Semiconductor device and method of manufacturing the same

Abstract

There is provided a semiconductor device which comprises a capacitor including a lower electrode, a dielectric film, and an upper electrode, a first protection film formed on the capacitor, a first wiring formed on the first protection film, a first insulating film formed on the first wiring, a second wiring formed on the first insulating film, a second insulating film formed on the second wiring, and at least one of a second protection film formed between the first insulating film and the first wiring to cover at least the capacitor and a third protection film formed on the second insulating film to cover the capacitor and set to an earth potential. Accordingly, the degradation of the ferroelectric capacitor formed under the multi-layered wiring structure can be suppressed.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a semiconductor device and to a method of manufacturing the same and, more particularly, to a semiconductor device having a nonvolatile semiconductor memory (FeRAM: Ferroelectric Random Access Memory) using ferroelectric material as a dielectric film of a capacitor and a nonvolatile semiconductor memory (DRAM: Dynamic Random Access Memory) using high-dielectric material as the dielectric film of the capacitor, and to a method of manufacturing the same.

2. Description of the Prior Art

In the FeRAM, the multi-layered wiring technology used in other electronic devices is being employed to meet the demand for the higher integration of the device. However, since the ferroelectric material used in the FeRAM is exposed to the reducing atmosphere containing the hydrogen when the interlayer insulating film, the tungsten plug, the cover film, etc. are formed, the ferroelectric material is easily damaged by the formation of the multi-layered wiring structure.

In order to suppress the degradation of the ferroelectric film constituting the capacitor due to the reducing reaction, several trials were carried out.

For example, in FIG. 1 of Patent Application Publication (KOKAI) Hei 7-111318, it is set forth that, the protection film made of aluminum nitride which is formed above the upper electrode of the capacitor is possible to prevent the reduction of the ferroelectric film of the capacitor caused when the reducing gas permeates through the upper electrode. Also, in FIG. 8 of the same Publication, it is set forth that the protection film is formed on the wiring connected to the upper electrode of the capacitor and on the insulating film covering the capacitor. But no recitation about constituent material and the particular operation of the insulating film is given.

Also, in Patent Application Publication (KOKAI) Hei 9-97883, it is set forth that the lower electrode and the dielectric film constituting the capacitor are formed, then the dielectric film is covered with the insulating film, then the opening for exposing the dielectric film is formed in the insulating film, then the upper electrode of the capacitor is formed in the opening and on the insulating film, and then the protection film having the double-layered structure consisting of titanium and titanium nitride is formed on the upper electrode. This protection film has a function for preventing the diffusion of the hydrogen into the capacitor and the permeation of the moisture into the capacitor.

In FIG. 1 of Patent Application Publication (KOKAI) Hei 7-235639, it is set forth that the lower electrode and the dielectric film constituting the capacitor are formed, then the capacitor is covered with the insulating film, then the opening for exposing the upper electrode is formed in the insulating film, and then the wiring having the double-layered structure containing the titanium tungsten film is formed in the opening and on the insulating film. Also, in FIG. 2 of the same Publication, it is set forth that the water resisting layer made of silicon nitride is formed on the titan tungsten film over the capacitor except an area of the upper electrode of the capacitor. This water-resistant layer is formed to shut off the permeation of the moisture from the region on which the wiring is not formed.

Also, it is set forth on 17-th Ferroelectric Material Application Conference, Preprint, pp. 17-18 that the metal wiring connected to the capacitor is formed and then the alumina (Al 2 O 3 ) film for covering the metal wiring is formed over the overall area of the substrate.

By the way, the structure in which the first level wiring is connected to the upper electrode of the capacitor is disclosed in above references, but it is not set forth to form further second and third level wirings above the capacitor.

Accordingly, since the capacitor is exposed further to the reducing atmosphere during the step of forming the multi-layered wiring above the capacitor, there is the possibility that the degradation of characteristics of the capacitor cannot be satisfactorily suppressed by the protection structure of the above capacitor in above references.

The degradation of the imprint characteristic becomes the greatest problem out of degradations of the ferroelectric capacitor due to the reducing atmosphere. The degradation of the imprint characteristic is such a problem that, if one signal (e.g., “1”) is written into the ferroelectric capacitor, then the ferroelectric capacitor is left for a certain time as it is, and then an opposite signal (e.g., “0”) is written into the capacitor, the opposite signal cannot be read out. In other words, the degradation of the imprint characteristic signifies such a situation that, since the signal in the one direction is imprinted into the capacitor, it is difficult to write the opposite signal into the capacitor.

In the 2-transisitors/2-capacitors type FeRAM, after the positive signal is written into one of two pair of ferroelectric capacitors and the negative signal is written into the other, a difference of the polarization charge between the two capacitors is set to Q.

Then, a difference of the polarization charge between the two ferroelectric capacitors obtained after the ferroelectric capacitors are baked at 150° C. for 88 hours is defined as Q (88) μC/cm 2 , and a degradation rate of the difference Q between the capacitors obtained after an e time (e=natural logarithm) lapsed is defined as a “Q rate”, both are used as indices of the imprint characteristic. In other words, it is understood that, as a value of Q (88) is increased larger and an absolute value of the Q rate is reduced smaller, the imprint characteristic becomes excellent much more.

Now the reason to evaluate the ferroelectric capacitor at 150° C. for 88 hours is to assure the 10-year use of the FeRAM under the circumstance of 55° C. Details are set forth in S. D. TRAYNOR, T. D. HADNAGY, and L. KAMMERDINER, Integrated Ferroelectrics, 1997, Vol.16, pp.63-76.

When the degradation of the characteristics of the capacitor due to difference in the wiring structure on the ferroelectric capacitor are evaluated based on the evaluation of the degradation of such ferroelectric capacitor, results shown in Table 1 were obtained by the present inventor's experiments.

TABLE 1
Imprint characteristic of the ferroelectric
capacitor according to steps (5 V evaluation)
steps	Q (88) [μC/cm 2 ]	Q rate [%]
After formation of the	24.6	−1.4
ferroelectric capacitor
After formation of the	24.0	−1.8
second layer metal wiring
After formation of the third	19.2	−5.0
layer metal wiring + the cover film

In Table 1, the state that the first metal wiring is connected to the upper electrode of the ferroelectric capacitor is shown as “after formation of the ferroelectric capacitor”. Also, the state that the second metal wiring is formed on the ferroelectric capacitor is shown as “after formation of the second layer metal wiring”. In addition, the state that the third layer metal wiring and the cover film are formed on the ferroelectric capacitor is shown as “after formation of the third layer metal wiring+the cover film”. The measurement of Q was carried out under the condition applying the voltage of 5 V to the ferroelectric capacitors.

According to Table 1, the Q rate is not so increased until the second layer metal is formed, and thus the degradation of the imprint characteristic is small. However, after the third layer metal wiring and the cover film are formed, the Q rate is increased and thus the degradation of the imprint characteristic appears.

The main causes of the degradation of the imprint characteristic are the CVD process applied to form the tungsten in the reducing atmosphere and the CVD process applied to form the cover film made of silicon nitride in the reduced atmosphere.

Accordingly, with the increase in the number of the wiring layer of the multi-layered wiring structure, the degradation of the imprint characteristic is also increased and thus the degradation of the capacitor characteristic is caused.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a semiconductor device having a structure that is capable of suppressing degradation of a ferroelectric or high-dielectric capacitor formed under a multi-layered structure, and a method of manufacturing the same.

According to the present invention, the semiconductor device has the first protection film which is covering the surface of the ferroelectric or high-dielectric capacitor, the second protection film formed to cover the capacitor through the first wiring formed above the capacitor, the second wiring formed over the second protection film, the third protection film formed to cover the capacitor over the second wiring, and the third protection film is set to the earth potential.

According to this, even when the insulating films and the conductive films are formed or etched over the ferroelectric or high-dielectric capacitors in the reducing atmosphere, the ferroelectric or high-dielectric films of the capacitors can be protected from the reducing atmosphere by the first protection film, the second protection film, and the third protection film underlying the films which are subjected to the forming or etching processes.

The second protection film or the third protection film can prevent the reducing gas over the second or third protection film from permeating in the ferroelectric or high-dielectric capacitor, but the second or third protection film cannot prevent the moisture or the hydrogen existing under the second or third protection film from entering into the capacitor.

Therefore, in order to prevent the reduction of the ferroelectric or high-dielectric capacitors, either a combination of the first protection film and the second protection film or a combination of the first protection film and the third protection film is indispensable. When all the first protection film, the second protection film, and the third protection film are provided, the reduction of the ferroelectric or high-dielectric capacitors can be excellently prevented. Then, these protection films improve the imprint characteristic of the ferroelectric capacitor satisfactorily and also improve the retention performance peculiar to the FeRAM.

Also, as the third protection film is set to the earth potential, the mutual induction between the second wirings under the third protection film, e.g., the bit lines, under the third protection film is prevented and also the fluctuation of the electlic potential of the second wirings can be suppressed, whereby the performance of FeRAM or DRAM can be improved. In addition, the third protection film that is set to the earth potential prevents the hydrogen ion over the third protection film from permeating into the ferroelectric capacitors. The hydrogen ion is generated when one film is growing over the third protection film.

In case the first protection film and the second protection film are formed of alumina, the imprint characteristic of the ferroelectric capacitors can be improved if the thickness is set to 15 to 100 nm or the helicon sputter method is employed.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A to 1 P are sectional views showing steps of manufacturing a semiconductor device according to an embodiment of the present invention, taken along the extending direction of the bit line;

FIGS. 2A to 2 M are sectional views showing steps of manufacturing a semiconductor device according to the embodiment of the present invention, taken along the extending direction of the word line;

FIG. 3 is view showing circuit arrangement of the semiconductor device according to the embodiment of the present invention; and

FIG. 4 , FIG. 5 , FIG. 6 , and FIG. 7 are plan views showing arrangement relationship between a third protection film and respective conductive patterns of the semiconductor device according to the embodiment of the present invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Embodiments of the present invention will be explained with reference to the accompanying drawings hereinafter.

FIG. 1A to FIG. 1P are sectional views showing steps of manufacturing a semiconductor device according to an embodiment of the present invention, taken along the extending direction of the bit line. FIG. 2A to FIG. 2M are sectional views showing steps of manufacturing a capacitor and its peripheral structure, taken along the extending direction of the word line of the semiconductor device according to the embodiment of the present invention.

First, steps required to get a sectional structure shown in FIG. 1 A and FIG. 2A will be explained hereunder.

In FIG. 1A , a device isolation insulating film 2 is formed on a surface of a p-type silicon (semiconductor) substrate 1 by the LOCOS (Local Oxidation of Silicon) method. In this case, in addition to the silicon oxide film formed by the LOCOS method, the STI (Shallow Trench Isolation) may be employed as the device isolation insulating film 2 .

After such device isolation insulating film 2 is formed, a p-type impurity and an n-type impurity are selectively introduced into predetermined active regions (transistor forming regions) in a memory cell region A and a peripheral circuit region B of the silicon substrate 1 respectively, whereby a first p-well 3 a is formed in the active region of the memory cell region A and also an n-well 4 is formed in the active region of the peripheral circuit region B. Also, as shown in FIG. 2A , a second p-well 3 b is formed in vicinity of the region of the memory cell region A, in which the capacitor is formed.

Although not shown in FIG. 1A , a p-well (not shown) is formed in the peripheral circuit region B to form the CMOS.

Then, a silicon oxide film used as a gate insulating film 5 is formed by thermally oxidizing surfaces of respective active regions of the silicon substrate 1 .

Then, an amorphous silicon film and a tungsten silicide film are formed in sequence on the overall surface of the silicon substrate 1 to cover the device isolation insulating film 2 and the gate insulating film 5 . Then, the amorphous silicon film and the tungsten silicide film are patterned into predetermined shapes by the photolithography method, whereby gate electrodes 6 a to 6 c are formed in the active regions and also a leading wiring 7 is formed on the device isolation insulating film 2 .

In the memory cell region A, two gate electrodes 6 a , 6 b are arranged on the first p-well 3 a in almost parallel with each other. These gate electrodes 6 a , 6 b are extended on the device isolation insulating film 2 and act as the word lines WL.

In this case, a polysilicon film may be formed in place of the amorphous silicon film constituting the gate electrodes 6 a to 6 c.

Then, n-type impurity diffusion regions 8 a , 8 b serving as source/drain of an n-channel MOS transistor are formed by ion-implanting the n-type impurity into the first p-well 3 a on both sides of the gate electrodes 6 a , 6 b in the memory cell region A. At the same time, n-type impurity diffusion regions are formed in a p-well (not shown) in the peripheral circuit region B.

Next, p-type impurity diffusion regions 9 serving as source/drain of a p-channel MOS transistor are formed by ion-implanting the p-type impurity into the n-well 4 on both sides of the gate electrode 6 c in the peripheral circuit region B. The n-type impurity and the p-type impurity are ion-implanted separately by using the resist patterns.

Then, an insulating film is formed on the overall surface of the silicon substrate 1 . Sidewall insulating films 10 are left on both side portions of the gate electrodes 6 a to 6 c and the leading wiring 7 respectively by etching-back the insulating film. The silicon oxide (SiO2) formed by the CVD method, for example, is used as the insulating film.

Next, a silicon oxide nitride (SiON) film (not shown) as a cover film may be formed on the overall surface of the silicon substrate 1 by the plasma CVD method.

Then, a silicon oxide (SiO2) film is grown up to about 1.0 μm thickness by the plasma CVD method using the TEOS gas. This silicon oxide film is used as a first interlayer insulating film 11 .

Then, as the densifying process of the first interlayer insulating film 11 , such first interlayer insulating film 11 is annealed at the temperature of 700° C. for 30 minutes at the atmospheric pressure in the nitrogen atmosphere. Then, an upper surface of the first interlayer insulating film 11 is planarized by polishing the first interlayer insulating film 11 by virtue of the CMP (Chemical Mechanical Polishing) method.

Then, steps required to form a sectional structure shown in FIG. 1 B and FIG. 2B will be explained hereunder.

First, holes 12 a to 12 d reaching the impurity diffusion regions 8 a , 8 b , 9 , a hole 12 e reaching the leading wiring 7 , and a hole 12 f reaching the second well 3 b are formed by patterning the first interlayer insulating film 11 by virtue of the photolithography method. Then, a Ti (titanium) film of 20 nm thickness and a TiN (titanium nitride) film of 50 nm thickness are formed in sequence on the first interlayer insulating film 11 and in the holes 12 a to 12 f by the sputter method. Then, a W (tungsten) film is grown on the TiN film by the CVD method to have a thickness that can perfectly bury the holes 12 a to 12 f.

Then, the W film, the TiN film, and the Ti film are polished sequentially by the CMP method until the upper surface of the first interlayer insulating film 11 is exposed. After this polishing, the W film, etc. left in the holes 12 a to 12 f are used as contact plugs 13 a to 13 f.

The first contact plug 13 a on the n-type impurity diffusion region 8 a put between two gate electrodes 6 a , 6 b is connected to the bit line, described later, in the first p-well 3 a of the memory cell region A. In addition, two second contact plugs 13 b are connected to the upper electrode of the capacitor, described later.

In this case, for the sake of the contact compensation, the impurity may be ion-implanted into the impurity diffusion regions 8 a , 8 b , 9 after the holes 12 a to 12 f are formed.

Then, as shown in FIG. 1C , in order to prevent the oxidation of the contact plugs 13 a to 13 f , an SiON film 14 of 100 nm thickness is formed on the first interlayer insulating film 11 and the contact plugs 13 a to 13 f by the plasma CVD method using silane (SiH 4 ). In addition, an SiO 2 film 15 of 150 nm thickness is formed on the SiON film 14 by the plasma CVD method using TEOS and the oxygen as the reaction gas. The SiON film 14 has a function for preventing entering of the moisture into the first interlayer insulating film 11 .

Then, in order to densify the SiON film 14 and the SiO 2 film 15 , these films are annealed at the temperature of 650° C. for 30 minutes at the atmospheric pressure in the nitrogen atmosphere.

Then, as shown in FIG. 1D , a Ti layer and a Pt (platinum) layer on the SiO 2 film 15 are formed in sequence to form a first conductive film 16 having a double-layered structure. The Ti layer and the Pt (platinum) layer are formed by the DC sputter method. In this case, a thickness of the Ti layer is set to about 10 to 30 nm, and a thickness of the Pt layer is set to about 100 to 300 nm. For example, the thickness of the Ti layer is set to 20 nm, and the thickness of the Pt layer is set to 175 nm. In this case, a film made of iridium, ruthenium, ruthenium oxide, iridium oxide, strontium ruthenium oxide (SrRuO 3 ), or the like may be formed as a first conductive film 16 .

Then, a lead zirconate titanate (PZT: Pb(Zr 1-x Ti x )O 3 ) film serving as a ferroelectric film 17 is formed on the first conductive film 16 by the RF sputter method to have a thickness of 100 to 300 nm, e.g. 200 nm.

Then, as the crystallizing process of PZT constituting the ferroelectric film 17 , RTA (Rapid Thermal Annealing) is carried out at the temperature of 650 to 850° C. for 30 to 120 seconds in the oxygen atmosphere. For example, PZT is annealed at the temperature of 700° C. for 60 seconds.

As the method of forming ferroelectric material, in addition to the above sputter method, there are the spin-on method, the sol-gel method, the MOD (Metal Organic Deposition) method, and the MOCVD method. Also, as the ferroelectric material, in addition to PZT, there are oxides such as the lead lanthanum zirconate titanate (PLZT), SrBi 2 (Ta x Nb 1-x ) 2 O 9 (where 0<x<1), Bi 4 Ti 2 O 12 , etc. In case DRAM is formed instead of the FeRAM, the high-dielectric material such as (BaSr)TiO 3 (BST), strontium titanate (STO), or the like may be employed in place of the above ferroelectric material

Then, an iridium oxide (IrO 2 ) film as a second conductive film 18 is formed on the ferroelectric film 17 by the sputter method to have a thickness of 100 to 300 nm. For example, the thickness of the second conductive film 18 is set to 200 nm. In this case, as the second conductive film 18 , platinum or strontium ruthenium oxide (SRO) may be employed.

Then, steps required to form a sectional structure shown in FIG. 1 E and FIG. 2C will be explained hereunder.

First, upper electrodes 18 a of a plurality of capacitors, which are arranged in a matrix fashion along the extending direction of the word lines WL and the extending direction of the bit lines, described later, are formed by patterning the second conductive film 18 . The upper electrodes 18 a are formed in vicinity of the p-well 3 a as many as the MOS transistors formed in the memory cell region A. FIG. 2C shows a sectional shape taken along a I—I line in FIG. 1 E.

Then, stripe-like dielectric films 17 a of the capacitors, which are connected in the direction of the word lines WL under a plurality of upper electrodes 18 a , are formed by patterning the ferroelectric film 17 .

The silicon substrate 1 is placed in the oxygen atmosphere, and then the oxygen preprocess annealing is applied by annealing the substrate 1 at the substrate temperature of 350° C. for 60 minutes.

Then, as shown in FIG. 1 F and FIG. 2D , a first protection film 19 made of alumina of 20 to 100 nm thickness, e.g., 50 nm thickness is formed on the upper electrodes 18 a , the dielectric films 17 a , and the first conductive film 16 by employing the RF sputter equipment. This alumina is formed by setting the RF power to 2 kW at the atmospheric pressure of 7.5 mTorr. In turn, the silicon substrate 1 is placed in the oxygen atmosphere, and then the oxygen process annealing is applied by annealing the substrate 1 at the substrate temperature of 700° C. for 60 minutes.

Then, stripe-like resist patterns (not shown) for covering the dielectric films 17 a and the upper electrodes 18 a in the direction of the word lines WL are formed on the first protection film 19 . Then, the first protection film 19 and the first conductive film 16 are etched in sequence by using the resist patterns as a mask. Accordingly, as shown in FIG. 1 G and FIG. 2E , lower electrodes 16 a of the capacitor, which are commonly used as wirings passing under a plurality of dielectric films 17 a , are formed of the stripe-like first conductive film 16 .

Each of the lower electrodes 16 a has a contact region 16 b that is not covered by the stripe-like dielectric film 17 a . Also, the first protection film 19 has a shape to cover the upper electrode 18 a , the dielectric film 17 a , and the lower electrode 16 a . The silicon substrate 1 is placed in the oxygen atmosphere after the lower electrodes 16 a are patterned, and then the process for improving the film quality of the ferroelectric film 17 is carried out at the substrate temperature of 650° C. for 60 minutes.

The lower electrode 16 a , the dielectric film 17 a , and the upper electrode 18 a formed by above steps constitute a ferroelectric capacitor 20 . In the memory cell region A, the ferroelectric capacitors 20 are formed in the same number as the MOS transistors.

Then, steps required to form a sectional structure shown in FIG. 1 H and FIG. 2F will be explained hereunder.

First, a second interlayer insulating film 21 of SiO 2 is formed on the overall surface to cover the ferroelectric capacitors 20 . The second interlayer insulating film 21 is formed by the steps of forming an SiO 2 film of 480 nm by a plasma-CVD using TEOS, forming an SOG (Spin-On-Glass) film of 100 nm, and then etching back 300 nm these films in depth.

Then, the second interlayer insulating film 21 and the first protection film 19 are patterned by the photolithography method. Thus, contact holes 21 a are formed on the upper electrodes 18 a of the ferroelectric capacitors 20 and also contact holes 21 b are formed on the contact regions 16 b of the lower electrodes 16 a , as shown in FIG. 2 F.

Then, the second interlayer insulating film 21 , the SiON film 14 , and the SiO 2 film 15 are patterned by the photolithography method, whereby contact holes 21 c are formed on the second contact plugs 13 b formed near both ends of the first p-well 3 a in the memory cell region A.

Then, a TiN film of 125 nm thickness, for example, is formed on the second interlayer insulating film 21 and in the contact holes 21 a to 21 c by the sputter method. Then, the TiN film is patterned by the photolithography method. Thus, as shown in FIG. 1H , in the memory cell region A, first local wirings 22 a , that electrically connect the second contact plugs 13 b formed near both ends of the first p-well 3 a and the upper electrodes 18 a via the contact holes 21 a , 21 c , are formed and also second local wirings 22 b , that are extended to peripheries of the lower electrodes 16 a via the contact holes 21 b formed on the contact regions 16 b of the lower electrodes 16 a are formed.

The first and second local wirings 22 a , 22 b are the first layer metal wirings.

Then, as shown in FIG. 1I , a second insulating protection film 23 made of alumina for covering entire surfaces of the first and second local wirings 22 a , 22 b and the second interlayer insulating film 21 is formed to have a thickness of 15 to 100 nm. If the film thickness of the second insulating protection film 23 becomes thicker, the imprint rate of the ferroelectric capacitor 20 can be improved. However, it becomes difficult to execute the etching containing the postprocessing when contact holes for connecting the second layer metal wirings, described later, and the substrate are formed in the second insulating protection film 23 . In this case, it is preferable that a film thickness of the second protection film 23 should be set to about 20 nm.

The second protection film 23 may be patterned by the photolithography method such that it has a shape for covering at least the upper electrodes 18 a , a shape for covering only the ferroelectric capacitor 20 , a shape for covering the overall region of the memory cell region A, or a shape for not-covering the peripheral circuit region B.

Then, as shown in FIG. 1 J and FIG. 2G , a third interlayer insulating film 24 made of SiO 2 is formed on the second protection film 23 by the plasma CVD method using the TEOS to have a thickness of 200 to 400 nm. Then, the dehydration process is carried out by annealing the third interlayer insulating film 24 at 350° C. in the plasma atmosphere using the N 2 O gas.

Then, respective films from the third interlayer insulating film 24 to the SiON film 14 are patterned in the memory cell region A by the photolithography method employing the resist pattern (not shown). Accordingly, as shown in FIG. 1K , a hole 24 a is formed on the first contact plug 13 a located at the center position of the first p-well 3 a , and also, as shown in FIG. 2H , a hole 24 b is formed on the second contact plugs 13 b located on the second p-well 3 b . At the same time, holes 24 c to 24 e are formed on respective contact plugs 13 c to 13 e in the peripheral circuit region B. In this case, as shown in FIG. 2H , a hole 24 f is formed on the second local wiring 22 b extended from the lower electrode 16 a of the ferroelectric capacitor 20 to the outside.

The holes 24 a to 24 f formed in the third interlayer insulating film 24 and the underlying films are formed by the step etching using the same dry etching equipment.

For example, the third interlayer insulating film 24 is etched under the conditions that the pressure in the etching atmosphere is set to 350 mTorr by introducing Ar, CF 4 , and C 4 F 8 into the atmosphere at 618 sccm, 67 sccm, and 32 sccm respectively, the RF power is set to 1 kW, and the etching time is set to 26 seconds. The second protection film 23 is etched under the conditions that the pressure in the etching atmosphere is set to 1000 mTorr by introducing Ar, CHF 3 , and CF 4 into the atmosphere at 596 sccm, 16 sccm, and 24 sccm respectively, the RF power is set to 900 W, and the etching time is set to 22 seconds. In addition, the second interlayer insulating film 21 , the SiON film 14 , and the SiO 2 film 15 are etched under the conditions that the pressure in the etching atmosphere is set to 350 mTorr by introducing Ar, CF 4 , and C 4 F 8 into the atmosphere at 618 sccm, 67 sccm, and 32 sccm respectively, the RF power is set to 1 kW, and the etching time is set to 60 seconds.

Since the second local wiring 22 b formed of TiN shown in FIG. 2H acts as the etching stopper in such etchings, the hole 24 f formed on the wiring 22 b becomes shallower than remaining holes 24 a to 24 e.

Then, a metal film having a quintuple-layered structure, that consists of a Ti film of 20 nm thickness, a TiN film of 50 nm thickness, an Al—Cu film of 600 nm thickness, a Ti film of 5 nm thickness, and a TiN film of 150 nm thickness in order of low level, is formed on the third interlayer insulating film 24 and in the holes 24 a to 24 f , and then this metal film is patterned by the photolithography method.

Accordingly, as shown in FIG. 1L , a bit line 25 a is formed in the memory cell region A, and also wirings 25 b to 25 d are formed in the peripheral circuit region B. The bit line 25 a in the memory cell region A is connected to the first contact plug 13 a on the first p-well 3 a via the hole 24 a . Also, the wirings 25 b to 25 d in the peripheral circuit region B are connected to the underlying contact plugs 13 c to 13 e via the holes 24 b to 24 d respectively. Also, as shown in FIG. 2I , a ground wiring 25 e is formed around the lower electrodes 16 a in the memory cell region A, and this ground wiring 25 e is connected to the contact plugs 13 f on the second p-well 3 b via the hole 24 b . In addition, as shown in FIG. 2I , a leading wiring 25 f is formed on the second local wiring 22 b extended from the contact region 16 b of the lower electrode 16 a , and this leading wiring 25 f is connected to the second local wiring 22 b via the hole 24 f.

The bit line 25 a , the wirings 25 b to 25 d , the ground wiring 25 e , and the leading wiring 25 f act as the second layer metal wiring.

Then, steps required to obtain a situation shown in FIG. 1 M and FIG. 2J will be explained hereunder.

First, a fourth interlayer insulating film 26 made of SiO 2 and having a thickness of 2.3 μm is formed on the third interlayer insulating film 24 , the bit line 25 a , the wirings 25 c to 25 d , etc. by the plasma CVD method using the TEOS gas and the oxygen (O 2 ) gas.

Then, an upper surface of the fourth interlayer insulating film 26 is planarized by polishing it by virtue of the CMP method.

In turn, the silicon substrate 1 is placed in the low pressure atmosphere, then the N 2 O gas and the N 2 gas are plasmanized in the atmosphere, and then the fourth interlayer insulating film 26 is exposed to the plasma for a time in excess of three minutes, preferably four minutes, while setting the substrate temperature to less than 450° C., e.g., 350° C. Accordingly, the moisture that entered into the fourth interlayer insulating film 26 in polishing can be discharged to the outside and the moisture is difficult to enter into the fourth interlayer insulating film 26 .

If cavities are generated in the fourth interlayer insulating film 26 , in some cases such cavities are exposed by polishing. Therefore, after the polishing, a cap layer (not shown) made of SiO 2 and having a thickness of more than 100 nm may be formed as an upper layer of the fourth interlayer insulating film 26 . The cap layer is formed by the plasma CVD method using the TEOS gas and then exposed to the N 2 O plasma while setting the substrate temperature to 350° C.

Then, holes 26 c , 26 e for upper plugs are formed on the second layer wiring 25 c in the peripheral circuit region B and the ground wiring 25 e in the memory cell region A by patterning the fourth interlayer insulating film 26 by virtue of the photolithography method.

Then, steps required to get a structure shown in FIG. 1 N and FIG. 2K will be explained hereunder.

First, a glue layer 27 having a double-layered structure consisting of Ti and TiN is formed by the sputter on the fourth interlayer insulating film 26 and in the plug holes 26 c , 26 e . Then, tungsten seeds (not shown) are formed on the glue layer 27 by the CVD method using the tungsten hexafluoride (WF 6 ) gas and the silane (SiH 4 ) gas. In addition, a tungsten film 28 is formed on the glue layer 27 by using the WF 6 gas, the SiH 4 gas, and the hydrogen (H 2 ) gas at the growth temperature of 430° C. Accordingly, the glue layer 27 and the tungsten film 28 are filled into the plug holes 26 c , 26 e.

Then, the tungsten film 28 is removed from an upper surface of the fourth interlayer insulating film 26 by the CMP method or the etching-back, but such tungsten film 28 is left only in the plug holes 26 c , 26 e . Here, it is not needed to remove the glue layer 27 on the fourth interlayer insulating film 26 . In FIG. 1 N and FIG. 2K , there is shown the case where the glue layer 27 is left on the fourth interlayer insulating film 26 .

As a result, an upper plug (via) 28 c is formed of the tungsten film 28 and the glue layer 27 that are left in the plug hole 26 c on the wiring 25 c in the peripheral circuit region B. Also, a plug 28 e is formed of the tungsten film 28 and the glue layer 27 that are left in the plug hole 26 e on the wiring 25 e in the memory cell region A.

Then, steps required to get a structure shown in FIG. 10 and FIG. 2L will be explained hereunder.

First, an Al—Cu film 29 a of 600 nm thickness and a TiN film 29 b of 100 nm thickness are formed in sequence on the glue layer 27 and the plugs 28 c , 28 e . In case the glue layer 27 is removed from the upper surface of the fourth interlayer insulating film 26 , a TiN film (not shown) is formed under the Al—Cu film 29 a.

Then, the TiN film 29 b , the the Al—Cu film 29 a , and the glue layer 27 are patterned. Thus, a wiring 30 a connected to the plug 28 c in the peripheral circuit region B is formed, while a third protection film 30 b for covering the ferroelectric capacitor 20 is formed in the memory cell region A. As shown in FIG. 2L , the third protection film 30 b is electrically connected to the silicon substrate 1 via the upper plug 26 e , the ground wiring 25 e , the contact plugs 13 f , and the second p-well 3 b . In this case, the wiring 30 a connected to the via 28 c in the peripheral circuit region B is the third layer metal wiring.

By the way, arrangement of the upper plug 28 e and the contact plugs 13 f connected to the third protection film 30 b is schematically shown in FIG. 3 .

In FIG. 3 , arrangement of the hybrid FeRAM containing the logic circuit formed in one chip is shown, wherein the peripheral circuit region B is arranged adjacent to the memory cell region A in the FeRAM circuit. The ferroelectric capacitor cell is partitioned into several blocks in the memory cell region A, and the upper plugs 28 e are arranged between the blocks. Sixteen upper plugs 28 e are formed in the FeRAM in the present embodiment. The upper plugs 28 c are connected to the ground wiring 25 e . The contact plugs 13 f connected under the ground wiring 25 e are formed at different positions from the upper plugs 28 e and are formed in the larger number than the upper plugs 28 e (e.g., about 250 pieces).

In the region of the memory cell region A indicated by a broken line shown in FIG. 3 , the third protection film 30 b has a planar shape shown in FIG. 4 , for example, to cover the entire of the memory cell region A. In this case, the insulating films are omitted in FIG. 4 .

After the third protection film 30 b is formed as above, as shown in FIG. 1 P and FIG. 2M , a first cover insulating film 31 made of SiO 2 and having a thickness of 200 nm thickness, for example, is formed by the plasma CVD method using the TEOS to cover the third protection film 30 b and the third layer wiring 30 a . In addition, a second cover insulating film 32 made of silicon nitride is formed on the first cover insulating film 31 by the plasma CVD method using the silane and the ammonium to have a thickness of 500 nm, for example.

According to the above steps, a basic structure of the FeRAM including the ferroelectric capacitor 20 is formed.

The constituent material of the first protection film 19 or the second protection film 23 is not limited to the alumina, and thus the insulating material that is hard to permeate the hydrogen, e.g., PZT, TiO 2 , AlN, Si 3 N 4 , or SiON may be employed. Also, a wiring having a two layers or more may be formed between the second protection film 23 and the third protection film 30 b.

According to the above embodiments, since the ferroelectric capacitor 20 is covered with the second protection film 23 formed of alumina when the tungsten film 28 constituting the upper plugs 28 c , 28 e is formed, the degradation due to the reducing gas employed in forming the tungsten can be prevented. Also, since the ferroelectric capacitor 20 is covered with the first and second protection films 19 , 23 and the third protection film 30 b made of metal during when the second cover insulating film 32 made of silicon nitride is formed, the degradation of the ferroelectric capacitor 20 due to the reducing gas used to form the silicon nitride can be prevented.

Effects for preventing the reduction of the ferroelectric capacitor 20 by the first, second, and third protection films 19 , 23 , 30 b , etc. will be explained in detail in the following.

(i) The Influence of Difference in the Layer Number of the Protection Films on the Imprint Characteristic of the Ferroelectric Capacitor

When it was examined how the imprint characteristic of the ferroelectric capacitor 20 is affected by changing the combination of the first, second, and third protection films 19 , 23 , 30 b , results given in Table 2 were derived.

TABLE 2
The imprint characteristic of the ferroelectric capacitor when the
layer number of the protection films is changed (5 V evaluation)
film structure                  ΔQ                                                                        (88)                                                                     [μC/cm                                                                        2                                                                    ]
Only the first protection film  5.4
Only the second protection film 9.8
Only the third protection film  18.0
The first protection film +     2.4
the second protection film
The first protection film +     2.5
the third protection film
The first protection film +     0.1
the second protection film +
the third protection film

In Table 2, ΔQ( 88 ) denotes a value obtained by subtracting the Q( 88 ) obtained after process-out from the Q( 88 ) obtained immediately after the ferroelectric capacitor 20 is formed. That is, the smaller ΔQ( 88 ) shows the less process degradation.

Where Q indicates difference in the polarization charge when the opposite signal is loaded into two pairs of ferroelectric capacitors of 2-transistor/2-capacitor type, and the Q( 88 ) indicates a Q value obtained after the ferroelectric capacitor is baked at 150° C. for 88 hours. This Q is measured by applying the voltage of 5 V to the ferroelectric capacitor 20 .

In Table 2, if only the first protection film 19 is employed without the formation of the second and third protection films 23 , 30 b , ΔQ( 88 ) becomes 5.4 μc/cm 2 . Thus, it can be understood that the process degradation is caused in the ferroelectric capacitor 20 .

Also, if the cases where any one of the first, second, and third protection films 19 , 23 , 30 b is employed respectively are compared with each other, ΔQ( 88 ) of the first protection film 19 is smallest and then ΔQ( 88 ) is increased in the order of the second protection film 23 and the third protection film 30 b , and thus the imprint characteristic becomes worse.

Therefore, it can be understood that, if only one protection film is selected, it is advantageous to form the protection film at as close the position as possible to the ferroelectric capacitor 20 .

In the case of the single layer, the third protection film 30 b that has the small process degradation suppressing effect can reduce ΔQ( 88 ) by 80 percent or more if it is used together with the first protection film 19 , and thus the imprint characteristic can be improved largely. Similarly, the second protection film 23 can also reduce ΔQ( 88 ) by 70 percent or more if it is used together with the first protection film 19 . In this manner, if the first protection film 19 as the protection film formed immediately on the ferroelectric capacitor 20 is combined with the second protection film 23 or the third protection film 30 b formed over the first protection film 19 , the large process degradation suppressing effect can be achieved respectively rather than the case of the single protection film.

In addition, if all the first, second, and third protection films 19 , 23 , 30 b are combined together, ΔQ( 88 ) becomes 0.1 μC/cm 2 . Thus, there can be achieved such a large merit that the process degradation in forming the multi-layered wiring on the ferroelectric capacitor 20 can be completely suppressed.

In the case of the single layer, ΔQ( 88 ) of the third protection film 30 b is about two times large the second protection film 23 .

Assume that the simple protection film combination effect can be achieved, it is expected that a Q value obtained when the first and third protection films 19 , 30 b are employed is larger than a Q value obtained when the first and second protection films 19 , 23 are employed.

However, the Q value obtained when the first and second protection films 19 , 23 are employed is almost similar to the Q value obtained when the first and third protection films 19 , 30 b are employed. Accordingly, there is caused the effect that cannot be simply expected based on the combination of the first, second, and third protection films 19 , 23 , 30 b.

Because the film formed under the protection films 19 , 23 , 30 b is brought into the baking-in-a-casserole state by the heat applied by the process for forming the overlying film on the protection films 19 , 23 , 30 b , the moisture contained in the interlayer insulating film is spread downward to cause the degradation of the ferroelectric capacitor 20 .

Accordingly, when two protection films are employed, the first protection film 19 that can block the moisture coming in from the immediate upper side of the ferroelectric capacitor 20 fills the important role. In addition, it is indispensable to dehydrate completely respective interlayer insulating films formed between the first protection film 19 and the second protection film 23 or the third protection film 30 b . Therefore, in the present embodiment, the N 2 O plasma annealing at 350° C. that has the excellent dehydration effect is employed in the dehydration process of the second interlayer insulating film 21 being put between first protection film 19 and the second protection film 23 and the dehydration process of the third and fourth interlayer insulating films 24 , 26 being put between first protection film 19 and the third protection film 30 b.

According to Table 2, when the first protection film 19 is formed under the third protection film 30 b , the effect for blocking the moisture contained in the interlayer insulating films by the first protection film 19 appears remarkably.

With the above, the reason for that ΔQ( 88 ) obtained when a combination of the first protection film 19 and the second protection film 23 is employed is almost similar to ΔQ( 88 ) obtained when a combination of the first protection film 19 and the third protection film 30 b is employed is that such an effect that the first protection film 19 can suppress the influence of the moisture contained in the interlayer insulating films between two protection films can be achieved.

However, the degradation of the ferroelectric capacitor 20 caused when the second cover film 31 is formed by the reducing gas cannot be satisfactorily prevented only by the first protection film 19 .

Therefore, in order to suppress satisfactorily the degradation of the ferroelectric capacitor 20 , the first protection film 19 is indispensable and also at least one of the second protection film 23 and the third protection film 30 b is needed.

(ii) The Influence of Difference in Methods of Forming a Film of Alumina Constituting the Protection Film on the Imprint Characteristic of the Ferroelectric Capacitor

The influence of difference in methods of forming the alumina constituting the first and second protection films 19 , 23 on the imprint characteristic was given in Table 3.

TABLE 3
The influence of difference in alumina film forming
methods on the imprint characteristic (3 V evaluation)
	RF alumina	Helicon alumina
	Q (88)	Q rate	Q (88)	Q rate
Film structure	[μC/cm 2 ]	[%]	[μC/cm 2 ]	[%]
The first protection film	11.6	−8.0	10.0	−9.3
The first protection film +	12.8	−7.0	16.7	−4.4
the second protection film
The first protection film +	16.4	−5.1	13.6	−6.2
the third protection film
The first protection film +	18.4	−3.1	19.9	−2.4
the second protection film +
the third protection film

In Table 3, the RF alumina means alumina that is formed by the RF sputter equipment, and the helicon alumina means alumina that is formed by the helicon sputter equipment. The helicon sputter equipment can form the dense alumina rather than the case where the RF sputter equipment is employed since it has a structure in which the RF coil is arranged over the target to enhance the plasma density.

As the alumina film forming conditions by the helicon sputter equipment used in the experiment of Table 3, the pressure in the chamber is set to 1 mTorr, the RF power applied to the target is set to 600 W, the power applied to the RF coil is set to 60 W, the aluminum target is used as the target, and an argon gas and an oxygen (O 2 ) gas are introduced into the chamber at flow rates of 20 sccm and 7.6 sccm respectively.

In Table 3, Q is measured by applying the low voltage of 3 V to the ferroelectric capacitor 20 . The evaluation at the low voltage of 3 V is performed in view of the low voltage drive of the FeRAM.

When the evaluation of the ferroelectric capacitor 20 by applying the low voltage of 3 V is carried out, in the case of the RF alumina, the imprint characteristic is not so improved even if the second protection film 23 is formed on the first protection film 19 . In contrast, in the case of the helicon alumina, the improvement of the imprint characteristic appears if the second protection film 23 is formed on the first protection film 19 .

In the structure in which all the first, second, and third protection films 19 , 23 , 30 b are employed, the helicon alumina is superior in the imprint characteristic of the ferroelectric capacitor 20 to the RF alumina.

According to the above, it can be understood that, when the FeRAM is operated at the low voltage, it is advantageous to form the first protection film 19 and the second protection film 23 by using not the RF alumina but the helicon alumina.

(iii) The Influence of Difference in Film Thickness of the Second Protection Film on the Imprint Characteristic of the Ferroelectric Capacitor

When the influence of difference in film thickness of the second protection film 23 on the imprint characteristic of the ferroelectric capacitor was examined, results shown in Table 4 were derived.

TABLE 4
The influence of difference in film thickness of the
second protection film on the imprint characteristic (5 V evaluation)
Film thickness of the
second protection film	Q (88) [μC/cm 2 ]	Q rate [%]
0 nm	16.3	−6.4
15 nm	19.5	−4.8
20 nm	19.8	−4.7
50 nm	20.3	−4.3
70 nm	20.9	−4.8

Table 4 shows the evaluation when the second protection film 23 is formed by the RF alumina. The third protection film 30 b is not employed. In addition, Q is measured by applying the voltage of 5 V to the ferroelectric capacitor 20 .

According to Table 4, if the film thickness exceeds 15 nm, the imprint characteristic of the ferroelectric capacitor 20 can be apparently improved in contrast to the case where the second protection film 23 is not formed (film thickness=0). The larger film thickness is better, but there is no significant difference in the values Of Q( 88 ) and Q rate. Also, since the coverage is not so good when the film thickness is set to 10 nm, there is the possibility that the protection film cannot completely cover the ferroelectric capacitor 20 that has a large level difference on the surfaces.

(iv) The Influence of Difference in Potential of the Third Protection Insulating Film on the Retention Characteristic of the Ferroelectric Capacitor

The influence of difference in potential of the third protection film 30 b made of the conductive film on the retention characteristic of the ferroelectric capacitor 20 is shown in Table 5.

TABLE 5
The influence of potential of the third protection
film on the retention characteristic
(plastic package evaluation)
150 □ high-temperature
holding test	24 h	72 h	168 h	288 h	504 h	1008 h
(A)	Defect number/	0/50	0/50	0/50	0/50	0/50	13/50
	test piece number
	Fraction defective	0.0%	0.0%	0.0%	0.0%	0.0%	26.0%
(B)	Defect number/	0/50	0/50	0/50	0/50	0/50	1/50
	test piece number
	Fraction defective	0.0%	0.0%	0.0%	0.0%	0.0%	2.0%
(A) Third protection film provided/floating potential
(B) Third protection film provided/earth potential

As described above, the third protection film 30 b is formed of the same metal film as the third layer wiring 30 a . Also, as shown in FIG. 3 , the ferroelectric capacitor 20 has a structure that it is covered with the third protection film 30 b.

The retention characteristic was evaluated under two states, i.e., the case where the third protection film 30 b is electrically connected to the silicon substrate 1 and is set to the earth potential, and the case where the third protection film 30 b is not connected to the via 28 e and is set to the floating potential.

The retention characteristic was evaluated depending upon whether or not the signal of the ferroelectric capacitor 20 being left at the high-temperature atmosphere of 150° C. can be normally loaded/read.

The actual devices of 50 chips having respective structures were incorporated into the plastic package (not shown). In the semiconductor memory device formed in the chip used in the experiment in Table 5, the structure in which the second protection film 23 is not formed is employed.

As shown in Table 5, under both conditions that the third protection film 30 b is set to the floating potential and the third protection film 30 b is set to the earth potential, the signal of the ferroelectric capacitor 20 was able to be loaded/read without problem in the high-temperature holding until 504 hours.

However, when the holding time exceeds 1000 hours, the fraction defective was abruptly increased in the case where the third protection film 30 b is set to the floating potential whereas one defective chip is produced in the case where the third protection film 30 b is set to the earth potential. But apparently the fraction defective could be reduced rather than the case where the third protection film 30 b is set to the floating potential.

It may be supposed that there are two reasons that the fraction defective of the ferroelectric capacitor 20 can be reduced by setting the third protection film 30 b to the earth potential in this manner.

First, as set forth in Patent Application Publication (KOKAI) Hei 7-153921 and Patent Application Publication (KOKAI) Hei 2-5416, the permeation of the moisture can be suppressed. Accordingly, the degradation of the ferroelectric capacitor 20 can be prevented. Exactly speaking, this is because entering of the hydrogen ion used in forming the second cover film 32 made of silicon nitride can be suppressed. The prevention of the entering of the “hydrogen ion” is different in the strict meaning from the prevention of the permeation of the “hydrogen atoms” and the “moisture”, as set forth in Patent Application Publication (KOKAI) Hei 7-153921.

Second, the mutual coupling that is caused between the bit line 25 a and the bit line 25 a via the third protection film 30 b can be eliminated by setting the third protection film 30 b to the earth potential, and thus the effect of suppressing the fluctuation in potential of the bit line 25 a can be achieved. In other words, when the third protection film 30 b is set to the floating potential, the large fraction defective of the ferroelectric capacitor 20 is caused by the fluctuation in potential of the bit line 25 a . In other words, the fluctuation in potential of the bit line 25 a affects the retention characteristic with the degradation of the ferroelectric capacitor 20 .

Therefore, the mechanism applied to suppress the mutual coupling between the bit lines 25 a by the third protection film 30 b when such third protection film 30 b is set to the earth potential is different to that applied “to remove the charged charges” and “to remove effectively the static electricity”, as set forth in Patent Application Publication (KOKAI) Hei 7-153921 and Patent Application Publication (KOKAI) Hei 2-5416.

In this case, in Patent Application Publication (KOKAI) Hei 7-153921 and Patent Application Publication (KOKAI) Hei 2-5416, no recitation about the dehydration process of the interlayer insulating films under the plate and the formation of the bit line under the plate that is kept as the earth potential is given.

As described above, it has been found that, in order to improve the retention performance of the ferroelectric capacitor 20 , it is effective to set the potential of the third protection film 30 b to the earth level.

(v) The Pattern of the Third Protection Film

In FIG. 4 , the third protection film 30 b has a shape to cover the entire region of the memory cell region A. However, since the third protection film 30 b must cover at least respective ferroelectric capacitors 20 , it may be formed to have shapes shown in FIG. 5 , FIG. 6 and FIG. 7 .

The third protection film 30 b made of metal shown in FIG. 5 is formed in parallel with the lower electrodes 16 a of the ferroelectric capacitors 20 to cover the upper electrodes 18 a . Then, the third protection film 30 b is at the earth potential by the structure shown in FIG. 2 M.

The third protection film 30 b made of metal shown in FIG. 6 is formed in parallel with the bit lines 25 a to cover at least the upper electrodes 18 a of the ferroelectric capacitors 20 . Then, the third protection film 30 b is at the earth potential by the structure shown in FIG. 2 M. FIG. 5 , FIG. 6 , and FIG. 7 are depicted while omitting the insulating films and the p-well.

The third protection film 30 b made of metal shown in FIG. 7 is formed to individually cover at least the upper electrodes 18 a of the ferroelectric capacitors 20 . Then, the third protection film 30 b is at the earth potential by the structure shown in FIG. 2 M.

According to the third protection films 30 b shown in FIG. 5 , FIG. 6 or FIG. 7 , the third protection films 30 b having all shapes are set to the earth potential. Therefore, like the case of the shape shown in FIG. 4 , the retention performance of the ferroelectric capacitors 20 can be improved by preventing the enter of the reducing gas into the ferroelectric capacitors 20 and the preventing the mutual coupling between the bit lines 25 a.

As described above, according to the present invention, the surface (at least the upper electrodes and their peripheral regions) of the ferroelectric or high-dielectric capacitors is covered with the first protection film, then the second protection film is formed to cover the capacitors on the first wirings formed on the capacitors, then the second wirings are formed over the second protection film, then the third protection film is formed to cover the ferroelectric capacitors over the second wirings, and then the third protection film is set to the earth potential, whereby the first protection film and the second protection film, or the first protection film and the third protection film are employed.

According to this, even when the insulating films and the conductive films are formed or etched over the ferroelectric or high-dielectric capacitors by using the reducing atmosphere, the ferroelectric or high-dielectric films of the capacitors can be protected from the reducing atmosphere by the first protection film, the second protection film, or the third protection film underlying the films which are subjected to these processes. Therefore, the imprint characteristic of the ferroelectric capacitors can be improved and also the retention performance peculiar to the FeRAM can be improved.

Claims

1. A semiconductor device comprising:a capacitor including a lower electrode formed over a semiconductor substrate, a dielectric film formed of ferroelectric material or high-dielectric material, and an upper electrode; a first capacitor protection film formed above the capacitor; a first insulating film formed above the first capacitor protection film; a first wiring formed above the first insulating film; a second insulating film formed above the first wiring and first insulating film; a second wiring formed above the second insulating film; a third insulating film formed above the second wiring and the second insulating film; and at least one of a second capacitor protection film and a third capacitor protection film, the second capacitor protection film formed above the first wiring to cover the capacitor and formed under the second insulating film, and the third capacitor protection film formed above the third insulating film and above at least the upper electrode of the capacitor and set to an earth potential.

2. A semiconductor device according to claim 1, wherein the first capacitor protection film is formed any one of alumina, PZT, titanium oxide, aluminum nitride, silicon nitride, and silicon nitride oxide.

3. A semiconductor device according to claim 1, wherein the second capacitor protection film is formed any one of alumina, PZT, titanium oxide, aluminum nitride, silicon nitride, and silicon nitride oxide.

4. A semiconductor device according to claim 3, wherein the alumina of the second capacitor protection film has a film thickness of more than 15 nm.

5. A semiconductor device according to claim 1, wherein the third capacitor protection film has a laminated layer structure consisting of aluminum containing films including a titanium film or a titanium nitride film.

6. A semiconductor device according to claim 1, wherein the second capacitor protection film covers an overall memory cell region in which the capacitors are arranged.

7. A semiconductor device according to claim 1, wherein the third capacitor protection film is formed in every block or an entire of a region in which the capacitors are arranged.

8. A semiconductor device according to claim 1, wherein a third wiring made of a same conductive film which is patterned as the third capacitor protection film is formed on the third insulating film.

9. A semiconductor device according to claim 8, wherein the third wiring is connected to the second wiring via a plug buried in the third insulating film.

10. A semiconductor device according to claim 1, wherein dehydration process is applied to the second insulating film and the third insulating film respectively.

11. A semiconductor device according to claim 10, wherein the dehydration process is executed by N2O plasma annealing.

12. A semiconductor device according to claim 1, wherein the capacitor is formed on a fourth insulating film that covers a transistor having an impurity diffusion layer formed in the semiconductor substrate, andthe upper electrode of the capacitor is connected to the impurity diffusion layer via the first wiring.