WO1997006564A1 - Semiconductor device and method for manufacturing the same - Google Patents
Semiconductor device and method for manufacturing the same Download PDFInfo
- Publication number
- WO1997006564A1 WO1997006564A1 PCT/JP1996/002184 JP9602184W WO9706564A1 WO 1997006564 A1 WO1997006564 A1 WO 1997006564A1 JP 9602184 W JP9602184 W JP 9602184W WO 9706564 A1 WO9706564 A1 WO 9706564A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- semiconductor device
- concentration
- region
- diffusion layer
- conductivity type
- Prior art date
Links
- 239000004065 semiconductor Substances 0.000 title claims description 317
- 238000004519 manufacturing process Methods 0.000 title claims description 79
- 238000000034 method Methods 0.000 title claims description 36
- 239000000758 substrate Substances 0.000 claims abstract description 162
- 230000006798 recombination Effects 0.000 claims abstract description 41
- 230000007246 mechanism Effects 0.000 claims abstract description 16
- 238000009792 diffusion process Methods 0.000 claims description 253
- 239000012535 impurity Substances 0.000 claims description 136
- 238000005468 ion implantation Methods 0.000 claims description 85
- 239000013078 crystal Substances 0.000 claims description 63
- 238000010438 heat treatment Methods 0.000 claims description 32
- 230000005669 field effect Effects 0.000 claims description 22
- 238000003860 storage Methods 0.000 claims description 15
- 238000005215 recombination Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000002347 injection Methods 0.000 claims description 8
- 239000007924 injection Substances 0.000 claims description 8
- 229910052751 metal Inorganic materials 0.000 claims description 8
- 239000002184 metal Substances 0.000 claims description 8
- 230000000295 complement effect Effects 0.000 claims description 7
- 230000000903 blocking effect Effects 0.000 claims description 5
- 239000003990 capacitor Substances 0.000 claims 2
- 210000004709 eyebrow Anatomy 0.000 claims 1
- 239000000969 carrier Substances 0.000 abstract description 15
- 230000003071 parasitic effect Effects 0.000 abstract description 15
- 230000000694 effects Effects 0.000 description 79
- 229910021421 monocrystalline silicon Inorganic materials 0.000 description 59
- 239000002585 base Substances 0.000 description 27
- 230000007547 defect Effects 0.000 description 26
- 238000010586 diagram Methods 0.000 description 25
- 150000002500 ions Chemical class 0.000 description 24
- 230000002829 reductive effect Effects 0.000 description 24
- 230000015572 biosynthetic process Effects 0.000 description 23
- 229910021420 polycrystalline silicon Inorganic materials 0.000 description 23
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 17
- 229910052814 silicon oxide Inorganic materials 0.000 description 17
- 238000009826 distribution Methods 0.000 description 13
- 208000024891 symptom Diseases 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 10
- 230000008030 elimination Effects 0.000 description 10
- 238000003379 elimination reaction Methods 0.000 description 10
- 238000002955 isolation Methods 0.000 description 10
- 238000000059 patterning Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 10
- 230000001681 protective effect Effects 0.000 description 10
- 238000000151 deposition Methods 0.000 description 9
- 238000006243 chemical reaction Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 8
- 230000005684 electric field Effects 0.000 description 8
- 229910052698 phosphorus Inorganic materials 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000007935 neutral effect Effects 0.000 description 6
- 229910052710 silicon Inorganic materials 0.000 description 6
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 5
- 230000015556 catabolic process Effects 0.000 description 5
- 238000001312 dry etching Methods 0.000 description 5
- 239000000463 material Substances 0.000 description 5
- 239000011574 phosphorus Substances 0.000 description 5
- 230000009467 reduction Effects 0.000 description 5
- 230000002441 reversible effect Effects 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 238000012546 transfer Methods 0.000 description 5
- 229910000577 Silicon-germanium Inorganic materials 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 4
- 229910052786 argon Inorganic materials 0.000 description 4
- 229910052799 carbon Inorganic materials 0.000 description 4
- 239000000919 ceramic Substances 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 229910052732 germanium Inorganic materials 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 229910052754 neon Inorganic materials 0.000 description 4
- 231100000989 no adverse effect Toxicity 0.000 description 4
- 230000002093 peripheral effect Effects 0.000 description 4
- 238000001953 recrystallisation Methods 0.000 description 4
- 229910052581 Si3N4 Inorganic materials 0.000 description 3
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 3
- 229910052796 boron Inorganic materials 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 239000002131 composite material Substances 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 3
- 238000002425 crystallisation Methods 0.000 description 3
- 238000005530 etching Methods 0.000 description 3
- 239000007789 gas Substances 0.000 description 3
- 229910052736 halogen Inorganic materials 0.000 description 3
- 150000002367 halogens Chemical class 0.000 description 3
- 238000002513 implantation Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 239000013307 optical fiber Substances 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 230000000717 retained effect Effects 0.000 description 3
- 230000008054 signal transmission Effects 0.000 description 3
- 239000010703 silicon Substances 0.000 description 3
- HQVNEWCFYHHQES-UHFFFAOYSA-N silicon nitride Chemical compound N12[Si]34N5[Si]62N3[Si]51N64 HQVNEWCFYHHQES-UHFFFAOYSA-N 0.000 description 3
- 238000004544 sputter deposition Methods 0.000 description 3
- 238000007740 vapor deposition Methods 0.000 description 3
- ZOXJGFHDIHLPTG-UHFFFAOYSA-N Boron Chemical compound [B] ZOXJGFHDIHLPTG-UHFFFAOYSA-N 0.000 description 2
- 230000003321 amplification Effects 0.000 description 2
- 229910052785 arsenic Inorganic materials 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 229910052800 carbon group element Inorganic materials 0.000 description 2
- 230000008021 deposition Effects 0.000 description 2
- 230000006866 deterioration Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 239000006104 solid solution Substances 0.000 description 2
- AFCARXCZXQIEQB-UHFFFAOYSA-N N-[3-oxo-3-(2,4,6,7-tetrahydrotriazolo[4,5-c]pyridin-5-yl)propyl]-2-[[3-(trifluoromethoxy)phenyl]methylamino]pyrimidine-5-carboxamide Chemical compound O=C(CCNC(=O)C=1C=NC(=NC=1)NCC1=CC(=CC=C1)OC(F)(F)F)N1CC2=C(CC1)NN=N2 AFCARXCZXQIEQB-UHFFFAOYSA-N 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 229910052783 alkali metal Inorganic materials 0.000 description 1
- 150000001340 alkali metals Chemical class 0.000 description 1
- 229910052784 alkaline earth metal Inorganic materials 0.000 description 1
- 150000001342 alkaline earth metals Chemical class 0.000 description 1
- 230000005260 alpha ray Effects 0.000 description 1
- 229910052787 antimony Inorganic materials 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- RQNWIZPPADIBDY-UHFFFAOYSA-N arsenic atom Chemical compound [As] RQNWIZPPADIBDY-UHFFFAOYSA-N 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000005229 chemical vapour deposition Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 229910021419 crystalline silicon Inorganic materials 0.000 description 1
- 230000008025 crystallization Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002542 deteriorative effect Effects 0.000 description 1
- MROCJMGDEKINLD-UHFFFAOYSA-N dichlorosilane Chemical compound Cl[SiH2]Cl MROCJMGDEKINLD-UHFFFAOYSA-N 0.000 description 1
- 230000005496 eutectics Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 229910052733 gallium Inorganic materials 0.000 description 1
- 229910000078 germane Inorganic materials 0.000 description 1
- GNPVGFCGXDBREM-UHFFFAOYSA-N germanium atom Chemical compound [Ge] GNPVGFCGXDBREM-UHFFFAOYSA-N 0.000 description 1
- 239000007943 implant Substances 0.000 description 1
- 239000012212 insulator Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 229910052700 potassium Inorganic materials 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 229910052708 sodium Inorganic materials 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 description 1
- 229910052721 tungsten Inorganic materials 0.000 description 1
- 239000010937 tungsten Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
- H10D30/00—Field-effect transistors [FET]
- H10D30/60—Insulated-gate field-effect transistors [IGFET]
- H10D30/67—Thin-film transistors [TFT]
- H10D30/6704—Thin-film transistors [TFT] having supplementary regions or layers in the thin films or in the insulated bulk substrates for controlling properties of the device
- H10D30/6708—Thin-film transistors [TFT] having supplementary regions or layers in the thin films or in the insulated bulk substrates for controlling properties of the device for preventing the kink effect or the snapback effect, e.g. discharging the minority carriers of the channel region for preventing bipolar effect
Definitions
- the present invention relates to a semiconductor device, and more particularly to a high-performance MOS field-effect transistor formed on an insulating film and capable of operating at high speed with low parasitic capacitance, and a method of manufacturing the same.
- MOS MOS field effect transistor
- Si single crystal silicon
- FIG. 1 4 is an isolation insulating film
- 5 is a gate insulating film
- 61 is a gate electrode
- 7 is a gate protective insulating film
- 8 is a gate sidewall insulating film
- 9 and 10 are n-type high concentration diffusion layers.
- the SO I'MOS as shown in Fig. 1 has a feature that the drain junction capacitance and wiring parasitic capacitance can be reduced to about 1Z10 compared to the conventional MOS because it has a thick insulating film 2 directly underneath. . Furthermore, since the MOS is insulated and separated from the substrate, it has features such as malfunction due to ⁇ -ray irradiation and radical phenomena can be fundamentally eliminated.
- the disadvantage of the conventional SOI'MOS is that the single-crystal Si film 3 is insulated from the supporting substrate 1, and thus is transiently accumulated in the minority carrier force single-crystal Si film 3 generated by a strong electric field at the drain.
- the threshold voltage fluctuates, which is the so-called substrate floating effect.
- This effect is also a parasitic bipolar effect in which minority carriers are accumulated in the single-crystal Si film 3 and the potential rises due to an increase in the source force.
- nMOS n-conductivity SOI • MOS
- holes are accumulated, the threshold voltage fluctuates in the negative direction, a specific bump is observed in the current-voltage characteristics, and the leakage current increases in the off state. Further, the breakdown voltage between the source and the drain is reduced.
- the substrate floating effect may be a fatal drawback for differential amplifiers and analog circuits that require detection of small current differences. There is.
- S 0 I ⁇ MOS in Fig. 1 is a structure proposed to eliminate the substrate floating effect, and the germanium (Ge) ion implantation in the source diffusion layer 9 is performed with the Ge component ratio. About 10% of SiGe mixed crystal 16 is formed.
- FIG. 2 is an energy band diagram along the channel in a state where the drain voltage Vds is applied in the SOI ′ MOS of FIG.
- E f n is the pseudo-Furmi level
- E i is the intrinsic Fermi level.
- the introduction of Si Ge mixed crystal 16 narrows the band gap by about 0.1 leV, and the valence band Ev at the source is configured as shown by the broken line.
- the diffusion potential difference for holes near the source is reduced.
- the conduction band Ec is not affected by the Si Ge mixed crystal, and it is said that the behavior of electrons as majority carriers has no adverse effect.
- An object of the present invention is to consider the fact that the structure shown in FIG. 1 is effective only for nMOS in eliminating the substrate floating effect of S 0 I ⁇ MOS, and therefore, p-type MOS (hereinafter abbreviated as pMOS), and An object of the present invention is to provide a substrate floating effect eliminating structure applicable to complementary MOS (hereinafter abbreviated as CMOS).
- CMOS complementary MOS
- Another object of the present invention is to solve the problem of the Ge ion implantation step which is essential for the formation of Si Ge mixed crystals.
- GeH 4 germane
- GeH 4 is usually used as an ion source for Ge ion implantation, but GeH 4 is a very easily decomposable substance, and it is difficult to supply implanted ions stably.
- An object of the present invention is to provide a substrate floating effect eliminating structure which can be manufactured only by an existing sharable semiconductor manufacturing apparatus, and is therefore manufactured at a low cost.
- the substrate floating effect is based on the parasitic bipolar effect using the drain and the substrate as the emitter and the base, the substrate has the current amplification factor larger than the bipolar current amplification factor, and the substrate is connected to the emitter.
- the formation of a second parasitic bipolar in the source diffusion layer eliminates the accumulation of minority carriers on the substrate.
- FIG. 3 is an impurity distribution diagram in the depth direction of the source diffusion layer in one embodiment of the present semiconductor device. It is assumed that the bottom of the source diffusion layer is separated from the outside by a buried oxide film.
- a high concentration p-type diffusion layer is formed in an n-type source.
- a relatively low-concentration first n-type impurity layer is formed by ion implantation of phosphorus (P) and subsequent heat treatment to reach the bottom of the source diffusion layer. From this state, ions of high concentration boron (B) and further high concentration of arsenic (As) are implanted to form an npn impurity distribution from the surface. The same impurity distribution is formed in the drain region as in the source.
- P phosphorus
- As arsenic
- FIG. 4A is an energy band diagram for the nMOS having the above impurity distribution, in which the drain voltage Vds is applied.
- the solid lines indicate the valence band Ev and the conduction band Ec in the channel formation region on the SOI surface, and the broken lines indicate the energy bands near the p-type diffusion layer separated from the outside.
- Holes generated near the drain due to the strong electric field at the drain and accumulated in the p-type single-crystal Si film under the channel pass through the low-concentration P substrate under the channel through the p-type diffusion formed in the emitter P-E and source.
- the conduction is controlled by a pnp transistor whose region is the collector P + c and whose n-type diffusion region is the base nB between the two regions.
- FIG. 4B is an equivalent circuit diagram showing the vicinity of the source for explaining the operation principle of the pnp transistor.
- the collector current density J is expressed by the following equation.
- Equation 1 N B - W B is 1 0 1 2 value of about the normal transistor is referred base region number of carriers and Gummel number.
- q is the electron charge amount
- D n are diffusion coefficients
- n is the intrinsic density
- N B is Akuseputa impurity density in the base
- W B Habe Ichisu width
- V BE Habe Ichisu-Emitta voltage
- k T Is thermal energy.
- Low concentration p is the potential of the substrate accumulation hole density is Emitta buried oxide film capacitance C B. It is determined by x, and V ds.
- V BE should be large enough for the diffusion potential difference between n + s -p + c when n + s ⁇ P + c can be represented by a diode, that is, when a normal pn junction is formed. Can not.
- n + s ⁇ c When n + s ⁇ c is short-circuited, the collector is at ground potential and V BE can take a sufficiently large value, and holes flow out to the source electrode.
- the short circuit between n + s and pc can be realized by the formation of crystal defects on the pn junction surface by high-concentration As ion implantation or by the direct formation of n-type doped polycrystalline Si on the p-type high concentration layer.
- the p-type high-concentration layer in the source is a low-concentration n-type diffusion layer, which covers the side and bottom surfaces facing the drain, and the surface is covered with the high-concentration n-type diffusion layer.
- the number of gummels on the side surface is set small, and the number of gummels on the front surface is increased. This has the effect of suppressing electron emission from the source to the substrate due to the rise in the substrate potential.
- the width of the low-concentration n-type diffusion layer on the side of the p-type high-concentration layer in the source be narrow for the same reason.
- an n-type diffusion layer is also formed at the bottom of the p-type diffusion layer to isolate the p-type diffusion layer from the substrate, so that the presence of the P-type diffusion layer is not affected by the drain-substrate capacitance. That is, the effect of reducing the parasitic capacitance by the thick buried oxide film, which is the largest feature of the conventional SOI * MOS, is maintained.
- the substrate floating effect can be similarly eliminated by replacing the impurity conductivity type described above. That is, the present invention is effective in eliminating the substrate floating effect of SOI ⁇ CMOS.
- FIG. 5A is an enlarged cross-sectional view of the vicinity of the source diffusion layer in another embodiment of the present semiconductor device
- FIG. 5B is an equivalent circuit diagram of the present semiconductor device
- FIG. 6B is a diagram showing the energy bands of the high-concentration P region and the high-concentration n region in the source region in an overlapping manner. It is assumed that the bottom of the source region is separated from the outside by a buried oxide film.
- the second parasitic pnp bipolar formed in the source region is based on the emitter p- in the p-type substrate region, the base is based on the n-type low-concentration source diffusion layer, and the collector p + is in the p-type high-concentration region in the source. Act as The base is connected to the high-concentration source region at the ground potential through a resistance component of the low-concentration source diffusion layer. In the second parasitic bipolar of the present semiconductor device, the base potential is fixed.
- the hole injection efficiency can be improved by reducing the diffusion potential difference between holes by bringing the emitter and collector closer together under the condition of sufficiently low impurity base concentration.
- a mechanism is needed to quickly eliminate holes injected into the collector.
- the present semiconductor device utilizes a recombination center based on a crystal defect or the like. It is well known that the hole annihilation mechanism due to the recombination center also depends on the impurity concentration, and it is desirable that the P-type impurity concentration be as high as possible.
- the holes generated by the strong electric field at the drain and accumulated in the P-type substrate region are drawn out to the source region by the bipolar formed in the source region, and are converted into the source current. Is eliminated.
- the above source current density J is expressed by the following equation.
- the method of forming the present semiconductor device is to use a gate electrode as a mask and implant a relatively low-concentration first n-type impurity layer to reach a thick silicon oxide film immediately below the SOI layer by ion implantation of phosphorus (P).
- a low-concentration n-type diffusion layer is first formed by a subsequent heat treatment. From this state, a gate sidewall insulating film is formed, and high-concentration boron (B) ions are implanted using the gate electrode and the gate sidewall insulating film as a mask to make the bottom portion of the SOI layer amorphous.
- the bottom of the amorphous layer is an oxide film even in a short-time high-temperature heat treatment, single-crystallization by recrystallization heat treatment is not performed except for the side surface of the single-crystal SOI layer immediately below the gate. Only crystallization proceeds.
- the above polycrystalline or amorphous properties can be controlled by heat treatment conditions, and control the recombination center characteristics.
- the width of the source low-concentration n-type diffusion layer left on the side surface and serving as a base region is controlled by the thickness of the gate sidewall insulating film.
- the n-type high-concentration region formed on the surface of the SOI layer is formed by ion implantation or by selectively removing a certain thickness of the surface of the p-type high-concentration region, and then forming the semiconductor film by deposition to leave. May be. In the latter case, a steeper high impurity distribution can be realized, and an impurity distribution having more excellent recombination characteristics and thus more effective in eliminating the substrate floating effect can be realized. Since the method of the present invention is performed using the gate electrode and the gate sidewall insulating film as a mask, a similar structure is formed in the drain region.
- the drain voltage is applied to the n-type high-concentration region, but since the relationship between the n-type high-concentration region and the bottom P-type high-concentration region in the drain has a reverse characteristic with respect to hole injection, the junction leakage The increase in the blocking current is negligible and does not hinder the operation as a NOS transistor.
- FIGS. 5A and 5B ie, a p-type low-concentration SOI substrate emitter, an n-type low-concentration diffusion layer as a base, a p-type high-concentration region as a collector, and an n-type high-concentration
- the n-type high-concentration region was set to the ground potential, and the hole current flowing through the emitter when a positive voltage was applied to the p-type low-concentration SOI substrate emitter was determined by numerical analysis using the base width parameter.
- Fig. 7 shows the results.
- the p-type S 0 I layer having a uniform concentration distribution of 50 mn and 4 ⁇ 10 17 / cm 3 was used as the emitter, and the thickness was 10 nm, which was configured to be in contact with the buried insulating film at the bottom of the SO I layer.
- the high-concentration region, n-type base has a Gaussian distribution with a maximum concentration of 5 XI 0 'Vcm 3 , and is configured so as to separate the collector and the n-type high-concentration region from the emitter.
- the recombination time 1/10 '° sec collector region, in other regions assumes 1Z10 4 seconds.
- Recombination time 1 1 0 1 The value of 1 second is a value normally observed in a high impurity concentration polycrystalline Si film.
- the electron current and the hole current having no p-type collector region and the normal source structure are also shown by thin lines.
- the electron current flowing into the emitter that of the present semiconductor device also matches that of the conventional structure, there is no difference, and it shows an increasing trend in proportion to the applied voltage with the exponential function.
- the hole current the emitter current flows out from the emitter up to the threshold voltage (negative hole current). From the threshold voltage onwards, the hole current becomes a positive hole current, which increases in proportion to the applied voltage in an exponential function. Even in the normal structure, the hole current shows a value about two orders of magnitude larger than the electron current.
- the base width is 4
- the hole current near the threshold voltage is about three orders of magnitude higher than that of the conventional structure, and the threshold voltage is also 0.
- a 2 V drop is seen.
- the above-mentioned meaning indicates that the source diffusion potential with respect to the hole current is reduced by 0.2 eV in the present semiconductor device structure as compared with the conventional structure.
- the above value is twice the reduced value of the source diffusion potential of 0.1 eV due to the known SiGe mixed crystal formation in the source, indicating that the elimination of the substrate floating effect is further improved.
- the decrease in the threshold voltage tends to be eliminated with an increase in the base width, a decrease of 0.04 V is observed even with a base width of 0.1 ⁇ m as compared with the conventional structure. That is, in the present semiconductor device structure, the base width is desirably 0.1 m or less.
- the semiconductor device has a symmetric structure with respect to the source and the drain, and is also effective for a so-called transfer MOS or the like in which the drain and the source are switched according to the circuit operating conditions to operate in both directions. Furthermore, an n-type diffusion layer is also formed at the bottom of the p-type high-concentration region in the source and drain to isolate the p-type diffusion layer from the substrate, or a configuration that limits the P-type high-concentration region to only the desired locations. By doing so, the presence of a p-type high concentration region in the drain-substrate capacitance can be reduced to a negligible level.
- n MOS holes generated near the drain and accumulated in the substrate are quickly injected into the source diffusion layer and annihilated.
- a region having a small diffusion potential difference formed between the P-type substrate region immediately below the channel and the source diffusion region, that is, a sufficiently low concentration n-type diffusion layer region is provided adjacent to the source high concentration diffusion layer.
- a region acting as a recombination center for holes is provided in the n-type low-concentration diffusion layer region to eliminate holes injected into the n-type low-concentration diffusion layer region.
- the electrons required for hole annihilation are supplied from the n-type high concentration diffusion layer region on the source surface.
- the recombination based on crystal defects and the like is used as a hole annihilation mechanism by the recombination center. Take advantage of the center.
- FIG. 8 is an enlarged cross-sectional view near the source diffusion layer in the present semiconductor device
- FIG. 9 is an energy band diagram near the source when the drain voltage is the ground voltage. It is assumed that the bottom of the source region is separated from the outside by a buried oxide film.
- the energy band in the case of the conventional source structure using the high-concentration source diffusion layer is also shown by broken lines, but in the structure of the present invention, the decrease in the diffusion potential difference with respect to holes is apparent.
- the impurity concentration of the low concentration diffusion layer is preferably as low as possible without changing the conductivity type, and is preferably 10 15 / cm 3 or more and 10 ′ 8 Zcm 3 or less.
- the formation of crystal defects acting as recombination centers is performed by ion implantation with an element that does not increase the diffusion potential difference with respect to holes in the n-type low-concentration diffusion layer, thereby making the bottom surface of the SOI layer amorphous.
- single-crystallization by recrystallization heat treatment is not performed except for the side surface of the single-crystal SOI layer immediately below the gate because the bottom is an oxide film. Only progresses.
- the above polycrystalline or amorphous properties can be controlled by heat treatment conditions, and the recombination center characteristics can be controlled.
- CMOS semiconductor devices which are the mainstream of MOS type semiconductor devices
- individually applying nMOS and pMOS will increase the number of manufacturing steps and decrease the yield of non-defective products. This leads to higher costs. Therefore, as the ion implantation source that forms the recombination center, an element that does not increase the diffusion potential difference with respect to a small number of carriers in the n-type and p-type low-concentration diffusion layers is used in the same process for each of the pMOS and nMOS source low-concentration diffusion regions. It is desirable to perform ion implantation to form recombination.
- an element other than an element that easily activates and forms n or p conductivity type such as P, B, As, Sb, and Ga, may be used as an ion implantation source.
- the element be an element that makes the semiconductor substrate amorphous by ion implantation, and an element having an atomic mass of 10 or less is not preferable.
- Alkali metals such as Na and K and alkaline earth metals including Mg, which have an unusually large diffusion coefficient in Si semiconductors and impair reliability, are also not preferred.
- a group 14 element such as Si, Ge and C, a halogen element such as F and C1, and a rare gas element such as Ne and Ar which constitute the semiconductor are preferable.
- elements such as Si, C, Ne, Ar, and C1, which are inexpensive, stable in supply, easy to ionize, and stable, are most preferable.
- a low-concentration and high-concentration source / drain diffusion layer is formed using a gate electrode as a mask.
- the low-concentration diffusion layer is formed by ion implantation and heat treatment after reaching the silicon oxide film having a thickness L just below the SOI layer.
- a gate sidewall insulating film is formed, and ion implantation of, for example, Si using the gate electrode and the gate sidewall insulating film as a mask is performed so as to reach a thick silicon oxide film immediately below the SOI layer, thereby forming the SOI layer in the oxide film interface region.
- the interface of the amorphous region is not single-crystallized by the subsequent heat treatment, but is polycrystallized by fine grain boundaries, and functions as a recombination center.
- the center between the recombination center region and the channel region The distance between the SOI substrates, that is, the width of the source low concentration n-type diffusion layer is controlled by the thickness of the gate sidewall insulating film.
- the source low-concentration n-type diffusion layer width is preferably lower than 100 ⁇ 0 ° so that minority carriers can easily reach the recombination center region and disappear.
- the n-type high-concentration source region formed on the surface of the SOI layer may be formed by forming the above-described recombination center region, selectively removing a certain thickness of the surface portion, and leaving the semiconductor film by the deposition method. ,.
- Another method of forming the semiconductor device is to form a source / drain region having a desired diffusion layer structure based on a conventional manufacturing method, and then form a contact hole for connection with a source electrode. Selectively perform ion implantation to form a low concentration diffusion layer that reaches the thick silicon oxide film just below the SOI layer.
- a side wall film is provided so as to reduce the contact hole dimension by a certain width, and ion implantation for forming a recombination center region using the side wall film as an implantation mask is performed so as to reach a thick silicon oxide film immediately below the SOI layer. May be.
- the method of the present invention is performed using the gate electrode and the gate side wall insulating film or the contact hole as an ion implantation mask, a similar structure is formed in the drain region.
- the drain voltage is applied to the n-type high-concentration region.However, the relationship between the n-type high-concentration region and the bottom recombination center region in the drain has a reverse characteristic with respect to hole injection. The increase in junction leakage current is negligible and does not hinder the operation as a MOS transistor.
- the n-type high-concentration region is set to the ground potential, and the hole current flowing when a positive voltage is applied to the p-type low-concentration SOI substrate is reduced by the distance between the recombination center region and the channel region S 0 I substrate.
- Figure 6 shows the results obtained by numerical analysis as parameters.
- the diffusion layer width was used as a parameter.
- the source low-concentration n-type diffusion layer has a Gaussian distribution with a maximum concentration of 1 ⁇ 10 16 / cm 3 and is configured to be in contact with the buried insulating film at the bottom of the SOI layer.
- S assumes a 1 Z 1 0 4 seconds in other areas.
- the recombination time of 1Z10 '° sec is the value normally observed for polycrystalline Si films.
- the electron current and the hole current of the normal source structure having no recombination center region are also shown by thin lines for reference.
- That of the semiconductor device of the present invention also matches that of the conventional structure, and no difference is observed.
- the increase in the applied current is proportional to the exponential function.
- For the hole current there is a threshold voltage at which a current proportional to the applied voltage starts to flow exponentially. Below the threshold voltage, the behavior is as if the current becomes zero at zero applied voltage. Even in the normal structure, the hole current shows a value about two orders of magnitude larger than the electron current.
- the hole current near the threshold voltage value is about three orders of magnitude higher than that of the conventional structure and the threshold voltage is 0.2 V when the source low-concentration n-type diffusion layer width is 4 Onm ⁇ or less. Is seen to decrease.
- the above means that the source diffusion potential with respect to the hole current is reduced by 0.2 eV in the present semiconductor device structure as compared with the conventional structure.
- the above value is twice the reduction value of the source diffusion potential of 0.1 eV due to the formation of the SiGe mixed crystal in the known source, indicating that the elimination of the substrate floating effect is further improved compared to the known method. I have.
- the width of the source low-concentration n-type diffusion layer is desirably 0.1 ⁇ m or less.
- the results in FIG. 10 relate to the forward characteristics when a positive voltage is applied to the channel region S 0 I substrate.
- Numerical analysis was also performed on the reverse characteristics in which a positive voltage was applied to the n-type high concentration region.However, in the analysis results up to 3 V, the current was a value between 1/10 13 and 1Z10 15 A within the calculation error range. However, there was no difference from the normal structure. This result indicates that even if the same structure as in the source is formed in the drain, no problem such as an increase in leakage current occurs.
- the semiconductor device according to the present invention has a symmetric structure with respect to the source and the drain, and is also effective for a so-called transfer MOS or the like in which the drain and the source are switched according to the circuit operating conditions to operate in both directions.
- the presence of the recombination center region in the source and drain does not affect the drain-substrate capacitance at all, and the parasitic capacitance reduction effect of the thick buried oxide film, which is the largest feature of conventional SOI and MOS, is maintained. Is done.
- the present invention is effective irrespective of the conductivity type of the semiconductor device, and is therefore effective for eliminating the substrate floating effect of SOI ⁇ CMOS.
- FIG. 1 is a cross-sectional view of a conventional semiconductor device.
- FIG. 2 is an explanatory view of an energy band of a substrate floating effect eliminating mechanism in a conventional semiconductor device.
- FIG. 3 is an impurity distribution diagram in the depth direction of the source diffusion layer in the semiconductor device of the present invention.
- 4A and 4B are an energy node diagram and a source equivalent circuit diagram illustrating a mechanism for eliminating a substrate floating effect in a semiconductor device of the present invention.
- 5A and 5B are a cross-sectional view near a source diffusion layer and an equivalent circuit diagram in the semiconductor device of the present invention.
- 6A and 6B are energy band diagrams illustrating a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention.
- FIG. 7 is an analysis result regarding a substrate floating effect eliminating mechanism by the semiconductor device of the present invention.
- FIG. 8 is a cross-sectional view of a source diffusion layer illustrating a mechanism for eliminating a substrate floating effect in a semiconductor device of the present invention.
- FIG. 9 is an energy band diagram for explaining a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention.
- FIG. 10 is an analysis result on a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention.
- FIG. 11 is a sectional view of the semiconductor device according to the first embodiment of the present invention.
- FIG. 12 is a sectional view of a first step in manufacturing a semiconductor device according to the first embodiment of the present invention.
- FIG. 13 is a sectional view of a second step in manufacturing the semiconductor device according to the first embodiment of the present invention.
- FIG. 14 is a sectional view of a third step of manufacturing the semiconductor device according to the first embodiment of the present invention.
- FIG. 15 is a sectional view of a semiconductor device according to a second embodiment of the present invention.
- FIG. 16 is a sectional view of a semiconductor device according to a third embodiment of the present invention.
- FIG. 17 is a sectional view of a semiconductor device according to a fourth embodiment of the present invention.
- FIG. 18 is a sectional view of a semiconductor device according to a fifth embodiment of the present invention.
- FIG. 19 is a sectional view of a semiconductor device according to a sixth embodiment of the present invention.
- FIG. 20 is a cross-sectional view of a first step in manufacturing a semiconductor device according to the seventh embodiment of the present invention.
- FIG. 21 is a sectional view of a second step of the semiconductor device according to the seventh embodiment of the present invention.
- FIG. 22 is a completed sectional view of a semiconductor device according to an eighth embodiment of the present invention.
- FIG. 23 is a cross-sectional view showing the order of the manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention c.
- FIG. 24 is a cross-sectional view showing the order of the manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention c 2 5 is a cross-sectional view c Figure 2 6 showing the manufacturing process sequence of the semiconductor device according to the eighth embodiment of the present invention, a ninth sectional view c showing the manufacturing process sequence of the semiconductor device according to an embodiment of the present invention 2 7 is a cross-sectional view c Figure 2 8 showing the manufacturing process sequence of the semiconductor device according to a ninth embodiment of the present invention, the finished cross-sectional view of a semiconductor device according to a ninth embodiment of the present invention.
- FIG. 29 is a completed sectional view of a semiconductor device according to a tenth embodiment of the present invention.
- FIG. 30 is a completed sectional view of the semiconductor device according to the eleventh embodiment of the present invention.
- FIG. 31 is a cross-sectional view of a completed semiconductor device according to the 12th embodiment of the present invention.
- FIG. 32 is a sectional view showing the order of manufacturing steps of the semiconductor device according to the thirteenth embodiment of the present invention.
- FIG. 33 is a completed sectional view of a semiconductor device according to a thirteenth embodiment of the present invention.
- FIG. 34 is a completed sectional view of the semiconductor device according to the fourteenth embodiment of the present invention.
- FIG. 35 is a sectional view showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention.
- FIG. 36 is a sectional view showing the order of manufacturing the semiconductor device according to the fourteenth embodiment of the present invention.
- FIG. 37 is a cross-sectional view showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention.
- FIG. 38 is a cross-sectional view showing the order of manufacturing steps of the semiconductor device according to the fifteenth embodiment of the present invention.
- FIG. 39 is a cross-sectional view showing a semiconductor device manufacturing process according to the fifteenth embodiment of the present invention.
- FIG. 40 is a completed sectional view of a semiconductor device according to a fifteenth embodiment of the present invention.
- FIG. 41 is a cross-sectional view showing the order of manufacturing the semiconductor device according to the sixteenth embodiment of the present invention.
- FIG. 42 is a completed sectional view of a semiconductor device according to a sixteenth embodiment of the present invention.
- FIG. 43 is a cross-sectional view showing the order of the manufacturing process of the semiconductor device according to the seventeenth embodiment of the present invention.
- FIG. 44 is a completed sectional view of a semiconductor device according to a seventeenth embodiment of the present invention.
- FIGS. 45A and 45B are explanatory diagrams of an as-necessary write / read storage device for explaining an application example of the embodiment of the present invention.
- FIGS. 46A and 46B are explanatory diagrams of a constant write / read storage device for explaining another application example of the embodiment of the present invention.
- FIG. 47 is a logic circuit diagram for explaining another application example of the embodiment of the present invention.
- FIG. 48 is an explanatory diagram of a computer configuration for explaining still another application example of the embodiment of the present invention.
- FIG. 49 is an explanatory diagram of an asynchronous transmission mode system for explaining another application example of the embodiment of the present invention.
- FIG. 50 is a sectional view of a semiconductor device according to an eighteenth embodiment of the present invention.
- FIG. 51 is a sectional view of a semiconductor device according to a ninth embodiment of the present invention.
- FIGS. 11 to 14 are cross-sectional views showing the steps of manufacturing a semiconductor device according to the first embodiment of the present invention, and FIG. 11 is a completed cross-sectional view thereof.
- ion implantation of p and subsequent heat treatment are performed using the gate protective insulating film 7 and the gate side wall insulating film 8 as an implantation preventing mask, the junction ends reach the oxide film 2, and the effective gate length is 15 O nm, and the maximum impurity concentration in the surface forming the source 9 and drain 1 0 by the low-concentration n-type diffusion layer so that Do a 1 x 1 0 1 Vcm 3.
- a high concentration p-type diffusion layer 11 having a maximum impurity concentration of 2 ⁇ 10 19 / cm 3 was formed inside the low concentration n-type source 9 and the drain 10 by ion implantation of BF 2 .
- High-concentration p-type diffusion layer 1 1 Minimum width between side junction and low-concentration source 9 or low-concentration drain junction is 50 nm, high-concentration P-type diffusion layer 11 It was about 3 O nm (FIG. 12).
- a W film 12 having a thickness of 15 O nm was deposited on the entire surface by sputtering, and was patterned so as to cover at least the surfaces of the high-concentration n-type diffusion layers 91 and 101.
- the W film 12 may be based on the selective chemical vapor deposition method on the Si surface (FIG. 14).
- a wiring protection insulating film 13 is deposited and an opening is formed at a desired position based on a known method of manufacturing a semiconductor device. Further, a source electrode 14 and a drain electrode 1 are formed by vapor deposition of wiring metal and patterning thereof. Wiring including 5 etc. was formed (Fig. 11).
- the withstand voltage between the source and the drain of the semiconductor device manufactured through this manufacturing process is 4.7 V, which is about 1 compared with the conventional structure SOI ⁇ MOS of the same size without the p-type diffusion layer 11 in the source. .5 V, and a breakdown voltage equivalent to that of MOS with the same dimensions normally manufactured on a semiconductor substrate could be secured.
- the current-voltage characteristics no abnormal bump-like characteristics called kink characteristics were observed, and the characteristics were normal.
- the existence of a leak current at a low gate voltage conventionally observed in SOI / MOS was not observed. No change was found in the leakage current and the threshold voltage even when the drain voltage was changed.
- the semiconductor device according to the present embodiment has been completely eliminated from the characteristics associated with the substrate floating effect. Since the current-voltage characteristics of the semiconductor device according to the present embodiment show normal characteristics, the semiconductor device shows high resistance homogeneity with the n-type high concentration diffusion layers 91 and 101 and is formed inside the source and the drain. It was also found that the p-type high-concentration diffused layer 11 had no adverse effect. This is because the region on the side of the p-type high-concentration diffusion layer 11 is composed of n-type low-concentration diffusion layers 9 and 10, and the surface channel region, which is the current path, is the n-type low-concentration diffusion layer 91 and 101.
- the holes that are minority carriers generated in the lower substrate of the channel are not in the n-type high-concentration diffusion layer 91 but in the p-type high-concentration diffusion layer 11 through the n-type low-concentration diffusion layer 9. It is thought to be injected. That is, it is considered that the presence of the n-type high-concentration diffusion layer 91 has no effect on the current transfer in the channel.
- the drain-to-substrate capacitance of the semiconductor device according to this example was also measured, but the capacitance value was equivalent to that of the conventional SOI'MOS, despite the presence of the p-type high-concentration diffusion layer 11. It was as small as about 1/10 of the value for the same size of MOS.
- the p-type high concentration diffusion layer 11 in the drain is surrounded by the n-type low concentration diffusion layer 10, and a depletion layer is formed in the p-type high concentration diffusion layer 11 by the drain electric field. It is considered that the capacitance is determined by the thick insulating film 2.
- the single-crystal Si layer 3 is extremely thin at 100 nm, and the substrate impurity concentration in the channel region is set as low as 1 ⁇ 10 17 / cm 3 . Therefore, the neutral region does not exist in the single crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single crystal Si layer 3, and the single crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is known that it is suitable for low voltage and high speed operation.
- a fully depleted MOS without a substrate floating phenomenon Indicate that it can be provided at a low price only by a conventional method of manufacturing a semiconductor device.
- FIG. 15 is a view showing a completed cross section of the semiconductor device according to the second embodiment of the present invention.
- the wiring protective film 13 is formed so that the high-concentration n-type diffusion layers 91 and 101 are not formed, and the source / drain electrodes 14 and 15 reach the high-concentration p-type diffusion layer 11.
- the single crystal Si film was slightly etched in the step of forming an opening in the substrate.
- the semiconductor device according to the present embodiment does not show a decrease in the breakdown voltage between the source and the drain, a bump-like characteristic in the current-voltage characteristic, a negative-direction fluctuation in the threshold voltage, and the like.
- FIG. 16 is a view showing a completed cross section of the semiconductor device according to the third embodiment of the present invention.
- the element isolation insulating film 4 was formed, and the active regions of the single-crystal Si film 3 were separated from each other. Region 31 was set.
- a gate oxide film 5, a gate electrode 61, a gate protection insulating film 7, and a gate sidewall insulating film 8 were formed on the low-concentration n-type region 31 and the low-concentration p-type region 3.
- a low-resistance polycrystalline Si film was used as the gate electrode 61.
- B ions are selectively implanted only into the low-concentration n-type region 31 and the heat treatment thereafter reaches the oxide film 2 and the surface impurity concentration is 5 XI 0
- Low-concentration p-type diffusion layers 90 and 100 of 1 cm 3 were formed.
- As ion implantation is performed so that the maximum impurity concentration is 1 ⁇ 10 ′ Vcm 3 inside the active region and the low impurity concentration p-type diffusion layers 90 and 100 are located inside the n-type diffusion layer 1. 10 were formed.
- P ions are selectively implanted into the low-concentration P-type region 3 using the gate electrode 61 and the gate side wall insulating film 8 as a mask, and the surface has a maximum concentration of 3 ⁇ 10 18 / cm 3 and an acid.
- a low-concentration n-type diffusion layer reaching the oxide film 2 was formed, and a source 9 and a drain 10 were formed.
- an ion implantation of Ge having a surface concentration of 1 ⁇ 10 2 Vcm 3 was performed in the source 9 and the drain 10 to form a Si ⁇ Ge eutectic layer 16 in the source 9 and the drain. .
- the wiring protection insulating film 13 is deposited, an opening is formed at a desired position, and the wiring is formed.
- a wiring including a ground potential line 17, an output terminal 18 and a power supply voltage line 19 was formed by vapor deposition and patterning of a metal film for use.
- CMOS of this example various symptoms caused by the substrate floating effect could not be observed for any of the pMOS and the nMOS. Furthermore, the through current due to the substrate floating effect unique to SOICMOS generated between the ground potential line 17 and the power supply line 19 due to the negative fluctuation of the nM0S threshold voltage and the positive fluctuation of the pMOS threshold voltage It was not observed.
- the absence of the substrate floating effect in the pMOS means that electrons, which are minority carriers generated in the single-crystal Si film 31 below the channel, are injected toward the n-type diffusion layer 110, and the source has a low concentration p. It is considered that the electrons reaching the diffusion layer 100 disappear by recombination in the depletion layer at the junction between the n-type diffusion layer 110 and the p-type diffusion layer 100.
- FIG. 17 is a view showing a completed cross section of a semiconductor device according to a fourth embodiment of the present invention.
- ion implantation of B was performed instead of ion implantation of Ge to form a high-concentration p-type diffusion layer 11 inside the low-concentration n-type diffusion layers 9 and 10.
- the high-concentration p-type diffusion layer 11 reached the oxide film 2, and the maximum impurity concentration of 2 ⁇ 10 ig / cm 3 was set so as to be located within 80 nm from the surface of the source 9.
- the pMOS is based on the same reason as in the third embodiment, and even in the nMOS, holes, which are a small number of carriers generated in the channel lower single-crystal Si film 3, are p-type diffusion layers 11 1 It is thought that the holes injected toward the source and reached the source low-concentration n-type diffusion layer 10 disappear by recombination at the depletion layer at the junction between the p-type diffusion layer 11 and the n-type diffusion layer 10 .
- the ion source is unstable, and it is not necessary to perform the ion implantation of Ge which requires a large ion current reaching a composition ratio of about 10%. Measures against the substrate floating effect were taken.
- FIG. 18 is a view showing a completed cross section of a semiconductor device according to a fifth embodiment of the present invention.
- an S0I substrate having a thickness of 300 nm of the single-crystal Si film 3 on the oxide film 2 was used, and the n-type diffusion layers 9 and 10 and the p-type diffusion layers 90 and 100 were formed. 150 nm junction depth Set to.
- the p-type diffusion layer 9 0 and 1 0 0 as interior positioned above the selectively formed n-type diffusion layer 1 1 0 high concentration p-type diffusion layer 9 2 and 1 0 2 BF 2
- the n-type diffusion layer 110 is formed!
- -Type diffusion layers 92 and 102 were electrically short-circuited at high piles.
- the maximum impurity concentration of the high-concentration p-type diffusion layers 92 and 102 was on the surface, and was set to 1 ⁇ 10 20 / cm 3 .
- the drain-to-substrate capacitance in the pMOS shows a capacitance value equivalent to that of the conventional SOI.MOS, which is about 1/10 of the value of the MOS of the same dimensions usually manufactured on the semiconductor substrate. It was small.
- the result is that the n-type high-concentration diffusion layer 110 in the drain is surrounded by the p-type low-concentration diffusion layer 100, and the n-type high-concentration diffusion layer 110 is depleted by the drain electric field. It is considered that the layer is not formed and the capacitance is determined by the thick insulating film 2.
- the single-crystal Si layers 3 and 31 are relatively thick at 300, In the substrate region, a depletion layer and a neutral region force ⁇ exist, that is, a so-called partially depleted structure is obtained even when a gate voltage higher than the threshold voltage is applied.
- the partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, but can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. It is shown that the semiconductor device according to the present embodiment can provide a countermeasure against the substrate floating of the partially depleted structure MOS at low cost.
- FIG. 19 is a view showing a completed cross section of the semiconductor device according to the sixth embodiment of the present invention.
- ion implantation of B was performed instead of ion implantation of Ge, and a high-concentration p-type diffusion layer 11 was formed inside the low-concentration n-type diffusion layers 9 and 10, and subsequently a high-concentration p-type diffusion layer 11 was formed.
- As ion implantation for forming n-type diffusion layers 91 and 101 was performed.
- the maximum impurity concentration was set to 5 ⁇ 10 20 / cm 3 at the surface part by As ion implantation, but this resulted in the generation of crystal defects at the upper junction of the p-type diffusion layer 11 and the P-type diffusion layer. 11 and the high concentration n-type diffusion layers 91 and 101 were electrically short-circuited with high resistance.
- the drain-to-substrate capacitance of both pM0S and nMOS shows the same capacitance value as that of the conventional S0IMOS, which is about the same as the value of the MOS of the same size usually manufactured on the semiconductor substrate. It was as small as 1 Z 10.
- the n-type high-concentration diffusion layer 110 in the drain is surrounded by the p-type low-concentration diffusion layer 100, and the p-type high-concentration diffusion layer 11 is surrounded by the low-concentration diffusion layer 10 It is considered that no depletion layer is formed in the n-type high-concentration diffusion layer 110 and the p-type high-concentration diffusion layer 11 due to the drain electric field, and the capacitance is determined by the thick insulating film 2.
- FIG. 20 is a sectional view showing a manufacturing process of the semiconductor device according to the seventh embodiment of the present invention
- FIG. 21 is a view showing a completed section thereof.
- a high-resistance polycrystalline Si film 20 having a thickness of 100 nm and a silicon oxide film 21 having a thickness of 10 were formed between the oxide film 2 and the single-crystal Si film 3.
- a multi-layer SOI substrate was used.
- the thickness of the single-crystal Si film 3 is 100 nm
- the impurity concentration of the p-type low concentration active region 3 and the n-type low concentration active region 31 is 1 ⁇ 10 16 / cm 3 , respectively.
- Set low that is, in this example, a fully depleted complementary MOS field effect transistor was manufactured.
- the high-resistance polycrystalline Si film immediately below the region where the nMOS gate electrode 7 is to be formed is formed.
- a high-concentration p-type impurity region 22 having an impurity concentration of 2 ⁇ 10 1 cm 3 is added to a high-resistance polycrystalline Si film 20 immediately below a region where a pMOS gate electrode 7 is to be formed.
- 1 0 l 8 / cm 3 comprising a high concentration of preformed ⁇ -type impurity region 2 3 (FIG. 2
- the semiconductor device was manufactured in the sixth embodiment from the state of FIG. 20, the high-concentration p-type diffusion layer 11 formed in the source of the nMOS, the high-concentration n-type diffusion layer 91 on the top, and the pMOS
- the conditions relating to the junction depth of the high-concentration n-type diffusion layer 110 in the source and the high-concentration p-type diffusion layer 92 in the upper part were as in Example 1.
- a W film 12 was selectively deposited on the entire surface of the source and drain except for the gate and the gate side wall insulating film portion according to Example 1.
- the source-drain current at a gate and drain voltage of 2 V of the semiconductor device according to the present embodiment was at least 1.4 times larger than that of pMOS and nMOS.
- an extremely steep drain conductance characteristic was obtained as compared with the semiconductor device of Example 6.
- the delay time per stage was 12 ps, which was 6 ps faster than the ring oscillator according to the sixth embodiment.
- Such ultra-high-speed, high-current characteristics are such that the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 is formed at a high temperature during the manufacturing process. Since it acts as an impurity diffusion mask for heat treatment, the impurity concentration of the P-type low-concentration active region 3 and the n-type low-concentration active region 31 forming the channel can be set to an extremely low concentration, and the mobility due to impurity scattering can be reduced. It is considered that the deterioration was prevented.
- FIGS. 23 to 25 are sectional views showing the order of manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention, and FIG. 22 is a completed sectional view.
- a silicon oxide film (hereinafter simply referred to as an oxide film) 2 having a thickness of 500 nm and a silicon oxide film having a thickness of 10 O nm are formed on a support substrate 1 made of a single crystal Si having a diameter of 12.5 cm.
- a S 0 I substrate consisting of a single crystal Si layer 3 with p conductivity type, resistivity of 10 ⁇ , and plane orientation of (100) by well-known M 0 S field effect transistor Film 4, a gate oxide film 5 having a thickness of 5 cm, a gate electrode 6 made of an n-type low-resistance polycrystalline Si film, and a gate protection insulating film 7 were formed.
- B ions Prior to the formation of the gate oxide film 5, B ions were implanted into the single-crystal Si layer 3 so that the threshold voltage became 0.1 V.
- the gate length is 20 O nm.
- the junction end reaches the oxide film 2
- the effective gate length is 15 O nm.
- the single-crystal Si layer 3 maintained the single-crystal property.
- a deposited insulating film having a thickness of 5 O nm is formed on the entire surface, and the insulating film is selectively left only on the side walls of the gate by anisotropic dry etching.
- a first sidewall insulating film 8 was formed.
- a semiconductor device based on the present example in which the thickness condition of the gate sidewall insulating film 8 was changed from 10 to 20 nm at an interval was also manufactured separately.
- a high-concentration p-type region 11 having a maximum impurity concentration of 2 ⁇ 10 19 / cm 3 was formed at the interface with the oxide film 2 inside the low-concentration n-type source 9 and the drain 10 by ion implantation of BF 2 .
- the width of the low-concentration n-type diffusion layer 9 remaining on the side surface of the high-concentration p-type region 11 was finally 4 O nm.
- the acceleration energy may be set so that the maximum impurity concentration is in oxide film 2.
- the high-concentration P-type region 11 was converted to amorphous by the above ion implantation (Fig. 23).
- the surface concentration 2 X 1 0 2 Vcm 3 becomes high-concentration n-type regions 9 1 and 1 0 1 a high concentration p-type diffusion layer 1 1 on the shape forming did.
- a short heat treatment at 100 ° C for 10 seconds was performed to activate the implanted ions, but the high-concentration P-type region 11 was fine except for the lateral region immediately below the gate. It was polycrystalline with a particle size of about 1 O nm in thickness.
- the junction depth of the high-concentration n-type regions 91 and 101 was about 40 nm (FIG. 24).
- a 15 O nm thick W film 12 was selectively deposited on the exposed Si surface by a chemical vapor reaction.
- the W film 12 may be formed by sputtering over the entire surface and by patterning so as to cover at least the high-concentration n-type regions 91 and 101 surfaces.
- a wiring protection insulating film 13 was deposited using a silicon oxide film to which phosphorus was added (FIG. 25).
- an opening is formed in a desired portion of the wiring protective insulating film 13, and further, a source electrode 14, a drain electrode 15 and the like are formed by vapor deposition of wiring metal and patterning thereof. Wiring was formed.
- the film thickness of the single crystal Si layer 3 in the final step was reduced to 5 O nm by the cleaning treatment in the manufacturing step (FIG. 22).
- the withstand voltage between the source and the drain of the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process is 4.7 V, which is smaller than that of the conventional structure SOIMOS having the same dimensions without the p-type diffusion layer 11 in the source.
- By about 1.5 V it was possible to secure a breakdown voltage equivalent to that of MOS transistors of the same dimensions normally manufactured on semiconductor substrates.
- no abnormal bump-like characteristics called kink characteristics were observed. showed that.
- the source-drain current-gate voltage characteristics are possible to secure.
- the interval from the high-concentration p-type region 1 1 end to the low-concentration n-type diffusion layer junction was observed from 10 ⁇ ⁇ ⁇ from bottom to 20. No observation was made, and it was clear that it was the most favorable. In the structure in which one end of the high-concentration p-type region 11 is beyond the low-concentration n-type diffusion layer junction, the substrate floating effect could not be eliminated.
- the semiconductor device according to the present embodiment is effective for eliminating the SOIMOS substrate floating effect, and the recrystallization of the polycrystalline injected holes of the high-concentration p-type region 11 formed in contact with the oxide film 2 It was speculated that it would work well as a center.
- the single-crystal Si layer 3 is as thin as 5 O nm, and the substrate impurity concentration in the channel region is set as low as 1 ⁇ 10 ′ Vcm 3 . Therefore, the neutral region does not exist in the single-crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single-crystal Si layer 3, and the single-crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is suitable for low voltage and high speed operation.
- FIG. 28 is a completed sectional view thereof.
- Si films 92 and 102 to which phosphorus was added at a high concentration were selectively formed in the above-mentioned etching region, and a heat treatment at 65 ° C. for 10 minutes was performed to activate the film.
- the source and drain consisted of a high-concentration n-type region.
- the formation of the Si films 92 and 102 may be performed under any of the conditions of single crystal epitaxial selective growth and polycrystalline selective growth. Further, the Si film deposited on the entire surface may be formed by patterning.
- the high-concentration n-type may be based on, for example, ion implantation instead of the simultaneous addition of impurities when forming the Si film.
- the above-mentioned Si films 92 and 102 are formed from the viewpoint of reducing the source resistance as if the Si films 92 and 102 had a surface formed above the initial surface before etching. The structure is desirable (Fig. 27).
- the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, no characteristic associated with the substrate floating effect was observed similarly to the semiconductor device according to the eighth embodiment, and a normal fully depleted SOI MOS characteristic was obtained.
- the high-concentration p-type region 11 including the interface with the high-concentration n-type regions 9 and 10 can be formed into a steep high-concentration distribution, so that the impurity concentration increases.
- the recombination time, which is reduced accordingly, can be further reduced, and the injected holes can be more efficiently eliminated as compared with the semiconductor device of the first embodiment. That is, it is more effective in eliminating the substrate floating effect.
- FIG. 29 is a completed sectional view of a semiconductor device according to another embodiment (No. 10) of the present invention.
- the high-concentration P-type region 11 is selectively removed except for a part adjacent to the gate sidewall insulating film 8, and then the Si film 92, and The semiconductor device was manufactured by performing the manufacturing steps after the selective formation of silicon and silicon.
- the selective removal of the high-concentration P-type region 11 was performed using a photomask, but a material different from the gate sidewall insulating film 8 from the state shown in Fig. 26 was used in order to guarantee self-alignment with the gate electrode.
- the second sidewall insulating film may be selectively formed by (silicon nitride film) and the high-concentration P-type region 11 may be selectively removed using the second sidewall insulating film as a mask. Further selective removal Alternatively, the high-concentration P-type region 11 may be reduced by performing high-concentration n-type ion implantation using the second sidewall insulating film as a mask on the high-concentration p-type region 11 (FIG. 29). ).
- the drain-substrate capacitance is reduced in proportion to the selective removal area of the high-concentration p-type region 11 with respect to the drain region having the same shape as the source region.
- the capacitance was reduced to the same level as that of a normal SOI MOS without the high-concentration p-type region 11.
- the capacitance value of a normal SOI MOS is about 1/10 of the value of a MOS of the same size usually manufactured on a semiconductor substrate.
- FIG. 30 is a completed sectional view of a semiconductor device according to another embodiment (11) of the present invention.
- Example 8 an SOI substrate having a thickness of 20 O nm was used as the single-crystal Si layer 3, the formation region of the high-concentration P-type region 11 was separated from the oxide film 2 interface, and the low-concentration n-type diffusion It was configured to be adjacent to the oxide film 2 via the layers 9 and 10. That is, the high-concentration p-type region 11 has a structure completely isolated from the outside by the low-concentration n-type diffusion layers 9 and 10. Other manufacturing steps were in accordance with Example 1 (FIG. 30).
- the semiconductor device according to the present embodiment manufactured through the above manufacturing process exhibits a low junction capacitance value equivalent to that of the conventional SOI * MOS, despite the presence of the high-concentration P-type region 11 in the drain region.
- the value was as small as about 1/10 of the value of MOS of the same size usually manufactured on a semiconductor substrate.
- the above result shows that the high-concentration p-type region 11 in the drain is surrounded by the low-concentration n-type diffusion layer 10, and no depletion layer is formed in the high-concentration p-type region 11 by the drain electric field. It is considered that the capacitance is determined by the thickness and the oxide film 2.
- FIG. 31 is a completed sectional view of a semiconductor device according to another embodiment (No. 12) of the present invention.
- Example 8 an S 0 I substrate having a thickness of 500 nm was used as the single crystal Si layer 3, and the active regions of the single crystal Si film 3 were separated from each other by forming an element isolation insulating film 4. After that, ion implantation was performed on a part of the active region according to a desired circuit configuration to obtain a low-concentration n-type region 31.
- the gate oxide film 5, the gate electrode 61, the gate protection insulating film 7, and the gate sidewalls were formed on the low-concentration n-type region 31 and the low-concentration p-type region 3.
- An edge film 8 was formed.
- a W film was used as the gate electrode 61.
- the gate electrode 61 and the gate side wall insulating film 8 as a mask, only the low-concentration n-type region 31 is selectively ion-implanted with B and then subjected to a heat treatment to have a junction depth of 20 O nm.
- Low-concentration p-type diffusion layers 90 and 100 having an impurity concentration of 5 ⁇ 10 18 / cm 3 were formed.
- As is ion-implanted so that the maximum impurity concentration is 5 ⁇ 10 2 cm 3 inside the active region and the low-concentration p-type diffusion layers 90 and 100 are located inside the high-concentration n-type region 1. 10 were formed.
- the high-concentration n-type region 110 was not single-crystallized even after the activation heat treatment due to the impurity concentration higher than the solid solution limit, and a polycrystalline crystal defect region was retained.
- P ions are selectively implanted into the low-concentration P-type region 3 using the gate electrode 61 and the gate sidewall insulating film 8 as a mask, and the maximum concentration is 3 ⁇ 10 18 / cm 3 at the surface , and 20
- a low-concentration n-type diffusion layer having a junction depth of O nm was formed, and was used as a source 9 and a drain 10.
- the ground potential line 17, the output terminal 18 and the power supply voltage line 19 are formed by depositing the wiring protective insulating film 13 and opening it at a desired location, depositing a wiring metal film and patterning the same. Was formed.
- the pMOS and the nMOS could not observe any symptoms caused by the substrate floating effect.
- SOI ⁇ CMOS penetrates due to the substrate floating effect peculiar to SOI and CMOS caused by the negative fluctuation of the nMOS threshold voltage and the positive fluctuation of the pMOS threshold voltage. No current was observed.
- the absence of the substrate floating effect in p MOS indicates that electrons, which are a small number of carriers generated in the channel single crystal Si film 31, were injected toward the high-concentration n-type region 110, and the crystal defect occurred. It is thought that it disappears due to recombination center based on the pit.
- the capacitance between the drain substrate in the PMOS shows a capacitance value equivalent to that of the conventional SOI ⁇ MOS, and is approximately 1 Z 1 of the value of the MOS of the same size usually manufactured on the semiconductor substrate. It was a small thing of 0.
- the above result is The high-concentration n-type region 110 in the rain is surrounded by the p-type low-concentration diffusion layer 100, and no depletion layer is formed in the high-concentration n-type region 110 due to the drain electric field. It is considered that the amount is determined by the thick insulating film 2.
- the single-crystal Si layers 3 and 31 are relatively thick at 50 O nm, and in the substrate region below the channel region, the depletion layer and the neutral region are applied even when a gate voltage higher than the threshold voltage is applied.
- a so-called partially depleted structure in which a region exists is obtained.
- the partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, it can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. This shows that the semiconductor device according to the present embodiment can provide a countermeasure against the substrate floating of the partially depleted structure MOS at low cost.
- FIG. 32 is a sectional view showing a manufacturing process of a semiconductor device according to another embodiment (the thirteenth) of the present invention
- FIG. 33 is a completed sectional view thereof.
- the thickness of the single-crystal Si film 3 was set to 200, and the high-resistance multilayer having a thickness of 100 nm was formed between the oxide film 2 and the single-crystal Si layer 3 having a thickness of 200 nm.
- a multi-layer SOI substrate having a crystalline Si film 20 and a 10 nm thick silicon oxide film 21 was used.
- the thickness of the single-crystal Si film 3 is 10 O nm, and the impurity concentration of the p-type low concentration active region 3 and the n-type low concentration active region 31 is 1 ⁇ 10 16 cm 3 , respectively.
- Set low That is, in this example, a completely depleted complementary MOS field effect transistor was manufactured.
- the high-resistance polycrystalline Si directly under the region where the nMOS gate electrode 7 is to be formed is formed prior to the formation of the gate oxide film 5, the gate electrode 6, the gate protection insulating film 7, and the gate sidewall insulating film 8, the high-resistance polycrystalline Si directly under the region where the nMOS gate electrode 7 is to be formed is formed.
- a high-concentration p-type impurity region 22 having an impurity concentration of 2 ⁇ 10 1 cm 3 is formed in the film 20, and an impurity concentration 2 is formed in the high-resistance polycrystalline Si film 20 immediately below the region where the p-MOS gate electrode 7 is to be formed.
- the X 1 0 1 8 / cm 3 comprising a high concentration n-type impurity region 2 3 preformed (FIG. 2).
- a semiconductor device was manufactured according to the embodiment 12 from the state of FIG. 32. However, a high-concentration n-type region 91 and a high-concentration n-type region 91 were formed above the high-concentration p-type region 11 formed in the source of nMOS. The high-concentration p-type regions 92 were respectively formed above the high-concentration n-type regions 110 in the source in FIG. Each of the above junction depths was the same as the condition in Example 11 described above. Further, a W film 12 was selectively deposited on the entire surface of the source and drain except for the gate and gate sidewall insulating films according to Example 4 (FIG. 33).
- the delay time per stage was 12 ps, which was 6 ps faster than the ring oscillator according to the fifth embodiment.
- the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 acts as an impurity diffusion mask for high-temperature heat treatment during the manufacturing process. Therefore, the ultra-high-speed and large-current characteristics of the semiconductor device according to the present embodiment can be set to extremely low impurity concentrations in the p-type low-concentration active region 3 and the n-type low-concentration active region 31 forming the channel. It is considered that the mobility was prevented from deteriorating due to impurity scattering.
- FIGS. 34 to 37 are sectional views showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention, and FIG. 34 is a completed sectional view thereof.
- An S0I substrate consisting of a single crystal Si layer 3 having a thickness of 100 Qcm and a plane orientation of (100) is formed on a S0I substrate by a known method of manufacturing an M0S field-effect transistor.
- a gate oxide film 5 a gate electrode 6 made of an n-type low-resistance polycrystalline Si film, and a gate protection insulating film 7 were formed. Prior to the formation of the gate oxide film 5, B ions were implanted into the single-crystal Si layer 3 so that the threshold voltage became 0.1 V.
- the gate length is 20 O nm.
- ion implantation of p and subsequent heat treatment are performed using the gate protection insulating film 7 and the gate electrode 6 as an implantation blocking mask, the junction ends reach the oxide film 2, the effective gate length is 15 O nm, and the surface is maximum impurity concentration formed a 2 XI 0, 6 / cm 3 and the source 9 and drain 1 0 by the low-concentration n-type diffusion layer as made in.
- ion implantation of As and subsequent heat treatment are performed, resulting in high-concentration n-type diffusion. Layers 91 and 101 were formed.
- the single-crystal Si layer 3 maintained single-crystallinity (FIG. 35).
- a 50-nm-thick deposited insulating film is formed on the entire surface, and the above-mentioned insulating film is selectively left only on the gate side wall by anisotropic dry etching to form a gate side wall insulating film 8.
- a semiconductor device based on the present embodiment in which the minimum film thickness is changed to 20 nm and the maximum film thickness is set to 0.5 / im at intervals of 10 to 100 is also used. Manufactured separately.
- Si ions were implanted under the conditions of a dose of 3 ⁇ 10 1 cm 2 so that the concentration was maximized at the interface between the low concentration n-type source 9 and the oxide film 2 inside the drain 10.
- a heat treatment was performed at 800 ° C. for 10 minutes.
- Observation of the cross section of a sample separately manufactured under the same conditions with a transmission electron microscope revealed that twins of fine crystal grains, that is, crystal defect regions 11 were formed near the oxide film 2 interface. became.
- the acceleration energy may be set so that the maximum concentration is in oxide film 2.
- ion implantation for forming the crystal defect region 11 does not need to be Si as an ion species, and rare gas elements such as Ne and Ar, halogen elements such as F and C1, It has been found that the same effect can be obtained even with a group 14 element such as C and Ge. However, it has been found that the ion implantation of the n-conductivity-type element in the Si single crystal like P has no effect as described later. ( Figure 36).
- a 150 nm thick W film 12 was selectively deposited on the exposed Si surface by a chemical vapor reaction.
- the W film 12 may be formed by sputtering over the entire surface and by patterning so as to cover at least the high-concentration n-type regions 91 and 101 surfaces.
- a wiring protection insulating film 13 was deposited using a silicon oxide film to which phosphorus was added (FIG. 37).
- openings are formed at desired locations in the wiring protection insulating film 13 based on a known method of manufacturing a semiconductor device, and further, a source electrode 14 and a drain electrode 15 are formed by depositing and patterning wiring metal. The wiring including was formed. ( Figure 34).
- the semiconductor device according to the present embodiment manufactured through the above manufacturing process has a source-drain withstand voltage of 4.7 V and a conventional structure SOI MOS having the same dimensions without the p-type diffusion layer 11 in the source. About 1.5 V higher than normal, usually manufactured on semiconductor substrates A breakdown voltage equivalent to that of MOS of the same dimensions could be secured. Also, in the current-voltage characteristics, no abnormal bump-like characteristics called kink characteristics were observed, and the characteristics were normal. Furthermore, in the source-drain current-gate voltage characteristics, the existence of a leak current at a low gate voltage conventionally observed in SOI / MOS was not observed. Also, no change was found in the leakage current and the threshold voltage even when the drain voltage was changed.
- the semiconductor device according to the present example was completely eliminated from the characteristics associated with the substrate floating effect.
- the current-voltage characteristics of the semiconductor device according to this example showed normal characteristics, and it was also found that the crystal defect region 11 formed inside the source and the drain had no adverse effect.
- the elimination of the above substrate floating effect according to the present embodiment is achieved by changing the thickness of the gate side wall insulating film 8 according to the present embodiment and changing the thickness of the gate defect side wall insulating film 8 from the end of the crystal defect region 11 to the low concentration n-type diffusion layer.
- the crystal defect region 11 was formed based on the P ion implantation, no elimination of the substrate floating effect was found.
- the semiconductor device according to the present embodiment is effective in eliminating the substrate floating effect of SOIMOS is effective in regenerating holes into which polycrystals in the crystal defect region 11 formed in contact with the oxide film 2 are injected. It was presumed that it sufficiently acted as a crystal center.
- the single-crystal Si layer 3 is relatively thick at 20 O nm, and in the substrate region below the channel region, even if a gate voltage equal to or higher than the threshold voltage is applied, the single-crystal Si layer 3 becomes in a depletion layer.
- a so-called partially depleted structure having a neutral region is obtained.
- the partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, it can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. This shows that the semiconductor device according to the present embodiment can provide a countermeasure against floating of the partially depleted structure MS substrate at low cost.
- FIG. 38 to 40 are sectional views showing the order of manufacturing steps of a semiconductor device according to another embodiment of the present invention, and FIG. 40 is a complete sectional view thereof.
- the gate protection insulating film 7 is used as a mask from the state in which the steps up to the formation of the element isolation insulating film 4, the gate oxide film 5, the gate electrode 6, and the gate protection insulating film 7 based on the embodiment 14 are performed.
- High-concentration ion implantation with low acceleration energy of 2 keV of As, junction depth 1 O nm, surface impurity concentration 1 X Shallow junction n-type high concentration diffusion layers 95 and 105 of 102 and / cm 3 were formed.
- a gate sidewall insulating film 8 having a thickness of 100 nm is formed in accordance with the first embodiment, and high-concentration ion implantation of P is performed using the gate sidewall insulating film 8 as a mask, and a junction depth of 10 O nm
- the low resistance source diffusion layer 91 and the low resistance drain diffusion layer 101 were formed (FIG. 38).
- a wiring protection insulating film 13 was deposited and an opening was formed at a desired location.
- the n-type low concentration diffusion layer 9 having a minimum impurity concentration of 1 ⁇ 10 16 / cm 3 contacting the lower part of the low resistance source diffusion layer 9 1 and reaching the base oxide film 2 is formed. Formed.
- the n-type low concentration diffusion layer 10 is simultaneously formed also at the bottom of the drain diffusion layer. After performing a heat treatment for adjusting the activation and diffusion depth of the n-type low concentration diffusion layers 9 and 10, a deposition film 13 1 made of a different material from the wiring protection insulating film 13 is formed on the side surface of the opening. It was selectively left using the dry etching technique.
- the opening side wall insulating film 13 1 has a meaning that a film serving as an ion implantation mask has a constant thickness from the opening on the side wall, and even if it exists on the bottom of the opening, there is no problem in the next step. Does not occur. From this state, high-energy ion implantation of Si using the opening side wall film 13 1 as a mask was performed based on the conditions of the above Example 14, and the crystal defect region 11 was formed in the n-type low concentration diffusion layer 9. An oxide film was formed near the interface between the two. In the above manufacturing process, the method of using the open side wall film 13 1 to adjust the distance from the end of the crystal defect region 11 to the junction of the n-type low concentration diffusion layer 9 has been described. It may be based on a method of adjusting the depth by heat treatment. In this case, the step of forming the side wall film 131 can be omitted (FIG. 39).
- FIG. 40 a cross-sectional view of a semiconductor device according to a step of forming a wiring material after removing the opening side wall film 131 is illustrated, but the opening side wall film is left as desired. This does not cause any problems (Fig. 40).
- the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, similar to the semiconductor device according to Embodiment 14 described above, no characteristic associated with the substrate floating effect is observed, and a normal partially depleted SOIMOS characteristic is obtained. I got it. Further, in the semiconductor device according to the present embodiment, the elimination of the substrate floating effect can be realized only in the contact hole region. Therefore, there is no restriction on the shape of the source / drain diffusion layer near the gate electrode end that determines the transistor characteristics. Therefore, according to this embodiment, desired transistor characteristics can be realized without the influence of the substrate floating effect.
- FIG. 41 is a sectional view showing a semiconductor device manufacturing process according to another embodiment (16th) of the present invention
- FIG. 42 is a completed sectional view thereof.
- Example 14 after the active region of the single-crystal Si film 3 was separated from each other by forming an element isolation insulating film 4 using a SOI substrate having a thickness of 50 O nm as the single-crystal Si layer 3 According to a desired circuit configuration, a part of the active region was ion-implanted to form a low-concentration n-type region 31.
- a gate oxide film 5, a gate electrode 61, and a gate protection insulating film 7 were formed on the low-concentration n-type region 31 and the low-concentration P-type region 3 according to Example 14 described above.
- a W film was used as the gate electrode 61. From this state, ion implantation was performed using the gate electrode 61 as a mask according to Example 14 described above. The above-described ion implantation is performed in the low-concentration P-type region 3 in accordance with the above-described embodiment, by performing P ion implantation for forming the low-concentration n-type diffusion layers 9 and 10 and forming the high-concentration n-type diffusion layer 9K101.
- low-concentration p-type diffusion layers 90 and 100 are formed by B ion implantation, and high-concentration P-type diffusion layers 92 and 100 are formed. Formed two.
- the low-concentration n-type diffusion layers 9 and 10 and the low-concentration P-type diffusion layers 90 and 100 are formed so as to reach the oxide film 2, and the maximum impurity concentration is finally 1 ⁇ 10 ′ Vcm.
- the silicon nitride film is selectively left in the nMOS formation region, and the silicon nitride film is formed only on the side wall of the nMOS gate by anisotropic dry etching.
- a gate sidewall insulating film 8 having a thickness of 10 O nm.
- an oxide film was selectively left in the pMOS formation region, and a 20-nm-thick gate sidewall insulating film 8 of an oxide film was formed only on the sidewall of the pMOS gate by anisotropic dry etching.
- the gate sidewall insulating films 8 of the pMOS and the nMOS may have the same thickness and the same material, if desired.
- Ar ions are implanted at a dose of 5 ⁇ 10 15 / cm 3 so as to reach the oxide film 2 interface, and the low-concentration n-type diffusion layer 9 is formed.
- a crystal defect region 11 was buried therein.
- the formation of the crystal defect region 11 is due to Ar Irrespective of the ion implantation, the ion implantation of Si or the like as in the first embodiment may be used. Due to the influence of the base oxide film 2, the crystal defect region 11 was not monocrystallized in the region in contact with the base oxide film 2 even by the recrystallization heat treatment, and a polycrystalline crystal defect region was retained (FIG. 4). 1).
- the wiring protection insulating film 13 is deposited and opened at a desired location, and the wiring metal film is deposited and patterned to form a ground potential line 17, an output terminal 18 and a power supply voltage line.
- a wiring including 19 was formed (FIG. 42).
- the CMOS, and the pMOS and the nMOS according to the present embodiment manufactured through the above manufacturing process no symptom caused by the substrate floating effect could be observed in any of the pMOS and the nMOS. Furthermore, a through current caused by the substrate floating effect peculiar to SO ICMOS generated between the ground potential line 17 and the power supply voltage line 19 due to the negative fluctuation of the nMOS threshold voltage and the positive fluctuation of the pMOS threshold voltage is also observed. Was not done.
- the substrate floating effect of the nMOS and the pMOS can be eliminated by the same ion implantation process, and the performance of the CMOS can be improved at low cost without complicating the manufacturing process.
- the substrate floating effect of the nMOS and the pMOS can be eliminated by the same ion implantation process, and the performance of the CMOS can be improved at low cost without complicating the manufacturing process.
- the crystal defect region 11 acting as a recrystallization center and the low concentration in the lower part of the channel! ) Type gate region 3 or gate sidewall insulating film 8 for determining the distance between low-concentration n-type regions 31 was formed so as to have a different thickness between nM0S and pM0S.
- the above is intended to correct the junction depth in consideration of the fact that the junction depth is different due to the difference in impurities between the low-concentration P-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10.
- the maximum impurity concentration of the low-concentration p-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10 is 1 ⁇ 10 15 / cm 3 or more, and 1 ⁇ 10 1 cm 3 It is desirable that the value be in the range of 1 ⁇ 10 16 to 5 ⁇ 10 ′ 7 / cm 3 . This in 5 X 1 0 'Vcm 3 or less, elimination of substrate floating phenomenon is particularly pronounced Chikaraku, whereas 1 X 1 0 16 / cm 3 1/1 0 12 A about the reverse characteristics of the pn junction in the following Small current is generated, which may result in transistor leakage current.
- the ion implantation process for forming the crystal defect region 11 may be performed only once in the CMOS, and the amorphous formation can be achieved at the interface with the base oxide film 2 in consideration of the thickness of the SOI layer. I just need.
- the ion implanted element may be any material that does not cause a change in conductivity type in the low-concentration P-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10. It is preferable to use halogen elements such as Si, Ge, C, etc., and F, C1, etc., and further, rare gas elements such as Ne, Ar, etc.
- FIG. 43 is a sectional view showing a manufacturing process of a semiconductor device according to another embodiment (17th) of the present invention
- FIG. 44 is a completed sectional view thereof.
- the thickness of the single-crystal Si film 3 was set to 100 nm, and the high-resistance of 100 nm was formed between the oxide film 2 and the single-crystal Si film 3 having a thickness of 200 nm.
- a multi-layer SOI substrate having a polycrystalline Si film 20 and a 10 nm thick silicon oxide film 21 was used.
- the impurity concentration of the p-type low-concentration active region 3 and the n-type low-concentration active region 31 in the single-crystal Si film 3 was set to an extremely low value of 1 ⁇ 10 ′ Vcm 3 , respectively. That is, in this example, a fully depleted phase-trapping MOS field-effect transistor was manufactured.
- the high-resistance polycrystalline Si immediately below the region where the nMOS gate electrode 7 is to be formed is formed prior to the formation of the gate oxide film 5, the gate electrode 6, the gate protection insulating film 7, and the gate sidewall insulating film 8, the high-resistance polycrystalline Si immediately below the region where the nMOS gate electrode 7 is to be formed is formed.
- a high-concentration P-type impurity region 22 having an impurity concentration of 2 ⁇ 10 1 cm 3 is formed in the film 20, and an impurity concentration 2 is formed in the high-resistance polycrystalline Si film 20 directly below the region where the pMOS gate electrode 7 is to be formed. It was preformed X 1 0 1 cm 3 comprising a high concentration n-type impurity region 2 3 (4 3).
- a semiconductor device was manufactured from the state shown in FIG. 43 according to Example 16 above. However, in the low-concentration n-type diffusion layer 9 adjacent to the lower part of the nMOS source high-concentration n-type diffusion layer 9 1, and the source in the pMOS A recombination center region 11 composed of a crystal defect layer is formed in the low-concentration p-type diffusion layer 90 adjacent to the lower portion of the high-concentration p-type diffusion layer 90 by an Ar ion implantation process so as to be in contact with the base oxide film 2. did.
- the junction depth of the source high-concentration n-type diffusion layer 91 and the high-concentration p-type diffusion layer 102 was set to about 5 Ornn.
- the low-concentration n-type diffusion layer 9 and the low-concentration P-type diffusion layer 90 are configured so that their bottoms are in contact with the base oxide film 2 and their maximum impurity concentrations are from 1 ⁇ 10 16 to 2 ⁇ 10,7 . It was set as become / cm 3. Further, a W film 12 having a thickness of 10 O nm based on a chemical vapor reaction was selectively deposited on the entire surface of the source and drain except for the gate and gate sidewall insulating film portions, and then wiring was performed in accordance with the third embodiment. By forming the protective insulating film 13 and opening it at a desired location, and further depositing a wiring metal film and patterning the same, wiring including the ground potential line 17, the output terminal 18 and the power supply voltage line 19 is formed. Formed (Fig. 44).
- the pMOS and the nMOS are not observed with any symptom caused by the floating substrate effect as in the above embodiment.
- the source / drain current at a gate and drain voltage of 2 V of the semiconductor device according to the present embodiment is at least 1.4 times larger than that of the semiconductor device of Embodiment 16 in both pMOS and nMOS. Currentization has been achieved. In the non-saturation characteristic region where the power is also low and the drain voltage is low, an extremely steep drain conductance characteristic was obtained as compared with the semiconductor device of Example 16 described above.
- the delay time per stage was 12 ps, which was as high as 6 ps as compared with the ring oscillator according to the third embodiment.
- the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 serves as an impurity diffusion mask for high-temperature heat treatment during the manufacturing process. Therefore, the ultra-high-speed and large-current characteristics of the semiconductor device according to the present embodiment make it possible to set the impurity concentration of the P-type low-concentration active region 3 and the n-type low-concentration active region 31 constituting the channel to an extremely low concentration. It is considered that the deterioration of mobility due to impurity scattering was prevented.
- the single-crystal Si layer 3 is extremely thin at 50 feet, and the substrate impurity concentration in the channel region is set as low as 1 ⁇ 10 ′ Vcm 3 . Therefore, the neutral region does not exist in the single-crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single-crystal Si layer 3, and the single-crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is suitable for low voltage and high speed operation.
- FIGS. 45 to 49 below illustrate examples of application of the semiconductor device disclosed in the embodiments described above.
- FIGS. 45A and 45B are diagrams showing application examples of the semiconductor device according to the embodiment of the present invention.
- a random access memory device (referred to as DRAM) is constituted by a semiconductor device based on the present invention.
- the memory cell that is one storage unit is
- the semiconductor device according to the present invention is formed by connecting one semiconductor device and one capacitance element Cs in series, and is connected to a bit line as a data transmission line and a word line for input / output control.
- the random access memory device is composed of a memory cell array in which memory cells are arranged in rows and columns and a control peripheral circuit.
- the peripheral circuit is also composed of the semiconductor device of the present invention.
- the column address signal and the row address signal are shifted to reduce the number of address signal terminals for memory cell selection, and the multiplexed and applied force ⁇ RAS and CAS are each pulse signal, and clock generators 1 and 2 are used. Under control, the address signal is distributed to the row decoder and the column decoder.
- Specific address lines and bit lines are selected according to address signals distributed to the row decoder and the column decoder by an address buffer as a buffer circuit.
- a sense amplifier using a flip-flop type amplifier is connected to each bit line, and amplifies a signal read from a memory cell.
- the pulse signal WE controls the switching between writing and reading by controlling the writing clock generator.
- D is a write and read signal.
- each of the semiconductor devices constituting this application example is constituted by the semiconductor device according to the present invention, a high-speed operation in which the access time can be reduced by 30% or more compared with the conventional device can be realized.
- the worst-case refresh characteristics were 0.5 seconds, which was a 10-fold improvement over the past.
- the high-speed operation described above is due to the effect of reducing the parasitic capacitance by the SOI structure, and the increase in current based on the seventh embodiment.
- the improvement of the refresh characteristics is based on the reduction of the junction area by the SOI structure and the elimination of the threshold voltage fluctuation by eliminating the substrate floating effect.
- FIGS. 46A and 46B are diagrams showing another application example of the semiconductor device according to the embodiment of the present invention.
- This example is an example in which a constant write / read type storage device (referred to as SRAM) is constituted by a semiconductor device based on the present invention.
- a memory cell as one storage unit is composed of two sets of complementary MOSs according to the present invention and two MOSs (referred to as transfer MOSs) for controlling signal input / output, as shown in FIG. 46B. It consists of.
- This SRAM is composed of a memory cell array in which memory cells are arranged in rows and columns and a control peripheral circuit, and the peripheral circuit is also composed of the semiconductor device of the present invention.
- Configuration of this example The configuration is basically the same as that of Figs. 45A and 45B, and an address transition detector is provided to achieve the high speed and low power consumption of the SRAM. Controlling.
- the row decoder is composed of two stages, a predecoder and a main decoder.
- the chip select is a circuit for avoiding contention between information writing and reading data by the signals CS and WE, and making the write cycle time and the read cycle time almost equal to enable high-speed operation.
- the power supply voltage can be reduced from 3.5 V to 2.0 V and the access time can be reduced by 30% or more compared to the conventional device, because each of the semiconductor devices constituting the semiconductor device of this example is a semiconductor device according to the present invention.
- the high speed that can be achieved was realized. This is due to the effect of reducing the parasitic capacitance by the SOI structure, the increase in current based on the seventh embodiment, and the drastic improvement in drain conductance at low voltage. Furthermore, the threshold voltage fluctuation due to the elimination of the substrate floating effect has been eliminated, and the operation range of the sense amplifier has been reduced to enable higher speed.
- FIG. 47 is a diagram showing still another application example of the semiconductor device according to the embodiment of the present invention.
- This example is an example in which a logic circuit device is configured using a semiconductor device according to the present invention.
- the figure shows an example of a composite gate circuit.
- the present invention is applied to a logic circuit including a NAND circuit and a NOR circuit in a composite gate circuit by a semiconductor device according to the present invention.
- each of the semiconductor devices constituting the semiconductor device of the present embodiment is constituted by the semiconductor device according to the present invention, the delay time can be reduced by 20% or more compared with the conventional logic circuit device. This is due to the effect of reducing the parasitic capacitance by the SOI structure, the increase in current according to the seventh embodiment, and the drastic improvement in drain conductance at low voltage.
- FIG. 1 Another application example will be described with reference to a computer configuration diagram in FIG. This example is an example in which the semiconductor device of the present invention is applied to a high-speed large-scale computer in which a plurality of processors 500 for processing instructions and operations are connected in parallel.
- the semiconductor device 500 of the present invention having 10 to 30 pixels on one side includes the device 500, the system control device 501, the main storage device 502, and the like.
- a processor 500 for processing these instructions and calculations, a system controller 501, and a data communication interface 503 composed of a compound semiconductor device are mounted on the same ceramic substrate 506. Further, the data communication interface 503 and the data communication control device 504 are mounted on the same ceramic substrate 507.
- the ceramic substrates on which the ceramic substrates 506 and 507 and the main storage device 502 were mounted were mounted on a substrate having a size of about 50 cm on a side or smaller, thereby forming a central processing unit 508 for a computer. .
- the data communication in the central processing unit 508, the data communication between a plurality of central processing units, or the data communication between the data communication interface 503 and the board 509 on which the input / output processor 505 is mounted are illustrated by arrows in FIG. The operation was performed through an optical fiber 510 indicated by a dotted line.
- a processor 500 for processing instructions and operations, a system controller 501, and a main storage device 502 are formed by using the semiconductor device according to the present invention in parallel, and operate at high speed, and Since communication is performed using light as a medium, the number of command processing operations per second could be greatly increased.
- FIG. 49 is a diagram showing still another application example of the semiconductor device according to the embodiment of the present invention.
- the present example is a signal transmission processing device constituted by the semiconductor device according to the present invention, particularly a signal transmission processing device relating to an asynchronous transmission system (referred to as an ATM switch).
- the information signal transmitted in series at a very high speed by an optical fiber is converted into an electric signal (0 / E conversion) and is converted into a parallel signal (SZP conversion). It was introduced into an integrated circuit (BFMLS I) composed of a semiconductor device based on the invention.
- the electrical signal that has been addressed by the integrated circuit is output via a serialized (P / S conversion) and optical signal (E / 0 conversion) optical fiber.
- the BFMLS I consists of a multiplexer (MUX), a buffer memory (BFM), and a separator (DMUX).
- MUX multiplexer
- BFM buffer memory
- DMUX separator
- the BFM LSI is controlled by a memory control LSI and an LSI (empty address FIFO memory LSI) having an empty address distribution control function.
- This signal transmission processor is an ultra-high-speed transmission signal sent regardless of the address to be transmitted.
- the BFM LSI is constituted by the semiconductor device based on the present invention, and the operation speed is three times faster and less expensive than the conventional BFMLSI. It has become possible to reduce the storage capacity of ML SI to about 1/3 that of the conventional model. As a result, the manufacturing cost of the ATM exchanger was reduced.
- FIG. 50 is a view showing a completed cross section of the semiconductor device according to the eighteenth embodiment of the present invention.
- Example 6 instead of the ion implantation for forming the high-concentration n-type diffusion layers 91 and 101, the selective removal of the Si layer on the high-concentration P-type diffusion layer 11 and the high-concentration n The polycrystalline Si film 24 to which the type impurity was added was selectively left.
- ion implantation for forming high-concentration p-type diffusion layers 92 and 102 selective removal of Si layer on high-concentration n-type diffusion layer 110 and high-concentration p-type impurities Selective remaining of the added polycrystalline Si film 25 was performed.
- a method of selectively depositing only the Si exposed surface based on a low-pressure chemical vapor reaction using dichlorosilane as a raw material was used for selectively leaving the polycrystalline Si film 25.
- it may be formed by depositing amorphous or polycrystalline Si over the entire surface, implanting P and n-type impurities at desired locations with high concentration ions, and then patterning the Si film. Good.
- CMOS of the present example no symptom caused by the substrate floating effect was observed for any of pMOS and nMoS as in the semiconductor device of Example 6.
- a polycrystalline Si film doped with a high concentration n-type impurity and a polycrystalline Si film doped with a high concentration P-type impurity Good ohmic characteristics were exhibited between 25 and the high-concentration n-type diffusion layer 110 due to the presence of the crystal grain boundaries of the polycrystalline film.
- the polycrystalline Si films 24 and 25 can be made thicker so as to reach the upper part of the monocrystalline Si film surface if necessary and to be located on the side of the gate electrode, so that the source resistance can be reduced. .
- FIG. 51 is a view showing a completed cross section of a semiconductor device according to another embodiment (No. 19) of the present invention. It is.
- Example 6 a single crystal Si substrate 30 having the same specification as the single crystal Si layer 3 and having a thickness of 62 ⁇ m was used in place of the SOI substrate as a substrate constituting pMOS and nMOS.
- c instead of the low-concentration n-type region 31 and the low-concentration p-type region 3, a low-concentration n-type well 32 and a low-concentration P-type well 33 are formed based on a known method for manufacturing a semiconductor device.
- a stacked film of the high-concentration n-type polycrystalline Si film 40 and the W film 12 is formed on the high-concentration n-type diffusion layers 91 and 101, and the high-concentration p-type diffusion layer 92 is formed.
- a stacked film of a high-concentration p-type polycrystalline Si film 41 and a W film 12 was formed to reduce source resistance.
- both the pMOS and the nMOS are caused by a cell potential floating effect such as a threshold voltage fluctuation, although no electrode for supplying a cell potential is provided. No cord symptoms were observed. In other words, the need for a fixed gel potential fixed electrode, which was required in the conventional complementary MOS, was eliminated, thereby contributing to a reduction in chip occupation area. Further, in the semiconductor device according to the present embodiment, the source resistance could be reduced by the effect of the laminated film of the low-resistance polycrystalline film and the W film stacked on the source.
- the substrate floating effect of pMOS on the SOI substrate which was impossible in the past, can be solved by a low-cost manufacturing method. Therefore, according to the present invention, the substrate floating effect can be completely eliminated by a low-cost manufacturing method for CMOS on the SOI substrate. Thus, a low-voltage, low-power, ultra-high-speed semiconductor device and a system including the semiconductor device can be provided.
Landscapes
- Thin Film Transistor (AREA)
Abstract
Description
明 細 書 半導体装置及びその製造方法 技術分野 Description Semiconductor device and method for manufacturing the same
本発明は半導体装置に係り、 特に絶縁膜上に構成された低寄生容量で高速動作 可能な高性能 MO S電界効果型トランジスタ及びその製造方法に関する。 The present invention relates to a semiconductor device, and more particularly to a high-performance MOS field-effect transistor formed on an insulating film and capable of operating at high speed with low parasitic capacitance, and a method of manufacturing the same.
背景技術 Background art
絶縁膜上の単結晶半導体層にトランジスタを構成する手法は S 0 I (シリコン ·オン ·インシユレ一タ: Silicon On Insulator)構造として公知であり、 図 1 に示されるような構造が 1995年春季応用物理学会講演予稿集 755ページ等 に記載されている。 MOS電界効果型トランジスタ (以降単に MOSと略記する) は支持基板 1から厚い絶縁膜 2により隔離された単結晶シリコン (以下 S i)膜 3に構成される。 図 1で、 4は素子間分離絶縁膜、 5はゲート絶縁膜、 61はゲ —ト電極、 7はゲート保護絶縁膜、 8はゲート側壁絶縁膜、 9及び 1 0は n型高 濃度拡散層で各々ソース, ドレイン領域である。 図 1のような SO I 'MOSは 直下に厚い絶縁膜 2を有しているためドレイン接合容量、 及び配線寄生容量が従 来 MOSに比べて 1Z1 0程度にまで低減できる特徴を有している。 更に MOS が基板から絶縁分離されているため α線照射による誤動作、 及びラツチアツプ現 象を根本的に解消できる等の特徴を有している。 The method of forming a transistor in a single crystal semiconductor layer on an insulating film is known as an S 0 I (Silicon On Insulator) structure, and the structure shown in FIG. It is described in 755 pages of proceedings of lectures of the Physical Society of Japan. A MOS field effect transistor (hereinafter simply abbreviated as MOS) is composed of a single crystal silicon (Si) film 3 separated from a supporting substrate 1 by a thick insulating film 2. In FIG. 1, 4 is an isolation insulating film, 5 is a gate insulating film, 61 is a gate electrode, 7 is a gate protective insulating film, 8 is a gate sidewall insulating film, and 9 and 10 are n-type high concentration diffusion layers. Are the source and drain regions, respectively. The SO I'MOS as shown in Fig. 1 has a feature that the drain junction capacitance and wiring parasitic capacitance can be reduced to about 1Z10 compared to the conventional MOS because it has a thick insulating film 2 directly underneath. . Furthermore, since the MOS is insulated and separated from the substrate, it has features such as malfunction due to α-ray irradiation and radical phenomena can be fundamentally eliminated.
従来 SOI 'MOSの欠点は単結晶 S i膜 3が支持基板 1から絶縁されている ため、 ドレイン強電界等により発生した少数キャリア力単結晶 S i膜 3内に過渡 的に蓄積され、 これにより閾電圧値が変動する、 いわゆる、 基板浮遊効果にある。 この効果は少数キャリアの単結晶 S i膜 3内蓄積による電位上昇に伴い、 ソース 力、らの多数キャリア流入が生じる寄生バイポーラ効果でもある。 n導電型 SOI •MOS (以下 nMOSと略記) では正孔が蓄積され、 閾電圧値は負値方向に変 動し、 電流電圧特性に特異なこぶが観測されたり、 オフ状態における漏洩電流の 増大、 更にはソース · ドレイン間耐圧の低下をもたらす。 基板浮遊効果は微小電 流差の検出を要する差動増幅器やアナログ回路にとっては致命的な欠点となる恐 れがある。 The disadvantage of the conventional SOI'MOS is that the single-crystal Si film 3 is insulated from the supporting substrate 1, and thus is transiently accumulated in the minority carrier force single-crystal Si film 3 generated by a strong electric field at the drain. The threshold voltage fluctuates, which is the so-called substrate floating effect. This effect is also a parasitic bipolar effect in which minority carriers are accumulated in the single-crystal Si film 3 and the potential rises due to an increase in the source force. In an n-conductivity SOI • MOS (hereinafter abbreviated as nMOS), holes are accumulated, the threshold voltage fluctuates in the negative direction, a specific bump is observed in the current-voltage characteristics, and the leakage current increases in the off state. Further, the breakdown voltage between the source and the drain is reduced. The substrate floating effect may be a fatal drawback for differential amplifiers and analog circuits that require detection of small current differences. There is.
図 1の S 0 I · MO Sは基板浮遊効果を解消するために提案された構造で、 ソ ース拡散層 9内にゲルマ二ユウム (以下 Ge) をイオン注入することにより G e 成分比で 10%程度の S i Ge混晶 16を構成している。 S 0 I · MOS in Fig. 1 is a structure proposed to eliminate the substrate floating effect, and the germanium (Ge) ion implantation in the source diffusion layer 9 is performed with the Ge component ratio. About 10% of SiGe mixed crystal 16 is formed.
図 2は図 1の SOI 'MOSでドレイン電圧 Vdsが印加された状態における チャネルに沿ったエネルギバンド図である。 E f nは擬フヱルミ一準位、 E iは 真性フヱルミ一準位である。 S i Ge混晶 16の導入によりバンドギャップは約 0. l eV狭まり、 ソースにおける価電子帯 Evは破線で示されるように構成さ れる。 ソース近傍における正孔に対する拡散電位差が低減される。 これによりド レイン近傍で発生し、 単結晶 S i膜 3に注入された正孔は容易にソース内に拡散 し、 消滅する。 伝導帯 Ecは S i Ge混晶により影響を受けず、 多数キャリアで ある電子の振舞には悪影響は無 、とされている。 FIG. 2 is an energy band diagram along the channel in a state where the drain voltage Vds is applied in the SOI ′ MOS of FIG. E f n is the pseudo-Furmi level, and E i is the intrinsic Fermi level. The introduction of Si Ge mixed crystal 16 narrows the band gap by about 0.1 leV, and the valence band Ev at the source is configured as shown by the broken line. The diffusion potential difference for holes near the source is reduced. As a result, the holes generated near the drain and injected into the single crystal Si film 3 easily diffuse into the source and disappear. The conduction band Ec is not affected by the Si Ge mixed crystal, and it is said that the behavior of electrons as majority carriers has no adverse effect.
発明の開示 Disclosure of the invention
本発明の課題は図 1で示される構造が S 0 I · MO Sの基板浮遊効果解消に有 効であるのは nMOSに限られる事実に鑑み、 p導電型 MOS (以下 pMOSと 略記) 、 及び相補型 MOS (以下 CMOSと略記) にも適用可能な基板浮遊効果 解消構造を提供することにある。 即ち、 S i Ge混晶の存在は pMOSでもソー ス拡散層近傍における価電子帯の拡散電位差を低減し、 伝導帯の拡散電位差が保 存される。 この状況は多数キャリアである正孔はパンチスルー現象を起こし、 ゲ —ト電位で制御できなくなること、 逆に単結晶 S i膜 3に注入された少数キヤリ ァの電子はソース内に注入できず基板浮遊効果を解消できないことを意味する。 本発明の他の課題は S i G e混晶形成に必須の G eイオン注入工程の問題点を 解消することにある。 即ち、 G eイオンの注入には通常 G eH4 (ゲルマン) を イオン源として用いるが GeH4 は極めて分解し易い物質であり、 注入イオンを 安定に供給することが難しく、 イオン電流の制御、 従ってイオン注入条件の制御 に困難を伴うこと、 及びィォン化室を汚染するため他のィォン注入との共用が難 しく G e専用のイオン注入装置の導入が不可欠等の問題がある。 An object of the present invention is to consider the fact that the structure shown in FIG. 1 is effective only for nMOS in eliminating the substrate floating effect of S 0 I · MOS, and therefore, p-type MOS (hereinafter abbreviated as pMOS), and An object of the present invention is to provide a substrate floating effect eliminating structure applicable to complementary MOS (hereinafter abbreviated as CMOS). In other words, the presence of the SiGe mixed crystal reduces the valence band diffusion potential difference near the source diffusion layer even in the pMOS and preserves the conduction band diffusion potential difference. In this situation, holes, which are majority carriers, cause a punch-through phenomenon and cannot be controlled by the gate potential. Conversely, minority carrier electrons injected into the single crystal Si film 3 cannot be injected into the source. This means that the substrate floating effect cannot be eliminated. Another object of the present invention is to solve the problem of the Ge ion implantation step which is essential for the formation of Si Ge mixed crystals. In other words, GeH 4 (germane) is usually used as an ion source for Ge ion implantation, but GeH 4 is a very easily decomposable substance, and it is difficult to supply implanted ions stably. There are problems that it is difficult to control the ion implantation conditions, and it is difficult to share the ion implantation chamber with other ion implantations due to the contamination of the ionization chamber, and it is necessary to introduce a dedicated ion implantation device for Ge.
本発明の目的は既存の共用可能な半導体製造装置のみで製造可能で、 従つて廉 価な製造方法により基板浮遊効果解消構造を提供することにある。 本発明の動作原理を要約すれば、 基板浮遊効果がドレインと基板をエミッタ及 びベースとする寄生バイポーラ効果に基づくため、 上記バイポーラの電流増幅率 より大きな電流増幅率を有し、 基板をエミッタとする第二の寄生バイポーラをソ —ス拡散層に形成することにより基板への少数キヤリァ蓄積を解消することであ る。 以下に上記課題を解消するための手段を更に詳細に説明するが理解を容易に するために nMOSを例として説明する。 SUMMARY OF THE INVENTION An object of the present invention is to provide a substrate floating effect eliminating structure which can be manufactured only by an existing sharable semiconductor manufacturing apparatus, and is therefore manufactured at a low cost. To summarize the operating principle of the present invention, since the substrate floating effect is based on the parasitic bipolar effect using the drain and the substrate as the emitter and the base, the substrate has the current amplification factor larger than the bipolar current amplification factor, and the substrate is connected to the emitter. The formation of a second parasitic bipolar in the source diffusion layer eliminates the accumulation of minority carriers on the substrate. Hereinafter, the means for solving the above-mentioned problem will be described in more detail. However, in order to facilitate understanding, description will be made using an nMOS as an example.
図 3は本半導体装置の一実施例におけるソース拡散層の深さ方向不純物分布図 である。 ソ一ス拡散層底部は埋込酸化膜により外部から分離された構成を仮定し ている。 n型ソース内に高濃度 p型拡散層を形成する。 形成方法はソース拡散層 底部に達するように比較的低濃度の第一の n型不純物層をリン (P) のイオン注 入とその後の熱処理によりまず形成する。 この状態より高濃度ボロン (B) 、 及 び更に高濃度の砒素 (As) のイオン注入を施して表面より n p n構成の不純物 分布を形成する。 ドレイン領域にもソースと同一の不純物分布が形成される。 図 4 Aは上記不純物分布を有する nMOSに関するエネルギ帯図であり、 ドレ イン電圧 Vd sが印加されている。 図で、 価電子帯 Ev、 及び伝導帯 E cとして 実線で表示されているのは SO I表面のチャネル形成領域の、 破線で示されるの は外部から分離された p型拡散層近傍におけるエネルギ帯図である。 チャネル領 域における電子、 即ち多数キヤリアは実線に示されるエネルギ帯図に従って伝導 し、 従来と変わらない。 FIG. 3 is an impurity distribution diagram in the depth direction of the source diffusion layer in one embodiment of the present semiconductor device. It is assumed that the bottom of the source diffusion layer is separated from the outside by a buried oxide film. A high concentration p-type diffusion layer is formed in an n-type source. First, a relatively low-concentration first n-type impurity layer is formed by ion implantation of phosphorus (P) and subsequent heat treatment to reach the bottom of the source diffusion layer. From this state, ions of high concentration boron (B) and further high concentration of arsenic (As) are implanted to form an npn impurity distribution from the surface. The same impurity distribution is formed in the drain region as in the source. FIG. 4A is an energy band diagram for the nMOS having the above impurity distribution, in which the drain voltage Vds is applied. In the figure, the solid lines indicate the valence band Ev and the conduction band Ec in the channel formation region on the SOI surface, and the broken lines indicate the energy bands near the p-type diffusion layer separated from the outside. FIG. Electrons in the channel region, that is, many carriers, conduct according to the energy band diagram shown by the solid line, and remain the same as before.
ドレイン強電界によりドレイン近傍で発生し、 チャネル下部の p型単結晶 S i 膜内に蓄積される正孔はチャネル下部の低濃度 P基板をエミッタ P— E , ソース 内に形成された p型拡散領域をコレクタ P+ c , 両領域に挟まれた n型拡散領域 をベース n B とする p n pトランジスタにより伝導が制御される。 Holes generated near the drain due to the strong electric field at the drain and accumulated in the p-type single-crystal Si film under the channel pass through the low-concentration P substrate under the channel through the p-type diffusion formed in the emitter P-E and source. The conduction is controlled by a pnp transistor whose region is the collector P + c and whose n-type diffusion region is the base nB between the two regions.
図 4 Bは p n pトランジスタの動作原理を説明するためのソース近傍を示す等 価回路図である。 図で、 ソース表面の高濃度 n型拡散層 n+ s とコレクタ p+ c 間は (1) ダイオードで表わせる場合と、 (2) 破線のように短絡してソース電 位に固定する場合の両構成が可能である。 コレクタ電流密度 Jは次式で表わされ 。 FIG. 4B is an equivalent circuit diagram showing the vicinity of the source for explaining the operation principle of the pnp transistor. In the figure, between the high-concentration n-type diffusion layer n + s on the source surface and the collector p + c, (1) the case where it can be represented by a diode, and (2) the case where it is fixed to the source potential by short-circuiting as shown by the broken line Both configurations are possible. The collector current density J is expressed by the following equation.
(q · D„ · n, 2 /N · WB ) - exp(q · V1¾K/kT) … (数 1 ) ここで NB - WB はガンメル数と称されるベース領域内キャリア数であり通常 のトランジスタでは 1 0 1 2程度の値である。 qは電子の電荷量、 D n は拡散係数、 n , は真性密度、 NB はベース内のァクセプタ不純物密度、 WB はべ一ス幅、 VB Eはべ一ス ·ェミッタ間電圧、 k Tは熱エネルギである。 ェミッタである低濃 度 p基板の電位は蓄積正孔密度、 埋込酸化膜容量 C B。x 、 及び V d sで決定され る。 VBEは n + s · P + c 間がダイオードで表せる場合、 即ち、 正常な p n接合 が形成されている場合、 n + s - p + c 間の拡散電位差のために十分大きな値を とることができない。 (q · D „· n, 2 / N · W B ) -exp (q · V 1¾K / kT)… ( Equation 1) Here N B - W B is 1 0 1 2 value of about the normal transistor is referred base region number of carriers and Gummel number. q is the electron charge amount, D n are diffusion coefficients, n, is the intrinsic density, N B is Akuseputa impurity density in the base, W B Habe Ichisu width, V BE Habe Ichisu-Emitta voltage, k T Is thermal energy. Low concentration p is the potential of the substrate accumulation hole density is Emitta buried oxide film capacitance C B. It is determined by x, and V ds. V BE should be large enough for the diffusion potential difference between n + s -p + c when n + s · P + c can be represented by a diode, that is, when a normal pn junction is formed. Can not.
従ってエミッタからコレクタに向けて注入される正孔はベースである低濃度型 拡散層に蓄積され、 ベース ' コレクタ間空乏層で速やかに再結合して消滅する。 n + s · c 間が短絡されている場合、 コレクタは接地電位であり VB Eは十分 に大きな値を取り得て正孔はソース電極へ流出する。 n + s と p c 間の短絡は 高濃度 A sイオン注入により p n接合面への結晶欠陥の形成、 又は p型高濃度層 上への n型不純物添加多結晶 S i直接形成により実現できる。 即ち、 チャネル下 部の p型単結晶 S i膜内に蓄積される正孔はソース拡散層内の p + n + の形成に よりソース拡散層内へ引抜くことができる。 従って、 本発明により S O I · ΜΟ Sの基板浮遊効果を解消することができる。 Therefore, holes injected from the emitter toward the collector are accumulated in the low-concentration diffusion layer, which is the base, and quickly recombine and disappear in the base-collector depletion layer. When n + s · c is short-circuited, the collector is at ground potential and V BE can take a sufficiently large value, and holes flow out to the source electrode. The short circuit between n + s and pc can be realized by the formation of crystal defects on the pn junction surface by high-concentration As ion implantation or by the direct formation of n-type doped polycrystalline Si on the p-type high concentration layer. That is, holes accumulated in the p-type single-crystal Si film below the channel can be extracted into the source diffusion layer by forming p + n + in the source diffusion layer. Therefore, according to the present invention, the substrate floating effect of SOIIS can be eliminated.
この構造で、 ソース内 p型高濃度層は低濃度 n型拡散層でドレインと相対する 側面と底面を、 表面部は高濃度 n型拡散層で覆う構造が望ましい。 これは側面に おける基板内蓄積正孔のソース内への引抜き効率を上げるため、 側面でのガンメ ル数を小さく設定し、 表面部でのガンメル数を大きく設定することにより表面チ ャネル部での基板電位の上昇に基づくソースから基板への電子放出を抑える効果 が生じる。 ソース内 p型高濃度層側面の低濃度 n型拡散層幅も同様の理由により 狭いことが望ましい。 In this structure, it is desirable that the p-type high-concentration layer in the source is a low-concentration n-type diffusion layer, which covers the side and bottom surfaces facing the drain, and the surface is covered with the high-concentration n-type diffusion layer. In order to increase the efficiency of drawing holes accumulated in the substrate into the source on the side surface, the number of gummels on the side surface is set small, and the number of gummels on the front surface is increased. This has the effect of suppressing electron emission from the source to the substrate due to the rise in the substrate potential. It is also desirable that the width of the low-concentration n-type diffusion layer on the side of the p-type high-concentration layer in the source be narrow for the same reason.
本発明の手法はゲ一ト電極又はゲート側壁絶縁膜をイオン注入マスクとして実 施するためドレイン拡散層内にも Ρ τ η +接合が形成される。 ドレイン電圧は ρ 型拡散層にも印加されるが周囲を η型拡散層に囲われ、 同様にドレイン電圧が印 加されるのでドレインからの正孔放出は生じない。 従って、 本発明に基づく S O I · MO Sは回路動作条件により ドレインとソースを入替えて両方向動作させる、 いわゆる、 トランスファ MO Sに対しても有効である。 更に p型拡散層の底面部 にも n型拡散層を形成し、 基板から p型拡散層を隔離する構成によりドレイン · 基板間容量に P型拡散層の存在は影響されない。 即ち、 従来 S O I * MO Sの最 大特徴である厚い埋込酸化膜による寄生容量低減効果は維持される。 なお、 p M 0 Sでは上記した不純物伝導型を入替えて実施すれば同様に基板浮遊効果が解消 される。 即ち、 本発明は S O I · CMO Sの基板浮遊効果の解消に有効である。 本発明にに基づけば従来 S O I · MO Sの最大の欠点であった基板浮遊効果に 基づく閾電圧の過渡的値変動、 異常電流電圧特性、 更にはソース · ドレイン耐圧 低下等の問題を解消することができる。 これにより低寄生容量、 製造工程数低減 等 S O I · MO Sの本来の特徵を活かした高速動作可能な半導体装置を専用のィ ォン注入装置等新規の半導体装置製造装置の導入無しに廉価に提供することがで さる。 Approach of the present invention also [rho tau eta + junction in real Hodokosuru for drain diffusion layer of the gate one gate electrode or a gate sidewall insulating film as an ion implantation mask is formed. The drain voltage is applied to the ρ-type diffusion layer, but the periphery is surrounded by the η-type diffusion layer. Similarly, the drain voltage is applied, so that no hole emission from the drain occurs. Therefore, the SOI / MOS according to the present invention performs the bidirectional operation by exchanging the drain and the source according to the circuit operating conditions. It is also effective for so-called transfer MOS. In addition, an n-type diffusion layer is also formed at the bottom of the p-type diffusion layer to isolate the p-type diffusion layer from the substrate, so that the presence of the P-type diffusion layer is not affected by the drain-substrate capacitance. That is, the effect of reducing the parasitic capacitance by the thick buried oxide film, which is the largest feature of the conventional SOI * MOS, is maintained. In addition, in the case of pM0S, the substrate floating effect can be similarly eliminated by replacing the impurity conductivity type described above. That is, the present invention is effective in eliminating the substrate floating effect of SOI · CMOS. Based on the present invention, it is necessary to eliminate the problems of transient fluctuation of threshold voltage, abnormal current-voltage characteristics, and source / drain withstand voltage drop due to the substrate floating effect, which are the biggest drawbacks of the conventional SOI and MOSS. Can be. As a result, semiconductor devices capable of high-speed operation utilizing the inherent features of SOI and MOS, such as low parasitic capacitance and reduction in the number of manufacturing steps, can be provided at low cost without introducing new semiconductor device manufacturing equipment such as dedicated ion implantation equipment. You can do it.
また、 図 5 Aは本半導体装置の他の実施例におけるソース拡散層近傍の拡大断 面図、 図 5 Bは本半導体装置の等価回路図であり、 図 6 Aはドレイン電圧が接地 電圧におけるソース近傍の二次元エネルギーバンド図、 図 6 Bはソース領域の高 濃度 P領域と高濃度 n領域のでエネルギーバンドを重合わせて示した図である。 ソ一ス領域底部は埋込酸化膜により外部から分離された構成を仮定している。 図 5 Aにおいて、 ソース領域に形成する第二の寄生 p n pバイポーラは p型基板領 域をェミッタ p -、 n型低濃度ソース拡散層をベース 、 ソース内の p型高濃 度領域をコレクタ p +として作用させる。 ベースは低濃度ソース拡散層による抵 抗成分を介して接地電位にある高濃度ソース領域に接続される。 本半導体装置の 第二の寄生バイポーラにおいてはベース電位は固定されている。 FIG. 5A is an enlarged cross-sectional view of the vicinity of the source diffusion layer in another embodiment of the present semiconductor device, FIG. 5B is an equivalent circuit diagram of the present semiconductor device, and FIG. A two-dimensional energy band diagram in the vicinity, and FIG. 6B is a diagram showing the energy bands of the high-concentration P region and the high-concentration n region in the source region in an overlapping manner. It is assumed that the bottom of the source region is separated from the outside by a buried oxide film. In FIG. 5A, the second parasitic pnp bipolar formed in the source region is based on the emitter p- in the p-type substrate region, the base is based on the n-type low-concentration source diffusion layer, and the collector p + is in the p-type high-concentration region in the source. Act as The base is connected to the high-concentration source region at the ground potential through a resistance component of the low-concentration source diffusion layer. In the second parasitic bipolar of the present semiconductor device, the base potential is fixed.
正孔の注入効率向上は図 6 Bに示すごとく、 十分に低不純物のベース濃度の条 件の基にエミッタ ·コレクタ間を接近させて正孔に対する拡散電位差を低減する ことにより可能となる。 上記寄生 p n pバイポーラを効率良く動作させるために はコレクタに注入された正孔を速やかに消滅させる機構が要求される。 上記機構 として、 本半導体装置においては結晶欠陥等に基づく再結合中心を利用する。 再 結合中心による正孔消滅機構は不純物濃度にも依存することが良く知られており、 P型不純物濃度は可能な限り高濃度であることが望ましい。 再結合中心による正 孔消滅に要する電子はソース表面の n型高濃度領域 n +と p型高濃度コレクタ p +間が電子注入に対して順方向の関係にあり、 十分に供給される。 As shown in Fig. 6B, the hole injection efficiency can be improved by reducing the diffusion potential difference between holes by bringing the emitter and collector closer together under the condition of sufficiently low impurity base concentration. In order for the parasitic pnp bipolar to operate efficiently, a mechanism is needed to quickly eliminate holes injected into the collector. As the above mechanism, the present semiconductor device utilizes a recombination center based on a crystal defect or the like. It is well known that the hole annihilation mechanism due to the recombination center also depends on the impurity concentration, and it is desirable that the P-type impurity concentration be as high as possible. Positive by recombination center The electrons required for hole annihilation are sufficiently supplied because the n-type high-concentration region n + on the source surface and the p-type high-concentration collector p + have a forward relationship with respect to electron injection.
即ち、 ドレイン強電界により発生し、 P型基板領域内に蓄積された正孔はソ一 ス領域内に構成されたバイポーラによりソース領域に引抜かれ、 ソース電流に変 換されることにより基板浮遊効果は解消される。 上記のソース電流密度 Jは次式 で表わされる。 In other words, the holes generated by the strong electric field at the drain and accumulated in the P-type substrate region are drawn out to the source region by the bipolar formed in the source region, and are converted into the source current. Is eliminated. The above source current density J is expressed by the following equation.
J % (q · D„ · n, 2 /NB · Wb ) - exp(q · VBE/kT) ここで qは電子の電荷量、 Dn は拡散係数、 n, は真性密度、 NB はベース内の ァクセプタ不純物密度、 WB はべ一ス幅、 VBEはべ一ス 'ェミッタ間電圧、 kT は熱エネルギーである。 NB · Wb はガンメル数と称されるベース領域内キヤリ ァ数であり通常のトランジスタでは 1 0 '2程度の値である。 ェミッタである低濃 度 P基板の電位は蓄積正孔密度、 埋込酸化膜容量 CBOX 、 及びドレイン電圧J% (q · D "· n, 2 / N B · W b) - exp (q · V BE / kT) where q is the electron charge amount, D n are diffusion coefficients, n, is the intrinsic density, N B is the acceptor impurity density in the base, W B is the base width, V BE is the base-to-emitter voltage, kT is the thermal energy, and N B · W b is the base region called the Gummel number. in is normal transistor is Kiyari number § 1 0 '2 a value of about. potential of the low concentration P substrate which is a Emitta accumulation hole density, buried oxide film capacitance CBOX, and the drain voltage
Vd sで決定される。 Determined by Vds.
本半導体装置の形成方法としてはゲ一ト電極をマスクとして、 S 0 I層直下の 厚いシリコン酸化膜に達するごとく比較的低濃度の第一の n型不純物層をリン (P) のィォン注入とその後の熱処理によりまず低濃度 n型拡散層を形成する。 この状態よりゲート側壁絶縁膜の形成し、 ゲート電極及びゲート側壁絶縁膜をマ スクとする高濃度ボロン (B) のイオン注入を施し、 SO I層底面部を非晶質化 させる。 The method of forming the present semiconductor device is to use a gate electrode as a mask and implant a relatively low-concentration first n-type impurity layer to reach a thick silicon oxide film immediately below the SOI layer by ion implantation of phosphorus (P). A low-concentration n-type diffusion layer is first formed by a subsequent heat treatment. From this state, a gate sidewall insulating film is formed, and high-concentration boron (B) ions are implanted using the gate electrode and the gate sidewall insulating film as a mask to make the bottom portion of the SOI layer amorphous.
上記非晶質はその後の短時間高温熱処理においても底部が酸化膜であるため再 結晶化熱処理による単結晶化はゲ一ト直下の単結晶 SO I層側面部を除いて行な われず、 多結晶化が進行するだけである。 上記多結晶、 又は非晶質性は熱処理条 件により制御可能であり、 再結合中心特性を制御する。 上記の p型高濃度領域の 形成において、 その側面に残置させ、 ベース領域として作用するソース低濃度 n 型拡散層幅はゲート側壁絶縁膜の膜厚により制御する。 SO I層表面に構成する n型高濃度領域はィォン注入法による形成か、 又は上記 p型高濃度領域表面の一 定厚さを選択除去した後、 堆積法による半導体膜を残置して形成しても良い。 後 者の場合、 より急峻な高不純物分布が実現でき、 より再結合特性に優れた、 従つ てより基板浮遊効果解消に有効な不純物分布が可能となる。 本発明の手法はゲ一ト電極、 及びゲート側壁絶縁膜をマスクとして実施するた め、 ドレイン領域にも同様の構造が形成される。 ドレインにおいてはドレイン電 圧が n型高濃度領域に印加されるが、 ドレインにおける n型高濃度領域と底部の P型高濃度領域の関係は正孔注入に関して逆方向特性の関係にあるため接合リー ク電流の増加は無視でき、 NOSトランジスタとしての動作に何ら支障は生じな い。 Since the bottom of the amorphous layer is an oxide film even in a short-time high-temperature heat treatment, single-crystallization by recrystallization heat treatment is not performed except for the side surface of the single-crystal SOI layer immediately below the gate. Only crystallization proceeds. The above polycrystalline or amorphous properties can be controlled by heat treatment conditions, and control the recombination center characteristics. In the formation of the p-type high-concentration region, the width of the source low-concentration n-type diffusion layer left on the side surface and serving as a base region is controlled by the thickness of the gate sidewall insulating film. The n-type high-concentration region formed on the surface of the SOI layer is formed by ion implantation or by selectively removing a certain thickness of the surface of the p-type high-concentration region, and then forming the semiconductor film by deposition to leave. May be. In the latter case, a steeper high impurity distribution can be realized, and an impurity distribution having more excellent recombination characteristics and thus more effective in eliminating the substrate floating effect can be realized. Since the method of the present invention is performed using the gate electrode and the gate sidewall insulating film as a mask, a similar structure is formed in the drain region. At the drain, the drain voltage is applied to the n-type high-concentration region, but since the relationship between the n-type high-concentration region and the bottom P-type high-concentration region in the drain has a reverse characteristic with respect to hole injection, the junction leakage The increase in the blocking current is negligible and does not hinder the operation as a NOS transistor.
なお、 上記の説明において、 p導電型と n導電型を置換えれば本半導体装置に よる基板浮遊効果解消が p M 0 Sに対しても有効であることは明らかである。 図 5A, 5 Bに示した構造、 即ち p型低濃度 SO I基板ェミッタ、 n型低濃度 拡散層をべ一ス、 p型高濃度領域をコレクタとし、 該コレクタ領域上に n型高濃 度領域を有する構造において、 n型高濃度領域を接地電位とし、 p型低濃度 SO I基板エミッ夕に正電圧を印加した時のエミッタを流れる正孔電流をベース幅パ ラメータとして数値解析により求めた結果を図 7に示す。 上記解析において、 厚 さ 50mn、 4 X 1017/cm3 なる一様濃度分布の p型 S 0 I層をエミッタとし、 SO I層底部の埋込絶縁膜に接するごとく構成された厚さ 10nm、 接合深さ 10 nm、 最大不純物濃度 2 X I 01 cm3 の p型コレクタ、 該コレクタ上に構成され、 表面で最大不純物濃度 2 X 102 Vcm3 を有し、 接合深さ 4 Onmの n型高濃度領 域、 n型ベースは 5 X I 0 'Vcm3 なる最大濃度のガウス分布を有し、 コレクタ、 及び n型高濃度領域とェミッタ間を分離するごとく構成している。 なお、 再結合 時間はコレクタ領域で 1/10'°秒、 他の領域では 1Z104 秒を仮定している。 再結合時間 1ノ1 01(1秒なる値は高不純物濃度多結晶 S i膜では通常観測される 値である。 In the above description, if the p-conductivity type and the n-conductivity type are replaced, it is clear that the substrate floating effect by the present semiconductor device is effective for pM0S. The structure shown in FIGS. 5A and 5B, ie, a p-type low-concentration SOI substrate emitter, an n-type low-concentration diffusion layer as a base, a p-type high-concentration region as a collector, and an n-type high-concentration In a structure with a region, the n-type high-concentration region was set to the ground potential, and the hole current flowing through the emitter when a positive voltage was applied to the p-type low-concentration SOI substrate emitter was determined by numerical analysis using the base width parameter. Fig. 7 shows the results. In the above analysis, the p-type S 0 I layer having a uniform concentration distribution of 50 mn and 4 × 10 17 / cm 3 was used as the emitter, and the thickness was 10 nm, which was configured to be in contact with the buried insulating film at the bottom of the SO I layer. N-type with a junction depth of 10 nm and a maximum impurity concentration of 2 XI 0 1 cm 3, a p-type collector formed on the collector and having a maximum impurity concentration of 2 X 10 2 Vcm 3 on the surface and a junction depth of 4 Onm The high-concentration region, n-type base has a Gaussian distribution with a maximum concentration of 5 XI 0 'Vcm 3 , and is configured so as to separate the collector and the n-type high-concentration region from the emitter. Incidentally, the recombination time 1/10 '° sec collector region, in other regions assumes 1Z10 4 seconds. Recombination time 1 1 0 1 (The value of 1 second is a value normally observed in a high impurity concentration polycrystalline Si film.
図 7には参考のために p型コレクタ領域を有しなレ、通常ソ一ス構造の電子電流、 及び正孔電流も細線で示した。 エミッタに流入する電子電流に関しては本半導体 装置のものも従来構造のものと一致し、 差は見られず、 印加電圧に対して指数関 数に比例した増加傾向を示す。 正孔電流に関しては閾電圧までエミッタカ、ら流出 する (負の正孔電流) 力 閾電圧以上から正の正孔電流となり、 印加電圧に対し て指数関数に比例して増加する。 通常構造においても正孔電流は電子電流より二 桁程度大きな値を示す。 本半導体装置構造の場合、 ベース幅が 4 Onm (下の条件 で閾電圧値近傍での正孔電流は従来構造に比べて三桁程度大きく、 閾電圧も 0 .In FIG. 7, for reference, the electron current and the hole current having no p-type collector region and the normal source structure are also shown by thin lines. Regarding the electron current flowing into the emitter, that of the present semiconductor device also matches that of the conventional structure, there is no difference, and it shows an increasing trend in proportion to the applied voltage with the exponential function. With regard to the hole current, the emitter current flows out from the emitter up to the threshold voltage (negative hole current). From the threshold voltage onwards, the hole current becomes a positive hole current, which increases in proportion to the applied voltage in an exponential function. Even in the normal structure, the hole current shows a value about two orders of magnitude larger than the electron current. In the case of this semiconductor device structure, the base width is 4 The hole current near the threshold voltage is about three orders of magnitude higher than that of the conventional structure, and the threshold voltage is also 0.
2 Vの低下が見られる。 上記の意味するところは本半導体装置構造では正孔電流 に対するソース拡散電位が従来構造に比べて 0 . 2 e V低下されたことを示して いる。 上記の値は公知のソース内 S i G e混晶形成によるソース拡散電位の低下 値 0 . 1 e Vの 2倍であり、 基板浮遊効果解消が更に改善されることを示してい る。 上記閾電圧の低下はベース幅の増加と共に解消される方向に向かうが、 0 . 1 〃mのべ一ス幅でも従来構造に比べて 0 . 0 4 Vの低下が観測される。 即ち、 本半導体装置構造において、 ベース幅は 0 . 1 m以下であることが望ましい。 図 7の結果はェミッタに正の電圧を印加する順方向特性に関するものである。 n型高濃度領域に正電圧を印加する逆方向特性に関しても数値解析を実施したが、 3 Vまでの解析結果では電流は計算誤差範囲内の 1 Z 1 0 1 3から 1ノ1 0 1 5 Aの 間の値で、 通常構造のものと差違が見られなかった。 この結果はソース内と同様 の構造をドレイン内に構成してもリーク電流の増加等の問題力生じないことを示 している。 A 2 V drop is seen. The above-mentioned meaning indicates that the source diffusion potential with respect to the hole current is reduced by 0.2 eV in the present semiconductor device structure as compared with the conventional structure. The above value is twice the reduced value of the source diffusion potential of 0.1 eV due to the known SiGe mixed crystal formation in the source, indicating that the elimination of the substrate floating effect is further improved. Although the decrease in the threshold voltage tends to be eliminated with an increase in the base width, a decrease of 0.04 V is observed even with a base width of 0.1 μm as compared with the conventional structure. That is, in the present semiconductor device structure, the base width is desirably 0.1 m or less. The results in FIG. 7 relate to the forward characteristics when a positive voltage is applied to the emitter. Numerical analysis was also performed on the reverse characteristics in which a positive voltage was applied to the n-type high-concentration region, but in the analysis results up to 3 V, the current was within the calculation error range from 1 Z 10 13 to 1 10 15 There was no difference between the values of A and those of the normal structure. This result indicates that even if a structure similar to that in the source is formed in the drain, no problem such as an increase in leakage current occurs.
本発明に基づく半導体装置はソース、 ドレインに関して対称な構造を有してお り、 回路動作条件によりドレインとソースを入替えて両方向動作させる所謂トラ ンスファ MO S等に対しても有効である。 更にソース ' ドレイン内 p型高濃度領 域の底面部にも n型拡散層を形成し、 基板から p型拡散層を隔離する構成、 又は P型高濃度領域を所望個所のみに限定するごとき構成にすることにより ドレイン ·基板間容量に p型高濃度領域の存在を無視できる程度に低減できる。 The semiconductor device according to the present invention has a symmetric structure with respect to the source and the drain, and is also effective for a so-called transfer MOS or the like in which the drain and the source are switched according to the circuit operating conditions to operate in both directions. Furthermore, an n-type diffusion layer is also formed at the bottom of the p-type high-concentration region in the source and drain to isolate the p-type diffusion layer from the substrate, or a configuration that limits the P-type high-concentration region to only the desired locations. By doing so, the presence of a p-type high concentration region in the drain-substrate capacitance can be reduced to a negligible level.
さらに、 本発明のもう一つ別の実施例の動作原理を要約する。 理解を容易にす る為に n MO Sを例として説明する。 本発明においてはドレイン近傍で発生し、 基板内に蓄積される正孔を速やかにソース拡散層内に注入、 消滅させる。 上記手 段として、 チャネル直下の P型基板領域とソース拡散眉間に形成される拡散電位 差が小さい領域、 即ち十分に低濃度の n型拡散層領域をソース高濃度拡散層に隣 接して設ける。 更に該 n型低濃度拡散層領域内に正孔に対して再結合中心として 作用する領域を設け、 n型低濃度拡散層領域内に注入された正孔を消滅させる。 正孔消滅に要する電子はソース表面の n型高濃度拡散層領域から供給される。 再 結合中心による正孔消滅機構として本発明においては結晶欠陥等に基づく再結合 中心を利用する。 Further, the operating principle of another embodiment of the present invention will be summarized. In order to facilitate understanding, n MOS will be described as an example. In the present invention, holes generated near the drain and accumulated in the substrate are quickly injected into the source diffusion layer and annihilated. As a method described above, a region having a small diffusion potential difference formed between the P-type substrate region immediately below the channel and the source diffusion region, that is, a sufficiently low concentration n-type diffusion layer region is provided adjacent to the source high concentration diffusion layer. Further, a region acting as a recombination center for holes is provided in the n-type low-concentration diffusion layer region to eliminate holes injected into the n-type low-concentration diffusion layer region. The electrons required for hole annihilation are supplied from the n-type high concentration diffusion layer region on the source surface. In the present invention, the recombination based on crystal defects and the like is used as a hole annihilation mechanism by the recombination center. Take advantage of the center.
図 8は本半導体装置におけるソース拡散層近傍の拡大断面図であり、 図 9はド レイン電圧が接地電圧におけるソース近傍のエネルギーバンド図である。 ソース 領域底部は埋込酸化膜により外部から分離された構成を仮定している。 図 8にお いて、 高濃度ソース拡散層による従来ソース構造の場合のエネルギーバンドも破 線で示したが本発明構造において、 正孔に対する拡散電位差の低下は明らかであ る。 低濃度拡散層の不純物濃度は伝導型が変わらない範囲で低い方が良く 1015 /cm3 以上、 10'8Zcm3 以下が望ましい。 再結合中心として作用する結晶欠陥 の形成は n型低濃度拡散層における正孔に対する拡散電位差を増大させない元素 によるイオン注入を施し、 SO I層底面部を非晶質化させる。 上記非晶質はその 後の短時間高温熱処理においても底部が酸化膜であるため再結晶化熱処理による 単結晶化はゲー卜直下の単結晶 SO I層側面部を除いて行なわれず、 多結晶化が 進行するだけである。 上記多結晶、 又は非晶質性は熱処理条件により制御可能で あり、 再結合中心特性を制御できる。 FIG. 8 is an enlarged cross-sectional view near the source diffusion layer in the present semiconductor device, and FIG. 9 is an energy band diagram near the source when the drain voltage is the ground voltage. It is assumed that the bottom of the source region is separated from the outside by a buried oxide film. In FIG. 8, the energy band in the case of the conventional source structure using the high-concentration source diffusion layer is also shown by broken lines, but in the structure of the present invention, the decrease in the diffusion potential difference with respect to holes is apparent. The impurity concentration of the low concentration diffusion layer is preferably as low as possible without changing the conductivity type, and is preferably 10 15 / cm 3 or more and 10 ′ 8 Zcm 3 or less. The formation of crystal defects acting as recombination centers is performed by ion implantation with an element that does not increase the diffusion potential difference with respect to holes in the n-type low-concentration diffusion layer, thereby making the bottom surface of the SOI layer amorphous. In the above amorphous state, even in the short-time high-temperature heat treatment, single-crystallization by recrystallization heat treatment is not performed except for the side surface of the single-crystal SOI layer immediately below the gate because the bottom is an oxide film. Only progresses. The above polycrystalline or amorphous properties can be controlled by heat treatment conditions, and the recombination center characteristics can be controlled.
MO S型半導体装置の主流である CMO S半導体装置に本発明手法を適用する に当り、 nMOSと pMOSを各々個別に適用することは製造工程数の増加、 及 び良品歩留りの低下を招き、 製造コストの上昇につながる。 従って、 再結合中心 を形成するィォン注入源としては n型、 及び p型低濃度拡散層における少数キヤ リァに対する拡散電位差を増大させない元素を同一工程で pMO S及び nMO S の各ソース低濃度拡散領域にイオン注入させ、 再結合を形成することが望ましい。 上記観点から S i半導体による MOS半導体装置においてはイオン注入源として P、 B、 As、 Sb、 Gaのごとく容易に活性化して n、 又は p導電型を形成す る元素以外の元素であることが望ましい。 更に、 イオン注入により半導体基板を 非晶質化する元素であることが望ましく、 原子質量が 10以下の元素は好ましく ない。 S i半導体内において拡散係数が異常に大きく信頼性を損なう Na、 Kの ごときアルカリ金属、 Mgを含むアルカリ土類金属も好ましくない。 本発明にお いては半導体を構成する S i、 Ge、 C等の 14族元素、 F、 C 1等のハロゲン 元素、 Ne、 A r等の希ガス元素が好ましい。 特に廉価で、 供給も安定し、 ィォ ン化が容易で且つ安定な S i、 C、 Ne、 Ar、 C 1等の元素が最も好ましい。 本半導体装置の一形成方法としてはまずゲ一ト電極をマスクとして低濃度、 及 び高濃度のソース · ドレイン拡散層を形成する。 上記低濃度拡散層は S 0 I層直 下の厚 L、シリコン酸化膜に達するごとィォン注入とその後の熱処理により形成す る。 この状態よりゲート側壁絶縁膜の形成し、 ゲート電極及びゲート側壁絶縁膜 をマスクとする例えば S iのイオン注入を S O I層直下の厚いシリコン酸化膜に 達するごとく施して上記酸化膜界面領域の S O I層を非晶質化させる。 上記非晶 質領域はその後の熱処理によっても界面部は単結晶化されず、 微少粒界よりなる 多結晶化され、 再結合中心として機能する。 再結合中心領域とチャネル領域 S O I基板間間隔、 即ちソース低濃度 n型拡散層幅はゲート側壁絶縁膜の膜厚により 制御する。 上記ソース低濃度 n型拡散層幅は少数キャリアが容易に再結合中心領 域に達し、 消滅するために 1 0 0 ηπ0¾下であることが望ましい。 S O I層表面に 構成する n型高濃度ソース領域は上記再結合中心領域の形成後、 その表面部の一 定厚さを選択除去した後、 堆積法による半導体膜を残置して形成しても良 、。 本半導体装置の他の形成方法としては従来製造方法に基づき所望の拡散層構造 を有するソース · ドレイン領域を形成した後、 ソース電極との接続の為のコンタ ク卜穴形成において、 上記コンタクト穴から選択的に S O I層直下の厚いシリコ ン酸化膜に達する低濃度拡散層形成のイオン注入を施す。 しかる後、 上記コンタ クト穴寸法を一定幅縮小するごとく側壁膜を設け、 該側壁膜を注入マスクとする 再結合中心領域形成のイオン注入を S 0 I層直下の厚いシリコン酸化膜に達する ごとく施しても良い。 When applying the method of the present invention to CMOS semiconductor devices, which are the mainstream of MOS type semiconductor devices, individually applying nMOS and pMOS will increase the number of manufacturing steps and decrease the yield of non-defective products. This leads to higher costs. Therefore, as the ion implantation source that forms the recombination center, an element that does not increase the diffusion potential difference with respect to a small number of carriers in the n-type and p-type low-concentration diffusion layers is used in the same process for each of the pMOS and nMOS source low-concentration diffusion regions. It is desirable to perform ion implantation to form recombination. In view of the above, in a MOS semiconductor device using an Si semiconductor, an element other than an element that easily activates and forms n or p conductivity type, such as P, B, As, Sb, and Ga, may be used as an ion implantation source. desirable. Further, it is preferable that the element be an element that makes the semiconductor substrate amorphous by ion implantation, and an element having an atomic mass of 10 or less is not preferable. Alkali metals such as Na and K and alkaline earth metals including Mg, which have an unusually large diffusion coefficient in Si semiconductors and impair reliability, are also not preferred. In the present invention, a group 14 element such as Si, Ge and C, a halogen element such as F and C1, and a rare gas element such as Ne and Ar which constitute the semiconductor are preferable. In particular, elements such as Si, C, Ne, Ar, and C1, which are inexpensive, stable in supply, easy to ionize, and stable, are most preferable. As one method of forming the present semiconductor device, first, a low-concentration and high-concentration source / drain diffusion layer is formed using a gate electrode as a mask. The low-concentration diffusion layer is formed by ion implantation and heat treatment after reaching the silicon oxide film having a thickness L just below the SOI layer. From this state, a gate sidewall insulating film is formed, and ion implantation of, for example, Si using the gate electrode and the gate sidewall insulating film as a mask is performed so as to reach a thick silicon oxide film immediately below the SOI layer, thereby forming the SOI layer in the oxide film interface region. Is made amorphous. The interface of the amorphous region is not single-crystallized by the subsequent heat treatment, but is polycrystallized by fine grain boundaries, and functions as a recombination center. The center between the recombination center region and the channel region The distance between the SOI substrates, that is, the width of the source low concentration n-type diffusion layer is controlled by the thickness of the gate sidewall insulating film. The source low-concentration n-type diffusion layer width is preferably lower than 100 ηπ0 ° so that minority carriers can easily reach the recombination center region and disappear. The n-type high-concentration source region formed on the surface of the SOI layer may be formed by forming the above-described recombination center region, selectively removing a certain thickness of the surface portion, and leaving the semiconductor film by the deposition method. ,. Another method of forming the semiconductor device is to form a source / drain region having a desired diffusion layer structure based on a conventional manufacturing method, and then form a contact hole for connection with a source electrode. Selectively perform ion implantation to form a low concentration diffusion layer that reaches the thick silicon oxide film just below the SOI layer. Thereafter, a side wall film is provided so as to reduce the contact hole dimension by a certain width, and ion implantation for forming a recombination center region using the side wall film as an implantation mask is performed so as to reach a thick silicon oxide film immediately below the SOI layer. May be.
本発明の手法はゲート電極及びゲート側壁絶縁膜、 又はコンタクト穴をイオン 注入マスクとして実施するため、 ドレイン領域にも同様の構造が形成される。 ド レインにおいてはドレイン電圧が n型高濃度領域に印加されるが、 ドレインにお ける n型高濃度領域と底部の再結合中心領域との関係は正孔注入に関して逆方向 特性の関係にあるため接合リーク電流の増加は無視でき、 MO S トランジスタと しての動作に何ら支障は生じない。 Since the method of the present invention is performed using the gate electrode and the gate side wall insulating film or the contact hole as an ion implantation mask, a similar structure is formed in the drain region. In the drain, the drain voltage is applied to the n-type high-concentration region.However, the relationship between the n-type high-concentration region and the bottom recombination center region in the drain has a reverse characteristic with respect to hole injection. The increase in junction leakage current is negligible and does not hinder the operation as a MOS transistor.
図 8に示した構造において、 n型高濃度領域を接地電位とし、 p型低濃度 S O I基板に正電圧を印加した時に流れる正孔電流を再結合中心領域とチャネル領域 S 0 I基板間間隔のパラメータとして数値解析により求めた結果を図 6に示す。 上記解析には厚さ 30 Onm、 4 X 1 01 Vcm3 なる一様濃度分布の p型 S 0 I層 を用い、 再結合中心領域とチャネル領域 SO I基板間間隔、 即ちソース低濃度 n 型拡散層幅をパラメータとした。 ソース低濃度 n型拡散層は最大濃度 1 X 1 016 /cm3 のガウス分布を有し、 SO I層底部の埋込絶縁膜に接するごとく構成した c 再結合時間は再結合中心領域で 1 Z 1 01。秒、 他の領域では 1 Z 1 04 秒を仮定 している。 再結合時間 1Z10 '°秒なる値は多結晶 S i膜では通常観測される値 であ o。 In the structure shown in FIG. 8, the n-type high-concentration region is set to the ground potential, and the hole current flowing when a positive voltage is applied to the p-type low-concentration SOI substrate is reduced by the distance between the recombination center region and the channel region S 0 I substrate. Figure 6 shows the results obtained by numerical analysis as parameters. The thickness 30 onm in the above analysis, 4 X 1 0 1 Vcm 3 with p-type S 0 I layer of uniform density distribution comprising, a recombination center region and the channel region SO I substrate spacing, i.e. the source low concentration n-type The diffusion layer width was used as a parameter. The source low-concentration n-type diffusion layer has a Gaussian distribution with a maximum concentration of 1 × 10 16 / cm 3 and is configured to be in contact with the buried insulating film at the bottom of the SOI layer. Z 1 0 1. S assumes a 1 Z 1 0 4 seconds in other areas. The recombination time of 1Z10 '° sec is the value normally observed for polycrystalline Si films.
図 1 0には参考のために再結合中心領域を有しない通常ソース構造の電子電流、 及び正孔電流も細線で示した。 電子電流に関しては本半導体装置のものも従来構 造のものと一致し、 差は見られず、 印加電圧に対して指数関数に比例した増加傾 向を示す。 正孔電流に関しては印加電圧に対して指数関数に比例した電流が流れ 始める閾電圧が存在し、 閾電圧以下では印加電圧ゼロで電流がゼロになるごとく 振舞う。 通常構造にお 、ても正孔電流は電子電流より二桁程度大きな値を示す。 本半導体装置構造の場合、 ソース低濃度 n型拡散層幅が 4 Onm^下の条件で閾電 圧値近傍での正孔電流は従来構造に比べて三桁程度大きく、 閾電圧も 0. 2Vの 低下が見られる。 上記の意味するところは本半導体装置構造では正孔電流に対す るソース拡散電位が従来構造に比べて 0. 2 eV低下されたことを示している。 上記の値は公知のソース内 S i G e混晶形成によるソース拡散電位の低下値 0. 1 eVの 2倍であり、 基板浮遊効果解消が公知手法に比べて更に改善されること を示している。 上記閾電圧の低下はべ一ス幅の増加と共に解消される方向に向か うが、 0. 1 /zmのソース低濃度 n型拡散層幅でも従来構造に比べて 0. 04 V の低下が観測される。 即ち、 本半導体装置構造において、 ソース低濃度 n型拡散 層幅は 0. 1〃m以下であることが望ましい。 In FIG. 10, the electron current and the hole current of the normal source structure having no recombination center region are also shown by thin lines for reference. Regarding the electron current, that of the semiconductor device of the present invention also matches that of the conventional structure, and no difference is observed. The increase in the applied current is proportional to the exponential function. For the hole current, there is a threshold voltage at which a current proportional to the applied voltage starts to flow exponentially. Below the threshold voltage, the behavior is as if the current becomes zero at zero applied voltage. Even in the normal structure, the hole current shows a value about two orders of magnitude larger than the electron current. In the case of this semiconductor device structure, the hole current near the threshold voltage value is about three orders of magnitude higher than that of the conventional structure and the threshold voltage is 0.2 V when the source low-concentration n-type diffusion layer width is 4 Onm ^ or less. Is seen to decrease. The above means that the source diffusion potential with respect to the hole current is reduced by 0.2 eV in the present semiconductor device structure as compared with the conventional structure. The above value is twice the reduction value of the source diffusion potential of 0.1 eV due to the formation of the SiGe mixed crystal in the known source, indicating that the elimination of the substrate floating effect is further improved compared to the known method. I have. The above-mentioned decrease in threshold voltage tends to be eliminated as the base width increases, but even at a source low-concentration n-type diffusion layer width of 0.1 / zm, a decrease of 0.04 V compared to the conventional structure is observed. Observed. That is, in this semiconductor device structure, the width of the source low-concentration n-type diffusion layer is desirably 0.1 μm or less.
図 10の結果はチャネル領域 S 0 I基板に正の電圧を印加する順方向特性に関 するものである。 n型高濃度領域に正電圧を印加する逆方向特性に関しても数値 解析を実施したが、 3 Vまでの解析結果では電流は計算誤差範囲内の 1 / 1013 から 1Z1015Aの間の値で、 通常構造のものと差違が見られなかった。 この結 果はソース内と同様の構造をドレイン内に構成してもリーク電流の増加等の問題 が生じないことを示している。 本発明に基づく半導体装置はソース、 ドレインに関して対称な構造を有してお り、 回路動作条件によりドレインとソースを入替えて両方向動作させる所謂トラ ンスファ MO S等に対しても有効である。 更にソース · ドレイン内における再結 合中心領域の存在はドレイン ·基板間容量に何ら影響を与えるものでなく、 従来 S O I · MO Sの最大特徵である厚い埋込酸化膜による寄生容量低減効果は維持 される。 更に、 本発明は半導体装置の導電型に関係無く有効であり、 従って S O I · C MO Sの基板浮遊効果の解消に有効である。 The results in FIG. 10 relate to the forward characteristics when a positive voltage is applied to the channel region S 0 I substrate. Numerical analysis was also performed on the reverse characteristics in which a positive voltage was applied to the n-type high concentration region.However, in the analysis results up to 3 V, the current was a value between 1/10 13 and 1Z10 15 A within the calculation error range. However, there was no difference from the normal structure. This result indicates that even if the same structure as in the source is formed in the drain, no problem such as an increase in leakage current occurs. The semiconductor device according to the present invention has a symmetric structure with respect to the source and the drain, and is also effective for a so-called transfer MOS or the like in which the drain and the source are switched according to the circuit operating conditions to operate in both directions. In addition, the presence of the recombination center region in the source and drain does not affect the drain-substrate capacitance at all, and the parasitic capacitance reduction effect of the thick buried oxide film, which is the largest feature of conventional SOI and MOS, is maintained. Is done. Further, the present invention is effective irrespective of the conductivity type of the semiconductor device, and is therefore effective for eliminating the substrate floating effect of SOI · CMOS.
図面の簡単な說明 Brief description of drawings
図 1は、 従来の半導体装置の断面図。 FIG. 1 is a cross-sectional view of a conventional semiconductor device.
図 2は、 従来の半導体装置における基板浮遊効果解消機構のエネルギバンドの 説明図。 FIG. 2 is an explanatory view of an energy band of a substrate floating effect eliminating mechanism in a conventional semiconductor device.
図 3は、 本発明の半導体装置におけるソース拡散層の深さ方向不純物分布図。 図 4 A及び 4 Bは、 本発明の半導体装置における基板浮遊効果解消機構を説明 するエネルギノ N-ンド図とソ一ス等価回路図。 FIG. 3 is an impurity distribution diagram in the depth direction of the source diffusion layer in the semiconductor device of the present invention. 4A and 4B are an energy node diagram and a source equivalent circuit diagram illustrating a mechanism for eliminating a substrate floating effect in a semiconductor device of the present invention.
図 5 A及び 5 Bは、 本発明の半導体装置におけるソース拡散層近傍断面図と等 価回路図。 5A and 5B are a cross-sectional view near a source diffusion layer and an equivalent circuit diagram in the semiconductor device of the present invention.
図 6 A及び 6 Bは、 本発明の半導体装置による基板浮遊効果解消機構を説明す るエネルギバンド図。 6A and 6B are energy band diagrams illustrating a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention.
図 7は、 本発明の半導体装置による基板浮遊効果解消機構に関する解析結果。 図 8は、 本発明の半導体装置における基板浮遊効果解消機構を説明するソース 拡散層断面図。 FIG. 7 is an analysis result regarding a substrate floating effect eliminating mechanism by the semiconductor device of the present invention. FIG. 8 is a cross-sectional view of a source diffusion layer illustrating a mechanism for eliminating a substrate floating effect in a semiconductor device of the present invention.
図 9は、 本発明の半導体装置による基板浮遊効果解消機構を説明するエネルギ バンド図。 FIG. 9 is an energy band diagram for explaining a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention.
図 1 0は、 本発明の半導体装置による基板浮遊効果解消機構に関する解析結果。 図 1 1は、 本発明の第一の実施例による半導体装置の断面図。 FIG. 10 is an analysis result on a mechanism for eliminating a substrate floating effect by the semiconductor device of the present invention. FIG. 11 is a sectional view of the semiconductor device according to the first embodiment of the present invention.
図 1 2は、 本発明の第一の実施例による半導体装置の製造第一工程の断面図。 図 1 3は、 本発明の第一の実施例による半導体装置の製造第二工程の断面図。 図 1 4は、 本発明の第一の実施例による半導体装置の製造第三工程の断面図。 図 1 5は、 本発明の第二の実施例による半導体装置の断面図。 図 1 6は、 本発明の第三の実施例による半導体装置の断面図。 FIG. 12 is a sectional view of a first step in manufacturing a semiconductor device according to the first embodiment of the present invention. FIG. 13 is a sectional view of a second step in manufacturing the semiconductor device according to the first embodiment of the present invention. FIG. 14 is a sectional view of a third step of manufacturing the semiconductor device according to the first embodiment of the present invention. FIG. 15 is a sectional view of a semiconductor device according to a second embodiment of the present invention. FIG. 16 is a sectional view of a semiconductor device according to a third embodiment of the present invention.
図 1 7は、 本発明の第四の実施例による半導体装置の断面図。 FIG. 17 is a sectional view of a semiconductor device according to a fourth embodiment of the present invention.
図 1 8は、 本発明の第五の実施例による半導体装置の断面図。 FIG. 18 is a sectional view of a semiconductor device according to a fifth embodiment of the present invention.
図 1 9は、 本発明の第六の実施例による半導体装置の断面図。 FIG. 19 is a sectional view of a semiconductor device according to a sixth embodiment of the present invention.
図 2 0は、 本発明の第七の実施例による半導体装置の製造第一工程の断面図。 図 2 1は、 本発明の第七の実施例による半導体装置の第二工程の断面図。 FIG. 20 is a cross-sectional view of a first step in manufacturing a semiconductor device according to the seventh embodiment of the present invention. FIG. 21 is a sectional view of a second step of the semiconductor device according to the seventh embodiment of the present invention.
図 2 2は、 本発明の第 8の実施例による半導体装置の完成断面図。 FIG. 22 is a completed sectional view of a semiconductor device according to an eighth embodiment of the present invention.
図 2 3は、 本発明の第 8の実施例による半導体装置の製造工程順を示す断面図 c 図 2 4は、 本発明の第 8の実施例による半導体装置の製造工程順を示す断面図 c 図 2 5は、 本発明の第 8の実施例による半導体装置の製造工程順を示す断面図 c 図 2 6は、 本発明の第 9の実施例による半導体装置の製造工程順を示す断面図 c 図 2 7は、 本発明の第 9の実施例による半導体装置の製造工程順を示す断面図 c 図 2 8は、 本発明の第 9の実施例による半導体装置の完成断面図。 FIG. 23 is a cross-sectional view showing the order of the manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention c. FIG. 24 is a cross-sectional view showing the order of the manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention c 2 5 is a cross-sectional view c Figure 2 6 showing the manufacturing process sequence of the semiconductor device according to the eighth embodiment of the present invention, a ninth sectional view c showing the manufacturing process sequence of the semiconductor device according to an embodiment of the present invention 2 7 is a cross-sectional view c Figure 2 8 showing the manufacturing process sequence of the semiconductor device according to a ninth embodiment of the present invention, the finished cross-sectional view of a semiconductor device according to a ninth embodiment of the present invention.
図 2 9は、 本発明の第 1 0の実施例による半導体装置の完成断面図。 FIG. 29 is a completed sectional view of a semiconductor device according to a tenth embodiment of the present invention.
図 3 0は、 本発明の第 1 1の実施例による半導体装置の完成断面図。 FIG. 30 is a completed sectional view of the semiconductor device according to the eleventh embodiment of the present invention.
図 3 1は、 本発明の第 1 2の実施例による半導体装置の完成断面図。 FIG. 31 is a cross-sectional view of a completed semiconductor device according to the 12th embodiment of the present invention.
図 3 2は、 本発明の第 1 3の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 32 is a sectional view showing the order of manufacturing steps of the semiconductor device according to the thirteenth embodiment of the present invention.
図 3 3は、 本発明の第 1 3の実施例による半導体装置の完成断面図。 FIG. 33 is a completed sectional view of a semiconductor device according to a thirteenth embodiment of the present invention.
図 3 4は、 本発明の第 1 4の実施例による半導体装置の完成断面図。 FIG. 34 is a completed sectional view of the semiconductor device according to the fourteenth embodiment of the present invention.
図 3 5は、 本発明の第 1 4の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 35 is a sectional view showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention.
図 3 6は、 本発明の第 1 4の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 36 is a sectional view showing the order of manufacturing the semiconductor device according to the fourteenth embodiment of the present invention.
図 3 7は、 本発明の第 1 4の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 37 is a cross-sectional view showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention.
図 3 8は、 本発明の第 1 5の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 38 is a cross-sectional view showing the order of manufacturing steps of the semiconductor device according to the fifteenth embodiment of the present invention.
図 3 9は、 本発明の第 1 5の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 39 is a cross-sectional view showing a semiconductor device manufacturing process according to the fifteenth embodiment of the present invention. FIG.
図 4 0は、 本発明の第 1 5の実施例による半導体装置の完成断面図。 FIG. 40 is a completed sectional view of a semiconductor device according to a fifteenth embodiment of the present invention.
図 4 1は、 本発明の第 1 6の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 41 is a cross-sectional view showing the order of manufacturing the semiconductor device according to the sixteenth embodiment of the present invention.
図 4 2は、 本発明の第 1 6の実施例による半導体装置の完成断面図。 FIG. 42 is a completed sectional view of a semiconductor device according to a sixteenth embodiment of the present invention.
図 4 3は、 本発明の第 1 7の実施例による半導体装置の製造工程順を示す断面 図。 FIG. 43 is a cross-sectional view showing the order of the manufacturing process of the semiconductor device according to the seventeenth embodiment of the present invention.
図 4 4は、 本発明の第 1 7の実施例による半導体装置の完成断面図。 FIG. 44 is a completed sectional view of a semiconductor device according to a seventeenth embodiment of the present invention.
図 4 5 A及び 4 5 Bは、 本発明の実施例の適用例を説明するための随時書込み 読出し記憶装置の説明図。 FIGS. 45A and 45B are explanatory diagrams of an as-necessary write / read storage device for explaining an application example of the embodiment of the present invention.
図 4 6 A及び 4 6 Bは、 本発明の実施例の別の適用例を説明するための常時書 込み読出し記憶装置の説明図。 FIGS. 46A and 46B are explanatory diagrams of a constant write / read storage device for explaining another application example of the embodiment of the present invention.
図 4 7は、 本発明の実施例の他の適用例を説明するための論理回路図。 FIG. 47 is a logic circuit diagram for explaining another application example of the embodiment of the present invention.
図 4 8は、 本発明の実施例の更に別の適用例を説明するための計算機構成の説 明図。 FIG. 48 is an explanatory diagram of a computer configuration for explaining still another application example of the embodiment of the present invention.
図 4 9は、 本発明の実施例のもう一つ別の適用例を説明するための非同期伝送 モ一ドシステムの説明図。 FIG. 49 is an explanatory diagram of an asynchronous transmission mode system for explaining another application example of the embodiment of the present invention.
図 5 0は、 本発明の第 1 8の実施例による半導体装置の断面図。 FIG. 50 is a sectional view of a semiconductor device according to an eighteenth embodiment of the present invention.
図 5 1は、 本発明の第 1 9の実施例による半導体装置の断面図。 FIG. 51 is a sectional view of a semiconductor device according to a ninth embodiment of the present invention.
発明を実施するための最良の形態 BEST MODE FOR CARRYING OUT THE INVENTION
以下、 本発明を実施例によりさらに詳細に説明する。 理解を容易にするため、 図面を用いて説明し、 要部は他の部分よりも拡大して示されている。 Hereinafter, the present invention will be described in more detail with reference to Examples. For ease of understanding, the explanation is made with reference to the drawings, and the main parts are shown larger than other parts.
図 1 1から図 1 4は本発明の第一の実施例による半導体装置の製造工程を示す 断面図、 図 1 1はその完成断面図である。 図 1 2において、 高抵抗単結晶 S iよ りなる支持基板 1上に 5 0 O nm厚のシリコン酸化膜 (単に酸化膜と称する) 2、 及び 1 0 O nm厚の p導電型, 抵抗率 1 0 Ω αη, 面方位 (1 0 0 ) , 直径 1 2 . 5 cmの単結晶 S i層 3からなる S 0 I基板に公知の M 0 S電界効果型トランジスタ の製造方法により素子間分離絶縁膜 4, 5 nm厚のゲート酸化膜 5, タングステン (W) よりなるゲート電極 6, ゲート保護絶縁膜 7, 1 0 O nm厚のゲート側壁絶 縁膜 8を形成した。 なお、 ゲート酸化膜 5の形成に先立って、 閾電圧値が 0 . 1 Vとなるように単結晶 S i層 3に Bのイオン注入を施した。 ゲート長は 2 0 O nm である。 FIGS. 11 to 14 are cross-sectional views showing the steps of manufacturing a semiconductor device according to the first embodiment of the present invention, and FIG. 11 is a completed cross-sectional view thereof. In FIG. 12, a 50-nm-thick silicon oxide film (simply referred to as an oxide film) 2 on a supporting substrate 1 made of a high-resistance single-crystal Si, and a p-type conductivity of 100-nm-thickness, resistivity Isolation and isolation between elements by the well-known method of manufacturing a M0S field-effect transistor on an S0I substrate composed of a single-crystal Si layer 3 having a thickness of 10 Ω αη, plane orientation (100), and diameter of 12.5 cm. Film 4, 5 nm thick gate oxide film 5, Tungsten (W) gate electrode 6, Gate protective insulating film 7, 100 nm thick gate sidewall An edge film 8 was formed. Prior to the formation of the gate oxide film 5, B ions were implanted into the single-crystal Si layer 3 so that the threshold voltage became 0.1 V. The gate length is 20 O nm.
この状態よりゲー卜保護絶縁膜 7とゲ一卜側壁絶縁膜 8を注入阻止マスクとし て pのイオン注入とその後の熱処理を施し、 接合端が酸化膜 2に達し、 且つ実効 ゲート長が 1 5 O nm, 及び表面における最大不純物濃度が 1 x 1 0 1 Vcm3 とな るように低濃度 n型拡散層によるソース 9及びドレイン 1 0を形成した。 弓 I続き、 BF2 のイオン注入により低濃度 n型ソース 9及びドレイン 1 0内部に最大不純物 濃度 2 X 1 0 1 9/cm3 なる高濃度 p型拡散層 1 1を形成した。 高濃度 p型拡散層 1 1側面接合と低濃度ソース 9、 又は低濃度ドレイン接合間の最小幅は 5 0 nm、 高濃度 P型拡散層 1 1底面に残置された低濃度 n型拡散層は約 3 O nmであった (図 1 2 ) 。 From this state, ion implantation of p and subsequent heat treatment are performed using the gate protective insulating film 7 and the gate side wall insulating film 8 as an implantation preventing mask, the junction ends reach the oxide film 2, and the effective gate length is 15 O nm, and the maximum impurity concentration in the surface forming the source 9 and drain 1 0 by the low-concentration n-type diffusion layer so that Do a 1 x 1 0 1 Vcm 3. Following the bow I, a high concentration p-type diffusion layer 11 having a maximum impurity concentration of 2 × 10 19 / cm 3 was formed inside the low concentration n-type source 9 and the drain 10 by ion implantation of BF 2 . High-concentration p-type diffusion layer 1 1 Minimum width between side junction and low-concentration source 9 or low-concentration drain junction is 50 nm, high-concentration P-type diffusion layer 11 It was about 3 O nm (FIG. 12).
図 1 2の状態より A sの低エネルギイオン注入を行い、 表面濃度 2 X I 0 2 1 / cm3 なる高濃度 n型拡散層 9 1及び 1 0 1を高濃度 p型拡散層 1 1上に形成した。 イオン注入工程に引続き 1 0 0 0 °C, 1 0秒なる短時間熱処理を施し、 注入ィォ ンの活性化を行ったが高濃度 P型拡散層 1 1上の接合は整流特性を示さず、 悪い ォーミック特性で高濃度 n型拡散層 9 1及び 1 0 1と高濃度 p型拡散層 1 1は接 続されていた。 高濃度 n型拡散層 9 1及び 1 0 1の接合深さは約 2 O nmであった (図 1 3 ) 。 From the state shown in Fig. 12, low energy ion implantation of As is performed, and the high-concentration n-type diffusion layers 91 and 101 having a surface concentration of 2 XI 0 2 / cm 3 are placed on the high-concentration p-type diffusion layer 11. Formed. Following the ion implantation process, a short heat treatment of 100 ° C and 10 seconds was performed to activate the implanted ions, but the junction on the high-concentration P-type diffusion layer 11 did not exhibit rectification characteristics. The high-concentration n-type diffusion layers 91 and 101 and the high-concentration p-type diffusion layer 11 were connected with poor ohmic characteristics. The junction depth of the high-concentration n-type diffusion layers 91 and 101 was about 2 O nm (FIG. 13).
次にスパッタリングによる 1 5 O nm厚の W膜 1 2を全面に被着し、 少なくとも 高濃度 n型拡散層 9 1及び 1 0 1表面を覆うようにパターニングした。 なお、 W 膜 1 2は S i面への選択化学気相堆積法に基づいても良い (図 1 4 ) 。 Next, a W film 12 having a thickness of 15 O nm was deposited on the entire surface by sputtering, and was patterned so as to cover at least the surfaces of the high-concentration n-type diffusion layers 91 and 101. The W film 12 may be based on the selective chemical vapor deposition method on the Si surface (FIG. 14).
図 1 4の状態より公知の半導体装置の製造方法に基づき配線保護絶縁膜 1 3の 堆積と所望個所への開口、 更には配線金属の蒸着とそのパターニングによるソ一 ス電極 1 4、 ドレイン電極 1 5等を含む配線を形成した (図 1 1 ) 。 From the state shown in FIG. 14, a wiring protection insulating film 13 is deposited and an opening is formed at a desired position based on a known method of manufacturing a semiconductor device. Further, a source electrode 14 and a drain electrode 1 are formed by vapor deposition of wiring metal and patterning thereof. Wiring including 5 etc. was formed (Fig. 11).
この製造工程を経て製造された半導体装置のソース ' ドレイン間耐圧は 4 . 7 Vとソース内の p型拡散層 1 1が構成されていない同一寸法の従来構造 S O I · MO Sに比べて約 1 . 5 V向上し、 通常半導体基板に製造された同一寸法の M O Sと同等の耐圧値を確保することができた。 また、 電流 ·電圧特性でもキンク特性と称される異常なこぶ状特性は観測され ず、 正常な特性を示した。 更に、 ソース · ドレイン電流'ゲート電圧特性で、 従 来 S O I · MO Sで観測された低ゲート電圧におけるリーク電流の存在も観測さ れなかった。 またリーク電流、 及び閾電圧値はドレイン電圧を変化させても変化 が見出せなかった。 The withstand voltage between the source and the drain of the semiconductor device manufactured through this manufacturing process is 4.7 V, which is about 1 compared with the conventional structure SOI · MOS of the same size without the p-type diffusion layer 11 in the source. .5 V, and a breakdown voltage equivalent to that of MOS with the same dimensions normally manufactured on a semiconductor substrate could be secured. In the current-voltage characteristics, no abnormal bump-like characteristics called kink characteristics were observed, and the characteristics were normal. Furthermore, in the source-drain current'gate voltage characteristics, the existence of a leak current at a low gate voltage conventionally observed in SOI / MOS was not observed. No change was found in the leakage current and the threshold voltage even when the drain voltage was changed.
これらの特性から、 本実施例に基づく半導体装置では基板浮遊効果に伴う諸特 性から完全に解消されたことが明らかとなつた。 本実施例に基づく半導体装置の 電流 ·電圧特性が正常な特性を示すため n型高濃度拡散層 9 1及び 1 0 1と高抵 抗ォ一ミック特性を示し、 ソース、 及びドレイン内部に形成された p型高濃度拡 散層 1 1は何ら悪影響を及ぼさないことも判明した。 これは p型高濃度拡散層 1 1側面の領域が n型低濃度拡散層 9、 及び 1 0で構成され、 電流経路である表面 チャネル領域が n型低濃度拡散層 9 1、 及び 1 0 1で構成されているためチヤネ ル下部基板に発生した少数キヤリァである正孔は n型高濃度拡散層 9 1ではなく、 n型低濃度拡散層 9を介して p型高濃度拡散層 1 1に注入されると考えられる。 即ち、 n型高濃度拡散層 9 1の存在によるチャネルでの電流移送には何の影響も 与えないと考えられる。 From these characteristics, it has become clear that the semiconductor device according to the present embodiment has been completely eliminated from the characteristics associated with the substrate floating effect. Since the current-voltage characteristics of the semiconductor device according to the present embodiment show normal characteristics, the semiconductor device shows high resistance homogeneity with the n-type high concentration diffusion layers 91 and 101 and is formed inside the source and the drain. It was also found that the p-type high-concentration diffused layer 11 had no adverse effect. This is because the region on the side of the p-type high-concentration diffusion layer 11 is composed of n-type low-concentration diffusion layers 9 and 10, and the surface channel region, which is the current path, is the n-type low-concentration diffusion layer 91 and 101. Therefore, the holes that are minority carriers generated in the lower substrate of the channel are not in the n-type high-concentration diffusion layer 91 but in the p-type high-concentration diffusion layer 11 through the n-type low-concentration diffusion layer 9. It is thought to be injected. That is, it is considered that the presence of the n-type high-concentration diffusion layer 91 has no effect on the current transfer in the channel.
本実施例に基づく半導体装置のドレイン基板間容量も測定したが p型高濃度拡 散層 1 1の存在にも係わらず、 従来 S O I ' MO Sと同等の容量値を示し、 通常 半導体基板に製造された同一寸法の MO Sにおける値の約 1 / 1 0と小さなもの であった。 この結果はドレイン内の p型高濃度拡散層 1 1は n型低濃度拡散層 1 0により周りを囲まれており、 p型高濃度拡散層 1 1にはドレイン電界による空 乏層が形成されず、 容量は厚い絶縁膜 2により決定されるためと考えられる。 本実施例に基づく半導体装置では単結晶 S i層 3が 1 0 0 nmと極めて薄く、 チ ャネル領域における基板不純物濃度も 1 X 1 0 1 7/cm3 と低く設定されている。 従って、 単結晶 S i層 3内の電荷量限定により閾電圧以上のゲート電圧条件では チャネル領域の単結晶 S i層 3に中性領域が存在せず、 完全空乏状態となってい る。 これは電流駆動源であるチヤネル中の可動電荷を効果的に誘起することがで き、 大電流化に適している。 即ち、 低電圧 ·高速動作に適していることが知られ ている。 本実施例に基づく半導体装置では基板浮遊現象のない完全空乏型 MO S が従来半導体装置の製造方法のみで廉価に提供できることを示している。 The drain-to-substrate capacitance of the semiconductor device according to this example was also measured, but the capacitance value was equivalent to that of the conventional SOI'MOS, despite the presence of the p-type high-concentration diffusion layer 11. It was as small as about 1/10 of the value for the same size of MOS. As a result, the p-type high concentration diffusion layer 11 in the drain is surrounded by the n-type low concentration diffusion layer 10, and a depletion layer is formed in the p-type high concentration diffusion layer 11 by the drain electric field. It is considered that the capacitance is determined by the thick insulating film 2. In the semiconductor device according to the present embodiment, the single-crystal Si layer 3 is extremely thin at 100 nm, and the substrate impurity concentration in the channel region is set as low as 1 × 10 17 / cm 3 . Therefore, the neutral region does not exist in the single crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single crystal Si layer 3, and the single crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is known that it is suitable for low voltage and high speed operation. In the semiconductor device according to the present embodiment, a fully depleted MOS without a substrate floating phenomenon Indicate that it can be provided at a low price only by a conventional method of manufacturing a semiconductor device.
図 1 5は本発明の第二実施例による半導体装置の完成断面を示す図である。 実 施例 1で、 低濃度 n型拡散層 9、 及び 1 0の最大不純物濃度が 5 X 1 0 ' 8/cm;! , 高濃度 P型拡散層 1 1の最大不純物濃度が 1 X 1 0 1 9/cm3 , 拡散層底部が酸化 膜 2に達するように設定した。 また、 高濃度 n型拡散層 9 1、 及び 1 0 1の形成 を行わず、 且つソースドレイン電極 1 4、 及び 1 5が高濃度 p型拡散層 1 1に達 するように配線保護膜 1 3への開口形成工程で、 単結晶 S i膜を僅かにエツチン グした。 FIG. 15 is a view showing a completed cross section of the semiconductor device according to the second embodiment of the present invention. In actual Example 1, the low-concentration n-type diffusion layer 9, and 1 maximum impurity concentration of 5 X 1 0 of 0 '8 / cm;!, 1 is the maximum impurity concentration of the high concentration P-type diffusion layer 1 1 X 1 0 It was set so that the bottom of the diffusion layer reached oxide film 2 at 19 / cm 3 . Further, the wiring protective film 13 is formed so that the high-concentration n-type diffusion layers 91 and 101 are not formed, and the source / drain electrodes 14 and 15 reach the high-concentration p-type diffusion layer 11. The single crystal Si film was slightly etched in the step of forming an opening in the substrate.
本実施例の半導体装置も実施例 1による半導体装置と同様にソース · ドレイン 間耐圧の低下、 電流電圧特性におけるこぶ状特性、 及び閾電圧の負方向変動等は 観測されず基板浮遊効果は見出せなかった。 Similarly to the semiconductor device according to the first embodiment, the semiconductor device according to the present embodiment does not show a decrease in the breakdown voltage between the source and the drain, a bump-like characteristic in the current-voltage characteristic, a negative-direction fluctuation in the threshold voltage, and the like. Was.
図 1 6は本発明の第三実施例による半導体装置の完成断面を示す図である。 実 施例 1で素子間分離絶縁膜 4を形成し、 単結晶 S i膜 3の活性領域を互いに分離 した後、 所望の回路構成に従って活性領域の一部にイオン注入を施して低濃度 n 型領域 3 1とした。 低濃度 n型領域 3 1と低濃度 p型領域 3上に実施例 1に従つ て、 ゲート酸化膜 5、 ゲート電極 6 1、 ゲート保護絶緣膜 7、 ゲート側壁絶縁膜 8を形成した。 FIG. 16 is a view showing a completed cross section of the semiconductor device according to the third embodiment of the present invention. In Example 1, the element isolation insulating film 4 was formed, and the active regions of the single-crystal Si film 3 were separated from each other. Region 31 was set. According to Example 1, a gate oxide film 5, a gate electrode 61, a gate protection insulating film 7, and a gate sidewall insulating film 8 were formed on the low-concentration n-type region 31 and the low-concentration p-type region 3.
本実施例ではゲート電極 6 1として低抵抗多結晶 S i膜を用いた。 ゲート電極 6 1ゲ一ト側壁絶縁膜 8をマスクとして低濃度 n型領域 3 1にのみ選択的に Bィ オンを注入し、 その後の熱処理により酸化膜 2に達し、 表面不純物濃度が 5 X I 0 1 cm3 の低濃度 p型拡散層 9 0及び 1 0 0を形成した。 続いて活性領域内部 で最大不純物濃度が 1 X 1 0 ' Vcm3 で、 低濃度 p型拡散層 9 0及び 1 0 0内部 に位置するように A sのイオン注入を施して n型拡散層 1 1 0を形成した。 In this embodiment, a low-resistance polycrystalline Si film was used as the gate electrode 61. Using the gate electrode 6 1 gate sidewall insulating film 8 as a mask, B ions are selectively implanted only into the low-concentration n-type region 31 and the heat treatment thereafter reaches the oxide film 2 and the surface impurity concentration is 5 XI 0 Low-concentration p-type diffusion layers 90 and 100 of 1 cm 3 were formed. Subsequently, As ion implantation is performed so that the maximum impurity concentration is 1 × 10 ′ Vcm 3 inside the active region and the low impurity concentration p-type diffusion layers 90 and 100 are located inside the n-type diffusion layer 1. 10 were formed.
低濃度 P型領域 3に対してはゲート電極 6 1、 ゲート側壁絶縁膜 8をマスクと して選択的に Pのイオン注入を行い、 表面で最大濃度 3 X 1 0 1 8/cm3 となり酸 化膜 2に達する低濃度 n型拡散層を形成し、 ソース 9、 及びドレイン 1 0とした。 続いて、 1 X 1 0 2 Vcm3 表面濃度の G eのイオン注入をソース 9、 及びドレイ ン 1 0内に施して S i · G e共晶層 1 6をソース 9及びドレイン内に形成した。 その後、 実施例 1に従い配線保護絶縁膜 1 3の堆積と所望個所への開口、 配線 用金属膜の蒸着とそのパターニングにより接地電位線 1 7、 出力端子 1 8、 及び 電源電圧線 1 9を含む配線を形成した。 P ions are selectively implanted into the low-concentration P-type region 3 using the gate electrode 61 and the gate side wall insulating film 8 as a mask, and the surface has a maximum concentration of 3 × 10 18 / cm 3 and an acid. A low-concentration n-type diffusion layer reaching the oxide film 2 was formed, and a source 9 and a drain 10 were formed. Subsequently, an ion implantation of Ge having a surface concentration of 1 × 10 2 Vcm 3 was performed in the source 9 and the drain 10 to form a Si · Ge eutectic layer 16 in the source 9 and the drain. . Then, according to the first embodiment, the wiring protection insulating film 13 is deposited, an opening is formed at a desired position, and the wiring is formed. A wiring including a ground potential line 17, an output terminal 18 and a power supply voltage line 19 was formed by vapor deposition and patterning of a metal film for use.
本実施例の CMOSでは pMOS, nMOSの何れに関しても基板浮遊効果に 起因する諸症状を観測することができなかった。 更に n M 0 S閾電圧値の負方向 変動、 pMOS閾電圧値の正方向変動によって接地電位線 1 7と電源電圧線 1 9 間に生じる SO I · CMOS特有の基板浮遊効果に基づく貫通電流も観測されな かった。 pMOSで、 基板浮遊効果が見られなかったことは、 チャネル下部単結 晶 S i膜 3 1に発生した少数キャリアである電子が n型拡散層 1 1 0に向かって 注入され、 ソース低濃度 p型拡散層 1 00に達した電子が n型拡散層 1 1 0と p 型拡散層 1 00間の接合における空乏層で再結合により消滅するためと考えられ る。 In the CMOS of this example, various symptoms caused by the substrate floating effect could not be observed for any of the pMOS and the nMOS. Furthermore, the through current due to the substrate floating effect unique to SOICMOS generated between the ground potential line 17 and the power supply line 19 due to the negative fluctuation of the nM0S threshold voltage and the positive fluctuation of the pMOS threshold voltage It was not observed. The absence of the substrate floating effect in the pMOS means that electrons, which are minority carriers generated in the single-crystal Si film 31 below the channel, are injected toward the n-type diffusion layer 110, and the source has a low concentration p. It is considered that the electrons reaching the diffusion layer 100 disappear by recombination in the depletion layer at the junction between the n-type diffusion layer 110 and the p-type diffusion layer 100.
図 1 7は本発明の第四実施例による半導体装置の完成断面を示す図である。 前 記第三実施例で G eのィォン注入に代えて Bのィォン注入を行い、 高濃度 p型拡 散層 1 1を低濃度 n型拡散層 9及び 1 0内部に形成した。 高濃度 p型拡散層 1 1 は酸化膜 2に達し、 2 X 1 0 ig/cm3 なる最大不純物濃度はソース 9表面から 8 0 nm内部に位置するように設定した。 FIG. 17 is a view showing a completed cross section of a semiconductor device according to a fourth embodiment of the present invention. In the third embodiment, ion implantation of B was performed instead of ion implantation of Ge to form a high-concentration p-type diffusion layer 11 inside the low-concentration n-type diffusion layers 9 and 10. The high-concentration p-type diffusion layer 11 reached the oxide film 2, and the maximum impurity concentration of 2 × 10 ig / cm 3 was set so as to be located within 80 nm from the surface of the source 9.
本実施例の CMOSでは、 pMOS, nMOSの何れに関しても基板浮遊効果 に起因する諸症状を観測することができなかった。 基板浮遊効果の解消に関し、 pMOSに関しては実施例 3と同様な理由に基づくと考えられ、 nMOSでもチ ャネル下部単結晶 S i膜 3に発生した少数キヤリァである正孔が p型拡散層 1 1 に向かい注入され、 ソース低濃度 n型拡散層 1 0に達した正孔が p型拡散層 1 1 と n型拡散層 1 0間の接合における空乏層で再結合により消滅するためと考えら れる。 In the CMOS of this example, various symptoms caused by the substrate floating effect could not be observed for any of the pMOS and the nMOS. With respect to the elimination of the substrate floating effect, it is considered that the pMOS is based on the same reason as in the third embodiment, and even in the nMOS, holes, which are a small number of carriers generated in the channel lower single-crystal Si film 3, are p-type diffusion layers 11 1 It is thought that the holes injected toward the source and reached the source low-concentration n-type diffusion layer 10 disappear by recombination at the depletion layer at the junction between the p-type diffusion layer 11 and the n-type diffusion layer 10 .
更に本実施例の製造工程ではイオン源が不安定で、 組成比で約 1 0%にも達す る大ィォン電流を要する G eのィォン注入を施す必要がないため実施例 3に比べ て廉価に基板浮遊効果に対する対策をすることができた。 Further, in the manufacturing process of the present embodiment, the ion source is unstable, and it is not necessary to perform the ion implantation of Ge which requires a large ion current reaching a composition ratio of about 10%. Measures against the substrate floating effect were taken.
図 1 8は本発明の第五実施例による半導体装置の完成断面を示す図である。 第 三実施例で酸化膜 2上の単結晶 S i膜 3の膜厚が 300 nmの S 0 I基板を用い、 n型拡散層 9及び 1 0、 更には p型拡散層 90及び 1 00の接合深さを 1 50 nm に設定した。 p型拡散層 9 0及び 1 0 0内部に選択的に形成された n型拡散層 1 1 0の上部に位置するように高濃度 p型拡散層 9 2及び 1 0 2を BF2 のイオン注 入により形成したが、 イオン注入により n型拡散層 1 1 0の上部接合部での結晶 欠陥発生により n型拡散層 1 1 0と!)型拡散層 9 2及び 1 0 2とは電気的に高抵 杭で短絡された。 高濃度 p型拡散層 9 2及び 1 0 2の最大不純物濃度は表面部で あり、 1 X 1 0 2 0/cm3 に設定した。 FIG. 18 is a view showing a completed cross section of a semiconductor device according to a fifth embodiment of the present invention. In the third embodiment, an S0I substrate having a thickness of 300 nm of the single-crystal Si film 3 on the oxide film 2 was used, and the n-type diffusion layers 9 and 10 and the p-type diffusion layers 90 and 100 were formed. 150 nm junction depth Set to. ion Note the p-type diffusion layer 9 0 and 1 0 0 as interior positioned above the selectively formed n-type diffusion layer 1 1 0 high concentration p-type diffusion layer 9 2 and 1 0 2 BF 2 However, due to crystal defects at the upper junction of the n-type diffusion layer 110 due to ion implantation, the n-type diffusion layer 110 is formed! ) -Type diffusion layers 92 and 102 were electrically short-circuited at high piles. The maximum impurity concentration of the high-concentration p-type diffusion layers 92 and 102 was on the surface, and was set to 1 × 10 20 / cm 3 .
半導体装置 C MO Sで p MO S, n MO Sの何れに関しても基板浮遊効果に起 因する諸症状を観測することができなかった。 更に本実施例の半導体装置は p M O Sにおけるドレイン基板間容量が従来 S O I · MO Sと同等の容量値を示し、 通常半導体基板に製造された同一寸法の MO Sにおける値の約 1 / 1 0と小さな ものであった。 この結果はドレイン内の n型高濃度拡散層 1 1 0が p型低濃度拡 散層 1 0 0により周りを囲まれており、 n型高濃度拡散層 1 1 0にはドレイン電 界による空乏層が形成されず、 容量は厚い絶縁膜 2により決定されるためと考え 本実施例に基づく半導体装置では単結晶 S i層 3及び 3 1が 3 0 0議と比較的 厚く、 チャネル領域下部における基板領域では閾電圧以上のゲート電圧印加によ つても空乏層と中性領域力 <存在する、 いわゆる、 部分空乏化構造となる。 部分空 乏化構造は低電圧 ·高速動作で完全空乏構造に比べてやや落ちるが製造条件で従 来の半導体基板を用し、た条件で容易に製造できる。 本実施例に基づく半導体装置 では廉価に部分空乏化構造 MO Sの基板浮遊対策を提供できることを示している。 図 1 9は本発明の第六実施例による半導体装置の完成断面を示す図である。 第 五実施例で G eのィォン注入に代えて Bのィォン注入を行 、、 高濃度 p型拡散層 1 1を低濃度 n型拡散層 9及び 1 0内部に形成し、 引続いて高濃度 n型拡散層 9 1、 及び 1 0 1形成の A sイオン注入を実施した。 A sイオン注入で最大不純物 濃度は表面部であり、 5 X 1 0 2 0/cm3 に設定したがこれにより p型拡散層 1 1 の上部接合部での結晶欠陥の発生により P型拡散層 1 1と高濃度 n型拡散層 9 1 及び 1 0 1とは電気的に高抵抗で短絡された。 In the semiconductor device CMOS, no symptoms caused by the substrate floating effect could be observed for either pMOS or nMOS. Furthermore, in the semiconductor device of this embodiment, the drain-to-substrate capacitance in the pMOS shows a capacitance value equivalent to that of the conventional SOI.MOS, which is about 1/10 of the value of the MOS of the same dimensions usually manufactured on the semiconductor substrate. It was small. The result is that the n-type high-concentration diffusion layer 110 in the drain is surrounded by the p-type low-concentration diffusion layer 100, and the n-type high-concentration diffusion layer 110 is depleted by the drain electric field. It is considered that the layer is not formed and the capacitance is determined by the thick insulating film 2. In the semiconductor device according to the present embodiment, the single-crystal Si layers 3 and 31 are relatively thick at 300, In the substrate region, a depletion layer and a neutral region force <exist, that is, a so-called partially depleted structure is obtained even when a gate voltage higher than the threshold voltage is applied. The partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, but can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. It is shown that the semiconductor device according to the present embodiment can provide a countermeasure against the substrate floating of the partially depleted structure MOS at low cost. FIG. 19 is a view showing a completed cross section of the semiconductor device according to the sixth embodiment of the present invention. In the fifth embodiment, ion implantation of B was performed instead of ion implantation of Ge, and a high-concentration p-type diffusion layer 11 was formed inside the low-concentration n-type diffusion layers 9 and 10, and subsequently a high-concentration p-type diffusion layer 11 was formed. As ion implantation for forming n-type diffusion layers 91 and 101 was performed. The maximum impurity concentration was set to 5 × 10 20 / cm 3 at the surface part by As ion implantation, but this resulted in the generation of crystal defects at the upper junction of the p-type diffusion layer 11 and the P-type diffusion layer. 11 and the high concentration n-type diffusion layers 91 and 101 were electrically short-circuited with high resistance.
本実施例に基づく半導体装置 C MO Sで p MO S, n M O Sの何れに関しても 基板浮遊効果に起因する諸症状を観測することができなかった。 更に本実施例の 製造工程ではイオン源が不安定で、 組成比で約 1 0 %にも達する大イオン電流を 要する G eのィォン注入を施す必要がないため実施例 3に比べて廉価に基板浮遊 効果に対する対策をすることができた。 本実施例の半導体装置では p M 0 S及び n MO S共にドレイン基板間容量が従来 S 0 I · M O Sと同等の容量値を示し、 通常半導体基板に製造された同一寸法の M O Sにおける値の約 1 Z 1 0と小さな ものであった。 この結果はドレイン内の n型高濃度拡散層 1 1 0が p型低濃度拡 散層 1 0 0により、 p型高濃度拡散層 1 1が低濃度型拡散層 1 0により周りを囲 まれる構成になっているため、 n型高濃度拡散層 1 1 0及び p型高濃度拡散層 1 1にはドレイン電界による空乏層が形成されず、 容量は厚い絶縁膜 2により決定 されるためと考えられる。 In the semiconductor device CMOS based on this example, no symptom caused by the substrate floating effect could be observed for any of pMOS and nMOS. Further, in this embodiment, In the manufacturing process, the ion source is unstable, and it is not necessary to perform Ge ion implantation, which requires a large ion current of about 10% in the composition ratio. We were able to. In the semiconductor device of the present embodiment, the drain-to-substrate capacitance of both pM0S and nMOS shows the same capacitance value as that of the conventional S0IMOS, which is about the same as the value of the MOS of the same size usually manufactured on the semiconductor substrate. It was as small as 1 Z 10. The result is that the n-type high-concentration diffusion layer 110 in the drain is surrounded by the p-type low-concentration diffusion layer 100, and the p-type high-concentration diffusion layer 11 is surrounded by the low-concentration diffusion layer 10 It is considered that no depletion layer is formed in the n-type high-concentration diffusion layer 110 and the p-type high-concentration diffusion layer 11 due to the drain electric field, and the capacitance is determined by the thick insulating film 2. Can be
図 2 0は本発明の第七実施例による半導体装置の製造工程を示す断面図、 図 2 1はその完成断面を示す図である。 第六の実施例で、 酸化膜 2と単結晶 S i膜 3 の間に 1 0 0 nm厚の高抵抗多結晶 S i膜 2 0と 1 0腦厚のシリコン酸化膜 2 1が 構成された多層構造 S O I基板を用いた。 ここで単結晶 S i膜 3の膜厚は 1 0 0 nmとし、 p型低濃度活性領域 3及び n型低濃度活性領域 3 1の不純物濃度は各々 1 X 1 0 1 6/cm3 と極めて低く設定した。 即ち、 本実施例では完全空乏型の相補 型 M 0 S電界効果トランジスタを製造した。 FIG. 20 is a sectional view showing a manufacturing process of the semiconductor device according to the seventh embodiment of the present invention, and FIG. 21 is a view showing a completed section thereof. In the sixth embodiment, a high-resistance polycrystalline Si film 20 having a thickness of 100 nm and a silicon oxide film 21 having a thickness of 10 were formed between the oxide film 2 and the single-crystal Si film 3. A multi-layer SOI substrate was used. Here, the thickness of the single-crystal Si film 3 is 100 nm, and the impurity concentration of the p-type low concentration active region 3 and the n-type low concentration active region 31 is 1 × 10 16 / cm 3 , respectively. Set low. That is, in this example, a fully depleted complementary MOS field effect transistor was manufactured.
本実施例ではゲート酸化膜 5、 ゲート電極 6、 ゲート保護絶縁膜 7、 ゲ一ト側 壁絶縁膜 8の形成に先立つて n M O Sのゲート電極 7形成予定領域直下の高抵抗 多結晶 S i膜 2 0に不純物濃度 2 X 1 0 1 cm3 なる高濃度 p型不純物領域 2 2 を、 p MO Sのゲート電極 7形成予定領域直下の高抵抗多結晶 S i膜 2 0に不純 物濃度 2 X 1 0 l 8/cm3 なる高濃度 η型不純物領域 2 3を予め形成した (図 2In this embodiment, prior to the formation of the gate oxide film 5, the gate electrode 6, the gate protective insulating film 7, and the gate side wall insulating film 8, the high-resistance polycrystalline Si film immediately below the region where the nMOS gate electrode 7 is to be formed is formed. A high-concentration p-type impurity region 22 having an impurity concentration of 2 × 10 1 cm 3 is added to a high-resistance polycrystalline Si film 20 immediately below a region where a pMOS gate electrode 7 is to be formed. 1 0 l 8 / cm 3 comprising a high concentration of preformed η-type impurity region 2 3 (FIG. 2
0 ) 0)
図 2 0の状態から第六実施例で半導体装置を製造したが、 n M O Sのソース内 に形成する高濃度 p型拡散層 1 1、 上部の高濃度 n型拡散層 9 1、 及び p M O S におけるソース内の高濃度 n型拡散層 1 1 0、 上部の高濃度 p型拡散層 9 2の各 接合深さに関する条件は実施例 1に従った。 更にゲート、 ゲート側壁絶縁膜部を 除くソース、 ドレイン全表面に実施例 1に従って W膜 1 2を選択的に被着した この本実施例の C MO Sで p MO S, n MO Sの何れに関しても基板浮遊効果 に起因する諸症状が観測されないことは、 実施例 6と同様であつたが、 実施例 6 の半導体装置に比べて本実施例に基づく半導体装置のゲ一ト、 ドレイン電圧 2 V におけるソース ' ドレイン電流は p MO S, n MO S共 1 . 4倍以上の大電流化 力達成された。 しかも低ドレイン電圧条件である非飽和特性領域で、 実施例 6の 半導体装置に比べて極めて急峻なドレインコンダク夕ンス特性が得られた。 本実施例に基づく半導体装置により構成されたリングオシレータでは一段当り の遅延時間が 1 2 p秒と、 実施例 6に基づくリングォシレー夕に比べて 6 p秒も 高速化が達成された。 このような超高速、 大電流特性は高濃度 p型不純物領域 2 2、 及び高濃度 n型不純物領域 2 3がパンチスルー防止機構として作用し、 且つ 薄いシリコン酸化膜 2 1が製造工程中の高温熱処理に対する不純物拡散マスクと して作用するため、 チャネルを構成する P型低濃度活性領域 3、 及び n型低濃度 活性領域 3 1の不純物濃度を極めて低濃度に設定でき、 不純物散乱による移動度 の劣化が防止できたためと考えられる。 Although the semiconductor device was manufactured in the sixth embodiment from the state of FIG. 20, the high-concentration p-type diffusion layer 11 formed in the source of the nMOS, the high-concentration n-type diffusion layer 91 on the top, and the pMOS The conditions relating to the junction depth of the high-concentration n-type diffusion layer 110 in the source and the high-concentration p-type diffusion layer 92 in the upper part were as in Example 1. Further, a W film 12 was selectively deposited on the entire surface of the source and drain except for the gate and the gate side wall insulating film portion according to Example 1. As in the sixth embodiment, in the CMOS of the present embodiment, no symptom caused by the substrate floating effect was observed in any of the pMOS and nMOS in the same manner as in the sixth embodiment. In comparison with this, the source-drain current at a gate and drain voltage of 2 V of the semiconductor device according to the present embodiment was at least 1.4 times larger than that of pMOS and nMOS. In addition, in the non-saturation characteristic region where the drain voltage is low, an extremely steep drain conductance characteristic was obtained as compared with the semiconductor device of Example 6. In the ring oscillator constituted by the semiconductor device according to the present embodiment, the delay time per stage was 12 ps, which was 6 ps faster than the ring oscillator according to the sixth embodiment. Such ultra-high-speed, high-current characteristics are such that the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 is formed at a high temperature during the manufacturing process. Since it acts as an impurity diffusion mask for heat treatment, the impurity concentration of the P-type low-concentration active region 3 and the n-type low-concentration active region 31 forming the channel can be set to an extremely low concentration, and the mobility due to impurity scattering can be reduced. It is considered that the deterioration was prevented.
図 2 3から図 2 5は本発明の第 8の実施例による半導体装置の製造工程順を示 す断面図、 図 2 2は完成断面図である。 図 2 3において、 直径 1 2 . 5 cmの単結 晶 S iよりなる支持基板 1上に 5 0 0腦厚のシリコン酸化膜 (単に酸化膜と称す る) 2、 及び 1 0 O nm厚の p導電型、 抵抗率 1 0 Ωαη、 面方位 ( 1 0 0 ) の単結 晶 S i層 3からなる S 0 I基板に公知の M 0 S電界効果型トランジスタの製造方 法により素子間分離絶縁膜 4、 5讓厚のゲート酸化膜 5、 n型低抵抗多結晶 S i 膜よりなるゲート電極 6、 ゲート保護絶縁膜 7を形成した。 なお、 ゲート酸化膜 5の形成に先立って、 閾電圧値が 0 . 1 Vとなるごとく単結晶 S i層 3に Bのィ オン注入を施した。 ゲート長は 2 0 O nmである。 この状態よりゲート保護絶縁膜 7及びゲート電極 6を注入阻止マスクとして pのィォン注入とその後の熱処理を 施し、 接合端が酸化膜 2に達し、 且つ実効ゲート長が 1 5 O nm. 及び表面におけ る最大不純物濃度が 5 X I 0 1 7/cm3 となるごとく低濃度 n型拡散層によるソー ス 9及びドレイン 1 0を形成した。 上記ィォン注入では単結晶 S i層 3は単結晶 性を維持していた。 しかる後、 5 O nm厚の堆積性絶縁膜を全面に形成し、 異方性 ドライエツチングによりゲ一ト側壁部にのみ上記絶縁膜を選択的に残置させてゲ 一ト側壁絶縁膜 8を形成した。 ゲ一ト側壁絶縁膜 8の膜厚条件に関して、 1 0か ら 2 O nm間隔で変化させた本実施例に基づく半導体装置も別途製造した。 引続き、 BF2 のイオン注入により低濃度 n型ソース 9及びドレイン 1 0内部で、 酸化膜 2 界面で最大不純物濃度 2 X 1 0 1 9/cm3 なる高濃度 p型領域 1 1を形成した。 酸 ィヒ膜 2界面において、 高濃度 p型領域 1 1側面に残存する低濃度 n型拡散層 9の 幅は最終的に 4 O nmであった。 B F 2 のイオン注入はその最大不純物濃度が酸化 膜 2内になるごとく加速エネルギーを設定してもよい。 上記ィォン注入により高 濃度 P型領域 1 1は非晶質に変換された (図 2 3 )。 FIGS. 23 to 25 are sectional views showing the order of manufacturing steps of the semiconductor device according to the eighth embodiment of the present invention, and FIG. 22 is a completed sectional view. In FIG. 23, a silicon oxide film (hereinafter simply referred to as an oxide film) 2 having a thickness of 500 nm and a silicon oxide film having a thickness of 10 O nm are formed on a support substrate 1 made of a single crystal Si having a diameter of 12.5 cm. Isolation and isolation between elements on a S 0 I substrate consisting of a single crystal Si layer 3 with p conductivity type, resistivity of 10 Ωαη, and plane orientation of (100) by well-known M 0 S field effect transistor Film 4, a gate oxide film 5 having a thickness of 5 cm, a gate electrode 6 made of an n-type low-resistance polycrystalline Si film, and a gate protection insulating film 7 were formed. Prior to the formation of the gate oxide film 5, B ions were implanted into the single-crystal Si layer 3 so that the threshold voltage became 0.1 V. The gate length is 20 O nm. From this state, ion implantation of p and subsequent heat treatment are performed using the gate protective insulating film 7 and the gate electrode 6 as an injection blocking mask, the junction end reaches the oxide film 2, the effective gate length is 15 O nm. maximum impurity concentration that put were formed source 9 and the drain 1 0 by 5 XI 0 1 7 / cm 3 and becomes as low concentration n-type diffusion layer. In the above ion implantation, the single-crystal Si layer 3 maintained the single-crystal property. Thereafter, a deposited insulating film having a thickness of 5 O nm is formed on the entire surface, and the insulating film is selectively left only on the side walls of the gate by anisotropic dry etching. A first sidewall insulating film 8 was formed. A semiconductor device based on the present example in which the thickness condition of the gate sidewall insulating film 8 was changed from 10 to 20 nm at an interval was also manufactured separately. Subsequently, a high-concentration p-type region 11 having a maximum impurity concentration of 2 × 10 19 / cm 3 was formed at the interface with the oxide film 2 inside the low-concentration n-type source 9 and the drain 10 by ion implantation of BF 2 . At the interface of the oxygen film 2, the width of the low-concentration n-type diffusion layer 9 remaining on the side surface of the high-concentration p-type region 11 was finally 4 O nm. In the ion implantation of BF 2, the acceleration energy may be set so that the maximum impurity concentration is in oxide film 2. The high-concentration P-type region 11 was converted to amorphous by the above ion implantation (Fig. 23).
図 2 3の状態より A sの低エネルギーィォン注入を行ない、 表面濃度 2 X 1 0 2 Vcm3 なる高濃度 n型領域 9 1及び 1 0 1を高濃度 p型拡散層 1 1上に形 成した。 上記イオン注入工程に引続き 1 0 0 0 °C、 1 0秒なる短時間熱処理を施 し、 注入イオンの活性化を行なったが高濃度 P型領域 1 1はゲート直下の横方向 領域を除き微細粒径の多結晶で、 厚さは約 1 O nmであった。 高濃度 n型領域 9 1 及び 1 0 1の接合深さは約 4 O nmであった (図 2 4 ) 。 2 3 state performs low energy I O emissions injection of A s from, the surface concentration 2 X 1 0 2 Vcm 3 becomes high-concentration n-type regions 9 1 and 1 0 1 a high concentration p-type diffusion layer 1 1 on the shape forming did. Following the above ion implantation process, a short heat treatment at 100 ° C for 10 seconds was performed to activate the implanted ions, but the high-concentration P-type region 11 was fine except for the lateral region immediately below the gate. It was polycrystalline with a particle size of about 1 O nm in thickness. The junction depth of the high-concentration n-type regions 91 and 101 was about 40 nm (FIG. 24).
図 2 4の状態よりソース抵抗の低減を図る目的で化学気相反応による 1 5 O nm 厚の W膜 1 2を露出されている S i面に選択的に堆積した。 上記 W膜 1 2はスパ ッタリングによる全面被着と、 少なくとも高濃度 n型領域 9 1及び 1 0 1表面を 覆うごとくパターニングして形成しても良い。 しかる後、 燐が添加されたシリコ ン酸化膜による配線保護絶縁膜 1 3の堆積を施した (図 2 5 ) 。 In order to reduce the source resistance from the state shown in FIG. 24, a 15 O nm thick W film 12 was selectively deposited on the exposed Si surface by a chemical vapor reaction. The W film 12 may be formed by sputtering over the entire surface and by patterning so as to cover at least the high-concentration n-type regions 91 and 101 surfaces. Thereafter, a wiring protection insulating film 13 was deposited using a silicon oxide film to which phosphorus was added (FIG. 25).
図 2 5の状態より公知の半導体装置の製造方法に基づき配線保護絶縁膜 1 3の 所望個所への開口、 更には配線金属の蒸着とそのパターニングによるソース電極 1 4、 ドレイン電極 1 5等を含む配線を形成した。 なお、 最終工程における単結 晶 S i層 3の膜厚は製造工程中の洗浄化処理等により減少し、 5 O nmであった (図 2 2 ) 。 From the state shown in FIG. 25, based on a known method for manufacturing a semiconductor device, an opening is formed in a desired portion of the wiring protective insulating film 13, and further, a source electrode 14, a drain electrode 15 and the like are formed by vapor deposition of wiring metal and patterning thereof. Wiring was formed. The film thickness of the single crystal Si layer 3 in the final step was reduced to 5 O nm by the cleaning treatment in the manufacturing step (FIG. 22).
上記製造工程を経て製造された本実施例に基づく半導体装置のソース · ドレイ ン間耐圧は 4 . 7 Vとソース内の p型拡散層 1 1が構成されていない同一寸法の 従来構造 S O I M O Sに比べて約 1 . 5 V向上し、 通常半導体基板に製造された 同一寸法の M O Sと同等の耐圧値を確保することができた。 また、 電流 ·電圧特 性においてもキンク特性と称される異常なこぶ状特性は観測されず、 正常な特性 を示した。 更に、 ソース · ドレイン電流 ·ゲ一ト電圧特性において、 従来 S O IThe withstand voltage between the source and the drain of the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process is 4.7 V, which is smaller than that of the conventional structure SOIMOS having the same dimensions without the p-type diffusion layer 11 in the source. By about 1.5 V, it was possible to secure a breakdown voltage equivalent to that of MOS transistors of the same dimensions normally manufactured on semiconductor substrates. In the current and voltage characteristics, no abnormal bump-like characteristics called kink characteristics were observed. showed that. Furthermore, in the source-drain current-gate voltage characteristics,
• M 0 Sで観測された低ゲート電圧におけるリーク電流の存在も観測されなかつ た。 また上記リーク電流、 及び閾電圧値はドレイン電圧を変化させても変化が見 出せなかった。 これらの特性から、 本実施例に基づく半導体装置では基板浮遊効 果に伴う緒特性から完全に解消されたことが明らかとなった。 本実施例に基づく 半導体装置の電流'電圧特性は正常な特性を示し、 ソース、 及びドレイン内部に 形成された P型高濃度領域 1 1は何ら悪影響を及ぼさないことも判明した。 本実 施例に基づく上記基板浮遊効果の解消は本実施例に基づきゲ一ト側壁絶縁膜 8の 膜厚を変化させて製造した半導体装置において、 • Leakage current at low gate voltage observed at M 0 S was not observed. Also, no change was found in the above leakage current and threshold voltage even when the drain voltage was changed. From these characteristics, it has been clarified that the semiconductor device according to the present example was completely eliminated from the characteristics associated with the substrate floating effect. It was also found that the current-voltage characteristics of the semiconductor device according to the present example showed normal characteristics, and that the P-type high-concentration region 11 formed inside the source and the drain had no adverse effect. The elimination of the substrate floating effect according to the present embodiment is achieved by changing the thickness of the gate sidewall insulating film 8 according to the present embodiment in a semiconductor device.
高濃度 p型領域 1 1端から低濃度 n型拡散層接合までの間隔が 1 0 Ο ηπ ^下から 2 0 まで観測されたが、 3 0から 5 O nmの間隔の場合が特性のばらつきも観測 されず、 最も好ましいことが明らかとなった。 なお、 高濃度 p型領域 1 1端が低 濃度 n型拡散層接合を越えて構成された構造においては基板浮遊効果解消は見出 せなかった。 The interval from the high-concentration p-type region 1 1 end to the low-concentration n-type diffusion layer junction was observed from 10 Ο ηπ ^ from bottom to 20. No observation was made, and it was clear that it was the most favorable. In the structure in which one end of the high-concentration p-type region 11 is beyond the low-concentration n-type diffusion layer junction, the substrate floating effect could not be eliminated.
本実施例に基づく半導体装置が S O I M O Sの基板浮遊効果解消に有効なこと 力、ら酸化膜 2に接して構成された高濃度 p型領域 1 1の多結晶性が注入された正 孔の再結晶中心として十分に作用することが推測された。 The fact that the semiconductor device according to the present embodiment is effective for eliminating the SOIMOS substrate floating effect, and the recrystallization of the polycrystalline injected holes of the high-concentration p-type region 11 formed in contact with the oxide film 2 It was speculated that it would work well as a center.
本実施例に基づく半導体装置においては単結晶 S i層 3が 5 O nmと極めて薄く、 チャネル領域における基板不純物濃度も 1 X 1 0 ' Vcm3 と低く設定されている。 従って、 単結晶 S i層 3内の電荷量限定により閾電圧以上のゲート電圧条件では チャネル領域の単結晶 S i層 3に中性領域は存在せず、 完全空乏状態となってい る。 これは電流駆動源であるチヤネル中の可動電荷を効果的に誘起することがで き、 大電流化に適している。 即ち、 低電圧'高速動作に適している。 本実施例に 基づく半導体装置においては基板浮遊効果を伴うことなく上記の完全空乏型 S 0 I M O Sが従来半導体装置の製造方法のみで廉価に提供できることを示している。 図 2 6から図 2 8は本発明の他の実施例 (第 9実施例) による半導体装置の製 造工程順を示す断面図、 図 2 8はその完成断面図である。 前記実施例 8における 図 2 4の状態からゲー卜保護絶縁膜 7、 ゲート側壁絶縁膜 8、 及び素子間分離絶 縁膜 4をエッチングマスクとして露出している低濃度 n型拡散層 9及び 1 0表面 の選択的エッチングを施し、 5 nm厚の高濃度 p型領域 1 1を残置させた (図 2 6 ) o In the semiconductor device according to the present embodiment, the single-crystal Si layer 3 is as thin as 5 O nm, and the substrate impurity concentration in the channel region is set as low as 1 × 10 ′ Vcm 3 . Therefore, the neutral region does not exist in the single-crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single-crystal Si layer 3, and the single-crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is suitable for low voltage and high speed operation. This shows that in the semiconductor device according to the present embodiment, the above-described fully depleted S 0 IMOS can be provided at a low cost only by the conventional method of manufacturing a semiconductor device without causing a substrate floating effect. 26 to 28 are sectional views showing the order of manufacturing steps of a semiconductor device according to another embodiment (a ninth embodiment) of the present invention, and FIG. 28 is a completed sectional view thereof. The low-concentration n-type diffusion layers 9 and 10 exposed from the state of FIG. 24 in Example 8 using the gate protective insulating film 7, the gate sidewall insulating film 8, and the device isolation insulating film 4 as an etching mask. surface Was selectively etched to leave a high-concentration p-type region 11 with a thickness of 5 nm (Figure 26).
しかる後、 燐が高濃度に添加された S i膜 9 2、 及び 1 0 2を上記エッチング 領域に選択形成させ 6 5 0 °C 1 0分なる熱処理を施し、 活性化させた。 高濃度 n 型領域からなるソース及びドレインとした。 S i膜 9 2、 及び 1 0 2の形成は単 結晶ェピタキシャル選択成長の条件でも、 多結晶選択成長の条件の何れであつて も良い。 更に全面堆積された S i膜をパターニングして形成しても良い。 高濃度 n型化は S i膜形成時の不純物同時添加によらず、 例えばイオン注入に基づいて も良い。 上記 S i膜 9 2、 及び 1 0 2の形成はソース抵抗低減の観点から S i膜 9 2、 及び 1 0 2エッチング前の初期表面より上部に表面が構成されるごとく、 所謂積上げソース ' ドレイン構造とすること力望ましい (図 2 7 ) 。 Thereafter, Si films 92 and 102 to which phosphorus was added at a high concentration were selectively formed in the above-mentioned etching region, and a heat treatment at 65 ° C. for 10 minutes was performed to activate the film. The source and drain consisted of a high-concentration n-type region. The formation of the Si films 92 and 102 may be performed under any of the conditions of single crystal epitaxial selective growth and polycrystalline selective growth. Further, the Si film deposited on the entire surface may be formed by patterning. The high-concentration n-type may be based on, for example, ion implantation instead of the simultaneous addition of impurities when forming the Si film. The above-mentioned Si films 92 and 102 are formed from the viewpoint of reducing the source resistance as if the Si films 92 and 102 had a surface formed above the initial surface before etching. The structure is desirable (Fig. 27).
図 2 7の状態から前記実施例 8に従つて、 配線保護絶縁膜 1 3、 ソース電極 1 4、 ドレイン電極 1 5等を含む配線を形成した (図 2 8 ) 。 From the state shown in FIG. 27, a wiring including the wiring protection insulating film 13, the source electrode 14, the drain electrode 15 and the like was formed according to Example 8 (FIG. 28).
上記製造工程を経て製造された本実施例に基づく半導体装置においては、 前記 実施例 8に基づく半導体装置と同様に基板浮遊効果に伴う緒特性が観測されず、 正常な完全空乏型 S O I MO S特性を得ることができた。 本実施例に基づく半導 体装置においては、 高濃度 n型領域 9及び 1 0との界面部分を含めた高濃度 p型 領域 1 1が急峻な高濃度分布に構成できるため不純物濃度の増加に伴って低減さ れる再結合時間をより低減することができ、 従つて注入される正孔を前記実施例 1の半導体装置に比べて更に効率良く消滅できる。 即ち、 更に基板浮遊効果の解 消に有効となる。 In the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, no characteristic associated with the substrate floating effect was observed similarly to the semiconductor device according to the eighth embodiment, and a normal fully depleted SOI MOS characteristic was obtained. Could be obtained. In the semiconductor device according to the present embodiment, the high-concentration p-type region 11 including the interface with the high-concentration n-type regions 9 and 10 can be formed into a steep high-concentration distribution, so that the impurity concentration increases. As a result, the recombination time, which is reduced accordingly, can be further reduced, and the injected holes can be more efficiently eliminated as compared with the semiconductor device of the first embodiment. That is, it is more effective in eliminating the substrate floating effect.
図 2 9は本発明の他の実施例 (第 1 0 ) による半導体装置の完成断面図である。 前記実施例 9における図 2 6の状態からゲ一ト側壁絶縁膜 8に隣接する一部を除 いて高濃度 P型領域 1 1を選択除去してから前記実施例に従って S i膜 9 2、 及 び 1 0 2の選択形成以降の製造工程を実施し、 半導体装置を製造した。 上記高濃 度 P型領域 1 1の選択除去はホトマスクを用いて実施したが、 ゲート電極との自 己整合性を保証するために図 2 6の状態からゲ一ト側壁絶縁膜 8と異なる材料 (シリコン窒化膜) による第二の側壁絶縁膜の選択形成と上記第二の側壁絶縁膜 をマスクとする高濃度 P型領域 1 1の選択除去を施しても良い。 更に選択除去に 代えて、 第二の側壁絶縁膜をマスクとする高濃度 n型ィォン注入を高濃度 p型領 域 1 1へ施すことで高濃度 P型領域 1 1の低減を図っても良い (図 2 9 ) 。 FIG. 29 is a completed sectional view of a semiconductor device according to another embodiment (No. 10) of the present invention. From the state of FIG. 26 in the ninth embodiment, the high-concentration P-type region 11 is selectively removed except for a part adjacent to the gate sidewall insulating film 8, and then the Si film 92, and The semiconductor device was manufactured by performing the manufacturing steps after the selective formation of silicon and silicon. The selective removal of the high-concentration P-type region 11 was performed using a photomask, but a material different from the gate sidewall insulating film 8 from the state shown in Fig. 26 was used in order to guarantee self-alignment with the gate electrode. The second sidewall insulating film may be selectively formed by (silicon nitride film) and the high-concentration P-type region 11 may be selectively removed using the second sidewall insulating film as a mask. Further selective removal Alternatively, the high-concentration P-type region 11 may be reduced by performing high-concentration n-type ion implantation using the second sidewall insulating film as a mask on the high-concentration p-type region 11 (FIG. 29). ).
上記製造工程を経て製造された本実施例に基づく半導体装置におし、ては前記実 施例 8及び 9に基づく半導体装置と同様に基板浮遊効果に起因する緒現象は観測 されなかった。 更に、 本実施例に基づく半導体装置に於いては、 ソース領域と同 一形状に構成されるドレイン領域に関して、 ドレイン ·基板間容量が高濃度 p型 領域 1 1の選択除去面積に比例して低減され、 高濃度 p型領域 1 1の残領域が 0 . 1 以下では高濃度 p型領域 1 1の無い通常の S O I MO Sの容量と同等にま で低減された。 通常の S O I MO Sの容量値は通常半導体基板に製造された同一 寸法の MO Sにおける値の約 1 / 1 0であった。 In the semiconductor device according to the present embodiment manufactured through the above manufacturing process, no phenomenon caused by the substrate floating effect was observed as in the semiconductor devices according to Embodiments 8 and 9. Further, in the semiconductor device according to the present embodiment, the drain-substrate capacitance is reduced in proportion to the selective removal area of the high-concentration p-type region 11 with respect to the drain region having the same shape as the source region. When the remaining region of the high-concentration p-type region 11 was 0.1 or less, the capacitance was reduced to the same level as that of a normal SOI MOS without the high-concentration p-type region 11. The capacitance value of a normal SOI MOS is about 1/10 of the value of a MOS of the same size usually manufactured on a semiconductor substrate.
図 3 0は本発明の他の実施例 (第 1 1 ) による半導体装置の完成断面図である。 前記実施例 8において、 単結晶 S i層 3として 2 0 O nmの厚さの S O I基板を用 い、 高濃度 P型領域 1 1の形成領域を酸化膜 2界面から離し、 低濃度 n型拡散層 9及び 1 0を介して酸化膜 2に隣接するごとく構成した。 即ち、 高濃度 p型領域 1 1は低濃度 n型拡散層 9及び 1 0により外部から完全に隔離された構造とした。 その他の製造工程は実施例 1に従った (図 3 0 )。 FIG. 30 is a completed sectional view of a semiconductor device according to another embodiment (11) of the present invention. In Example 8, an SOI substrate having a thickness of 20 O nm was used as the single-crystal Si layer 3, the formation region of the high-concentration P-type region 11 was separated from the oxide film 2 interface, and the low-concentration n-type diffusion It was configured to be adjacent to the oxide film 2 via the layers 9 and 10. That is, the high-concentration p-type region 11 has a structure completely isolated from the outside by the low-concentration n-type diffusion layers 9 and 10. Other manufacturing steps were in accordance with Example 1 (FIG. 30).
上記の製造工程を経て製造された本実施例による半導体装置はドレイン領域に おいて高濃度 P型領域 1 1の存在にも係らず、 従来 S O I * MO Sと同等の低接 合容量値を示し、 通常半導体基板に製造された同一寸法の MO Sにおける値の約 1 / 1 0と小さなものであった。 上記結果はドレイン内の高濃度 p型領域 1 1は 低濃度 n型拡散層 1 0により回りを囲まれており、 高濃度 p型領域 1 1にはドレ ィン電界による空乏層が形成されず、 容量は厚 、酸化膜 2により決定されるため と考えられる。 The semiconductor device according to the present embodiment manufactured through the above manufacturing process exhibits a low junction capacitance value equivalent to that of the conventional SOI * MOS, despite the presence of the high-concentration P-type region 11 in the drain region. However, the value was as small as about 1/10 of the value of MOS of the same size usually manufactured on a semiconductor substrate. The above result shows that the high-concentration p-type region 11 in the drain is surrounded by the low-concentration n-type diffusion layer 10, and no depletion layer is formed in the high-concentration p-type region 11 by the drain electric field. It is considered that the capacitance is determined by the thickness and the oxide film 2.
図 3 1は本発明の他の実施例 (第 1 2 ) による半導体装置の完成断面図である。 前記実施例 8において、 単結晶 S i層 3として 5 0 0 nmの厚さの S 0 I基板を用 い、 素子間分離絶縁膜 4の形成により単結晶 S i膜 3の活性領域を互いに分離し た後、 所望の回路構成に従つて該活性領域の一部にィォン注入を施して低濃度 n 型領域 3 1とした。 低濃度 n型領域 3 1と低濃度 p型領域 3上に前記実施例 8に 従って、 ゲート酸化膜 5、 ゲート電極 6 1、 ゲート保護絶縁膜 7、 ゲート側壁絶 縁膜 8を形成した。 本実施例においてはゲート電極 6 1として W膜を用いた。 ゲ ート電極 6 1、 ゲート側壁絶縁膜 8をマスクとして低濃度 n型領域 3 1にのみ選 択的に Bのイオン注入とその後の熱処理により 2 0 O nmの接合深さを有し、 表面 不純物濃度が 5 X 1 0 1 8/cm3 の低濃度 p型拡散層 9 0及び 1 0 0を形成した。 続いて活性領域内部で最大不純物濃度が 5 X 1 0 2 cm3 で、 低濃度 p型拡散層 9 0及び 1 0 0内部に位置するごとく A sのイオン注入を施して高濃度 n型領域 1 1 0を形成した。 上記高濃度 n型領域 1 1 0は活性化熱処理後も固溶限界以上 の不純物濃度のため単結晶化されず、 多結晶状の結晶欠陥領域が保持されていた。 低濃度 P型領域 3に対してはゲート電極 6 1、 ゲート側壁絶縁膜 8をマスクとし て選択的に Pのイオン注入を行ない、 表面で最大濃度 3 X 1 0 1 8/cm3 となり 2 0 O nmの接合深さを有する低濃度 n型拡散層を形成し、 ソース 9、 及びドレイン 1 0とした。 続いて、 活性領域内部で最大不純物濃度が 5 X 1 0 2 0/cm3 で、 低 濃度 n型拡散層 9及び 1 0内部に位置するごとく B F 2 のイオン注入を施して高 濃度 P型領域 1 1を形成した。 上記高濃度 p型領域 1 1も活性化熱処理後も固溶 限界以上の不純物濃度のため単結晶化されず、 多結晶状の結晶欠陥領域が保持さ れていた。 しかる後、 前記実施例 8に従い配線保護絶縁膜 1 3の堆積と所望個所 への開口、 配線用金属膜の蒸着とそのパターニングにより接地電位線 1 7、 出力 端子 1 8、 及び電源電圧線 1 9を含む配線を形成した。 FIG. 31 is a completed sectional view of a semiconductor device according to another embodiment (No. 12) of the present invention. In Example 8, an S 0 I substrate having a thickness of 500 nm was used as the single crystal Si layer 3, and the active regions of the single crystal Si film 3 were separated from each other by forming an element isolation insulating film 4. After that, ion implantation was performed on a part of the active region according to a desired circuit configuration to obtain a low-concentration n-type region 31. According to Example 8, the gate oxide film 5, the gate electrode 61, the gate protection insulating film 7, and the gate sidewalls were formed on the low-concentration n-type region 31 and the low-concentration p-type region 3. An edge film 8 was formed. In the present embodiment, a W film was used as the gate electrode 61. Using the gate electrode 61 and the gate side wall insulating film 8 as a mask, only the low-concentration n-type region 31 is selectively ion-implanted with B and then subjected to a heat treatment to have a junction depth of 20 O nm. Low-concentration p-type diffusion layers 90 and 100 having an impurity concentration of 5 × 10 18 / cm 3 were formed. Subsequently, As is ion-implanted so that the maximum impurity concentration is 5 × 10 2 cm 3 inside the active region and the low-concentration p-type diffusion layers 90 and 100 are located inside the high-concentration n-type region 1. 10 were formed. The high-concentration n-type region 110 was not single-crystallized even after the activation heat treatment due to the impurity concentration higher than the solid solution limit, and a polycrystalline crystal defect region was retained. P ions are selectively implanted into the low-concentration P-type region 3 using the gate electrode 61 and the gate sidewall insulating film 8 as a mask, and the maximum concentration is 3 × 10 18 / cm 3 at the surface , and 20 A low-concentration n-type diffusion layer having a junction depth of O nm was formed, and was used as a source 9 and a drain 10. Subsequently, ion implantation of BF 2 is performed so that the maximum impurity concentration is 5 × 10 20 / cm 3 inside the active region and is located inside the low-concentration n-type diffusion layers 9 and 10. 1 formed 1 Even after the activation heat treatment, the high-concentration p-type region 11 was not single-crystallized due to the impurity concentration exceeding the solid solution limit, and a polycrystalline crystal defect region was retained. Thereafter, according to the eighth embodiment, the ground potential line 17, the output terminal 18 and the power supply voltage line 19 are formed by depositing the wiring protective insulating film 13 and opening it at a desired location, depositing a wiring metal film and patterning the same. Was formed.
上記製造工程を経て製造された本実施例に基づく半導体装置、 C MO S、 にお いて p MO S、 n MO Sの何れに関しても基板浮遊効果に起因する緒症状を観測 することができなかった。 更に n MO S閾電圧値の負方向変動、 p MO S閾電圧 値の正方向変動によって接地電位線 1 7と電源電圧線 1 9間に生じる S O I · C MO S特有の基板浮遊効果に基づく貫通電流も観測されなかった。 p MO Sにお いて、 基板浮遊効果が見られなかったことはチャネル部単結晶 S i膜 3 1に発生 した少数キヤリァである電子が高濃度 n型領域 1 1 0に向かい注入され、 結晶欠 陥に基づく再結合中心により消滅するためと考えられる。 In the semiconductor device according to the present example manufactured through the above-described manufacturing process, and in the CMOS, the pMOS and the nMOS could not observe any symptoms caused by the substrate floating effect. . Furthermore, SOI · CMOS penetrates due to the substrate floating effect peculiar to SOI and CMOS caused by the negative fluctuation of the nMOS threshold voltage and the positive fluctuation of the pMOS threshold voltage. No current was observed. The absence of the substrate floating effect in p MOS indicates that electrons, which are a small number of carriers generated in the channel single crystal Si film 31, were injected toward the high-concentration n-type region 110, and the crystal defect occurred. It is thought that it disappears due to recombination center based on the pit.
更に本実施例に基づく半導体装置においては P MO Sにおけるドレイン基板間 容量が従来 S O I · MO Sと同等の容量値を示し、 通常半導体基板に製造された 同一寸法の M O Sにおける値の約 1 Z 1 0の小さなものであった。 上記結果はド レイン内の高濃度 n型領域 1 1 0が p型低濃度拡散層 1 0 0により周りを囲まれ ており、 高濃度 n型領域 1 1 0にはドレイン電界による空乏層が形成されず、 容 量は厚い絶縁膜 2により決定されるためと考えられる。 Further, in the semiconductor device according to this embodiment, the capacitance between the drain substrate in the PMOS shows a capacitance value equivalent to that of the conventional SOI · MOS, and is approximately 1 Z 1 of the value of the MOS of the same size usually manufactured on the semiconductor substrate. It was a small thing of 0. The above result is The high-concentration n-type region 110 in the rain is surrounded by the p-type low-concentration diffusion layer 100, and no depletion layer is formed in the high-concentration n-type region 110 due to the drain electric field. It is considered that the amount is determined by the thick insulating film 2.
本実施例に基づく半導体装置においては単結晶 S i層 3及び 3 1が 5 0 O nmと 比較的厚く、 チャネル領域下部における基板領域では閾電圧以上のゲート電圧印 加によっても空乏層と中性領域が存在する所謂部分空乏化構造となる。 部分空乏 化構造は低電圧 ·高速動作で完全空乏構造に比べてやや落ちるが製造条件におい て従来の半導体基板を用 、た条件で容易に製造できる。 本実施例に基づく半導体 装置において廉価に部分空乏化構造 MO Sの基板浮遊対策を提供できることを示 している。 In the semiconductor device according to the present embodiment, the single-crystal Si layers 3 and 31 are relatively thick at 50 O nm, and in the substrate region below the channel region, the depletion layer and the neutral region are applied even when a gate voltage higher than the threshold voltage is applied. A so-called partially depleted structure in which a region exists is obtained. Although the partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, it can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. This shows that the semiconductor device according to the present embodiment can provide a countermeasure against the substrate floating of the partially depleted structure MOS at low cost.
図 3 2は本発明の他の実施例 (第 1 3 ) による半導体装置の製造工程を示す断 面図、 図 3 3はその完成断面図である。 前記実施例 1 2において、 単結晶 S i膜 3の膜厚は 2 0 0 とし、 酸化膜 2と 2 0 0 nm厚の単結晶 S i層 3の間に 1 0 0 nm厚の高抵抗多結晶 S i膜 2 0と 1 0 nm厚のシリコン酸化膜 2 1が構成された多 層構造 S O I基板を用いた。 ここで単結晶 S i膜 3の膜厚は 1 0 O nmとして、 p 型低濃度活性領域 3及び n型低濃度活性領域 3 1の不純物濃度は各々 1 X 1 0 1 6 ノ cm3 と極めて低く設定した。 即ち、 本実施例においては完全空乏型の相補型 M O S電界効果トランジスタを製造した。 本実施例においてはゲート酸化膜 5、 ゲ ート電極 6、 ゲート保護絶縁膜 7、 ゲート側壁絶縁膜 8の形成に先立って n MO Sのゲート電極 7形成予定領域直下の高抵抗多結晶 S i膜 2 0に不純物濃度 2 X 1 0 1 cm3 なる高濃度 p型不純物領域 2 2を、 p MO Sのゲート電極 7形成予 定領域直下の高抵抗多結晶 S i膜 2 0に不純物濃度 2 X 1 0 1 8/cm3 なる高濃度 n型不純物領域 2 3を予め形成した (図 3 2 ) 。 FIG. 32 is a sectional view showing a manufacturing process of a semiconductor device according to another embodiment (the thirteenth) of the present invention, and FIG. 33 is a completed sectional view thereof. In Example 12 described above, the thickness of the single-crystal Si film 3 was set to 200, and the high-resistance multilayer having a thickness of 100 nm was formed between the oxide film 2 and the single-crystal Si layer 3 having a thickness of 200 nm. A multi-layer SOI substrate having a crystalline Si film 20 and a 10 nm thick silicon oxide film 21 was used. Here, the thickness of the single-crystal Si film 3 is 10 O nm, and the impurity concentration of the p-type low concentration active region 3 and the n-type low concentration active region 31 is 1 × 10 16 cm 3 , respectively. Set low. That is, in this example, a completely depleted complementary MOS field effect transistor was manufactured. In this embodiment, prior to the formation of the gate oxide film 5, the gate electrode 6, the gate protection insulating film 7, and the gate sidewall insulating film 8, the high-resistance polycrystalline Si directly under the region where the nMOS gate electrode 7 is to be formed is formed. A high-concentration p-type impurity region 22 having an impurity concentration of 2 × 10 1 cm 3 is formed in the film 20, and an impurity concentration 2 is formed in the high-resistance polycrystalline Si film 20 immediately below the region where the p-MOS gate electrode 7 is to be formed. the X 1 0 1 8 / cm 3 comprising a high concentration n-type impurity region 2 3 preformed (FIG. 2).
図 3 2の状態から前記実施例 1 2に従って半導体装置を製造したが、 n MO S のソース内に形成する高濃度 p型領域 1 1の上部に高濃度 n型領域 9 1、 及び p MO Sにおけるソース内の高濃度 n型領域 1 1 0の上部に高濃度 p型領域 9 2を 各々形成した。 上記の各接合深さは前記実施例 1 1の条件と同一とした。 更にゲ ート、 ゲート側壁絶縁膜を除くソース、 ドレイン全表面に実施例 4に従って W膜 1 2を選択的に被着した (図 3 3 ) 。 上記製造工程を経て製造された本実施例に基づく半導体装置、 C MO S、 にお いて p MO S、 n MO Sの何れに関しても基板浮遊効果に起因する緒症状を観測 されないことは前記実施例 1 2と同様であつたが、 前記実施例 1 2の半導体装置 に比べて本実施例に基づく半導体装置のゲート、 ドレイン電圧 2 Vにおけるソ一 ス · ドレイン電流は p MO S、 n MO S共 1 . 4倍以上の大電流化が達成された c しかも低ドレイン電圧条件である非飽和特性領域において、 前記実施例 1 2の半 導体装置に比べて極めて急峻なドレインコンダクタンス特性が得られた。 本実施 例に基づく半導体装置により構成されたリングォシレータにおいては一段当りの 遅延時間は 1 2 p秒と前記実施例 5に基づくリングオシレー夕に比べて 6 p秒も 高速化が達成された。 これは高濃度 p型不純物領域 2 2、 及び高濃度 n型不純物 領域 2 3がパンチスルー防止機構として作用し、 且つ薄いシリコン酸化膜 2 1が 製造工程中の高温熱処理に対する不純物拡散マスクとして作用するため、 本実施 例に基づく半導体装置の超高速、 大電流特性はチャネルを構成する p型低濃度活 性領域 3、 及び n型低濃度活性領域 3 1の不純物濃度を極めて低濃度に設定でき、 不純物散乱による移動度の劣化が防止できた為と考えられる。 A semiconductor device was manufactured according to the embodiment 12 from the state of FIG. 32. However, a high-concentration n-type region 91 and a high-concentration n-type region 91 were formed above the high-concentration p-type region 11 formed in the source of nMOS. The high-concentration p-type regions 92 were respectively formed above the high-concentration n-type regions 110 in the source in FIG. Each of the above junction depths was the same as the condition in Example 11 described above. Further, a W film 12 was selectively deposited on the entire surface of the source and drain except for the gate and gate sidewall insulating films according to Example 4 (FIG. 33). In the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, and in the C MOS, neither the p MOS nor the n MON S observed any symptom caused by the substrate floating effect. 12 was the same as that of Example 12, except that the source and drain currents at a gate and drain voltage of 2 V of the semiconductor device according to the present example were both pMOS and nMOS compared to the semiconductor device of Example 12 above. 1. in the non-saturation characteristic region is four times more c moreover low drain voltage condition larger current is achieved, a very steep drain conductance characteristics were obtained in comparison with the semi-conductor device of example 1 2. In the ring oscillator constituted by the semiconductor device according to the present embodiment, the delay time per stage was 12 ps, which was 6 ps faster than the ring oscillator according to the fifth embodiment. This is because the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 acts as an impurity diffusion mask for high-temperature heat treatment during the manufacturing process. Therefore, the ultra-high-speed and large-current characteristics of the semiconductor device according to the present embodiment can be set to extremely low impurity concentrations in the p-type low-concentration active region 3 and the n-type low-concentration active region 31 forming the channel. It is considered that the mobility was prevented from deteriorating due to impurity scattering.
図 3 4から図 3 7は本発明の第 1 4の実施例による半導体装置の製造工程順を 示す断面図、 図 3 4はその完成断面図である。 直径 1 2 . 5 cmの単結晶 S iより なる支持基板 1上に 5 0 O nm厚のシリコン酸化膜 (単に酸化膜と称する) 2、 及 び 2 0 O nm厚の p導電型、 抵抗率 1 0 Q cm、 面方位 (1 0 0 ) の単結晶 S i層 3 からなる S 0 I基板に公知の M 0 S電界効果型トランジスタの製造方法により素 子間分離絶縁膜 4、 5 nm厚のゲート酸化膜 5、 n型低抵抗多結晶 S i膜よりなる ゲート電極 6、 ゲート保護絶縁膜 7を形成した。 なお、 ゲート酸化膜 5の形成に 先立って、 閾電圧値が 0 . 1 Vとなるごとく単結晶 S i層 3に Bのイオン注入を 施した。 ゲート長は 2 0 O nmである。 この状態よりゲート保護絶縁膜 7及びゲー ト電極 6を注入阻止マスクとして pのイオン注入とその後の熱処理を施し、 接合 端が酸化膜 2に達し、 且つ実効ゲート長が 1 5 O nm, 及び表面における最大不純 物濃度が 2 X I 0 , 6/cm3 となるごとく低濃度 n型拡散層によるソース 9及びド レイン 1 0を形成した。 引続いて加速エネルギー 2 5 kev、 ドーズ量 3 x 1 0 1 5 /cm2 の条件により A sのイオン注入とその後の熱処理を施し、 高濃度 n型拡散 層 9 1、 及び 1 0 1を形成した。 上記の各 n型不純物イオン注入では単結晶 S i 層 3は単結晶性を維持していた (図 3 5 ) 。 FIGS. 34 to 37 are sectional views showing the order of manufacturing steps of the semiconductor device according to the fourteenth embodiment of the present invention, and FIG. 34 is a completed sectional view thereof. A 50-nm-thick silicon oxide film (simply called an oxide film) 2 on a supporting substrate 1 made of single-crystal Si with a diameter of 12.5 cm 2, and a p-type conductivity with a thickness of 20-nm and resistivity An S0I substrate consisting of a single crystal Si layer 3 having a thickness of 100 Qcm and a plane orientation of (100) is formed on a S0I substrate by a known method of manufacturing an M0S field-effect transistor. A gate oxide film 5, a gate electrode 6 made of an n-type low-resistance polycrystalline Si film, and a gate protection insulating film 7 were formed. Prior to the formation of the gate oxide film 5, B ions were implanted into the single-crystal Si layer 3 so that the threshold voltage became 0.1 V. The gate length is 20 O nm. From this state, ion implantation of p and subsequent heat treatment are performed using the gate protection insulating film 7 and the gate electrode 6 as an implantation blocking mask, the junction ends reach the oxide film 2, the effective gate length is 15 O nm, and the surface is maximum impurity concentration formed a 2 XI 0, 6 / cm 3 and the source 9 and drain 1 0 by the low-concentration n-type diffusion layer as made in. Subsequently, under the conditions of an acceleration energy of 25 kev and a dose of 3 × 10 15 / cm 2 , ion implantation of As and subsequent heat treatment are performed, resulting in high-concentration n-type diffusion. Layers 91 and 101 were formed. In each of the above-described n-type impurity ion implantations, the single-crystal Si layer 3 maintained single-crystallinity (FIG. 35).
図 3 5の状態より、 5 0 nm厚の堆積性絶縁膜を全面に形成し、 異方性ドライエ ツチングによりゲート側壁部にのみ上記絶縁膜を選択的に残置させてゲート側壁 絶縁膜 8を形成した。 ゲ一ト側壁絶縁膜 8の膜厚条件に関して、 最小膜厚 2 0 nm、 最大膜厚 0 . 5 /imまで 1 0から 1 0 0腦間隔で変化させた本実施例に基づく半 導体装置も別途製造した。 引続き、 低濃度 n型ソース 9及びドレイン 1 0内部の 酸化膜 2界面で濃度が最大となるごとく ドーズ量 3 X 1 0 1 cm2 なる条件で S iのイオン注入を施した。 上記イオン注入の後、 8 0 0 °C、 1 0分の熱処理を 施した。 同一条件で別途製造した試料についてその断面を透過型電子顕微鏡によ り観察した結果、 酸化膜 2界面近傍では微細結晶粒の双晶、 即ち結晶欠陥領域 1 1が形成されていることが明らかとなった。 S iのイオン注入はその最大濃度が 酸化膜 2内になるごとく加速エネルギーを設定してもよい。 さらに別途実験の結 果、 結晶欠陥領域 1 1の形成の為のイオン注入はイオン種として S iでなくとも 良く、 N e、 A r等の希ガス元素、 F、 C 1等のハロゲン元素、 及び C、 G e等 の 1 4族元素であっても同等の効果が得られることが判明した。 しかしながら、 Pのごとく S i単結晶中で n伝導型を構成する元素のイオン注入では後述のごと く効果がないことが判明した。 (図 3 6 ) 。 From the state shown in Fig. 35, a 50-nm-thick deposited insulating film is formed on the entire surface, and the above-mentioned insulating film is selectively left only on the gate side wall by anisotropic dry etching to form a gate side wall insulating film 8. did. Regarding the thickness condition of the gate sidewall insulating film 8, a semiconductor device based on the present embodiment in which the minimum film thickness is changed to 20 nm and the maximum film thickness is set to 0.5 / im at intervals of 10 to 100 is also used. Manufactured separately. Subsequently, Si ions were implanted under the conditions of a dose of 3 × 10 1 cm 2 so that the concentration was maximized at the interface between the low concentration n-type source 9 and the oxide film 2 inside the drain 10. After the ion implantation, a heat treatment was performed at 800 ° C. for 10 minutes. Observation of the cross section of a sample separately manufactured under the same conditions with a transmission electron microscope revealed that twins of fine crystal grains, that is, crystal defect regions 11 were formed near the oxide film 2 interface. became. In the ion implantation of Si, the acceleration energy may be set so that the maximum concentration is in oxide film 2. Furthermore, as a result of another experiment, ion implantation for forming the crystal defect region 11 does not need to be Si as an ion species, and rare gas elements such as Ne and Ar, halogen elements such as F and C1, It has been found that the same effect can be obtained even with a group 14 element such as C and Ge. However, it has been found that the ion implantation of the n-conductivity-type element in the Si single crystal like P has no effect as described later. (Figure 36).
図 3 6の状態によりソース抵抗の低減を図る目的で化学気相反応による 1 5 0 nm厚の W膜 1 2を露出している S i面に選択的に堆積した。 上記 W膜 1 2はスパ ッタリングによる全面被着と、 少なくとも高濃度 n型領域 9 1及び 1 0 1表面を 覆うごとくパターニングして形成しても良い。 しかる後、 燐が添加されたシリコ ン酸化膜による配線保護絶縁膜 1 3の堆積を施した (図 3 7 ) 。 In order to reduce the source resistance in the state shown in FIG. 36, a 150 nm thick W film 12 was selectively deposited on the exposed Si surface by a chemical vapor reaction. The W film 12 may be formed by sputtering over the entire surface and by patterning so as to cover at least the high-concentration n-type regions 91 and 101 surfaces. Thereafter, a wiring protection insulating film 13 was deposited using a silicon oxide film to which phosphorus was added (FIG. 37).
図 3 7の状態により公知の半導体装置の製造方法に基づき配線保護絶縁膜 1 3 の所望個所への開口、 更には配線金属の蒸着とそのパターニングによるソース電 極 1 4、 ドレイン電極 1 5等を含む配線を形成した。 (図 3 4 ) 。 Based on the state shown in FIG. 37, openings are formed at desired locations in the wiring protection insulating film 13 based on a known method of manufacturing a semiconductor device, and further, a source electrode 14 and a drain electrode 15 are formed by depositing and patterning wiring metal. The wiring including was formed. (Figure 34).
上記製造工程を経て製造された本実施例に基づく半導体装置のソース · ドレイ ン間耐圧は 4 . 7 Vとソース内の p型拡散層 1 1が構成されていない同一寸法の 従来構造 S O I MO Sに比べて約 1 . 5 V向上し、 通常半導体基板に製造された 同一寸法の M O Sと同等の耐圧値を確保することができた。 また、 電流 ·電圧特 性においてもキンク特性と称される異常なこぶ状特性は観測されず、 正常な特性 を示した。 更に、 ソースドレイン電流 ·ゲート電圧特性において、 従来 S O I · MO Sで観測された低ゲート電圧におけるリーク電流の存在も観測されなかつた。 また上記リーク電流、 及び閾電圧値はドレイン電圧を変化させても変化が見出せ なかった。 これらの特性から、 本実施例に基づく半導体装置では基板浮遊効果に 伴う緒特性から完全に解消されたことが明らかとなつた。 本実施例に基づく半導 体装置の電流 ·電圧特性は正常な特性を示し、 ソース、 及びドレイン内部に形成 された結晶欠陥領域 1 1は何ら悪影響を及ぼさないことも判明した。 本実施例に 基づく上記基板浮遊効果の解消は本実施例に基づきゲー卜側壁絶縁膜 8の膜厚を 変化させて製造した半導体装置において、 結晶欠陥領域 1 1端から低濃度 n型拡 散層接合までの間隔が 5 0 O nmJ¾下から 2 0 nmまで観測された力 特に 1 0 0 nm 以下の間隔の場合が特性のばらつきも観測されず、 最も好ましいことが明らかと なった。 なお、 結晶欠陥領域 1 1形成を Pのイオン注入に基づいて形成した試料 においては基板浮遊効果解消が全く見出せなかった。 The semiconductor device according to the present embodiment manufactured through the above manufacturing process has a source-drain withstand voltage of 4.7 V and a conventional structure SOI MOS having the same dimensions without the p-type diffusion layer 11 in the source. About 1.5 V higher than normal, usually manufactured on semiconductor substrates A breakdown voltage equivalent to that of MOS of the same dimensions could be secured. Also, in the current-voltage characteristics, no abnormal bump-like characteristics called kink characteristics were observed, and the characteristics were normal. Furthermore, in the source-drain current-gate voltage characteristics, the existence of a leak current at a low gate voltage conventionally observed in SOI / MOS was not observed. Also, no change was found in the leakage current and the threshold voltage even when the drain voltage was changed. From these characteristics, it became clear that the semiconductor device according to the present example was completely eliminated from the characteristics associated with the substrate floating effect. The current-voltage characteristics of the semiconductor device according to this example showed normal characteristics, and it was also found that the crystal defect region 11 formed inside the source and the drain had no adverse effect. The elimination of the above substrate floating effect according to the present embodiment is achieved by changing the thickness of the gate side wall insulating film 8 according to the present embodiment and changing the thickness of the gate defect side wall insulating film 8 from the end of the crystal defect region 11 to the low concentration n-type diffusion layer. The force observed from the point where the distance to the junction is below 50 O nmJ¾ to 20 nm, especially when the distance is less than 100 nm, no variation in the characteristics was observed. In the sample in which the crystal defect region 11 was formed based on the P ion implantation, no elimination of the substrate floating effect was found.
本実施例に基づく半導体装置が S 0 I MO Sの基板浮遊効果解消に有効なこと 力、ら酸化膜 2に接して構成された結晶欠陥領域 1 1の多結晶が注入された正孔の 再結晶中心として十分に作用することが推測された。 The fact that the semiconductor device according to the present embodiment is effective in eliminating the substrate floating effect of SOIMOS is effective in regenerating holes into which polycrystals in the crystal defect region 11 formed in contact with the oxide film 2 are injected. It was presumed that it sufficiently acted as a crystal center.
本実施例に基づく半導体装置においては単結晶 S i層 3が 2 0 O nmと比較的厚 く、 チャネル領域下部における基板領域では閾電圧以上のゲ一ト電圧印加によつ ても空乏層と中性領域が存在する所謂部分空乏化構造となる。 部分空乏化構造は 低電圧 ·高速動作で完全空乏構造に比べてやや落ちるが製造条件において従来の 半導体基板を用 、た条件で容易に製造できる。 本実施例に基づく半導体装置にお I、ては廉価に部分空乏化構造 M 0 S基板浮遊対策を提供できることを示している。 図 3 8から図 4 0は本発明の他の実施例による半導体装置の製造工程順を示す 断面図、 図 4 0はその完全断面図である。 前記実施例 1 4に基づいて素子間分離 絶縁膜 4、 ゲ一ト酸化膜 5、 ゲート電極 6、 ゲ一ト保護絶縁膜 7の形成まで行な つた状態から上記ゲート保護絶縁膜 7をマスクとする A sの 2 keVなる低加速ェ ネルギ一で高濃度イオン注入を行ない、 接合深さ 1 O nm、 表面不純物濃度 1 X 1 0 2 , /cm3 なる浅接合 n型高濃度拡散層 9 5、 及び 1 0 5を形成した。 続いて、 前記実施例 1に従い 1 0 0 nm厚のゲ一ト側壁絶縁膜 8を形成し、 ゲート側壁絶縁 膜 8をマスクとする Pの高濃度イオン注入を行ない、 接合深さ 1 0 O nmの低抵抗 ソース拡散層 9 1、 及び低抵抗ドレイン拡散層 1 0 1を形成した (図 3 8 ) 。 図 3 8の状態より公知の半導体装置製造方法に基づき、 配線保護絶縁膜 1 3の 堆積と、 所望個所への開口を施した。 上記開口より Pのイオン注入を施し、 低抵 抗ソース拡散層 9 1下部に接し、 下地酸化膜 2に達するごとく最小不純物濃度 1 X 1 0 1 6/cm3 なる n型低濃度拡散層 9を形成した。 上記工程において、 ドレイ ン拡散層底部にも n型低濃度拡散層 1 0が同時に形成される。 上記 n型低濃度拡 散層 9及び 1 0の活性化と拡散深さを調整する熱処理を施した後、 前記開口側面 に配線保護絶縁膜 1 3と材料を異にする堆積膜 1 3 1をドライエッチングの手法 を用いて選択的に残置させた。 開口側壁絶縁膜 1 3 1は開口から一定膜厚でィォ ン注入マスクとなる膜が側壁部に存在することに意味があり、 開口底面部に存在 していても次工程には何ら問題は生じない。 この状態から開口側壁膜 1 3 1をマ スクとする S iの高エネルギーイオン注入を前記実施例 1 4の条件に基づき実施 し、 結晶欠陥領域 1 1を上記 n型低濃度拡散層 9内の酸化膜 2界面近傍に形成し た。 上記製造工程において、 結晶欠陥領域 1 1端から n型低濃度拡散層 9接合ま での間隔の調整に開口側壁膜 1 3 1を用いる手法について説明したが、 上記は n 型低濃度拡散層接合深さを熱処理により調整する手法に基づレ、ても良い。 この場 合、 開口側壁膜 1 3 1の形成工程は省略できる (図 3 9 )。 In the semiconductor device according to the present embodiment, the single-crystal Si layer 3 is relatively thick at 20 O nm, and in the substrate region below the channel region, even if a gate voltage equal to or higher than the threshold voltage is applied, the single-crystal Si layer 3 becomes in a depletion layer. A so-called partially depleted structure having a neutral region is obtained. Although the partially depleted structure is slightly lower than the fully depleted structure at low voltage and high speed operation, it can be easily manufactured under conventional manufacturing conditions using a conventional semiconductor substrate. This shows that the semiconductor device according to the present embodiment can provide a countermeasure against floating of the partially depleted structure MS substrate at low cost. 38 to 40 are sectional views showing the order of manufacturing steps of a semiconductor device according to another embodiment of the present invention, and FIG. 40 is a complete sectional view thereof. The gate protection insulating film 7 is used as a mask from the state in which the steps up to the formation of the element isolation insulating film 4, the gate oxide film 5, the gate electrode 6, and the gate protection insulating film 7 based on the embodiment 14 are performed. High-concentration ion implantation with low acceleration energy of 2 keV of As, junction depth 1 O nm, surface impurity concentration 1 X Shallow junction n-type high concentration diffusion layers 95 and 105 of 102 and / cm 3 were formed. Subsequently, a gate sidewall insulating film 8 having a thickness of 100 nm is formed in accordance with the first embodiment, and high-concentration ion implantation of P is performed using the gate sidewall insulating film 8 as a mask, and a junction depth of 10 O nm The low resistance source diffusion layer 91 and the low resistance drain diffusion layer 101 were formed (FIG. 38). Based on a known semiconductor device manufacturing method from the state shown in FIG. 38, a wiring protection insulating film 13 was deposited and an opening was formed at a desired location. P ions are implanted from the above opening, and the n-type low concentration diffusion layer 9 having a minimum impurity concentration of 1 × 10 16 / cm 3 contacting the lower part of the low resistance source diffusion layer 9 1 and reaching the base oxide film 2 is formed. Formed. In the above step, the n-type low concentration diffusion layer 10 is simultaneously formed also at the bottom of the drain diffusion layer. After performing a heat treatment for adjusting the activation and diffusion depth of the n-type low concentration diffusion layers 9 and 10, a deposition film 13 1 made of a different material from the wiring protection insulating film 13 is formed on the side surface of the opening. It was selectively left using the dry etching technique. The opening side wall insulating film 13 1 has a meaning that a film serving as an ion implantation mask has a constant thickness from the opening on the side wall, and even if it exists on the bottom of the opening, there is no problem in the next step. Does not occur. From this state, high-energy ion implantation of Si using the opening side wall film 13 1 as a mask was performed based on the conditions of the above Example 14, and the crystal defect region 11 was formed in the n-type low concentration diffusion layer 9. An oxide film was formed near the interface between the two. In the above manufacturing process, the method of using the open side wall film 13 1 to adjust the distance from the end of the crystal defect region 11 to the junction of the n-type low concentration diffusion layer 9 has been described. It may be based on a method of adjusting the depth by heat treatment. In this case, the step of forming the side wall film 131 can be omitted (FIG. 39).
図 3 9の状態より公知の半導体装置の製造方法に基づき、 配線金属材料による ソース電極 1 4、 ドレイン電極 1 5を含む電極、 及び配線を形成した。 図 4 0に おいては開口側壁膜 1 3 1を除去してから配線材料を形成する工程に従った半導 体装置断面図を図示したが、 上記開口側壁膜は所望により残置させたままであつ ても何ら問題は生じない (図 4 0 ) 。 From the state shown in FIG. 39, an electrode including a source electrode 14 and a drain electrode 15 and a wiring were formed by a wiring metal material based on a known method for manufacturing a semiconductor device. In FIG. 40, a cross-sectional view of a semiconductor device according to a step of forming a wiring material after removing the opening side wall film 131 is illustrated, but the opening side wall film is left as desired. This does not cause any problems (Fig. 40).
上記製造工程を経て製造された本実施例に基づく半導体装置においては、 前記 実施例 1 4に基づく半導体装置と同様に基板浮遊効果に伴う緒特性が観測されず、 正常な部分空乏型 S O I M O S特性を得ることができた。 更に、 本実施例に基づ く半導体装置においては基板浮遊効果解消がコンタクト孔領域だけで実現できる ため、 トランジスタ特性を決定するゲート電極端近傍におけるソース、 ドレイン 拡散層形状に何らの制約を生じない。 従って、 本実施例に基づけば所望のトラン ジスタ特性を基板浮遊効果の影響なく実現することができる。 In the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, similar to the semiconductor device according to Embodiment 14 described above, no characteristic associated with the substrate floating effect is observed, and a normal partially depleted SOIMOS characteristic is obtained. I got it. Further, in the semiconductor device according to the present embodiment, the elimination of the substrate floating effect can be realized only in the contact hole region. Therefore, there is no restriction on the shape of the source / drain diffusion layer near the gate electrode end that determines the transistor characteristics. Therefore, according to this embodiment, desired transistor characteristics can be realized without the influence of the substrate floating effect.
図 4 1は本発明の他の実施例 (第 1 6 ) による半導体装置の製造工程を示す断 面図、 図 4 2はその完成断面図である。 前記実施例 1 4において、 単結晶 S i層 3として 5 0 O nmの厚さの S O I基板を用い、 素子間分離絶縁膜 4の形成により 単結晶 S i膜 3の活性領域を互いに分離した後、 所望の回路構成に従って該活性 領域の一部にイオン注入を施して低濃度 n型領域 3 1とした。 低濃度 n型領域 3 1と低濃度 P型領域 3上に前記実施例 1 4に従って、 ゲ一卜酸化膜 5、 ゲート電 極 6 1、 ゲ一ト保護絶縁膜 7を形成した。 本実施例においてはゲート電極 6 1と して W膜を用いた。 この状態より前記実施例 1 4に従い、 ゲート電極 6 1をマス クとしたイオン注入を施した。 上記イオン注入は低濃度 P型領域 3では前記実施 例に従い低濃度 n型拡散層 9、 及び 1 0の形成のための Pイオン注入と、 高濃度 n型拡散層 9 K 1 0 1の形成のための A sイオン注入を行ない、 低濃度 n型領 域 3 1では Bイオン注入による低濃度 p型拡散層 9 0、 及び 1 0 0の形成と高濃 度 P型拡散層 9 2及び 1 0 2を形成した。 低濃度 n型拡散層 9、 及び 1 0と低濃 度 P型拡散層 9 0、 及び 1 0 0は酸化膜 2に達するごとく形成し、 最大不純物濃 度は最終的に 1 X 1 0 ' Vcm3 になるごとく設定した。 上記の各イオン注入とそ の後の活性化熱処理の後、 n MO S形成領域にはシリコン窒化膜を選択残置し、 異方性ドライエッチングにより、 n MO Sゲートの側壁部にのみシリコン窒化膜 による 1 0 O nm厚のゲート側壁絶縁膜 8を形成した。 同様に p MO S形成領域に は酸化膜を選択残置し、 異方性ドライエッチングにより、 p MO Sゲートの側壁 部にのみ酸化膜による 2 0 O nm厚のゲート側壁絶縁膜 8を形成した。 上記 p MO S、 及び n MO Sのゲート側壁絶縁膜 8は所望により同一材料による同一膜厚で あっても良い。 しかる後、 ゲート電極 6 1、 ゲート側壁絶縁膜 8をマスクとして 酸化膜 2界面に達するごとく ドーズ量 5 X 1 0 l 5/cm3 の A rをイオン注入し、 低濃度 n型拡散層 9、 及び 1 0と低濃度 p型拡散層 9 0、 及び 1 0 0内部で高濃 度 n型拡散層 9 1、 1 0 1及び高濃度 p型拡散層 9 2及び 1 0 2から隔離された 領域に結晶欠陥領域 1 1を埋込形成した。 結晶欠陥領域 1 1の形成は A rのィォ ン注入によらず、 前記実施例 1のごとく S iのイオン注入等によっても何ら差し 支えない。 上記結晶欠陥領域 1 1は下地酸化膜 2の影響により再結晶化熱処理に よっても下地酸化膜 2に接する領域は単結晶化されず、 多結晶状の結晶欠陥領域 が保持されていた (図 4 1)。 FIG. 41 is a sectional view showing a semiconductor device manufacturing process according to another embodiment (16th) of the present invention, and FIG. 42 is a completed sectional view thereof. In Example 14, after the active region of the single-crystal Si film 3 was separated from each other by forming an element isolation insulating film 4 using a SOI substrate having a thickness of 50 O nm as the single-crystal Si layer 3 According to a desired circuit configuration, a part of the active region was ion-implanted to form a low-concentration n-type region 31. A gate oxide film 5, a gate electrode 61, and a gate protection insulating film 7 were formed on the low-concentration n-type region 31 and the low-concentration P-type region 3 according to Example 14 described above. In the present embodiment, a W film was used as the gate electrode 61. From this state, ion implantation was performed using the gate electrode 61 as a mask according to Example 14 described above. The above-described ion implantation is performed in the low-concentration P-type region 3 in accordance with the above-described embodiment, by performing P ion implantation for forming the low-concentration n-type diffusion layers 9 and 10 and forming the high-concentration n-type diffusion layer 9K101. In the low-concentration n-type region 31, low-concentration p-type diffusion layers 90 and 100 are formed by B ion implantation, and high-concentration P-type diffusion layers 92 and 100 are formed. Formed two. The low-concentration n-type diffusion layers 9 and 10 and the low-concentration P-type diffusion layers 90 and 100 are formed so as to reach the oxide film 2, and the maximum impurity concentration is finally 1 × 10 ′ Vcm. Set as 3 After each of the above ion implantations and subsequent activation heat treatment, the silicon nitride film is selectively left in the nMOS formation region, and the silicon nitride film is formed only on the side wall of the nMOS gate by anisotropic dry etching. To form a gate sidewall insulating film 8 having a thickness of 10 O nm. Similarly, an oxide film was selectively left in the pMOS formation region, and a 20-nm-thick gate sidewall insulating film 8 of an oxide film was formed only on the sidewall of the pMOS gate by anisotropic dry etching. The gate sidewall insulating films 8 of the pMOS and the nMOS may have the same thickness and the same material, if desired. Thereafter, using the gate electrode 6 1 and the gate sidewall insulating film 8 as a mask, Ar ions are implanted at a dose of 5 × 10 15 / cm 3 so as to reach the oxide film 2 interface, and the low-concentration n-type diffusion layer 9 is formed. And 10 and low-concentration p-type diffusion layers 90 and 100, and regions isolated from high-concentration n-type diffusion layers 91 and 101 and high-concentration p-type diffusion layers 92 and 102 inside A crystal defect region 11 was buried therein. The formation of the crystal defect region 11 is due to Ar Irrespective of the ion implantation, the ion implantation of Si or the like as in the first embodiment may be used. Due to the influence of the base oxide film 2, the crystal defect region 11 was not monocrystallized in the region in contact with the base oxide film 2 even by the recrystallization heat treatment, and a polycrystalline crystal defect region was retained (FIG. 4). 1).
しかる後、 前記実施例 1 4に従い配線保護絶縁膜 1 3の堆積と所望個所への開 口、 配線用金属膜の蒸着とそのパターニングにより接地電位線 1 7、 出力端子 1 8、 及び電源電圧線 1 9を含む配線を形成した (図 42 )。 Thereafter, according to the embodiment 14, the wiring protection insulating film 13 is deposited and opened at a desired location, and the wiring metal film is deposited and patterned to form a ground potential line 17, an output terminal 18 and a power supply voltage line. A wiring including 19 was formed (FIG. 42).
上記製造工程を経て製造された本実施例に基づく半導体装置、 CMOS、 にお いて pMOS、 nMOSの何れに関しても基板浮遊効果に起因する緒症状を観測 することができなかった。 更に nMOS閾電圧値の負方向変動、 pMOS閾電圧 値の正方向変動によって接地電位線 1 7と電源電圧線 1 9間に生じる SO I · C MO S特有の基板浮遊効果に基づく貫通電流も観測されなかった。 pMO Sにお いて、 基板浮遊効果が見られなかったことはチヤネノレ部単結晶 S i膜 3 1に発生 した少数キャリアである電子が低濃度 P型拡散層 1 00内を移動し、 結晶欠陥領 域 1 1内の再結合中心により消滅するためと考えられる。 本実施例に基づく半導 体装置においては、 nMOSと pMOSの基板浮遊効果を同一のィォン注入工程 で解消することができ、 製造工程を複雑にすることなく、 従って廉価に CMOS の高性能化を達成することができた。 In the semiconductor device, the CMOS, and the pMOS and the nMOS according to the present embodiment manufactured through the above manufacturing process, no symptom caused by the substrate floating effect could be observed in any of the pMOS and the nMOS. Furthermore, a through current caused by the substrate floating effect peculiar to SO ICMOS generated between the ground potential line 17 and the power supply voltage line 19 due to the negative fluctuation of the nMOS threshold voltage and the positive fluctuation of the pMOS threshold voltage is also observed. Was not done. The absence of the substrate floating effect in pMOS indicates that electrons, which are minority carriers generated in the single-crystal Si film 31 of the channel portion, move within the low-concentration P-type diffusion layer 100, and the crystal defect region It is considered that this disappears due to the recombination center in region 11. In the semiconductor device according to the present embodiment, the substrate floating effect of the nMOS and the pMOS can be eliminated by the same ion implantation process, and the performance of the CMOS can be improved at low cost without complicating the manufacturing process. Could be achieved.
本実施例において、 再結晶中心として作用する結晶欠陥領域 1 1とチャネル下 部の低濃度!)型領域 3、 或いは低濃度 n型領域 3 1間間隔を決定するためのゲー ト側壁絶縁膜 8の膜厚を n M 0 Sと p M 0 Sで異なるごとく形成した。 上記は低 濃度 P型拡散層 90及び 1 00と、 低濃度 n型拡散層 9及び 1 0の不純物の違い により接合深さが異なる点を考慮し、 その捕正を目的とするものである。 In this embodiment, the crystal defect region 11 acting as a recrystallization center and the low concentration in the lower part of the channel! ) Type gate region 3 or gate sidewall insulating film 8 for determining the distance between low-concentration n-type regions 31 was formed so as to have a different thickness between nM0S and pM0S. The above is intended to correct the junction depth in consideration of the fact that the junction depth is different due to the difference in impurities between the low-concentration P-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10.
本実施例において、 低濃度 p型拡散層 9 0及び 1 00と、 低濃度 n型拡散層 9 及び 1 0の最大不純物濃度は 1 X 1 015/cm3 以上、 1 x 1 01 cm3 以下であ ることが望ましく、 特に 1 X 1 016から 5 X 1 0 '7/cm3 の範囲であることが望 ましい。 これは 5 X 1 0 'Vcm3 以下において、 基板浮遊現象の解消は特に顕著 である力く、 一方 1 X 1 016/cm3 以下において p n接合の逆方向特性に 1 / 1 012 A程度の微小電流が生じ、 トランジスタのリーク電流となる恐れがある。 更に、 結晶欠陥領域 1 1形成のためのイオン注入工程は C MO Sにおいても一 回でよく、 S 0 I層の厚さを考慮し、 下地酸化膜 2界面で非晶質形成が達成でき ればよい。 又、 イオン注入元素も低濃度 P型拡散層 9 0及び 1 0 0と、 低濃度 n 型拡散層 9及び 1 0内で導電型に変動をきたさない材料であればよく、 1 4族元 素の S i、 G e、 C等、 及び F、 C 1等のハロゲン元素、 更には N e、 A r等の 希ガス元素等が望ましい。 In this embodiment, the maximum impurity concentration of the low-concentration p-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10 is 1 × 10 15 / cm 3 or more, and 1 × 10 1 cm 3 It is desirable that the value be in the range of 1 × 10 16 to 5 × 10 ′ 7 / cm 3 . This in 5 X 1 0 'Vcm 3 or less, elimination of substrate floating phenomenon is particularly pronounced Chikaraku, whereas 1 X 1 0 16 / cm 3 1/1 0 12 A about the reverse characteristics of the pn junction in the following Small current is generated, which may result in transistor leakage current. Further, the ion implantation process for forming the crystal defect region 11 may be performed only once in the CMOS, and the amorphous formation can be achieved at the interface with the base oxide film 2 in consideration of the thickness of the SOI layer. I just need. Also, the ion implanted element may be any material that does not cause a change in conductivity type in the low-concentration P-type diffusion layers 90 and 100 and the low-concentration n-type diffusion layers 9 and 10. It is preferable to use halogen elements such as Si, Ge, C, etc., and F, C1, etc., and further, rare gas elements such as Ne, Ar, etc.
図 4 3は本発明の他の実施例 (第 1 7 ) による半導体装置の製造工程を示す断 面図、 図 4 4はその完成断面図である。 前記実施例 1 6において、 単結晶 S i膜 3の膜厚は 1 0 0 nmとし、 酸化膜 2と 2 0 0 nm厚の単結晶 S i膜 3の間に 1 0 0 nm厚の高抵抗多結晶 S i膜 2 0と 1 0 nm厚のシリコン酸化膜 2 1が構成された多 層構造 S O I基板を用いた。 ここで単結晶 S i膜 3における p型低濃度活性領域 3及び n型低濃度活性領域 3 1を不純物濃度は各々 1 X 1 0 ' Vcm3 と極めて低 く設定した。 即ち、 本実施例においては完全空乏型の相捕型 MO S電界効果トラ ンジスタを製造した。 本実施例においてはゲート酸化膜 5、 ゲート電極 6、 ゲ一 卜保護絶縁膜 7、 ゲート側壁絶縁膜 8の形成に先立って n MO Sのゲート電極 7 形成予定領域直下の高抵抗多結晶 S i膜 2 0に不純物濃度 2 X 1 0 1 cm3 なる 高濃度 P型不純物領域 2 2を、 p MO Sのゲート電極 7形成予定領域直下の高抵 抗多結晶 S i膜 2 0に不純物濃度 2 X 1 0 1 cm3 なる高濃度 n型不純物領域 2 3を予め形成した (図 4 3 ) 。 FIG. 43 is a sectional view showing a manufacturing process of a semiconductor device according to another embodiment (17th) of the present invention, and FIG. 44 is a completed sectional view thereof. In Example 16 described above, the thickness of the single-crystal Si film 3 was set to 100 nm, and the high-resistance of 100 nm was formed between the oxide film 2 and the single-crystal Si film 3 having a thickness of 200 nm. A multi-layer SOI substrate having a polycrystalline Si film 20 and a 10 nm thick silicon oxide film 21 was used. Here, the impurity concentration of the p-type low-concentration active region 3 and the n-type low-concentration active region 31 in the single-crystal Si film 3 was set to an extremely low value of 1 × 10 ′ Vcm 3 , respectively. That is, in this example, a fully depleted phase-trapping MOS field-effect transistor was manufactured. In this embodiment, prior to the formation of the gate oxide film 5, the gate electrode 6, the gate protection insulating film 7, and the gate sidewall insulating film 8, the high-resistance polycrystalline Si immediately below the region where the nMOS gate electrode 7 is to be formed is formed. A high-concentration P-type impurity region 22 having an impurity concentration of 2 × 10 1 cm 3 is formed in the film 20, and an impurity concentration 2 is formed in the high-resistance polycrystalline Si film 20 directly below the region where the pMOS gate electrode 7 is to be formed. It was preformed X 1 0 1 cm 3 comprising a high concentration n-type impurity region 2 3 (4 3).
図 4 3の状態から前記実施例 1 6に従って半導体装置を製造したが、 n MO S のソース高濃度 n型拡散層 9 1下部に隣接した低濃度 n型拡散層 9内、 及び p M O Sにおけるソース高濃度 p型拡散層 1 0 2下部に隣接した低濃度 p型拡散層 9 0内に下地酸化膜 2に接するごとく結晶欠陥層よりなる再結合中心領域 1 1を A rのイオン注入工程により形成した。 ソース高濃度 n型拡散層 9 1、 及び高濃 度 p型拡散層 1 0 2の接合深さは約 5 O rnnに設定した。 又、 低濃度 n型拡散層 9 及び低濃度 P型拡散層 9 0はその底部を下地酸化膜 2と接する如く構成し、 その 最大不純物濃度は 1 X 1 0 1 6から 2 X 1 0 , 7/cm3 になるごとく設定した。 更に ゲート、 ゲート側壁絶縁膜部を除くソース、 ドレイン全表面には化学気相反応に 基づく 1 0 O nm厚の W膜 1 2を選択的に被着した後、 前記実施例 3に従って配線 保護絶縁膜 1 3の形成、 及びその所望個所への開口、 さらには配線用金属膜の蒸 着とそのパターニングにより接地電位線 1 7、 出力端子 1 8、 及び電源電圧線 1 9を含む配線を形成した (図 4 4 ) 。 A semiconductor device was manufactured from the state shown in FIG. 43 according to Example 16 above. However, in the low-concentration n-type diffusion layer 9 adjacent to the lower part of the nMOS source high-concentration n-type diffusion layer 9 1, and the source in the pMOS A recombination center region 11 composed of a crystal defect layer is formed in the low-concentration p-type diffusion layer 90 adjacent to the lower portion of the high-concentration p-type diffusion layer 90 by an Ar ion implantation process so as to be in contact with the base oxide film 2. did. The junction depth of the source high-concentration n-type diffusion layer 91 and the high-concentration p-type diffusion layer 102 was set to about 5 Ornn. The low-concentration n-type diffusion layer 9 and the low-concentration P-type diffusion layer 90 are configured so that their bottoms are in contact with the base oxide film 2 and their maximum impurity concentrations are from 1 × 10 16 to 2 × 10,7 . It was set as become / cm 3. Further, a W film 12 having a thickness of 10 O nm based on a chemical vapor reaction was selectively deposited on the entire surface of the source and drain except for the gate and gate sidewall insulating film portions, and then wiring was performed in accordance with the third embodiment. By forming the protective insulating film 13 and opening it at a desired location, and further depositing a wiring metal film and patterning the same, wiring including the ground potential line 17, the output terminal 18 and the power supply voltage line 19 is formed. Formed (Fig. 44).
上記製造工程を経て製造された本実施例に基づく半導体装置、 C MO S、 にお いて p M O S、 n M O Sの何れに関しても基板浮遊効果に起因する緒症状を観測 されないことは前記実施例と同様であつたが、 前記実施例 1 6の半導体装置に比 ベて本実施例に基づく半導体装置のゲート、 ドレイン電圧 2 Vにおけるソース · ドレイン電流は p M O S、 n M O S共 1 . 4倍以上の大電流化が達成された。 し 力、も低ドレイン電圧条件である非飽和特性領域において、 前記実施例 1 6の半導 体装置に比べて極めて急峻なドレインコンダクタンス特性が得られた。 本実施例 に基づく半導体装置により構成されたリングオシレータにおいては一段当りの遅 延時間は 1 2 p秒と前記実施例 3に基づくリングオシレータに比べて 6 p秒も高 速化が達成された。 これは高濃度 p型不純物領域 2 2、 及び高濃度 n型不純物領 域 2 3がパンチスルー防止機構として作用し、 且つ薄いシリコン酸化膜 2 1が製 造工程中の高温熱処理に対する不純物拡散マスクとして作用するため、 本実施例 に基づく半導体装置の超高速、 大電流特性はチャネルを構成する P型低濃度活性 領域 3、 及び n型低濃度活性領域 3 1の不純物濃度を極めて低濃度に設定でき、 不純物散乱による移動度の劣化が防止できた為と考えられる。 In the semiconductor device according to the present embodiment manufactured through the above-described manufacturing process, and in the CMOS, the pMOS and the nMOS are not observed with any symptom caused by the floating substrate effect as in the above embodiment. However, the source / drain current at a gate and drain voltage of 2 V of the semiconductor device according to the present embodiment is at least 1.4 times larger than that of the semiconductor device of Embodiment 16 in both pMOS and nMOS. Currentization has been achieved. In the non-saturation characteristic region where the power is also low and the drain voltage is low, an extremely steep drain conductance characteristic was obtained as compared with the semiconductor device of Example 16 described above. In the ring oscillator constituted by the semiconductor device according to the present embodiment, the delay time per stage was 12 ps, which was as high as 6 ps as compared with the ring oscillator according to the third embodiment. This is because the high-concentration p-type impurity region 22 and the high-concentration n-type impurity region 23 act as a punch-through prevention mechanism, and the thin silicon oxide film 21 serves as an impurity diffusion mask for high-temperature heat treatment during the manufacturing process. Therefore, the ultra-high-speed and large-current characteristics of the semiconductor device according to the present embodiment make it possible to set the impurity concentration of the P-type low-concentration active region 3 and the n-type low-concentration active region 31 constituting the channel to an extremely low concentration. It is considered that the deterioration of mobility due to impurity scattering was prevented.
本実施例に基づく半導体装置においては単結晶 S i層 3が 5 0腿と極めて薄く、 チャネル領域における基板不純物濃度も 1 X 1 0 ' Vcm3 と低く設定されている。 従って、 単結晶 S i層 3内の電荷量限定により閾電圧以上のゲート電圧条件では チャネル領域の単結晶 S i層 3に中性領域は存在せず、 完全空乏状態となってい る。 これは電流駆動源であるチヤネル中の可動電荷を効果的に誘起することがで き、 大電流化に適している。 即ち、 低電圧 ·高速動作に適している。 本実施例に 基づく半導体装置においては基板浮遊効果を伴うことなく上記の完全空乏型 S 0 I M O Sが従来半導体装置の製造方法のみで廉価に提供できることを示している。 以下の図 4 5〜図 4 9までに、 これまでの実施例で開示された半導体装置の適 用例について説明する。 In the semiconductor device according to the present embodiment, the single-crystal Si layer 3 is extremely thin at 50 feet, and the substrate impurity concentration in the channel region is set as low as 1 × 10 ′ Vcm 3 . Therefore, the neutral region does not exist in the single-crystal Si layer 3 in the channel region under the gate voltage condition higher than the threshold voltage due to the limitation of the charge amount in the single-crystal Si layer 3, and the single-crystal Si layer 3 is in a completely depleted state. This can effectively induce mobile charges in the channel, which is the current drive source, and is suitable for increasing the current. That is, it is suitable for low voltage and high speed operation. This shows that in the semiconductor device according to the present embodiment, the above-described fully depleted S 0 IMOS can be provided at a low cost only by the conventional method of manufacturing a semiconductor device without causing a substrate floating effect. FIGS. 45 to 49 below illustrate examples of application of the semiconductor device disclosed in the embodiments described above.
図 4 5 A , 4 5 Bは本発明の実施例による半導体装置の適用例を示す図である。 この例は、 随時書込み読出し型記憶装置 (D R AMと称される) を本発明に基づ く半導体装置により構成した例である。 図で、 一記憶単位であるメモリセルは図FIGS. 45A and 45B are diagrams showing application examples of the semiconductor device according to the embodiment of the present invention. In this example, a random access memory device (referred to as DRAM) is constituted by a semiconductor device based on the present invention. In the figure, the memory cell that is one storage unit is
4 5 Aのように本発明による一つの半導体装置と一つの容量素子 C sの直列接続 により構成され、 データ伝達線であるビット線、 及び入出力制御のワード線に接 続される。 本随時書込み読出し型記憶装置はメモリセルが行列状に配置されたメ モリセルアレイと制御用周辺回路で構成されるが、 周辺回路も本発明の半導体装 置により構成した。 メモリセル選択のァドレス信号端子数を低減するため列ァド レス信号と行アドレス信号をずらし、 多重化して印加する力 \ R A Sと C A Sは 各各パルス信号であり、 クロック発生器 1、 及び 2を制御してアドレス信号を行 デコーダと列デコーダに振分けている。 緩衝回路であるアドレスバッファにより 行デコーダ及び列デコーダに振分られたァドレス信号に従って特定のヮ一ド線、 及びビット線を選択する。 各ビット線にはフリップフ口ップ型増幅器によるセン スアンプが接続され、 メモリセルから読出された信号を増幅する。 パルス信号 W Eは書込みク口ック発生器を制御することにより書込みと読出しの切換えを制御 する。 Dは書込み、 及び読出し信号である。 As in the case of 45A, the semiconductor device according to the present invention is formed by connecting one semiconductor device and one capacitance element Cs in series, and is connected to a bit line as a data transmission line and a word line for input / output control. The random access memory device is composed of a memory cell array in which memory cells are arranged in rows and columns and a control peripheral circuit. The peripheral circuit is also composed of the semiconductor device of the present invention. The column address signal and the row address signal are shifted to reduce the number of address signal terminals for memory cell selection, and the multiplexed and applied force \ RAS and CAS are each pulse signal, and clock generators 1 and 2 are used. Under control, the address signal is distributed to the row decoder and the column decoder. Specific address lines and bit lines are selected according to address signals distributed to the row decoder and the column decoder by an address buffer as a buffer circuit. A sense amplifier using a flip-flop type amplifier is connected to each bit line, and amplifies a signal read from a memory cell. The pulse signal WE controls the switching between writing and reading by controlling the writing clock generator. D is a write and read signal.
本適応例を構成する各半導体装置が本発明に基づく半導体装置よりなることに より、 アクセス時間を従来比で 3 0 %以上低減できる高速性を実現できた。 さら に、 リフレッシュ特性も 1 6メガビッ 卜メモリ構成で、 最悪で 0 . 5秒と従来に 比べて 1 0倍に向上することができた。 上記の高速動作化は S O I構造による寄 生容量低減効果、 及び実施例 7に基づく大電流化による。 また、 リフレッシュ特 性の向上は S O I構造による接合面積の低減、 基板浮遊効果解消による閾電圧変 動の解消に基づく。 Since each of the semiconductor devices constituting this application example is constituted by the semiconductor device according to the present invention, a high-speed operation in which the access time can be reduced by 30% or more compared with the conventional device can be realized. In addition, with a 16-megabit memory configuration, the worst-case refresh characteristics were 0.5 seconds, which was a 10-fold improvement over the past. The high-speed operation described above is due to the effect of reducing the parasitic capacitance by the SOI structure, and the increase in current based on the seventh embodiment. The improvement of the refresh characteristics is based on the reduction of the junction area by the SOI structure and the elimination of the threshold voltage fluctuation by eliminating the substrate floating effect.
図 4 6 A, 4 6 Bは本発明の実施例による半導体装置の別の適用例を示す図で ある。 本例は、 常時書込み読出し型記憶装置 (S R AMと称される) を本発明に 基づく半導体装置により構成した例である。 図で、 一記憶単位であるメモリセル は図 4 6 Bのように、 本発明による二組の相補型 MO Sと信号の入出力を制御す る二つの M O S (トランスファ MO Sと称される) で構成される。 FIGS. 46A and 46B are diagrams showing another application example of the semiconductor device according to the embodiment of the present invention. This example is an example in which a constant write / read type storage device (referred to as SRAM) is constituted by a semiconductor device based on the present invention. In the figure, a memory cell as one storage unit is composed of two sets of complementary MOSs according to the present invention and two MOSs (referred to as transfer MOSs) for controlling signal input / output, as shown in FIG. 46B. It consists of.
本 S R AMはメモリセルが行列状に配置されたメモリセルアレイと制御用周辺 回路で構成されるが、 周辺回路も本発明の半導体装置により構成した。 本例の構 成は基本的に図 45A, 45 Bのものとほぼ同一である力く、 SRAMの高速性、 低消費電力性を図るためにァドレス遷移検出器を設け、 これにより発生するパル スによって内部回路を制御している。 更に、 アドレスバッファからデコーダまで の回路の高速化を図るため行デコーダをプリデコーダと主デコーダのニ段により 構成している。 チップセレクトは信号 CS、 及び WEにより情報の書込み、 及び 読出し時のデータを競合を避け、 且つ書込みサイクル時間と読出サイクル時間を ほぼ同じにして高速動作を可能にするための回路である。 This SRAM is composed of a memory cell array in which memory cells are arranged in rows and columns and a control peripheral circuit, and the peripheral circuit is also composed of the semiconductor device of the present invention. Configuration of this example The configuration is basically the same as that of Figs. 45A and 45B, and an address transition detector is provided to achieve the high speed and low power consumption of the SRAM. Controlling. Furthermore, in order to speed up the circuit from the address buffer to the decoder, the row decoder is composed of two stages, a predecoder and a main decoder. The chip select is a circuit for avoiding contention between information writing and reading data by the signals CS and WE, and making the write cycle time and the read cycle time almost equal to enable high-speed operation.
本例の半導体装置を構成する各半導体装置が本発明に基づく半導体装置よりな ることにより、 電源電圧を 3. 5Vから 2. 0 Vと低減でき、 且つアクセス時間 を従来比で 30%以上低減できる高速性を実現できた。 これは SO I構造による 寄生容量低減効果、 及び実施例 7に基づく大電流化、 及び低電圧におけるドレイ ンコンダクタンスの大幅な向上によるものである。 更に、 基板浮遊効果解消によ る閾電圧変動が解消され、 センスアンプの動作範囲の縮小による高速化が可能に なったためである。 The power supply voltage can be reduced from 3.5 V to 2.0 V and the access time can be reduced by 30% or more compared to the conventional device, because each of the semiconductor devices constituting the semiconductor device of this example is a semiconductor device according to the present invention. The high speed that can be achieved was realized. This is due to the effect of reducing the parasitic capacitance by the SOI structure, the increase in current based on the seventh embodiment, and the drastic improvement in drain conductance at low voltage. Furthermore, the threshold voltage fluctuation due to the elimination of the substrate floating effect has been eliminated, and the operation range of the sense amplifier has been reduced to enable higher speed.
図 47は本発明の実施例による半導体装置の更に別の適用例を示す図である。 本例は、 本発明に基づく半導体装置を用 、て論理回路装置を構成した例である。 図は複合ゲート回路の例であるが、 本発明に基づく半導体装置により複合ゲ一ト 回路に NAN D回路と NOR回路を含む論理回路に適用した。 図の複合回路は Vout = V, · ν2 +V3 · ν4 なる論理演算を行う回路であり、 この演算を ΝΑ ND回路と NOR回路の組合せで構成するより トランジスタ数を 1/2に低減で さ 。 FIG. 47 is a diagram showing still another application example of the semiconductor device according to the embodiment of the present invention. This example is an example in which a logic circuit device is configured using a semiconductor device according to the present invention. The figure shows an example of a composite gate circuit. The present invention is applied to a logic circuit including a NAND circuit and a NOR circuit in a composite gate circuit by a semiconductor device according to the present invention. The composite circuit shown in the figure is a circuit that performs the logical operation of Vout = V, · ν 2 + V 3 · ν 4 , and this operation is reduced to half the number of transistors compared to the case where this operation is configured with a combination of ND circuit and NOR circuit In.
本実施例の半導体装置を構成する各半導体装置が本発明に基づく半導体装置か らなることにより、 従来の論理回路装置に比べて遅延時間で 20%以上の低減が 図られた。 これは SO I構造による寄生容量低減効果、 及び実施例 7に基づく大 電流化、 及び低電圧におけるドレインコンダクタンスの大幅な向上による。 他の適用例を図 4 8の計算機構成図で説明する。 本例は本発明の半導体装置を 命令や演算を処理するプロセッサ 5 00が複数個並列に接続された高速大型計算 機に適用した例である。 本例では半導体装置が従来のバイポーラトランジスタを 用いた集積回路よりも集積度が高く廉価なため、 命令や演算を処理するプロセッ サ 500、 システム制御装置 50 1、 及び主記憶装置 502等を 1辺が 1 0から 30画の本発明の半導体装置で構成した。 Since each of the semiconductor devices constituting the semiconductor device of the present embodiment is constituted by the semiconductor device according to the present invention, the delay time can be reduced by 20% or more compared with the conventional logic circuit device. This is due to the effect of reducing the parasitic capacitance by the SOI structure, the increase in current according to the seventh embodiment, and the drastic improvement in drain conductance at low voltage. Another application example will be described with reference to a computer configuration diagram in FIG. This example is an example in which the semiconductor device of the present invention is applied to a high-speed large-scale computer in which a plurality of processors 500 for processing instructions and operations are connected in parallel. In this example, since the semiconductor device has a higher degree of integration and is less expensive than an integrated circuit using conventional bipolar transistors, a processor for processing instructions and operations is used. The semiconductor device 500 of the present invention having 10 to 30 pixels on one side includes the device 500, the system control device 501, the main storage device 502, and the like.
これら命令や演算を処理するプロセッサ 500、 システム制御装置 50 1、 及 び化合物半導体装置からなるデータ通信インタフヱース 503を同一セラミック 基板 506に実装した。 また、 データ通信インタフヱース 503、 及びデータ通 信制御装置 504を同一セラミック基板 507に実装した。 これらセラミック基 板 506、 及び 507と主記憶装置 502が実装されたセラミック基板を大きさ が 1辺約 50cm程度、 あるいはそれ以下の基板に実装し、 計算機の中央処理ュニ ット 508を形成した。 この中央処理ュニット 508内データ通信や、 複数の中 央処理ュニット間データ通信、 あるいはデータ通信インタフヱ一ス 503と入出 力プロセッサ 505を実装した基板 509との間のデータの通信は図中の両端矢 印線で示される光ファイバ 5 1 0を介して行われた。 A processor 500 for processing these instructions and calculations, a system controller 501, and a data communication interface 503 composed of a compound semiconductor device are mounted on the same ceramic substrate 506. Further, the data communication interface 503 and the data communication control device 504 are mounted on the same ceramic substrate 507. The ceramic substrates on which the ceramic substrates 506 and 507 and the main storage device 502 were mounted were mounted on a substrate having a size of about 50 cm on a side or smaller, thereby forming a central processing unit 508 for a computer. . The data communication in the central processing unit 508, the data communication between a plurality of central processing units, or the data communication between the data communication interface 503 and the board 509 on which the input / output processor 505 is mounted are illustrated by arrows in FIG. The operation was performed through an optical fiber 510 indicated by a dotted line.
この計算機では命令や演算を処理するプロセッサ 500、 システム制御装置 5 0 1、 及び主記憶装置 502等が本発明による半導体装置を並列に用いて形成さ れており、 高速に動作し、 またデータの通信が光を媒体に行われるため、 1秒間 当たりの命令処理回数を大幅に増加することができた。 In this computer, a processor 500 for processing instructions and operations, a system controller 501, and a main storage device 502 are formed by using the semiconductor device according to the present invention in parallel, and operate at high speed, and Since communication is performed using light as a medium, the number of command processing operations per second could be greatly increased.
図 49は本発明の実施例による半導体装置の更にまた別の適用例を示す図であ る。 本例は、 本発明に基づく半導体装置により構成された信号伝送処理装置であ り、 特に非同期伝送方式 (ATM交換器と称される) に関する信号伝送処理装置 である。 図 49に於いて、 光ファイバにより超高速で直列的に伝送されてきた情 報信号は電気信号に変換し (0/E変換) 、 且つ並列化 (SZP変換) させる装 置を介して、 本発明に基づく半導体装置により構成される集積回路 (BFMLS I) に導入した。 集積回路で番地付処理された電気信号は直列化 (P/S変換) 及び光信号化 (E/0変換) された光ファイバで出力される。 FIG. 49 is a diagram showing still another application example of the semiconductor device according to the embodiment of the present invention. The present example is a signal transmission processing device constituted by the semiconductor device according to the present invention, particularly a signal transmission processing device relating to an asynchronous transmission system (referred to as an ATM switch). In FIG. 49, the information signal transmitted in series at a very high speed by an optical fiber is converted into an electric signal (0 / E conversion) and is converted into a parallel signal (SZP conversion). It was introduced into an integrated circuit (BFMLS I) composed of a semiconductor device based on the invention. The electrical signal that has been addressed by the integrated circuit is output via a serialized (P / S conversion) and optical signal (E / 0 conversion) optical fiber.
BFMLS Iは多重器 (MUX) 、 バッファメモリ (BFM) 、 及び分離器 (DMUX) により構成される。 BFMLS Iはメモリ制御 LS I、 及び空ァド レス振分け制御の機能を有する LS I (空アドレス F I FOメモリ LS I) によ り制御される。 The BFMLS I consists of a multiplexer (MUX), a buffer memory (BFM), and a separator (DMUX). The BFM LSI is controlled by a memory control LSI and an LSI (empty address FIFO memory LSI) having an empty address distribution control function.
本信号伝送処理装置は伝送すベき番地と無関係に送られてくる超高速伝送信号 を所望番地に超高速で伝送するスィツチの機能を有する装置である。 B F M L S Iは入力光信号の伝送速度に比べて著しく動作速度が遅いため、 入力信号を直接 スイッチングできず、 入力信号を一時記憶させ、 記憶された信号をスイッチング してから超高速な光信号に変換して所望番地に伝送する方式を用いている。 B F ML S Iの動作速度が遅ければ、 大きな記憶容量が要求される。 This signal transmission processor is an ultra-high-speed transmission signal sent regardless of the address to be transmitted. Is a device having a function of a switch for transmitting data to a desired address at a very high speed. Since the operation speed of BFMLSI is remarkably slower than the transmission speed of the input optical signal, the input signal cannot be directly switched.The input signal is temporarily stored, the stored signal is switched, and then converted to an ultra-high-speed optical signal. A method of transmitting data to a desired address is used. If the operation speed of the BF ML SI is slow, a large storage capacity is required.
本例に基づく A TM交換器に於いては B FM L S Iが本発明に基づく半導体装 置により構成され、 従来の B FML S Iに比べて動作速度が三倍と高速で且つ廉 価なため、 B F ML S Iの記憶容量を従来比で約 1 / 3に低減することが可能と なった。 これにより A TM交換器の製造原価を低減することができた。 In the ATM switch based on this example, the BFM LSI is constituted by the semiconductor device based on the present invention, and the operation speed is three times faster and less expensive than the conventional BFMLSI. It has become possible to reduce the storage capacity of ML SI to about 1/3 that of the conventional model. As a result, the manufacturing cost of the ATM exchanger was reduced.
図 5 0は本発明の第 1 8実施例による半導体装置の完成断面を示す図である。 実施例 6で、 高濃度 n型拡散層 9 1、 及び 1 0 1形成のイオン注入に代えて高濃 度 P型拡散層 1 1上の S i層の選択除去と除去領域への高濃度 n型不純物が添加 された多結晶 S i膜 2 4の選択残置を実施した。 同様に高濃度 p型拡散層 9 2、 及び 1 0 2形成のイオン注入に代えて高濃度 n型拡散層 1 1 0上の S i層の選択 除去と除去領域への高濃度 p型不純物が添加された多結晶 S i膜 2 5の選択残置 を実施した。 本実施例では上記多結晶 S i膜 2 5の選択残置はジクロルシランを 原料とする低圧化学気相反応に基づいて S i露出面上にのみ選択堆積する手法を 用いた。 この手法に代えて非晶質、 又は多結晶質 S iの全面堆積と、 所望個所へ の P、 及び n型不純物の高濃度イオン注入、 その後の S i膜のパターニングによ り形成してもよい。 FIG. 50 is a view showing a completed cross section of the semiconductor device according to the eighteenth embodiment of the present invention. In Example 6, instead of the ion implantation for forming the high-concentration n-type diffusion layers 91 and 101, the selective removal of the Si layer on the high-concentration P-type diffusion layer 11 and the high-concentration n The polycrystalline Si film 24 to which the type impurity was added was selectively left. Similarly, instead of ion implantation for forming high-concentration p-type diffusion layers 92 and 102, selective removal of Si layer on high-concentration n-type diffusion layer 110 and high-concentration p-type impurities Selective remaining of the added polycrystalline Si film 25 was performed. In the present embodiment, a method of selectively depositing only the Si exposed surface based on a low-pressure chemical vapor reaction using dichlorosilane as a raw material was used for selectively leaving the polycrystalline Si film 25. Instead of this method, it may be formed by depositing amorphous or polycrystalline Si over the entire surface, implanting P and n-type impurities at desired locations with high concentration ions, and then patterning the Si film. Good.
本実施例の C MO Sでは、 p MO S、 n M o Sの何れに関しても基板浮遊効果 に起因する諸症状は実施例 6の半導体装置と同様に観測されなかった。 本実施例 による半導体装置で、 高濃度 n型不純物が添加された多結晶 S i膜 2 4と高濃度 P型拡散層 1 1間、 及び高濃度 P型不純物が添加された多結晶 S i膜 2 5と高濃 度 n型拡散層 1 1 0間は多結晶膜の結晶粒界の存在により良好なォーミク特性を 示した。 多結晶 S i膜 2 4と 2 5は所望により単結晶 S i膜表面よりも上部に達 し、 ゲート電極側部に位置するように厚く構成し、 ソース抵抗の低減を図ること が可能である。 In the CMOS of the present example, no symptom caused by the substrate floating effect was observed for any of pMOS and nMoS as in the semiconductor device of Example 6. In the semiconductor device according to the present embodiment, a polycrystalline Si film doped with a high concentration n-type impurity and a polycrystalline Si film doped with a high concentration P-type impurity Good ohmic characteristics were exhibited between 25 and the high-concentration n-type diffusion layer 110 due to the presence of the crystal grain boundaries of the polycrystalline film. The polycrystalline Si films 24 and 25 can be made thicker so as to reach the upper part of the monocrystalline Si film surface if necessary and to be located on the side of the gate electrode, so that the source resistance can be reduced. .
図 5 1は本発明の他の実施例 (第 1 9 ) による半導体装置の完成断面を示す図 である。 実施例 6で、 p MO Sと n MO Sを構成する基板として S O I基板に代 えて単結晶 S i層 3と同一仕様で 6 2 0〃m厚さの単結晶 S i基板 3 0を用いた c また、 低濃度 n型領域 3 1と低濃度 p型領域 3に代えて低濃度 n型ゥエル 3 2と 低濃度 P型ゥエル 3 3を公知の半導体装置の製造方法に基づいて形成した。 更に、 本実施例では高濃度 n型拡散層 9 1、 及び 1 0 1上に高濃度 n型多結晶 S i膜 4 0と W膜 1 2の積層膜を、 高濃度 p型拡散層 9 2、 及び 1 0 2上に高濃度 p型多 結晶 S i膜 4 1と W膜 1 2の積層膜を形成してソース抵抗の低減を図った。 FIG. 51 is a view showing a completed cross section of a semiconductor device according to another embodiment (No. 19) of the present invention. It is. In Example 6, a single crystal Si substrate 30 having the same specification as the single crystal Si layer 3 and having a thickness of 62 μm was used in place of the SOI substrate as a substrate constituting pMOS and nMOS. c Further , instead of the low-concentration n-type region 31 and the low-concentration p-type region 3, a low-concentration n-type well 32 and a low-concentration P-type well 33 are formed based on a known method for manufacturing a semiconductor device. Further, in this embodiment, a stacked film of the high-concentration n-type polycrystalline Si film 40 and the W film 12 is formed on the high-concentration n-type diffusion layers 91 and 101, and the high-concentration p-type diffusion layer 92 is formed. , And 102, a stacked film of a high-concentration p-type polycrystalline Si film 41 and a W film 12 was formed to reduce source resistance.
本実施例の C MO Sでは、 ゥェル電位供給用の電極を配置しなかったにも係わ らず、 p MO S、 n MO Sの何れにも閾電圧変動等のゥヱル電位浮遊効果に起因 する緒症状は観測されなかった。 即ち、 従来相補型 MO Sで、 必須であったゥェ ル電位固定電極が不要となり、 チップ占有面積の低減に寄与することができた。 更に本実施例による半導体装置ではソース上に積上げ構成された低抵抗多結晶膜 と W膜の積層膜の効果によりソース抵抗を低減することができた。 In the CMOS of the present embodiment, both the pMOS and the nMOS are caused by a cell potential floating effect such as a threshold voltage fluctuation, although no electrode for supplying a cell potential is provided. No cord symptoms were observed. In other words, the need for a fixed gel potential fixed electrode, which was required in the conventional complementary MOS, was eliminated, thereby contributing to a reduction in chip occupation area. Further, in the semiconductor device according to the present embodiment, the source resistance could be reduced by the effect of the laminated film of the low-resistance polycrystalline film and the W film stacked on the source.
産業上の利用可能性 Industrial applicability
以上のように、 本発明によれば、 S O I基板上に構成された半導体装置の最大 の欠点であつた基板浮遊効果に基づく閾電圧の変動や電流電圧特性上の異常なこ ぶ状特性の発生などの欠点を、 ィォン源が不安定でかつ専用装置を要する G eィ オン注入等の方法によることなく、 既存の半導体装置によって、 解消することが できる。 従って本発明によれば、 専用装置の付加に伴う半導体装置の占有面積や 費用の増大を必要とせず、 上記特性の改善を実現することができる。 As described above, according to the present invention, fluctuations in the threshold voltage due to the substrate floating effect and the occurrence of abnormal bump-like characteristics in the current-voltage characteristics, which are the biggest disadvantages of the semiconductor device formed on the SOI substrate, are The disadvantage of the above can be solved by an existing semiconductor device without using a method such as Ge ion implantation in which an ion source is unstable and requires a dedicated device. Therefore, according to the present invention, the above characteristics can be improved without requiring an increase in the area occupied by the semiconductor device and the cost associated with the addition of the dedicated device.
さらに、 従来不可能であった S O I基板上の p MO Sの基板浮遊効果に対して も廉価な製造方法により解消することができる。 従って、 本発明によれば S O I 基板上の C MO Sに対して廉価な製造方法により基板浮遊効果を完全に解消する ことができる。 これにより低電圧、 低電力で且つ超高速の半導体装置、 及びそれ により構成されるシステムを提供することができる。 Furthermore, the substrate floating effect of pMOS on the SOI substrate, which was impossible in the past, can be solved by a low-cost manufacturing method. Therefore, according to the present invention, the substrate floating effect can be completely eliminated by a low-cost manufacturing method for CMOS on the SOI substrate. Thus, a low-voltage, low-power, ultra-high-speed semiconductor device and a system including the semiconductor device can be provided.
Claims
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP20065795 | 1995-08-07 | ||
JP7/200657 | 1995-08-07 | ||
JP21041295 | 1995-08-18 | ||
JP7/210412 | 1995-08-18 | ||
JP21041095 | 1995-08-18 | ||
JP7/210410 | 1995-08-18 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1997006564A1 true WO1997006564A1 (en) | 1997-02-20 |
Family
ID=27327846
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP1996/002184 WO1997006564A1 (en) | 1995-08-07 | 1996-08-02 | Semiconductor device and method for manufacturing the same |
Country Status (2)
Country | Link |
---|---|
TW (1) | TW302538B (en) |
WO (1) | WO1997006564A1 (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2358084A (en) * | 2000-01-07 | 2001-07-11 | Seiko Epson Corp | Field effect transistors |
US6770517B2 (en) | 1997-06-19 | 2004-08-03 | Matsushita Electric Industrial Co., Ltd. | Semiconductor device and method for fabricating the same |
JP2008192760A (en) * | 2007-02-02 | 2008-08-21 | Oki Electric Ind Co Ltd | Semiconductor device, manufacturing method of semiconductor device, and method of using semiconductor device |
WO2016166930A1 (en) * | 2015-04-15 | 2016-10-20 | 信越半導体株式会社 | Method for manufacturing semiconductor device and method for evaluating semiconductor device |
JP2018018872A (en) * | 2016-07-26 | 2018-02-01 | 信越半導体株式会社 | Method of manufacturing semiconductor device, and method of evaluating semiconductor device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62271472A (en) * | 1986-05-20 | 1987-11-25 | Toshiba Corp | Semiconductor device |
JPS6489464A (en) * | 1987-09-30 | 1989-04-03 | Toshiba Corp | Semiconductor device and manufacture thereof |
JPH04313242A (en) * | 1991-04-10 | 1992-11-05 | Sony Corp | Manufacture of thin-film semiconductor device |
JPH06291142A (en) * | 1993-03-31 | 1994-10-18 | Sony Corp | Soi mos fet and fabrication thereof |
JPH08213622A (en) * | 1994-09-13 | 1996-08-20 | Toshiba Corp | Semiconductor device and manufacturing method thereof |
-
1996
- 1996-08-02 WO PCT/JP1996/002184 patent/WO1997006564A1/en active Application Filing
- 1996-08-06 TW TW085109520A patent/TW302538B/zh active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62271472A (en) * | 1986-05-20 | 1987-11-25 | Toshiba Corp | Semiconductor device |
JPS6489464A (en) * | 1987-09-30 | 1989-04-03 | Toshiba Corp | Semiconductor device and manufacture thereof |
JPH04313242A (en) * | 1991-04-10 | 1992-11-05 | Sony Corp | Manufacture of thin-film semiconductor device |
JPH06291142A (en) * | 1993-03-31 | 1994-10-18 | Sony Corp | Soi mos fet and fabrication thereof |
JPH08213622A (en) * | 1994-09-13 | 1996-08-20 | Toshiba Corp | Semiconductor device and manufacturing method thereof |
Non-Patent Citations (1)
Title |
---|
KIYOO ITO, "Advanced Electronics I-9, VLSI Memory", 5 November 1994, BAIFUKAN, p. 337. * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6770517B2 (en) | 1997-06-19 | 2004-08-03 | Matsushita Electric Industrial Co., Ltd. | Semiconductor device and method for fabricating the same |
GB2358084A (en) * | 2000-01-07 | 2001-07-11 | Seiko Epson Corp | Field effect transistors |
US6528830B1 (en) | 2000-01-07 | 2003-03-04 | Seiko Epson Corporation | Thin film transistor |
GB2358084B (en) * | 2000-01-07 | 2004-02-18 | Seiko Epson Corp | Semiconductor transistor |
JP2008192760A (en) * | 2007-02-02 | 2008-08-21 | Oki Electric Ind Co Ltd | Semiconductor device, manufacturing method of semiconductor device, and method of using semiconductor device |
WO2016166930A1 (en) * | 2015-04-15 | 2016-10-20 | 信越半導体株式会社 | Method for manufacturing semiconductor device and method for evaluating semiconductor device |
JP2016207682A (en) * | 2015-04-15 | 2016-12-08 | 信越半導体株式会社 | Semiconductor device manufacturing method and semiconductor device evaluation method |
JP2018018872A (en) * | 2016-07-26 | 2018-02-01 | 信越半導体株式会社 | Method of manufacturing semiconductor device, and method of evaluating semiconductor device |
WO2018020961A1 (en) * | 2016-07-26 | 2018-02-01 | 信越半導体株式会社 | Semiconductor device manufacturing method and semiconductor device evaluation method |
US10886129B2 (en) | 2016-07-26 | 2021-01-05 | Shin-Etsu Handotai Co., Ltd. | Method for manufacturing semiconductor device and method for evaluating semiconductor device |
Also Published As
Publication number | Publication date |
---|---|
TW302538B (en) | 1997-04-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO1999005715A1 (en) | Semiconductor device and method of producing the same | |
US5886385A (en) | Semiconductor device and manufacturing method thereof | |
KR0141519B1 (en) | High performance bipolar differential sense amplifier in a bicmos sram | |
US7630235B2 (en) | Memory cells, memory devices and integrated circuits incorporating the same | |
US20020158254A1 (en) | Memory system capable of operating at high temperatures and method for fabricating the same | |
US7504286B2 (en) | Semiconductor memory devices and methods for fabricating the same | |
KR101519573B1 (en) | Gated lateral thyristor-based random access memory (gltram) cells with separate read and write access transistors, memory devices and integrated circuits incorporating the same | |
JPH08213564A (en) | Semiconductor device | |
US5693975A (en) | Compact P-channel/N-channel transistor structure | |
US6091654A (en) | Circuit and method for low voltage, current sense amplifier | |
JP2005251776A (en) | Semiconductor device and its manufacturing method | |
JPH09139434A (en) | Semiconductor device and manufacturing method thereof | |
WO1997006564A1 (en) | Semiconductor device and method for manufacturing the same | |
JP3845272B2 (en) | SRAM and manufacturing method thereof | |
KR101631634B1 (en) | Methods for fabricating gated lateral thyristor-based random access memory(gltram) cells | |
US20030132452A1 (en) | Semiconductor device and method of producing the same | |
JPH10189959A (en) | Semiconductor device | |
JP3980670B2 (en) | Semiconductor device | |
JP2637186B2 (en) | Semiconductor device | |
JPH10189771A (en) | Semiconductor device and manufacturing method thereof | |
Maeda et al. | High performance BiCMOS technology design for sub-10 ns 4 Mbit BiCMOS SRAM with 3.3 V operation | |
JP2006066928A (en) | Method of manufacturing semiconductor device | |
KR960003863B1 (en) | Semiconductor device having buried layer and the manufacturing method thereof | |
JPH0214567A (en) | Semiconductor integrated circuit device | |
JPS6333863A (en) | Manufacture of semiconductor device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN JP KR SG US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
122 | Ep: pct application non-entry in european phase |