Patent Grant
8334532
This application claims priority under 35 USC 119 from Japanese Patent Application No. 2009-147931, filed on Jun. 22, 2009, the disclosure of which is incorporated by reference herein.
1. Field of the Invention
The invention relates to an IGZO-based oxide material and a method of producing an IGZO-based oxide material.
2. Description of the Related Art
In recent years, amorphous In—Ga—Zn—O-based homologous oxide materials (hereinafter, referred to as “IGZO-based oxide materials” or simply as “IGZO” sometimes), represented by a composition formula In2-xGaxO3(ZnO)m (0<x<2 and m is a natural number) in a crystalline state thereof, have been attracting attention.
Hosono et al. from Tokyo Institute of Technology reports that the amorphous IGZO-based oxide materials exhibit a value of electrical resistivity similar to that of a semiconductor and can be formed into a film at room temperature, and that these materials can achieve a mobility that is equal to or higher than that of amorphous silicon (Hosono et al., Non-patent Document 1, Nature, 432 (2004) pp. 488-492).
In particular, amorphous IGZO-based oxide materials represented by the above composition formula in which m=1 are highly promising material systems because of their great amount of ratio of overlapping of electron orbits between In—In which is considered to contribute to the conduction of electrons.
Research and development on the amorphous IGZO-based oxide materials are being intensely conducted with respect to the use of these materials as, for example, an effective material for an active layer of a thin film transistor (hereinafter, referred to as “TFT” sometimes).
On the other hand, IGZO-based oxide materials having a crystalline structure are described, for example, in the following literatures.
Non-patent Document 2 (Journal of the American Ceramic Society, 82 (1999) pp. 2705-2710) describes a method of producing a crystalline IGZO-based oxide material in which m=1, in which a mixture of raw materials including In, Ga and Zn is annealed at 1350° C. or higher, and then rapidly cooled down from this extremely high annealing temperature. The document also discloses that the solid solution range of Ga (range of x) is from 0.66 to 1.06.
Further, Japanese Patent No. 3947575 discloses a process of subjecting a crystalline IGZO-based oxide material in which m=1, which is obtained by annealing at certain conditions, to a reduction heat treatment at a certain temperature in a hydrogen or argon atmosphere.
In this regard, in many of electron-conductive oxide materials properties including IGZO, properties that are unique to these materials are significantly affected by the value of an amount of oxygen vacancy δ. When the value of δ is great, a large number of carriers (electrons) are generated and a “degenerate semiconductor”, having a Fermi level within the conductive band, is formed. In other words, the oxide material in this state is a conductor that exhibits metallic conductivity. On the other hand, when the value of δ is small, generation of carriers can be suppressed and the oxide material can exist as a semiconductor. The above fact indicates that the nature of an oxide material may greatly vary from a conductor to a semiconductor, depending on the value of oxide deficit amount δ.
The present invention has been made in view of the above circumstances and provides an IGZO-based oxide material and a method of producing an IGZO-based oxide material.
A first aspect of the invention provides an IGZO-based oxide material represented by a composition formula of In2-xGaxZnO4-δ, where 0.75<x<1.10 and 0 <δ≦1.29161×exp(−x/0.11802)+0.00153, and formed from a single phase of IGZO having a crystal structure of YbFe2O4.
A second aspect of the present invention provides a method of producing the IGZO-based oxide material according to the first aspect of the present invention, the method comprising annealing, in an oxygen-containing atmosphere, a mixed material including In, Ga and Zn, under conditions of a maximum annealing temperature of from 1200° C. to 1400° C. and an average rate of temperature decrease from the maximum annealing temperature to 300° C. of from 50° C./hr to 500° C./hr.
A third aspect of the present invention provides a method of producing the IGZO-based oxide material according to the first aspect of the present invention, the method comprising:
producing an IGZO-based oxide represented by a composition formula of In2-xGaxZnO4-δ, where 0.75<x<1.10 and δ>0, and formed from a single phase of IGZO having a crystal structure of YbFe2O4; and
controlling the range of δ in the composition formula of the IGZO-based oxide to 0<δ≦1.29161×exp(−x/0.11802)+0.00153 by subjecting the IGZO-based oxide to post-annealing in an oxidizing atmosphere that contains oxygen.
Exemplary embodiments of the present invention will be described in detail based on the following figures, wherein:
In the following, an exemplary embodiment of the IGZO-based oxide material and the method of producing the IGZO-based oxide material is described with reference to the drawings. In the drawings, components having the substantially same function as that of the components that have been previously explained are shown with the same symbol, and explanation thereof may be omitted sometimes.
(Details of IGZO-Based Oxide Material)
First, details of the IGZO-based oxide material are described.
The IGZO-based oxide material according to the first exemplary embodiment of the invention has a crystal IGZO phase represented by a composition formula of In2-xGaxO3(ZnO)m in which m=1. Specifically, in consideration of oxygen vacancy, the material is formed from an IGZO phase represented by a composition formula of In2-xGaxZnO4-δ.
The IGZO phase that forms the IGZO-based oxide material may be single-crystalline or polycrystalline, and has a crystal structure of YbFe2O4 as shown in
Further, the IGZO-based oxide material is formed from a single phase of IGZO, as mentioned above. In the present specification, the term “single phase” refers to the case in which all peaks are derived from the IGZO phase when the peaks are confirmed from the result of measurement carried out by using a powder X-ray diffraction measurement device (RINT-ULTIMA III, trade name, manufactured by Rigaku Corporation), and no peak derived from impurities is observed. Accordingly, even if an IGZO-based oxide material contains a small amount of impurities that cannot be determined by powder X-ray diffraction measurement, this material may be considered to be formed from a single phase of IGZO.
As shown above, when an IGZO-based oxide material is formed from a single phase of IGZO, occurrence of electron scattering at grain boundaries or the like can be suppressed, and the mobility of carriers (electrons) can be maintained, as compared with the case in which multiple crystalline phases exist in combination. Therefore, this material can be effectively applied to an active layer of a TFT or the like.
The value of resistivity of an IGZO-based oxide material is in the range of from 1×102 Ω·cm to 1×109 Ω·cm when measured by a resistivity measurement device (AC Hall measurement device, RESITEST 8300, trade name, manufactured by Toyo Corporation), which falls within a range of the value of resistivity that is typically suitable for an active layer of a TFT, and this range indicates that the material is a semiconductor.
When an IGZO-based oxide material is used for an active layer of a TFT, the value of resistivity thereof is preferably in the range of from 1×104 Ω·cm to 1×105 Ω·cm, since the Vgs-Id curve rises at around 0 V.
In the composition formula In2-xGaxZnO4-δ, the range of x is 0.75<x<1.10, preferably 0.80≦x≦1.05 and more preferably 0.80≦x≦1.00 in view of reliability. This range of x represents the solid-solution range of Ga. When the amount of Ga in an IGZO-based oxide material is outside this solid-solution range, the IGZO-based oxide material is not formed only from an IGZO phase (single phase), but is in a mixed state of an IGZO phase with an impurity phase of In2O3, ZnGa2O4 or the like.
In the composition formula In2-xGaxZnO4-δ, the range of an amount of oxygen vacancy δ is 0<δ≦1.29161×exp(−x/0.11802)+0.00153, preferably 0.0002≦δ≦1.29161×exp(−x/0.11802)+0.00153. The term “exp” refers to an exponential function where the base is e.
However, it is generally difficult to precisely determine the absolute value or the range of an amount of oxygen vacancy of an oxide such as IGZO, since a high degree of measurement accuracy is required and it is not easy to differentiate from moisture, impurities or the like. Therefore, δ may be satisfied with a value at which the value of resistivity of an IGZO-based oxide material is within a range of from 1×102 Ω·cm to 1×109 Ω·cm.
The upper limit of an amount of oxygen vacancy δ (δ max) is the value of an amount of oxygen vacancy at which the value of resistivity of In2-xGaxZnO4-δ is 1×102 Ω·cm or higher, which is calculated from the result of thermogravimetric analysis and the results of measuring the resistivity and the Hall measurement. Details of the calculation method are described later.
The reason why the lower limit of an amount of oxygen vacancy δ is higher than 0 is that when δ=0, the IGZO-based oxide material is presumed to be an insulator since carriers (electrons) are not generated.
In this regard, for example, Japanese Patent No. 3644647 describes that conductivity can be imparted to an IGZO-based oxide material by performing element substitution, even if the amount of oxygen vacancy is zero. It is true that a dope effect can be achieved by element substitution, but in most cases carriers are generated by substituting an element with another element having a different valency, and in principle, carriers are not generated by substituting an element with another element having the same valency. If it is desired to effectively conduct electron doping while the amount of oxygen vacancy being zero, the doping is conducted by substituting an element with another element having a different valency, for example, substituting a trivalent site of In or Ga with a tetravalent element or substituting a divalent site of Zn with a trivalent element.
(Method of Producing IGZO-Based Oxide Material)
In the following, the method of producing the IGZO-based oxide material is described.
Exemplary methods of producing the IGZO-based oxide material according to the first exemplary embodiment of the invention include a method of producing powder, such as a solid-phase reaction method, a sol-gel method, an oxalate method, an alkoxide method or a coprecipitation method; a method of producing a single crystal, such as a flux method, a zone melting method, a CZ method or a glass annealing method via a glass precursor; and a method of producing a thin film, such as a sputtering method, a laser abrasion method, a chemical vapor deposition (CVD) method or a metal organic decomposition (MOD) method. In the following, details of the solid-phase reaction method are described.
1. Preparation of Raw Materials
First, a compound including In, a compound including Ga and a compound including Zn are prepared as the raw materials for the IGZO-based oxide material according to the first exemplary embodiment of the invention.
Examples of the compound including In include In2O3, In(NO3)3, In(NO3)3.nH2O, In(CH3COO)3, In(CH3COO)2(OH), In2O3.nH2O, InN, In(OH)3, InO(OH), In2(C2O4)3, In2(C2O4)3.4H2O, In(C2H5OCS2)3 and In2(SO4)3.nH2O.
Examples of the compound including Ga include, similarly to the compound including In, Ga2O3, Ga(NO3)3, Ga(NO3)3.nH2O, Ga(CH3COO)3, Ga(CH3COO)2(OH), Ga2O3.nH2O, GaN, Ga(OH)3, GaO(OH), Ga2(C2O4)3, Ga2(C2O4)3.4H2O, Ga(C2H5OCS2)3 and Ga2(SO4)3.nH2O.
Examples of the compound including Zn include ZnO, Zn(C2H3O2)2, Zn(C2H3O2)2.2H2O, ZnBr2, ZnCO3, ZnS, ZnCl2, ZnF2, ZnF2.4H2O, Zn(OH)2, ZnI2, Zn(NO3)2.6H2O, ZnC2O4, Zn(CN)2, Zn(OCH3)2, Zn(OC2H5)2 and ZnSO4.
2. Measurement and Mixing of Raw Materials
The raw materials as prepared above are measured using an electronic balance so that the finally obtained IGZO-based oxide material has a desired composition ratio of the raw materials. Then, the raw materials are uniformly mixed using a ceramic pot mill, or a mortar and a pestle, thereby obtaining a mixed material including In, Ga and Zn.
3. Drying and Shaping of Mixed Material
The mixed material including In, Ga and Zn is appropriately dried, and shaped. These processes of drying and shaping can be skipped.
4. Pre-Annealing and Main-Annealing of Mixed Material
Subsequently, the mixed material including In, Ga and Zn is subjected to pre-annealing. The pre-annealing is conducted in order to promote the reaction by maintaining the activity of particles in the unreacted phase by treating the mixed material at a temperature immediately under the temperature at which the desired crystal phase is obtained as a single phase. Another reason of conducing the pre-annealing is to remove carbon that is included in the mixed material in a small amount. The pre-annealing may be conducted, for example, at a maximum annealing temperature (Tmax) of from 400° C. to 1200° C., a time for maintaining Tmax of from 1 hour to 24 hours, an average rate of temperature increase of from 100° C. to 1000° C./hr (hr represents hour), and an average rate of temperature decrease of from 50° C./hr to 1000° C./hr. However, the conditions for the pre-annealing are not particularly limited thereto.
Thereafter, the mixed material that has been subjected to pre-annealing is preferably subjected to main-annealing under the conditions that include all of the following (A) to (D).
(A) The annealing atmosphere for the main-annealing is an atmosphere containing oxygen in order to allow the sintered body to absorb oxygen, during the step of temperature decrease. The oxygen concentration in the atmosphere is preferably 10% or higher in view of starting the intake of oxygen, and the oxygen concentration is more preferably equal to or higher than that of the air (approximately 21% or higher) in view of further promoting the intake of oxygen.
(B) The time for maintaining the maximum annealing temperature Tmax (details thereof are described later) is, for example, from 0.5 hours to 24 hours in view of sufficiently causing the solid-phase reaction of the raw materials, improving productivity, or the like.
(C) The average rate of temperature decrease from Tmax to 300° C. is from 50° C./hr to 500° C./hr. This rate is typically referred to as “slowly-cooling”.
The reason why the average rate of temperature decrease is 50° C./hr or higher is to suppress the reduction of an amount of oxygen vacancy δ down to 0 in the sintered body In2-xGaxZnO4-δ obtained after the main-annealing, as a result of excessive absorption of oxygen during the temperature decrease, thereby suppressing the formation of an insulator from the sintered body.
The reason why the average rate of temperature decrease is 500° C./hr or lower is to suppress the increase of an amount of oxygen vacancy δ in the sintered body In2-xGaxZnO4-δ obtained after the main-annealing, as a result of excessively suppressing the absorption of oxygen during the temperature decrease at a rate of higher than 500° C./hr that is typically referred to as “quenching”, thereby maintaining the sintered body to be a semiconductor.
In order to control the value of 6 to be within a more suitable range, the average rate of temperature decrease is preferably from 100° C./hr to 200° C./hr.
The reason why the lower limit of temperature decrease is 300° C. is to prevent the amount of oxygen vacancy δ from being significantly changed.
In this regard, the amount of oxygen vacancy δ may change depending on the conditions of annealing atmosphere even if the lower limit of temperature decrease is lower than 300° C., such as 100° C. Therefore, the lower limit of temperature decrease is preferably 100° C., more preferably room temperature (25° C.).
(D) The maximum annealing temperature Tmax is not particularly limited as long as the sintered body obtained after the main-annealing forms a single phase of IGZO, and may change depending on the type or the particle size of the raw materials. For example, when In2O3, Ga2O3 and ZnO are used as the starting materials, the maximum annealing temperature Tmax is in the range of from 1200° C. to 1400° C., more preferably from 1350° C. to 1400° C.
The range of the maximum annealing temperature Tmax as mentioned in (D) is based on the following experimental results.
An IGZO-based oxide represented by In2-xGaxZnO4-δ in which x=1, i.e., InGaZnO4-δ, was prepared by mixing oxide powders of In2O3, Ga2O3 and ZnO at a desired molar ratio, shaping the mixed material, and annealing the same. The annealing was conducted in the air under the conditions in which the temperature was increased at a rate of 500° C./hr to a desired temperature (1000° C., 1100° C., 1200° C., 1300° C., 1350° C., 1400° C. or 1450° C.), which was maintained for 2 hours, and then the temperature was slowly decreased to room temperature.
The samples annealed at respective annealing temperatures as described above were subjected to powder X-ray diffraction measurement using a powder X-ray diffraction measurement device (RINT-ULTIMA III, trade name, manufactured by Rigaku Corporation). The results of the measurement are shown in
From the results as shown in
In view of the above results, in order to obtain a single phase of IGZO in a solid-phase reaction method in which the temperature decrease is carried out in a manner of “slowly cooling” as described in (C), the maximum annealing temperature Tmax during the main-annealing is preferably within the range of from 1200° C. to 1400° C., as mentioned above. This is because precipitation of impurities can be suppressed when the annealing is conducted at a maximum annealing temperature Tmax within this range.
Further, it was proved that the c-axis lattice constant calculated from the peak (006) in the X-ray diffraction pattern of the IGZO single phase region at from 1200° C. to 1400° C. was decreased as the annealing temperature was increased up to 1350° C., while the c-axis lattice constant was constant at 1350° C. or higher (see
As a result, it was proved that the Tmax in the main-annealing was more preferably from 1350° C. to 1400° C. when the temperature was decreased in a manner of “slowly cooling” as described in (C) in a solid-phase reaction method.
By subjecting the pre-annealed material to main-annealing under the conditions that satisfy all of the aforementioned (A) to (D), the IGZO-based oxide material according to the first exemplary embodiment of the invention can be produced.
<Variations>
In the above sections, the exemplary embodiments of the invention are explained. However, the invention is not limited to these exemplary embodiments.
For example, the maximum annealing temperature Tmax, which is one of the conditions for conducting the main-annealing, is not particularly limited to the range of from 1200° C. to 1400° C. if the synthesis is conducted by a method other than the solid-phase reaction method as described in the first exemplary embodiment. For example, when a low-temperature synthesis method such as a sol-gel method or a coprecipitation method is employed, the range of maximum annealing temperature Tmax may be set to a lower level than that required in the solid-phase reaction method.
Further, the IGZO-based oxide material according to the first and second exemplary embodiments of the invention is produced by employing a technique of “slowly cooling” in a cooling step during the main-annealing. However, this technique of “slowly cooling” may not be employed in the cooling step during heating if the value of 6 in the composition formula of IGZO-based oxide material (In2-xGaxZnO4-δ), which is obtained by heating the same at certain conditions, is controlled to 0<δ<1.29161×exp(−x/0.11802)+0.00153, by further subjecting the IGZO-based oxide material to post-annealing under an oxygen-containing atmosphere.
This technique of post-annealing is particularly effective when, for example, the IGZO-based oxide material with the amount of oxygen vacancy δ of not less than 1.29161×exp(−x/0.11802)+0.00153 is a conductor, in order to change the IGZO-based oxide material to a semiconductor by controlling the value of δ to the range of 0<δ<1.29161×exp(−x/0.11802)+0.00153. Moreover, this technique is effective also in the case in which the IGZO-based oxide material is a semiconductor with δ in the range of 0<δ<1.29161×exp(−x/0.11802)+0.00153, in order to change the value of resistivity of the IGZO-based oxide material to a desired value by changing the value of δ to a different value within the range of 0<δ<1.29161×exp(−x/0.11802)+0.00153.
In the following, the IGZO-based oxide material and a method of producing an IGZO-based oxide material according to the invention are described with reference to the examples. However, the invention is not limited to these examples.
1. Preparation of Raw Materials
As the raw materials for the IGZO-based oxide material according to Example 1 of the invention, an oxide powder of In2O3 (manufactured by Kojundo Chemical Lab. Co., Ltd., purity: 99.99%), an oxide powder of Ga2O3 (manufactured by Kojundo Chemical Lab. Co., Ltd., purity: 99.99%), and an oxide powder of ZnO (manufactured by Kojundo Chemical Lab. Co., Ltd., purity: 99.99%) were prepared.
2. Measurement and Mixing of Raw Material Powders
These oxide powders were measured using an electronic balance such that the molar ratio of In:Ga:Zn is 2−x:x:1 (0.5≦x≦1.2). For example, when the molar ratio of In:Ga:Zn is 1:1:1 (x=1), the mass of In2O3 is 30.0000 g, the mass of Ga2O3 is 20.2430 g, and the mass of ZnO is 17.5758 g.
These oxide powders were placed in a polyethylene wide-month bottle together with 100 ml of ethyl alcohol and 150 alumina balls having a diameter of 10 mm. This wide-mouth bottle is sealed so that the liquid was not spilled out, and the oxide powders were uniformly mixed by a wet method by rotating the wide-mouth bottle at 120 to 130 rpm for 12 hours using a desk-top pot mill rotating table.
3. Drying and Shaping of Powder Mixture
The alumina balls were taken out from the slurry of powder mixture obtained by mixing the same by a wet-method, and ethyl alcohol was removed therefrom using a rotary evaporator. The obtained powder mixture was dried at 100° C. for 12 hours and roughly crumbled in a mortar, and was then shaped into pellets having a square shape of 6 mm2 and a thickness of 2 mm, by a uniaxial shaping method at a shaping pressure of 100 MPa.
4. Pre-Annealing and Main-Annealing of Powder Mixture
A Pt sheet was placed in an alumina sagger (SSA-S, trade name, manufactured by Nikkato Corporation), and the shaped product obtained by a uniaxial shaping method was placed on the Pt sheet. The alumina sagger on which the shaped product was placed was placed in a high-temperature furnace (FD41) and the shaped product was subjected to pre-annealing under an atmosphere by increasing the temperature at a rate of 500° C./hr, maintaining the maximum annealing temperature at 1100° C. for 2 hours, and then slowly cooling at a rate of 100° C./hr. After the pre-annealing, the pre-annealed body was pulverized in an agate mortar, and the obtained powder was again shaped into pellets by a uniaxial shaping method at a shaping pressure of 100 MPa.
Subsequently, the pre-sintered body shaped in the form of pellets was subjected to main-annealing at a maximum annealing temperature Tmax of about 1350° C. Other conditions for the main-annealing were the same as that of the pre-annealing.
The aforementioned processes (1. preparation of raw materials, 2. measurement and mixing of raw material powders, 3. drying and shaping of powder mixture and 4. pre-annealing and main-annealing) were repeated while changing the input molar ratio x of Ga in the raw materials, as shown in the following Table 2. In Table 2, sample numbers are given to the samples as prepared when the value of input molar ratio x of Ga in the raw material is given as a predetermined value.
5. X-Ray Diffraction Measurement
The pellets of samples 1 to 15 obtained by performing the main-annealing were pulverized in an agate mortar, respectively, and X-ray diffraction measurement of these samples was carried out using an X-ray diffraction measurement device (RINT-ULTIMA III, trade name, manufactured by Rigaku Corporation).
Specifically, the measurement was conducted with the above device using CuKα rays, at a tube voltage of 40 kV, a tube current of 40 mA, a scanning speed of 2.0 deg/min, a light-receiving slit of 0.15 mm, and a scanning range of from 2θ: 5° to 135°. X-ray diffraction patterns of samples 1 to 15 were thus obtained.
From the results shown in
Moreover, in samples 11 to 15 (1.10≦x≦1.50), it was confirmed that a ZnGa2O4 phase (JCPDS#38-0416) was precipitated in addition to an InGaZnO4 phase (JCPDS#38-1104). The above results are shown in Table 3.
From the results described in Table 3, it is proved that the solid-solution range of Ga in In2-xGaxZnO4-δ is 0.80≦x≦1.05. Further, although not shown in Table 3, there is a possibility that a single phase of InGaZnO4 is obtained also in the ranges of 0.75≦x≦0.80 and 1.05≦x≦1.10. Accordingly, the solid-solution range of Ga is 0.75≦x≦1.10, preferably 0.80≦x≦1.05.
The solid-solution range shown above is narrower as compared with that of 0.66≦x≦1.06 as described in Non-patent Document 2. This is because the oxide materials represented by In2-xGaxZnO4-δ obtained in the Examples are in thermal equilibrium at room temperature, as a result of decreasing the temperature in the main-annealing in a manner of “slowly cooling”. Since the temperature decrease in the main-annealing is carried out in a manner of “quenching” in the method of Non-patent Document 2, the solid-solution range described therein is that under the high-temperature conditions. Since the solid-solution range is considered to be generally broader at a higher temperature, the solid-solution range described in Non-patent Document 2 is broader than that shown in the results of the Examples of the invention.
6. Calculation of Lattice Constant
In order to calculate the lattice constant of samples 1 to 15, the values are plotted according to the values of the Nelson-Riley function ½{(cos θ)2/sin θ+(cos θ)2/θ} calculated from the angle 2θ at each diffraction peak in the X-diffraction pattern as the x axis and the values of the lattice constant at each index obtained from Bragg's diffraction condition as the y axis. Subsequently, the value of the y-intercept of the line obtained by a leastsquares method was calculated as the true lattice constant.
More specifically, since InGaZnO4 is a hexagonal crystal, the value of c-axis lattice constant was calculated prior to the calculation of the value of a-axis lattice constant. The true c-axis lattice constant was calculated using a Nelson-Riley function from the values of lattice constant obtained from the diffraction lines at (0012), (0015) and (0018).
The true a-axis lattice constant was calculated using a Nelson-Riley function from the values of lattice constant which were calculated by using the obtained true c-axis lattice constant at 2θ=80° to 135°.
As shown in
Further, it was proved that as the input molar ratio x of Ga was increased, the a-axis lattice constant and the c-axis lattice constant were both decreased. This result is considered to be a behavior within a predictable range according to Vegard's law that is derived from the fact that the ion radius of Ga is smaller than that of In.
7. Evaluation of Electrical Characteristics
An Au electrode was formed at a Van der Pauw position using the pellets having the size of 6 mm3 of samples 1 to 15 after being subjected to main-annealing. Thereafter, the values of resistivity of samples 1 to 15 were measured using an AC Hall measurement device (RESITEST 8300, trade name, manufactured by Toyo Corporation), and the values of carrier concentration and mobility were calculated by carrying out a Hall measurement. These measurements were conducted under a nitrogen gas atmosphere.
From the measurement results shown in
Further, the values of resistivity obtained in the invention are more than three orders of magnitude higher than those described in Non-patent Document 2. This is because the production method described in Non-patent Document 2 includes a process of performing quenching from a high temperature, and the amount of oxygen vacancy δ is increased as a result of performing the quenching. In the method of Non-patent Document 2, the amount of oxygen vacancy δ is purposely increased by performing quenching with the vision of development as conductive materials. However, for example, for producing a semiconductor layer (active layer) of a TFT, the material preferably has a small amount of oxygen vacancy δ and a high degree of resistivity. Therefore, it can be said that changes in the electrical characteristics of an IGZO-based oxide material, which has a small amount of oxygen vacancy and is at thermal equilibrium at room temperature, were shown for the first time by the present examples.
From the calculation results shown in
Moreover, as shown in
8. Evaluation of an Amount of Oxygen Vacancy
The amount of oxygen vacancy δmax of In2-xGaxZnO4-δ at which the value of resistivity was ρ≧102 Ω·cm, which was suitable for an active layer of a TFT, was calculated. Specifically, the value of δmax was calculated based on a combination of the results of the measurement of temperature dependency of the value of resistivity, Hall measurement and thermogravimetric analysis, in a manner as shown below.
(1) Measurement of Temperature Dependency of Value of Resistivity
The temperature dependency of the value of resistivity of sample 5 (x=0.80), sample 7 (x=0.90) and sample 9 (x=1.00) as shown in Table 2 were measured in an argon atmosphere.
(2) Hall Measurement
The Hall measurement of sample 5 (x=0.80), sample 7 (x=0.90) and sample 9 (x=1.00) was conducted in an argon atmosphere while changing the temperature.
Subsequently, the difference Δn in the carrier concentration nTρ at a temperature Tp and the career concentration nRT at room temperature TRT (25° C.), expressed by Δn=nTρ−nRT, was calculated. The value of Δn where x=0.80 was 3.28×1016 (cm−3), the value of Δn where x=0.90 was 1.28×1016 (cm−3), and the value of Δn where x=1.00 was 1.33×1016 (cm−3).
The difference Δn indicates the amount of carriers generated while changing the temperature of In2-xGaxZnO4-δ from room temperature TRT to Tρ, which corresponds to the amount of carriers generated by the relative amount of change of oxygen vacancy Δδ as described later. In the present examples, all of the increase or decrease in carrier concentration are assumed to be due to the oxygen vacancy.
(3) Thermogravimetric Analysis
The thermogravimetric analysis of sample 5 (x=0.80), sample 7 (x=0.90) and sample 9 (x=1.00) was conducted using a measurement device (PYRIS 1 TGA, trade name, manufactured by PerkinElmer Inc.) Specifically, the measurement was conducted under an argon atmosphere at a flow rate of 40 cc/mm, by increasing the temperature up to 400° C. at a rate of 15° C./min.
From the measurement results obtained in the above process, the relative amount of change Δδ of each sample at Tρ was calculated (star symbols shown in the drawing). As a result, the Δδ of sample 5 (x=0.80) was 0.00121, the Δδ of sample 7 (x=0.90) was 0.00132, and the Δδ of sample 9 (x=1.00) was 0.00153.
(4) Calculation of an Amount of Oxygen Vacancy δmax
The value of an amount of oxygen vacancy δmax can be calculated by adding the amount of oxygen vacancy δRT of samples 5, 7 and 9 that already existed at room temperature TRT (before annealing) and the relative amount of change of oxygen vacancy Δδ after annealing the samples to a temperature of Tρ (refer to the following expression (1)).
δmax=δRT+Δδ (1)
Since there is a correlation between the amount of oxygen vacancy and the carrier concentration, the following expressions (2) and (3) are derived when the conversion constant for converting the carrier concentration to the amount of oxygen vacancy is given as α.
nRT×α=δRT (2)
Δn×α=Δδ (3)
In order to calculate the value of δRT, conversion constant α of each sample was obtained using expression (3). Then, the obtained a was substituted into expression (2) together with nRT, thereby calculating δRT.
As a result of the above calculation, the δRT of sample 5 (x=0.80) was 0.00179, the δRT of sample 7 (x=0.90) was 0.00084, and the δRT of sample 9 (x=1.00) was 0.00027. As shown in
Finally, the obtained Δδ and the calculated δRT were substituted into expression (1), thereby calculating δmax.
As a result of the above calculation, the δmax of sample 5 (x=0.80) was 0.00300, the δmax of sample 7 (x=0.90) was 0.00216, and the δmax of sample 9 (x=1.00) was 0.00180.
The relational expression of molar ratio x and δmax was obtained from three data points shown in
As a result, the following relational expression (4) was obtained.
δmax=1.29161×exp(−x/0.11802)+0.00153 (4)
From the above results, it was derived that in the single-phase region of In2-xGaxZnO4-δ, the range of δ≦δmax=1.29161×exp(−x/0.11802)+0.00153 was δ in which the value of resistivity p was 102 Ω·cm or higher, considering the fact that the carrier concentration can be more suppressed when the amount of oxygen vacancy is smaller.
In Example 2, a sample in which the input molar ratio x of Ga was 1.00 and a sample in which the input molar ratio x of Ga was 0.80 were produced in accordance with the method and conditions similar to that of Example 1. However, the average rate of temperature decrease during the main-annealing was 500° C./h in Example 2, instead of 100° C./h in Example 1.
Subsequently, X-ray diffraction measurement, calculation of lattice constant and evaluation of electrical characteristics of these two samples (x=1.00, 0.80) were conducted in accordance with the method and conditions similar to that of Example 1.
As a result of the X-ray diffraction measurement, it was confirmed that a single phase of InGaZnO4 (JCPDS#38-1104) was obtained both in these two samples (x=1.00, 0.80).
As a result of the calculation of lattice constant, the lattice constants of the sample (x=1.00) were a=0.3294 nm and c=2.6036 nm, which were relatively closer to the lattice constants of a=0.3295 nm and c=2.6070 nm indicated by a JCPDS card of InGaZnO4 (No. 38-1104). Further, the lattice constants of the sample (x=0.80) were a=0.3320 nm and c=2.6272 nm. From these results, it was proved that as the input molar ratio x of Ga having a small ionic radius was increased, the lattice constants on the a-axis and the c-axis were also decreased, as was the case with Example 1.
As a result of measuring the value of resistivity of each sample, the value of resistivity p of the sample (x=1.00) was 9.83×102 Ω·cm, and the value of resistivity ρ of the sample (x=0.80) was 1.12×102 Ω·cm. From these results, it was proved that as the input molar ratio x of Ga was increased, the value of resistivity was also increased, as was the case with Example 1.
Then, Hall measurement of each sample was conducted to calculate the mobility and carrier concentration thereof. As a result, the mobility (μ) and the carrier concentration (n) of the sample (x=1.00) were μ=0.83 cm2/V.S and n=3.50×1015 cm−3, respectively; and the mobility (μ) and the carrier concentration (n) of the sample (x=0.80) were μ=0.74 cm2/V.S and n=6.21×1016 cm−3, respectively. In view of the above, it was proved that the carrier concentration was the dominant factor for causing changes in the value of resistivity, as was the case with Example 1.
In Example 3, a sample in which the input molar ratio x of Ga was 1.00 and a sample in which the input molar ratio x of Ga was 0.80 were produced in accordance with the method and conditions similar to that of Example 1. However, the average rate of temperature decrease during the main-annealing was 50° C./h in Example 3, instead of 100° C./h in Example 1.
Subsequently, X-ray diffraction measurement, calculation of lattice constant and evaluation of electrical characteristics of these two samples (x=1.00, 0.80) were conducted in accordance with the method and conditions similar to that of Example 1.
As a result of the X-ray diffraction measurement, it was confirmed that a single phase of InGaZnO4 (JCPDS#38-1104) was obtained in these two samples (x=1.00, 0.80).
As a result of the calculation of lattice constant, the lattice constants of the sample (x=1.00) were a=0.3295 nm and c=2.6038 nm, which were relatively closer to the lattice constants of a=0.3295 nm and c=2.6070 nm indicated by a JCPDS card of InGaZnO4 (No. 38-1104). Further, the lattice constants of the sample (x=0.80) were a=0.3321 nm and c=2.6275 nm. From these results, it was proved that as the input molar ratio x of Ga having a small ionic radius was increased, the lattice constants on the a-axis and the c-axis were also decreased, as was the case with Example 1 and Example 2.
As a result of measuring the value of resistivity of each sample, the value of resistivity p of the sample (x=1.00) was 5.21×103 Ω·cm, and the value of resistivity p of the sample (x=0.80) was 3.54×102 Ω·cm. From these results, it was proved that as the input molar ratio x of Ga was increased, the value of resistivity was also increased, as was the case with Example 1.
Then, Hall measurement of each sample was conducted to calculate the mobility and carrier concentration thereof. As a result, the mobility (μ) and the carrier concentration (n) of the sample (x=1.00) were μ=0.67 cm2/V.S and n=9.81×1014 cm−3, respectively; and the mobility (μ) and the carrier concentration (n) of the sample (x=0.80) were μ=0.78 cm2/V.S and n=5.23×1016 cm−3, respectively. In view of the above, it was proved that the carrier concentration was the dominant factor for causing changes in the value of resistivity, as was the case with Example 1 and Example 2.
In Example 4, a sample in which the input molar ratio x of Ga was 1.00 was prepared by a method and conditions similar to that of Example 1. However, Example 4 employed furnace cooling, which was capable of rapid cooling by turning off the power of an electric furnace to allow natural cooling from the maximum annealing temperature (Tmax=1350° C.), instead of “slowly cooling” employed in Example 1 for temperature decrease during the main-annealing.
Further, the sample was subjected to post-annealing by increasing the temperature up to 500° C. in an atmosphere of oxygen gas 100% at a rate of temperature increase of 500° C./h, maintaining this temperature for 2 hours, and then cooling the same at an average rate of temperature decrease of 100° C./h.
X-ray measurement, calculation of lattice constant and evaluation of electrical characteristics of the sample was conducted by the same method and conditions to that of Example 1, both before and after the post-annealing.
As a result of the X-ray diffraction measurement, it was confirmed that a single phase of InGaZnO4 (JCPDS#38-1104) was obtained in the sample in both cases of before and after the post-annealing.
The calculation result of the lattice constant was that the lattice constants of the sample before the post-annealing were a=0.3291 nm and c=2.6040 nm, and the lattice constants of the sample after the post-annealing were a=0.3291 nm and c=2.6040 nm.
As a result, it was proved that the lattice constants of the sample in both cases of before and after the post-annealing were relatively similar to a=0.3321 nm and c=2.6275 nm shown by a JCPDS card of InGaZnO4 (No. 38-1104).
The result of measuring the value of resistivity of the sample before the post-annealing was ρ=3.21×103 Ω·cm before the post-annealing, whereas the value of resistivity of each sample after the post-annealing was ρ=5.31×103 Ω·cm after the post-annealing. Therefore, it was confirmed that the value of resistivity was increased by performing post-annealing. As a result, it was confirmed that the value of resistivity of an IGZO oxide material could be increased up to a value at which the material could operate as an active layer of a TFT, or the like, by performing post-annealing in an oxidizing atmosphere that contains oxygen.
The invention includes the following exemplary embodiments.
producing an IGZO-based oxide represented by a composition formula of In2-xGaxZnO4-δ, where 0.75<x<1.10 and δ>0, and formed from a single phase of IGZO having a crystal structure of YbFe2O4; and
controlling the range of 6 in the composition formula of the IGZO-based oxide to 0<δ≦1.29161×exp(−x/0.11802)+0.00153 by subjecting the IGZO-based oxide to post-annealing in an oxidizing atmosphere that contains oxygen.
All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2009-147931 | Jun 2009 | JP | national |
| Number | Date | Country |
|---|---|---|
| 3947575 | Jul 2007 | JP |
| 2007-223886 | Sep 2007 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20100320458 A1 | Dec 2010 | US |
