Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

# Direct penetration of spin-triplet superconductivity into a ferromagnet in Au/SrRuO3/Sr2RuO4 junctions

## Abstract

Efforts have been ongoing to establish superconducting spintronics utilizing ferromagnet/superconductor heterostructures. Previously reported devices are based on spin-singlet superconductors (SSCs), where the spin degree of freedom is lost. Spin-polarized supercurrent induction in ferromagnetic metals (FMs) is achieved even with SSCs, but only with the aid of interfacial complex magnetic structures, which severely affect information imprinted to the electron spin. Use of spin-triplet superconductors (TSCs) with spin-polarizable Cooper pairs potentially overcomes this difficulty and further leads to novel functionalities. Here, we report spin-triplet superconductivity induction into a FM SrRuO3 from a leading TSC candidate Sr2RuO4, by fabricating microscopic devices using an epitaxial SrRuO3/Sr2RuO4 hybrid. The differential conductance, exhibiting Andreev-reflection features with multiple energy scales up to around half tesla, indicates the penetration of superconductivity over a considerable distance of 15 nm across the SrRuO3 layer without help of interfacial complex magnetism. This demonstrates potential utility of FM/TSC devices for superspintronics.

## Introduction

Superconducting Cooper pairs are known to penetrate a normal metal (N) adjacent to a superconductor. This phenomenon, called the superconducting proximity effect, can occur even in FMs, where the strong ferromagnetic exchange field is expected to destroy the Cooper pairs. For FM/SSC interfaces, the proximity effect arises with the induction of spin-singlet Cooper pairs (pairs with the total spin S=0) into the FM over a coherence length , where ħ, DF, Eex are the reduced Planck constant, diffusion coefficient and exchange energy, respectively. Importantly, ξF is no more than a few nanometres even for weak FMs1. However, spin-triplet Cooper pairs (pairs with S=1) can also emerge at a FM/SSC interface if appropriate magnetic inhomogeneity, such as the ferromagnetic domain walls or noncollinear magnetization, exists at the interface2,3. Since spin-triplet Cooper pairs with parallel spins (Sz=±1) are insensitive to exchange field, they can penetrate the FM over a distance much longer than ξF. This is called the ‘long-range proximity effect’ (LRPE) and it has actively been investigated3,4,5,6,7,8. However, to exploit the SSC-based LRPE in spintronics devices, the necessity for magnetic inhomogeneity can be a technical issue, since magnetic inhomogeneity itself complicates the fabrication and can even disturb the spin information.

In contrast, there is a straightforward means of inducing the LRPE. This involves using TSCs instead of SSCs. Favourably, even without magnetic inhomogeneity, both charge and spin supercurrents can be generated at the FM/TSC interface9,10,11,12. The essential simplicity and the utility of the spin-degree of freedom in a FM/TSC junction in comparison with a FM/SSC junction is depicted in Fig. 1a,b. The absence of magnetic inhomogeneity is advantageous to transfer spin information with less disturbance. Furthermore, by rotating the magnetization of the FM relative to the spin of the spin-triplet Cooper pairs by applying an external magnetic field, the induction and functionality of the LRPE can be controlled. To date, the proximity effects at a FM/TSC interface have not been realized experimentally.

In case of FM/TSC junctions, the orbital symmetry of the induced superconducting correlation is also important. In any superconductor-based junctions, because of the lack of inversion symmetry near the interface, pair amplitudes with various orbital symmetries (s-wave, p-wave, d-wave and f-wave) can in principle emerge at the interface13. Regardless the nature of the bulk superconductor, for a diffusive and clean junction the s-wave odd frequency and p-wave even frequency Cooper pairs dominate, respectively. So far, experimentally, s-wave odd frequency correlation has been observed in Josephson junctions (in diffusive limit) fabricated using various ferromagnets and s-wave SSCs. On the other hand, the p-wave correlation has never been realized. Thus, FM/TSC junctions provide opportunity to study properties of the p-wave induced pair amplitude14, in particular toward utilization of the orbital degree of freedom in combination with the spins.

Sr2RuO4 (SRO214) with the superconducting critical temperature (Tc) of 1.5 K is one of the best-candidates TSCs15. Extensive experimental and theoretical studies16 indicate that SRO214 exhibits most likely the chiral p-wave spin-triplet state with broken time-reversal symmetry17,18,19,20,21,22, although there are still unresolved issues23,24. Recently, SRO214 attracts interest for exploring topological superconducting phenomena originating from its orbital phase winding16. Investigations of FM/SRO214 would also provide definitive evidence for its superconducting order parameter.

Since high-quality superconducting SRO214 thin films are not available, we recently developed an alternative method to obtain FM/SRO214 heterostructures. We use SRO214 single crystals as substrates and epitaxially deposit ferromagnetic SrRuO3 (SRO113) thin films with an atomically smooth and highly conductive interface25. By utilizing SRO113(15-nm)/SRO214 heterostructures and ex-situ depositing 600-nm-thick Au layer as an N layer, we fabricate N/FM/TSC junctions (see Methods).

In this Letter, we present results demonstrating for the first time the emergence of the spin-triplet LRPE in a FM through a simple epitaxial interface. It is essentially different from previous FM/SSC junctions, for which magnetic inhomogeneity is always required (Fig. 1a). The voltage-dependent differential conductance (dI/dV(V)) exhibits multiple characteristic features, persisting up to as high as 0.5 T without affected by ferromagnetic domain rearrangement. This result indicates that the superconducting correlations are induced even in the N-layer beyond the 15-nm FM layer, evidencing LRPE.

## Results

### Transport behaviour of a superconducting junction

For an N/SSC interface with negligible interfacial electronic-transport barrier strength Z, the differential conductance GS of the interface is twice the normal-state conductance GN for a bias voltage within the superconducting-gap energy. This is due to the transmission of an electron as a Cooper pair with a reflected hole via the Andreev reflection26. As Z increases, GS decreases within the gap and approaches zero for Z→∞. For an FM/SSC interface, GS falls with an increase in the spin-polarization of the FM and reaches zero for a 100% spin-polarized FM27, even for Z≈0. Our Au/SRO113/SRO214 junctions consist of two important interfaces: SRO113/SRO214 and Au/SRO113 (Fig. 1c–e). For such junctions, multiple features in the conductance are expected if the superconducting correlation reaches the Au/SRO113 interface as a result of LRPE.

### Temperature dependent transport properties

Figure 2a shows the temperature dependent resistance of Junction A measured with different applied currents Ia=0.1–1.0 mA. We observe the decrease in the resistance (increase in conductance) due to superconductivity. Obviously, there are multiple superconducting transitions. The first transition occurs at 1.22 K, which corresponds to the bulk superconducting transition of the SRO214 neck part immediately below the SRO113 layer (Fig. 1d). Indeed, the drop in the resistance accompanying this first transition is robust against Ia up to 1 mA. In addition, a resistance drop of 0.5 mΩ closely corresponds to the estimated value of the resistance of the SRO214 neck part (see Supplementary Figs 1 and 2 and Supplementary Note 1). The second transition emerges at 1.1 K for Ia=0.1 mA. The superconducting feature of the second transition is significantly suppressed with increasing Ia. By taking the temperature derivative of the resistance measured at 0.4 mA, we clearly see in addition to two peaks corresponding to the first and second transitions, another broad peak at lower temperatures (Fig. 2b). The suppression of the second transition and the existence of the third peak in dR/dT implies multiple characteristic energies induced within the junctions.

Next, we extract characteristic voltages associated with these three features. Figure 2c summarizes their temperature dependencies, as obtained from the dI/dV(V) at different temperatures (Fig. 2d). Note that we take the derivative of IV curves to obtain the dI/dV curves (Supplementary Figs 3 and 4; Supplementary Note 2). At 1.12 K, immediately below the first transition, we observe strong dips at a characteristic voltage V1≈±14 μV (blue arrow) and a small conductance enhancement within ±V1. The corresponding current flow of ≈2 mA yields current density of ≈107 A m−2 through the SRO214 neck (Fig. 1b inset). Thus, the feature of V1 originates from the destruction of the superconductivity at the neck by a current exceeding the critical current density of SRO214 (ref. 28). Interestingly, with decreasing temperature, another dip at V2 (purple arrow) and the enhancement of the conductance within ±V2 become evident (Fig. 2d, lower panel). The appearance of V2 provides a signature of the Andreev reflection with a small Z at the SRO113/SRO214 interface, indicating that superconductivity is induced in SRO113 via the proximity effect. At 1.11 K, an additional enhancement starts to appear within ±V3. Although V3 is close to zero at around 1.11 K, it increases to 5 μV at 0.3 K (Fig. 2d, green arrows). Because of the enhancement of the conductance with a flat-top peak at low temperatures, the V3 anomaly is attributable to an additional Andreev reflection in the junction. The temperature dependence of characteristic voltages are shown in Fig. 2c. Interestingly, V3(T) is described well with the temperature dependence of the superconducting gap of Sr2RuO4 obtained by taking into account the p-wave paring and two dimensionality29. This fact supports our interpretation that V3 represent superconducting-gap-like features.

We should comment here on the chiral p-wave px+ipy order parameter of SRO214. In some types of experiments, superconducting contribution in transport along the c axis should vanish because the px+ipy order parameter is averaged to zero. For example, in the c-axis oriented Josephson junctions between SRO214 and an s-wave superconductor, critical current is expected to be cancelled out to be zero. In contrast, the Andreev reflection in a metal attached along the c axis is certainly allowed even for the px+ipy state because the magnitude of the Andreev reflection is not given by the simple integration of the order-parameter phase14.

Let us consider where the Andreev reflection related to V3 takes place. As mentioned above, V2 should arise from the proximity induction at the SRO113/SRO214 interface. Considering the two interfaces within our junction, V3 should arise at the other interface, that is, the Au/SRO113 interface. For this to occur, any superconducting correlations penetrating from the SRO113/SRO214 interface must reach the Au layer across the 15-nm SRO113 layer, as illustrated in Fig. 2e. Other possible explanations of these multiple features are discussed in Supplementary Note 3.

### Field dependent transport properties

To investigate the magnetic field effect on these three features, we measured dI/dV(V) at 0.3 K with the field applied along the ab-surface (in-plane), after zero-field cooling. The obtained data (Supplementary Fig. 5) are presented in Fig. 3a as a three-dimensional colour map, which clearly shows the three features. The field dependences of V1 to V3 plotted in Fig. 3b exhibit a smooth decrease as the field increases. In particular, V2 and V3 exhibit behaviour at high fields (Supplementary Fig. 6), as expected for gap-like features30. The critical fields Hc for V2 and V3 reach as high as ≈550 mT and ≈650 mT, respectively, indicating that the proximity effect is limited by the bulk superconducting gap but not by the interfacial magnetic structure. The smooth variations of V2 and V3 are also surprising because the ferromagnetic domain walls within the device change their configuration at ≈200 mT, as verified by the sharp peaks in the normal-state magnetoresistance at 2 K (Fig. 3c). These peaks are associated with enhanced domain-wall scattering. Thus, the domain walls do not play significant roles in the generation of the proximity effect.

## Discussion

Briefly summarizing the experimental observations, we observe three superconducting features V1, V2 and V3 in dI/dV(V). Conductance enhancement within ±V3 strongly suggests the penetration of superconducting correlations through the 15-nm SRO113 layer leading to the Andreev reflection at the Au/SRO113 interface. This penetration is found to persist even above 0.5 T and to be unaffected by the ferromagnetic domain-wall configuration of the SRO113 layer.

The observed superconducting proximity penetration of 15 nm in SRO113 is actually much stronger than the expected penetration for spin-singlet Cooper pairs. The ferromagnetic coherence length for SRO113 (ξF-113) of the spin-singlet proximity induction is 3 nm for the clean limit (1 nm for the diffusive limit)31, which is five (fifteen) times smaller than the observed value. Therefore, only spin-triplet Cooper pairs can be responsible for the superconducting penetration within SRO113 of the present devices. Thus, this observation marks a new type of LRPE. Similar results are obtained from another junction (Junction B: 5 × 5 μm2; Supplementary Fig. 7). For comparison, various obtained parameters are given in Supplementary Table 1 (also see Supplementary Note 4).

The characteristic decay length of this new LRPE can be estimated by assuming that the superconducting gap decreases as , where x is the distance in the SRO113 layer from the SRO214/SRO113 interface. From the relationship V2Δ(x=0) and V3Δ(x=15 nm), we evaluate ξ*113=7 nm at 1 K and zero field and ξ*113=9 nm at 0.3 K and 350 mT. These values are about three times larger than ξF-113 for spin-singlet superconductivity in the clean limit but smaller than the thickness of the SRO113 layer. As shown in Fig. 4, we found that ξ*113 increases as the temperature decrease; at 0.3 K, ξ*113 is field-independent between 350 and 500 mT, as would be expected for the LRPE.

This penetration of the spin-triplet pairs is naturally explained by assuming SRO214 to be a TSC, as already evidenced by many experiments14,32. On the other hand, it should be noted that spin-triplet pairs can be generated even from a SSC provided magnetic inhomogeneity exists at the FM/SSC interface3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20. The ferromagnetic domain walls also provide this magnetic inhomogeneity. In this case, we would expect the domain-wall configuration to severely affect junction properties; the alignment of the domains by external fields should strongly suppress the proximity effect. However, we observed that superconducting transport properties of our junctions are insensitive to the domain-wall configurations and thus LRPE in SRO113 should be induced without magnetic inhomogeneity. Therefore, a hypothetical scenario that SRO214 is an SSC and the LRPE is induced by magnetic inhomogeneity is unlikely. The present results constitute additional new evidence for spin-triplet superconductivity in SRO214. To support this scenario, we performed theoretical calculations explained in Supplementary Note 5 (also see Supplementary Fig. 8).

Generally, in addition to ordinary even-frequency spin-triplet p-wave Cooper pairs, odd-frequency spin-triplet s-wave pairs can be generated at the interface9,10,11. The former cannot survive over a distance greater than the electron mean free path le, whereas the latter remains stable over a greater distance. SRO113 is a relatively good metal at low temperatures and for thin films with residual resistivity of around 10–20 μΩ cm yielding, le≈15 to 30 nm (ref. 33). This is of the same order or greater than the thickness of the SRO113 layer used in our junctions. In such a case, the even-frequency p-wave correlation amplitude may dominate. Theoretically, it has been demonstrated that, for diffusive N/TSC junction, the odd-frequency s-wave correlation can emerge at the interface and lead to a sharp zero-bias peak in the conductance spectrum originating from the induced mid-gap Andreev resonant state34. We anticipate that a similar zero-bias peak would be observed also in FM/TSC junctions if the odd-frequency pairs were dominant. On the other hand, the conductance of the present junctions at lower temperatures and zero field (Fig. 2d) exhibits a flat-top peak around zero bias voltage, without any sharp anomalies. These observations also support the penetration of the p-wave (even-frequency) amplitude.

The LRPE can be also sensitive to spin-flip scattering events. Therefore, the Cooper pair penetration is also limited by the spin-flip length , where τsf is the spin-flip time. Both s-wave (odd frequency) and p-wave (even frequency) correlation cannot extend over this length, whereas the latter can also be limited by the le. For SRO113 thin films, τsf is approximately ≈300 fs as revealed by an optical Kerr effect experiment35, yielding Lsf≈30 nm. This estimated Lsf is three times larger than the value of the decay length ξ*11310 nm obtained from experiment. This comparison suggests that the proximity effect arising in our devices decays due to potential scatterings rather than spin-flip scatterings, in consistent with the scenario for the penetration of p-wave correlations.

Nevertheless, it is actually impossible to completely deny contributions of the odd-frequency s-wave pairs. To examine the orbital characteristics of the dominating Cooper pairs, further studies using junctions with different thickness of SRO113 and a tunnel barrier between N and F layers are needed.

In summary, spin-triplet Cooper-pair penetration occurs in the superconducting Au/SRO113/SRO214 junctions without additional inhomogeneous magnetic layer. The results are attributed to the direct penetration of even-frequency p-wave spin-triplet pairs from a TSC into a FM through a simple epitaxial interface and persist even at 0.5 T. This observation opens new possibilities to transfer p-wave superconductor across an interface without losing spin information. Thus, our research offers new directions for the utilization of LRPE in FM/TSC systems in a new research area termed Superspintronics.

## Methods

### SrRuO3 thin films deposition

High-quality SRO214 single crystals were grown by using a floating-zone method. SRO214 crystals with the intrinsic Tc≈1.5 K require the elimination of Ru deficiencies; as a result, some parts of the crystals tend to contain SRO327, SRO113, as well as Ru-metal inclusions. In this study, we carefully choose SRO214 crystals without any such inclusions, but with a slightly lower Tc≈1.22 K. Ferromagnetic SRO113 thin films were epitaxially deposited by means of pulsed laser deposition on the cleaved ab-surface of the SRO214 substrates (3 × 3 × 0.5 mm3)25. Immediately after the deposition of the SRO113 thin film, a 20-nm thick Au layer with a 5-nm thick Ti adhesive layer was deposited ex-situ as a capping layer by dc-sputtering. Transmission electron microscope image given in Fig. 1c shows that SrRuO3/Sr2RuO4 interface is atomically smooth.

### Device fabrication

Afterwards, laser ultraviolet maskless photolithography was utilized to fabricate the Au/SRO113/SRO214 junctions. First, micron-sized square pads of SRO113 of the area of 25 × 25 μm2 (Junction A) and 10 × 10 μm2 (Junction B) were etched out using a AZ1500 photoresist (baked at 110 °C for 2 min) and ion-beam etching (1 kV, 100 W). Next, a 300-nm SiO2-layer was deposited (at a backing pressure of 10−9 mbar) over the samples, while maintaining 20 × 20 μm2 and 5 × 5 μm2 windows over 25 × 25 μm2 and 10 × 10 μm2 SRO113 pads, respectively, using a bilayer photoresist (AZ1500 and LOR15) and a lift-off technique. Note that the junction area is made smaller than the overall SRO113 pad area to avoid direct contact between the Au and SRO214. Top Au electrodes were also prepared by applying a lift-off technique. Finally, a 600-nm thick Au-layer was sputtered at a backing pressure of 10−9 mbar. Figure 1e shows optical microscope image of a device with six such junctions.

### Measurements of transport properties

Electrical transport measurements were performed using the four-point technique with two contacts on the Au top electrode and the other two contacts connected directly to the side of the SRO214 crystal, as shown in Fig. 1d. Transport properties were studied down to 300 mK using a 3He cryostat equipped with a superconducting magnet. The temperature-dependent resistance in the normal state (300–4 K) of Au/SRO113/SRO214 junctions is dominated by the interfacial resistance, with additional contribution of the c-axis (out-of-plane) resistance behaviour of the bulk SRO214 substrate (Supplementary Fig. 2a). Furthermore, the normal-state magnetoresistance at 2 K exhibits two sharp peaks at 200 mT (Fig. 3d). These facts reveal that the applied current (Ia) properly passes through the interface.

### Data availability

The data corresponding to this study are available from the corresponding author on request.

How to cite this article: Anwar, M. S. et al. Direct penetration of spin-triplet superconductivity into a ferromagnet in Au/SrRuO3/Sr2RuO4 junctions. Nat. Commun. 7, 13220 doi: 10.1038/ncomms13220 (2016).

## References

1. 1

Buzdin, A. I. Proximity effects in superconductor-ferromagnet heterostructures. Rev. Mod. Phys. 77, 935–976 (2005).

2. 2

Linder, J. & Robinson, J. W. A. Superconducting spintronics. Nat. Phys. 11, 307–315 (2015).

3. 3

Bergeret, F. S., Volkov, A. F. & Efetov, K. B. Long-range proximity effects in superconductor-ferromagnet structures. Phys. Rev. Lett. 86, 4096–4099 (2001).

4. 4

Keizer, R. S. et al. A spin triplet supercurrent through the half-metallic ferromagnet CrO2 . Nature 439, 825–827 (2006).

5. 5

Khaire, T. S., Khasawneh, M. A., Pratt, W. P. Jr & Birge, N. O. Observation of spin-triplet superconductivity in Co-based Josephson junctions. Phys. Rev. Lett. 104, 137002 (2010).

6. 6

Robinson, J. W. A., Witt, J. D. S. & Blamire, M. G. Controlled injection of spin-triplet supercurrents into a strong ferromagnet. Science 329, 59–61 (2010).

7. 7

Anwar, M. S., Czeschka, F., Hesselberth, M., Porcu, M. & Aarts, J. Long-range supercurrents through half-metallic ferromagnetic CrO2 . Phys. Rev. B 82, 100501(R) (2010).

8. 8

Anwar, M. S., Veldhorst, M., Brinkman, A. & Aarts, J. Long range supercurrents in ferromagnetic CrO2 using a multilayer contact structure. Appl. Phys. Lett. 100, 052602 (2012).

9. 9

Tanaka, Y., Golubov, A. A., Kashiwaya, S. & Ueda, M. Anomalous Josephson effect between even- and odd-frequency superconductors. Phys. Rev. Lett. 99, 037005 (2007).

10. 10

Brydon, P. M. R. & Manske, D. 0-π transition in magnetic triplet superconductor Josephson junctions. Phys. Rev. Lett. 103, 147001 (2009).

11. 11

Gentile, P. et al. Spin-orbital coupling in a triplet superconductor-ferromagnet junction. Phys. Rev. Lett. 111, 097003 (2013).

12. 12

Cheng, Q. & Jin, B. Quantum transport in normal-metal/ferromagnet/spin-triplet superconductor junctions. Phys. B: Cond. Matter 426, 40–44 (2013).

13. 13

Eschrig, M. & Löfwander, T. Triplet supercurrents in clean and disordered half-metallic ferromagnets. Nat. Phys. 4, 138–143 (2008).

14. 14

Terrade, D., Gentile, P., Cuoco, M. & Manske, D. Proximity effects in spin-triplet superconductor-ferromagnet heterostucture with spin-active interface. Phys. Rev. B 88, 054516 (2013).

15. 15

Maeno, Y. et al. Superconductivity in a layered perovskite without copper. Nature 372, 532–534 (1994).

16. 16

Maeno, Y., Kittaka, S., Nomura, T., Yonezawa, S. & Ishida, K. Evaluation of spin-triplet superconductivity in Sr2RuO4 . J. Phys. Soc. Jpn 81, 011009 (2012).

17. 17

Luke, G. M. et al. Time-reversal symmetry-breaking superconductivity in Sr2RuO4 . Nature 394, 558–561 (1998).

18. 18

Ishida, K. et al. Spin-triplet superconductivity in Sr2RuO4 identified by 17O knight shift. Nature 396, 658–660 (1998).

19. 19

Nelson, K. D., Mao, Z. Q., Maeno, Y. & Liu, Y. Odd-parity superconductivity in Sr2RuO4 . Science 306, 1151–1154 (2004).

20. 20

Xia, J., Maeno, Y., Beyersdorf, P. T., Fejer, M. M. & Kapitulnik, A. High resolution polar Kerr effect measurements of Sr2RuO4: evidence for broken time-reversal symmetry in the superconducting state. Phys. Rev. Lett. 97, 167002 (2006).

21. 21

Kidwingira, F., Strand, J. D., Harlingen, D. J. V. & Maeno, Y. Dynamical superconducting order parameter domains in Sr2RuO4 . Science 314, 1267–1271 (2006).

22. 22

Anwar, M. S. et al. Anomalous switching in Nb/Ru/Sr2RuO4 topological junctions by chiral domain wall motion. Sci. Rep. 3, 2480 (2013).

23. 23

Hicks, C. W. et al. Limits on superconductivity-related magnetization in Sr2RuO4 and PrOs4Sb12 from scanning SQUID microscopy. Phys. Rev. B 81, 214501 (2010).

24. 24

Yonezawa, S., Kajikawa, T. & Maeno, Y. First order superconducting transition of Sr2RuO4 . Phys. Rev. Lett. 110, 077003 (2013).

25. 25

Anwar, M. S. et al. Ferromagnetic SrRuO3 thin-film deposition on a spin-triplet superconductor Sr2RuO4 with a highly conducting interface. Appl. Phys. Express 8, 015502 (2015).

26. 26

Blonder, G. E., Tinkham, M. & Klapwijk, T. M. Transition from metallic to tunneling regimes in superconducting microconstrictions: excess current, charge imbalance, and supercurrent conversion. Phys. Rev. B 25, 4515 (1982).

27. 27

Soulen, R. J. et al. Measuring the spin polarization of a metal with a superconducting point contact. Science 282, 85–88 (1998).

28. 28

Saitoh, K. et al. High-supercurrent-density contacts and Josephson effect of strontium ruthenate. Appl. Phys. Exp. 5, 113101 (2012).

29. 29

Nomura, T. & Yamada, K. Detailed investigation of gap structure and specific heat in the p-wave superconductor Sr2RuO4 . J. Phys. Soc. Jpn 71, 404–407 (2002).

30. 30

Morenzoni, E. et al. Field dependence of the superconducting gap in YPd2Sn: A μSR and NMR study. J. Phys.: Conf. Ser. 551, 012027 (2014).

31. 31

Asuin, I., Yuli, O., Koren, G. & Millo, O. Evidence for crossed Andreev reflections in bilayer of (100) YBa2Cu3O7−δ and the itinerant ferromagnet SrRuO3 . Phys. Rev. B 74, 092501 (2006).

32. 32

Ishida, K. et al. Spin polarization enhanced by spin-triplet pairing in Sr2RuO4 probed by NMR. Phys. Rev. B 92, 100502(R) (2015).

33. 33

Koster, G. et al. Structure, physical properties, and applications of SrRuO3 thin films. Rev. Mod. Phys. 84, 253–298 (2012).

34. 34

Tanaka, Y., Asano, Y., Golubov, A. A. & Kashiwaya, S. Anomalous features of the proximity effect in triplet superconductors. Phys. Rev. B 72, 140503 (2005).

35. 35

Kantner, C. L. S. et al. Determination of the spin-flip time in ferromagnetic SrRuO3 from time resolved Kerr measurements. Phys. Rev. B 83, 134432 (2011).

## Acknowledgements

We are grateful to Y. Tanaka, J. Aarts, S. Kashiwaya, Jian Wei and Ying Liu for the useful discussions. We are thankful to M. Cuoco, D. Terrade, P. Gentile and M. Kunieda for their help on theoretical calculations. M.S.A. is supported as an International Research Fellow of the Japan Society for the Promotion of Science (JSPS). This work was supported by Grants-in-Aid for Scientific Research on Innovative Areas on ‘Topological Quantum Phenomena’ (JSPS KAKENHI JP22103002) and ‘Topological Materials Science’ (JSPS KAKENHI JP15H05851 and JP15H05852); and by Grant-in-Aid JSPS KAKENHI JP26287078. This work was also supported by IBS-R009-D1 and by Kyoto University Research Development Program Migaki'' 2015.

## Author information

Authors

### Contributions

M.S.A. devised the concept, designed the experiments and prepared the SRO214 substrates from crystals grown in Y.M.’s group at Kyoto University. S.R.L. and Y.J.S. prepared the SRO113 thin films and S.J.K. examined the films using a transmission electron microscope, under the supervision of T.W.N. at Seoul National University. R.I., Y.T. and M.S.A. fabricated the devices under the supervision of H.T. at RIKEN and the Tokyo University of Science. M.S.A., Y.S. and S.Y. performed the transport measurements under the supervision of Y.M. at Kyoto. D.M. provided the theoretical input on the results. M.S.A., S.Y., Y.S. and Y.M. wrote the paper and all authors contributed in the discussions.

### Corresponding author

Correspondence to M. S. Anwar.

## Ethics declarations

### Competing interests

The authors declare no competing financial interests.

## Supplementary information

### Supplementary Information

Supplementary Figures 1-8, Supplementary Table 1, Supplementary Notes 1-5 and Supplementary References. (PDF 1438 kb)

## Rights and permissions

Reprints and Permissions

Anwar, M., Lee, S., Ishiguro, R. et al. Direct penetration of spin-triplet superconductivity into a ferromagnet in Au/SrRuO3/Sr2RuO4 junctions. Nat Commun 7, 13220 (2016). https://doi.org/10.1038/ncomms13220

• Accepted:

• Published:

• ### Fermi surface and kink structures in $$\hbox {Sr}_{{4}}\hbox {Ru}_{{3}}\hbox {O}_{{10}}$$ revealed by synchrotron-based ARPES

• Prosper Ngabonziza
• Emanuela Carleschi
• Bryan Patrick Doyle

Scientific Reports (2020)

• ### Anomalous anisotropic behaviour of spin-triplet proximity effect in Au/SrRuO3/Sr2RuO4 junctions

• M. S. Anwar
• M. Kunieda
• Y. Maeno

Scientific Reports (2019)

• ### Triplet superconductivity in coupled odd-gon rings

• Sahinur Reja
• Satoshi Nishimoto

Scientific Reports (2019)

• ### Even odder after twenty-three years: the superconducting order parameter puzzle of Sr2RuO4

• Andrew P. Mackenzie
• Thomas Scaffidi
• Yoshiteru Maeno

npj Quantum Materials (2017)