CN105702753B - A kind of ferroelectric thin film device with bulk photovoltaic effect - Google Patents

Abstract

The invention discloses a ferroelectric thin film device with volume photovoltaic effect, which comprises a substrate, on which a ferroelectric thin film is deposited, and the polarization direction of the ferroelectric thin film is perpendicular to its surface, eliminating the photovoltaic effect of the ferroelectric thin film device. The contribution of the depolarization field in the effect; the ferroelectric film is deposited with a regular polygonal central electrode, and the ferroelectric film is also deposited with a plurality of regular polygonal edge electrodes, and the number of edge electrodes is equal to the number of sides of the central electrode , the edge electrode array is around the center electrode, each side of the center electrode is parallel to the side of the adjacent edge electrode and the distance is L, the material of the edge electrode and the center electrode is the same and prepared at the same time, eliminating the ferroelectric thin film device The contribution of the Schottky barrier in the photovoltaic effect. Especially when the ferroelectric thin film is in T phase, the contribution of the domain wall in the photovoltaic effect of the ferroelectric thin film device can also be eliminated, thereby obtaining a ferroelectric thin film device with bulk photovoltaic effect.

Description

A Ferroelectric Thin Film Device with Bulk Photovoltaic Effect

technical field

The invention belongs to the field of ferroelectric materials, in particular to a ferroelectric thin film device with bulk photovoltaic effect.

Background technique

Ferroelectric materials have attracted much attention due to their abnormal photovoltaic effect (the photovoltaic voltage is not limited by the crystal band gap (E _g ), and can even be 2 to 4 orders of magnitude higher than E _g , reaching 10 ³ to 10 ⁵ V/cm). focus on.

Half a century ago, ferroelectric photovoltaic effect has been found in various ferroelectric materials with non-centrosymmetric, which can produce stable photovoltaic effect along the direction of polarization. It is generally believed that the photovoltaic effect of ferroelectric materials originates from its spontaneous polarization. One of the remarkable characteristics of ferroelectric photovoltaics is that when the polarization direction changes under the action of an electric field, the photo-generated current also changes accordingly, and in ferroelectric materials The direction of the internal photogenerated current is always opposite to the polarization direction. The difference between the ferroelectric photovoltaic effect and the traditional pn junction is that in the traditional pn junction, the electron-hole pairs excited by light are rapidly separated by the built-in field in the pn junction, drift in the opposite direction, and finally reach the electrode , and then collected by the electrodes. Therefore, in theory, the photovoltaic voltage generated by pn junction solar cells is limited by the width of the semiconductor band gap, generally less than 1V. For the ferroelectric photovoltaic effect, the experimentally obtained photovoltage is proportional to the polarization intensity and the distance between electrodes, and is not limited by the band gap width, which can reach 10 ⁴ V. The higher the photovoltaic voltage of a solar cell, the more electricity it generates and the higher its efficiency.

Although the research on the ferroelectric photovoltaic effect has been done for decades, until now, no one has been able to pinpoint the principle of the photovoltaic process of this material, and there has been controversy about the origin of the anomalous photovoltaic effect of ferroelectric materials. In general, there are many factors that affect the photogenerated voltage of ferroelectric materials, such as the distance between two electrodes, the intensity of light, the conductivity of the material, the remanent polarization, crystal orientation, grain size, oxygen vacancies, domain walls and interfaces. But in essence, the mechanism of ferroelectric photovoltaic effect mainly has the following types:

(1) Bulk photovoltaic effect

This mechanism believes that the photo-generated voltage is generated inside the ferroelectric material, so it is called "bulk photovoltaic effect", and the ferroelectric material acts as a "current source". The stable current generated under light irradiation (photo-induced current: J _s ) is related to the properties of ferroelectric materials having non-centrosymmetry. In non-centrosymmetric crystals, the probability of an electron transitioning from a state with momentum k to a state with momentum k' is different from the probability of electrons transitioning from a state with momentum k' to a state with momentum k, which leads to the photogenerated carrier The momentum distribution is asymmetric, so that a stable current can be formed under light. The total current density (J) through the ferroelectric material can be expressed as:

J＝J _s +(σ _d +σ _ph )E (1)

In the formula, _{σd and σph represent} the conductance of the ferroelectric material in dark field and bright field, respectively, that is, dark conductance and photoconductance; E=V/d is the electric field inside the ferroelectric material under light, which depends on the applied voltage (V ) and the distance between the two electrodes (d). Since the distance between electrodes is usually relatively large, and the dark conductance and photoconductivity of most ferroelectric materials are very low, solar photovoltaic devices composed of ferroelectric materials can be regarded as current sources. In ferroelectric materials, the open circuit voltage V _oc under illumination can be expressed as:

It can be seen from the above formula that if the total conductivity (σ _d +σ _ph ) does not depend significantly on the light intensity, the open circuit voltage V _oc increases linearly with the short-circuit photocurrent density J _sc , indicating that the open circuit voltage is proportional to the short-circuit photocurrent I _oc (because the short-circuit photocurrent I _oc is equal to the short-circuit photocurrent density J _s multiplied by the area where the current flows), the ratio of the open-circuit voltage to the short-circuit photocurrent is equal to the thickness of the sample. That is to say, if the solar photovoltaic device made of ferroelectric materials can be regarded as a current source, the photocurrent is constant, and the value of the short-circuit photovoltage is proportional to the thickness of the material.

(2) Domain wall theory

When Yang et al. studied the photovoltaic effect of bismuth ferrite (BiFeO ₃ , abbreviated as BFO) thin films, they found that the photogenerated voltage in BFO increased linearly with the increase of the number of domain walls in the polarization direction, and no obvious observation was observed perpendicular to the polarization direction. the photovoltaic effect. According to the domain wall theory, since the polarization intensity will produce a component perpendicular to the domain wall, the potential generated at the domain wall is ~10mV, and the domain wall width is about 2nm, so the electric field generated by the polarization at the domain wall is as high as 5×10 ⁶ V/m, which is much larger than the internal electric field in the pn junction, is considered to be the origin of the anomalous photovoltaic effect of ferroelectric materials, and is also the main driving force for the separation of photogenerated carriers. Since there are many electric domains in ferroelectric materials, the domains are connected end to end after being polarized, and the domain walls are like nanoscale photovoltaic generators connected in series, and the photogenerated voltage gradually accumulates along the polarization direction. This mechanism is similar to the concept of connecting solar cells in series, where the output voltage is the sum of each unit. If the distance between the two electrodes is larger, there will be more electric domains, and the photovoltage generated between the two electrodes under light will be higher. This model can well explain the anomalous photovoltaic effect. In addition, due to the continuous _photocurrent generated under illumination, domain walls are regarded as current sources in some literatures. The distance J _sc decides. Different from the bulk photovoltaic effect, the domain wall theory attributes the anomalous photovoltaic effect to the excitation of carriers at the domain wall. It is believed that the recombination speed of photo-excited carriers outside the domain wall is very fast, and the bulk photovoltaic effect can be ignored.

Although the domain wall theory can be used to explain the anomalous photovoltaic effect, that is, the photogenerated voltage can be much larger than the forbidden band width, however, there are some experimental phenomena that cannot be explained only by the magnetic domain wall theory, and the bulk photovoltaic effect theory must be taken into account. For example, according to the domain wall model, since the potential drop at the domain wall is caused by polarized charges, the photocurrent does not depend on the polarization direction of light. However, researchers have observed that the photocurrent changes with the polarization direction of incident light in ferroelectric materials such as BFO, indicating that the origin of the anomalous photovoltaic effect of ferroelectric materials is more complicated than everyone expected. In the ferroelectric photovoltaic effect, since both the electric domain and the body effect contribute to the photo-generated current, if the two are constructive, the photo-generated current will be larger, otherwise, the photo-generated current will be smaller, which can explain why in Yang et al. No photocurrent was observed in the direction parallel to the domain wall in the experiment.

(3) Schottky junction effect

When the ferroelectric material is in contact with the electrode to form a Schottky barrier, the energy band at the interface will bend, and the electron-hole pairs generated under the light will be driven by the local electric field near the electrode, and the photocurrent generated is largely driven by the Schottky Depths of the base barrier and depletion layer are determined. According to this model, the magnitude of the photo-generated voltage generated inside the Schottky barrier is still limited to the band gap of the ferroelectric material, and the voltage caused by the Schottky effect is often ignored in the early stage of studying the ferroelectric photovoltaic effect, which is Because it is much lower than the anomalous photogenerated voltage in most ferroelectric crystals. But the Schottky effect is becoming more and more important in ferroelectric thin-film photovoltaic devices, because the photovoltaic voltage output in these devices is usually relatively small. Generally speaking, in a ferroelectric photovoltaic device with a sandwich structure composed of the same electrode and ferroelectric material, the contribution of the photocurrent generated by the Schottky barrier does not exist, because the upper and lower two identical electrodes and ferroelectric material The two Schottky junctions formed are back-to-back and contain each other, so the photo-generated voltage and current generated cancel each other out. However, the enhancement of the photovoltaic effect in ferroelectric photovoltaic devices with vertical structures can be achieved if different types of electrodes are used. Since the Schottky junction effect has nothing to do with the polarization direction of ferroelectric materials, according to this feature, the contribution of the Schottky junction and the bulk photovoltaic effect to the photocurrent can be distinguished. However, some researchers believe that the height of the Schottky barrier can be tuned by changing the polarization direction of the ferroelectric material by applying an electric field. Moreover, when the polarization direction of the Schottky barrier and the ferroelectric material changes, the sign of the photogenerated voltage changes accordingly. For example, in a ferroelectric diode with a vertical structure composed of Au/BFO/Au, both the photo-generated current and the photo-generated voltage change with the change of the polarization direction. Initially, the bulk photovoltaic effect of BFO films was considered to be the main reason for this phenomenon, but subsequent studies showed that the Schottky barrier change of BFO films during the polarization process was mainly caused by the migration of oxygen vacancies, while when When the oxygen vacancy migration is frozen at low temperature, the photovoltaic effect no longer changes with the polarization direction.

(4) Depolarization field effect

For a ferroelectric film in a polarized state, the surface of the film has a high concentration of polarized charges. If the shielding effect is not considered, these high-density polarized charges will generate a huge electric field in the ferroelectric layer. Taking BFO film as an example, its remnant polarization is about 100μC/cm ² , and the electric field generated by unshielded polarized charges can reach 3×10 ¹⁰ V/m. When the ferroelectric film is in contact with a metal or semiconductor, the surface charge caused by remanent polarization will be partially shielded by the free charge in the metal or semiconductor. Usually, the reason why the surface charge is not completely shielded is that the center of gravity of the polarized charge and the free compensation charge do not coincide, and an electric field is generated inside the entire ferroelectric film, that is, a depolarization field. The depolarization field can be very large, for example, for a BTO film with a thickness of 10 to 30 nm, the depolarization field in a sandwich structure composed of BTO and SrRuO ₃ electrodes is about 45×10 ⁶ V/m. Such a high depolarization field is considered to be the main driving force for the separation of photogenerated carriers, and also indicates that the anomalous photovoltaic effect is closely related to the shielding degree of polarized charges. The distribution of shielding charges depends on the properties of ferroelectric materials and metals (or semiconductors), such as remanent polarization, free charge density, and dielectric constant. On the other hand, the influence of unshielded polarized charges on the depolarization field depends mainly on the thickness of the ferroelectric layer: a thinner ferroelectric layer results in a larger depolarization field. In general, the depolarization field generated by the contact between a semiconductor and a ferroelectric material is larger than that generated by a metal in contact with a ferroelectric material, because the semiconductor material has a smaller free charge density and a larger dielectric constant, resulting in a weaker shielding effect.

In general, there are many mechanisms that affect the photovoltaic effect of ferroelectric thin films. Factors such as bulk effect, depolarization field, electric domain, and interface barriers have simultaneous effects on the photovoltaic effect, and there is a certain relationship between them. Therefore, it is of great significance to distinguish the contribution of each mechanism to the ferroelectric photovoltaic effect and to find out which mechanism is dominant to the photovoltaic effect in ferroelectric materials.

Among many ferroelectric materials, bismuth ferrite (BFO) has attracted much attention due to its large polarization and relatively small optical bandgap. Due to the lattice mismatch between the BFO film and the substrate, the BFO film will be stressed. When the stress is relatively small, BFO belongs to the rhombohedral structure (R phase). If the in-plane compressive stress of the R-phase BFO film continues to increase, its lattice constant and structure will become different, thereby affecting its physical properties. The lattice constant in the c direction increases as the stress increases, and when the stress reaches a certain level, the R phase transforms into a distorted tetragonal crystal structure (tetragonal structure, ie T phase). The in-plane and out-of-plane lattice constants are c/a=1.26, which belongs to the P4mm point group.

For R-phase BFO, its polarization direction is along the diagonal. Therefore, when studying the BFO photovoltaic effect, since the polarization direction (the opposite direction of the depolarization field) has components in the plane (parallel to the surface) and out of the plane (perpendicular to the surface), there is a contribution to the photovoltaic effect. . In this way, it is difficult to obtain the positive value of the bulk photovoltaic effect, because the measured bulk photovoltaic effect includes the contribution of the depolarization field. In addition, if the electrodes on both sides of the BFO (upper and lower electrodes or left and right electrodes) are asymmetrical, the resulting photovoltaic effect includes the contribution of the interface barrier in addition to the bulk photovoltaic effect. Different from the R-phase, the polarization direction of the T-phase BFO is perpendicular to its surface.

Contents of the invention

In order to measure the intrinsic photovoltaic effect of ferroelectric thin film, the invention provides a kind of ferroelectric thin film device with bulk photovoltaic effect, on the ferroelectric thin film deposit the central electrode of regular polygon, the surrounding array of central electrode and central electrode The edge electrodes with the same number of sides, each side of the center electrode is parallel to the side of the adjacent edge electrode and separated by a certain distance, and the materials of the center electrode and the edge electrode are the same and the photovoltaic effect of the ferroelectric film is measured at the same time. There is no Schottky barrier (that is, the interface barrier); at the same time, the polarization direction of the ferroelectric film is perpendicular to the surface of the ferroelectric film, and the connection line between the center electrode and the edge electrode is perpendicular to the depolarization field, then the measured The photovoltaic effect of the ferroelectric film does not include the contribution of the depolarization field. When the ferroelectric film is in the T phase, there are only 180-degree domain walls, and the domain walls are also perpendicular to the surface of the ferroelectric film. The photovoltaic effect does not include the contribution of domain walls, but only the intrinsic bulk photovoltaic effect of ferroelectric thin films.

The present invention is realized through the following technical solutions:

A ferroelectric thin film device with bulk photovoltaic effect, comprising a substrate on which a ferroelectric thin film is deposited, the polarization direction of the ferroelectric thin film is perpendicular to its surface; A central electrode, a plurality of regular polygonal edge electrodes are also deposited on the ferroelectric film, the number of edge electrodes is equal to the number of sides of the central electrode, the array of edge electrodes is around the central electrode, and each side of the central electrode is connected to the adjacent The sides of the edge electrodes are parallel and separated by a distance L, and the materials of the edge electrodes and the center electrodes are the same and are prepared at the same time.

Further, the ferroelectric film is T-phase, and the T-phase ferroelectric film has only 180-degree domain walls, and the domain walls are also perpendicular to the surface of the ferroelectric film, so it is measured that the photovoltaic effect of the ferroelectric film does not contain the contribution of domain walls .

Further, the side length d1 of the center electrode, the side length d2 of the edge electrode, d1≥d2>L, the side length d1 of the center electrode and the side length d2 of the edge electrode are far greater than the parallel direction of the center electrode and the edge electrode The distance L between the sides, the center electrode and the adjacent edge electrodes can be regarded as plate capacitors, and the electric field between them is the parallel electric field.

Further, the side length d2 of the edge electrodes is >0.3mm, and the distance between the central electrode and the side parallel to the adjacent edge electrodes is 0<L<0.1mm.

Further, the materials of the center electrode and the edge electrodes are the same as one of Ag, Au, Cu, Pt, and ITO.

Beneficial effects of the present invention:

The ferroelectric thin film device with bulk photovoltaic effect of the present invention deposits the same central electrode and a plurality of edge electrodes of the material simultaneously on the ferroelectric thin film, and the polarization direction of the ferroelectric thin film is perpendicular to the surface of the ferroelectric thin film, when the ferroelectric thin film When the film is in the T phase, there are only 180 degree domain walls, and the domain walls are also perpendicular to the surface of the ferroelectric film, so the measured photovoltaic effect of the ferroelectric film does not include Schottky barriers (ie, interface barriers), regression The contribution of the polarization field and the domain wall is only the intrinsic bulk photovoltaic effect. Since the bulk photovoltaic effect is related to the polarization direction of the incident light, the ferroelectric thin film device of the present invention arrays a plurality of edge electrodes parallel to the sides of the central electrode around the central electrode, and can measure the center respectively without changing the incident light. The photovoltaic effect between each surface of the electrode and the adjacent edge electrodes, thereby obtaining the relationship between the photovoltaic effect and the angle (and the relationship between the photovoltaic effect and the polarization direction of the incident light).

Description of drawings

Fig. 1 is the structural representation of ferroelectric thin film device of the present invention;

Fig. 2 is the top view of Fig. 1;

Fig. 3 is the XRD diffraction pattern of T phase bismuth ferrite film;

Fig. 4 is to measure the numbering of the photovoltaic effect of ferroelectric thin film device to electrode;

Fig. 5 is a trend graph of short-circuit photocurrent values between the center electrode and different edge electrodes.

reference sign

1-substrate; 2-bismuth ferrite film; 3-center electrode; 4-edge electrode.

detailed description

The present invention will be further described below in conjunction with drawings and embodiments.

As shown in Fig. 1 and Fig. 2, a kind of ferroelectric thin film device with bulk photovoltaic effect comprises substrate 1, deposits ferroelectric thin film 2 on described substrate 1, and the polarization direction of this ferroelectric thin film 2 is perpendicular to its surface, the measured photovoltaic effect of the ferroelectric thin film does not include the contribution of the depolarization field. Optimally, the ferroelectric film 2 is in T phase, with only 180-degree domain walls, and the domain walls are also perpendicular to the surface of the ferroelectric film, so it is measured that the photovoltaic effect of the ferroelectric film does not include the contribution of domain walls. . The ferroelectric thin film 2 is deposited with a regular polygonal center electrode 3, and a plurality of regular polygonal edge electrodes 4 are also deposited on the ferroelectric thin film 2. The number of edge electrodes 4 is equal to the number of sides of the central electrode 3, and the edge The electrodes 4 are arrayed around the center electrode 3, each side of the center electrode 3 is parallel to the side of the adjacent edge electrode 4 and the distance is L, and the materials of the edge electrode 4 and the center electrode 3 are the same and prepared at the same time, then The measured photovoltaic effect of the ferroelectric thin film does not have the contribution of the Schottky barrier (that is, the interface barrier), so the measured photovoltaic effect of the ferroelectric thin film device is only the intrinsic bulk photovoltaic effect of the ferroelectric thin film. A plurality of edge electrode arrays are on the edge of the center electrode and the sides of the edge electrodes are parallel to the adjacent sides of the center electrode and separated by a certain distance. The photovoltaic effect between the electrodes, thus obtaining the relationship between the photovoltaic effect and the angle (and the relationship between the photovoltaic effect and the polarization direction of the incident light), has a simple structure and is convenient for measurement. Optimally, the side length d1 of the center electrode 3, the side length d2 of the edge electrode 4, d1≥2>L, the side length d1 of the center electrode and the side length d2 of the edge electrode are far greater than the center electrode and the edge The separation distance L between the electrodes to the parallel sides. The side length d2 of the edge electrode 4 is >0.3 mm, and the distance between the center electrode 3 and the side parallel to the adjacent edge electrode 4 is 0<L<0.1 mm. The material of the center electrode 3 and the edge electrode 4 is one of Ag, Au, Cu, Pt and ITO. In this embodiment, the ferroelectric thin film 2 is a BFO thin film, and the substrate 1 selects one of LaAlO ₃ and YAlO ₃ whose lattice constant is smaller than that of the BFO thin film, so that four-directional bismuth ferrite is easily formed. In other cases, diamond-shaped bismuth ferrite is easy to grow; d1=d2=0.3mm, L<0.1mm; d1=d2=5mm, L<1mm; d1=d2=0.1mm, L<0.01mm.

The bismuth ferrite thin film device with bulk photovoltaic effect in the present embodiment is carried out the test of photovoltaic effect: Fig. 3 is the XRD diffraction pattern of T phase bismuth ferrite thin film, BFO thin film is T phase, is epitaxial growth, only (00L) wherein L can remove 1, 2, 3... diffraction peaks, and no other impurity phases are produced. The wavelength of the laser used in the test of the photovoltaic effect is 408nm, and the power is linearly polarized light of 50mW. The instrument used for measurement is Keithley2611 Digital Source Meter. Fig. 4 is the numbering of the electrodes for measuring the photovoltaic effect of the ferroelectric thin film device, and the measured results are shown in Fig. 5 . Figure 5 shows the short circuit between the center electrode and the corresponding edge electrode at different angles (that is, a-b1 (0° to the horizontal direction), a-b2 (45°), ... a-b8 (315°)) The magnitude of the photocurrent. It can be seen that, without changing the incident light, the obtained short-circuit photocurrent has a certain dependence on the angle (equivalent to changing the polarization direction of the incident light without changing the angle). Since the polarization direction and domain wall in the ferroelectric thin film device are perpendicular to the surface of the bismuth ferrite film, and the electrodes on the surface of the bismuth ferrite film are the same electrode, the experimental results directly prove that the photocurrent is generated by the bulk photovoltaic effect.

Claims (5)

1. a kind of ferroelectric thin film device with bulk photovoltaic effect, it is characterised in that：Including substrate (1), sunk on the substrate (1) Product ferroelectric thin film (2), the polarised direction of the ferroelectric thin film (2) is perpendicular to its surface；There is in just deposition on the ferroelectric thin film (2) Also deposition has the edge electrodes (4) of multiple regular polygons, edge electrodes on polygonal central electrode (3), ferroelectric thin film (2) (4) quantity is equal with the side number of the central electrode (3), and edge electrodes (4) array is around central electrode (3), center The each edge of electrode (3) is parallel with the side of adjacent edge electrodes (4) and spacing distance is L, the edge electrodes (4) and center The material of electrode (3) is identical and prepares simultaneously.

2. a kind of ferroelectric thin film device with bulk photovoltaic effect according to claim 1, it is characterised in that：The ferroelectricity Film (2) is T-phase.

3. a kind of ferroelectric thin film device with bulk photovoltaic effect according to claim 1 or 2, it is characterised in that：It is described The length of side d1, the length of side d2, d1 >=d2 of the edge electrodes (4) of central electrode (3)>L.

4. a kind of ferroelectric thin film device with bulk photovoltaic effect according to claim 3, it is characterised in that：Edge electrodes (4) length of side d2>0.3mm, the spacing distance 0 on central electrode (3) side parallel with neighboring edge electrode (4)<L<0.1mm.

5. a kind of ferroelectric thin film device with bulk photovoltaic effect according to claim 4, it is characterised in that：The center Electrode (3), edge electrodes (4) material are all mutually Ag, Au, Cu, Pt, ITO one of which.