Effects of Surface Electron Doping and Substrate on the

Subscriber access provided by UNIV OSNABRUECK

Communication

Effects of surface electron doping and substrate on the superconductivity of epitaxial FeSe films Wenhao Zhang, Xi Liu, Chenhaoping Wen, Rui Peng, Shiyong Tan, Binping Xie, Tong Zhang, and Donglai Feng Nano Lett., Just Accepted Manuscript • Publication Date (Web): 09 Feb 2016 Downloaded from http://pubs.acs.org on February 9, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Effects of surface electron doping and substrate on the superconductivity of epitaxial FeSe films W. H. Zhang†, X. Liu†, C. H. P. Wen†, R. Peng†, S. Y. Tan†, B. P. Xie†,‡, T. Zhang*,†,‡, D. L. Feng†,‡ †

State Key Laboratory of Surface Physics, Department of Physics, and Advanced Materials Laboratory, Fudan University, Shanghai 200433, China ‡ Collaborative Innovation Center of Advanced Microstructures, Fudan University, Shanghai 200433, China *E-mail address: [email protected]

Superconductivity in FeSe is greatly enhanced in films grown on SrTiO3 substrates, although the mechanism behind remains unclear. Recently, surface potassium (K) doping has also proven able to enhance the superconductivity of FeSe. Here by using scanning tunneling microscopy, we compare the K doping dependence of the superconductivity in FeSe films grown on two substrates: SrTiO3 (001) and graphitized SiC (0001). For thick films (20 unit cells (UC)), the optimized superconducting (SC) gaps are of similar size (~9 meV) regardless of the substrate. However, when the thickness is reduced to a few UC, the optimized SC gap is increased up to ~15meV for films on SrTiO3, while it remains unchanged for films on SiC. This clearly indicates that the FeSe/SrTiO3 interface can further enhance the superconductivity, beyond merely doping electrons. Intriguingly, we found that this interface enhancement decays exponentially as the thickness increases, with a decay length of 2.4 UC which is much shorter than the lengthscale for relaxation of the lattice strain, pointing to interfacial electron-phonon coupling as the likely origin. KEYWORDS: iron-based superconductor, FeSe, interfacial effect, scanning tunneling microscopy, electron-phonon coupling Recently single-layer FeSe/SrTiO3 is found to have much enhanced superconductivity (Tc > 60K) than the bulk FeSe (Tc ~ 8K), which has greatly stimulated people’s interest1~10. The central issue is to understand the pairing enhancement mechanism in this interfacial system. It is believed that the SrTiO3 substrate should play an important role in the pairing enhancement, because it introduces lattice strain and electron doping, and possibly enhanced electron-phonon coupling2, 6,11-13. On the other hand, recent studies show that just through heavy electron doping, via surface K dosing14, liquid gating15-17 or intercalation18, one can also enhance the Tc above 40K in multi-layer or bulk FeSe. Therefore, to clarify the role of interfacial effects in the FeSe/SrTiO3 system, further effort is needed. It is

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

necessary to investigate the film thickness dependence of the Tc enhancement in FeSe/SrTiO3, and more directly, to examine interfaces other than FeSe/SrTiO3. In this work, by using low temperature scanning tunneling microscopy (STM), we studied the superconductivity of K dosed FeSe films grown on two different substrates – SrTiO3 (001) and graphitized SiC (0001) – with variable film thicknesses from single layer to 20 UC. Compared with the undosed case, the superconductivity in all the films is enhanced. For 20 UC thick films, either grown on SrTiO3 or SiC, the maximum superconducting gaps observed (taken to be 'optimal' doping) are of similar size (~9 meV). However, when the thickness is reduced to a few layers (1~4 UC), the SC gap in films on SrTiO3 is further enhanced to 15 meV, while the gap in thin films on graphitized SiC remains the same or is slightly weakened. This clearly indicates that interfacial effects from SrTiO3 can further enhance the superconductivity in FeSe. Moreover, this further enhancement effect is found to decay exponentially when film thickness increases, with a rather short decay length of 2.4 UC. In combination with ARPES results, we found that this behavior cannot be ascribed to strain effects, and is more likely induced by other interfacial effects such as electron-phonon coupling. The experiment was conducted in a low-temperature STM at 4.2 K. The SrTiO3 (001) substrates (0.5% Nb doped) were cleaned by direct heating at 1250K in vacuum. The graphitized SiC (0001) substrates were prepared by direct heating of SiC (0001) at 1650K. FeSe thin films were grown by co-deposition of high purity Se (99.999%) and Fe (99.995%) on substrates held at 620 K, followed by post-annealing at 670K. For multi-layer FeSe films on SrTiO3, the interfacial FeSe layer was grown first, followed by annealing at 800 K, then the remaining layers were grown on top. This procedure ensures that the interfacial FeSe layer has an enhanced Tc. K atoms were deposited on the sample surface at low temperature (~100K), with a growth rate of 0.075 ML/min (calibrated by counting deposited atoms, see Fig.1e and Supplementary Fig. s1) . The samples were transferred to the STM module in vacuum immediately after growth. PtIr tips were used in all the measurements; the tunneling conductance (dI/dV) was collected by lock-in amplifier with a modulation frequency of 713 Hz. Fig. 1a shows the typical surface of a cleaned SrTiO3 (001) substrate, which is composed of atomically flat terraces. Fig. 1b shows the graphitized SiC used here, which is mostly covered by single-layer graphene. The 6 × 6 reconstruction of the underlying SiC can also be observed. It is known that epitaxial graphene layers on SiC are strongly bonded to the substrate, making them electronically different from the free-standing case19. Fig. 1c shows the typical topography of 2 UC FeSe films on SrTiO3 (001), where irregular domain boundaries can be seen. These structures are likely inherited from the interfacial FeSe layer, which has been observed to exhibit domain structure due to the existence of two epitaxial sites1,10. We also found that films with different thickness can coexist in one sample, for example there is 1UC region in Fig. 1c (see also Supplementary Fig.s2 for a 2.8 UC film in which 1~4 UC thickness coexist). This enables us to measure the thickness dependence of the gap in one sample. K atoms were deposited on the surfaces with the coverage (referred as Kc in the


Page 2 of 12

Page 3 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

following) varied from 0.01ML to 0.3ML. Here a monolayer (ML) is defined by the areal density of Fe atoms in single-layer FeSe (~1.41×1015/cm2). We found that in large scale over tens of nanometers, the K coverage is homogeneous (and regardless of film thickness), but at the small scale of few nanometers, K atoms are usually randomly distributed (see Fig. 1e and Supplementary Fig. s3(a~c)). We noticed that at some coverages such as 0.125 ML and 0.16 ML, K atoms can locally form ordered structures, such as 5 × 5 (with respect to the FeSe unit cell, Fig. 1f), or a six-fold close-packed structure with a lattice constant of 0.76 nm (Fig. 1g). When Kc < 0.2 ML, K atoms adsorb individually, and we calculate Kc by counting atoms over a 20×20 nm2 region. At coverages higher than 0.2ML, K atoms form disordered clusters and individual atoms can no longer be resolved (Fig. 1h). The coverage is then estimated by the growth rate and deposition time. Without K doping, multi-layer FeSe films on SrTiO3 (001) do not show the enhanced superconductivity observed in the interfacial layer. On films thicker than 20 UC, a superconducting gap of about 2 meV can be observed, which is attributed to the bulk superconductivity of FeSe. After appropriate surface K dosing, the multi-layer films, or at least their topmost layer, also become superconducting, evidenced by the gap opening at the Fermi energy EF. Fig. 2a shows the typical dI/dV spectra of 2 UC FeSe/SrTiO3 as a function of K coverage. Before K dosing, the spectrum is semiconducting-like. Starting from Kc ~ 0.013ML, a superconducting gap appears to open. The gap magnitude is then determined by one half of the energy between two coherence peaks (or kink-like feature at the gap edge, as indicated by the short bars in Fig. 2a). It reaches a maximum value of 14 meV at Kc = 0.15 ML, and closes at Kc ~ 0.26 ML, giving a dome-like doping dependence. The influence of the K dosing on the FeSe band structure is indicated by the dI/dV spectra in Fig. 2b. The peak below EF in each spectrum is likely from the top of a hole band (at the Γ point of the Brillouin Zone (BZ)), which shifts systematically to lower energy as Kc increases. The overall shift from Kc =0 to Kc ~ 0.15 ML is about 30 meV, as shown in the Fig. 2b insert. Fig. 3 summarizes the doping dependence of the SC gap for different film thicknesses. We observed that the gap always has some spatial inhomogeneity, as shown in Supplementary Fig.s3. The gap inhomogeneity could be due to the locally random distribution of K atoms/clusters, since they may cause local variation of the doping and the coherence length of FeSe superconductors are also small (about 2~3 nm for 1UC FeSe/STO10). Therefore we determine the gap size at each Kc by averaging 5~10 spectra taken at different locations within a 20×20 nm2 region. One see that for all the films, the gap size has a dome-shaped dependence on Kc. However, the maximum gap value increases as the film thickness decreases. The 2 UC film has a SC gap of 14 meV, which is quickly reduced to 9 meV for 20 UC films. For comparison, the 1 UC FeSe film has a SC gap of 15 meV, as shown in Supplementary Fig.s4 (K dosing cannot further enhance the gap of the 1 UC film but slightly suppress it). The current observations indicate electron doping is certainly required to achieve a high Tc. However, the SC gap shows a further enhancement as the film thickness is


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reduced, which is unexpected for ordinary superconducting systems. This unusual behavior implies the SrTiO3 substrate or the interface must play a crucial role. To further clarify the substrate effect, we studied the K doping dependence of FeSe films grown on graphitized SiC. Fig. 1d shows the topography of single-layer FeSe on 1ML graphene/SiC, with atomic resolution. We found that layer-by-layer growth of FeSe films could only be achieved on SiC with single-layer epitaxial graphene, while on multi-layer graphene/SiC, FeSe tends to form islands. The easier nucleation of FeSe on 1ML graphene/SiC may be due to the remaining weak polarization from the underlying SiC interface. From Fig. 1d, the lattice constant of a 1 UC film is measured to be 0.37 nm, very close to the bulk value. Therefore, unlike single-layer FeSe on SrTiO3 (which has an expanded lattice parameter of 0.39 nm), single-layer FeSe on graphitized SiC has no significant epitaxial strain. The growth behavior and relaxed lattice also indicate that the bonding between FeSe film and graphitized SiC is rather weak. It has been reported that as-grown FeSe thick films on graphitized SiC show similar superconductivity to bulk FeSe20,21. In particular, the SC gap decreases as the film thickness decreases. Here we found that upon K dosing, all the FeSe films on graphitized SiC exhibit enhanced superconductivity (with respect to bulk FeSe). Fig. 4b shows the typical dI/dV spectra of 1 UC, 4 UC, and 20 UC FeSe films at the optimal doping level (Kc~0.15 ML). A superconducting gap of about 9 meV (from averaging 5~10 spectra) is observed in all such films. The K doping dependence of the 20 UC film, as shown in Fig. 4a, also displays a dome-like structure (gap size vs. Kc is summarized in Fig. 3). Here we note that the gap bottom of the 1 UC and 4 UC films does not reach zero conductance, which could be due to the increased thermal or quantum fluctuations in these thin films, which are thought to be weakly bound to the SiC substrate. Although the K dosed FeSe films on graphitized SiC also have enhanced superconductivity relative to bulk FeSe, their pairing gaps are still smaller than that of single (or few) layer FeSe on SrTiO3. This is consistent with recent ARPES and transport studies, which show that single layer FeSe/STO has a Tc of 65K or 110K (refs. 2~4,9), while K dosed FeSe films have a Tc of about 46K (ref 14). This further confirms that the FeSe/SrTiO3 interface should do more than merely doping electrons. As shown in Fig. 4c, the extra enhancement of Tc decreases rapidly as the thickness increases. An exponential fit (black dashed line, see also Fig. 5) yields a decay length L∆=2.4 UC (~1.2 nm). To date, two candidate effects for the interfacial enhancement have been observed in FeSe/SrTiO3: the lattice strain3 and interfacial electron-phonon (e-ph) coupling6. First, the lattice constant of single-layer FeSe/SrTiO3 is expanded from 0.37 nm to 0.39 nm (ref 3), which may enhance the antiferromagnetic interactions22. To clarify whether the strain is relevant here, we measured the thickness dependence of the in-plane lattice constant by ARPES (ref 3), as plotted in Fig. 5. The lengthscale for the relaxation of this strain, La=16.5 UC, far exceeds the 2.4 UC relaxation length for the superconducting gap. In addition, we note that even for a 1 UC film with unstrained lattice (0.37nm, grown on BaTiO3 (ref 5)), the Tc is still similar to that with the expanded lattice. Therefore, the strain cannot account for


Page 4 of 12

Page 5 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

the enhanced SC gap observed here. The other possible interfacial effect is the recently reported e-ph coupling. ARPES studies6,23, accompanied by theoretical calculations11~13, have shown pronounced e-ph coupling due to the optical phonon mode of SrTiO3. Although the details of the coupling may not be very clear, one can expect the polar field generated by the optical phonons will be quickly screened out as the film thickness increases, thus the e-ph coupling strength will be suppressed too. Therefore we speculate that the exponential decay of the gap magnitude vs. thickness on SrTiO3 is governed by the decay of the effects from interfacial e-ph coupling. The epitaxial graphene layer on SiC is unlikely to support high frequency out-of-plane phonon modes, thus a similar interfacial enhancement is not expected in the FeSe/SiC system. Finally, the superconducting proximity effect may play some role in the Tc enhancement in FeSe/ SrTiO3. The high Tc of the interfacial FeSe may enhance the Tc of FeSe overlayers on top of it, with the prerequisite that the overlayers be electron doped. Therefore proximity effect could be effective in K doped 2 UC film. However, since only the topmost layer of the film is actually doped by K atoms, as demonstrated by recent ARPES measurements24, it is unlikely that the proximity effect would still be effective for 3 or 4 UC films, as there are undoped FeSe layers between the interfacial layer and the topmost layer in these cases. In summary, we have compared the superconductivity of surface K dosed FeSe films grown on SrTiO3 and graphitized SiC. For thick films (20 UC), the optimized SC gap magnitudes are independent of the substrate. When the thickness is reduced to few UC, however, the optimized gap is further increased for films on SrTiO3 but not for films on SiC. The lengthscale over which the SrTiO3 interface enhances Tc is inconsistent with that of the lattice expansion, and we argue that a superconducting proximity effect from the interfacial layer is also incapable of explaining the observed decay. This leaves electron-phonon coupling at the interface as the key factor governing the anomalous Tc enhancement on SrTiO3 substrates. Upon finishing this work, we noticed two related works on K dosed FeSe films, in Refs. 25 and 26. Particularly in Ref. 26 Song et al. observed SC gap on K dosed FeSe/SiC with double coherence peaks and larger size than we observed. We think the discrepancy in gap shape and size could be due to some difference in growth conditions of the FeSe film (such as post-annealing temperature), which may result different defect densities in the film and eventually affects the superconductivity after K dosing. Nevertheless, the FeSe films studied here are all grown under the same condition. We thank Darren Peets for useful discussion and help on the writing. This work is supported by the National Science Foundation of China and the National Basic Research Program of China (973 Program) under grant No. 2012CB921402. Supporting Information: Including: The calibration of K growth rate; large scale STM image of 2.8 UC FeSe on STO; spatial inhomogeneity of the superconducting gap in K dosed FeSe; the K dosing dependence of the SC gap of 1UC FeSe/STO


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References: 1.

2.

3.

4.

5. 6.

7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Wang, Q. Y.; Li, Z.; Zhang, W. H.; Zhang, Z. C.; Zhang, J. S.; Li, W.; Ding, H.; Ou, Y. B.; Deng, P.; Chang, K.; Wen, J.; Song, C. L.; He, K.; Jia, J. F.; Ji, S. H.; Wang, Y. Y.; Wang, L. L.; Chen, X.; Ma, X. C.; and Xue, Q. K. Chin. Phys. Lett. 2012, 29, 037402. Liu, D. F.; Zhang, W. H.; Mou, D. X.; He, J. F.; Ou, Y. B.; Wang, Q. Y.; Li, Z.; Wang, L.; Zhao, L.; He, S. L.; Peng, Y. Y.; Liu, X.; Chen, C. Y.; Yu, L.; Liu, G. D.; Dong, X. L.; Zhang, J.; Chen, C. T.; Xu, Z. Y.; Hu, J. P.; Chen, X.; Ma, X. C.; Xue, Q. K.; and Zhou, X. J. Nat. Commun. 2012, 3, 931. Tan, S. Y.; Zhang, Y.; Xia, M.; Ye, Z.; Chen, F.; Xie, X.; Peng, R.; Xu, D. F.; Fan, Q.; Xu, H. C.; Jiang, J.; Zhang, T.; Lai, X. C.; Xiang, T.; Hu, J. P.; Xie, N. P.; and Feng, D. L. Nat. Mater. 2013, 12, 634. He, S. L.; He, J. F.; Zhang, W. H.; Zhao, L.; Liu, D. F.; Liu, X.; Mou, D. X.; Ou, Y. B.; Wang, Q. Y.; Li, Z.; Wang, L. L.; Peng, Y. Y.; Liu, Y.; Chen, C. Y.; Yu, L.; Liu, G. D.; Dong, X. L.; Zhang, J.; Chen, C. T.; Xu, Z. Y.; Chen, X.; Ma, X. C.; Xue, Q. K.; and Zhou, X. J. Nat. Mater. 2013, 12, 605. Peng, R.; Xu, H. C.; Tan, S. Y.; Xia, M.; Shen, X. P.; Huang, Z. C.; Wen, C. H. P.; Song, Q.; Zhang, T.; Xie, B. P.; Feng, D. L. Nat. Commun. 2014, 5, 5044. Lee, J. J.; Schmitt, F. T.; Moore, R. G.; Johnston, S.; Cui, Y.-T.; Li, W.; Yi, M.; Liu, Z. K.; Hashimoto, M.; Zhang, Y.; Lu, D. H.; Devereaux, T. P.; Lee, D.-H. & Shen, Z.-X. Nature 2014, 515, 245. Zhang, W. H.; Sun, Y.; Zhang, J. S.; Li, F. S.; Guo, M. H.; Zhao, Y. F.; Zhang, H. M.; Peng, J. P.; Xing, Y.; Wang, H. C.; Fujita, T.; Hirata, A.; Li, Z.; Ding, H.; Tang, C. J.; Wang, M.; Wang, Q. Y.; He, K.; Ji, S. H.; Chen, X.; Wang, J. F.; Xia, Z. C.; Li, L.; Wang, Y. Y.; Wang, J.; Wang, L. L.; Chen, M. W.; Xue, Q. K.; and Ma, X. C. Chin. Phys. Lett. 2014, 31, 017401. Zhang, W. H.; Li, Z.; Li, F. S.; Zhang, H. M.; Peng, J. P.; Tang, C. J.; Wang, Q. Y.; He, K.; Chen, X.; Wang, L.; Ma, X. C.; and Xue, Q. K. Phys. Rev. B 2014, 89, 060506(R). Ge, J. F.; Liu, Z. L.; Liu, C.; Gao, C. L.; Qian, D.; Xue, Q. K.; Liu, Y.; and Jia, J. F. Nat. Mater. 2014, 14, 285. Fan, Q.; Zhang, W. H.; Liu, X.; Yan, Y. J.; Ren, M. Q.; Peng, R.; Xu, H. C.; Xie, B. P.; Hu, J. P.; Zhang, T.; and Feng, D. L. Nat. Phys. 2015, 11, 946. Coh, S.; Cohen, M. L and Louie, S. G. New J. Phys. 2015, 17, 073027. Xie, Y.; Cao, H. Y.; Zhou, Y.; Chen, S.; Xiang, H.; Gong, X. G. Sci. Rep. 2015, 5, 10011. Lee, D. H. Chin. Phys. Lett. 2015, 24, 117405. Miyata, Y.; Nakayama, K.; Sugawara, K.; Sato, T.; and Takahashi, T. Nat. Mater. 2015, 14, 775. Hanzawa, K.; Sato, H.; Hiramatsu, H.; Kamiya, T.; and Hosono, H. arXiv:1508.07689, 2015. Lei, B.; Cui, J. H.; Xiang, Z. J.; Shang, C.; Wang, N. Z.; Ye, G. J.; Luo, X. G.; Wu, T.; Sun, Z.; and Chen, X. H. arXiv:1509.00620, 2015. Shiogai, J.; Ito, Y.; Mitsuhashi, T.; Nojima, T. and Tsukazaki, A. Nature Phys. DOI: 10.1038/NPHYS3530, 2015. Lu, X. F.; Wang, N. Z.; Wu, H.; Wu, Y. P.; Zhao, D.; Zeng, X. Z.; Luo, X. G.; Wu, T.; Bao, W.; Zhang, G. H.; Huang, F. Q.; Huang, Q. Z. & Chen, X. H. Nat. Mater. 2015, 14, 325.


Page 6 of 12

Page 7 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

19. Hass, J.; Heer, W. A de and Conrad, E. H. J. Phys.: Condens. Matter. 2008, 20, 323202. 20. Song, C. L.; Wang, Y. L.; Cheng, P.; Jiang, Y. P.; Li, W.; Zhang, T.; Li, Z.; He, K.; Wang, L.; Jia, J. F.; Hung, H.-H.; Wu, C. J.; Ma, X. C.; Chen, X.; and Xue, Q. K. Science 2011, 332,1410. 21. Song, C. L.; Wang, Y. L.; Jiang, Y. P.; Li, Z.; Wang, L.; He, K.; Chen, X.; Ma, X. C.; and Xue, Q. K. Phys. Rev. B 2011, 84, 020503. 22. Cao, H.-Y.; Tan, S.; Xiang, H.; Feng, D. L.; and Gong, X.-G. Phys. Rev. B 2014, 89, 014501. 23. Yang, S.; Sobota, J. A.; Leuenberger, D.; Kemper, A. F.; Lee, J. J.; Schmitt, F. T.; Li, W.; Moore, R.G.; Kirchmann, P. S.; and Shen, Z.-X. Nano Lett. 2015, 15, 4150. 24. Wen, C. H. P.; Xu, H. C.; Chen, C.; Huang, Z. C.; Pu, Y. J.; Song, Q.; Xie, B. P.; M. Abdel-Hafiez; Chareev, D. A.; Vasiliev, A. N.; Peng, R.; and Feng, D. L. arXiv:1508.05848, 2015. 25. Tang, C. J.; Liu, C.; Zhou, G. Y.; Li, F. S.; Zhang, D.; Li, Z.; Song, C. L.; Ji, S. H.; He, K.; Chen, X.; Wang, L. L.; Ma, X. C.; and Xue Q. K., Phys. Rev. B, 93, 020507(R) 2016 26. Song, C. L.; Zhang, H. M.; Zhong, Y.; Hu, X. P.; Ji, S. H.; Wang, L.; He, K.; Ma, X. C. and Xue, Q. K. arXiv:1511.02007, 2015.

FIG. 1 Topographic images of (a) cleaned SrTiO3 substrate (Vb = 3.0 V, I = 10 pA), (b) graphitized SiC with single-layer graphene (Vb = 0.1 V, I = 50 pA), (c) 2 UC FeSe films on SrTiO3 (Vb = 3.0 V, I = 10 pA), (d) atomic-resolution image of 1 UC FeSe on graphitized SiC (Vb = 0.1 V, I = 30 pA). (e) - (h) Topographic images of 2 UC FeSe/SrTiO3 films with surface K adsorption at various coverages: (e) 0.045 ML, (f) 0.125 ML with a √5 × √5 superstructure, (g) 0.163 ML with a close-packed structure, and (h) 0.20 ML.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

FIG. 2 (a) Typical dI/dV curves taken on the 2 UC FeSe/SrTiO3 films with various K coverages as indicated; The short bars indicate the position of coherence peaks or kinks near the gap edge. The superconducting gap emerges starting from Kc = 0.013 ML. (b) Large scale tunneling spectra with Kc ranging from 0.013 ML to 0.163 ML, showing a distinctive peak below EF, which shifts to lower energy at large Kc. Insert: The peak position’s evolution as a function of K coverage. Each peak position is obtained by averaging 5~10 spectra taken at different locations. Error bars represent the spatial inhomogeneity.


Page 8 of 12

Page 9 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

FIG. 3 The summarized dependence of the superconducting (SC) gap on K coverage for different film thicknesses and substrates. The gap size is obtained by averaging 5~10 spectra taken at different locations for each Kc and film thickness. Error bars represent the spatial inhomogeneity of the SC gap.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

FIG. 4 (a) dI/dV spectra of 20 UC FeSe/SiC films with various K coverages. (b) SC gap of FeSe/SiC for different thicknesses. The short bars in (a) and (b) mark the position of coherence peaks or kinks near the gap edge. (c) Evolution of optimal SC gaps for different thickness of FeSe films on SrTiO3 and graphitized SiC. The gap size is obtained by averaging 5~10 spectra taken at different locations. Green dashed line indicates the SC gap of bulk FeSe (2.2 meV).


Page 10 of 12

Page 11 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

FIG. 5 The film thickness dependence of the in-plane lattice constant (red dots) and optimized SC gap size (black squares) of the FeSe/SrTiO3 system. The dashed lines are exponential fits.


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

50x25mm (300 x 300 DPI)


Page 12 of 12

Effects of Surface Electron Doping and Substrate on the

Recommend Documents