JP2013128077A - Interface passivation structure, back passivation solar cell, and manufacturing method of interface passivation structure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an interface passivation technique which achieves excellent characteristics by suppressing occurrence of parasitic shunting.SOLUTION: A passivation film 202 is composed by laminating two kinds of passivation film having different interface fixed charge density, i.e., SiNx films 202a, 202b, on a p-type silicon substrate 101. The relation of Qf1<Qf2 is satisfied between the interface fixed charge density Qf1 when the SiNx film 202a is laminated alone on a silicon substrate 201 in contact therewith, and the interface fixed charge density Qf2 when the SiNx film 202b is laminated alone on the silicon substrate 201 in contact therewith. With such a configuration, the interface fixed charge density Qf of the whole passivation film is held down, and the interface state density can also be held down.

Description

  The present invention relates to an interface passivation technique, and more specifically to an interface passivation structure and a manufacturing method thereof, and a back surface passivation type solar cell.

  A solar cell which is a kind of photoelectric conversion element basically has a pn junction of a p-type semiconductor layer that collects positively charged holes generated by light reception and an n-type semiconductor layer that collects negatively charged electrons. Specifically, the pn junction is formed by forming an n-type impurity diffusion layer on the light-receiving surface side of the p-type silicon substrate, and electrodes are respectively formed on the light-receiving surface side and the back surface side of the p-type silicon substrate. Is provided.

  The electrode formed on the back surface generally has a two-layer structure of a p + layer formed on a p-type silicon substrate and a metal layer formed on the layer.

  Conventionally, aluminum is used as a material for forming the metal layer. When heat of 700 ° C. or higher is applied in a state where aluminum is in contact with the p-type silicon substrate, a diffusion region in which aluminum diffuses in the p-type silicon substrate is formed, and an alloy of aluminum and silicon is generated. Since aluminum serves as a dopant for supplying holes to silicon, a p + layer is formed in the diffusion region. Thus, a part of aluminum contributes to p + layer formation, and the remaining aluminum becomes a metal electrode as it is.

  In this type of solar cell, an electric field derived from a potential difference is formed at the interface (p / p + interface) between the p-type silicon substrate and the p + layer. This electric field generated by the p + layer is mainly generated in the p-type silicon substrate and diffused to the back surface, and reflects the electrons into the p-type silicon substrate and selectively passes the holes to the p + layer. It has a function to make it. That is, this action brings about an effect (field passivation effect) of eliminating carriers and reducing carrier loss due to recombination of holes and electrons at the back surface interface of the solar cell.

  This is because the electric field generated at the p / p + interface spatially separates the distributions of holes and electrons in the vicinity of the back surface and causes a density difference on the back surface, thereby significantly reducing the probability of coupling between the two. Because. As a result, this contributes to improvement of the conversion efficiency of the solar cell.

  A conventional solar cell having such an aluminum alloy layer on the back surface is widely used as a mass production technique because it is relatively easy to manufacture. However, this method has the following problems in view of the ultimate high efficiency of solar cells.

  That is, as described above, most of the electrons are rejected by the electric field created by the p + layer, but still a few electrons with kinetic energy higher than the potential difference generated at the p / p + interface will overcome this potential difference. This leads to the p + layer / back electrode interface. However, in the conventional method, no deactivation treatment is performed on the p + layer / back electrode interface to reduce the interface recombination rate. In addition, since the p + layer itself is formed by highly doped aluminum to form recombination centers, the density of recombination centers is high and the quality as a semiconductor is poor.

  Therefore, a small number of electrons that have overcome the barrier are recombined with high probability and disappear in the p + layer or at the p + layer / back electrode interface. This electron annihilation causes the loss of the electrical output of the solar cell. Become.

  Further, it is desirable that a part of the light incident from the front surface of the solar cell and reaching the back surface is reflected on the back surface to return to the inside of the solar cell again and contribute to power generation. However, in the conventional method, the light absorption rate in the p + layer is large. For this reason, most of the light reaching the back surface of the solar cell is absorbed by the p + layer, and the electron-hole pairs generated at that time immediately recombine and disappear, so that they do not contribute to the electrical output. . Furthermore, since the physical shape of the interface between the p + layer and the electrode is generally uneven and is not a smooth flat surface, the reflectance of the reaching light decreases due to this uneven shape.

  As described above, the conventional solar cell has a problem that the light reaching the back surface of the solar cell cannot be effectively used for power generation.

  In response to this problem, development of backside passivation solar cells is underway with the aim of future replacement of the conventional solar cells. The present solar cell is different from the conventional solar cell. Specifically, by covering the back surface of the solar cell with a passivation film, the dangling bonds that inherently exist at the interface between the substrate and the passivation film and cause recombination are terminated. That is, the back surface passivation type solar cell does not attempt to reduce the carrier recombination velocity by the electric field generated at the p / p + interface, but rather reduces the density of the recombination centers on the back surface itself and reduces the carrier recombination. Based on the idea.

  Therefore, in this solar cell, when the performance of the back surface passivation film is sufficiently high, a low-quality p + layer as a semiconductor becomes unnecessary, and recombination loss caused by the p + layer and large light absorption in the p + layer are eliminated. be able to. Furthermore, the interface between the passivation film and the substrate can be produced smoothly, and the light reflectance at the interface can be increased.

  The back surface passivation film used for the back surface passivation type solar cell contains silicon and hydrogen. The dangling bonds that are inherently present at the interface between the p-type silicon substrate and the back surface passivation film and cause recombination are terminated by the silicon element and the hydrogen element of the back surface passivation film, so that they are not recombination centers. Therefore, the holes and electrons that have reached the back surface are not captured by dangling bonds, but are reflected at the interface as if nothing happened, and return to the p-type silicon substrate.

  This idea looks very promising from one point of view of interface deactivation. However, there is a problem when applied to actual solar cells. A portion covered with a passivation film and an electrode are adjacent to each other on the front or back surface of the solar cell. When an interaction between two parties is considered, a simple scenario as described above is not necessarily established. That is, in the backside passivation type solar cell, there is a problem that a characteristic deterioration phenomenon known as parasaimatic shunting occurs in the world, and the improvement in characteristics is limited.

  Here, the principle of the occurrence of the parasitic shunting phenomenon will be described with reference to FIGS.

  FIG. 8 is a schematic view showing a structural example of a back surface passivation type solar cell. In the figure, an n + layer 82 and an antireflection film 83 are formed on the surface of a P-type silicon substrate 81. In order to take out an electrical output also in the back surface passivation type solar cell, the back surface electrode layer 85 is required in addition to the front surface electrode layer 84. In this figure, the back electrode layer 85 is in contact with the P-type silicon substrate 81, and the opening region of the passivation film 86 for taking out electrical output from the P-type silicon substrate 81 is referred to as a back electrode 89.

  Since it is necessary to take out an electrical output from the P-type silicon substrate 81 at a portion where the back electrode 89 is formed, a passivation film 86 (in many cases, an insulator film) cannot be deposited. Since the back surface electrode 89 is provided in a dot shape or a stripe shape on a part of the back surface, it occupies several percent of the entire back surface. As a result, the portion covered with the passivation film and the back electrode portion are adjacent to each other on the back surface of the back surface passivation type solar cell.

  At this time, in the case of the back surface passivation type solar cell using the P type silicon substrate 81, if a positive fixed charge is present at the interface between the passivation film 86 and the P type silicon substrate 81 (hereinafter referred to as the back surface interface). For example, in the P-type silicon substrate 81 immediately below the passivation film 86, an electric field is generated by this fixed charge. As a result, holes are expelled from the interface and electrons are attracted to the interface. As a result, an inversion layer 87 in which electrons are higher than the hole density appears although it is p-type silicon immediately below the passivation film 86.

The appearance of the inversion layer 87 has the following two effects.
(1) The movement of electrons in the back surface interface in the lateral direction is facilitated.
(2) The electron density is increased in the contact region between the back electrode 89 and the passivation film 86.

  As described above, since the holes selectively passed from the P-type silicon substrate 81 to the p + layer gather in the contact region described in the above (2), that is, the peripheral portion of the back electrode 89, The hole density is higher than that inside the substrate.

  In the contact region between the back electrode 89 and the passivation film 86, the electron density is increased by the above (2), so that both the electron and hole densities are increased. Further, since this contact region is a materially discontinuous region in which the electrode material and the passivation film material are adjacent to each other, a high inaccurate level region 88 is generated due to the discontinuity.

  Therefore, as shown in FIG. 9 (a schematic diagram showing an enlarged contact region between the passivation film and the electrode portion in the structure example of FIG. 8), in the vicinity of the boundary between the back electrode 89 and the passivation film 86, the electrons 91 and Since the value of the density product with the holes 92 is large, the probability of collision recombination of these increases, and in addition, the illegal level density in the high illegal level region 88 also increases, resulting in a large carrier loss due to the interaction. Occurs.

  According to the above (1), electrons that are commensurate with such carrier loss are supplied by the inversion layer at the interface between the passivation film 86 and the p-type silicon substrate 81.

  That is, (2) creates a soil with high carrier recombination loss, and (1) establishes a transport path for efficiently transporting electrons to compensate for the lost electrons. This phenomenon is known as a parasite shunting phenomenon, and causes a serious deterioration in various characteristics of the back surface passivation type solar cell.

In the back side passivation solar cell, in order to suppress the occurrence of para Sai tick shunting phenomenon reduces the positive interface fixed charge density Q _f at the interface between the passivation film and the silicon substrate, preferably a negative value Is required.

As a specific countermeasure, Non-Patent Document 1 discloses a technique for suppressing the occurrence of the parasitic shunting phenomenon by devising the passivation film material and setting the interface fixed charge density _Qf to a negative value. Has been. In the method according this publication, as the passivation film material, by typically employing alumina interfacial fixed charge density Q _f at the interface between the passivation film and the silicon substrate becomes a negative value, the interface fixed charge density Q _f attracts holes in the silicon to the interface. Then, since the high-density carriers are the same type (holes) in both the vicinity of the electrode part where holes are collected in the p + part immediately below and the passivation film part, recombination of electrons and holes is caused. It becomes difficult to occur, and it is possible to avoid the occurrence of the parasitic shunting phenomenon and to suppress the deterioration of the characteristics of the back surface passivation type solar cell.

B. Hoex, J. Schmidt, R. Bock, PP Altermatt, MCM van de Sanden, and WMM Kessels, “Excellent passivation of highly doped p-type Si surfaces by the negative-charge-dielectric Al2O3”, Appl. Phys. Lett 91, 112107 (2007)

However, if alumina is used as the passivation film material as described above, the interface fixed charge density _Qf becomes a negative value, and even if it is possible to avoid the occurrence of the parasite shunting phenomenon, alumina is a solar cell. It is not often used as a deactivation film material. This is because it is necessary to deposit using a special method such as ALD (Atomic Layer Deposition). In general, the deposition rate of alumina by ALD is extremely slow, so the throughput is low and it is not suitable for mass production as a passivation film.

Therefore, have inherently positive interfacial fixed charge density Q _f, if avoid the above problems by using the SiNx familiar as a solar cell material, it is the most simple method, so that the effect is high.

In using the SiNx having a positive Q _f, as described above, to reduce as much as possible positive Q _f at the interface between SiNx film (passivation film) and p-type silicon substrate (p-type semiconductor layer) Needs to be devised. The method is already well known.

For example, when forming a SiNx film, the fraction of ammonia gas contained in the gas flowing into the plasma is increased, the RF output for exciting the plasma is increased, the substrate temperature is decreased during film formation, or the pressure during film formation Q _f can be reduced by using a means such as adjusting Q.

However, when Q _f is reduced by the above means, there arises a problem that the interface state density D _it increases. Especially when applying heat to the SiNx film, at the interface between SiNx film and the silicon substrate applied serious damage, sometimes D _it increases.

In other words, when adopting film having a low Q _f (low Q _f film) as a back passivation film solar cell, when annealed at high temperature, compared to a film having a high Q _f (high Q _f film) _Therefore, the increase in Dit is likely to occur. Therefore, when the low _Qf film is used as the interface passivation of the solar cell, the film characteristics may be deteriorated through an annealing process (high temperature process such as electrode formation).

The present invention solves this problem. That is, the more it is a significant reduction in the para site tick shunting phenomenon lower while have a surface fixed charge density Q _f, even at a high temperature anneal, such as when making a back electrode, interface state density D _it It is an object of the present invention to provide a passivation film that can be sufficiently reduced.

  That is, an object of the present invention is to provide a solar cell that suppresses the occurrence of the parasite shunting phenomenon and realizes good characteristics.

The interface passivation structure according to the present invention is:
(1) a p-type semiconductor layer;
(2) a first passivation film stacked on and in contact with the p-type semiconductor layer;
(3) at least a second passivation film stacked on and in contact with the first passivation film,
(4) Interfacial fixed charge density Q _f1 when the first passivation film is laminated on and in contact with the p-type semiconductor layer, and the second passivation film alone on the p-type semiconductor layer. A relationship of Q _f1 <Q _f2 is established with the interface fixed charge density Q _f2 when stacked in contact with each other.

In the above configuration, when the relationship of Q _f1 <Q _f2 is established, the first passivation film functions to maintain the interface fixed charge density Q _f at the interface between the p-type semiconductor layer and the passivation film at a low level. To do. The second passivation film when annealed at high temperatures and serves to reduce the interface state density D _it.

Due to the combined action of the two types of passivation films having different properties, the interface fixed charge density _Qf can be kept low, and the interface state density _Dit is not deteriorated even by high-temperature annealing.

The hypothesis of the mechanism explaining the above effect is as follows. The passivation film contains hydrogen. Then, by exposing the passivation film to heat, the hydrogen is desorbed from the inside of the passivation film, and a part of the hydrogen reaches the p-type semiconductor layer interface. The hydrogen that has reached the interface terminates the dangling bonds present therein, to reduce the D _it by reducing carrier traps.

Here, at the interface fixed charge density Q _f taller Q _f film, but does not occur elimination of hydrogen when he does not have a high temperature, in the low interfacial fixed charge density Q _f low Q _f membranes, desorption of hydrogen Happens at low temperatures. Therefore, unlike the conventional, if the passivation film is a structure of one layer of low-Q _f film only, when annealing the film at a high temperature of about 750 ° C., since hydrogen in the film is lost already eliminated, Similarly, a lot of hydrogen does not exist at the interface between the passivation film and the p-type semiconductor layer. As a result, the number of dangling bonds is increased, and the interface state density _Dit is increased.

On the other hand, the passivation film of the present invention, together with a low Q _f film of the first layer, since it contains a high Q _f film of the second layer, even at a high temperature of about 750 ° C., high Q _f of the second layer Hydrogen can be sufficiently supplied from the membrane.

That is, according to the present invention, since the passivation film includes two passivation films having different film qualities, the first passivation film removes hydrogen deficient at the interface of the p-type semiconductor layer during high-temperature annealing. Even if the supply from is impossible, it can be compensated by supplying from the second passivation film resistant to high temperature annealing. Thus, the high-temperature annealing, suppressing the interfacial fixed charge density Q _f of the entire passivation film, and can be kept low even the interface state density D _it.

In the interface passivation structure according to the present invention,
(1) The p-type semiconductor layer is a silicon layer,
(2) The first passivation film and the second passivation film are silicon nitride films.

  Thereby, the silicon nitride film is widely used as a constituent material of the solar cell, and the positive interface fixed charge inherently included in the silicon nitride film can be kept low as already described. It becomes easy to cope with mass production of solar cells with good characteristics.

  The back surface passivation type solar cell according to the present invention includes the above-mentioned interface passivation structure as a back surface passivation structure.

The by interfacial passivation structure, it is possible to suppress the interfacial fixed charge density Q _f and interface state density D _it, it is possible to improve the characteristics of the back passivation solar cell.

The manufacturing method of the interface passivation structure according to the present invention is as follows:
(1) A first silicon nitride film is stacked on and in contact with the silicon layer by a vapor phase growth method using plasma, and then a second silicon nitride film is stacked on and in contact with the first silicon nitride film. Including steps,
(2) Regarding the value of the SiH ₄ / NH ₃ flow rate ratio as the flow rate ratio of each source gas used in the vapor phase growth method, the second silicon nitride film is obtained from the value at the time of stacking the first silicon nitride film. The value at the time of stacking is increased.

As described above, regarding the value of the SiH ₄ / NH ₃ flow rate ratio as the flow rate ratio of each source gas used in the vapor phase growth method, the second nitridation is based on the value at the time of stacking the first silicon nitride film. By increasing the value at the time of stacking the silicon films, an interface passivation structure that satisfies the relationship of Q _f1 <Q _f2 described above can be obtained.

  By forming the passivation film laminated on the p-type semiconductor layer to include two layers of passivation films having different properties, it is possible to provide a solar cell with excellent characteristics.

It is a block diagram which shows the interface passivation structure which concerns on embodiment of this invention. In a high-Q _f film and a low Q _f film is a graph showing the results of measuring the relationship between the annealing temperature and D _it. It is a graph showing the results of measurement of Q _f and D _it interfacial passivation film formed by changing the flow rate ratio of the mixed gas flowing through the plasma. It is a block diagram which shows the measurement sample of the solar cell which concerns on this invention. The interfacial passivation structure according to the conventional interfacial passivation structure and the present embodiment is a graph showing the results of measurement of Q _f and D _it before and after annealing. It is a schematic diagram which shows the structure of the solar cell which concerns on embodiment of this invention. It is a figure which compares and shows the various characteristics of the solar cell of a conventional structure, and the solar cell which concerns on this embodiment. It is a schematic diagram which shows the structural example of a back surface passivation type solar cell. It is a schematic diagram which expands and shows a part of structural example of the solar cell shown in FIG.

  Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention only to those unless otherwise specified. This is just an example.

(Interface passivation structure)
An interface passivation structure according to this embodiment is shown in FIG. The interface passivation structure 200 shown in FIG. 1 can be applied to, for example, a back surface passivation type solar cell that is one form of a photoelectric conversion element having a pn junction.

  As shown in the figure, the interface passivation structure 200 includes a p-type silicon substrate (hereinafter abbreviated as a silicon substrate) 201 as a p-type semiconductor layer, and a passivation film 202. The passivation film 202 includes a silicon nitride film (hereinafter abbreviated as SiNx film) 202a as a first passivation film stacked in contact with the silicon substrate 201, and a second passivation film stacked in contact with the SiNx film 202a. It has a two-layer structure including a SiNx film 202b as a film.

However, the properties of the SiNx film 202a and the SiNx film 202b are different as follows. That is, as will be described later, the interface fixed charge density Q _f1 when the SiNx film 202a is laminated on the silicon substrate 201 alone, and the SiNx film 202b is laminated on the silicon substrate 201 alone. The relationship of Q _f1 <Q _f2 is established between the interface fixed charge density Q _f2 of the above.

With the above configuration, even when the SiNx film 202a and the SiNx film 202b are annealed at a high temperature of about 750 ° C., the interface fixed charge density Q _f (hereinafter simply abbreviated as Q _f ) of the entire passivation film is suppressed, In addition, the interface state density D _it (hereinafter simply abbreviated as D _it ) can be kept low.

  Therefore, the characteristics of the solar cell including the interface passivation structure 200 as the back surface passivation film can be improved.

(The relationship between the annealing temperature and the _{D it)}
In the conventional method of reducing a positive Q _f at the interface between SiNx film and the p-type silicon substrate, already described that the problem of D _it is increased occurs.

Here, when a solar cell is manufactured, if heat is applied to the SiNx film through a high temperature process such as electrode formation, serious damage may be applied to the interface, and _Dit may increase. As a typical example, one experimental result is shown in FIG.

In the same experiment, two types of SiNx films were produced by changing the fraction of ammonia gas and silane gas that flow into plasma when forming the SiNx film. One is a passivation film having 3.0E + 12cm ^-2 or more high _{Q f} (high _{Q f} film), the other, a passivation film _{(low-Q f} film) having 1.9E + 12cm ^-2 following relatively low _{Q f} It is.

Note that 3.0E + 12 represents 3.0 × 10 ¹² , and 1.9E + 12 represents 1.9 × 10 ¹² . _The unit of _{D it} is ^{min ·} cm -2 ^{· eV -1.}

According to the figure, it can be seen that there is a common behavior in the two types of films. This is because, if slide into higher annealing temperature above 500 ° C. in the high-temperature process, D _it is once decreased, as increases again, is to show the behavior of the U-shape. However, the difference between the two is that, in the high Q _f film, the temperature range in which D _it is minimal exists up to 700 ° C. or more, whereas in the low Q _f film, D _it is around 600 ° C. It has become extremely small and has already started increasing at 700 ° C.

In other words, when employing the low Q _f film as a back passivation film solar cell, when annealed at high temperature, compared to the high Q _f films, the increase in D _it results that easily occurs is obtained. Therefore, when using a low-Q _f film interface passivation of the solar cell, by going through the high temperature processes such as electrode formation, it caused an increase in interfacial D _it, there is a possibility that exacerbate characteristics.

  The present inventor has already proposed a preferable range of Qf in which Dit can be suitably reduced by a general annealing process performed after the formation of the SiNx film in consideration of the problem of increase in Dit due to the high temperature process. (Japanese Patent Application No. 2010-172659). Here is a brief description of the proposal.

  The inventor of the present application has found that when Qf becomes smaller than a certain value, Dit starts to increase and the characteristics of the solar cell deteriorate due to the annealing process. That is, there is a lower limit point in Qf that can suitably reduce Dit by the annealing process.

  (A) and (b) of FIG. 3 are part of the measurement results related to the proposed content, and show the measurement results derived from the lower limit point of Qf.

  FIG. 3 (a) shows the result of measuring Qf by changing the flow rate ratio (SiH4 / NH3) of the source gas, and FIG. 3 (b) shows the result of changing the flow rate ratio (SiH4 / NH3) of the source gas. The result of having measured the change of Dit is shown.

  Note that sccm (standard cubic centimeter per minute) is a unit of flow rate, and represents a gas volume (cm 3) per minute normalized under conditions of 1 atm and constant temperature.

  From FIG. 3A, it can be seen that when the flow rate ratio (SiH 4 / NH 3) of the source gas is decreased, Qf generally decreases, but there is almost no difference in Qf depending on the presence or absence of the annealing step.

  On the other hand, looking at Dit shown in FIG. 3B, there is no significant difference even if the SiH 4 / NH 3 flow rate ratio is changed without the annealing step.

  However, when the annealing process is performed, when the SiH 4 / NH 3 flow rate ratio is 25/200 or more, Dit is reduced more favorably than when the annealing process is not performed, whereas the SiH 4 / NH 3 flow rate ratio is 19 / It can be seen that when it becomes smaller than 200, Dit does not decrease and increases conversely, compared with the case without the annealing step, and as a result, approaches Dit without the annealing step.

  From this measurement result, Qf = 1.3E + 12 cm −2 corresponding to the SiH 4 / NH 3 flow rate ratio of 19/200 was derived as the lower limit point of Qf.

  This means that in order to suppress the parasite shunting phenomenon, it is desired to reduce Qf, but a SiNx film whose Qf is smaller than the lower limit is weak against annealing, and this is used for backside passivation of a backside passivation type solar cell. This means that it is difficult to use as a film.

(Demonstration of effect of the present invention by experiment)
In order to verify the effect of the present invention, a sample shown in FIG. 4 was prepared as the interface passivation structure 200 and a verification experiment was performed. In this experiment, the SiNx films 202a and 202b were produced under the following conditions using a plasma CVD apparatus using a parallel plate plasma source.

Pressure: 100Pa
Substrate temperature: 450 ° C
RF power density: 0.074 W / cm ²
Distance between electrodes: 30mm
Under these conditions, the SiNx film 202b having a high Q _f (≈2.5E + 12 cm ⁻² ) was obtained by setting the SiH ₄ / NH ₃ flow rate ratio to 25 sccm / 50 sccm. Also, the SiNx film 202a having a low Q _f (≈9.8E + 11 cm ⁻² ) was obtained by setting the SiH ₄ / NH ₃ flow rate ratio to 13 sccm / 200 sccm.

The SiNx films 202a and 202b were formed on a 2.0 Ωcm silicon substrate 201 manufactured using an Fz (Floating-Zone) method so that the entire film thickness was a constant value of 306 nm. The first of the passivation film (low _{Q f} film) thickness x of the SiNx film 202a as was deposited varied in the range of x = 40~160nm. Next, the SiNx film 202b as a second passivation film (high _{Q f} film) is deposited only remaining thickness min (306-x) nm, to prepare a passivation film 202.

This sample was put into a firing furnace and annealed under the general electrode firing conditions of a solar cell (750 ° C., 12 seconds). Thereafter, the interface between the passivation film 202 and the silicon substrate 201 of this sample was evaluated using a CV (Capacitance-Voltage) method. The evaluation _was carried out by measuring the _{D it} and _{Q f.}

  Note that three types of samples were prepared by changing the thickness x of the SiNx film 202a to three types of 40, 80, and 160 (nm).

(A) of FIG. 5 is a measurement result of Q _f in three types of samples. Note that for reference, SiNx film 202a equivalent SiNx film as a low _{Q f} membranes, and the SiNx film 202b equivalent SiNx film as a high _{Q f} films were deposited on the silicon substrate 201 singly sample The measurement results are also shown.

From the drawing, on the three samples according to the present invention, different SiNx film 202a of the nature of the passivation film 202 is a two-layer, despite the structure comprising 202b, _{Q f} after annealing, the SiNx film 202a It can be seen that the value is close to Q _f of the sample deposited alone.

Thus, the Q _f of the composite film type passivation film 202 is substantially equal to the Q _f value of the first layer passivation film (202 a) in contact with the interface of the silicon substrate 201.

Next, consider the results of measurement of _{D it} shown in (b) of FIG. The sample is deposited on the silicon substrate 201 a SiNx film 202b equivalent SiNx film alone, by executing the _{annealing, D it} is reduced to 1/3 or less, which is about 1E ^{+ 11cm} -2 ^{eV -1} .

On the other hand, in the sample deposited on the silicon substrate 201 a SiNx film 202a equivalent SiNx film alone, _{D it} is not reduced even after the annealing, resulting in _{D it} is remains high level of 3E ^{+ 11cm} -2 ^{eV -1} Yes.

In contrast, in the three samples with a passivation film 202 which the value of _{Q f} consists height two SiNx film, _{D it} becomes 7E ^{+ 10cm} -2 ^{eV -1} after annealing, the _annealing is _{D it} 1 It can be seen that it is greatly reduced to about / 5. Percentage of reduction, also by changing the thickness of the SiNx film 202a of the low _{Q f} in the range of 40～160Nm, showed no change.

Conventionally, due to the reasons described above, in an interface passivation structure composed of a single layer of a low _Qf film, it was impossible to reduce _Dit when high-temperature annealing at 750 ° C. was performed. However, it has been found that the interface passivation structure in which two types of passivation films having a high and low Q _f value are stacked as in the present invention can achieve a low Q _f that maintains a low _Dit even when annealing is performed at 750 ° C. It was.

(Configuration of solar cell)
In the present embodiment, a solar cell 100 having a structure shown in FIG. 6 is employed as a back surface passivation type solar cell that is one form of the photoelectric conversion element. In addition, this solar cell 100 is used also as a model structure for numerical model calculation mentioned later.

  In the figure, on the surface side (light incident side) of the solar cell 100, an n layer 102 is formed on the surface of a p-type silicon substrate 101 (hereinafter referred to as a silicon substrate 101) as a p-type semiconductor layer. . On the other hand, on the back surface side of solar cell 100, p + layer 114 is formed immediately above the back electrode forming portion. The solar cell 100 further includes an antireflection film 103 and a back surface passivation film 106 (hereinafter referred to as a passivation film 106). The antireflection film 103 is formed so as to cover the surface of the n layer 102 and also has a function as a passivation film on the surface side. The passivation film 106 is formed so as to cover a part of the surface of the silicon substrate 101 to the p + layer 114.

  That is, the solar cell 100 has a structure in which the p + layer 114 and the silicon substrate 101 that is n-doped by the positive charge 112 in the passivation film 106 are in contact with each other at the interface of the silicon substrate 101 on the back surface side of the solar cell 100.

  Further, a region where the antireflection film 103 is partially removed is formed on the surface of the solar cell 100, and a surface electrode layer 104 is provided in the region. A back electrode layer 109 is formed on the back surface side of the passivation film 106. That is, the back electrode layer 109 is formed by removing a part of the passivation film 106 in advance at a position facing the front electrode layer 104 across the silicon substrate 101. The positive charge 112 in the figure schematically shows the state of the interface fixed charge that exists fixedly at the interface between the silicon substrate 101 and the passivation film 106.

In the passivation film 106, by applying a two-layer structure of the passivation film 202, to achieve a low Q _f to maintain a low D _it, it is possible to obtain a solar cell 100 having good properties.

(Numerical model calculation)
Based on the realistic design of the solar cell, the inventor of the present application calculates a numerical model using a computer for both the case where the stacked passivation film according to the present invention is applied to the solar cell and the case where it is not applied to the solar cell. Went. A detailed procedure for deriving the solar cell characteristics by the numerical model calculation will be described below.

  The basic principle of the numerical model calculation is based on the Poisson equation that governs carrier transport inside the substrate and the current continuity formula. It is to solve the substituted basic equation. In order to execute the numerical model calculation, the basic equation is linearized and then discretized to obtain simultaneous equations that can be numerically calculated. Then, by solving the simultaneous equations, the potential distribution and the carrier density distribution inside the solar battery cell can be obtained, so that the current flowing between the electrodes can be obtained.

  In performing the numerical model calculation related to the model structure of the solar cell 100, carrier movement is performed in two directions: a direction parallel to the interface between the passivation film 106 and the silicon substrate 101 and a direction perpendicular to the interface. There is a need to consider. Therefore, a general-purpose device simulator (for example, Sentraus manufactured by Synopsys) that can handle carrier movement in two dimensions was used.

  Further, in the specific calculation, the substrate thickness, impurity concentration, electron lifetime, surface recombination velocity, surface impurity concentration, back surface passivation film contact width (in FIG. The width of the passivation film 106 and the silicon substrate 101 in contact with each other) Parameters such as sunlight spectrum / illuminance, surface reflectance, surface internal reflectance, back surface internal reflectance, and shadow loss directly under the electrode are actual values. It was set to be appropriate.

  However, among the various parameters described above, the width (Wtot) of one section formed by the adjacent back surface electrode 105 on the back surface side of the solar cell 100 was set to 200 μm. However, in the numerical model calculation, Wtot / 2 = 100 μm was used as shown in FIG. 6 in view of the symmetry of the solar cell structure.

  Further, the thickness of the silicon substrate 101 including the n layer 102 was set to a relatively small value of 50 μm in order to strongly reflect the influence of the back surface passivation structure in the calculation result.

The configuration of the SiNx film deposited on the back surface was set to the following three types based on the data obtained in the verification experiment.
Single layer construction of <A> high _{Q f} film Q _f ~2.5E + 12cm ^-2, annealed ^{_{D it = 1.0E + 11cm -2 eV}} -1
<B> single layer structure of a _{low-Q f} film Q _f ~9.8E + 11cm ^-2, annealed ^{_{D it = 3.0E + 11cm -2 eV}} -1
<C> Two-layer configuration of A and B, Q _f ˜9.8E + 11 cm ⁻² , D _it after annealing = 7.0E + 10 cm ⁻² eV ⁻¹
A and B are conventional passivation films, and C is a passivation film according to the present invention.

  FIG. 7 is a table showing numerical model calculation results of various characteristics of the solar cell when the passivation films having the above-described configurations A to C are used. Each characteristic shown on the first line of the figure is preferable as the numerical value is higher.

First, A and B are compared. Each characteristic of B is worse than that of A. Since B has a smaller _Qf than A, the parasitic shunting phenomenon does not occur, and the characteristics should be improved. However, since B has a worse value of D _it than A, the effect of improving the characteristics due to the small Q _f is negated, and the characteristics are deteriorated.

On the other hand, for C, all the characteristics are improved by two effects that Q _f is equal to B, smaller than A, and _Dit is low.

  As described above, the numerical model calculation also proves that the characteristics of the solar cell can be improved by the interface passivation structure of the present invention.

(Example of manufacturing process of solar cell according to the present invention)
The following explanations from the first to the ninth show examples of manufacturing processes of the solar cell 100 according to the present embodiment.

  First, the silicon substrate 101 is cleaned by an RCA cleaning method developed by RCA. Subsequently, etching is performed at a liquid temperature of about 90 ° C. using a mixed solution of NaOH aqueous solution and isopropyl alcohol, so that the surface of the silicon substrate 101 (the light receiving surface side in the solar cell 100) has a minute height of several μm. Unevenness is formed. Thus, forming a fine concavo-convex structure on the surface of the silicon substrate 101 is called texture etching. By performing texture etching in this manner, light on the surface of the silicon substrate 101 is scattered and reflection can be suppressed, so that the light utilization efficiency of the solar cell 100 can be increased.

  The silicon substrate 101 can be obtained, for example, by adding a small amount of a trivalent element such as boron, aluminum, or gallium to a silicon substrate. A silicon substrate using single crystal silicon is also an object to which the present invention is applied.

  Moreover, you may use the reactive ion etching method for the said texture etching.

Second, the silicon substrate 101 was placed in a high-temperature gas containing POCl ₃ , phosphorus was thermally diffused, and n + layers containing impurities were formed on the light receiving surface (front surface) side and the back surface side of the silicon substrate 101.

As a method for diffusing phosphorus on the surface of the silicon substrate 101, for example, in addition to the above-mentioned vapor phase diffusion method using POCl ₃ , a coating diffusion method using P ₂ O ₅ , ion implantation for directly diffusing P ions. There are laws.

  Third, a SiNx film was deposited as an antireflection film 103 on the light-receiving surface side (light incident side of the solar cell) of the silicon substrate 101 by plasma CVD (Chemical Vapor Deposition). The SiNx film also has a function as a surface passivation film.

  Here, as the antireflection film 103 having a passivation effect and also serving as an inactivating film, an aluminum oxide film, a silicon oxide film, a titanium oxide film, or the like is used in addition to the SiNx film. However, in the case of a solar cell using a polycrystalline silicon substrate, it is preferable to use a SiNx film containing hydrogen from the viewpoint of improving the conversion efficiency.

  Further, as a method of forming the antireflection film 103, in addition to the plasma CVD method, a CVD method such as a catalytic CVD method, an atmospheric pressure thermal CVD method, a low pressure thermal CVD method or a photo CVD method, a vacuum deposition method or a sputtering method. A PVD (Physical Vapor Deposition) method such as can be used. In the case where a SiNx film is used as the antireflection film 103, it is preferably formed by a plasma CVD method from the viewpoint of easy control of the film thickness.

  Fourth, a protective tape was applied to the surface (light receiving surface) of the silicon substrate 101 and immersed in a solution of nitric acid: hydrofluoric acid = 3: 1. Thereby, while the n + layer existing on the surface of the silicon substrate 101 remains, the n + layer existing on the back surface of the silicon substrate 101 is removed, and the back surface of the silicon substrate 101 is exposed. Thus, the n layer 102 is formed.

  Fifth, a process corresponding to the method for manufacturing the interface passivation structure of the present invention is performed. That is, as a vapor phase growth method using plasma, a plasma CVD method is used to form a SiNx film 202a (first silicon nitride film) and a SiNx film 202b (second silicon nitride film) that form the passivation film 106 on the back surface. Were deposited in this order. In the present embodiment, SiNx containing hydrogen is used for the passivation film 106 from the viewpoint of improving conversion efficiency in a solar cell using a polycrystalline silicon substrate.

In the plasma CVD method of this embodiment, the SiNx films 202a and 202b are formed by using two types of monosilane (SiH ₄ ) and ammonia (NH ₃ ) as source gases in addition to nitrogen (N ₂ ) as a dilution gas. The gas was used.

As described above, during lamination of the SiNx film 202a of the low _{Q f} was set to _SiH 4 / NH ₃ flow ratio for example 13 sccm / 200 sccm. Also, during lamination of the SiNx film 202b of the high and low _{Q f,} from the value at the time of stacking of the SiNx film 202a, a larger value at the time of stacking of the SiNx film 202b, setting the _SiH 4 / NH ₃ flow ratio for example 25 sccm / 50 sccm .

Generally, Q _f is small at the interface between SiNx film and the silicon containing a large amount of nitrogen, it is known to increase at the interface between SiNx film and the silicon rich silicon. Therefore, in the step of forming the SiNx films 202a and 202b, the flow ratio or mixing ratio (SiH ₄ / NH ₃ ) of monosilane (SiH ₄ ) and ammonia (NH ₃ ) used in the vapor phase growth method using plasma is set. by varying, by varying the composition ratio of nitrogen and silicon SiNx film can be adjusted to control the value of Q _f.

In addition to the above method, any one of the methods such as changing the pressure during vapor phase growth using plasma, RF power, discharge gap, temperature of the silicon substrate 101, and adding hydrogen to the source gas, or also by any combination of multiple techniques, changing the composition ratio, it is possible to control the value of Q _f.

In the above description, SiNx is used as the material of the layer constituting the passivation film 106, but the type of the passivation film 106 that can be used in the present invention is not limited to this, and other materials such as SiO ₂ are used. It may be adopted.

  Sixth, a hole corresponding to the back electrode layer 109 was formed in the passivation film 106 by using a method such as photolithography.

  Seventh, an aluminum film was deposited on the entire back surface to form the back electrode layer 109. The film thickness of the aluminum to vapor-deposit is 2 micrometers, for example. As a method for forming the back electrode layer 105, a printing method using a paste material mainly composed of aluminum and glass frit is preferable from the viewpoint of cost. In addition to the vacuum deposition method, a sputtering method can also be used.

  Eighth, an annealing process was performed. By this annealing step, aluminum diffuses from the back electrode layer 109 to the silicon substrate 101 as a p-type impurity, and an aluminum alloy part (p + layer 114) is formed in the silicon substrate 101 at a contact portion with the back electrode layer 109. The This process produced a good p + layer 114 with a high hole density. This contributes to increasing the photoelectric conversion efficiency of the solar cell 100.

  Ninth, the surface electrode 104 was printed on the light-receiving surface side of the silicon substrate 101 using a conductive paste and annealed. Due to the fire-through phenomenon that occurs at this time, the surface electrode 104 penetrates the antireflection film 103 and reaches the n layer 102. As a result, an electrical output can be taken out from the surface electrode 104.

  The material which comprises the surface electrode layer 104 is not specifically limited, For example, materials, such as aluminum, silver, titanium, palladium, or gold conventionally used in the field of a solar cell, can be used. Further, the method for forming the surface electrode layer 104 is not particularly limited, and for example, a screen printing method or a vacuum deposition method can be used. However, the screen printing method is used from the viewpoint of improving mass productivity and reducing manufacturing costs. Is preferred.

  The annealing conditions for performing the above-described annealing step are the same as the general annealing conditions used in the back surface passivation solar cell manufacturing process. For example, the annealing step is performed at a temperature of about 700 ° C. to 800 ° C.

  The solar cell 100 can be manufactured as described above.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims, and other embodiments obtained by appropriately combining the technical means disclosed in the above-described embodiments. The form is also included in the technical scope of the present invention.

  The present invention relates to a solar cell constituting a photoelectric conversion element, and more specifically, can be suitably used for, for example, a back surface passivation type solar cell.

100 solar cell 101 p-type silicon substrate (p-type semiconductor layer, silicon layer)
106 Back surface passivation film 200 Interface passivation structure 201 p-type silicon substrate (p-type semiconductor layer, silicon layer)
202a SiNx film (first passivation film, first silicon nitride film)
202b SiNx film (second passivation film, second silicon nitride film)

Claims (4)

a p-type semiconductor layer;
A first passivation film stacked on and in contact with the p-type semiconductor layer;
A second passivation film stacked on and in contact with the first passivation film,
Interfacial fixed charge density Q _f1 when the first passivation film is laminated on and in contact with the p-type semiconductor layer, and the second passivation film is laminated on and in contact with the p-type semiconductor layer. An interface passivation structure characterized in that a relationship of Q _f1 <Q _f2 is established between the fixed interface charge density Q _f2 and the interface fixed charge density Q _f2 . The p-type semiconductor layer is a silicon layer;
The interface passivation structure according to claim 1, wherein the first passivation film and the second passivation film are silicon nitride films. A back surface passivation type solar cell comprising the interface passivation structure according to claim 1 or 2 as a back surface passivation structure. A step of laminating a first silicon nitride film in contact with the silicon layer by a vapor phase growth method using plasma, and then laminating a second silicon nitride film in contact with the first silicon nitride film. ,
Regarding the value of the SiH ₄ / NH ₃ flow rate ratio as the flow rate ratio of each source gas used in the vapor phase growth method, the value at the time of stacking the second silicon nitride film is greater than the value at the time of stacking the first silicon nitride film. A method for manufacturing an interface passivation structure, wherein the value of is increased.

Images