Patent Application
20160079454
The invention relates to the field of industrial processes for forming a semiconductor layer, in particular for photovoltaic applications.
The invention relates more particularly to a process for forming a type I-III-VI2 semiconductor layer by heat treatment and chalcogenization, in at least one furnace chamber, of a type I-III metal precursor deposited on a substrate.
As illustrated in
As shown in the photograph of
In addition, as represented in
In addition, as represented in
In this context, the present invention improves the situation by overcoming one or more of the limitations mentioned above.
To this end, the process of the invention, which is otherwise in accordance with the preamble provided above, is essentially such that it comprises:
The process thus advantageously enables formation of a semiconductor layer offering a gain of around 4% in conversion efficiency compared to a semiconductor layer formed according to the formation process illustrated in
According to one feature of the invention, the first temperature is between 480° C. and 520° C.
According to another feature, the first temperature is equal to 505° C.
The formation process is thus advantageously optimized according to the temperature at which the chalcogenization step begins.
According to another feature, during the heating step, the temperature increases at a rate of 3.5° C./sec, plus or minus 1° C./sec.
The process thus advantageously enables finely tuned control of densification of the metal precursor.
According to another feature, the chalcogenization step consists of a selenization step by injection of a gaseous mixture of selenium and dinitrogen into at least one furnace chamber.
The process thus advantageously enables selenization of the metal precursor at a chosen moment in the evolution of the temperature over time.
According to another feature of this chalcogenization step, the gaseous mixture of selenium and nitrogen is obtained by heating selenium to a temperature of 500° C., plus or minus 20° C., to obtain a high partial pressure of the selenium.
The process thus advantageously makes it possible to optimize the amount of selenium that can be captured in the semiconductor layer formed, relative to the amount of copper in the metal precursor, and enables formation of a semiconductor layer at an industrial rate.
In another feature of this chalcogenization step, injection of the gaseous mixture of selenium and dinitrogen occurs at a volumetric flow rate of 13 standard liters per minute, plus or minus 3 standard liters per minute.
According to another feature, the chalcogenization step lasts 5 minutes, plus or minus 1 minute.
The process thus advantageously enables formation of a semiconductor layer at an industrial rate.
According to another feature, the ratio of the total amount of chalcogen incorporated into the substrate and precursor to the amount of metal precursor is between 1.4 and 2.2.
The process advantageously provides satisfactory stability of the formed semiconductor layer within this range of values for said ratio.
According to another feature, the furnace comprises at least one series of chambers, and the heating step is carried out in a first chamber of the series and the chalcogenization step is carried out in a second chamber of the series.
According to another feature, at least the second chamber of the furnace is maintained at a pressure that is 20 to 200 Pa below atmospheric pressure.
The process thus advantageously ensures a satisfactory level of safety.
According to another feature, the second temperature (the stabilization temperature) is between 570° C. and 590° C.
The invention also relates to a type I-III-VI2 semiconductor layer obtained by the process according to any of the features set forth above.
According to one feature of said semiconductor layer, it has a microstructure composed of grains of different sizes corresponding to a full width at half maximum of the {112} XRD peak of the CIGSe of between 0.16° and 0.18°.
The semiconductor layer, or equivalently the absorber, thus preferably provides a gain in conversion efficiency of about 4% compared to a semiconductor layer formed according to the formation process illustrated in
According to another feature of the semiconductor layer, it comprises several layers of different compositions, including a lower layer which is a CuGaSe2 layer.
The semiconductor layer thus advantageously grips the carrier layers more securely, and in particular a layer of MoSe2 composition.
The invention further relates to a furnace for carrying out the process according to one of its features set forth above.
Said furnace comprises:
the latter communicating the temperature measurements for each chamber to the control means for the purposes of controlling each heating device, so as to ensure a uniform increase in temperature in the first chamber to a first temperature of between 460° C. and 540° C., and the maintaining of the temperature in the second chamber at a second temperature, for stabilization, of between 550° C. and 600° C., said furnace further comprising injection means for injecting an inert gas into the first chamber, and
said furnace further comprising injection means for injecting, into the second chamber, a gaseous mixture of selenium and dinitrogen having a temperature of between 480° C. and 520° C.
Other features and advantages of the invention will become apparent from the following description provided for indicative and non-limiting purposes, with reference to the accompanying drawings, wherein:
a to 2d show different stacks of layers corresponding to different phases of the formation process according to the invention,
a and 3b are graphs showing the change in temperature over time within the furnace, illustrating the start and end of the chalcogenization step of the formation process according to the invention and to the prior art respectively,
a and 4b are photographs, obtained by microscopy, of the microstructures formed by the formation processes according to the invention and to the prior art respectively,
a and 8b are adjoining graphs, each representing the photovoltaic cell conversion efficiency as a function of the ratio of the molar amount of copper to the molar amount of gallium and indium in the metal precursor, in particular when this ratio varies between 0.6 and 1.2, said photovoltaic cells being respectively obtained by the formation process of the invention and by the formation process of the prior art,
In the following description, each layer is described as formed or deposited “on” or “under” another layer or component, which means that this layer may be formed “directly” or “indirectly” (with another layer or component between) on or under another layer or another component. In addition, relative criteria such as “lower”, “upper”, or “intermediate” define each layer as illustrated in the accompanying drawings.
In the drawings, the thickness or size of each layer is exaggerated or omitted or only schematically represented for the purposes of convenience and clarity of the explanation. In addition, the thickness or size of each layer does not reflect its actual size or thickness.
The formation process S firstly comprises the providing of a substrate 3. The substrate has, for example, a width and length equal to 60 cm and 120 cm to provide a surface area of 7200 cm2.
As illustrated in
Next, the formation process S comprises a step of depositing on the substrate 3 a stack of layers of group IB and IIIA elements, such as copper (Cu) and indium (In) respectively. Another group IIIA element, more particularly gallium, can also be used in association with the indium and copper. The use of gallium makes it possible to increase the energy band, the open circuit voltage (OCV), and the conversion efficiency of the photovoltaic cells formed. In addition, it should be noted that gallium has a melting point of 29.8° C., which is close to room temperature, meaning it has high diffusion; its concentration profile in the semiconductor layer 1 to be formed must therefore be closely controlled, which is what the present process intends to achieve, particularly by continuous control of the temperatures to which the different stacks of layers are exposed during the various phases of the formation process, as illustrated in
As illustrated in
By way of non-limiting example, the deposition step consists of a step of electrodeposition of at least one of the layers of the stack. All the layers of group IB and IIIA elements can advantageously be electrodeposited, electrodeposition being an industrial deposition technique that is particularly fast and inexpensive.
In addition, it should be noted that the layers of the stack are preferably electrodeposited, at least in the sense that the values of the parameters of the various heat treatments said below to be preferred are particularly suited to this technique. Depositing at least one layer of the stack by physical vapor deposition for example is likely to result in having to specifically identify other preferred values for these parameters, although these are likely to remain within the range of values shown below, in particular by retaining the principle in the sense of the invention of uniformly increasing the temperature, followed by a plateau, during the heating S1 and chalcogenization S2 steps.
Next, the formation process S comprises an annealing step to form the type I-III metal precursor 2 on the substrate 3.
The annealing step at least consists of heating the stack of layers of group IB and IIIA elements on substrate 3 to a temperature between 80° C. and 110° C., preferably 90° C., which is maintained for 20 to 40 minutes, preferably 30 minutes, to allow interdiffusion between the layers.
This annealing is said to be “soft” because the maximum annealing temperature is relatively low and therefore its duration can be relatively long. For example, adequate diffusion of the gallium layer through the indium layer to the substrate 3 is thus achieved.
As shown in
The “soft” annealing step can end with a phase of cooling to ambient temperature.
As illustrated in
Densification of the metal precursor is understood here to mean a reorganization of the metal atoms, resulting in a mixture of dense alloys which contain phases containing only I and III elements as well as mixed phases of I-III elements, without creating porosities.
The present invention further relates to a furnace 4 for carrying out at least the heating S1 and chalcogenization S2 steps described below.
As illustrated in
at least a first chamber 400 and second chamber 410,
means of conveyance 40, or conveyor, from one chamber to the next,
a heating device 42 for each chamber,
control means 44, or controller, for each heating device 42, and
measurement means 46, or sensors, for measuring the temperature in each chamber 400, 410.
The temperature measurement means 46 communicate the temperature measurements for each chamber 400, 410, 420 to the control means 44. The control means control each heating device 42 to ensure at least, in the first chamber 400, a uniform increase in temperature to a first temperature T1 of between 460° C. and 540° C., and in the second chamber 410, that the temperature is maintained at a second temperature T2, for stabilization, of between 550° C. and 600° C.
In its broadest sense, the heating step S1 under an inert atmosphere consists of a step in which the temperature increases uniformly to the first temperature T1 of between 460° C. and 540° C. The first temperature T1 may more particularly be between 480° C. and 520° C., and is preferably equal to 505° C.
More particularly, the fact that the heating step S1 is carried out under an inert atmosphere is understood to mean that the chamber 400 or chambers of the furnace in which the heating step S1 is performed is or are filled with an inert gas such as dinitrogen, of formula N2, and contain no selenium.
So that the heating step S1 is carried out under an inert atmosphere, the furnace 4 may comprise injection means 48, or an injector, for injecting inert gas into the first chamber 400.
It is possible to achieve a uniform temperature increase in a furnace 4 with multiple chambers which an object to be heated travels between. The temperature of each chamber is controlled, for example via the control means 44, to have an appropriate thermal profile. In practice, step S1 occurs in a chamber 400 or in a series of multiple chambers.
It should be noted, by way of illustrative examples, that the heating step S1 starts at the final “soft” annealing temperature, meaning at a temperature of between 80° C. and 110° C., preferably equal to 90° C., if the “soft” annealing does not include a cooling phase, or at ambient temperature if the “soft” annealing includes a cooling phase to ambient temperature. In the heat treatment example illustrated in
The fact that the temperature increases uniformly is understood here to mean that the temperature increases according to an increasing function that is continuous and differentiable at all points of the time interval considered.
After that, the temperature does not increase according to an increasing function and continues in segments and plateaus for the interval concerned.
According to a particular embodiment of the heating step S2, the temperature increases at a rate between 2.5° C./sec and 4.5° C./sec, and preferably at a rate of 3° C./sec. This rate is either an average rate for the time interval concerned or is an instantaneous rate at a point in this interval, within the limitations of the uniform increase in temperature as defined above.
In the examples illustrated in
In its broadest sense, the chalcogenization step S2 starts at said first temperature T1 and, during this step S2, the temperature continues to increase to a second temperature T2, the stabilization temperature, of between 550° C. and 600° C. Stabilization temperature is understood to mean a temperature which, once reached, remains constant for a defined time.
Thus, in the example illustrated in
The second temperature T2 is more particularly between 570° C. and 590° C., and is preferably equal to 580° C.
According to one embodiment of the chalcogenization step S2, the chalcogen is selenium and the chalcogenization step S2 is a selenization step. The use of another chalcogen such as sulfur is also possible.
According to one feature of this embodiment, the selenization step consists of injecting a gaseous mixture of selenium and dinitrogen, also called selenium vapor, into the second chamber 410 of the furnace 4 for the example shown in
To inject the selenium vapor, the furnace 4 may comprise injection means 48 for injecting into the second chamber 410 a gaseous mixture of selenium and dinitrogen having a temperature of between 480° C. and 520° C.
According to another feature of this embodiment, the injection of the gaseous mixture of selenium and dinitrogen is performed at a volumetric flow rate of 13 standard liters per minute (SLM), plus or minus 3 standard liters per minute.
According to another feature of this embodiment, the mixture of selenium and dinitrogen comes from a source heated to 500° C., plus or minus 20° C.
It should be noted that said injection is the only addition of selenium in the formation process S according to the invention, which, unlike certain formation processes of the prior art, does not include any step of depositing a layer of selenium, for example whether by electrodeposition or by physical vapor deposition.
Selenium is particularly toxic, and particularly in the vapor phase, it is advantageous that at least the second chamber 410 of the furnace 4 is maintained at slightly below atmospheric pressure, more particularly at a pressure of 20 to 200 Pa below atmospheric pressure, because then any release of toxic vapors into the environment outside the preferably sealed second chamber 410 becomes improbable, which ensures the safety of personnel.
In addition, due to the nearly atmospheric pressure prevailing at least in the second chamber 410 of the furnace, the formation process S advantageously allows limiting the duration of the chalcogenization step S2, and more particularly the selenium vapor injection step, to 5 minutes plus or minus 1 minute, as shown in
That the first temperature T1 at which the selenization step begins is fixed in the manner described above is a choice which stems from observations made by the inventors. These observations are essentially related to measurements on photovoltaic cells based on semiconductor layers formed according to formation processes comprising a heating step S1 and a chalcogenization step S2. These measurements are compiled in
The inventors have observed a strong dependency of the average yield, or average conversion efficiency, of the photovoltaic cells produced, on the temperature at which the chalcogenization step S2 begins. The relevant measurements are compiled in
Presumably in a manner that is correlated to chalcogenization reaction kinematics, optimizing the rising slope of the temperature prepares the material for the actual chalcogenization reaction, in particular with atomic mobility at temperatures facilitating the incorporation of selenium into the structure of the metal precursor 2.
As illustrated in
The inventors have also observed that, after selenization, a given value for the ratio of the molar amount of copper to the molar amount of gallium and indium in the metal precursor 2 (this ratio sometimes denoted below as Cu/(In+Ga) for clarity) can have two corresponding values for the ratio of the total molar amount of selenium incorporated into the substrate and precursor to the molar amount of metal precursor 2 (this ratio sometimes denoted below as Se/(Cu+In+Ga) for clarity), depending on the injection temperature of the gaseous mixture of selenium and dinitrogen.
Thus, as illustrated in
In addition, and again as illustrated in
with, between these two phases, and more particularly for an injection temperature of the gaseous mixture of between 475° C. and 540° C., a reversal of the behavior of the amount of copper in the metal precursor 2 necessary to capture an only slightly varying amount of selenium to form the semiconductor layer 1.
Thus, the measurements represented in
It should be noted that the total molar amount of selenium incorporated into the substrate and precursor is greater than the molar amount of selenium incorporated into the precursor alone, provided that the substrate effectively captures a certain molar amount of selenium. Therefore, in this case, the ratio of the molar amount of selenium incorporated into the precursor to the molar amount of metal precursor 2 is within a range of values that are less than the specified range of from 140% to 220%.
In addition, the measurements represented in
After the chalcogenization step S2 is completed, it is important to eliminate the selenium “dust”. For this purpose, the formation process S of the invention comprises, after the chalcogenization step S2, a step of injecting a neutral gas such as dinitrogen into the second chamber 410. This injection may for example last 50 seconds.
As illustrated in
The evolution of the temperature over time during these cooling steps may be controlled by: the control means 44 of the heating device 42, based on measurements made by the measurement means 46 in the second chamber 410 of the furnace 4, for example together with at least one injection of dinitrogen at a predefined temperature and for a given duration; or by the arrangement of a series of chambers at the exit from the furnace 4, including a third chamber 420 illustrated in
The cooling in successive stages occurs for example under an inert atmosphere in successive chambers, to optimize the production rate of the formation process.
The formation process described above allows the formation of a type I-III-VI2 semiconductor layer 1 whose characteristics are discussed below, particularly in comparison to the characteristics of a semiconductor layer obtained by a formation process comprising a chalcogenization step beginning at 580° C., as discussed in the introduction and illustrated in
Firstly, the semiconductor layer 1 obtained by the formation process according to the invention has a microstructure 10 with improved crystallinity compared to the semiconductor layer obtained by the formation process shown in
More specifically, this microstructure 10 is composed of well-defined grains 100, as illustrated in the photograph of
This improvement in the size of the grains of the absorber 100 is achieved by the introduction of selenium vapor when the first temperature T1 is reached, meaning at densification of the metal precursor 2.
Secondly, the grains 100 of the microstructure 10 have different sizes which are proportional to the full width at half maximum of the XRD peak of the CIGSe semiconductor layer 1 for the crystallographic planes identified by Miller indices {112}. As illustrated in
As illustrated in
In addition, the grain size distribution is smaller and better controlled, than what can be obtained by the formation process shown in
This assertion is founded on an analysis of the compiled measurements in
It is immediately apparent, particularly within the two horizontal lines joined by a vertical double arrow on each graph, that even by limiting the range of values for the ratio of Cu/(In+Ga) to values below 0.9 for the purposes of this analysis, the dispersion in the conversion efficiency is markedly reduced by the formation process according to the invention in comparison to the formation process shown in
Thirdly, the semiconductor layer 1 comprises several layers of different compositions. More particularly, it can advantageously consist of a mixture of three phases, while the semiconductor layer formed by the process illustrated in
For example, and as illustrated in
It is therefore observed that the gallium continued to diffuse to the lower layers of the stack constituting the metal precursor 2, during the heating S1 and chalcogenization S2 steps.
In addition, it should be noted that a certain amount of selenium was captured by the molybdenum initially constituting the substrate 3, to form a layer of composition MoSe2 under the lower layer 11, as illustrated in
The formation of the lower layer of composition CuGaSe2 is therefore advantageous in that the adhesion of the semiconductor layer 1 to the layers on which it sits, particularly to the layer of composition MoSe2 shown in
It will be understood from the foregoing that different ranges for the temperature, temperature increase slope, volumetric flow rate and/or duration of injection, as presented above, may be variable depending on the parameters to be considered such as layer thicknesses, and/or substrate dimensions, and/or composition of the inert atmosphere, and/or the amount of gallium.
| Number | Date | Country | Kind |
|---|---|---|---|
| 13 54112 | May 2013 | FR | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/FR2014/051030 | 4/30/2014 | WO | 00 |
