative MR for temperature below 40 K and a positive MR above 160 K. In the intermediate temperature region (40–160 K), the MR contains two contributions: a positive MR appears at low magnetic field, followed by a negative slope at high field. From previous reports29,43, such negative-to-positive MR transition could be attributed to two competitive effects: the spin-dependent scattering by ferromagnetic clusters29,44 or magnetic polarons29,44 gave rise to the negative MR, while their spatial fluctuations led to the positive MR29,44. Since the presence of ferromagnetic clusters in our nanomesh sample has been excluded from the above detailed microstructural analysis and ZFC and FC measurement, the two competitive effects (illustrated in Fig. 4i) most likely come from the magnetic polarons 44,45,46, which are the origin of the ferromagnetism in the FMS. The radius of the polaron, rh, grows with decreasing temperature defined by rh(ab/2)ln(sS|J0|/(kBT)), where ab is the decay length of the localized carrier, s and S are the absolute values of the hole and impurity spin, respectively and J0 is the strength of the exchange coupling between carriers and magnetic ions47. As a magnetic field is applied, the carrier mobility increases due to the suppression of the spin-dependent scattering by the magnetic polarons, giving rise to the negative MR, which is proportional to the susceptibility (χ(H))29,43 of the sample. However, the spin-dependent scattering effect by the magnetic polarons decreases with increasing temperature43, which makes the positive MR from the spatial fluctuations of magnetic polarons gradually show up. Due to the strong pd exchange coupling6 in MnxGe1−x, the magnetic polarons will cause a strongly localized valence band splitting into two spin subbands ɛSJM(r)43,48, where J is the exchange coupling energy, S is the spin of electron and M(r) is the local magnetic moment of the magnetic polaron. This can form a complicated landscape for the valence band with hills and valleys, serving as the hole traps. With increasing the magnetic field, the hole traps become deeper, the number of itinerant holes decreases, and thus the MR increases. This positive MR is proportional to the square of the magnetization (M2(H))29, which can be reflected from the steep slope of the positive MR. As the temperature further increases, the contribution from the spin-dependent scattering becomes weaker, and finally only positive MR dominates the transport.

Gate control of MR in the MnxGe1−x nanomesh

For low-power applications, it is desirable to exploit the electric-field control of ferromagnetism in our MnxGe1−x nanomesh samples. For the measurement, an Al2O3 layer was deposited by atomic layer deposition on the nanomesh surface as the gate dielectric, followed by the e-beam evaporation of Cr/Au as the gate metal contact. Due to the unique nanomesh structure, the almost-wrap-around gate can be realized to provide a three-dimensional electric-field control of the conduction channel, thereby giving a very efficient and robust carrier modulation. The schematic illustration and optical micrograph of the gated nanomesh device are shown in Fig. 5a,b, respectively. Figure 5c–f shows the gate bias-dependent MR of the Mn0.05Ge0.95 nanomesh sample measured at 40, 60 and 100 K, while the comparison results on the Mn0.03Ge0.97 nanomesh sample are shown in Supplementary Fig. 4. When the gate bias changes from −8 to 8 V, a clear negative-to-positive MR transition is observed at 40 K. As described above, this phenomenon should also stem from the competitive effect between the spin-dependent scattering-induced negative MR and the spatial fluctuations-induced positive MR. As already reported, hole-mediated ferromagnetism in Mn-doped Ge6 has the ability to control the magnetic phase transition from a strong ferromagnetism to a soft ferromagnetism, when changing the gate bias from negative to positive. Therefore, as a negative gate bias is applied on our sample, the enhanced ferromagnetism with larger susceptibility (χ(H)) in the Mn0.05Ge0.95 nanomesh can provide a strong spin-dependent scattering, consequently giving rise to a negative MR. When the gate bias is swept to positive, the weakened ferromagnetism gives rise to a weaker spin-dependent scattering. From another point of view, the decreased density of magnetic polarons due to reduced carrier density probably makes them from an overlap-together state to a disconnecting state, which thus increases the polaron fluctuation, as the gate bias goes to the positive range. And hence the positive MR dominates. Additionally, from the zoom-in MR curves (shown in Fig. 5d), the biggest coercivity happens at −8 V with the highest hole density, which further confirms the hole-mediated ferromagnetism in our Mn0.05Ge0.95 nanomesh. To further illustrate the hole-mediated ferromagnetism, gate-dependent Hall measurements were performed on the Mn0.05Ge0.95 nanomesh, and the results are shown in Supplementary Fig. 5a,b. From the gate-dependent anomalous Hall term (RxyA) over longitudinal resistance (Rxx), a hole-mediated ferromagnetism can be clearly observed, (see the details in Supplementary Note 3).

Figure 5: Electric-field-controlled magnetoresistance (MR) of MnxGe1−x nanomeshes.

(a) Schematic illustration and (b) Optical micrograph of the Mn0.05Ge0.95 nanomesh device for electric-field control of MR measurement, respectively. Scale bar, 50 μm. (c) Electric-field controlled MR at 40 K, clearly showing a transition from negative to positive when the gate was biased from negative to positive. (d) Magnified MR curves, clearly showing the biggest coercivity at −8 V. (e,f) Electric-field controlled MR transition at 60 and 100 K, respectively.

In addition, gate-dependent MR curves at 60 and 100 K are shown in Fig. 5e,f, respectively. At 60 K, the MR transition from the negative to positive value can still be clearly observed when tuning the gate bias. However, the control effect is not as strong as that at 40 K in the negative bias range. This effect can be explained by the fact that even at 0 V, the hole density at 60 K is already high enough to align most of the Mn ions along one direction11. Further increasing the negative bias does not significantly enhance the ferromagnetism. At 100 K, it shows a similar effect, that is, an even weaker control effect in the negative bias range. The gate-dependent Hall measurement at 100 K is shown in Supplementary Fig. 5c,d, which again shows the ferromagnetism modulation by the gate bias. Therefore, it is shown that by manipulating the Mn-doping concentration, not only giant MR but also electric-field controlled MR can be engineered. This property may offer a great advantage for designing future spintronics devices with voltage controlled, low-power and non-volatile functions.


To summarize, wafer-scale MnxGe1−x nanomeshes with controlled Mn-doping concentration were successfully grown by MBE on a pre-patterned Ge substrate fabricated by nanosphere lithography. Comprehensive TEM characterization revealed its single crystallinity without observable intermetallic compounds formation. Extensive magnetic property measurements disclosed that the nanomesh exhibited a Tc over 400 K. Through magnetotransport measurement, two intriguing phenomena were found in such materials when adjusting the Mn-doping concentration: one was that the nanomesh exhibited a giant MR even up to 8,000%, and the other was that the MR could be engineered by external electric field. Both these features were attributed to the unique meshed nanostructures. The discovery of the extraordinary property in MnxGe1−x nanomeshes points out an effective approach to synthesize high-quality and high-Tc FMS material. It may lay the foundation to build room-temperature spintronics devices, such as spinFET9, magnetic random-access memory27 and magnetic-field sensor49.


MBE growth

A 100 nm-thick SiO2 thin film was firstly deposited on a Ge (111) wafer by plasma enhanced chemical vapour deposition, followed by the formation of a large-scale and closely-packed hexagonal single layer of nanospheres on the SiO2 substrate (Fig. 1a). By adjusting O2 plasma etching time, the nanospheres were successfully shrunk to the desired size with appropriate gaps between the nanospheres as shown in Fig. 1b. Using the nanospheres as the mask, the pattern was transferred to the bottom SiO2 layer by two-step etching. The dry etching was firstly employed to etch SiO2 layer and stopped with 10 nm SiO2 left. Then, wet etching was hired to remove the left SiO2 layer. After dissolving the nanospheres, periodic SiO2 nanopillars were obtained on the substrate (shown in Fig. 1c). For the MnxGe1−x nanomesh growth, the SiO2 nanopillars pattern was cleaned by O2 plasma to remove any organic residue induced by dry etching. Followed by rinsing in acetone, isopropyl alcohol and diluted hydrofluoric acid, the substrate was loaded into the MBE chamber for the MnxGe1−x growth. High-purity Ge and Mn sources were evaporated from Knudsen effusion cells in MBE growth. After degassing at 600 °C for 30 min, the patterned substrate was in-situ cooled down to 160 °C for the MnxGe1−x nanomesh growth with a Ge growth rate of 0.2 Å s−1 and a controlled Mn flux as the dopant source.


The morphologies of the patterns and formed nanomeshes were investigated by SEM. The microstructure and composition were comprehensively characterized by TEM equipped with EDS. SQUID was employed to measure their magnetic properties. Furthermore, the magnetotransport studies were carried out in the Quantum Design PPMS.

Data availability

The data that support the findings of this study are available from the corresponding author upon request.

Additional information

How to cite this article: Nie, T. et al. Enhancing electric-field control of ferromagnetism through nanoscale engineering of high-Tc MnxGe1−x nanomesh. Nat. Commun. 7:12866 doi: 10.1038/ncomms12866 (2016).


  1. 1

    Xiu, F. et al. Synthesis of high-Curie-temperature Fe0.02Ge0.98 quantum dots. J. Am. Chem. Soc. 132, 11425–11427 (2010).

    CAS  Article  Google Scholar 

  2. 2

    Matsukura, F., Tokura, Y. & Ohno, H. Control of magnetism by electric fields. Nat. Nanotechnol. 10, 209–220 (2015).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Ohno, H. Making nonmagnetic semiconductors ferromagnetic. Science 281, 951–956 (1998).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Ikeda, S. et al. A perpendicular-anisotropy CoFeB–MgO magnetic tunnel junction. Nat. Mater. 9, 721–724 (2010).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Ohno, H. et al. Electric-field control of ferromagnetism. Nature 408, 944–946 (2000).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Park, Y. D. et al. A group-IV ferromagnetic semiconductor:MnxGe1−x . Science 295, 651–654 (2002).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Žutić, I., Fabian, J. & Das Sarma, S. Spintronics: fundamentals and applications. Rev. Mod. Phys. 76, 323–410 (2004).

    ADS  Article  Google Scholar 

  8. 8

    Jamet, M. et al. High-Curie-temperature ferromagnetism in self-organized Ge1−xMnx nanocolumns. Nat. Mater. 5, 653–659 (2006).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Tang, J., Nie, T. & Wang, K. L. Spin transport in Ge nanowires for diluted magnetic semiconductor-based nonvolatile transpinor. ECS Trans. 64, 613–623 (2014).

    CAS  Article  Google Scholar 

  10. 10

    Tang, J. et al. Electrical spin injection and detection in Mn5Ge3/Ge/Mn5Ge3 nanowire transistors. Nano Lett. 13, 4036–4043 (2013).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Xiu, F. et al. Electric-field-controlled ferromagnetism in high-Curie-temperature Mn0.05Ge0.95 quantum dots. Nat. Mater. 9, 337–344 (2010).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Thi-Giang, L. et al. Growth competition between semiconducting Ge1−xMnx nanocolumns and metallic Mn5Ge3 clusters. Adv. Nat. Sci: Nanosci. Nanotechnol. 3, 025007–025011 (2012).

    ADS  Google Scholar 

  13. 13

    Park, Y. D. et al. Magnetoresistance of Mn:Ge ferromagnetic nanoclusters in a diluted magnetic semiconductor matrix. Appl. Phys. Lett. 78, 2739–2741 (2001).

    ADS  CAS  Article  Google Scholar 

  14. 14

    Faxian, X. et al. MnGe magnetic nanocolumns and nanowells. Nanotechnology 21, 255602–255606 (2010).

    Article  Google Scholar 

  15. 15

    van der Meulen, M. I. et al. Single crystalline Ge1−xMnx nanowires as building blocks for nanoelectronics. Nano Lett. 9, 50–56 (2008).

    ADS  Article  Google Scholar 

  16. 16

    Cho, Y. J. et al. Ferromagnetic Ge1−xMx (M=Mn, Fe, and Co) nanowires. Chem. Mater. 20, 4694–4702 (2008).

    CAS  Article  Google Scholar 

  17. 17

    Holub, M. et al. Mn-doped InAs self-organized diluted magnetic quantum-dot layers with Curie temperatures above 300 K. Appl. Phys. Lett. 85, 973–975 (2004).

    ADS  CAS  Article  Google Scholar 

  18. 18

    Abolfath, R. M., Hawrylak, P. & Žutić, I. Tailoring magnetism in quantum dots. Phys. Rev. Lett. 98, 207203–207206 (2007).

    ADS  Article  Google Scholar 

  19. 19

    Mackowski, S. et al. Optically-induced magnetization of CdMnTe self-assembled quantum dots. Appl. Phys. Lett. 84, 3337–3339 (2004).

    ADS  CAS  Article  Google Scholar 

  20. 20

    Zeng, Z. et al. Fabrication of graphene nanomesh by using an anodic aluminum oxide membrane as a template. Adv. Mater. 24, 4138–4142 (2012).

    CAS  Article  Google Scholar 

  21. 21

    Bai, J., Zhong, X., Jiang, S., Huang, Y. & Duan, X. Graphene nanomesh. Nat. Nanotechnol. 5, 190–194 (2010).

    ADS  CAS  Article  Google Scholar 

  22. 22

    Wang, M. et al. CVD growth of large area smooth-edged graphene nanomesh by nanosphere lithography. Sci. Rep. 3, 1238–1243 (2013).

    Article  Google Scholar 

  23. 23

    Ding, J. et al. Transfer patterning of large-area graphene nanomesh via holographic lithography and plasma etching. J. Vac. Sci. Technol. B 32, 06FF01–06FF07 (2014).

    ADS  Article  Google Scholar 

  24. 24

    Qiu, T. et al. Au@MnO2 core–shell nanomesh electrodes for transparent flexible supercapacitors. Small 10, 4136–4141 (2014).

    CAS  PubMed  Google Scholar 

  25. 25

    Chen, Y. W. et al. Enhanced photoluminescence due to lateral ordering of GeSi quantum dots on patterned Si(001) substrates. Nanotechnology 21, 175701–175704 (2010).

    ADS  Article  Google Scholar 

  26. 26

    Yuldashev, S. U. et al. Enhanced positive magnetoresistance effect in GaAs with nanoscale magnetic clusters. J. Appl. Phys. 90, 3004–3006 (2001).

    ADS  CAS  Article  Google Scholar 

  27. 27

    Yuasa, S., Nagahama, T., Fukushima, A., Suzuki, Y. & Ando, K. Giant room-temperature magnetoresistance in single-crystal Fe/MgO/Fe magnetic tunnel junctions. Nat. Mater. 3, 868–871 (2004).

    ADS  CAS  Article  Google Scholar 

  28. 28

    Gunnella, R. et al. Localization of the dopant in Ge: Mn diluted magnetic semiconductors by x-ray absorption at the Mn K edge. J. Phys.: Condens. Matter 22, 216006–216015 (2010).

    ADS  CAS  Google Scholar 

  29. 29

    Li, A. P. et al. Dopant segregation and giant magnetoresistance in manganese-doped germanium. Phys. Rev. B 75, 201201–201204 (2007).

    ADS  Article  Google Scholar 

  30. 30

    Nie, T. et al. Superlattice of FexGe1−x nanodots and nanolayers for spintronics application. Nanotechnology 25, 505702–505708 (2014).

    ADS  Article  Google Scholar 

  31. 31

    Seong, H.-K. et al. Magnetic and electrical properties of single-crystalline Mn-doped Ge nanowires. J. Phys. Chem. C 113, 10847–10852 (2009).

    CAS  Article  Google Scholar 

  32. 32

    Dalmas de Réotier, P. et al. Core–shell nanostructure in a Ge0.9Mn0.1 film observed via structural and magnetic measurements. Phys. Rev. B 91, 245408–245416 (2015).

    ADS  Article  Google Scholar 

  33. 33

    Lauhon, L. J., Gudiksen, M. S., Wang, D. & Lieber, C. M. Epitaxial core–shell and core–multishell nanowire heterostructures. Nature 420, 57–61 (2002).

    ADS  CAS  Article  Google Scholar 

  34. 34

    Nie, T.-X. et al. Thermally oxidized formation of new Ge dots over as-grown Ge dots in the Si capping layer. J. Appl. Phys. 110, 114304–114308 (2011).

    ADS  Article  Google Scholar 

  35. 35

    De Padova, P. et al. Mn0.06Ge0.94 diluted magnetic semiconductor epitaxially grown on Ge(001): influence of Mn5Ge3 nanoscopic clusters on the electronic and magnetic properties. Phys. Rev. B 77, 045203–045209 (2008).

    ADS  Article  Google Scholar 

  36. 36

    Chen, L. et al. Enhancing the Curie temperature of ferromagnetic semiconductor (Ga,Mn)As to 200 K via nanostructure engineering. Nano Lett. 11, 2584–2589 (2011).

    ADS  CAS  Article  Google Scholar 

  37. 37

    Erwin, S. C. & Petukhov, A. G. Self-compensation in manganese-doped ferromagnetic semiconductors. Phys. Rev. Lett. 89, 227201–227204 (2002).

    ADS  Article  Google Scholar 

  38. 38

    Chang, L.-T. et al. Electric-field control of ferromagnetism in Mn-doped ZnO nanowires. Nano Lett. 14, 1823–1829 (2014).

    ADS  CAS  Article  Google Scholar 

  39. 39

    Tsui, F. et al. Novel germanium-based magnetic semiconductors. Phys. Rev. Lett. 91, 177203–177206 (2003).

    ADS  CAS  Article  Google Scholar 

  40. 40

    Solin, S. A., Thio, T., Hines, D. R. & Heremans, J. J. Enhanced room-temperature geometric magnetoresistance in inhomogeneous narrow-gap semiconductors. Science 289, 1530–1532 (2000).

    ADS  CAS  Article  Google Scholar 

  41. 41

    Heimbrodt, W. et al. Magnetic interactions in granular paramagnetic–ferromagnetic GaAs: Mn/MnAs hybrids. J. Supercond. 18, 315–320 (2005).

    ADS  CAS  Article  Google Scholar 

  42. 42

    Iye, Y. et al. Metal–insulator transition and magnetotransport in III–V compound diluted magnetic semiconductors. Mater. Sci. Eng. B 63, 88–95 (1999).

    Article  Google Scholar 

  43. 43

    Haas, C. Spin-disorder scattering and magnetoresistance of magnetic semiconductors. Phys. Rev. 168, 531–538 (1968).

    ADS  Article  Google Scholar 

  44. 44

    Li, A. P. et al. Magnetism in MnxGe1−x semiconductors mediated by impurity band carriers. Phys. Rev. B 72, 195205–195213 (2005).

    ADS  Article  Google Scholar 

  45. 45

    Xiu, F. et al. Room-temperature electric-field controlled ferromagnetism in Mn0.05Ge0.95 quantum dots. ACS Nano 4, 4948–4954 (2010).

    CAS  Article  Google Scholar 

  46. 46

    Kumar, S. et al. Magnetic polarons and large negative magnetoresistance in GaAs nanowires implanted with Mn ions. Nano Lett. 13, 5079–5084 (2013).

    ADS  CAS  Article  Google Scholar 

  47. 47

    Kaminski, A. & Das Sarma, S. Polaron percolation in diluted magnetic semiconductors. Phys. Rev. Lett. 88, 247202–247204 (2002).

    ADS  CAS  Article  Google Scholar 

  48. 48

    Riss, O. et al. Magnetization-driven metal–insulator transition in strongly disordered Ge:Mn magnetic semiconductors. Phys. Rev. B 79, 241202–241204 (2009).

    ADS  Article  Google Scholar 

  49. 49

    Wolf, S. A. et al. Spintronics: a spin-based electronics vision for the future. Science 294, 1488–1495 (2001).

    ADS  CAS  Article  Google Scholar 

Download references


This work is supported in part by the FAME Center, one of the six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA. We also gratefully acknowledge the financial support from the National Science Foundation through grant ECCS 1308358.

Author information




T.N. and K.L.W. jointly conceived the ideas and designed the experiments. T.N. contributed to MBE growth, fabricate and measure the devices with contributions from J.T, X.K. X.Z. L.-T.C. and K.M. S.L. performed the TEM characterization. T.N. wrote the paper. All authors discussed the results and commented on the manuscript. The study was performed under the supervision of K.L.W.

Corresponding authors

Correspondence to Tianxiao Nie or Kang L. Wang.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Information

Supplementary Figures 1-5, Supplementary Notes 1-3 and Supplementary References (PDF 2191 kb)

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Nie, T., Tang, J., Kou, X. et al. Enhancing electric-field control of ferromagnetism through nanoscale engineering of high-Tc MnxGe1−x nanomesh. Nat Commun 7, 12866 (2016). https://doi.org/10.1038/ncomms12866

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing