1. Introduction

Manganites have attracted a great deal of attentions due to the fascinating physical properties, including ferromagnetism, antiferromagnetism, metal-insulator transition, charge ordering, colossal magnetoresistance (MR), and electron-lattice interaction and potential technological applications to new devices such as magnetic read heads, spin-based magnetic sensors and memories in MR devices [1–5]. During the last few years, there was an increasing interest in the fabrication and investigation of manganite-based heterojunctions, which not only demonstrated excellent rectifying behavior [6] but also displayed large MR [7–10] and photovoltages [11–13]. The transient photovoltaic characteristics enable us to develop ultrafast optoelectronic devices, such as ultraviolet and infrared photodetectors and photoswitches.

Compared with conventional semiconductor, the manganites show unique features, such as a strongly correlated electronic state and a magnetic state dependent band structure. Hence, one may anticipate that the photovoltaic properties of manganite-based heterojunctions can be tuned magnetically and electrically, which is of special interest from the application viewpoint. In this paper, to improve the photovoltaic responsivity of manganite-based photodetectors and also to meet the demand for useful photoelectronic devices operated at higher temperature even near room temperature, we perform comprehensive studies on the effect of external magnetic field H and bias current I_b on the transient photovoltaic effect in a manganite heterojunction on a n-type Si substrate at room temperature.

2. Experimental details

We fabricated the junction by depositing a 100-nm-thickness La_2/3Ca_1/3MnO₃ (LCMO) layer on a 0.5-mm-thick n-type Si (001) substrate with a resistivity of 4 Ωcm and a carrier concentration of ~10¹⁶ cm⁻³ using facing target sputtering technique from stoichiometry targets [14,15]. The substrate temperature was kept at 680 °C, and the oxygen partial pressure at 30 mTorr during deposition. Immediately after deposition, the vacuum chamber was back-filled with 1 atm oxygen gas. The deposited film was then cooled to room temperature with the substrate heater power cut off. For the electronic measurements, silver electrodes with separation of 2 mm were prepared on the LCMO and Si surfaces shown in the inset of Fig. 1(a) . A 248 nm KrF pulse laser in duration of 20 ns was used as the light source and the energy density was 0.61 mJ/mm². The waveform was recorded by a sampling oscilloscope terminated into 50 Ω. To test whether the transient photovoltage in the LCMO/Si heterojunction can be influenced by the applied magnetic field H, we applied external +/−H ( + represents towards right, - represents opposite direction) on the heterojunction as shown in Fig. 2 .

 

Fig. 1 A typical I-V curve with the strong asymmetry at room temperature measured by the four-point method with two electrodes on LCMO (electrode separation about 2 mm) and the other two on Si. The inset shows the schematic diagram of experiments.

Download Full Size | PDF

 

Fig. 2 (a) TPV under pulsed laser illumination in selected H (0, −6.4 and 6.4 kOe) with electrode setting at LCMO and Si. Fitting curves result from (TPV_6.4 + TPV_-6.4)/2 which agrees with TPV₀. The on sample energy density is 0.61 mJ/mm². (b) TPV variation defined as ΔTPV_H/TPV₀ = (TPV_H – TPV₀)/TPV₀. Inset shows the schematic diagram of experimental set up.

Download Full Size | PDF

3. Results and discussion

Sufficient rectifying behavior was observed in LCMO/Si heterojunction. A typical current–voltage (I-V) characteristic curve is shown in Fig. 1 with the strong asymmetry measured by the four-point method at room temperature. Leakage current occurring in the reverse bias region is very small (−2.5 μA at −0.52 V) and the ratio of forward current and reverse current approaches 100 in the applied voltage region from −0.15 to + 0.15 V. Diffusion potential arising from the mutual diffusion of holes and electrons between LCMO and Si is about 0.1 V, from which point the current began to increase obviously in the forward bias region.

To study the tunable effect on transient photovoltaic effect by the applied magnetic field H, we measured the transient photovoltaic responses (TPV) with the H perpendicular to the interface under lateral laser irradiations as shown in Fig. 2. Figure 2(a) shows the TPV under laser irradiations at selected H (0, −6.4 and 6.4 kOe). The magnetic field of −6.4 kOe dramatically reduces the TPV, which increases when 6.4 kOe is applied. The fitting curve resulting from (TPV_H + TPV_-H)/2 quite agrees well with TPV₀. The output peak value of TPV induced by pulsed laser is TPV₀^P ~26 mV without an applied magnetic field. And it can be modulated to TPV_H^P ~138 mV when a magnetic field 6.4 kOe is applied and to TPV_-H^P ~-114 mV with reversal magnetic field of −6.4 kOe. The relative variation, defined as ΔTPV_H/TPV₀ = (TPV_H – TPV₀)/TPV₀, can reach up to about 1000% as shown in Fig. 2(b).

We also measured TPV by gradually varying the external fields for constant light power density. From Fig. 3 , we can see that the peak photovoltage TPV^P has a nearly linear response to H when the relatively applied external magnetic field is larger than a certain value. These results unambiguously reveal that the external magnetic fields can manipulate the TPV of the LCMO/Si heterojunction. To further analysis the tunable effect by magnetic field, the dependence of the modulation coefficient Δλ on the external magnetic fields was calculated as shown in the inset of Fig. 3. Δλ is defined as (TPV_H^P – TPV₀^P)/H. It is showed that a small applied magnetic field can generate a big tunable effect on the TPV. With increasing magnetic field, the modulation coefficient gradually decreases to a constant value.

 

Fig. 3 TPV^P as a function of selected H. The on sample energy density is 0.61 mJ/mm². The inset shows the dependence of the modulation coefficient Δλ on the external magnetic fields.

Download Full Size | PDF

It can be also seen in Fig. 2(b) that the response time of TPV have been greatly shortened in external magnetic field indicating a faster photoelectric process different from conventional photovoltaic effect. Therefore, the waveform of TPV consist of two photovoltage from different processes (TPV = TPV1 + TPV2). TPV1 induced by build-in field and TPV2 induced by external magnetic field, which act independently. For the TPV1 process, photoinduced carriers separated by the built-in field derived from p-n junction and generated a photovoltaic response TPV₀ (as shown in Fig. 2(a)). For the TPV2 process, the photoresponse is from the separation of electrons and holes by external magnetic field. Under the ultrafast laser illumination, the photon absorption at the vicinity of the illuminated LCMO/Si profile results in carrier accumulation on the side surfaces and carrier concentration with gradient distribution along the direction of the radiation. Therefore, the electrons and holes will diffuse from side surfaces to internal sample along the gradient direction (also the direction of the radiation). Due to different charge polarity of electrons and holes, the photo-generated electrons and holes could be deflected toward opposite directions by the Lorentz force while diffusing into internal LCMO/Si bulks, which produces a voltage TPV2 along the film surface. When the direction of the magnetic field is reversed, the carriers are deflected in the opposite direction, forming reversal TPV2. TPV2_-H and TPV2_H induced by H and -H are equal in magnitude and opposite in direction which can also be demonstrated by the axial symmetrical Δλ with respect to H = 0, as shown in the inset of Fig. 3. Thus, in our experiments, the (TPV_H + TPV_-H)/2 = [(TPV1 + TPV2_H) + (TPV1-TPV2_-H)]/2 = TPV₀, which is in good agreement with the result of experiment as shown in Fig. 2(a). On the basis of these experiment results, the LCMO/Si heterojunction has potential utilization for magnetically tunable PV photodetectors which have a very wide range of linear dependence of the signal on the magnetic field larger than a certain value. In addition, such photodetectors are characterized by a short response time and can be used in studies of fast processes.

Figure 4 shows the effect of the bias currents I_b on the TPV. The applied bias currents are −350, 0, 350 μA and applied external magnetic fields are −6.4, 0, 6.4 kOe, respectively. The TPV are strongly dependent on the external magnetic fields as well as bias currents. At H = 0, the TPV change from positive to negative with the bias currents change from −350 to 350 μA. The bias current strongly affect the electric field distribution in the sample as well as the TPV1 process. The photoinduced carriers were separated by the combination of applied electric field and built-in field and generated a photovoltaic response TPV1 (I_b). At I_b = 0, the peak values change from negative to positive with the magnetic fields change from −6.4 to 6.4 kOe, which is similar to the results in Fig. 2. Furthermore, the first extreme value of TPV (I_b, H) occurs at t = t₁ = 11 ns while that of TPV (I_b, 0) occurs at t = t₀ = 24 ns, which indicates magnetic field highly shortened the TPV response times. Thus magnetic field has potential application in manganite-based heterojunctions based ultrafast photodetectors. Meanwhile the TPV (I_b, H) are affected by the bias current. At H = 6.4 kOe, the value of the TPV peak at t = t₁ decreases from 0.08 to 0.04 V with bias current from −350 to 350 μA, while at H = −6.4 kOe the first extreme value of TPV at t = t₁ changes from −0.035 to −0.067 V with bias current from −350 to 350 μA. It is obvious that TPV (I_b, H) = TPV1 (I_b) + TPV2 (H) as shown in Fig. 4. Moreover, when the bias current ± 350 μA is applied, the second extreme value appears in temporal photoresponse waveform at t = t₂ ~90 ns, and in a fixed magnetic field the second extreme value can be also strongly modulated from negative to positive while bias current changing from 350 to −350 μA. Combining the above two tunable effects on TPV, we can get the three-dimensional graph about the peak photovoltage TPV^P, bias current I_b and magnetic field H, as shown in Fig. 5 .

 

Fig. 4 TPV to pulsed laser illumination as a function of external magnetic field 6.4, 0, −6.4 kOe and bias current −350, 0, 350 μA simultaneously. The first extreme value of TPV (I_b, H) occur at t = t₁ = 11 ns while that of TPV (I_b, 0) occur at t = t₀ = 24 ns. The on sample energy density is 0.16 mJ/mm².

Download Full Size | PDF

 

Fig. 5 The three-dimensional graph about the TPV^P, bias current I_b and magnetic field H.

Download Full Size | PDF

4. Conclusion

In summary, magnetically and electrically tuned TPV were investigated in a manganite heterojunction LCMO/Si. The TPV are tuned by external magnetic field and bias current in a vast range. Magnetic field can shorten the response time of photovoltaic responses. The experiment results show that the manganite heterojunctions possesses potential application for magnetically and electrically tunable transient ultrafast photovoltaic detectors at room temperature.