JPH06204160A - Adding method of impurities - Google Patents

Abstract

PURPOSE:To make it unnecessary to consider the range of dopants desides objective impurities, and dope a specimen, with high throughput and low cost, without additionally damaging the specimen, by doping the specimen with separation by packing fraction. CONSTITUTION:In a specimen, a gate electrode 2, an insulating layer 3, and a I-type amorphous silicon layer 4a are laminated on a glass substrate 1, and a protecting insulative film 5 which is to be used as a mask at the time of doping is formed on the layer 4a. PFx0-3<+> 6a and Fx1-3<+>6b as the specimen is doped with no-mass separation, thereby forming an N layer 7a in a source/ drain region. The ion implantation range of ions 6b relative to F is nearly equal to that of ions 6a relative to P. Hence it is not necessary to especially consider the range of ions 6b relative to F, and impurities can be added, so that it is not necessary to change forming conditions and doping conditions of the specimen. Further, the specimen can be doped with high throughput and low cost, without superfluously damaging the specimen.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an impurity addition method for forming an arbitrary impurity layer by adding impurity ions to a sample by non-mass separation.

[0002]

2. Description of the Related Art Conventionally, an ion implantation method has been widely used as a technique for forming an impurity layer in a semiconductor element such as a MOS transistor. Recently, even in a thin film field effect transistor (TFT) used as a driving element of an active matrix liquid crystal display, the ion implantation method has come to be used due to the miniaturization of the TFT accompanying the higher definition of the liquid crystal display (LCD). ing. As an example thereof, the TFT described in JP-A-3-004566, "Thin-film field effect transistor and method for manufacturing the same" is known. This TFT has a channel protection type inverted stagger structure as shown in FIG. 4, has a gate electrode 2 made of a conductive material such as metal or its silicide on a glass substrate 1, and has silicon nitride or silicon oxide above it. And the like, and an i-type amorphous silicon layer (i layer) 4a are laminated, and further formed on the i layer 4a immediately above the gate electrode 2 with silicon nitride or silicon oxide used as a mask for ion implantation. An n-type or p-type amorphous silicon layer (n layer or p) formed by ion implantation of phosphorus (P) or boron (B) on the protective insulating film 5 and both sides thereof.
The source / drain region 7c is formed of an impurity layer (layer).

When forming an impurity layer in such a TFT arrayed in a matrix in a large area by ion implantation, the ion species generated from the impurity source are mass-separated and only arbitrary ions are selected for ion implantation. In the conventional ion implantation method, there is a problem that (1) low throughput and (2) high cost because the ion beam current cannot be taken sufficiently and the implantation process takes time.

On the other hand, at present, a method for forming an arbitrary impurity layer by adding impurities to a target sample without mass-separating the generated ionic species has been realized.
This method decomposes phosphine (PH ₃ ) which is a compound of phosphorus (P) and hydrogen (H) or diborane (B ₂ H ₆ ) which is a compound of B and H into a plasma state as an impurity source. Then, the generated ion species are accelerated by an electric field without mass separation, and impurities are added to the sample as they are. In contrast to the conventional ion implantation method (mass separation type ion implantation method), an impurity addition method ( Non-mass separation type ion implantation method). This method can add impurities to a large area at once and with good uniformity.
Furthermore, the processing time is extremely short, and it is a new technology that is expected as a technology that enables high throughput and low cost of the process.

[0005]

However, in this impurity addition method, since impurities are added by non-mass separation using a hydride-based impurity gas as an impurity source, impurities other than the intended impurities, that is, hydrogen is added. I will end up.

PH ₃ and B ₂ H ₆ were added to a sample as shown in FIG.
When the n-layer or the p-layer 7c is formed by adding impurities by the non-mass separation type ion implantation method using hydrogen as an impurity gas, H-related ions (H _{X = 1-3}
⁺ , H _{X = 1-6} ⁺ ) 6e, 6h are P-related ions (PH _{X = 0-3} ⁺ ) 6d and B-related ions (BH _{X = 0-6} ⁺ , B ₂ ) under the same ion acceleration conditions. H _{X = 0-6} ⁺ ) 6f, 6g, etc. has a range of ion implantation range several times to several tens of times larger than the target impurity under the same ion acceleration conditions. Invades. Therefore, the H-related ions 6e and 6h penetrate the protective insulating film 5 used as a mask at the time of adding impurities, and damage the sample as shown in FIG. 4 and deteriorate the characteristics of the sample. was there. In order to avoid this problem, it is necessary to take measures such as (1) increasing the mask film thickness, (2) decreasing the acceleration energy,
In the case of (1), the throughput is lowered, and in the case of (2), only the impurity distribution limited to the range of hydrogen is obtained,
There were problems such as having to reexamine the optimum preparation conditions for the sample. An object of the present invention is to solve the above-mentioned problems and to provide an impurity addition method with high throughput and low cost.

[0007]

The impurity addition method of the present invention comprises at least an element capable of controlling a valence electron and an element having an ion implantation range equal to or less than that of the element capable of controlling a valence electron. Impurity gas composed of two or more kinds of elements is decomposed into plasma, the generated ion species are used as an impurity source to accelerate the ion species by an electric field, and impurities are added to the sample to form an impurity layer on the sample. It is characterized by forming.

[0008]

The impurity gas composed of at least two kinds of elements including an element capable of controlling valence electrons and an element having an ion implantation range equal to or less than that of the elements capable of controlling valence electrons By decomposing into a plasma and accelerating the ion species using the generated ion species as an impurity source by an electric field to add impurities to the sample, all of the generated ion species have the same range as the target impurity range or Since it is less than that, it is possible to prevent impurities other than the intended impurities from penetrating the mask used when adding the impurities, and prevent damage to the sample. Therefore, it is not necessary to change the mask film thickness, and the impurity distribution can be set on the basis of the target range of impurities. Therefore, it is not necessary to change the sample preparation conditions such as the impurity addition conditions. For this reason, it is possible to save the trouble of changing the process, and it is possible to add impurities with high throughput and at low cost.

[0009]

EXAMPLE FIG. 1 shows an example of the impurity addition method according to the first and second aspects. The sample as shown in FIG. 1 is manufactured through the following steps. First, a gate electrode 2 made of a metal such as chromium (Cr) or aluminum (Al) or its silicide is 50 nm on a glass substrate 1.
Formed with a thickness of. Then, an insulating layer 3 made of silicon nitride, silicon oxide or the like and an i-type amorphous silicon layer (i layer) 4a are laminated on the upper part thereof with a thickness of 300 nm and 100 nm, respectively. Next, on the i layer 4a immediately above the gate electrode 2,
Protective insulating film 5 used as a mask when adding impurities
Is formed with a thickness of 200 nm. This thickness is determined by the ion acceleration energy condition described later. The protective insulating film 5 may be an insulating film such as silicon nitride or silicon oxide, or an organic polymer film such as a resist. However, in this case, the film thickness must be changed depending on the material used, and when a resist is used, It is necessary to make the film thickness more than double that of the silicon nitride or silicon oxide film. However, even in this case, there is almost no difference in the process.

In the sample manufactured up to the above-mentioned state, impurities are added to the source / drain regions 7a by the impurity addition method described in claim 1 thereafter to form an arbitrary impurity layer.
Impurities are added by adding phosphorus trifluoride (P), which is a compound of P and fluorine (F), which is a fluorine-based impurity gas as a pentavalent impurity gas.
F ₃ ), plasma-decomposed by a high frequency electric field of 13.56 MHz, and P-related ions (P
The F _{x = 0-3} ⁺ ) 6a and the F related ion (F _{x = 0-3} ⁺ ) 6b are added to the source / drain regions by impurity addition to the above-mentioned use at an acceleration energy of 40 keV by an electric field by non-mass separation.
Form the layer 7a. The ranges of P and F for the silicon nitride film at this acceleration energy according to the LSS theory are 30.0 nm and 51.7 nm, respectively, and the ranges of P and F for the silicon oxide film are 38.8 nm and 6 respectively.
The ranges of 6.9 nm and P and F with respect to the resist film are 165.4 nm and 260.9 nm, which are close to each other, and can be sufficiently blocked by the mask film thickness under the conventional conditions. Therefore, the F-related ions 6b do not penetrate the mask to damage the sample and deteriorate the characteristics. Therefore, it is not necessary to change the manufacturing conditions for use and the impurity addition conditions, and it becomes possible to add impurities as in the conventional ion implantation method. However, depending on the generated ion species and its ratio, the distribution of P deviates from the LSS theory, but this is essential in the non-mass separation type impurity addition method, and this deviation is corrected in advance. Predetermined impurity addition can be performed by performing impurity addition under such conditions. In this case, there is no particular adverse effect on the sample.

In the embodiment of FIG. 1, PF ₃ which is a fluoride type impurity gas is used as the pentavalent impurity gas, but phosphorus pentafluoride (PF ₅ ) is also used. In addition, as pentavalent impurity gases, arsenic (As) and F, which are fluoride-based impurity gases, are used.
Compounds such as arsenic trifluoride (AsF ₃ ) and arsenic pentafluoride (AsF ₃ )
F ₅ ) can be used. Also, a compound of P and chlorine (Cl), which is a chloride impurity gas, is phosphorus trichloride (PCl ₃ ),
Similar effects can be obtained by the impurity addition method according to the first and _third aspects using phosphorus pentachloride (PCl ₅ ) and arsenic trichloride (AsCl ₃ ). For example, when PCl ₃ is used as a pentavalent impurity gas, P and C with respect to the silicon nitride film are added under the impurity addition conditions of the above-described embodiment.
The ranges of l are 30.0 nm and 26.6 nm, respectively, and the ranges of P and Cl for the silicon oxide film are 3 respectively.
8.8 nm and 34.4 nm, and P for the resist film
And Cl have a range of 165.4 nm and 151.
Since the values are almost equal at 0 nm, not only the effect obtained in the embodiment of FIG. 1 but also the damage caused by Cl to other parts of the sample is less than that when using a fluoride-based impurity gas. An almost equal impurity distribution can be realized. Further, the impurity addition method of claims 1 and 4 using a mixed halide-based impurity gas composed of F and Cl and an element capable of controlling valence electrons is also the same. In this case, as the pentavalent mixed halide-based impurity gas, phosphorus difluoride monochloride (PClF ₂ ) and phosphorus monofluoride dichloride (PCl ₂ F) are used.
Etc. can be used.

In the embodiment shown in FIG. 1, the i-type amorphous silicon layer 4a is used as the i-type semiconductor layer, but the same applies if an i-type polycrystalline silicon layer is separately used. In this case, the embodiment is as shown in FIG. 2 and has a structure in which the i-type amorphous silicon layer 4a in FIG. 1 is replaced by the i-type polycrystalline silicon layer 4b. In the case of the embodiment of FIG. 2, there is a high temperature process in the middle, so the electrode material is molybdenum (M
o), tantalum (Ta), tungsten (W), or refractory metals such as silicide thereof are preferable. Other sample preparation conditions, impurity addition conditions, and the impurity gas used can of course be used as they are as in the embodiment of FIG.

FIG. 3 shows another embodiment of the impurity addition method according to the first and second aspects. The sample as shown in FIG. 3 is manufactured through the following steps. An insulating layer 3 made of silicon oxide, silicon nitride, or the like is formed on the n-type single crystal silicon substrate 8 to have a thickness of 100 nm. Then, the gate electrode 2 made of a metal such as Al or Cr or a silicide thereof is formed on the insulating layer 3 with a thickness of 300 nm. This gate electrode is also used as a mask when impurities are added.

In the sample manufactured up to the above-mentioned state, impurities are added to the source / drain regions 9 through the insulating film by the impurity adding method according to claims 1 and 2, and an arbitrary impurity layer is added to the insulating layer. It is formed in the lower part of 3. For the impurity addition, boron trifluoride (BF ₃ ) which is a compound of B and F which is a fluoride-based impurity gas is used as a trivalent impurity gas.
Plasma-decomposed by a high-frequency electric field of 6 MHz, and B-related ions (BF _{X = 0-3} ⁺ ) 6c and F-related ions (F _{X = 0-3} ⁺ ) 6b generated in the plasma are subjected to non-mass separation to generate an electric field of 60 keV. Impurities are added to the above-described sample with acceleration energy to form a p-type single crystal silicon layer (p layer) 9. LS
The ranges of B and F for the Al electrode at this acceleration energy according to S theory are 195.2 nm and 110. nm, respectively.
4 nm, which is smaller than that for the Cr electrode. As described above, since B is larger than the range of F, F can be sufficiently blocked if the electrode film thickness can block B. When B ₂ H ₆ which is a conventional hydride-based impurity gas is used, the range of H for Al by LSS theory is 1077.9 nm under the above-mentioned conditions. There is a problem in terms of throughput in that the thickness must be increased by about 5 times. But,
The above-mentioned problem is solved by the impurity addition method according to claim 1, and a predetermined impurity addition process can be performed without impairing the throughput and without significantly changing the sample preparation conditions. As a result, a dramatic throughput is achieved. Can be improved.

On the other hand, as a method for manufacturing a semiconductor device in which a BF ₂ ion, which is one of the B-related ions 6c, is ion-implanted to form a p-type impurity layer on the semiconductor device, Japanese Patent Laid-Open No. 63-155720. Publication "Method of manufacturing semiconductor device"
The contents described in are known. The contents of the publication are:
After the amorphous layer is formed on the semiconductor element by ion implantation of ions, the second ions are ion-implanted at a depth shallower than the depth of the amorphous layer formed by the first ion implantation. The most characteristic feature is that the impurities introduced by ion implantation are diffused to make the depth of the impurity layer slightly deeper than the depth of the amorphous layer. The BF ₂ ion used here is for obtaining low-energy B ion, and is a conventional mass separation type ion implantation method for the purpose of realizing a shallow junction. Therefore, the influence of an element ion having an ion implantation range several times larger than the impurity element capable of controlling the valence electrons in the non-mass separation type ion implantation method, which is a feature of the present invention, is controlled by the impurity element capable of controlling the valence electrons. Is fundamentally different from the object of the present invention, which is to be eliminated by substituting elemental ions having the same or smaller ion implantation range as in (1).

In the embodiment of FIG. 3, BF ₃ which is a fluoride type impurity gas is used as the trivalent impurity gas, but boron tetrafluoride (B ₂ F ₄ ) is also used. Besides, boron trichloride (BCl ₃ ) and tetrachloride boron (B ₂ Cl ₄ ) can also be used as the chloride-based impurity gas. Indium chloride (InC) which is a compound of indium (In) and Cl
1), indium dichloride (InCl ₂ ) and indium trichloride (InCl ₃ ) can be used to obtain the same effect. Further, a mixed halide-based impurity gas composed of F and Cl and the target impurity can also be used. In this case, boron trifluoride monochloride (BClF ₂ ) or boron monofluoride dichloride (BCl ₂ F) can be used as the trivalent mixed halide-based impurity gas.

[0017]

According to the present invention, even in the same ion accelerating energy condition in the impurity addition method in which impurities are added by non-mass separation, the impurity distribution of impurities other than the intended impurity can be similarly controlled. There is no need to change the process such as changing the mask film thickness or changing the implantation conditions in order to avoid the problem caused by the large range of H, which is a problem when using the system impurity gas. It is possible to add impurities to a sample while realizing high throughput and low cost without deteriorating the characteristics. Further, in a TFT using hydrogenated amorphous silicon as a semiconductor region, when a fluorine-based impurity gas is used as an impurity source and impurities are added to the semiconductor region, unbonded hands of silicon in the aforementioned semiconductor region are used. When the terminating H is replaced by F, the Si-F bond has a larger binding energy than the Si-H bond and is thermally stable, so that higher reliability of the sample is expected than in the case of H. it can.

[Brief description of drawings]

FIG. 1 is a sectional view showing an embodiment of the present invention.

FIG. 2 is a sectional view showing another embodiment of the present invention.

FIG. 3 is a sectional view showing another embodiment of the present invention.

FIG. 4 is a cross-sectional view showing a conventional impurity addition method.

[Explanation of symbols]

1 Glass Substrate 2 Gate Electrode 3 Insulating Layer 4a i-type Amorphous Silicon Layer (i-layer) 4b i-type Polycrystalline Silicon Layer (i-layer) 5 Protective Insulating Film 6a PF _{X = 0-3} ⁺ 6b F _{X = 1- 3} ⁺ 6c BF _{X = 0-3} ⁺ 6d PH _{X = 0-3} ⁺ 6e H _{X = 1-3} ⁺ 6f BH _{X = 0-6} ⁺ 6g B ₂ H _{X = 0-6} ⁺ 6h H _{X = 1- 6} ⁺ 7a n type amorphous silicon layer (n layer): source / drain region 7b n type polycrystalline silicon layer (n layer): source / drain region 7c n type or p type amorphous silicon layer (n layer or p layer): source / drain region 8 n-type single crystal silicon substrate 9 p type single crystal silicon layer: source / drain region 10 H _{X = 1-3} ⁺ ion damage layer

Claims (4)

[Claims] 1. An impurity gas composed of at least two kinds of elements including an element capable of controlling valence electrons and an element having an ion implantation range equal to or less than that of the elements capable of controlling valence electrons. Is decomposed into a plasma, the generated ion species are used as an impurity source to accelerate the ion species by an electric field, and impurities are added to the sample to form an impurity layer on the sample. 2. The impurity addition method according to claim 1, wherein a compound gas with fluorine is used as the impurity gas. 3. The impurity addition method according to claim 1, wherein a compound gas with chlorine is used as the impurity gas. 4. The impurity addition method according to claim 1, wherein a mixed halogen compound gas of fluorine and chlorine is used as the impurity gas.