Controllable Magnetic Doping of the Surface State of a Topological Insulator

A combined experimental and theoretical study of doping individual Fe atoms into ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ is presented. It is shown through a scanning tunneling microscopy study that single Fe atoms initially located at hollow sites on top of the surface (adatoms) can be incorporated into subsurface layers by thermally activated diffusion. Angle-resolved photoemission spectroscopy in combination with ab initio calculations suggest that the doping behavior changes from electron donation for the Fe adatom to neutral or electron acceptance for Fe incorporated into substitutional Bi sites. According to first principles calculations within density functional theory, these Fe substitutional impurities retain a large magnetic moment, thus presenting an alternative scheme for magnetically doping the topological surface state. For both types of Fe doping, we see no indication of a gap at the Dirac point.

Figure 1

Constant current STM images of differently prepared samples of Fe on ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ [tunneling current $I_{\mathrm{t}}=50\mathrm{pA}$, sample bias voltage $V_{\mathrm{b}}=0.25\mathrm{V}$ (a)–(d), $V_{\mathrm{b}}=1\mathrm{V}$ (e)]. (a) Cold deposited Fe ($\approx 1\%$ of a ML) on pristine ${\mathrm{Bi}}_2{\mathrm{Se}}_3$. Inset: magnified view showing 8 individual Fe adatoms $10\mathrm{nm}\times 10\mathrm{nm}$, $V_{\mathrm{b}}=-0.3\mathrm{V}$, $I_{\mathrm{t}}=1\mathrm{nA}$). (b), (c) Samples prepared as in (a) annealed to $T_{\mathrm{A}}\approx 260$ and 370 K, respectively. (d) Cold deposited Fe (0.5% ML) on Ca-doped ${\mathrm{Bi}}_2{\mathrm{Se}}_3$. (e) Sample prepared as in (d) but with 6% ML Fe and annealed to $T_{\mathrm{A}}\approx 370\mathrm{K}$. (f) Number of Fe adatoms and subsurface Fe atoms relative to initial Fe adatom number as a function of $T_{\mathrm{A}}$. Emergent clusters were not counted in the statistics.

Figure 2

(a), (b) Atomically resolved STM images of two different subsurface Fe defects [$I_{\mathrm{t}}=50\mathrm{pA}$, $V_{\mathrm{b}}=1\mathrm{V}$ (a), $V_{\mathrm{b}}=0.25\mathrm{V}$ (b)]. (c)–(h) Bias dependent constant current images of a subsurface Fe atom (red arrows), 3 Fe adatoms and a Se vacancy at $V_{\mathrm{b}}$ as indicated. ($I_{\mathrm{t}}=100\mathrm{pA}$, $\Delta z=35\mathrm{pm}$).

Figure 3

Two series of ARPES measurements for pristine (a)–(e) and Ca-doped (f)–(j) ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ samples after different Fe deposition and annealing steps. (a), (f) Initial spectra after in situ cleaving. Subsequent cycles of cold deposition of Fe ($T\approx 150\mathrm{K}$, % ML Fe as indicated) and annealing at $T_{\mathrm{A}}\approx 370\mathrm{K}$ followed. The DP energy extracted from the spectra is marked by a horizontal line. The small arrows in (d) indicate QWS induced from band bending [13]. (k) The resulting binding energy of the DP is plotted in dependence of the preparation step number for a pristine (circles) and two differently Ca-doped samples (up and down triangles). Preparation step number 0 indicates the freshly cleaved substrate without Fe, preparation step 1 (1%), 3 (2.5%), 5 (5%), 7 (10%), and 2, 4, 6, 8 subsequent annealing steps.

Figure 4

(a)–(f) Side views of ball and stick models of the different Fe doped ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ systems considered in the ab initio calculations: (a), (b) Fe adatoms on the two different hollow sites fcc (Fe_T1) and hcp (Fe_T2), (c), (d) Fe substituting Bi atoms in the 2nd (Fe_2Bi) and 4th (Fe_4Bi) subsurface layer, and (e), (f) Fe in two different interstitial sites (Fe_I1, Fe_I2) of the ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ lattice. Fe is shown in black, Bi blue, and Se orange. (g) Calculated total DOS for the different considered sites. The resulting energy shift $\Delta E$ of the DOS due to the doping is indicated. The magnetic moments in the Fe $d$ shell and the total magnetic moments ($m_{\mathrm{Fe}}$, $m_{\mathrm{tot}}$) for Fe at these different sites are given in the legend in units of ${\mu }_B$.