Abstract
The helimagnet FeP is part of a family of binary pnictide materials with the MnPtype structure, which share a nonsymmorphic crystal symmetry that preserves generic band structure characteristics through changes in elemental composition. It shows many similarities, including in its magnetic order, to isostructural CrAs and MnP, two compounds that are driven to superconductivity under applied pressure. Here we present a series of high magnetic field experiments on highquality single crystals of FeP, showing that the resistance not only increases without saturation by up to several hundred times its zerofield value by 35 T, but that it also exhibits an anomalously linear field dependence over the entire range when the field is aligned precisely along the crystallographic caxis. A close comparison of quantum oscillation frequencies to electronic structure calculations links this orientation to a semiDirac point in the band structure, which disperses linearly in a single direction in the plane perpendicular to field, a symmetryprotected feature of this entire material family. We show that the two striking features of magnetoresistance—large amplitude and linear field dependence—arise separately in this system, with the latter likely due to a combination of ordered magnetism and topological band structure.
Similar content being viewed by others
Introduction
The orthorhombic MnPtype (or B31) family of materials has been under study for several decades^{1,2,3,4} but its diverse range of properties has recently been the subject of renewed attention. Aside from a peripheral connection to ironbased hightemperature superconductivity^{5}, novel magnetism^{6,7,8}, quantum criticality^{9,10}, metal–insulator transitions^{11}, and indications of nontrivial electronic topology^{12,13,14} have all been reported in a series of binary transition metalpnictides. Two members of the B31 family, CrAs and MnP itself, have also been shown to superconduct upon suppression of helimagnetic order under applied pressure^{9,15,16,17}, suggesting a novel interplay of ground states. Furthermore, a linear magnetoresistance (MR) was observed in CrAs near the magnetic quantum critical point and was attributed to the presence of a “semiDirac” point in the band structure (one which disperses linearly along a single direction in momentum space)^{14}. It has recently been shown that the nonsymmorphic Pnma structure of the MnP family preserves many specific band structure features, including the semiDirac point, across different members^{18}. CrAs, MnP, and paramagnetic WP (ambient pressure T_{c} = 0.8 K)^{19} are all predicted to be unconventional topological superconductors as a result, suggesting a possible connection between helimagnetism, superconductivity, and nontrivial topological features that deserves further attention.
MR has been key to revealing topological properties in many other materials. Semimetals such as WTe_{2}^{20}, Cd_{3}As_{2}^{21}, or NbP^{13} have shown extreme, nonsaturating increases of the resistivity in field on the order of 10^{5} times greater than ρ(0 T). This has been attributed to high mobility, massless carriers that result from the linear dispersion. However, for these materials semiclassical explanations could also be valid for the observed phenomena^{22}. For this reason, it is important to find materials with topological band structure elements but with other characterstics that differentiate them from the typical Weyl/Dirac semimetal. The very good metallic behavior and ordered magnetism found in the MnPtype family provide such an opportunity.
Here we present electrical resistance measurements of highquality single crystals of FeP, a B31 family member isostructural to CrAs and MnP, which orders magnetically below T_{N} = 120 K^{2}, in a state which was shown in closely related FeAs to feature a noncollinear spindensity wave order^{7}. Following up on basic transport and physical property measurements^{3,23}, we focus on transport and fermiology under high magnetic fields. We observe large, nonsaturating MR reaching values of several hundred times, which we attribute to a shift in carrier mobility below ~50 K that causes increased compensation of the Fermi surface (FS), an explanation that likely extends to other B31 compounds. Studying the fieldangle dependence, we observe features of a complex FS but also a singular linear MR when field is directed precisely along the crystallographic caxis. A careful comparison of quantum oscillations (QOs) data with calculated band structure directly confirms the location of the semiDirac point in this system, and more importantly, its role in the anomalous linear MR that would only occur in a certain field orientation. Seizing on new theoretical work^{24}, we propose a link between linear MR, topological band structure, and the magnetically ordered state, which combine to unique effect in FeP.
Results
Magnetotransport
The electrical resistivity (Fig. 1) and magnetic susceptibility^{3} of FeP are very similar to that of FeAs^{25}, with Sshaped curvature and distinct kink at T_{N} (which is 70 K in the arsenide), especially noticeable in the derivative, suggesting that itinerant spindensity wave magnetism is very similar in both compounds. However, even compared to the highest quality crystals of FeAs^{8}, the residual resistivity of FeP is extremely small (as low as 0.2 μΩ cm), with a residual resistivity ratio (RRR = ρ_{300K}/ρ_{1.8K}) of up to 1500, much larger than that of CrP and CrAs^{16,26,27}, and rivaled only by MnP^{28}. The trend of higher RRR in phosphides appears generic to this family, as it is also observed in CoAs^{29} and CoP (Supplementary Fig. 1). This makes FeP an ideal candidate for high fidelity measurements of magnetotransport and QOs, particularly at the high fields available with resistive magnets.
We focus on two singlecrystal samples, S1 and S2, for magnetotransport measurements with fields rotated through different crystallographic orientations as shown in Fig. 2. Both S1 and S2 exhibit very large and nonsaturating MR at all angles, as well as multifrequency QOs. For both, the largest MR is observed when B ∥ [100]. It is slightly larger for S2, likely due to its higher RRR^{20,30,31}. Most angles show generally similar behavior, with a superlinear field dependence that becomes more linear at high applied field. Power law fits to MR data below 15 T have a maximum n = 1.5 for the angles exhibiting the largest MR, which is nearly 300 times the zerofield resistivity. This increase is more than two orders of magnitude larger than that of FeAs in high field^{8}. Given the close structural and magnetic parallels between the two materials, it seems that the substitution of P for As and related significant decrease in residual resistivity produce a bigger infield response. Indeed, the B31 materials with the largest MR are all phosphides. MnP^{4} has only been measured up to 8 T, but at some angles actually shows a larger increase than FeP up to that field. CrP^{26} has been measured up to 58 T with an MR of about 350. All of these have a significant angular variation of MR. There are other examples of large MR in transition metal pnictide binaries: the four (Nb/Ta) (P/As) combinations, which form in a cubic structure, can reach values exceeding 1000 by 10 T. A comparison of the maximum reported MR for a variety of transition metalpnictogen binaries is given in Supplementary Table I.
The most striking aspect of the FeP angular MR data is found when the magnetic field is aligned along the crystallographic caxis. As shown in Fig. 2c, the MR data are linear from zero field (Fig. 2c, inset) up to the highest measured field of 35 T. The lowfield behavior was verified in the same sample after measurements at National High Magnetic Field Laboratory (NHMFL), confirming the linearity when sweeping the field through zero (see Supplementary Fig. 2). This orientation also has the lowest MR of any of the angles measured in either rotation plane. While MR tends toward linear behavior at high fields for all angles, there is no smooth decrease in crossover field, as none of the other curves are truly linear below 10 T. Therefore the caxis MR linearity must be closely linked to a particular property of the FS at that orientation.
The anisotropy is seen more clearly in a measurement of MR upon field rotation at constant fields, shown in Fig. 3, which also reveals intricate features similar to those observed in MnP^{4}. Both these features and the underlying πperiodic dependence indicate an anisotropic band structure with complex Fermi pocket shapes. Interestingly, as shown in Fig. 3a the maximum in MR actually occurs at an angle 5° away from the aaxis, where there is a narrow spike. This is not due to misalignment of the rotator, because the resistance minimum lies near 90° and is smoother and more symmetric. In fitting a cosine to the data and leaving all parameters except the π periodicity free, the maximum is still within 2° of the expected location.
To further investigate the detailed MR behavior, temperature sweeps at constant fields up to 14 T were made for specific angles, showing similar behavior to that seen in CrP and several extreme MR rareearth pnictide binaries, with a “turn on” temperature, T*, below which MR increases significantly^{20,26}. Figure 4a presents data for the angle with maximum MR in Fig. 3a, 5° away from B ∥ aaxis. Here we define T* as the temperature of the resistance minimum in the 14T sweep, roughly 35 K. As demonstrated in Fig. 4b, below this temperature, the angular dependence of the MR becomes noticeable, similar to that of previously mentioned materials.
As can be seen in both Figs. 1 and 4a, the zerofield resistance changes little below about 20 K. Our observation of a constant linear increase in MR down to 2 K means that in this temperature range the MR at constant B is no longer solely a function of ρ_{0}, in violation of Kohler’s rule^{32,33}. In other words, a different scattering mechanism has emerged below T*, or different parts of the FS are contributing to scattering. This can be compared to the data from other angles (Fig. 4b), all of which have a minimum at a similar T* at 14 T and at least sublinear behavior at lowest temperature, indicating an approach to saturation, mimicking zerofield behavior. This includes the data set taken at 0°, a small shift from that in Fig. 4a. Thus the Kohler’s rule violation does not occur at all angles (see Supplementary Fig. 2, which shows Kohler’s rule being more closely followed for B ∥ caxis).
Field sweeps in a Hall geometry for field along the [101], [010], and [001] directions are negative at all temperatures, and linear except for the [101] data (Supplementary Fig. 3a, c) below 50 K. Fits to a twoband model show that the most significant drop in the holeelectron mobility ratio is in the 25–50K range (Supplementary Fig. 3b), i.e., around T*, further proof of a change in the electronic properties of FeP at low temperatures.
Electronic structure and QOs
To understand the role of FS geometry and complexity in the observed anomalous MR field and angle dependence, we made a comparison between QO data and the calculated FS. Figure 5a presents the electronic structure calculated along the highsymmetry lines in reciprocal space. We assumed a paramagnetic configuration because of the small magnetic moment seen in susceptibility^{3}, which is a consequence of the octahedral crystal field that inverts the 4s and 3d energetic levels and changes the Fe configuration from commonly observed d^{6} to d^{8}. As will be seen, a comparison to experiment shows good agreement with the paramagnetic FS at low temperature. The band structure is similar to that recently reported^{34}, though our Fermi level appears to be at a slightly lower energy, which somewhat alters the appearance of the FS (Fig. 5b). As expected from the angular dependence of MR, the shapes of the FS pockets are not simple. The nonsymmorphic symmetry of the MnPtype structure implies the presence of several linear bands and semiDirac points. While the eightfolddegenerate anisotropic Dirac points at R and S are split via spinorbit coupling (SOC) in analogy to other topological pnictides^{35,36,37}, the fourfolddegenerate points at X, Y, and Z are protected by the nonsymmorphic symmetry^{18}.
The calculated FS is visualized in Fig. 5b. As expected from the angular dependence of MR, there are several pockets (of both hole and electron nature) and they have a complicated geometry. As the nonsymmorphic symmetry induces nontrivial features that may strongly affect transport properties, it is particularly important to examine the agreement between the theoretical FS and experimental data. To this end, we have calculated the frequencies of QOs which arise from orbits of carriers perpendicular to an applied magnetic field around the perimeters of FS pockets.
The extremal orbital frequency, with units of magnetic field, is related to the area enclosed by an orbit, thus the shape of the FS determines QO frequencies. While QOs are visible in the MR data, we found them easier to detect and analyze with torque magnetometry, which has a higher sensitivity in materials with low resistance and simpler background; our previous QO work on this family showed torque oscillations as low as 5 T when they did not appear in resistance up to 30 T^{8,29}. Magnetic torque was measured in parallel with MR measurements of sample S1, but on a different sample (S3). An example of the raw torque signal is shown in Fig. 6a for B ∥ a and B ∥ [011]. Because the torque amplitude becomes very small when the field is aligned with crystal axes, the B ∥ a data have been multiplied by a factor of five. Oscillations are clear down to about 7 T for both angles. A polynomial fit was subtracted to remove the nonoscillatory background, and a fast Fourier transform was performed on the residual signal. The frequency spectra for the same two angles are in Fig. 6b. The Greek letters in that panel correspond to orbits identified in a previous QO paper that rotated between the three principal axes^{38}, based on their having similar frequency values. Our experiment reproduces the same reported frequencies at the two common angles, but there is one (marked with an asterisk) not previously observed. The extreme similarity at both angles justifies using the previous results as a second verification of our band structure calculations. We also tracked the change in amplitude of the most prominent peaks with temperature in order to calculate the effective mass for field along [011] and [100]. Those data (with, where possible, a comparison to previous results and theoretical predictions) are in the SI (Supplementary Table II).
Figure 6c, d presents a comparison between theoretical and experimental frequencies identified by angular sweeps in this (c) and in the prior (d) study^{38}. In the Supercell Kspace Extremal Area Finder (SKEAF) calculations, the Fermi level was tuned over a wide range of energies and the best agreement was found when E_{F} was set to only 15 meV above the original density functional theory (DFT) value. The agreement in a 5meV window around this value was noticeably better than at any alternative energy setting.
We note that only the fundamental frequencies were included in the plot; as noticed both by us and in the prior work, there are many higher frequencies in FeP that can be attributed to harmonics or magnetic breakdown, and which SKEAF would therefore not predict. Similarly, predicted frequencies that were unlikely to be observed, either because the predicted effective mass was too large, they were very close in frequency to another band, or they existed over a narrow angular range, have been excluded. In (c) some of the bands have been rigidly shifted by up to 200 T, but this does not change the qualitative angular dependence. It is possible in both cases to identify similarities in the angular dependence of the theoretical and experimental frequencies, confirming in multiple dimensions general agreement between the theoretical and experimental FSs, with minimal adjustment needed. This supports our use of the paramagnetic FS as a reference point. The agreement seems overall to be much better with the predictions for electron (green lines) rather than hole (blue) bands.
Discussion
There are two distinct interesting aspects of the magnetotransport of FeP: (1) that it is very large and (2) that it follows a completely linear field dependence only for B ∥ caxis, and that it does so from very low field up to more than 30 T, an unprecedented range. These phenomena do not seem intertwined, since the MR is large and nonsaturating at all angles. Linear MR, meanwhile, occurs only at a specific orientation. However, studies of topological semimetals have shown that the two can have roots in the same physics. Electronhole compensation generally leads to a large, nonsaturating MR^{31}, and can occur when the valence and conduction bands touch close to E_{F}. These small pockets near the Fermi level can often have a linear dispersion, i.e., the touching points are topological Weyl or Dirac points. Cd_{3}As_{2}^{21} and TlBiSSe^{39} are two materials that combine large MR, linear MR, and topological band structure features.
Large MR is seen in many more MnPtype materials than linear MR. As noted, the MR of FeP is comparable to observations in MnP^{4} and CrP^{26}. While smaller, CoP (Supplementary Fig. 1) still has a sizable increase in comparison to CoAs. None of these compounds show any sign of MR saturation. In the semiclassical picture of MR, resistivity should stop increasing with field only at very high B, if at all, for compensated or nearly compensated materials, where electron and hole transport is balanced^{31}. Analyzed with a twoband model, Hall effect results for S1 show a two order of magnitude change in the electronhole mobility ratio toward parity starting at 50 K (Supplementary Fig. 3), roughly in the region of T*. The preservation of band structure features by the space group means other B31 pnictides will have a similar dispersion, providing the conditions for large MR in those that have low residual resistivities (generally, the phosphides).
There are not many cases of linear MR over such a wide range of field. Very few have as large of an increase, and of those none show linearity to as low of field. Recently, Ru_{2}Sn_{3} was reported to have a linear MR starting from low field at certain orientations^{40}. However, there is still a visible initial quadratic component not seen in FeP, meaning that there is a different process at work (the authors suspect a gap opening in the low carrier, semimetallic band structure). Bilayer graphene^{41} and silver chalcogenides^{42} can in fact exhibit such behavior beyond 60 T. But the weak temperature dependence and 2D nature of the graphene samples likewise point to a wholly different origin. The highT_{c} cuprates are another example^{43}, but the low field behavior is obscured by the superconducting state, so it is difficult the point to where normal state linear MR begins. The lack of linear ρ(T) in FeP also prohibits a quantum criticality explanation like that applied to the cuprates. Linear MR has been seen in transition metal pnictides^{21,44}, even at pulsed field in Cd_{3}As_{2}^{45}, but the explanation in that case is mobility fluctuations due to disorder. The very low residual resistivity of S1 rules this out, and we still see linearity (with a smaller MR) in a sample with much lower RRR (Supplementary Fig. 5), showing that the field dependence is preserved against disorder even when the magnitude is not. Weyl semimetal candidates PrAlSi and LaAlSi show quasilinear MR up to 9 T^{46}. Both show similar behavior, despite a difference in magnetism and the appearance of QOs. Analysis of the Hall data of PrAlSi shows that this can likely be attributed to compensated behavior and is not inherently topological.
The explanation given for linearity under pressure in CrAs (where MR roughly doubles up to 14 T) was based on the assumption that only a single, small FS pocket contributes to magnetoconductivity for B ∥ caxis. This is the Abrikosov quantum linear MR picture, which says such a pocket can very quickly be reduced to the lowest Landau level in field. The linear dispersion leads to a vanishing effective mass^{47}, producing a linear MR from low field. This theory cannot be applied to FeP, as our DFT calculations revealed the semiDirac point about 71 meV above the Fermi level (Fig. 5a) while based on the QO comparison, the Fermi level in the analyzed samples seems to be shifted up by 15 meV. This means that it is further away than the semiDirac point in CrP (found at −47 meV), which showed large but nonlinear MR^{14,26}, whereas the same point was <10 meV below E_{F} in CrAs at the pressure where MR is most linear.
It does not seem possible to explain linear MR in FeP by analogy to previous experiments. We look instead to a recent theoretical work, which stated that in systems with both topological band structure features and long range magnetic order, a positive linear MR can emerge from low field^{24} and be maintained in higher fields. This effect is the result of intraband scattering of the topological band, and is predicted to be comparable to and potentially even larger than the interband scattering contribution. A large amount of this intraband scattering would explain the linear MR in FeP, with more leeway for the location of the halflinear dispersion, and its absence in paramagnetic CrP^{26}, CoAs^{29}, and CoP (Supplementary Fig. 1). The more stringent requirements on the magnetic state explain why linear MR is not as widespread as large MR in the B31 class, in spite of crystallographic protection of topological points. The relevance of linear MR only appearing for field along the helimagnetic propagation direction is unclear, though it could also be that this effect appears at more orientations, but is obscured by other, larger contributions. Relating back to the intrascattering idea, the exact angle between applied field, magnetic moments, and topological dispersions could affect the ratio of linear contributions to others.
We note that the MR is minimized for field along the caxis, and still tends toward high field linearity at other angles, a possible sign of dominance of the semiDirac pointdriven MR at high field. Whether this explanation could also apply to CrAs is uncertain, as MR becomes closest to linear only after suppression of magnetic order^{14}. However, even at ambient pressure, the MR in CrAs is approximately linear by 5 T. The low field behavior changes from sublinear to roughly quadratic with higher pressure. This could also be an example of changing weight of multiple MR contributions with different pressure dependencies. Linear MR has been seen in the other two magnetic B31 compounds in certain field ranges: FeAs shows a quadraticlinear crossover around 10 T^{8}, while MnP can have quasilinear MR from below 2 T^{4}. However, interpretation of the latter data is complicated by the presence of metamagnetic transitions, which may alter the relationship between the semiDirac point and magnetic order.
The MR of FeP is large and nonsaturating at all angles, but with significant changes in magnitude and field dependence based on orientation. At most angles, this large MR has a superlinear dependence that straightens out by 35 T. However, with field along the caxis, MR is linear from very low field. We believe that this can result from the combination of ordered magnetism and topological band structure present in FeP, which produces an anisotropic linear response. A key finding is that large and linear MR in FeP have separate origins. Other materials in this family display large (e.g., CrP) or linear (CrAs) MR, but not both; in combining the two, FeP exhibits impressive low temperature, high field behavior. The field dependence is large, robust to disorder, and simple (lacking even QOs at low temperature), which could be very useful for future application. Given what has already been achieved with CrAs and MnP, the behavior of the magnetic state and linear MR under applied pressure is an interesting path for followup work.
Methods
Crystal growth
Samples were grown by the chemical vapor transport (CVT) method using I_{2} as the transport agent^{3,38,48,49,50}. For the CVT growth, a single zone horizontal tube furnace was used with the middle at about 900 °C, and the end about 200 °C cooler due to the natural gradient of the furnace. Fe and P powder along with I_{2} polycrystals were sealed under vacuum in a tube half the length of the furnace, arranged such that the material was initially at the hot end of the tube (the furnace center). Single crystals of FeP were found at the cold end of the furnace after 10–14 days, often with FeP powder; on occasion some would also be found at the hot end. The crystals grown with CVT were polyhedral with dimensions of roughly 0.5–3 mm in each direction, but nevertheless showed a clear preference for growing along the baxis, the shortest crystallographic axis. Xray diffraction (XRD) of ground single crystals showed single phase FeP with lattice parameters of a = 5.10 Å, b = 3.10 Å, and c = 5.79 Å, in line with previous reports^{23,38}. We also tried growing crystals with prereacted FeP powder and a similar technique, but the samples seemed to be of lower quality than those grown from the raw elements, as they had lower RRR and MR values.
Physical property measurements
Single crystals were aligned with a combination of singlecrystal XRD and Laue photography. Inhouse electrical transport measurements up to 14 T were taken in a Quantum Design Physical Properties Measurement System. Those to 31 and 35 T were made using two different resistive magnets at the NHMFL and a rotating sample platform, as were torque measurements done via piezoresistive cantilever.
DFT calculations
Firstprinciples calculations based on DFT were performed using the QUANTUM ESPRESSO package^{51,52}. We treated the exchange and correlation interaction within the generalized gradient approximation^{53}, and the ion–electron interaction with the projector augmentedwave pseudopotentials from the pslibrary database^{54,55}. The electron wave functions were expanded in a plane wave basis set with the cutoff of 50 Ry. The FeP structure was modeled by the orthorhombic unit cell with lattice constants set to those obtained via xray. The internal degrees of freedom were relaxed until the forces on each atom became smaller than 10^{−3} Ry/bohr. The Brillouin zone sampling at the level of DFT was performed following the Monkhorst–Pack scheme using a 12 × 16 × 10 kpoints grid. SOC was taken into account selfconsistently. The FSs were calculated on the interpolated mesh of 60 × 80 × 50 using the PAOFLOW code and visualized with FermiSurfer^{56,57}. The QO frequencies expected for the calculated FS were evaluated using the SKEAF code^{58}.
Data availability
All relevant data are available from the authors upon reasonable request.
References
Selte, K., Kjekshus, A. & Andresen, A. Magnetic structure and properties of FeAs. Acta Chem. Scand. 26, 3101–3113 (1972).
Kallel, A., Boller, H. & Bertaut, E. Helimagnetism in MnPtype compounds: MnP, FeP, CrAs and CrAs_{1−x}Sb_{x} mixed crystals. J. Phys. Chem. Solids 35, 1139–1152 (1974).
Westerstrandh, B., Lundgren, L., Gäfvert, U. & Carlsson, B. Magnetic susceptibility resistivity and thermal expansion measurements on FeP. Phys. Scripta 15, 276 (1977).
Takase, A. & Kasuya, T. High field magnetoresistance in MnP. J. Phys. Soc. Jpn. 49, 489–492 (1980).
Paglione, J. & Greene, R. L. Hightemperature superconductivity in ironbased materials. Nat. Phys. 6, 645 (2010).
Parker, D. & Mazin, I. I. Nonnesting spindensitywave antiferromagnetism in FeAs from first principles. Phys. Rev. B 83, 180403 (2011).
Rodriguez, E. E. et al. Noncollinear spindensitywave antiferromagnetism in FeAs. Phys. Rev. B 83, 134438 (2011).
Campbell, D. J. et al. Quantum oscillations in the anomalous spin density wave state of FeAs. Phys. Rev. B 96, 075120 (2017).
Cheng, J. & Luo, J. Pressureinduced superconductivity in CrAs and MnP. J. Phys.Condens. Mat. 29, 383003 (2017).
Matsuda, M. et al. Evolution of magnetic double helix and quantum criticality near a dome of superconductivity in CrAs. Phys. Rev. X 8, 031017 (2018).
Hirai, D., Takayama, T., Hashizume, D. & Takagi, H. Metalinsulator transition and superconductivity induced by Rh doping in the binary pnictides RuPn (Pn = P, As, Sb). Phys. Rev. B 85, 140509 (2012).
Lv, B. Q. et al. Experimental discovery of Weyl semimetal TaAs. Phys. Rev. X 5, 031013 (2015).
Shekhar, C. et al. Extremely large magnetoresistance and ultrahigh mobility in the topological Weyl semimetal candidate NbP. Nat. Phys. 11, 645–649 (2015).
Niu, Q. et al. Quasilinear quantum magnetoresistance in pressureinduced nonsymmorphic superconductor chromium arsenide. Nat. Comm. 8, 15358 (2017).
Kotegawa, H., Nakahara, S., Tou, H. & Sugawara, H. Superconductivity of 2.2 K under pressure in helimagnet CrAs. J. Phys. Soc. Jpn. 83, 093702 (2014).
Wu, W. et al. Superconductivity in the vicinity of antiferromagnetic order in CrAs. Nat. Comm. 5, 5508 (2014).
Cheng, J. G. et al. Pressure induced superconductivity on the border of magnetic order in MnP. Phys. Rev. Lett. 114, 117001 (2015).
Cuono, G. et al. Multiple band crossings and Fermi surface topology: role of double nonsymmorphic symmetries in MnPtype crystal structures. Phys. Rev. Mater. 3, 095004 (2019).
Liu, Z. et al. Superconductivity in WP single crystals. Phys. Rev. B 99, 184509 (2019).
Ali, M. N. et al. Correlation of crystal quality and extreme magnetoresistance of WTe_{2}. Europhys. Lett. 110, 67002 (2015).
Feng, J. et al. Large linear magnetoresistance in Dirac semimetal Cd_{3}As_{2} with Fermi surfaces close to the Dirac points. Phys. Rev. B 92, 081306 (2015).
Fauqué, B. et al. Magnetoresistance of semimetals: the case of antimony. Phys. Rev. Mater. 2, 114201 (2018).
Selte, K., Kjekshus, A. & Andresen, A. Structural and magnetic properties of FeP. Acta Chem. Scand. 26, 1276–1277 (1972).
Xiao, C. et al. Linear magnetoresistance induced by intrascattering semiclassics of Bloch electrons. Phys. Rev. B 101, 201410 (2020).
Segawa, K. & Ando, Y. Magnetic and transport properties of FeAs single crystals. J. Phys. Soc. Jpn. 78, 104720 (2009).
Niu, Q. et al. Nonsaturating large magnetoresistance in the high carrier density nonsymmorphic metal CrP. Phys. Rev. B 99, 125126 (2019).
Nigro, A. et al. Resistivity measurements unveil microscopic properties of CrAs. Europhys. Lett. 125, 57002 (2019).
Takase, A. & Kasuya, T. Temperature dependences of electrical resistivity in MnP. J. Phys. Soc. Jpn. 48, 430–434 (1980).
Campbell, D. J. et al. CoAs: The line of 3d demarcation. Phys. Rev. B 97, 174410 (2018).
Chen, X. H. et al. Correlation between the residual resistance ratio and magnetoresistance in mgb_{2}. Phys. Rev. B 65, 024502 (2001).
Pippard, A. B. Magnetoresistance in Metals, Vol. 2 (Cambridge University Press, 1989).
Wang, Y. L. et al. Origin of the turnon temperature behavior in WTe_{2}. Phys. Rev. B 92, 180402 (2015).
Jo, N. H. et al. Extremely large magnetoresistance and Kohler’s rule in PdSn_{4}: a complete study of thermodynamic, transport, and bandstructure properties. Phys. Rev. B 96, 165145 (2017).
Chernyavskii, I. O. et al. Incommensurate magnet iron monophosphide FeP: Crystal growth and characterization. Phys. Rev. Mater. 4, 083403 (2020).
Young, S. M. & Kane, C. L. Dirac semimetals in two dimensions. Phys. Rev. Lett. 115, 126803 (2015).
Cuono, G. et al. Spin–orbit coupling effects on the electronic properties of the pressureinduced superconductor CrAs. Eur. Phys. J. Spec. Top. 228, 631–641 (2019).
Autieri, C. & Noce, C. First principles study of structural, magnetic and electronic properties of CrAs. Philos. Mag. 97, 3276–3295 (2017).
Nozue, T. et al. de Haasvan Alphen effect of FeP in double helical magnetic state. J. Phys. Soc. Jpn. 70, 192–198 (2001).
Novak, M., Sasaki, S., Segawa, K. & Ando, Y. Large linear magnetoresistance in the Dirac semimetal TlBiSSe. Phys. Rev. B 91, 041203 (2015).
Wu, B. et al. Linear nonsaturating magnetoresistance in the Nowotny chimney ladder compound Ru_{2}Sn_{3}. Phys. Rev. B 101, 205123 (2020).
Kisslinger, F. et al. Linear magnetoresistance in mosaiclike bilayer graphene. Nat. Phys. 11, 650–653 (2015).
Husmann, A. et al. Megagauss sensors. Nature 417, 421–424 (2002).
GiraldoGallo, P. et al. Scaleinvariant magnetoresistance in a cuprate superconductor. Science 361, 479–481 (2018).
Xu, R. et al. Large magnetoresistance in nonmagnetic silver chalcogenides. Nature 390, 57–60 (1997).
Narayanan, A. et al. Linear magnetoresistance caused by mobility fluctuations in ndoped Cd_{3}As_{2}. Phys. Rev. Lett. 114, 117201 (2015).
Lyu, M. et al. Nonsaturating magnetoresistance, anomalous hall effect, and magnetic quantum oscillations in the ferromagnetic semimetal PrAlSi. Phys. Rev. B 102, 085143 (2020).
Abrikosov, A. A. Quantum magnetoresistance. Phys. Rev. B 58, 2788–2794 (1998).
Bellavance, D., Vlasse, M., Morris, B. & Wold, A. Preparation and properties of iron monophosphide. J. Solid State Chem. 1, 82–87 (1969).
Richardson, M. W. & Noläng, B. I. Predicting the rate of chemical transport using the flux function method. J. Cryst. Growth 42, 90–97 (1977).
Binnewies, M., Glaum, R., Schmidt, M. & Schmidt, P. Chemische Transportreaktionen (Walter de Gruyter, 2011).
Giannozzi, P. et al. QUANTUM ESPRESSO: a modular and opensource software project for quantum simulations of materials. J. Phys.: Condens. Matter 21, 395502 (2009).
Giannozzi, P. et al. Advanced capabilities for materials modelling with QUANTUM ESPRESSO. J. Phys.: Condens. Matter 29, 465901–4659031 (2017).
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmentedwave method. Phys. Rev. B 59, 1758–1775 (1999).
Corso, A. D. Pseudopotentials periodic table: from H to Pu. Comput. Mater. Sci. 95, 337–350 (2014).
Buongiorno Nardelli, M. et al. PAOFLOW: a utility to construct and operate on ab initio Hamiltonians from the projections of electronic wavefunctions on atomic orbital bases, including characterization of topological materials. Comput. Mater. Sci. 143, 462–472 (2018).
Kawamura, M. FermiSurfer: Fermisurface viewer providing multiple representation schemes. Comput. Phys. Commun. 239, 197–203 (2019).
Julian, S. Numerical extraction of de Haas–van Alphen frequencies from calculated band energies. Comput. Phys. Commun. 183, 324–332 (2012).
Momma, K. & Izumi, F. VESTA3 for threedimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr. 44, 1272–1276 (2011).
Acknowledgements
This work was supported by the National Science Foundation Division of Materials Research award no. DMR1905891 and the Gordon and Betty Moore Foundation’s EPiQS Initiative through Grant No. GBMF9071. A portion of this work was performed at the National High Magnetic Field Laboratory, which is supported by National Science Foundation Cooperative Agreement nos. DMR1157490 and 1644779 as well as the state of Florida. D.J.C. acknowledges the support of the Anne G. Wylie Dissertation Fellowship. We also acknowledge the support of the Maryland NanoCenter and its FabLab. C.A. is supported by the Foundation for Polish Science through the International Research Agendas program cofinanced by the European Union within the Smart Growth Operational Programme. The computational resources at the Texas Advanced Computing Center of the University of Texas, Austin are gratefully acknowledged.
Author information
Authors and Affiliations
Contributions
D.J.C. and J.C. grew the samples and conducted zero and low magnetic field measurements. D.J.C., K.W., B.W., Y.S.E., P.N., and D.G. made the high field measurements. J.S., C.A., and L.W. performed theoretical calculations. M.B.N. and J.P. supervised the project. D.J.C., J.S., and J.P. wrote the manuscript with input from all authors.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Campbell, D.J., Collini, J., Sławińska, J. et al. Topologically driven linear magnetoresistance in helimagnetic FeP. npj Quantum Mater. 6, 38 (2021). https://doi.org/10.1038/s41535021003372
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41535021003372
This article is cited by

Anisotropic positive linear and sublinear magnetoresistivity in the cubic typeII Dirac metal Pd3In7
npj Quantum Materials (2023)

Robust threedimensional typeII Dirac semimetal state in SrAgBi
npj Quantum Materials (2023)