Abstract
The interaction of (quasi)particles with a periodic potential arises in various domains of science and engineering, such as solidstate physics, chemical physics, and communication theory. An attractive test ground to investigate this interaction is represented by superconductors with artificial pinning sites, where magnetic flux quanta (Abrikosov vortices) interact with the pinning potential U(r) = U(r + R) induced by a nanostructure. At a combination of microwave and dc currents, fluxons act as mobile probes of U(r): The ac component shakes the fluxons in the vicinity of their equilibrium points which are unequivocally determined by the local pinning force counterbalanced by the Lorentz force induced by the dc current, linked to the curvature of U(r) which can then be used for a successful fitting of the voltage responses. A good correlation of the deduced dependences U(r) with the cross sections of the nanostructures points to that pinning is primarily caused by vortex length reduction. Our findings pave a new route to a nondestructive evaluation of periodic pinning in superconductor thin films. The approach should also apply to a broad class of systems whose evolution in time can be described by the coherent motion of (quasi)particles in a periodic potential.
Introduction
The interaction of (quasi)particles with a periodic potential arises in various domains of science and engineering, such as solidstate physics, chemical physics, and communication theory. Exemplary systems include Josephson junctions^{1}, superionic conductors^{2}, ring laser gyroscopes^{3}, the dynamics of spin^{4} and chargedensity^{5} waves, phaselocking loops^{6} in radioengineering, the motion of domain walls^{7}, the magnetization dynamics of interacting spins^{8} and the diffusion of colloidal particles in periodic structures^{9}. An interesting analogy can be found in superconductivity: Type II superconductors with periodic pinning sites substantiate an attractive test ground to investigate this interaction, thus allowing one to study dynamic properties of the vortex system and scrutinize its forcevelocity (currentvoltage) response.
The symmetry, shape and intensity of the pinning potential strongly affect the dynamics of Abrikosov vortices in superconductors, as reflected in both, the dc resistance and the microwave (mw) loss^{10}. In this regard, tailoring periodic pinning sites has been revealed to be most efficient for controlling the resistive response of superconductors^{11,12,13}, dynamic modelocking^{14}, stimulation of superconductivity at microwave frequencies^{15}, enhancing speed limits for the dynamics of Abrikosov vortices^{16,17,18,19} as well as enabling functionality of such fluxonic devices as rectifiers^{20}, transistors^{21}, and highfrequency filters^{22}. The availability of highresolution nanofabrication tools has advanced the use of more and more sophisticated, asymmetric pinning geometries (vortex ratchets)^{11,20}. For instance, in a vortex ratchet device (diode), the pinning site array may consist of elements of different sizes and shapes, such as triangles^{20}, grading circles^{23}, and arrowshaped wedged cages^{24}. The coordinate dependence of the resulting pinning potential U(x, y) in these is rather complex, which is why for a quantitative interpretation of data an assumption is usually made for U(x, y) a priori. In the case of a simpler washboard pinning potential (WPP)^{25}, U(x, y) ≡ U(x) = U(x + a), where a is the period, an asymmetry can be induced by predefining the steepness of the left and right slopes of linearextended pinning “sites” differently. The respective 3D washboard ratchet nanostructures can be realized, e.g., by milling of uniaxial nanogrooves by a focused ion beam (FIB) into the film^{26}, or via ferromagnetic decoration of the films by focused electron beam induced deposition (FEBID)^{27}. Since the geometrical appearance of an asymmetric washboard nanostructure does not reveal the asymmetry and the shape of the resulting WPP, nondestructive approaches for the determination of U(x, y) are needed.
The ideas for the determination of U(x, y) from experiment have been attracting attention of researchers for half a century^{28,29,30,31,32}. Early approaches for the determination of the distribution, density and strength of pinning sites included proposals to quantify the pinning force from measurements implying a small ripple magnetic field superimposed on a larger dc magnetic field^{28} and from measurements with a small ac current^{29}. Recently, elastic properties of individual vortices were investigated by magnetic force microscopy^{30}. Further, scanning SQUID microscopy has been used to probe the dynamics and pinning of single vortices under combined dc and small ac drives, and the dependence of the elementary pinning force of multiple defects on the vortex displacement has been measured^{31}. It has also been shown^{32} that vortices respond to local mechanical stress applied in the vicinity of a vortex thus allowing one to manipulate individual vortices without magnetic field or current. In all these experiments the samples contained randomly distributed pinning sites and either the vortex response in a noncoherent regime^{28,29} or that of an individual vortex^{30,31,32} was probed.
An intriguing situation ensues in a washboard pinning landscape at a particular value of the magnetic field H _{ m } when each row of vortices is pinned at the bottom of the linearlyextended pinning sites (nanogrooves) and there are neither vacant nanogrooves nor vortices pinned between them. This field H _{ m } is the fundamental matching field at which the vortex lattice is commensurate with the pinning landscape. Computer simulations revealed^{33} that in this case the vortex lattice has a crystalline structure, the effective intervortex interaction is cancelled, and each vortex experiences the same pinning force. An important consequence emerges from this: In the coherent regime, the moving ensemble of vortices in a WPP can be regarded as a moving vortex crystal, thus allowing one to interpret data employing singlevortex mechanistic models^{34,35,36,37,38}.
Here, in contradistinction to previous works^{28,29,30,31,32}, we use the coherent vortex dynamics at the fundamental matching field to examine a theoretical mechanistic approach^{39} for the determination of the coordinate dependence of a periodic pinning potential in superconductors under combined dc and mw current stimuli. Specifically, we evaluate the dc currentinduced reduction of the depinning frequency in nanopatterned Nb films with different groove slope asymmetries and determine the dependences U(x) from ensembleintegrated microwave power absorption data. Further, using the pinning asymmetry parameters thus deduced, we augment the validity of the presented approach by a good fitting of modelocking steps in the electrical voltage response in the presence of an ac drive to expressions derived within the framework of a stochastic model^{40} of anisotropic pinning.
Results
We studied the vortexstate resistive response of Nb microstrips by combined microwave transmission spectroscopy and electrical voltage measurements, Fig. 1. The 500 μm long and 150 μm wide microstrips were patterned with arrays of 500 nmspaced nanogrooves milled by FIB. The nanogrooves were parallel to the direction of the transport current density j which, in a perpendicular magnetic field B = μ _{0} H, exerted a transverse Lorentz force F _{ L } = j × B on vortices leading to their dissipative motion in the direction perpendicular to the grooves. The nanogroove arrays induced a periodic pinning potential of the washboard type U(x, y) ≡ U(x) = U(x + a). Sample S was patterned with nanogrooves having a symmetric crosssection, sample A1 with nanogrooves having a weak asymmetry of the groove slopes, and the control sample A2 with nanogrooves having a strong asymmetry of the slopes. Atomic force microscope images of the nanogrooves are presented further in the Discussion section. The asymmetric grooves are oriented in such a way that a positive dc bias makes the vortices to probe the gentle slope of the WPP: The vortices are shaken at the bottom of the pinning potential, which is gradually shifted in the negative x direction as the dc bias magnitude is increased. Samples S, A1, and A2 are 40 nm, 56 nm and 70 nm thick and have a superconducting transition temperature T _{ c } of 8.66 K, 8.85 K, and 8.94 K, respectively. The upper critical field for all samples at zero temperature is estimated as H _{ c2}(0) = H _{ c2}(T)/[1 − (T/T _{ c })^{2}] ≈ 1 T corresponding to a coherence length of ξ(0) = [Φ_{0}/2πH _{ c2}(0)]^{1/2} ≈ 17 nm. The diameter of the vortex core \(\simeq \,2\xi \) at T = 0.3T _{ c } is by a factor of 6 smaller than the full width at half depth of the grooves in samples A1 and A2, and by a factor of 3 for sample S.
“Ratchet window” in microwave power absorption
We investigated the frequency dependence of the mw power absorbed by vortices in Nb microstrips with symmetric and asymmetric nanogrooves at the fundamental matching field H _{ m } = 7.2 mT and T/T _{ c } = 0.3 for a series of dc bias values of both polarities. The data for the absorbed mw power were acquired in terms of the relative change of the absolute value of the forward transmission coefficient ΔS _{21}(f) ≡ S _{21} − S _{21ref}, where S _{21ref} is the reference mw loss in the transmission line (all cables, connectors etc.) and ΔS _{21} is a measure for the mw loss due to vortex motion in the sample under study. Here all quantities are expressed in dB since it is the ratio of the transmission coefficients with and without the microstrip device which is a quantity of our interest.
In what follows we focus on the data acquired at a mw excitation power of P = −20 dBm, leaving an analysis of nonlinear effects observed at higher power levels beyond the scope of this work. In the order of magnitude, the amplitude of the vortex displacement at P = −20 dBm at a frequency of 1 GHz can be estimated as 10 nm on the basis of complementary fluctuation spectroscopy measurements on a series of narrow Nb microstrips fabricated from asgrown flat and grooved Nb films. The data at P = −30 dBm lead to essentially the same results but the data are more noisy. Contrarily, further effects emerge at high excitation power levels of P = −6 dBm and −3 dBm in the nonlinear regime which should be analyzed with an indepth analysis of overheating effects and the associated with them quasiparticle escape from the vortex cores^{15}. In this way, in the absence of the dc current, the acdriven vortex dynamics is substantially in the linear regime (vortex displacements are small) and it transits into the dcinduced nonlinear regime as the dc bias is increasing so that the vortices are shaken while overcoming the WPP barriers.
Figure 2(a,b) presents the raw data ΔS _{21}(f) for sample A1, while the respective data for samples S and A2 are reported in Supplementary. The arrangement of vortices at 7.2 mT in the pinning nanolandscape is shown in Fig. 2(c) for the assumed triangular vortex lattice with lattice parameter \({a}_{{\rm{\Delta }}}={(2{{\rm{\Phi }}}_{0}/B\sqrt{3})}^{\mathrm{1/2}}\) and the matching condition \({a}_{{\rm{\Delta }}}=2a/\sqrt{3}\). In Fig. 2 ΔS _{21}(f) has a smooth crossover from the weaklydissipative regime at low frequencies (ΔS _{21}(f) = 0 corresponds to the minimal mw loss) to the stronglydissipative regime at high frequencies (ΔS _{21}(f) ≈ −6.3 dB is the maximum mw excess loss due to vortices). The crossover takes place at the depinning frequency f _{ d } determined at the point where the phase difference between the viscous and pinning forces amounts to π/2^{39} and corresponds to about −3 dB excess loss level in our measurements, which is indicated as a determination criterion for f _{ d } in Fig. 2(a). At low frequencies the vortices visit many potential wells (delocalized regime), whereas at high frequencies the vortices are shaken at the groove bottoms (localized regime). We note that the pinning force \({F}_{p}=\nabla U(x)\) attains a maximum at the groove slopes, whereas F _{ p } ≈ 0 in the vicinity of the groove bottoms where U(x) is nearly constant. Accordingly, overcoming of the pinning barriers by vortices at low frequencies causes the pinning force to dominate the viscous force and the response is weakly dissipative, whereas in the highfrequency regime the viscous force dominates the pinning force, thus leading to a strongly dissipative response^{34}. The zerobias depinning frequency f _{0} ≡ f _{ d }(j = 0) of samples S, A1 and A2 at T/T _{ c } = 0.3 and H _{ m } = 7.2 mT amounts to 5.72, 3.95, and 3.12 GHz, respectively.
For both dc bias polarities f _{ d } shifts towards low frequencies with increasing dc bias value, Fig. 2(a,b). Under dc polarity reversal, the magnitudes of the f _{ d }(j) shifts noticeably differ for sample A1, even more so for sample A2, but nearly coincide for sample S, see Supplementary. The reduction of the depinning frequency upon increasing the dc bias can be understood as a consequence of the effective lowering of one of the pinning potential wells due to its tilt caused by the dc bias^{41}. The mechanistic consideration of a vortex as a particle leads to the conclusion that during an ac semiperiod, while the dctilted pinning potential well is broadening^{41}, with increasing frequency f the vortex has no longer time ∝1/f to reach the groove slopes where the pinning forces dominate. Accordingly, the response becomes stronger dissipative already at lower frequencies, as compared with the zerobias curve.
The difference in the magnitudes of the currentinduced shifts of f _{ d } of samples A1 and A2 is best seen in Fig. 3 displaying contour plots of the difference signal \({\rm{\Delta }}{S}_{21}^{\pm }={\rm{\Delta }}{S}_{21}({j}^{+}){\rm{\Delta }}{S}_{21}({j}^{})\) which serves as a measure of the ratchet properties of the samples. The bright areas in the contour plots in Fig. 3(b,c) represent the “ratchet windows” in the microwave power absorption, i.e. the frequency  current range where the system exhibits rectifying properties, i.e. works as a diode. As expected, sample S with symmetric grooves does not exhibit ratchet properties, as the absolute value of the depinning current density does not change under dc bias polarity reversal in this sample. The “ratchet window” of samples A1 and A2 is bound by the gentleslope \({j}_{d}^{+}\) and the steepslope \({j}_{d}^{}\) depinning current densities (extrapolated to f → 0 corresponding to the dc regime), as well as by the value of the zerobias depinning frequency f _{ d }(j = 0), as indicated in Fig. 3. The depinning current density j _{ d } amounts to 0.75 MA/cm^{2} for both CVC branches of sample S, 0.91 MA/cm^{2} and 0.7 MA/cm^{2} for the steepslope and the gentleslope direction of the vortex motion in sample A1, and 1.25 MA/cm^{2} and 0.52 MA/cm^{2} for the steepslope and the gentleslope direction of the vortex motion in sample A2, respectively. The depinning current densities \({j}_{d}^{\pm }\) are determined by the 10 μV/cm electric field strength criterion from the currentvoltage curves (CVCs) discussed next.
Modelocking steps in the currentvoltage curves
Figure 4(a–c) displays the dc electrical voltage as a function of the normalized dc density and the ac amplitude at T = 0.98T _{ c } and the fundamental matching field H _{ m } = 7.2 mT. In all panels f = 0.3f _{ d }, with f _{ d } amounting to 444, 306, and 242 MHz, for samples S, A1, and A2, respectively. In the absence of microwave excitation, the CVC of Sample S in Fig. 4(a) is symmetric, while the CVCs of Sample A1 and A2 demonstrate two different absolute values of the depinning current density \({j}_{d}^{+} > {j}_{d}^{}\) for the positive and the negative branch. The addition of the microwave stimulus leads to the appearance of Shapiro steps in the CVC. The steps occur at voltages^{42} V = nV _{0} ≡ nNΦ_{0} f, where n is an integer, N is the number of vortex rows between the voltage leads, f is the microwave frequency, and Φ_{0} = 2.07 × 10^{−15} Vs is the magnetic flux quantum. The steps in the CVCs, which are best seen in the conventional currentvoltage representation in Fig. 5(a), arise when the hopping distance of Abrikosov vortices during one ac halfwave coincides with one or a multiple of the nanostructure period. These interference steps are a fingerprint of the coherent vortex dynamics. They remain visible in the field range from 7 to 7.5 mT and disappear as H is tuned further away from the fundamental matching field H _{ m } = 7.2 mT.
We note that temperatures close to T _{ c } are fortunate for the observation of Shapiro steps which are best seen in the flux flow regime at j > j _{ d }. This is because of the depinning current decreasing with increasing temperature and the flux flow setting on at smaller dc densities j _{ d } < j < j ^{*} thus allowing one to operate in an extended current range where overheating effects are negligible. Here j ^{*} is the current density corresponding to an abrupt transition of the sample into the normal state due to the LarkinOvchinnikov instability^{16,18,19,43,44,45,46,47,48}.
In Fig. 4(a–c) one sees that j _{ d } nearly linearly decreases with increasing microwave amplitude and at some point the lines j _{ d }(j ^{mw}) for different polarities cross, thus allowing one to deduce a dc bias value for which an effective “symmetrization” of the WPP is predicted^{40}: At this dc bias value the internal asymmetry of the pinning potential is effectively compensated by the extrinsic asymmetry of the WPP caused by the dc bias. Here we refer to Fig. 2 in Supplementary for a sketch. Quantitatively, these dc values characterize the loading capability of the ratchet^{49} and amount to j/j _{ d } ≈ 0.1 for sample A1 and j/j _{ d } ≈ 0.25 for sample A2. Obviously, the lines \({j}_{d}^{\pm }({j}^{mw})\) cross at j/j _{ d } ≈ 0 for sample S as this sample has symmetric grooves. As the ac amplitude increases, the dc voltage for all samples exhibits modelocking peculiarities (“fringes”). Up to four lowestorder fringes for sample A2 are exemplified in Fig. 5(b) for a series of subdepinning dc biases of both polarities. Both, Shapiro steps and modelocking fringes were theoretically predicted earlier for symmetric^{38} and asymmetric^{40} WPPs. To examine whether the voltage responses can be described theoretically, we determine the coordinate dependence U(x) of the pinning potential from the microwave power absorption and use the deduced expressions for U(x) for modeling of the voltage responses in Fig. 4(d–f), as detailed next.
Discussion
Determination of the coordinate dependence of the pinning potential
Figure 6(a–c) displays the dependences f _{ d }/f _{0} versus j/j _{ d } deduced from the raw experimental data ΔS _{21}(f), as shown in Fig. 2 for a series of dc bias values of both polarities for all samples. The data acquired for positive dc biases are shown by solid red symbols, while those for negative dc biases by open blue ones. The relative uncertainty in the determination of the depinning frequency does not exceed 7% at j → j _{ d } and it is smaller than 3% at dc current densities j < 0.8j _{ d }. This is indicated by the error bars in Fig. 6(a–c), which become smaller than the symbol size at j < 0.8j _{ d }. The larger error in the determination of f _{ d } at j → j _{ d } is associated with a smeared functional shape of the curves ΔS _{21}(f) in Fig. 2 upon vanish of the WPP barriers due to their tilt by the dc current.
To interpret the dcinduced reduction of the depinning frequency f _{ d }/f _{0} in Fig. 6(a–c) as a function of the dc current density j/j _{ d } at 0.3T _{ c } for the coherent vortex at the fundamental matching field H _{ m } = 7.2 mT, we employ a singlevortex theoretical approach outlined in Supplementary. This approach is based on the GittlemanRosenblum model^{34} generalized for the presence of a dc current bias^{39,50,51}. The procedure of determination of the coordinate dependence of the pinning potential from the microwave power absorbed by vortices requires to approximate the reduction of the depinning frequency for both dc polarities by some expression. The experimental data for all our samples fit to expressions of the universal “meanfield” type
with the exponents k/l and m/n labeled at the respective fit curves in Fig. 6(a–c). For consistency, we also plot the respective fit curves in the contour plots in Fig. 3 and note that they nicely encage the “ratchet windows” quantitatively.
The employment of the procedure outlined in Supplementary to the fit curves in Fig. 6(a–c) results in the U(x) curves shown in Fig. 6(g–i) by solid lines. The curves U(x) in Fig. 6(g–i) are plotted for one period of the pinning potential scaled to its depth U _{0} and period a. The line thickness reflects the uncertainty in the determination of U(x), which does not exceed 10% and 5% on the dome and at the bottom of the potential, respectively. Analytically, the dependences U(x) deduced from the mw data can be approximated by expressions of the general form
with e being the asymmetry parameter and amounting to 0, 0.13 and 0.56 for samples S, A1, and A2, respectively.
Figure 6(g–i) displays the crosssections of the nanogrooves measured by atomic force microscopy (AFM) in comparison with the dependences U(x) deduced from the mw data and the respective approximations by Eq. (2). Obviously, the deduced coordinate dependence of the pinning potential U(x) in Sample A1 corresponds quite well to the coordinate dependence of the film thickness d(x) modulated by the nanogrooves, as measured by AFM. The AFM profile for Sample A2 can also be fitted satisfactorily to the same expression as U(x). At the same time, U(x) for sample S reproduces the AFM profile qualitatively, but does not follow it quantitatively. The modulated thickness d(x) in Sample S can be fitted to
The stronger discrepancy between the dependences U(x) and d(x) in sample S may be caused by a factor of about two smaller groove width in Sample S as compared to Samples A1 and A2. In all, we draw the conclusion that the employed procedure^{39} is not only sensitive to the asymmetry of the pinning potential, but it also reveals a correlation between U(x) and d(x) thus indicating that pinning in the two thicker samples is primarily caused by vortex length reduction. This can be understood as the vortex energy ε(x) is proportional to its length L(x) determined by the film thickness ε(x) ∝ L(x) = d(x).
Modeling of electrical resistance responses
To augment the validity of the employed approach for the determinations of the coordinate dependences U(x) of the pinning potential from the microwave power absorption at T = 0.3T _{ c } we employ a stochastic model of uniaxial periodic pinning^{37,38,40} to fit the peculiarities observed in electrical voltage responses at noticeably higher temperatures T = 0.98T _{ c }. The theoretical treatment relies upon the Langevin equation for a vortex moving with velocity v in a magnetic field B = z B, where B ≡ B, and z is the unit vector in the z direction
where F _{ L } = (Φ_{0}/c)j × z is the Lorentz force, Φ_{0} is the magnetic flux quantum, and c is the speed of light. In Eq. (4) j ≡ j(t) = j + j ^{mw} cos ωt, where j and j ^{mw} are the dc and ac current density amplitudes and ω is the angular frequency. \({{\bf{F}}}_{p}=\nabla {U}_{p}(x)\) is the anisotropic pinning force, where U _{ p }(x) is a ratchet WPP. F _{ th } is the thermal fluctuation force represented by Gaussian white noise and η is the vortex viscosity.
The ratchet WPP is modeled by
where k = 2π/a with a being the period and U _{ p } the depth of the WPP. In Eq. (5) e is the asymmetry parameter allowing for tuning the asymmetry strength. It is this parameter which must be determined from experiment.
The asymmetry parameter deduced from the experimental data amounts to e = 0 for Sample S, e = 0.13 for Sample A1, and e = 0.56 for Sample A2. These values of e are used in the simulations of the electric field response given by^{40}
Here ν is the (j, j ^{mw}, f, t)dependent effective nonlinear mobility of the vortex under the influence of the Lorentz force and it is expressed in terms of matrix continued fractions^{40}. In Eq. (6), the electric field is scaled to the electric field at the first Shapiro step, while the dc density j and the mw density amplitude j ^{mw} to the depinning current density, frequency f to the depinning frequency, coordinate x to the WPP period, and temperature t to the WPP depth.
We use the meansquare parameter \({j}_{d}=\sqrt{{j}_{d}^{+}{j}_{d}^{}}\) for the presentation of the data in dimensionless form allowing for a direct comparison of experiment with theory in Fig. 4. The voltage responses are calculated by Eq. (6) for a series of values of the asymmetry parameter e at the reduced temperature t = T/U _{ p } = 0.002 which corresponds to the pinning activation energy U _{ p } of about 5000 K^{52} and the temperature T = 0.98T _{ c } at which the experiment is conducted.
We now proceed to a deeper analysis of the modelocking fringes reported in Fig. 5(b). The appearance of these voltage peaks (for the positive dc polarity) and dips (for the negative dc polarity) can be explained by higherorder ratchet effects (labeled with “1” to “4” in the plot) as follows: We first consider the curve \(V/{V}_{0}({j}^{mw}/{j}_{d}^{mw})\) in Fig. 5(b) at j/j _{ d } = 0.05, that is in the limit of very small dc biases. In the conventional ratchet effect resulting in the appearance of the “1” voltage peak, at a given ac amplitude \({j}^{mw}\mathop{ > }\limits_{ \tilde {}}{j}_{d\,{\rm{gentle}}}^{mw}\), a vortex overcomes the gentle slope of the WPP during one half of the ac period. However, the amplitude j ^{mw} is yet smaller than the strongslope depinning current density and the vortex can not return into its original WPP well. This results in a net motion of the vortex by one WPP period and the associated rectified voltage. As the ac amplitude increases, the vortex can overcome the steep barrier of the WPP and the net motion disappears thus resulting in an almost full suppression of the rectified voltage. With a further increase of the ac amplitude the vortex can consequentially overcome two gentle barriers of WPP, and two options appear for its backward motion depending on the strength of the WPP asymmetry and the ac frequency, please refer to Fig. 2 of Supplementary: (i) The vortex can only overcome one barrier in the backward direction, and after one ac period it appears in the WPP well in one period away from the original WPP well. (ii) The vortex can not overcome the barrier in the backward direction and it remains in the WPP well in two periods away from the original WPP well. Obviously, with a further increase of the ac amplitude the “splitting” of ways increases for the vortex in what WPP well it will appear after one ac period and this leads to a smearing of the higherorder ratchet peaks in the dc voltage.
The curve at j = 0.25 in Fig. 5(b) corresponds to the particular interesting case of effective compensation of the intrinsic WPP asymmetry due to the difference in the steepness of the groove slopes by the external asymmetry of the potential induced by the dc bias. In consequence of this effective “compensation of the anisotropies” the “floor” of the rectified voltage can be adjusted to almost zero which is why the loading capability of the ratchet amounts to about j/j _{ d } = 0.25 for sample A2. In all, the simulation results allow us not only to explain the main features observed in the experiment, but we find a very good quantitative agreement between the experimental data and theoretical modeling.
In conclusion, we have presented an approach allowing for the determination of a periodic pinning potential from the microwave power absorbed by vortices under dc bias reversal in superconductors with periodic pinning at low temperatures. For thicker films, the deduced coordinate dependences of the washboard pinning potentials U(x) largely mimic the coordinate dependences of the film thickness, as visible from crosssectional AFM line scans. The presented procedure allowed us to directly determine the asymmetry of the pinning potential and to further use it for modeling of voltage responses at T = 0.98T _{ c }. Our findings pave a new route to the nondestructive evaluation of periodic pinning in superconductor thin films. Moreover, the reported approach should also apply to a broad class of systems whose evolution in time can be described by the coherent motion of (quasi)particles in a periodic potential.
Methods
Film growth and characterization
The 150 × 500 μm^{2} microstrips were fabricated by photolithography and Ar etching from epitaxial (110) Nb films on acut sapphire substrates. The films were grown by dc magnetron sputtering in a setup with a base pressure in the 10^{−8} mbar range. In the sputtering process the substrate temperature was 850 °C, the Ar pressure 4 × 10^{−3} mbar, and the growth rate was about 1 nm/s. Xray diffraction measurements revealed the (110) orientation of the films^{53}. The epitaxy of the films has been confirmed by reflection highenergy electron diffraction. The asgrown films have a smooth surface with an rms surface roughness of less than 0.5 nm, as deduced from AFM scans in the range 1 μm × 1 μm. The background pinning in the epitaxial (110) Nb films is very weak as compared to the anisotropic pinning induced by nanogrooves milled by FIB. The parameters of the samples are compiled in Table 1.
Fabrication of nanogrooves
Patterning of the samples was done in a highresolution dualbeam scanning electron microscope (FEI, Nova Nanolab 600). In the patterning process, the asymmetry of the groove slopes was achieved by defining the grooves in the FIB bitmap file for sample S as a single line for the beam to pass, whereas a stepwise increasing number of FIB beam passes was assigned to each groove defined as a 3step and a 5step “stair” for samples A1 and A2, respectively. Due to blurring effects, the symmetric grooves in Sample S have rounded corners while smoothed straight slopes resulted instead of the “stairs” in samples A1 and A2. For all samples the beam parameters were 30 kV/50 pA, 1 μs dwell time and 50 nm pitch. The grooves are parallel to the microstrip edges with a misalignment of less than 0.2°. The microstrip width is an integer multiple number (N = 300) of the nanopattern period to prevent possible ratchet effects due to the edge barrier asymmetry^{13}.
Atomic force microscopy
A Nanosurf easyScan 2 atomic force microscope (AFM) under ambient conditions in noncontact, dynamic force mode was used. The cantilever tip was shaped like a polygonbased pyramid, with a tip radius of less than 7 nm (Nanosensors PPPNCLR). Convolution effects due to the finite tip radius can be neglected, as is corroborated by the invariance of the AFM images taken with the cantilever scanning at different angles with respect to the grooves. The crosssectional AFM profiles are the results of averaging of 250 line scans acquired for a scanning field of 500 × 500 nm^{2}.
Microwave spectroscopy
Combined broadband mw and dc electrical measurements were done in a ^{4}He cryostat with magnetic field H directed perpendicular to the film surface. A custommade cryogenic sample probe with coaxial cables was employed^{54}. The mw signal was generated and analyzed by an Agilent E5071C vector network analyzer (VNA). The mw and dc signals were superimposed and uncoupled by using two biastees mounted at the VNA ports. The VNA operated in the frequency sweep mode, with 1548 frequency points scanned with an exponentially growing increment between 300 KHz and 14 GHz. For all frequencies f, the mw excitation power at the sample was P = −20 dBm (10 μW) kept by the VNA in accordance with the presaved calibration data for S _{21ref }(f, T). At small fields, as investigated by us, S _{21ref } does not depend on the magnetic field value H.
Modeling of voltage responses
The infinite matrix continued fractions for the calculation of the voltage responses by Eq. (29) of ref.^{40} were approximated by matrix continued fractions of finite order. This has been done by putting Q _{ m } = 0 at some m = M, whereas the dimension of the submatrices Q _{ m } and the vectors C _{ m } was confined to some finite number K. Both M and K depend on temperature and mw density and on the number of harmonics to be taken into account. These numbers were chosen as K = 51 and M = 500 for the reliable calculation of the components \({F}_{k}^{1}(\omega )\) for up to 10 harmonics, for j ^{mw} up to 5, and for t ≡ T/U _{0} = 0.002, respectively. This ensures a calculation accuracy of at least three digits. The Hall effect was neglected in the calculations.
References
Barone, A. & Patterno, G. Physics and Applications of the Josephson Effect (John Wiley & Sons, New York, 1982).
Fulde, P., Pietronero, L., Schneider, W. R. & Strässler, S. Problem of brownian motion in a periodic potential. Phys. Rev. Lett. 35, 1776–1779, https://doi.org/10.1103/PhysRevLett.35.1776 (1975).
Chow, W. W. et al. The ring laser gyro. Rev. Mod. Phys. 57, 61–104, https://doi.org/10.1103/RevModPhys.57.61 (1985).
Barthel, E. et al. Conduction noise and motional narrowing of the nuclear magnetic resonance line in sliding spindensity waves. Phys. Rev. Lett. 71, 2825–2828, https://doi.org/10.1103/PhysRevLett.71.2825 (1993).
Zybtsev, S. G. & Pokrovskii, V. Y. Coherence of a sliding chargedensity wave stimulated by asynchronous rf irradiation. Phys. Rev. B 88, 125144, https://doi.org/10.1103/PhysRevB.88.125144 (2013).
Risken, H. The FokkerPlanck Equation (Springer, Berlin, 1989).
PérezJunquera, A. et al. Crossedratchet effects for magnetic domain wall motion. Phys. Rev. Lett. 100, 037203, https://doi.org/10.1103/PhysRevLett.100.037203 (2008).
Titov, S. V., Kachkachi, H., Kalmykov, Y. P. & Coffey, W. T. Magnetization dynamics of two interacting spins in an external magnetic field. Phys. Rev. B 72, 134425, https://doi.org/10.1103/PhysRevB.72.134425 (2005).
Evstigneev, M. et al. Diffusion of colloidal particles in a tilted periodic potential: Theory versus experiment. Phys. Rev. E 77, 041107, https://doi.org/10.1103/PhysRevE.77.041107 (2008).
Brandt, E. H. The fluxline lattice in superconductors. Rep. Progr. Phys. 58, 1465–1594, http://stacks.iop.org/00344885/58/i=11/a=003 (1995).
Plourde, B. L. T. Nanostructured superconductors with asymmetric pinning potentials: Vortex ratchets. IEEE Trans. Appl. Supercond. 19, 3698–3714, http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5232856, https://doi.org/10.1109/TASC.2009.2028873 (2009).
CuadraSolis, P.dJ. et al. Observation of commensurability effects in a patterned thin superconducting pb film using microwave reflection spectrometry. Phys. Rev. B 89, 054517, https://doi.org/10.1103/PhysRevB.89.054517 (2014).
Pryadun, V. V., Sierra, J., Aliev, F. G., Golubovic, D. S. & Moshchalkov, V. V. Plain superconducting films as magnetic field tunable twodimensional rectifiers. Appl. Phys. Lett. 88, https://doi.org/10.1063/1.2171788 (2006).
Jelic, Z. L., Milosevic, M. V., Van de Vondel, J. & Silhanek, A. V. Stroboscopic phenomena in superconductors with dynamic pinning landscape. Sci. Rep. 5, 14604 EP–, https://doi.org/10.1038/srep14604, Article (2015).
Lara, A., Aliev, F. G., Silhanek, A. V. & Moshchalkov, V. V. Microwavestimulated superconductivity due to presence of vortices. Sci. Rep. 5, 9187, https://doi.org/10.1038/srep09187 (2015).
Silhanek, A. V. et al. Influence of artificial pinning on vortex lattice instability in superconducting films. New J. Phys. 14, 053006, http://stacks.iop.org/13672630/14/i=5/a=053006 (2012).
Grimaldi, G. et al. Speed limit to the abrikosov lattice in mesoscopic superconductors. Phys. Rev. B 92, 024513, https://doi.org/10.1103/PhysRevB.92.024513 (2015).
Shklovskij, V. A., Nazipova, A. P. & Dobrovolskiy, O. V. Pinning effects on selfheating and fluxflow instability in superconducting films near T _{ c }. Phys. Rev. B 95, 184517, https://doi.org/10.1103/PhysRevB.95.184517 (2017).
Dobrovolskiy, O. V. et al. Pinning effects on flux flow instability in epitaxial nb thin films. Supercond. Sci. Technol. 30, 085002, http://stacks.iop.org/09532048/30/i=8/a=085002 (2017).
Villegas, J. E. et al. A superconducting reversible rectifier that controls the motion of magnetic flux quanta. Science 302, 1188–1191, http://www.sciencemag.org/content/302/5648/1188.abstract, https://doi.org/10.1126/science.1090390 (2003).
VlaskoVlasov, V. K., Colauto, F., Benseman, T., Rosenmann, D. & Kwok, W.K. Triode for magnetic flux quanta. Sci. Rep. 6, 36847 EP–, https://doi.org/10.1038/srep36847, Article (2016).
Dobrovolskiy, O. V. & Huth, M. Dual cutoff direct currenttunable microwave lowpass filter on superconducting nb microstrips with asymmetric nanogrooves. Appl. Phys. Lett. 106, 142601–1–5, https://doi.org/10.1063/1.4917229 (2015).
Gillijns, W., Silhanek, A. V., Moshchalkov, V. V., Reichhardt, C. J. O. & Reichhardt, C. Origin of reversed vortex ratchet motion. Phys. Rev. Lett. 99, 247002, https://doi.org/10.1103/PhysRevLett.99.247002 (2007).
Togawa, Y. et al. Direct observation of rectified motion of vortices in a niobium superconductor. Phys. Rev. Lett. 95, 087002, https://doi.org/10.1103/PhysRevLett.95.087002 (2005).
Dobrovolskiy, O. V. Abrikosov fluxonics in washboard nanolandscapes. Physica C 533, 80–90, http://www.sciencedirect.com/science/article/pii/S092145341630096X, https://doi.org/10.1016/j.physc.2016.07.008 (2017).
Dobrovolskiy, O. V., Huth, M. & Shklovskij, V. A. Alternating currentdriven microwave loss modulation in a fluxonic metamaterial. Appl. Phys. Lett. 107, 162603–1–5, https://doi.org/10.1063/1.4934487 (2015).
Dobrovolskiy, O. V., Huth, M. & Shklovskij, V. A. Anisotropic magnetoresistive response in thin Nb films decorated by an array of Co stripes. Supercond. Sci. Technol. 23, 125014–1–5, https://doi.org/10.1088/09532048/23/12/125014 (2010).
Campbell, A. M. & Evetts, J. E. Flux vortices and transport currents in type ii superconductors. Adv. Phys. 21, 199–428, https://doi.org/10.1080/00018737200101288 (1972).
Lowell, J. Investigation of pinning sites in superconductors by means of alternating currents. ii. experiments. J. Phys. F: Metal Phys. 2, 559, http://stacks.iop.org/03054608/2/i=3/a=023 (1972).
Auslaender, O. M. et al. Mechanics of individual isolated vortices in a cuprate superconductor. Nat Phys 5, 35–39, https://doi.org/10.1038/nphys1127 (2009).
Embon, L. et al. Probing dynamics and pinning of single vortices in superconductors at nanometer scales. Sci Rep. 5, 7598, https://doi.org/10.1038/srep07598 (2015).
Kremen, A. et al. Mechanical control of individual superconducting vortices. Nano Letters 0, null, https://doi.org/10.1021/acs.nanolett.5b04444, PMID: 26836018 (2016).
Lu, Q., Reichhardt, C. J. O. & Reichhardt, C. Reversible vortex ratchet effects and ordering in superconductors with simple asymmetric potential arrays. Phys. Rev. B 75, 054502, https://doi.org/10.1103/PhysRevB.75.054502 (2007).
Gittleman, J. I. & Rosenblum, B. Radiofrequency resistance in the mixed state for subcritical currents. Phys. Rev. Lett. 16, 734–736, https://doi.org/10.1103/PhysRevLett.16.734 (1966).
Coffey, M. W. & Clem, J. R. Unified theory of effects of vortex pinning and flux creep upon the rf surface impedance of typeII superconductors. Phys. Rev. Lett. 67, 386–389, https://doi.org/10.1103/PhysRevLett.67.386 (1991).
Pompeo, N. & Silva, E. Reliable determination of vortex parameters from measurements of the microwave complex resistivity. Phys. Rev. B 78, 094503–1–10, https://doi.org/10.1103/PhysRevB.78.094503 (2008).
Shklovskij, V. A. & Dobrovolskiy, O. V. ACdriven vortices and the Hall effect in a superconductor with a tilted washboard pinning potential. Phys. Rev. B 78, 104526–1–12, https://doi.org/10.1103/PhysRevB.78.104526 (2008).
Shklovskij, V. A. & Dobrovolskiy, O. V. Frequencydependent ratchet effect in superconducting films with a tilted washboard pinning potential. Phys. Rev. B 84, 054515–1–12, https://doi.org/10.1103/PhysRevB.84.054515 (2011).
Shklovskij, V. A. & Dobrovolskiy, O. V. Microwave Absorption by Vortices in Superconductors with a Washboard Pinning Potential, chap. 11, 263–288 (InTech, Rijeka, 2012).
Shklovskij, V. A., Sosedkin, V. V. & Dobrovolskiy, O. V. Vortex ratchet reversal in an asymmetric washboard pinning potential subject to combined dc and ac stimuli. J. Phys.: Cond. Matt. 26, 025703, http://stacks.iop.org/09538984/26/i=2/a=025703 (2014).
Shklovskij, V. A. & Hop, D. T. B. The Hall effect and microwave absorption by vortices in an anisotropic superconductor with a periodic pinning potential. Low Temp. Phys. 36, 71–80, https://doi.org/10.1063/1.329293 (2010).
Fiory, A. T. Quantum interference effects of a moving vortex lattice in al films. Phys. Rev. Lett. 27, 501–503, https://doi.org/10.1103/PhysRevLett.27.501 (1971).
Larkin, A. I. & Ovchinnikov, Y. N. Nonlinear conductivity of superconductors in the mixed state. J. Exp. Theor. Phys. 41, 960–965, http://www.jetp.ac.ru/cgibin/index/e/41/5/p960?a=list (1975).
Musienko, L. E., Dmitrenko, I. M. & Volotskaya, V. G. JETP Lett. 31, 567 (1980).
Klein, W., Huebener, R. P., Gauss, S. & Parisi, J. Nonlinearity in the fluxflow behavior of thinfilm superconductors. J. Low Temp. Phys. 61, 413–432, https://doi.org/10.1007/BF00683694 (1985).
Volotskaya, V. G., Dmitrenko, I. M., Koretskaya, O. A. & Musienko, L. E. Fiz. Nizk. Temp. 18, 973 (1992).
Peroz, C. & Villard, C. Flux flow properties of niobium thin films in clean and dirty superconducting limits. Phys. Rev. B 72, 014515–1–6, https://doi.org/10.1103/PhysRevB.72.014515 (2005).
Leo, A. et al. Competition between intrinsic and extrinsic effects in the quenching of the superconducting state in fe(se,te) thin films. Phys. Rev. B 93, 054503, https://doi.org/10.1103/PhysRevB.93.054503 (2016).
Knufinke, M. et al. Deterministic josephson vortex ratchet with a load. Phys. Rev. E 85, 011122–1–9, https://doi.org/10.1103/PhysRevE.85.011122 (2012).
Shklovskij, V. A. Determination of coordinate dependence of the washboard pinning potential from the dynamic experiment with vortices. In Procedings of the Fifth International Conference on Mathematical Modeling and Computer Simulation of Materials Technologies MMT2008, Ariel, Israel (2008).
Shklovskij, V. A. & Dobrovolskiy, O. V. Determination of coordinate dependence of a pinning potential from a microwave experiment with vortices. Low Temp. Phys. 39, 120–124, https://doi.org/10.1063/1.4791773 (2013). http://link.aip.org/link/?LTP/39/120/1.
Dobrovolskiy, O. V., Begun, E., Huth, M. & Shklovskij, V. A. Electrical transport and pinning properties of Nb thin films patterned with focused ion beammilled washboard nanostructures. New J. Phys. 14, 113027–1–27, http://stacks.iop.org/13672630/14/i=11/a=113027 (2012).
Dobrovolskiy, O. V. & Huth, M. Crossover from dirty to clean superconducting limit in dc magnetronsputtered thin Nb films. Thin Solid Films 520, 5985–5990, http://www.sciencedirect.com/science/article/pii/S0040609012005718, https://doi.org/10.1016/j.tsf.2012.04.083 (2012).
Dobrovolskiy, O. V., Franke, J. & Huth, M. ^{4}He sample probe for combined microwave and dc transport measurements. Meas. Sci. Technol. 26, 035502, http://stacks.iop.org/09570233/26/i=3/a=035502 (2015).
Acknowledgements
Roland Sachser is acknowledged for automating the data acquisition and his help with nanopatterning. This work was supported through DFG project DO1511/31 and conducted within the framework of the COST Action CA16218 (NANOCOHYBRI) of the European Cooperation in Science and Technology. Further, funding from the European Commission in the framework of the program Marie SklodowskaCurie Actions  Research and Innovation Staff Exchange (MSCARISE) under Grant Agreement No. 644348 (MagIC) is acknowledged.
Author information
Authors and Affiliations
Contributions
O.V.D. conducted the experiment and wrote the manuscript. O.V.D. and M.H. designated the samples. V.A.S. suggested the concept of the experiment and analyzed the data. R.V.V. took part in the discussion of the results and analyzed the electrical voltage responses. M.H., V.A.S., and R.V.V. participated in the manuscript writing.
Corresponding author
Ethics declarations
Competing Interests
The authors declare that they have no competing interests.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
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
Dobrovolskiy, O.V., Huth, M., Shklovskij, V.A. et al. Mobile fluxons as coherent probes of periodic pinning in superconductors. Sci Rep 7, 13740 (2017). https://doi.org/10.1038/s4159801714232z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s4159801714232z
This article is cited by

Composition variation and electron irradiation effects on the fluctuation conductivity in Y1–zPrzBa2Cu3O7−δ single crystals
Journal of Materials Science: Materials in Electronics (2020)

Magnon–fluxon interaction in a ferromagnet/superconductor heterostructure
Nature Physics (2019)

Effect of annealing on a pseudogap state in untwinned YBa2Cu3O7−δ single crystals
Scientific Reports (2019)

Incoherent charge transport induced by irradiation of YBCO single crystals with MeV electrons
Journal of Materials Science: Materials in Electronics (2019)

Tuning electric charge scattering in YBCO single crystals via irradiation with MeV electrons
Journal of Materials Science: Materials in Electronics (2019)
Comments
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.