WO2008023409A1 - Method for manufacturing semiconductor device - Google Patents

Method for manufacturing semiconductor device Download PDF

Info

Publication number
WO2008023409A1
WO2008023409A1 PCT/JP2006/316423 JP2006316423W WO2008023409A1 WO 2008023409 A1 WO2008023409 A1 WO 2008023409A1 JP 2006316423 W JP2006316423 W JP 2006316423W WO 2008023409 A1 WO2008023409 A1 WO 2008023409A1
Authority
WO
WIPO (PCT)
Prior art keywords
film
insulating film
capacitor
interlayer insulating
forming
Prior art date
Application number
PCT/JP2006/316423
Other languages
French (fr)
Japanese (ja)
Inventor
Katsuyoshi Matsuura
Original Assignee
Fujitsu Microelectronics Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujitsu Microelectronics Limited filed Critical Fujitsu Microelectronics Limited
Priority to PCT/JP2006/316423 priority Critical patent/WO2008023409A1/en
Publication of WO2008023409A1 publication Critical patent/WO2008023409A1/en

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L28/00Passive two-terminal components without a potential-jump or surface barrier for integrated circuits; Details thereof; Multistep manufacturing processes therefor
    • H01L28/40Capacitors
    • H01L28/55Capacitors with a dielectric comprising a perovskite structure material
    • HELECTRICITY
    • H10SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10BELECTRONIC MEMORY DEVICES
    • H10B53/00Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory capacitors
    • HELECTRICITY
    • H10SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10BELECTRONIC MEMORY DEVICES
    • H10B53/00Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory capacitors
    • H10B53/30Ferroelectric RAM [FeRAM] devices comprising ferroelectric memory capacitors characterised by the memory core region

Definitions

  • the present invention relates to a method for manufacturing a semiconductor device.
  • Flash memories and ferroelectric memories are known as nonvolatile memories that can store information even when the power is turned off.
  • the flash memory has a floating gate embedded in the gate insulating film of an insulated gate field effect transistor (IGFET), and information is stored by storing charges representing stored information in the floating gate.
  • IGFET insulated gate field effect transistor
  • the ferroelectric memory is also called FeRAM (Ferroelectric Random Access Memory), and stores information using the hysteresis characteristic of the ferroelectric film provided in the ferroelectric capacitor.
  • FeRAM Feroelectric Random Access Memory
  • the ferroelectric film generates polarization according to the voltage applied between the upper electrode and the lower electrode of the capacitor, and the spontaneous polarization remains even if the voltage is removed.
  • this spontaneous polarization is also reversed, and information is written to the ferroelectric film by making the direction of the spontaneous polarization correspond to “1” and “0”.
  • FeARM has the advantage that the voltage required for writing is lower than in flash memory and that writing can be performed faster than in flash memory.
  • PZT materials such as La-doped PZT (PLZT),
  • Bi layer structure compounds such as SBT (SrBi Ta O) and SBTN (SrBi (Ta, Nb) O) are used.
  • ferroelectric thin films are easily reduced by hydrogen in the process of forming an interlayer insulating film on the capacitor and the like, and the ferroelectric characteristics such as the residual polarization charge amount are likely to deteriorate.
  • the ferroelectric film In order to prevent such deterioration of the ferroelectric film, it normally has a key functioning as a hydrogen barrier film.
  • the capacitor is covered with a capacitor protection insulating film.
  • 0.35 m FeRAM designed with a minimum dimension of 0.35 ⁇ m uses an alumina (A1 0) film formed by sputtering as such a capacitor protective insulating film.
  • Patent Document 1 discloses that an alumina film having a film density exceeding 2.7 gZcm 3 is formed so as to cover all the capacitors covered by a platter structure, thereby providing hydrogen or the like. It is described that the reducing gas is prevented from entering from the lateral direction of the capacitor and the reduction of the ferroelectric film is prevented. In this case, by adopting RF (Radio Frequency) sputtering using an alumina target, an amorphous alumina film with fewer particles can be formed. Further, since the gas is not contained in the film formation atmosphere, the ferroelectric film due to the film formation of the alumina film is not deteriorated.
  • RF Radio Frequency
  • the CVD method provides better step coverage of the alumina film than the sputtering method, an alumina film having a sufficient thickness on the side surface of the capacitor even if the miniaturization advances and the interval between the capacitors becomes narrower.
  • the lateral force of the capacitor can also effectively prevent hydrogen from entering.
  • the alumina film formed by the CVD method is preferably formed as a capacitor protective insulating film in a stacked FeRAM that is advantageous for miniaturization of capacitors.
  • the alumina film formed by the CVD method as described above Is considered essential.
  • TMA Tri-Methyl Aluminum
  • H 0 water
  • FIG. 1 is a schematic diagram showing a mechanism for forming an alumina film by atomic layer deposition (ALD), which is a kind of CVD method.
  • ALD atomic layer deposition
  • An alumina film is formed by repeating a series of cycles A) to D).
  • the capacitor dielectric film may be reduced and deteriorated by moisture during heat treatment such as the recovery annealing of the capacitor.
  • Patent Document 2 when an alumina film is formed by an ALD method as a capacitor protective insulating film of FeRAM, ozone (0
  • ozone is a hydrogen-free oxidant, it is possible to prevent hydrogen and moisture from adsorbing to the capacitor dielectric and moisture from remaining in the alumina film when the alumina film is formed.
  • an alumina film is formed along the surface of the capacitor.
  • an alumina film is also formed on the interlayer insulating film covering the capacitor.
  • an alumina film is formed as a capacitor protection insulating film on the planarized surface of an interlayer insulating film. Since the alumina film is formed on a flat surface, it does not require excellent coverage characteristics and can be formed by sputtering.
  • a reaction bias is applied by applying a large bias voltage to a semiconductor substrate. Sputtering and film-forming at the same time proceed, and the space between narrow capacitors can be filled with an interlayer insulating film while suppressing the generation of “soot”.
  • Patent Document 1 Japanese Patent Laid-Open No. 2001-44375
  • Patent Document 2 JP 2004-193280 A
  • Patent Document 3 Japanese Unexamined Patent Publication No. 2006-49795
  • An object of the present invention is to provide a method for manufacturing a semiconductor device capable of preventing deterioration of a capacitor dielectric film.
  • a method of manufacturing a semiconductor device includes a step of forming an interlayer insulating film by a CVD method and a step of annealing the interlayer insulating film in an atmosphere containing ozone or oxygen.
  • the interlayer insulating film is formed by a plasma CVD method (HDPC VD method) in which a bias voltage is applied to the semiconductor substrate side.
  • the interlayer insulating film formed in this way is excellent in embeddability, and can fill a narrow space between capacitors without generating “soot”, while water that can deteriorate the capacitor dielectric film. Contains a lot in the film.
  • the interlayer dielectric film is annealed in an atmosphere containing ozone or oxygen, and the interlayer dielectric film is dehydrated, whereby the capacitor dielectric film is reduced by the moisture in the interlayer dielectric film. Prevent deterioration.
  • the oxidizing power of the atmosphere increases, so that the effect of dehydration can be expected even when the annealing substrate temperature is lowered, and the substrate temperature is increased. This can suppress the deterioration of the capacitor dielectric film.
  • the first capacitor protective insulating film formed so as to cover the capacitor functions to prevent a reducing substance such as hydrogen from entering the capacitor. If an alumina film is formed as the first capacitor protective insulating film by the ALD method with excellent coverage characteristics, a sufficient first capacitor protective insulating film should be formed on the side surface of the capacitor even if the semiconductor device is miniaturized. The lateral force can effectively prevent the reducing substance from entering the capacitor.
  • the annealing for the interlayer insulating film described above is preferably performed after the step of planarizing the surface of the interlayer insulating film by a CMP (Chemical Mechanical Polishing) method.
  • CMP Chemical Mechanical Polishing
  • the damage received by the capacitor dielectric film during the process is recovered by annealing the capacitor dielectric film in an oxygen-containing atmosphere after forming the above holes. Even if such a recovery annealing is performed, the interlayer insulating film is annealed for the purpose of dewatering as described above, and therefore the interlayer insulating film sandwiched between the first and second capacitor dielectric films. However, it does not become steamed due to moisture in the film. Therefore, the capacitor dielectric film deteriorates due to the moisture, or the second capacitor protective insulating film cracks. Can be prevented.
  • FIG. 1 is a schematic diagram showing a mechanism for forming an alumina film by the ALD method.
  • FIGS. 2 (a) to 2 (c) are cross-sectional views (part 1) in the middle of the manufacture of the semiconductor device according to the embodiment of the present invention.
  • 3 (a) and 3 (b) are cross-sectional views (part 2) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
  • 4 (a) and 4 (b) are cross-sectional views (part 3) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
  • FIGS. 5 (a) and 5 (b) are cross-sectional views (part 4) of the semiconductor device according to the embodiment of the present invention during manufacture.
  • 6 (a) and 6 (b) are cross-sectional views (part 5) in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
  • FIGS. 7A and 7B are cross-sectional views (part 6) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
  • FIGS. 8A and 8B are cross-sectional views (part 7) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
  • FIGS. 9 (a) and 9 (b) are cross-sectional views (part 8) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
  • FIG. 10 is a sectional view (No. 9) in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
  • FIG. 11 is a configuration diagram of an HDPCVD apparatus used in the embodiment of the present invention.
  • FIG. 12 is a diagram (No. 1) showing a result of TDS analysis performed in order to investigate the dehydration effect of an acid silicon film by annealing in the embodiment of the present invention.
  • FIG. 13 is a diagram (No. 2) showing a result of TDS analysis performed for examining the effect of dehydration of an oxide silicon film by annealing in the embodiment of the present invention.
  • FIG. 14 shows the removal of an oxide silicon film by annealing in an embodiment of the present invention. It is the figure (the 3) which shows the result of the TDS analysis performed in order to investigate the water effect.
  • FIG. 15 was conducted in order to investigate how the water content of the acid silicon film formed by the HDPCVD method depends on the film formation temperature in the embodiment of the present invention. It is the figure (the 1) which shows the result of TDS analysis.
  • FIG. 16 was carried out in order to investigate how the moisture content of an acid silicon film formed by the HDPCVD method depends on the film formation temperature in the embodiment of the present invention. It is the figure which shows the result of TDS analysis (the 2).
  • 2 to 10 are cross-sectional views in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
  • This semiconductor device is a stack type FeRAM advantageous for miniaturization, and is manufactured as follows.
  • a trench for STI Shallow Trench Isolation
  • an active region of a transistor is formed on the surface of an n-type or p-type silicon (semiconductor) substrate 1, and an insulating material such as silicon oxide is formed therein.
  • the element isolation insulating film 2 is formed by embedding the film.
  • the element isolation structure is not limited to STI, and the element isolation insulating film 2 may be formed by a LOCOS (Local Oxidation of Silicon) method.
  • an amorphous or polycrystalline silicon film is formed on the entire upper surface of the silicon substrate 1, and these films are patterned by photolithography to form two gate electrodes 5.
  • the two gate electrodes 5 described above are arranged in parallel at a distance from each other, and these gate electrodes 5 constitute a part of the word line.
  • n-type impurities are introduced into the silicon substrate 1 beside the gate electrode 5 by ion implantation using the gate electrode 5 as a mask to form first and second source / drain extensions 6a and 6b.
  • an insulating film is formed on the entire upper surface of the silicon substrate 1, and the insulating film is etched back to form an insulating sidewall 7 next to the gate electrode 5.
  • an oxide silicon film is formed by a CVD method.
  • n-type impurities are ion-implanted again into the silicon substrate 1 while using the insulating sidewalls 7 and the gate electrode 5 as a mask, so that the surface layer of the silicon substrate 1 on the side of the two gate electrodes 5 is obtained.
  • first and second source Z drain regions 8a and 8b spaced apart from each other are formed.
  • the active region of the silicon substrate 1 includes the first and second MOS transistors configured by the gate insulating film 4, the gate electrode 5, and the first and second source / drain regions 8a 8b.
  • TR TR is formed.
  • the refractory metal layer is heated to react with silicon.
  • a melting point metal silicide layer 9 is formed.
  • the refractory metal silicide layer 9 is also formed on the surface layer portion of the gate electrode 5, thereby reducing the resistance of the gate electrode 5.
  • a silicon nitride (SiN) film is formed to a thickness of about 200 on the entire upper surface of the silicon substrate 1 by plasma CVD, and this is used as the cover insulating film 10.
  • an acidic silicon film is formed on the cover insulating film 10 as a base insulating film 11 with a thickness of about 100 nm by a plasma CVD method using TEOS gas.
  • the upper surface of the base insulating film 11 is polished and planarized by a CMP (Chemical Mechanical Polishing) method.
  • the thickness of the base insulating film 11 is about 700 nm on the flat surface of the silicon substrate 1.
  • the cover insulating film 10 and the base insulating film 11 are patterned by photolithography to form contact holes on the first and second source / drain regions 8a 8b. Further, after forming a glue film (adhesion film) and a tungsten film in this contact hole in order, the excess glue film and tungsten film on the underlying insulating film 11 are polished and removed by CMP, and these are removed. The first and second conductive plugs 12a and 12b are left only in the contact holes. [0052] These first and second conductive plugs 12a and 12b are electrically connected to the first and second source / drain regions 8a and 8b, respectively.
  • the glue film is formed by forming a titanium film having a thickness of about 30 nm and a titanium nitride film having a thickness of about 50 nm in this order.
  • a titanium film is formed to a thickness of about 20 by sputtering on each of the base insulating film 11 and the first and second conductive plugs 12a and 12b.
  • the titanium film 20 ⁇ m is used as the base conductive film 21.
  • the substrate temperature is set at 650 ° C in a nitrogen atmosphere for the purpose of easily oxidizing and improving the acid resistance of the underlying conductive film 21 made of titanium.
  • RTA Rapid Thermal Anneal
  • the flow rate of nitrogen in this RTA is not particularly limited, but in this embodiment it is 10 slm (standard litter I min).
  • the unit slm represents a flow rate in a standard state (1.013 ⁇ 10 5 Pa, 0 ° C.).
  • the base conductive film 21 after nitriding in this way has a good crystallinity and is a film formed later on the base conductive film 21 It will function as a crystallinity improving film that improves the crystallinity of the film.
  • TiAIN titanium aluminum nitride
  • the conductive oxygen nore film 22 made of titanium aluminum nitride has an excellent oxygen permeation preventing function, and the first and second conductive plugs 12a and 12b below the surface are oxidized to cause contact failure. Play a role in preventing life.
  • an iridium film is formed as a first conductive film 23 on the conductive oxygen noria film 22 by a sputtering method to a thickness of about lOOnm.
  • the first PZT film and the second PZT film are respectively formed on the first conductive film 23 by MOCVD (Metal Organic CVD) with a substrate temperature of 620 ° C and a pressure of 5 Torr. Thickness 5nm, These PZT films are formed as a ferroelectric film 24 having a total film thickness of 120 nm.
  • MOCVD Metal Organic CVD
  • the first and second PZT films have the same composition. However, since the crystallinity of the PZT film becomes better when the film is formed at a low oxygen partial pressure, the first layer should be formed with the oxygen partial pressure in the film formation atmosphere lower than that of the second layer. To improve the crystallinity. However, if the second layer is deposited at a low oxygen partial pressure, the leakage current increases due to the increase in oxygen vacancies in the PZT film. Therefore, priority is given to reducing the leakage current for the second layer, giving higher oxygen than the first layer. Film is formed with partial pressure.
  • THF Tetra Hydro Furan
  • an iridium oxide film having a thickness of about 150 is formed on the ferroelectric film 24 by sputtering, and this iridium oxide film is used as the second conductive film 25.
  • An iridium film as a conductivity enhancement film 26 is formed on the film 25 by sputtering to a thickness of about 50 mm.
  • a titanium nitride film having a thickness of about 200 ⁇ m is formed on the conductivity improving film 26 by sputtering, and the titanium nitride film is used as the first mask material layer. 27. Further, an oxide silicon film is formed as a second mask material layer 28 on the first mask material layer 27 to a thickness of about 700 using a plasma CVD method using TEOS gas.
  • the second hard mask 28a is formed by patterning the second mask material layer 28 in an island shape.
  • the first hard mask 27a is formed by etching the first mask material layer 27 using the second hard mask 28a as a mask.
  • the dry etching gas is not particularly limited, but a mixed gas of HBr and oxygen is used as an etching gas for the first conductive film 23, the second conductive film 25, and the conductivity improving film 26. .
  • a mixed gas of chlorine and argon is used as an etching gas for the ferroelectric film 24 as an etching gas for the ferroelectric film 24 as an etching gas for the ferroelectric film 24 as an etching gas for the ferroelectric film 24, a mixed gas of chlorine and argon is used.
  • the conductive oxygen barrier film 22 has an etching resistance to the etching gas for the first conductive film 23 a conductive oxygen barrier is formed on the entire surface of the base conductive film 21 even after the etching is completed.
  • the membrane 22 remains.
  • the second hard mask 28a is removed by dry etching.
  • Etch 1 and the conductive oxygen noble film 22 to leave these films only under the capacitor dielectric film 24a.
  • This etching is performed by dry etching, and as the etching gas, for example, a mixed gas of argon and chlorine is used.
  • the first hard mask 27a is also etched by this etching gas, the first hard mask 27a is removed at the end of etching.
  • the lower electrode 23a composed of the base conductive film 21, the conductive oxygen barrier film 22, and the first conductive film 23 is connected to the first conductive plug 12a. It is formed to be electrically connected. And this lower electrode 23a, capacitor dielectric film
  • Capacitor Q is constituted by 24a and upper electrode 25a.
  • the hard mask 27a common to the lower electrode 23a, the capacitor dielectric film 24a, and the upper electrode 25a is used as an etching mask. There is no. Therefore, the capacitor Q can be miniaturized and high integration of FeRAM can be achieved.
  • an alumina film covering the capacitor Q is formed to a thickness of about 40 nm, and the alumina film is used as the first capacitor protection insulating film 39.
  • Alumina constituting the first capacitor protective insulating film 39 is excellent in hydrogen permeation preventing ability. Therefore, external hydrogen is blocked by the first capacitor protection insulating film 39, and deterioration of the capacitor dielectric film 24a due to hydrogen can be prevented.
  • the method for forming the first capacitor protective insulating film 39 is not particularly limited. However, since the interval between the adjacent capacitors Q is narrow, it is preferable to form the first capacitor protection insulating film 39 by the ALD method capable of forming a film having excellent coverage characteristics. By adopting the ALD method in this way, the first capacitor protective insulating film 39 having a sufficient thickness can be formed on the side surface of the capacitor Q, and the lateral force is also easy to prevent hydrogen from entering the capacitor Q. Become.
  • trimethylaluminum is used as the organoaluminum compound.
  • Hydrogen-free ozone is used as the oxidizing agent used with the organoaluminum compound.
  • the amount of hydrogen in the atmosphere of ALD decreases when ozone is used, and the capacitor dielectric film 24a deteriorates due to hydrogen when the first capacitor protection insulating film 39 is formed. Can be suppressed.
  • the first capacitor protective insulating film 39 may be formed by sputtering.
  • FIG. 11 is a configuration diagram of an HDPCVD apparatus used in this process.
  • a first high-frequency power source 104 is connected to a coil 103 provided above the chamber 100, and a second high-frequency power source 102 is further connected to the substrate platform 101.
  • the coil 103 is wound in a plane parallel to the main surface of the silicon substrate 1, and the cross section is shown in the drawing.
  • an oxide silicon film having a thickness that fills the space between the capacitors Q is formed on the first capacitor protection insulating film 39 as an interlayer insulating film 40. Since the interlayer insulating film 40 formed by HD PCVD has good embedding properties, even if the high integration progresses and the interval between the capacitors Q is narrowed, "s" is formed in the interlayer insulating film 40 between them. There is nothing to do.
  • the deposition temperature of the interlayer insulating film 40 is not limited to the above 300 ° C, but 250 ° C or more 350
  • the upper surface of the interlayer insulating film 40 is polished and flattened by the CMP method.
  • the film thickness of the interlayer insulating film 40 after CMP is about 300 nm on the upper electrode 25a.
  • the interlayer insulating film 40 formed by the HDPCVD method contains a large amount of moisture.
  • the survey results will be described later.
  • the moisture in the slurry permeates the interlayer insulating film 40 and the interlayer insulating film 40 contains a large amount of moisture.
  • the capacitor dielectric film 24a may be deteriorated.
  • the interlayer insulating film 40 is annealed and the interlayer insulating film 40 is dehydrated.
  • the annealing conditions are not particularly limited! In this embodiment, however, a mixed gas of ozone and oxygen generated by an ozonizer (ozone generator) is supplied to a chamber (not shown) at a flow rate of lOslm. Ozone concentration in the mixture gas is 200gZNm 3. Then, the annealing is performed for 30 minutes under a pressure of 0.133 kPa and a substrate temperature of 350 ° C. to 600 ° C., for example, 400 ° C.
  • the substrate temperature during annealing be higher than 600 ° C. because a dehydration reaction occurs remarkably in the interlayer insulating film 40 and a large amount of H 0 is generated.
  • the interlayer insulating film 40 once heated at the time of film formation is annealed at a substrate temperature lower than the film formation temperature, the dehydration effect due to the annealing cannot be expected.
  • the deposition temperature of the interlayer insulating film 40 is 250 ° C. to 350 ° C., the interlayer insulating film 40 is effectively formed by setting the substrate temperature at the above annealing to 350 ° C. or higher. Can be dehydrated.
  • a flat alumina film is formed on the interlayer insulating film 40 by sputtering to a thickness of about 50 mm, and the alumina film is formed as a second capacitor protective insulating film.
  • the film 41 is used.
  • the second capacitor protection insulating film 41 is formed on the planarized second interlayer insulating film 40, excellent coverage characteristics are not required, and can be formed by an inexpensive sputtering method as described above. .
  • the alumina constituting the second capacitor protective insulating film 41 is difficult to etch by a chemical reaction. Therefore, in order to facilitate the formation of holes in the second capacitor protective insulating film 41 in the etching process described later and to improve the hydrogen blocking performance, a dense and thin alumina film is used to form the second capacitor protective insulating film 41 using the ALD method. It is preferable to form as. In this case, the second capacitor protection insulating film 41 is formed as thin as about 30 °.
  • a plasma using TEOS gas is formed on the second capacitor protective insulating film 41.
  • a silicon oxide film is formed to a thickness of about 200 by CVD, and this silicon oxide film is used as a cap insulating film 42.
  • the first holes 40a are formed in these films on the upper electrode 25a by patterning each of the films 39-42.
  • the silicon substrate 1 is placed in a furnace (not shown) and the substrate temperature is set to 500 ° C in an oxygen atmosphere. I do.
  • the escape path of moisture in the interlayer insulating film 40 vaporized by this recovery annealing Is limited to the first hole 40a. Therefore, if the annealing process in FIG. 7B is omitted, a large amount of moisture in the interlayer insulating film 40 is vaporized during the recovery annealing, and the interlayer insulating film 40 and the capacitor Q below it are steamed by the vaporized moisture. As a result, the second capacitor protective insulating film 41 is cracked or the capacitor Q deteriorates.
  • the capacitor Q does not deteriorate even if the recovery annealing is performed! ,.
  • the substrate temperature in the annealing process of FIG. 7B is too low, the interlayer insulating film 40 is insufficiently dehydrated, and the second capacitor protective insulating film 41 is cracked during this recovery annealing. There is a fear. If this is a concern, it is preferable to set the substrate temperature of the annealer in Fig. 7 (b) as high as possible, for example, 500 ° C! /.
  • the second holes 40b are formed in these films on the second conductive plug 12b by patterning the films 39 to 42.
  • the second hole 40b made of alumina, which is difficult to chemically etch, is used.
  • Capacitor protection Insulating film 41 is etched.
  • the second capacitor protection insulating film 41 is thick, this etching becomes difficult, and it becomes difficult to process the holes 40a and 40b with high accuracy by the etching. Therefore, if this is a concern, as described above, the second capacitor protective insulating film 41 can be thinned to about 30 mm by the ALD method, which can form a dense film thinly. Preferably formed.
  • a titanium nitride film is formed as a glue film on the cap insulating film 42 and in the first and second holes 40a, 40b by a sputtering method.
  • tungsten film is formed on the glue film by the CVD method, and the first and second holes 40a and 40b are completely filled with this tungsten film.
  • the third conductive plug 45a is electrically connected to the upper electrode 25a of the capacitor Q.
  • the fourth conductive plug 45b is electrically connected to the second conductive plug 12b and constitutes a part of the bit line together with the second conductive plug 12b.
  • a metal multilayer film is formed on the cap insulating film 42 and each of the conductive plugs 45a and 45b by sputtering, and this metal multilayer film is patterned.
  • the metal wiring 47a and the conductive pad 47b for the bit line are formed.
  • a titanium film having a thickness of about 60, a titanium nitride film having a thickness of about 30, a copper-containing aluminum film having a thickness of about 400 nm, a titanium film having a thickness of about 5 nm, and a thickness of about A 70 nm titanium nitride film is formed in this order.
  • the cap insulating film 42 functions as an etching stopper film, thereby preventing the thin second capacitor protective insulating film 41 from being etched and diminishing the hydrogen blocking effect. can do.
  • the metal wiring 47a is formed on the second capacitor protection insulating film 41 without the cap insulating film 42, it is caused by the difference in stress between the second capacitor protection insulating film 41 and the metal wiring 47a. As a result, stress migration may occur in the metal wiring 47a.
  • the metal wiring 47a is formed on the cap insulating film 42 made of silicon oxide as described above, such stress migration can be suppressed and the reliability of the metal wiring 47a can be improved. Can do.
  • the interlayer insulating film 40 is formed by the HDPCVD method in the step of FIG. 6B, and the interlayer insulating film 40 is removed in the ozone-containing atmosphere in the step of FIG. 7B. Nealed.
  • TDS thermal desorption spectra
  • samples A and B are obtained by forming a silicon oxide film with a thickness of 50 Onm on a silicon wafer by a plasma CVD method using TEOS gas.
  • the CVD method is not an HDP CVD method but a normal plasma CVD method.
  • Sample A and sample B have different film formation conditions.
  • sample A was formed as an interlayer insulating film 40, it was a condition that had cracked during the recovery annealing.
  • Sample B is a condition that does not cause such cracks.
  • Sample C is a film in which a 500 nm thick silicon oxide film is formed on a silicon wafer by HDPCVD using the same film formation conditions as those of the interlayer insulating film 40 described above.
  • Sample D is obtained by annealing the same silicon oxide film as Sample C under the same conditions as in Fig. 7 (b).
  • Sample E was examined in the absence of a silicon wafer. Represents the background of
  • each sample A to D was obtained by cutting a silicon wafer into lcm x lcm squares for analysis.
  • Figure 12 shows only the spectrum with a mass number of 18 corresponding to H 0 in the TDS method.
  • the horizontal axis represents the temperature at TDS, and the vertical axis represents the ionic strength. However, the temperature on the horizontal axis indicates the stage temperature when the sampnore is installed, and does not represent the surface temperature of the sampnore.
  • sample C using HDPCVD is included in the film compared to samples A and B using normal plasma CVD. It can be seen that the amount of water produced is large.
  • FIG. 13 is a graph obtained by TDS analysis of samples A to E having the same structure as in FIG. 12, after leaving a silicon oxide film for one week.
  • Torn paper It can be seen that the dehydration peaks at around ° C and around 350 ° C increased. This is presumed to be because H 0 was further occluded in the macropores in the silicon oxide silicon film by leaving it to stand.
  • sample B without the above cracks, it is a dense film in which the dehydration peak does not increase even if left as it is. Therefore, the sample B silicon oxide film is suitable for the FeRAM interlayer insulation film, except that it has poor embedding characteristics.
  • FIG. 14 is a graph showing the results of the TDS analysis.
  • samples Dl and D2 were both subjected to annealing for dehydration on a silicon oxide film formed by the HDPCVD method.
  • the annealing conditions are different between sample D1 and sample D2.
  • sample D1 the substrate temperature was set to 500 ° C higher than that of sample D, and the silicon oxide film was annealed in an ozone-containing atmosphere.
  • the gas flow rate, pressure, and processing time are the same as Sample D.
  • Sample D2 was obtained by annealing an oxide silicon film at a substrate temperature of 500 ° C in a 100% oxygen atmosphere.
  • Sample Dl does not show a significant increase in the amount of dehydration on the high temperature side like Sample D. Dehydration on the high temperature side is caused by a dehydration reaction between OH groups of Si—OH bonds that exist in the silicon oxide film with little force. Therefore, it can be inferred from the above results that the Si—OH bonds in the film are reduced by raising the substrate temperature during annealing of the silicon oxide film.
  • ozone having a higher acid power than oxygen is added to the annealing atmosphere for the silicon oxide film as in this embodiment. Seems to be preferred.
  • sample D1 using ozone with sample D2 using oxygen, there is no significant difference between the profiles. This is presumably because, in sample D1, the substrate temperature during annealing was as high as 500 ° C, so ozone was killed during annealing of the silicon oxide film.
  • the substrate temperature during annealing when the substrate temperature during annealing is increased, the dehydration effect of the interlayer insulating film 40 is improved, while the moisture in the interlayer insulating film 40 passes through the first capacitor protective insulating film 39 and forms the capacitor dielectric film 24a. It is conceivable to diffuse to Therefore, the substrate temperature during annealing is preferably as low as possible in view of preventing deterioration of the capacitor dielectric film 24a during annealing. If ozone is added to the annealing atmosphere as described above, it is considered that the dehydration effect on the interlayer insulating film 40 is maintained by the strong oxidizing power of ozone even if the substrate temperature is lowered in this way.
  • FIGS. 15 and 16 are graphs obtained by investigating the moisture content of the silicon oxide film formed at various film formation temperatures by TDS analysis using the HDPCVD method. [0157] In these figures, sample E was measured in the absence of a silicon wafer, as in Figs.
  • Samples D3 to D5 had film formation temperatures of 250 ° C and 30 respectively in the HDPCVD method.
  • a silicon oxide film is formed on a silicon wafer at 0 ° C and 400 ° C.
  • sample D5 which has the highest film formation temperature among the three samples, has the smallest amount of water released from the film.
  • the silicon oxide film used as the interlayer insulating film 40 preferably has a low moisture content. However, if the deposition temperature is increased to 400 ° C in order to reduce the moisture content, Therefore, the capacitor dielectric film 24a may be deteriorated by the hydrogen activated.
  • the film forming temperature of the interlayer insulating film 40 is as low as possible, for example, 350 ° C. or less. Is preferred.
  • the film forming temperature is too low, the amount of moisture contained in the interlayer insulating film 40 is increased as shown in FIGS. 15 and 16.
  • the amount of moisture increases in this way, when the interlayer insulating film 40 is annealed, the capacitor dielectric film 24a is deteriorated by moisture in the film.
  • the film formation temperature of the interlayer insulating film 40 is too low.
  • the film formation temperature is 250 ° C or higher. preferable.
  • the film formation temperature is made very low so that the capacitor dielectric film 24a is not deteriorated by hydrogen, and By annealing and dehydrating the interlayer insulating film 40 after the film formation, the damage to the capacitor Q during the process can be most effectively reduced.
  • the interlayer insulating film can be formed at a higher deposition temperature than that of FeRAM. Since the interlayer insulating film obtained at such a high deposition temperature has a small amount of moisture in the film as described above, it is not necessary to perform annealing for dehydration as in this embodiment. .
  • annealing was performed on the interlayer insulating film 40 in a mixed atmosphere of oxygen and ozone, and the interlayer insulating film 40 was dehydrated.
  • the interlayer insulating film 40 is annealed in an atmosphere containing other oxidizing gas, for example, nitrogen dioxide (N 0), in addition to the mixed atmosphere of oxygen and ozone.
  • other oxidizing gas for example, nitrogen dioxide (N 0)
  • the silicon oxide film formed by the low-temperature HDPCVD method suitable as an interlayer insulating film of FeRAM is effective by annealing in a mixed atmosphere of ozone and oxygen as in this embodiment. Can be dehydrated.
  • annealing force is obtained by annealing the interlayer insulating film 40 in a non-plasma atmosphere, and the annealing may be performed in a plasma atmosphere!

Abstract

[PROBLEMS] To provide a method for manufacturing a semiconductor device which can prevent the deterioration of a capacitor dielectric film. [MEANS FOR SOLVING PROBLEMS] A method for manufacturing a semiconductor device comprising the steps of forming a base insulating film (11) above a silicon substrate (1), forming a capacitor (Q) provided with a lower electrode (23a), a capacitor dielectric film (24a) formed of a ferroelectric material, and an upper electrode (25a) on the base insulating film (11), forming a first capacitor protection insulating film (39) covering the capacitor (Q), forming an interlayer insulating film (40) on the first capacitor protection insulating film (39) by a plasma CVD method in which bias voltage is applied to the side of the silicon substrate (1), and annealing the interlayer insulating film (40) in an ozone- or oxygen-containing atmosphere.

Description

明 細 書  Specification
半導体装置の製造方法  Manufacturing method of semiconductor device
技術分野  Technical field
[0001] 本発明は、半導体装置の製造方法に関する。  The present invention relates to a method for manufacturing a semiconductor device.
背景技術  Background art
[0002] 電源を切っても情報を記憶することができる不揮発性メモリとして、フラッシュメモリ や強誘電体メモリが知られている。  [0002] Flash memories and ferroelectric memories are known as nonvolatile memories that can store information even when the power is turned off.
[0003] このうち、フラッシュメモリは、絶縁ゲート型電界効果トランジスタ(IGFET)のゲート 絶縁膜中に埋め込んだフローティングゲートを有し、記憶情報を表す電荷をこのフロ 一ティングゲートに蓄積することによって情報を記憶する。しかし、このようなフラッシ ュメモリでは、情報の書き込みや消去の際に、ゲート絶縁膜にトンネル電流を流す必 要があり、比較的高 、電圧が必要であると 、う欠点がある。  Among these, the flash memory has a floating gate embedded in the gate insulating film of an insulated gate field effect transistor (IGFET), and information is stored by storing charges representing stored information in the floating gate. Remember. However, such a flash memory has a drawback in that it requires a tunnel current to flow through the gate insulating film when writing or erasing information, and a relatively high voltage is required.
[0004] これに対し、強誘電体メモリは、 FeRAM(Ferroelectric Random Access Memory)とも 呼ばれ、強誘電体キャパシタが備える強誘電体膜のヒステリシス特性を利用して情報 を記憶する。その強誘電体膜は、キャパシタの上部電極と下部電極の間に印加され る電圧に応じて分極を生じ、その電圧を取り去っても自発分極が残留する。印加電 圧の極性を反転すると、この自発分極も反転し、その自発分極の向きを「1」と「0」に 対応させることで、強誘電体膜に情報が書き込まれる。この書き込みに必要な電圧は フラッシュメモリにおけるよりも低ぐまた、フラッシュメモリよりも高速で書き込みができ るという利点が FeARMにはある。  [0004] On the other hand, the ferroelectric memory is also called FeRAM (Ferroelectric Random Access Memory), and stores information using the hysteresis characteristic of the ferroelectric film provided in the ferroelectric capacitor. The ferroelectric film generates polarization according to the voltage applied between the upper electrode and the lower electrode of the capacitor, and the spontaneous polarization remains even if the voltage is removed. When the polarity of the applied voltage is reversed, this spontaneous polarization is also reversed, and information is written to the ferroelectric film by making the direction of the spontaneous polarization correspond to “1” and “0”. FeARM has the advantage that the voltage required for writing is lower than in flash memory and that writing can be performed faster than in flash memory.
[0005] 上記の強誘電体膜としては、 PZT(Pb(Zr, Ti)0  [0005] As the above ferroelectric film, PZT (Pb (Zr, Ti) 0
3 )、 Laドープ PZT(PLZT)等の PZT系材 料や、  3), PZT materials such as La-doped PZT (PLZT),
SBT(SrBi Ta O ), SBTN(SrBi (Ta, Nb) O )等の Bi層状構造化合物等が用いられてい  Bi layer structure compounds such as SBT (SrBi Ta O) and SBTN (SrBi (Ta, Nb) O) are used.
2 2 9 2 2 9  2 2 9 2 2 9
る。  The
[0006] 但し、これらの強誘電体薄膜は、キャパシタ上に層間絶縁膜を形成する工程等に おいて、水素により還元されて残留分極電荷量等の強誘電体特性が劣化し易い。こ のような強誘電体膜の劣化を防止するため、通常は、水素バリア膜として機能するキ ャパシタ保護絶縁膜によりキャパシタを覆うことが行われている。 However, these ferroelectric thin films are easily reduced by hydrogen in the process of forming an interlayer insulating film on the capacitor and the like, and the ferroelectric characteristics such as the residual polarization charge amount are likely to deteriorate. In order to prevent such deterioration of the ferroelectric film, it normally has a key functioning as a hydrogen barrier film. The capacitor is covered with a capacitor protection insulating film.
[0007] 従来、最小寸法が 0. 35 μ mで設計された 0. 35 m FeRAMでは、このようなキヤ パシタ保護絶縁膜として、スパッタ法により形成されたアルミナ (A1 0 )膜が使用されて  [0007] Conventionally, 0.35 m FeRAM designed with a minimum dimension of 0.35 μm uses an alumina (A1 0) film formed by sputtering as such a capacitor protective insulating film.
2 3  twenty three
いる。  Yes.
[0008] 例えば、特許文献 1には、ひな壇構造にカ卩ェされたキャパシタの全てを覆うように、 2. 7gZcm3を超える膜密度を有したアルミナ膜を成膜することにより、水素等の還元 性ガスがキャパシタの横方向から侵入するのが阻止され、強誘電体膜の還元が防止 されると記載されている。この場合、アルミナターゲットを用いた RF(Radio Frequency) スパッタを採用することにより、パーティクルが少なぐアモルファスなアルミナ膜の成 膜が可能になる。また、成膜雰囲気にガスが含まれないため、アルミナ膜の成膜時に よる強誘電体膜が劣化することも無 ヽ。 [0008] For example, Patent Document 1 discloses that an alumina film having a film density exceeding 2.7 gZcm 3 is formed so as to cover all the capacitors covered by a platter structure, thereby providing hydrogen or the like. It is described that the reducing gas is prevented from entering from the lateral direction of the capacitor and the reduction of the ferroelectric film is prevented. In this case, by adopting RF (Radio Frequency) sputtering using an alumina target, an amorphous alumina film with fewer particles can be formed. Further, since the gas is not contained in the film formation atmosphere, the ferroelectric film due to the film formation of the alumina film is not deteriorated.
[0009] 上記したアルミナ膜の成膜方法としては、スパッタ法の他に CVD法もある。  [0009] As a method for forming the alumina film, there is a CVD method in addition to the sputtering method.
[0010] CVD法では、スパッタ法と比べてアルミナ膜のステップカバレッジが良好になるので 、微細化が進んでキャパシタの間隔が狭くなつても、キャパシタの側面に十分な厚さ を持ったアルミナ膜を形成することができ、キャパシタの横方向力も水素が侵入する のを効果的に阻止することができる。  [0010] Since the CVD method provides better step coverage of the alumina film than the sputtering method, an alumina film having a sufficient thickness on the side surface of the capacitor even if the miniaturization advances and the interval between the capacitors becomes narrower. The lateral force of the capacitor can also effectively prevent hydrogen from entering.
[0011] このような特性から、 CVD法により形成されたアルミナ膜は、キャパシタの微細化に 有利なスタック型の FeRAMにおいてキャパシタ保護絶縁膜として形成するのが好まし い。特に、現状よりも微細化が進んだ次世代の FeRAM、例えば最小寸法が 0. 18 mで設計された 0. 18 mのスタック型 FeRAMでは、上記のように CVD法により形成さ れたアルミナ膜が必須であると考えられる。  [0011] Because of these characteristics, the alumina film formed by the CVD method is preferably formed as a capacitor protective insulating film in a stacked FeRAM that is advantageous for miniaturization of capacitors. In particular, in the next generation FeRAM, which has been further miniaturized compared to the current situation, for example, a 0.18 m stack type FeRAM designed with a minimum dimension of 0.18 m, the alumina film formed by the CVD method as described above. Is considered essential.
[0012] CVD法によるアルミナ膜の成膜では、通常、有機アルミニウム化合物としてトリメチ ルアルミニウム〔A1(CH ) ] (TMA: Tri-Methyl Aluminum)を用い、酸化剤として水 (H 0  In the formation of an alumina film by the CVD method, trimethylaluminum [A1 (CH 3)] (TMA: Tri-Methyl Aluminum) is usually used as the organoaluminum compound, and water (H 0) is used as the oxidant.
3 3 2 3 3 2
)を用いる。 ) Is used.
[0013] 図 1は、 CVD法の一種である原子層堆積法 (ALD: Atomic Layer Deposition)により アルミナ膜が形成されるメカニズムを示す模式図である。  FIG. 1 is a schematic diagram showing a mechanism for forming an alumina film by atomic layer deposition (ALD), which is a kind of CVD method.
[0014] 図 1に示されるように、 ALDによる成膜では、まず、 A)、基板 200の表面の全てを覆 うに H 0を吸着させた後、 B)、余分な H 0を真空排気してパージし、それ力も C)、 TM Aを流して吸着している OH群と反応させて原子層の Al 0を形成し、 D)、余分な TMA [0014] As shown in FIG. 1, in film formation by ALD, A) first adsorbs H 0 so as to cover the entire surface of the substrate 200, and then B) evacuates excess H 0. Purge, and force C), TM A) is flown to react with the adsorbed OH groups to form Al 0 in the atomic layer, and D) extra TMA
2 3  twenty three
を真空排気してパージする。そして、これら A)〜D)の一連のサイクルを繰り返すこと によりアルミナ膜が形成される。  Is evacuated and purged. An alumina film is formed by repeating a series of cycles A) to D).
[0015] 但し、このように酸化剤として水を使用すると、アルミナ膜の成膜時に強誘電体膜に 水素又は水分が吸着されたり、アルミナ膜に水分が残留したりする。こうなると、キヤ パシタの回復ァニール等の熱処理時にぉ 、て、水分によってキャパシタ誘電体膜が 還元して劣化する恐れがある。 However, when water is used as the oxidizing agent in this way, hydrogen or moisture is adsorbed on the ferroelectric film or moisture remains on the alumina film when the alumina film is formed. In this case, the capacitor dielectric film may be reduced and deteriorated by moisture during heat treatment such as the recovery annealing of the capacitor.
[0016] この点に鑑み、特許文献 2では、 FeRAMのキャパシタ保護絶縁膜として ALD法によ りアルミナ膜を形成する際、酸化剤としてオゾン (0 In view of this point, in Patent Document 2, when an alumina film is formed by an ALD method as a capacitor protective insulating film of FeRAM, ozone (0
3 )を使用している。これによれば、 オゾンが水素フリーな酸化剤であるため、アルミナ膜の成膜時にキャパシタ誘電体に 水素や水分が吸着したり、アルミナ膜中に水分が残留したりすることが抑制される。  3) is used. According to this, since ozone is a hydrogen-free oxidant, it is possible to prevent hydrogen and moisture from adsorbing to the capacitor dielectric and moisture from remaining in the alumina film when the alumina film is formed.
[0017] ここで、上記の特許文献 1及び特許文献 2では、キャパシタの表面に沿ってアルミ ナ膜を形成して 、るが、キャパシタへの水素や水分の侵入をより効果的に阻止する には、キャパシタを覆う層間絶縁膜の上にもアルミナ膜を形成するのが好ましい。  [0017] Here, in Patent Document 1 and Patent Document 2 described above, an alumina film is formed along the surface of the capacitor. However, in order to more effectively prevent hydrogen and moisture from entering the capacitor. Preferably, an alumina film is also formed on the interlayer insulating film covering the capacitor.
[0018] 例えば、特許文献 3では、層間絶縁膜の平坦化された表面の上に、キャパシタ保護 絶縁膜としてアルミナ膜を形成している。そのアルミナ膜は、平坦面上に形成される ため優れたカバレッジ特性が要求されず、スパッタ法で形成することができる。  For example, in Patent Document 3, an alumina film is formed as a capacitor protection insulating film on the planarized surface of an interlayer insulating film. Since the alumina film is formed on a flat surface, it does not require excellent coverage characteristics and can be formed by sputtering.
[0019] ところで、 FeRAMがより微細化されると、キャパシタ間のスペースを埋める層間絶縁 膜に"す"が発生する恐れがある。キャパシタ幅に対して大きな"す"が入ると、キャパ シタの横を押さえる層間絶縁膜の膜厚が薄くなり、上部電極や下部電極が熱膨張す る際に"す"部分力 層間絶縁膜にクラックが走り、層間絶縁膜上の金属配線の信頼 性を低下させてしまう。このような問題は、層間絶縁膜の成膜方法として、 TEOS(Tetr aEthOxySilane: Si(OC H ) )ガスを使用するプラズマ CVD法を採用する場合に顕著に  By the way, when FeRAM is further miniaturized, there is a possibility that “soot” is generated in the interlayer insulating film filling the space between the capacitors. When a large “soot” enters the capacitor width, the film thickness of the interlayer insulation film that holds the side of the capacitor decreases, and the “slow” partial force is applied to the interlayer insulation film when the upper and lower electrodes thermally expand. Cracks run and reduce the reliability of the metal wiring on the interlayer insulating film. Such a problem is prominent when the plasma CVD method using TEOS (TetraEthOxySilane: Si (OC H)) gas is adopted as a method for forming an interlayer insulating film.
2 5 4  2 5 4
見られる。  It can be seen.
[0020] そこで、このような"す"の発生を抑えるために、キャパシタ上の層間絶縁膜の成膜 方法として HDPCVD(High Density Plasma CVD)法を採用することが検討されている  [0020] Therefore, in order to suppress the occurrence of such "soot", it has been studied to adopt an HDPCVD (High Density Plasma CVD) method as a method of forming an interlayer insulating film on the capacitor.
[0021] HDPCVD法では、半導体基板に大きなバイアス電圧を印加することにより、反応ガ スによるスパッタと成膜とが同時に進行し、 "す"の発生を抑えながら、狭いキャパシタ 間を層間絶縁膜で埋めることができる。 [0021] In the HDPCVD method, a reaction bias is applied by applying a large bias voltage to a semiconductor substrate. Sputtering and film-forming at the same time proceed, and the space between narrow capacitors can be filled with an interlayer insulating film while suppressing the generation of “soot”.
[0022] 但し、この HDPCVD法では、バイアス電圧によって H+イオン等の還元性物質が半導 体基板に引き込まれ、その還元性物質によってキャパシタ誘電体膜が劣化する恐れ がある。 However, in this HDPCVD method, a reducing substance such as H + ions is attracted to the semiconductor substrate by the bias voltage, and the capacitor dielectric film may be deteriorated by the reducing substance.
[0023] そのため、 HDPCVDにより層間絶縁膜を形成する場合には、キャパシタの表面を覆 うキャパシタ保護絶縁膜として、既述の ALD法により緻密でカバレッジ特性が良好な アルミナ膜を形成するのが好まし 、。  [0023] Therefore, when an interlayer insulating film is formed by HDPCVD, it is preferable to form a dense alumina film having good coverage characteristics by the above-described ALD method as a capacitor protective insulating film that covers the surface of the capacitor. Better ,.
特許文献 1:特開 2001—44375号公報  Patent Document 1: Japanese Patent Laid-Open No. 2001-44375
特許文献 2 :特開 2004— 193280号公報  Patent Document 2: JP 2004-193280 A
特許文献 3 :特開 2006— 49795号公報  Patent Document 3: Japanese Unexamined Patent Publication No. 2006-49795
発明の開示  Disclosure of the invention
[0024] 本発明の目的は、キャパシタ誘電体膜の劣化を防止することが可能な半導体装置 の製造方法を提供することにある。  An object of the present invention is to provide a method for manufacturing a semiconductor device capable of preventing deterioration of a capacitor dielectric film.
[0025] 本発明の一観点によれば、半導体基板の上方に下地絶縁膜を形成する工程と、前 記下地絶縁膜の上に、下部電極、強誘電体材料よりなるキャパシタ誘電体膜、及び 上部電極を備えたキャパシタを形成する工程と、前記キャパシタを覆う第 1キャパシタ 保護絶縁膜を形成する工程と、前記第 1キャパシタ保護絶縁膜の上に、前記半導体 基板側にバイアス電圧を印加するプラズマ CVD法により、層間絶縁膜を形成するェ 程と、オゾン又は酸素を含む雰囲気中で前記層間絶縁膜をァニールする工程とを有 する半導体装置の製造方法が提供される。  According to one aspect of the present invention, a step of forming a base insulating film above a semiconductor substrate, a lower electrode, a capacitor dielectric film made of a ferroelectric material on the base insulating film, and Forming a capacitor having an upper electrode; forming a first capacitor protective insulating film covering the capacitor; and plasma for applying a bias voltage to the semiconductor substrate side on the first capacitor protective insulating film A method of manufacturing a semiconductor device is provided, which includes a step of forming an interlayer insulating film by a CVD method and a step of annealing the interlayer insulating film in an atmosphere containing ozone or oxygen.
[0026] 本発明によれば、半導体基板側にバイアス電圧を印加するプラズマ CVD法 (HDPC VD法)により層間絶縁膜を形成する。このように形成された層間絶縁膜は、埋め込み 性に優れており、 "す"が発生すること無しにキャパシタ間の狭いスペースを埋めるこ とができる一方、キャパシタ誘電体膜を劣化させ得る水分を膜中に多く含んで ヽる。  According to the present invention, the interlayer insulating film is formed by a plasma CVD method (HDPC VD method) in which a bias voltage is applied to the semiconductor substrate side. The interlayer insulating film formed in this way is excellent in embeddability, and can fill a narrow space between capacitors without generating “soot”, while water that can deteriorate the capacitor dielectric film. Contains a lot in the film.
[0027] そこで、本発明では、オゾン又は酸素を含む雰囲気中でこの層間絶縁膜をァニー ルし、層間絶縁膜を脱水することで、層間絶縁膜中の水分によってキャパシタ誘電体 膜が還元されて劣化するのを防止する。 [0028] このようにァニール雰囲気中にオゾンを添加することで、雰囲気の酸化力が増大す るため、ァニール時の基板温度を低温にしても脱水の効果が期待できると共に、高 い基板温度に起因してキャパシタ誘電体膜が劣化するのを抑制できる。 Therefore, in the present invention, the interlayer dielectric film is annealed in an atmosphere containing ozone or oxygen, and the interlayer dielectric film is dehydrated, whereby the capacitor dielectric film is reduced by the moisture in the interlayer dielectric film. Prevent deterioration. [0028] By adding ozone to the annealing atmosphere in this manner, the oxidizing power of the atmosphere increases, so that the effect of dehydration can be expected even when the annealing substrate temperature is lowered, and the substrate temperature is increased. This can suppress the deterioration of the capacitor dielectric film.
[0029] また、キャパシタを覆うように形成される第 1キャパシタ保護絶縁膜は、水素等の還 元性物質がキャパシタに侵入するのを防止するように機能する。この第 1キャパシタ 保護絶縁膜として、カバレッジ特性に優れた ALD法でアルミナ膜を形成すると、半導 体装置の微細化が進んでも、キャパシタの側面に十分な第 1キャパシタ保護絶縁膜 を形成することができ、横方向力もキャパシタに還元性物質が侵入するのを効果的に 防止することができる。  [0029] The first capacitor protective insulating film formed so as to cover the capacitor functions to prevent a reducing substance such as hydrogen from entering the capacitor. If an alumina film is formed as the first capacitor protective insulating film by the ALD method with excellent coverage characteristics, a sufficient first capacitor protective insulating film should be formed on the side surface of the capacitor even if the semiconductor device is miniaturized. The lateral force can effectively prevent the reducing substance from entering the capacitor.
[0030] 更に、その ALD法において、水素フリーなオゾンを酸化剤として用いることで、第 1 キャパシタ保護絶縁膜の成膜時に水素によってキャパシタ誘電体膜が劣化するのを 防ぐことが可能となる。  [0030] Furthermore, in the ALD method, by using hydrogen-free ozone as an oxidizing agent, it is possible to prevent the capacitor dielectric film from being deteriorated by hydrogen during the formation of the first capacitor protection insulating film.
[0031] また、上記した層間絶縁膜に対するァニールは、 CMP(Chemical Mechanical Polishi ng)法により層間絶縁膜の表面を平坦ィ匕する工程の後に行うのが好ましい。このように することで、 CMPのスラリーに含まれる水分が層間絶縁膜に吸収されたとしても、了二 ールによってその水分を除去することができる。更に、 CMPによって層間絶縁膜の体 積が減ることから、ァニール時に層間絶縁膜から脱ガスする水分量を低減することが でき、層間絶縁膜から出る水分によってキャパシタが劣化するのが抑制されるという 禾 IJ点ち得られる。  In addition, the annealing for the interlayer insulating film described above is preferably performed after the step of planarizing the surface of the interlayer insulating film by a CMP (Chemical Mechanical Polishing) method. In this way, even if the moisture contained in the CMP slurry is absorbed by the interlayer insulating film, the moisture can be removed by sealing. Furthermore, since the volume of the interlayer insulating film is reduced by CMP, the amount of water degassed from the interlayer insulating film during annealing can be reduced, and the deterioration of the capacitor due to the moisture released from the interlayer insulating film is suppressed.禾 IJ points can be obtained.
[0032] そして、層間絶縁膜をァニールする工程の後に、層間絶縁膜の上に第 2キャパシタ 保護絶縁膜を形成する工程と、上部電極の上の第 1キャパシタ保護絶縁膜、層間絶 縁膜、及び第 2キャパシタ保護絶縁膜にホールを形成する工程を行ってもょ 、。  [0032] Then, after the step of annealing the interlayer insulating film, a step of forming a second capacitor protective insulating film on the interlayer insulating film, a first capacitor protective insulating film on the upper electrode, an interlayer insulating film, Also, do the process of forming holes in the second capacitor protective insulating film.
[0033] その場合、プロセス中にキャパシタ誘電体膜が受けたダメージは、上記のホールを 形成した後に酸素含有雰囲気中にお!/ヽてキャパシタ誘電体膜をァニールすることで 回復される。このような回復ァニールを行っても、既述のように層間絶縁膜に対して脱 水を目的としたァニールを施してあるため、第 1、第 2キャパシタ誘電体膜に挟まれた 層間絶縁膜が膜中の水分によって蒸し焼きの状態にならない。よって、その水分によ つてキャパシタ誘電体膜が劣化したり、第 2キャパシタ保護絶縁膜にクラックが入った りするのを防止することが可能となる。 In this case, the damage received by the capacitor dielectric film during the process is recovered by annealing the capacitor dielectric film in an oxygen-containing atmosphere after forming the above holes. Even if such a recovery annealing is performed, the interlayer insulating film is annealed for the purpose of dewatering as described above, and therefore the interlayer insulating film sandwiched between the first and second capacitor dielectric films. However, it does not become steamed due to moisture in the film. Therefore, the capacitor dielectric film deteriorates due to the moisture, or the second capacitor protective insulating film cracks. Can be prevented.
図面の簡単な説明 Brief Description of Drawings
[図 1]図 1は、 ALD法によりアルミナ膜が成膜されるメカニズムを示す模式図である。 FIG. 1 is a schematic diagram showing a mechanism for forming an alumina film by the ALD method.
[図 2]図 2 (a)〜 (c)は、本発明の実施の形態に係る半導体装置の製造途中の断面 図(その 1)である。 [FIG. 2] FIGS. 2 (a) to 2 (c) are cross-sectional views (part 1) in the middle of the manufacture of the semiconductor device according to the embodiment of the present invention.
[図 3]図 3 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 2)である。  3 (a) and 3 (b) are cross-sectional views (part 2) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
[図 4]図 4 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 3)である。  4 (a) and 4 (b) are cross-sectional views (part 3) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
[図 5]図 5 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 4)である。  FIGS. 5 (a) and 5 (b) are cross-sectional views (part 4) of the semiconductor device according to the embodiment of the present invention during manufacture.
[図 6]図 6 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 5)である。  6 (a) and 6 (b) are cross-sectional views (part 5) in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
[図 7]図 7 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 6)である。  FIGS. 7A and 7B are cross-sectional views (part 6) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
[図 8]図 8 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 7)である。  FIGS. 8A and 8B are cross-sectional views (part 7) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
[図 9]図 9 (a)、(b)は、本発明の実施の形態に係る半導体装置の製造途中の断面図 (その 8)である。  FIGS. 9 (a) and 9 (b) are cross-sectional views (part 8) of the semiconductor device according to the embodiment of the present invention in the middle of manufacture.
[図 10]図 10は、本発明の実施の形態に係る半導体装置の製造途中の断面図(その 9)である。  FIG. 10 is a sectional view (No. 9) in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
[図 11]図 11は、本発明の実施の形態において使用される HDPCVD装置の構成図で ある。  FIG. 11 is a configuration diagram of an HDPCVD apparatus used in the embodiment of the present invention.
[図 12]図 12は、本発明の実施の形態において、ァニールによる酸ィ匕シリコン膜の脱 水効果を調べるために行った TDS分析の結果を示す図(その 1)である。  [FIG. 12] FIG. 12 is a diagram (No. 1) showing a result of TDS analysis performed in order to investigate the dehydration effect of an acid silicon film by annealing in the embodiment of the present invention.
[図 13]図 13は、本発明の実施の形態において、ァニールによる酸ィ匕シリコン膜の脱 水効果を調べるために行った TDS分析の結果を示す図(その 2)である。  FIG. 13 is a diagram (No. 2) showing a result of TDS analysis performed for examining the effect of dehydration of an oxide silicon film by annealing in the embodiment of the present invention.
[図 14]図 14は、本発明の実施の形態において、ァニールによる酸ィ匕シリコン膜の脱 水効果を調べるために行った TDS分析の結果を示す図(その 3)である。 [FIG. 14] FIG. 14 shows the removal of an oxide silicon film by annealing in an embodiment of the present invention. It is the figure (the 3) which shows the result of the TDS analysis performed in order to investigate the water effect.
[図 15]図 15は、本発明の実施の形態において、 HDPCVD法により形成された酸ィ匕シ リコン膜の水分含有量が成膜温度にどのように依存するのかを調べるために行われ た TDS分析の結果を示す図(その 1)である。  [FIG. 15] FIG. 15 was conducted in order to investigate how the water content of the acid silicon film formed by the HDPCVD method depends on the film formation temperature in the embodiment of the present invention. It is the figure (the 1) which shows the result of TDS analysis.
[図 16]図 16は、本発明の実施の形態において、 HDPCVD法により形成された酸ィ匕シ リコン膜の水分含有量が成膜温度にどのように依存するのかを調べるために行われ た TDS分析の結果を示す図(その 2)である。  [FIG. 16] FIG. 16 was carried out in order to investigate how the moisture content of an acid silicon film formed by the HDPCVD method depends on the film formation temperature in the embodiment of the present invention. It is the figure which shows the result of TDS analysis (the 2).
発明を実施するための最良の形態  BEST MODE FOR CARRYING OUT THE INVENTION
[0035] 次に、本発明の実施の形態について、添付図面を参照しながら詳細に説明する。 Next, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
[0036] 図 2〜図 10は、本発明の実施の形態に係る半導体装置の製造途中の断面図であ る。 2 to 10 are cross-sectional views in the middle of manufacturing the semiconductor device according to the embodiment of the present invention.
[0037] この半導体装置は、微細化に有利なスタック型の FeRAMであり、以下のようにして 作製される。  This semiconductor device is a stack type FeRAM advantageous for miniaturization, and is manufactured as follows.
[0038] 最初に、図 2 (a)に示す断面構造を得るまでの工程について説明する。  [0038] First, steps required until a sectional structure shown in FIG.
[0039] まず、 n型又は p型のシリコン(半導体)基板 1表面に、トランジスタの活性領域を画 定する STI(Shallow Trench Isolation)用の溝を形成し、その中に酸化シリコン等の絶 縁膜を埋め込んで素子分離絶縁膜 2とする。なお、素子分離構造は STIに限られず、 LOCOS(Local Oxidation of Silicon)法で素子分離絶縁膜 2を形成してもよい。 [0039] First, a trench for STI (Shallow Trench Isolation) that defines an active region of a transistor is formed on the surface of an n-type or p-type silicon (semiconductor) substrate 1, and an insulating material such as silicon oxide is formed therein. The element isolation insulating film 2 is formed by embedding the film. The element isolation structure is not limited to STI, and the element isolation insulating film 2 may be formed by a LOCOS (Local Oxidation of Silicon) method.
[0040] 次いで、シリコン基板 1の活性領域に p型不純物を導入して pゥエル 3を形成した後、 その活性領域の表面を熱酸化することにより、ゲート絶縁膜 4となる熱酸化膜を形成 する。 [0040] Next, after p-type impurities are introduced into the active region of the silicon substrate 1 to form the p-well 3, the surface of the active region is thermally oxidized to form a thermal oxide film that becomes the gate insulating film 4 To do.
[0041] 続いて、シリコン基板 1の上側全面に非晶質又は多結晶のシリコン膜を形成し、こ れらの膜をフォトリソグラフィによりパターユングして二つのゲート電極 5を形成する。  Subsequently, an amorphous or polycrystalline silicon film is formed on the entire upper surface of the silicon substrate 1, and these films are patterned by photolithography to form two gate electrodes 5.
[0042] pゥエル 3上には、上記の 2つのゲート電極 5が間隔をおいて平行に配置され、それ らのゲート電極 5はワード線の一部を構成する。 [0042] On the p-well 3, the two gate electrodes 5 described above are arranged in parallel at a distance from each other, and these gate electrodes 5 constitute a part of the word line.
[0043] 次いで、ゲート電極 5をマスクにするイオン注入により、ゲート電極 5の横のシリコン 基板 1に n型不純物を導入し、第 1、第 2ソース/ドレインエクステンション 6a、 6bを形 成する。 [0044] その後に、シリコン基板 1の上側全面に絶縁膜を形成し、その絶縁膜をエッチバック してゲート電極 5の横に絶縁性サイドウォール 7を形成する。その絶縁膜として、例え ば CVD法により酸ィ匕シリコン膜を形成する。 Next, n-type impurities are introduced into the silicon substrate 1 beside the gate electrode 5 by ion implantation using the gate electrode 5 as a mask to form first and second source / drain extensions 6a and 6b. Thereafter, an insulating film is formed on the entire upper surface of the silicon substrate 1, and the insulating film is etched back to form an insulating sidewall 7 next to the gate electrode 5. As the insulating film, for example, an oxide silicon film is formed by a CVD method.
[0045] 続いて、絶縁性サイドウォール 7とゲート電極 5をマスクにしながら、シリコン基板 1に n型不純物を再びイオン注入することにより、二つのゲート電極 5の側方のシリコン基 板 1の表層に、互いに間隔がおかれた第 1、第 2ソース Zドレイン領域 8a 8bを形成 する。  Subsequently, n-type impurities are ion-implanted again into the silicon substrate 1 while using the insulating sidewalls 7 and the gate electrode 5 as a mask, so that the surface layer of the silicon substrate 1 on the side of the two gate electrodes 5 is obtained. Then, first and second source Z drain regions 8a and 8b spaced apart from each other are formed.
[0046] ここまでの工程により、シリコン基板 1の活性領域には、ゲート絶縁膜 4、ゲート電極 5、及び第 1、第 2ソース/ドレイン領域 8a 8bによって構成される第 1、第 2MOSトラ ンジスタ TR TRが形成されたことになる。  Through the above steps, the active region of the silicon substrate 1 includes the first and second MOS transistors configured by the gate insulating film 4, the gate electrode 5, and the first and second source / drain regions 8a 8b. TR TR is formed.
1 2  1 2
[0047] 次に、シリコン基板 1の上側全面に、スパッタ法によりコバルト層等の高融点金属層 を形成した後、この高融点金属層を加熱してシリコンと反応させ、シリコン基板 1上に 高融点金属シリサイド層 9を形成する。その高融点金属シリサイド層 9はゲート電極 5 の表層部分にも形成され、それによりゲート電極 5が低抵抗ィ匕されることになる。  [0047] Next, after forming a refractory metal layer such as a cobalt layer on the entire upper surface of the silicon substrate 1 by sputtering, the refractory metal layer is heated to react with silicon. A melting point metal silicide layer 9 is formed. The refractory metal silicide layer 9 is also formed on the surface layer portion of the gate electrode 5, thereby reducing the resistance of the gate electrode 5.
[0048] その後、素子分離絶縁膜 2の上等で未反応となっている高融点金属層をウエットェ ツチングして除去する。  [0048] Thereafter, the unreacted refractory metal layer on the element isolation insulating film 2 or the like is removed by wet etching.
[0049] 続いて、プラズマ CVD法により、シリコン基板 1の上側全面に窒化シリコン (SiN)膜を 厚さ約 200 に形成し、それをカバー絶縁膜 10とする。次いで、このカバー絶縁膜 1 0の上に、 TEOSガスを使用するプラズマ CVD法により下地絶縁膜 11として酸ィ匕シリ コン膜を厚さ約 lOOOnmに形成する。  Subsequently, a silicon nitride (SiN) film is formed to a thickness of about 200 on the entire upper surface of the silicon substrate 1 by plasma CVD, and this is used as the cover insulating film 10. Next, an acidic silicon film is formed on the cover insulating film 10 as a base insulating film 11 with a thickness of about 100 nm by a plasma CVD method using TEOS gas.
[0050] 次!、で、下地絶縁膜 11の上面を CMP(Chemical Mechanical Polishing)法により研磨 して平坦化する。この CMPの結果、下地絶縁膜 11の厚さは、シリコン基板 1の平坦面 上で約 700nmとなる。  Next, the upper surface of the base insulating film 11 is polished and planarized by a CMP (Chemical Mechanical Polishing) method. As a result of this CMP, the thickness of the base insulating film 11 is about 700 nm on the flat surface of the silicon substrate 1.
[0051] そして、フォトリソグラフィによりカバー絶縁膜 10と下地絶縁膜 11とをパターユングし て第 1、第 2ソース/ドレイン領域 8a 8bの上にコンタクトホールを形成する。更に、こ のコンタクトホール内にグルー膜 (密着膜)とタングステン膜とを順に形成した後、下 地絶縁膜 11上の余分なグルー膜とタングステン膜とを CMP法により研磨して除去し、 これらの膜をコンタクトホール内にのみ第 1、第 2導電性プラグ 12a 12bとして残す。 [0052] これらの第 1、第 2導電性プラグ 12a、 12bは、それぞれ第 1、第 2ソース/ドレイン 領域 8a、 8bと電気的に接続される。 Then, the cover insulating film 10 and the base insulating film 11 are patterned by photolithography to form contact holes on the first and second source / drain regions 8a 8b. Further, after forming a glue film (adhesion film) and a tungsten film in this contact hole in order, the excess glue film and tungsten film on the underlying insulating film 11 are polished and removed by CMP, and these are removed. The first and second conductive plugs 12a and 12b are left only in the contact holes. [0052] These first and second conductive plugs 12a and 12b are electrically connected to the first and second source / drain regions 8a and 8b, respectively.
[0053] なお、上記のグルー膜は、厚さ約 30nmのチタン膜と厚さ約 50nmの窒化チタン膜と をこの順に形成してなる。 [0053] The glue film is formed by forming a titanium film having a thickness of about 30 nm and a titanium nitride film having a thickness of about 50 nm in this order.
[0054] 次に、図 2 (b)に示すように、下地絶縁膜 11と第 1、第 2導電性プラグ 12a、 12bの それぞれの上にスパッタ法によりチタン膜を厚さ約 20應に形成し、このチタン膜 20η mを下地導電膜 21とする。 Next, as shown in FIG. 2B, a titanium film is formed to a thickness of about 20 by sputtering on each of the base insulating film 11 and the first and second conductive plugs 12a and 12b. The titanium film 20 η m is used as the base conductive film 21.
[0055] その下地導電膜 21を構成するチタンは自己配向性を有するため、下地導電膜 21 の結晶'性は良好となる。 [0055] Since titanium constituting the underlying conductive film 21 has self-orientation, the crystallinity of the underlying conductive film 21 is good.
[0056] 続、て、図 2 (c)に示すように、酸化し易!、チタンよりなる下地導電膜 21の耐酸ィ匕性 を向上させる目的で、窒素雰囲気中において基板温度を 650°C、処理時間を 120秒 とする RTA(Rapid Thermal Anneal)を下地導電膜 21に対して行い、下地導電膜 21を 窒化させる。なお、この RTAにおける窒素流量は特に限定されないが、本実施形態 では 10slm(standard litter I min)とする。ここで、単位 slmは、標準状態(1. 013 X 10 5Pa、 0°C)での流量を表す。  [0056] Subsequently, as shown in FIG. 2 (c), the substrate temperature is set at 650 ° C in a nitrogen atmosphere for the purpose of easily oxidizing and improving the acid resistance of the underlying conductive film 21 made of titanium. Then, RTA (Rapid Thermal Anneal) with a processing time of 120 seconds is performed on the base conductive film 21 to nitride the base conductive film 21. The flow rate of nitrogen in this RTA is not particularly limited, but in this embodiment it is 10 slm (standard litter I min). Here, the unit slm represents a flow rate in a standard state (1.013 × 10 5 Pa, 0 ° C.).
[0057] 窒化前の下地導電膜 21の結晶性が良好なため、このように窒化された後の下地導 電膜 21は、その結晶性が良好であり、その上に後で形成される膜の結晶性を向上さ せる結晶性向上膜として機能するようになる。  Since the crystallinity of the base conductive film 21 before nitriding is good, the base conductive film 21 after nitriding in this way has a good crystallinity and is a film formed later on the base conductive film 21 It will function as a crystallinity improving film that improves the crystallinity of the film.
[0058] 次に、図 3 (a)に示す断面構造を得るまでの工程について説明する。  Next, steps required until a sectional structure shown in FIG.
[0059] まず、下地導電膜 21の上に導電性酸素ノリア膜 22として窒化チタンアルミニウム( TiAIN)膜を反応性スパッタ法で約 lOOnmの厚さに形成する。  First, a titanium aluminum nitride (TiAIN) film is formed as a conductive oxygen noria film 22 on the base conductive film 21 by a reactive sputtering method to a thickness of about lOOnm.
[0060] 窒化チタンアルミニウムよりなる導電性酸素ノリア膜 22は、酸素透過防止機能に優 れており、その下の第 1、第 2導電性プラグ 12a、 12bが酸ィ匕してコンタクト不良が発 生するのを防止する役割を担う。  [0060] The conductive oxygen nore film 22 made of titanium aluminum nitride has an excellent oxygen permeation preventing function, and the first and second conductive plugs 12a and 12b below the surface are oxidized to cause contact failure. Play a role in preventing life.
[0061] 次いで、この導電性酸素ノリア膜 22の上に第 1導電膜 23としてイリジウム膜をスパ ッタ法で厚さ約 lOOnmに形成する。  Next, an iridium film is formed as a first conductive film 23 on the conductive oxygen noria film 22 by a sputtering method to a thickness of about lOOnm.
[0062] そして、第 1導電膜 23の上に、基板温度を 620°C、圧力を 5Torrとする MOCVD(Me tal Organic CVD)法により、一層目の PZT膜と二層目の PZT膜をそれぞれ厚さ 5nm、 115nmに形成し、これらの PZT膜を全膜厚が 120nmの強誘電体膜 24とする。 [0062] Then, the first PZT film and the second PZT film are respectively formed on the first conductive film 23 by MOCVD (Metal Organic CVD) with a substrate temperature of 620 ° C and a pressure of 5 Torr. Thickness 5nm, These PZT films are formed as a ferroelectric film 24 having a total film thickness of 120 nm.
[0063] 一層目と二層目の PZT膜は同じ組成である。但し、低酸素分圧で成膜した方が PZT 膜の結晶性が良好になるため、一層目については、成膜雰囲気中の酸素分圧を二 層目のそれよりも下げて成膜することによりその結晶性を向上させる。しかし、二層目 も低酸素分圧で成膜すると、 PZT膜中の酸素欠損が多くなつてリーク電流が増大する ので、二層目についてリーク電流の低減を優先させて一層目よりも高酸素分圧で成 膜する。 [0063] The first and second PZT films have the same composition. However, since the crystallinity of the PZT film becomes better when the film is formed at a low oxygen partial pressure, the first layer should be formed with the oxygen partial pressure in the film formation atmosphere lower than that of the second layer. To improve the crystallinity. However, if the second layer is deposited at a low oxygen partial pressure, the leakage current increases due to the increase in oxygen vacancies in the PZT film. Therefore, priority is given to reducing the leakage current for the second layer, giving higher oxygen than the first layer. Film is formed with partial pressure.
[0064] なお、この MOCVD法では、 Pb(DPM) (化学式 Pb(C H 0 ) ))、 Zr(dmhd) (化学式 Z  [0064] In this MOCVD method, Pb (DPM) (chemical formula Pb (C H 0))), Zr (dmhd) (chemical formula Z
2 11 19 2 2 4 r(C H O ) )、及び Ti(0— iOr) (DPM) (化学式 Ti(C H O) (C H O ) )のそれぞれを 2 11 19 2 2 4 r (C H O)) and Ti (0—iOr) (DPM) (chemical formula Ti (C H O) (C H O))
9 15 2 4 2 2 3 7 2 11 19 2 2 9 15 2 4 2 2 3 7 2 11 19 2 2
THF(Tetra Hydro Furan: C H O)溶媒中に溶解したものをそれぞれ Pb、 Zr、及び Ti  Dissolved in THF (Tetra Hydro Furan: C H 2 O) solvent are Pb, Zr and Ti, respectively.
4 8  4 8
の原料として使用する。  Used as a raw material.
[0065] 続いて、この強誘電体膜 24の上に、スパッタ法により酸化イリジウム膜を厚さ約 150 應に形成し、この酸化イリジウム膜を第 2導電膜 25とする。  Subsequently, an iridium oxide film having a thickness of about 150 is formed on the ferroelectric film 24 by sputtering, and this iridium oxide film is used as the second conductive film 25.
[0066] そして、第 2導電膜 25だけでは不足しがちな導電性を向上させるため、第 2導電膜[0066] Then, in order to improve conductivity, which is often insufficient only with the second conductive film 25, the second conductive film
25の上に導電性向上膜 26としてイリジウム膜をスパッタ法で厚さ約 50應に形成する An iridium film as a conductivity enhancement film 26 is formed on the film 25 by sputtering to a thickness of about 50 mm.
[0067] 次に、図 3 (b)に示すように、導電性向上膜 26の上にスパッタ法により厚さ約 200η mの窒化チタン膜を形成し、その窒化チタン膜を第 1マスク材料層 27とする。更に、 T EOSガスを使用するプラズマ CVD法を用いて、第 1マスク材料層 27の上に第 2マスク 材料層 28として酸ィ匕シリコン膜を厚さ約 700應に形成する。 Next, as shown in FIG. 3 (b), a titanium nitride film having a thickness of about 200 ηm is formed on the conductivity improving film 26 by sputtering, and the titanium nitride film is used as the first mask material layer. 27. Further, an oxide silicon film is formed as a second mask material layer 28 on the first mask material layer 27 to a thickness of about 700 using a plasma CVD method using TEOS gas.
[0068] 次いで、図 4 (a)に示すように、第 2マスク材料層 28を島状にパターユングすること により第 2ハードマスク 28aを形成する。  Next, as shown in FIG. 4 (a), the second hard mask 28a is formed by patterning the second mask material layer 28 in an island shape.
[0069] 次に、図 4 (b)に示す断面構造を得るまでの工程について説明する。  [0069] Next, steps required until a sectional structure shown in FIG.
[0070] まず、第 2ハードマスク 28aをマスクにして第 1マスク材料層 27をエッチングすること により第 1ハードマスク 27aを形成する。  [0070] First, the first hard mask 27a is formed by etching the first mask material layer 27 using the second hard mask 28a as a mask.
[0071] 次いで、第 1、第 2ハードマスク 27a、 28aで覆われていない領域の膜 23〜26ドライ エッチングする。これにより、導電性向上膜 26と第 2導電膜 25とで構成される上部電 極 25aが形成されると共に、強誘電体膜 24がキャパシタ誘電体膜 24aとされる。更に 、第 1導電膜 23がキャパシタ誘電体膜 24aの下にのみ残存する。 Next, dry etching is performed on the films 23 to 26 in regions not covered with the first and second hard masks 27a and 28a. Thus, the upper electrode 25a composed of the conductivity improving film 26 and the second conductive film 25 is formed, and the ferroelectric film 24 is used as the capacitor dielectric film 24a. More The first conductive film 23 remains only under the capacitor dielectric film 24a.
[0072] そのドライエッチングのガスは特に限定されないが、第 1導電膜 23、及び第 2導電 膜 25、及び導電性向上膜 26に対するエッチングガスとしては HBrと酸素との混合ガ スが使用される。一方、強誘電体膜 24に対するエッチングガスとしては塩素とァルゴ ンとの混合ガスが使用される。 [0072] The dry etching gas is not particularly limited, but a mixed gas of HBr and oxygen is used as an etching gas for the first conductive film 23, the second conductive film 25, and the conductivity improving film 26. . On the other hand, as an etching gas for the ferroelectric film 24, a mixed gas of chlorine and argon is used.
[0073] また、第 1導電膜 23用のエッチングガスに対して導電性酸素ノ リア膜 22はエツチン グ耐性を有するので、このエッチングを終了した後でも下地導電膜 21の全面に導電 性酸素バリア膜 22は残存する。 [0073] In addition, since the conductive oxygen barrier film 22 has an etching resistance to the etching gas for the first conductive film 23, a conductive oxygen barrier is formed on the entire surface of the base conductive film 21 even after the etching is completed. The membrane 22 remains.
[0074] 続、て、図 5 (a)に示すように、ドライエッチングにより第 2ハードマスク 28aを除去す る。 [0074] Subsequently, as shown in FIG. 5A, the second hard mask 28a is removed by dry etching.
[0075] 次に、図 5 (b)に示す断面構造を得るまでの工程について説明する。  Next, steps required until a sectional structure shown in FIG.
[0076] まず、第 1ハードマスク 27a (図 5 (a)参照)をマスクとして用いながら、下地導電膜 2 First, while using the first hard mask 27a (see FIG. 5 (a)) as a mask, the underlying conductive film 2
1と導電性酸素ノ リア膜 22とをエッチングし、これらの膜をキャパシタ誘電体膜 24aの 下にのみ残す。このエッチングはドライエッチングにより行われ、そのエッチングガスと しては例えばアルゴンと塩素との混合ガスが使用される。 Etch 1 and the conductive oxygen noble film 22 to leave these films only under the capacitor dielectric film 24a. This etching is performed by dry etching, and as the etching gas, for example, a mixed gas of argon and chlorine is used.
[0077] また、このエッチングガスに対し第 1ハードマスク 27aもエッチングされるため、エツ チングの終了時には第 1ハードマスク 27aは除去される。 [0077] Further, since the first hard mask 27a is also etched by this etching gas, the first hard mask 27a is removed at the end of etching.
[0078] これにより、キャパシタ誘電体膜 24aの下には、下地導電膜 21、導電性酸素バリア 膜 22、及び第 1導電膜 23で構成される下部電極 23aが、第 1導電性プラグ 12aと電 気的に接続されるように形成される。そして、この下部電極 23a、キャパシタ誘電体膜Thus, under the capacitor dielectric film 24a, the lower electrode 23a composed of the base conductive film 21, the conductive oxygen barrier film 22, and the first conductive film 23 is connected to the first conductive plug 12a. It is formed to be electrically connected. And this lower electrode 23a, capacitor dielectric film
24a、及び上部電極 25aによりキャパシタ Qが構成される。 Capacitor Q is constituted by 24a and upper electrode 25a.
[0079] このように、スタック型の FeRAMでは、下部電極 23a、キャパシタ誘電体膜 24a、及 び上部電極 25aに共通のハードマスク 27aをエッチングマスクとして使用するので、 各層の位置合わせ余裕を取る必要が無い。従って、キャパシタ Qの微細化が可能と なり、 FeRAMの高集積ィ匕を図ることができるようになる。 [0079] As described above, in the stack type FeRAM, the hard mask 27a common to the lower electrode 23a, the capacitor dielectric film 24a, and the upper electrode 25a is used as an etching mask. There is no. Therefore, the capacitor Q can be miniaturized and high integration of FeRAM can be achieved.
[0080] 続!、て、図 6 (a)に示すように、キャパシタ Qを覆うアルミナ膜を厚さ約 40nmに形成 し、そのアルミナ膜を第 1キャパシタ保護絶縁膜 39とする。 Next, as shown in FIG. 6A, an alumina film covering the capacitor Q is formed to a thickness of about 40 nm, and the alumina film is used as the first capacitor protection insulating film 39.
[0081] 第 1キャパシタ保護絶縁膜 39を構成するアルミナは、水素の透過防止能力に優れ ているため、外部の水素はこの第 1キャパシタ保護絶縁膜 39によってブロックされ、 水素によるキャパシタ誘電体膜 24aの劣化を防止することができる。 [0081] Alumina constituting the first capacitor protective insulating film 39 is excellent in hydrogen permeation preventing ability. Therefore, external hydrogen is blocked by the first capacitor protection insulating film 39, and deterioration of the capacitor dielectric film 24a due to hydrogen can be prevented.
[0082] その第 1キャパシタ保護絶縁膜 39の成膜方法は特に限定されない。但し、隣接す るキャパシタ Qの間隔が狭いので、カバレッジ特性に優れた膜を形成することが可能 な ALD法で第 1キャパシタ保護絶縁膜 39を形成するのが好ましい。このように ALD法 を採用することにより、キャパシタ Qの側面に十分な厚さの第 1キャパシタ保護絶縁膜 39を形成することができ、横方向力も水素がキャパシタ Qに侵入するのを防止し易く なる。 The method for forming the first capacitor protective insulating film 39 is not particularly limited. However, since the interval between the adjacent capacitors Q is narrow, it is preferable to form the first capacitor protection insulating film 39 by the ALD method capable of forming a film having excellent coverage characteristics. By adopting the ALD method in this way, the first capacitor protective insulating film 39 having a sufficient thickness can be formed on the side surface of the capacitor Q, and the lateral force is also easy to prevent hydrogen from entering the capacitor Q. Become.
[0083] また、 ALD法を採用する場合、有機アルミニウム化合物としてトリメチルアルミニウム が使用される。そして、有機アルミニウム化合物と共に使用される酸化剤としては、水 素フリーなオゾンを使用する。酸化剤として水を使用する場合と比較して、オゾンを用 いると ALDの雰囲気中の水素量が少なくなり、第 1キャパシタ保護絶縁膜 39の成膜 時に水素によってキャパシタ誘電体膜 24aが劣化するのを抑制することができる。  [0083] When the ALD method is employed, trimethylaluminum is used as the organoaluminum compound. Hydrogen-free ozone is used as the oxidizing agent used with the organoaluminum compound. Compared with the case where water is used as the oxidizer, the amount of hydrogen in the atmosphere of ALD decreases when ozone is used, and the capacitor dielectric film 24a deteriorates due to hydrogen when the first capacitor protection insulating film 39 is formed. Can be suppressed.
[0084] なお、上記のように良好なカバレッジ特性が求められな 、場合は、スパッタ法により 第 1キャパシタ保護絶縁膜 39を形成してもよい。 Note that if good coverage characteristics are not required as described above, the first capacitor protective insulating film 39 may be formed by sputtering.
[0085] 次に、図 6 (b)に示す断面構造を得るまでの工程について説明する。 [0085] Next, steps required until a sectional structure shown in FIG.
[0086] 図 11は、この工程で使用される HDPCVD装置の構成図である。 FIG. 11 is a configuration diagram of an HDPCVD apparatus used in this process.
[0087] その装置では、チャンバ 100の上方に設けられたコイル 103に第 1高周波電源 104 が接続され、更に、基板載置台 101に第 2高周波電源 102が接続される。なお、コィ ル 103は、シリコン基板 1の主面と平行な面内において巻かれており、図ではその断 面が示されている。 In the apparatus, a first high-frequency power source 104 is connected to a coil 103 provided above the chamber 100, and a second high-frequency power source 102 is further connected to the substrate platform 101. The coil 103 is wound in a plane parallel to the main surface of the silicon substrate 1, and the cross section is shown in the drawing.
[0088] このように基板載置台 101に高周波電源を印加することで、シリコン基板 1にバイァ ス電圧が印加されるので、プラズマ化した反応ガスがシリコン基板 1に引き込まれる。 そのような反応ガスの中には、膜の堆積に寄与するものの他に、堆積した膜をスパッ タするものもある。このスパッタ作用により、キャパシタ Qの肩部では膜の堆積とスパッ タとが同時に行われ、該肩部に膜が厚く形成されるのが防がれる。これにより、キャパ シタ Qの側面の膜厚が均一にならされて、高アスペクトレシオのキャパシタ Q間に埋め 込み性の良い絶縁膜を形成することができる。 [0089] 本実施形態では、このような HDPCVD装置において、 SiH、 0、及び Arの混合ガス [0088] By applying a high frequency power supply to the substrate mounting table 101 in this way, a bias voltage is applied to the silicon substrate 1, so that the plasmad reaction gas is drawn into the silicon substrate 1. In addition to those that contribute to film deposition, some of these reactive gases spatter the deposited film. By this sputtering action, film deposition and sputtering are simultaneously performed on the shoulder portion of the capacitor Q, thereby preventing the film from being thickly formed on the shoulder portion. As a result, the film thickness of the side surface of the capacitor Q is made uniform, and an insulating film with good embedding can be formed between the capacitors Q having a high aspect ratio. In this embodiment, in such an HDPCVD apparatus, a mixed gas of SiH, 0, and Ar is used.
4 2  4 2
を反応ガスとして用い、以下の条件下で成膜を行う。  Is used as a reaction gas, and film formation is performed under the following conditions.
•SiH流 · · · ,Osccm  • SiH flow ···, Osccm
4  Four
•0流直 · · · 525sccm  • 0 flow straight · · · 525sccm
2  2
•Ar流量 · · ^Osccm  Ar flow · ^ Osccm
り土力 · · · 15mTorr  15mTorr
•第 1高周波電源 64の周波数 · ' · 13. 56MHz  • 1st high frequency power supply 64 frequency · '· 13. 56MHz
'第 1高周波電源 64のパヮ 3500W  '1st high frequency power supply 64 parts 3500W
'第 2高周波電源 62の周波数' · ·4ΜΗζ  'Frequency of second high frequency power supply 62'
'第 2高周波電源 62のパヮ 2400W  'Second high frequency power supply 62 2400W
•成膜温度 (キャパシタ Q直上の狙い温度) · · ' 300°C  • Deposition temperature (target temperature just above capacitor Q) · · '300 ° C
これにより、図 6 (b)に示すように、キャパシタ Q間のスペースを埋める厚さの酸ィ匕シ リコン膜が第 1キャパシタ保護絶縁膜 39の上に層間絶縁膜 40として形成される。 HD PCVD法で形成された層間絶縁膜 40は、埋め込み性が良いので、高集積ィ匕が進ん でキャパシタ Q間の間隔が狭くなつても、その間の層間絶縁膜 40に"す"が形成され ることがない。  As a result, as shown in FIG. 6B, an oxide silicon film having a thickness that fills the space between the capacitors Q is formed on the first capacitor protection insulating film 39 as an interlayer insulating film 40. Since the interlayer insulating film 40 formed by HD PCVD has good embedding properties, even if the high integration progresses and the interval between the capacitors Q is narrowed, "s" is formed in the interlayer insulating film 40 between them. There is nothing to do.
[0090] なお、層間絶縁膜 40の成膜温度は上記の 300°Cに限定されず、 250°C以上 350 [0090] Note that the deposition temperature of the interlayer insulating film 40 is not limited to the above 300 ° C, but 250 ° C or more 350
°C以下であってよ!、。この温度範囲の意義にっ 、ては後述する。 It must be below ° C! The significance of this temperature range will be described later.
[0091] 次に、図 7 (a)に示すように、層間絶縁膜 40の上面を CMP法により研磨して平坦ィ匕 する。 CMP後の層間絶縁膜 40の膜厚は、上部電極 25aの上で約 300nmである。 Next, as shown in FIG. 7A, the upper surface of the interlayer insulating film 40 is polished and flattened by the CMP method. The film thickness of the interlayer insulating film 40 after CMP is about 300 nm on the upper electrode 25a.
[0092] ところで、本願発明者が行った調査によれば、 HDPCVD法により形成された層間絶 縁膜 40には水分が多量に含まれていることが明ら力となった。その調査結果につい ては後述する。 By the way, according to the investigation conducted by the inventor of the present application, it has become clear that the interlayer insulating film 40 formed by the HDPCVD method contains a large amount of moisture. The survey results will be described later.
[0093] また、上記の CMPにおいても、スラリー中の水分が層間絶縁膜 40に浸透し、層間 絶縁膜 40に多くの水分が含有されることになる。  Also in the CMP described above, the moisture in the slurry permeates the interlayer insulating film 40 and the interlayer insulating film 40 contains a large amount of moisture.
[0094] このように層間絶縁膜 40に水分が多量に含まれていると、たとえ第 1キャパシタ保 護絶縁膜 39を形成しても、キャパシタ誘電体膜 24aが劣化する恐れがある。 As described above, if the interlayer insulating film 40 contains a large amount of moisture, even if the first capacitor protective insulating film 39 is formed, the capacitor dielectric film 24a may be deteriorated.
[0095] そこで、本実施形態では、図 7 (b)に示すように、オゾンと酸素との混合雰囲気中に お!、て層間絶縁膜 40をァニールし、層間絶縁膜 40中を脱水する。 Therefore, in the present embodiment, as shown in FIG. 7B, in a mixed atmosphere of ozone and oxygen. Then, the interlayer insulating film 40 is annealed and the interlayer insulating film 40 is dehydrated.
[0096] そのァニール条件は特に限定されな!、が、本実施形態では、ォゾナイザー (オゾン 発生器)で発生したオゾンと酸素との混合ガスを lOslmの流量でチャンバ(不図示)に 供給する。その混合ガスにおけるオゾン濃度は 200gZNm3である。そして、圧力が 0 . 133kPaの下で、基板温度を 350°C以上 600°C以下、例えば 400°Cにして 30分間 このァニーノレを行う。 The annealing conditions are not particularly limited! In this embodiment, however, a mixed gas of ozone and oxygen generated by an ozonizer (ozone generator) is supplied to a chamber (not shown) at a flow rate of lOslm. Ozone concentration in the mixture gas is 200gZNm 3. Then, the annealing is performed for 30 minutes under a pressure of 0.133 kPa and a substrate temperature of 350 ° C. to 600 ° C., for example, 400 ° C.
[0097] なお、ァニール時の基板温度を 600°Cよりも高くすると、層間絶縁膜 40において脱 水反応が顕著に起こり、大量の H 0が発生するので好ましくない。  Note that it is not preferable that the substrate temperature during annealing be higher than 600 ° C. because a dehydration reaction occurs remarkably in the interlayer insulating film 40 and a large amount of H 0 is generated.
2  2
[0098] また、成膜の際に一度加熱されている層間絶縁膜 40に対し、その成膜温度よりも 低 、基板温度でァニールをしたのでは、ァニールによる脱水効果が期待できな 、。 本実施形態では、層間絶縁膜 40の成膜温度が 250°C〜350°Cであるから、上記の ァニール時の基板温度を 350°C以上とすることで、層間絶縁膜 40を効果的に脱水 することができる。  Further, if the interlayer insulating film 40 once heated at the time of film formation is annealed at a substrate temperature lower than the film formation temperature, the dehydration effect due to the annealing cannot be expected. In this embodiment, since the deposition temperature of the interlayer insulating film 40 is 250 ° C. to 350 ° C., the interlayer insulating film 40 is effectively formed by setting the substrate temperature at the above annealing to 350 ° C. or higher. Can be dehydrated.
[0099] 更に、このァニールは、 CMPによって体積が減少した層間絶縁膜 40に対して行わ れるので、 CMP前に行う場合と比較して、層間絶縁膜 40から脱ガスする水分量を低 減することができ、層間絶縁膜 40から出る水分に起因してキャパシタ Qが劣化するの を抑制できる。  Furthermore, since this annealing is performed on the interlayer insulating film 40 whose volume has been reduced by CMP, the amount of moisture degassed from the interlayer insulating film 40 is reduced as compared with the case where it is performed before CMP. Therefore, it is possible to suppress the deterioration of the capacitor Q due to moisture from the interlayer insulating film 40.
[0100] 続いて、図 8 (a)に示すように、層間絶縁膜 40の上にスパッタ法により平坦なアルミ ナ膜を厚さ約 50應に形成し、そのアルミナ膜を第 2キャパシタ保護絶縁膜 41とする。  [0100] Subsequently, as shown in FIG. 8 (a), a flat alumina film is formed on the interlayer insulating film 40 by sputtering to a thickness of about 50 mm, and the alumina film is formed as a second capacitor protective insulating film. The film 41 is used.
[0101] この第 2キャパシタ保護絶縁膜 41は、平坦化された第 2層間絶縁膜 40上に形成さ れるため優れたカバレッジ特性が要求されず、上記のように安価なスパッタ法で形成 され得る。  [0101] Since the second capacitor protection insulating film 41 is formed on the planarized second interlayer insulating film 40, excellent coverage characteristics are not required, and can be formed by an inexpensive sputtering method as described above. .
[0102] ここで、第 2キャパシタ保護絶縁膜 41を構成するアルミナは、化学反応によるエッチ ングが困難である。従って、後述のエッチング工程で第 2キャパシタ保護絶縁膜 41に ホールを形成し易くし、且つ水素のブロック性能を高めるには、 ALD法を用いて緻密 で薄いアルミナ膜を第 2キャパシタ保護絶縁膜 41として形成するのが好ましい。その 場合、第 2キャパシタ保護絶縁膜 41は 30應程度に薄く形成される。  Here, the alumina constituting the second capacitor protective insulating film 41 is difficult to etch by a chemical reaction. Therefore, in order to facilitate the formation of holes in the second capacitor protective insulating film 41 in the etching process described later and to improve the hydrogen blocking performance, a dense and thin alumina film is used to form the second capacitor protective insulating film 41 using the ALD method. It is preferable to form as. In this case, the second capacitor protection insulating film 41 is formed as thin as about 30 °.
[0103] その後に、この第 2キャパシタ保護絶縁膜 41の上に、 TEOSガスを使用するプラズ マ CVD法により酸ィ匕シリコン膜を厚さ約 200應程度に形成し、この酸ィ匕シリコン膜を キャップ絶縁膜 42とする。 [0103] After that, a plasma using TEOS gas is formed on the second capacitor protective insulating film 41. A silicon oxide film is formed to a thickness of about 200 by CVD, and this silicon oxide film is used as a cap insulating film 42.
[0104] 続いて、図 8 (b)に示すように、各膜 39〜42をパターユングすることにより、上部電 極 25a上のこれらの膜に第 1ホール 40aを形成する。  Subsequently, as shown in FIG. 8 (b), the first holes 40a are formed in these films on the upper electrode 25a by patterning each of the films 39-42.
[0105] 次いで、ここまでの工程でキャパシタ誘電体膜 24aが受けたダメージを回復させる ため、不図示の炉内にシリコン基板 1を入れ、酸素雰囲気中で基板温度を 500°Cと する回復ァニールを行う。  [0105] Next, in order to recover the damage received by the capacitor dielectric film 24a in the steps so far, the silicon substrate 1 is placed in a furnace (not shown) and the substrate temperature is set to 500 ° C in an oxygen atmosphere. I do.
[0106] ここで、層間絶縁膜 40の上面と下面にそれぞれ第 1、第 2キャパシタ保護絶縁膜 39 、 41が形成されているので、この回復ァニールによって気化した層間絶縁膜 40中の 水分の逃げ道は第 1ホール 40aに限られる。従って、図 7 (b)のァニール工程を省くと 、回復ァニールの際に層間絶縁膜 40中の多量の水分が気化し、気化した水分によ つて層間絶縁膜 40やその下のキャパシタ Qが蒸し焼きの状態になり、第 2キャパシタ 保護絶縁膜 41にクラックが入ったりキャパシタ Qが劣化したりするという問題が発生す る。  Here, since the first and second capacitor protective insulating films 39 and 41 are formed on the upper surface and the lower surface of the interlayer insulating film 40, respectively, the escape path of moisture in the interlayer insulating film 40 vaporized by this recovery annealing Is limited to the first hole 40a. Therefore, if the annealing process in FIG. 7B is omitted, a large amount of moisture in the interlayer insulating film 40 is vaporized during the recovery annealing, and the interlayer insulating film 40 and the capacitor Q below it are steamed by the vaporized moisture. As a result, the second capacitor protective insulating film 41 is cracked or the capacitor Q deteriorates.
[0107] これに対し、本実施形態では、図 7 (b)のァニール工程により層間絶縁膜 40を予め 脱水してあるので、回復ァニールを行ってもキャパシタ Qが劣化することは無!、。  On the other hand, in this embodiment, since the interlayer insulating film 40 is dehydrated in advance by the annealing process of FIG. 7B, the capacitor Q does not deteriorate even if the recovery annealing is performed! ,.
[0108] 但し、図 7 (b)のァニール工程における基板温度が低すぎると、層間絶縁膜 40の脱 水が不十分となり、この回復ァニールの際に第 2キャパシタ保護絶縁膜 41にクラック が入る恐れがある。この点が懸念される場合には、図 7 (b)のァニールの基板温度を なるべく高ぐ例えば 500°Cに設定するのが好まし!/、。  However, if the substrate temperature in the annealing process of FIG. 7B is too low, the interlayer insulating film 40 is insufficiently dehydrated, and the second capacitor protective insulating film 41 is cracked during this recovery annealing. There is a fear. If this is a concern, it is preferable to set the substrate temperature of the annealer in Fig. 7 (b) as high as possible, for example, 500 ° C! /.
[0109] 次に、図 9 (a)に示すように、各膜 39〜42をパターユングすることにより、第 2導電 性プラグ 12bの上のこれらの膜に第 2ホール 40bを形成する。  Next, as shown in FIG. 9 (a), the second holes 40b are formed in these films on the second conductive plug 12b by patterning the films 39 to 42.
[0110] ところで、第 2ホール 40bの形成工程(図 9 (a) )や第 1ホール 40a (図 8 (b) )の形成 工程では、化学的にエッチングするのが困難なアルミナよりなる第 2キャパシタ保護 絶縁膜 41をエッチングする。ここで、その第 2キャパシタ保護絶縁膜 41が厚いと、こ のエッチングが困難となり、エッチングにより各ホール 40a、 40bを精度良く加工する のが難しくなる。よって、この点が懸念される場合は、既述のように、緻密な膜を薄く 成膜することが可能な ALD法により第 2キャパシタ保護絶縁膜 41を 30應程度に薄く 形成するのが好ましい。 Incidentally, in the formation process of the second hole 40b (FIG. 9 (a)) and the formation process of the first hole 40a (FIG. 8 (b)), the second hole 40b made of alumina, which is difficult to chemically etch, is used. Capacitor protection Insulating film 41 is etched. Here, if the second capacitor protection insulating film 41 is thick, this etching becomes difficult, and it becomes difficult to process the holes 40a and 40b with high accuracy by the etching. Therefore, if this is a concern, as described above, the second capacitor protective insulating film 41 can be thinned to about 30 mm by the ALD method, which can form a dense film thinly. Preferably formed.
[0111] 次に、図 9 (b)に示す断面構造を得るまでの工程について説明する。  [0111] Next, steps required until a sectional structure shown in FIG.
[0112] まず、キャップ絶縁膜 42上と第 1、第 2ホール 40a、 40b内に、グルー膜としてスパッ タ法により窒化チタン膜を形成する。  [0112] First, a titanium nitride film is formed as a glue film on the cap insulating film 42 and in the first and second holes 40a, 40b by a sputtering method.
[0113] 更に、 CVD法によりグルー膜の上にタングステン膜を形成し、このタングステン膜で 第 1、第 2ホール 40a、 40bを完全に埋め込む。 Further, a tungsten film is formed on the glue film by the CVD method, and the first and second holes 40a and 40b are completely filled with this tungsten film.
[0114] そして、キャップ絶縁膜 42上の不要なグルー膜とタングステン膜とを CMP法により 研磨して除去し、これらの膜を第 1、第 2ホール 40a、 40b内にのみ第 3、第 4導電性 プラグ 45a、 45bとして残す。 [0114] Then, unnecessary glue films and tungsten films on the cap insulating film 42 are polished and removed by CMP, and these films are removed only in the first and second holes 40a and 40b. Leave as conductive plugs 45a and 45b.
[0115] これらのプラグのうち、第 3導電性プラグ 45aは、キャパシタ Qの上部電極 25aと電 気的に接続される。 Of these plugs, the third conductive plug 45a is electrically connected to the upper electrode 25a of the capacitor Q.
[0116] 一方、第 4導電性プラグ 45bは、第 2導電性プラグ 12bに電気的に接続され、その 第 2導電性プラグ 12bと共にビット線の一部を構成する。このようにビット線を二段の 導電性プラグ 12b、 45bで構成することにより、それぞれのプラグ 12b、 45bが埋め込 まれるホールのアスペクト比を小さくすることができ、エッチングにより各ホールを容易 に形成することが可能となる。そのような二段のプラグによるコンタクト構造は、は via - to - viaコンタクトとち呼ばれる。  On the other hand, the fourth conductive plug 45b is electrically connected to the second conductive plug 12b and constitutes a part of the bit line together with the second conductive plug 12b. By configuring the bit line with the two-stage conductive plugs 12b and 45b in this way, the aspect ratio of the holes embedded in the plugs 12b and 45b can be reduced, and each hole can be easily etched. It becomes possible to form. Such a two-stage plug contact structure is called a via-to-via contact.
[0117] ここで、第 3導電性プラグ 45aのグルー膜を構成する窒化チタン膜が、上部電極 25 aを構成する第 1導電膜 25に触れると、上部電極 25aと第 3導電性プラグ 45aとの間 のコンタクト抵抗が高くなるという不都合がある。この点に鑑み、本実施形態では、上 部電極 25aの最上層に、イリジウムよりなる導電性向上膜 26を形成したので、上部電 極 25aと第 3導電性プラグ 45aとの間のコンタクト抵抗を低くすることができる。  Here, when the titanium nitride film constituting the glue film of the third conductive plug 45a touches the first conductive film 25 constituting the upper electrode 25a, the upper electrode 25a and the third conductive plug 45a There is a disadvantage that the contact resistance between the two becomes high. In view of this point, in this embodiment, since the conductivity improving film 26 made of iridium is formed on the uppermost layer of the upper electrode 25a, the contact resistance between the upper electrode 25a and the third conductive plug 45a is reduced. Can be lowered.
[0118] その後に、図 10に示すように、キャップ絶縁膜 42と各導電性プラグ 45a、 45bのそ れぞれの上にスパッタ法で金属積層膜を形成し、この金属積層膜をパターニングし て金属配線 47aとビット線用の導電性パッド 47bとを形成する。  Then, as shown in FIG. 10, a metal multilayer film is formed on the cap insulating film 42 and each of the conductive plugs 45a and 45b by sputtering, and this metal multilayer film is patterned. Thus, the metal wiring 47a and the conductive pad 47b for the bit line are formed.
[0119] その金属積層膜として、厚さ約 60應のチタン膜、厚さ約 30應の窒化チタン膜、厚 さ約 400nmの銅含有アルミニウム膜、厚さ約 5nmのチタン膜、及び厚さ約 70nmの窒 化チタン膜をこの順に形成する。 [0120] ここで、金属積層膜をパターユングする際、キャップ絶縁膜 42がエッチングのストツ パ膜として機能するので、薄い第 2キャパシタ保護絶縁膜 41がエッチングされて水素 ブロック効果が薄れるのを防止することができる。 [0119] As the metal laminated film, a titanium film having a thickness of about 60, a titanium nitride film having a thickness of about 30, a copper-containing aluminum film having a thickness of about 400 nm, a titanium film having a thickness of about 5 nm, and a thickness of about A 70 nm titanium nitride film is formed in this order. [0120] Here, when the metal laminated film is patterned, the cap insulating film 42 functions as an etching stopper film, thereby preventing the thin second capacitor protective insulating film 41 from being etched and diminishing the hydrogen blocking effect. can do.
[0121] 更に、キャップ絶縁膜 42を省いて第 2キャパシタ保護絶縁膜 41の上に金属配線 47 aを形成すると、第 2キャパシタ保護絶縁膜 41と金属配線 47aとのストレスの違いに起 因して金属配線 47aにストレスマイグレーションが発生する恐れがある。これに対し、 上記のように酸ィ匕シリコンよりなるキャップ絶縁膜 42の上に金属配線 47aを形成すれ ば、このようなストレスマイグレーションを抑制することができ、金属配線 47aの信頼性 を高めることができる。  [0121] Further, when the metal wiring 47a is formed on the second capacitor protection insulating film 41 without the cap insulating film 42, it is caused by the difference in stress between the second capacitor protection insulating film 41 and the metal wiring 47a. As a result, stress migration may occur in the metal wiring 47a. On the other hand, if the metal wiring 47a is formed on the cap insulating film 42 made of silicon oxide as described above, such stress migration can be suppressed and the reliability of the metal wiring 47a can be improved. Can do.
[0122] 以上により、本実施形態に係る半導体装置の基本構造が完成したことになる。  As described above, the basic structure of the semiconductor device according to this embodiment is completed.
[0123] 上記した本実施形態では、図 6 (b)の工程で HDPCVD法により層間絶縁膜 40を形 成し、図 7 (b)の工程において、オゾン含有雰囲気中において層間絶縁膜 40をァニ ールした。 In the present embodiment described above, the interlayer insulating film 40 is formed by the HDPCVD method in the step of FIG. 6B, and the interlayer insulating film 40 is removed in the ozone-containing atmosphere in the step of FIG. 7B. Nealed.
[0124] 本願発明者は、このようなオゾン含有雰囲気中でのァニールによる脱水効果を調 ベるため、種々のサンプルに対して昇温脱離ガス分析 (Thermal Desorption Spectres copy: TDS)を行った。  [0124] In order to investigate the dehydration effect due to annealing in such an ozone-containing atmosphere, the inventor of the present application performed thermal desorption spectra (TDS) on various samples. .
[0125] その結果を図 12に示す。  The results are shown in FIG.
[0126] 図 12において、サンプル A、 Bは、 TEOSガスを使用するプラズマ CVD法で厚さ 50 Onmの酸化シリコン膜をシリコンウェハ上に形成したものである。その CVD法は、 HDP CVD法ではなく、通常のプラズマ CVD法である。  In FIG. 12, samples A and B are obtained by forming a silicon oxide film with a thickness of 50 Onm on a silicon wafer by a plasma CVD method using TEOS gas. The CVD method is not an HDP CVD method but a normal plasma CVD method.
[0127] また、サンプル Aとサンプル Bとでは成膜条件を変えて 、る。サンプル Aは、層間絶 縁膜 40として形成した場合、回復ァニール時にクラックが入ったことのある条件であ る。そして、サンプル Bは、そのようなクラックが入らない条件である。  [0127] Sample A and sample B have different film formation conditions. When sample A was formed as an interlayer insulating film 40, it was a condition that had cracked during the recovery annealing. Sample B is a condition that does not cause such cracks.
[0128] また、サンプル Cは、上記した層間絶縁膜 40と同じ成膜条件を採用して、 HDPCVD 法によりシリコンウェハ上に厚さ 500nmの酸ィ匕シリコン膜を形成したものである。  [0128] Sample C is a film in which a 500 nm thick silicon oxide film is formed on a silicon wafer by HDPCVD using the same film formation conditions as those of the interlayer insulating film 40 described above.
[0129] そして、サンプル Dは、サンプル Cと同じ酸ィ匕シリコン膜に対し、図 7 (b)と同じ条件 でァニールを行ったものである。  [0129] Sample D is obtained by annealing the same silicon oxide film as Sample C under the same conditions as in Fig. 7 (b).
[0130] 更に、サンプル Eは、シリコンウェハが無い状態で調べたものであり、 TDS分析装置 のバックグラウンドを表して 、る。 [0130] Furthermore, Sample E was examined in the absence of a silicon wafer. Represents the background of
[0131] なお、この調査では、シリコンウェハを分析用に lcm X lcmの正方形に切り出すこと で各サンプル A〜Dとした。  [0131] In this investigation, each sample A to D was obtained by cutting a silicon wafer into lcm x lcm squares for analysis.
[0132] 図 12は、その TDS法において、 H 0に相当する質量数が 18のスペクトルのみをプ [0132] Figure 12 shows only the spectrum with a mass number of 18 corresponding to H 0 in the TDS method.
2  2
ロットしたものである。そして、横軸が TDSにおける温度、縦軸がイオン強度を表して いる。但し、横軸の温度は、サンプノレを設置してレ、るステージ温度を示すものであつ て、サンプノレの表面温度を表している訳ではない。  Lot. The horizontal axis represents the temperature at TDS, and the vertical axis represents the ionic strength. However, the temperature on the horizontal axis indicates the stage temperature when the sampnore is installed, and does not represent the surface temperature of the sampnore.
[0133] まず、 TEOSガスを使用する通常のプラズマ CVD法で形成されたサンプル Aとサン プル Bに着目すると、過去に回復ァニール時にクラックが入った条件で成膜されたサ ンプル Aでは、 250°Cあたりに脱水のピークがあることが分かる。これは、酸化シリコン 膜中のマクロ細孔に吸蔵されていた H Oが加熱によって外部に出たためと推測され、 [0133] First, focusing on Sample A and Sample B, which were formed by the usual plasma CVD method using TEOS gas, Sample A, which had been cracked during the recovery annealing in the past, had 250 It can be seen that there is a dehydration peak around ° C. This is presumed to be because H 2 O stored in the macropores in the silicon oxide film came out to the outside by heating,
2  2
サンプル Aの酸化シリコン膜がポーラスな膜であることを示している。  This shows that the silicon oxide film of Sample A is a porous film.
[0134] これに対して、本実施形態と同様の条件の HDPCVD法を採用したサンプル Cでは、 250°Cあたりから脱水が始まり、ピークを持つことなく高温側にかけて増大している。 このように高温側で増大するのは、酸ィ匕シリコン膜中に少な力 ず存在する Si—〇H 結合の OH基同士が脱水反応することにより出来た H 0が膜外に出たためと推測され [0134] On the other hand, in sample C employing the HDPCVD method under the same conditions as in the present embodiment, dehydration started around 250 ° C and increased toward the high temperature side without having a peak. The increase on the high temperature side is presumed to be because H 0 produced by the dehydration reaction between the OH groups of Si—OH bonds that existed in the silicon oxide film with little force has come out of the film. Is
2  2
る。  The
[0135] また、装置のパックグラウンドを示すサンプル Eとの差を考えると、 HDPCVD法を採 用したサンプル Cは、通常のプラズマ CVD法を採用したサンプル A、 Bと比較して膜 中に含まれる水分量が多レ、ことが分かる。  [0135] Considering the difference from sample E, which shows the device package ground, sample C using HDPCVD is included in the film compared to samples A and B using normal plasma CVD. It can be seen that the amount of water produced is large.
[0136] この結果より、 HDPCVD法で形成された酸化シリコン膜を層間絶縁膜 40として採用 する本実施形態では、層間絶縁膜 40に対する脱水処理が必須であることが理解で きる。 From this result, it can be understood that in this embodiment in which a silicon oxide film formed by the HDPCVD method is adopted as the interlayer insulating film 40, the dehydration treatment for the interlayer insulating film 40 is essential.
[0137] そこで、その脱水の効果を確認するため、脱水を行ったサンプル Dに着目すると、 ステージ温度力 S450°C程度においてサンプル Dから出る脱水量が激減している。  [0137] Thus, in order to confirm the effect of the dehydration, when attention is paid to the dehydrated sample D, the dehydration amount from the sample D is drastically reduced at a stage temperature force of about S450 ° C.
[0138] 図 13は、図 12におけるのと同じ構造のサンプル A〜Eについて、酸化シリコン膜を 成膜後、 1週間放置した後に TDS分析して得られたグラフである。  FIG. 13 is a graph obtained by TDS analysis of samples A to E having the same structure as in FIG. 12, after leaving a silicon oxide film for one week.
[0139] 図 12と比較して考えると、回復ァニール時にクラックが入ったサンプル Aでは、 250  [0139] Compared to Fig. 12, Sample A cracked during the recovery annealing was 250
れた捃紙 ( ) °Cあたりと 350°Cあたりの脱水ピークが増大していることが分かる。これは、放置させ ることで、酸ィ匕シリコン膜中のマクロ細孔にさらに H 0が吸蔵されたためと推測される Torn paper () It can be seen that the dehydration peaks at around ° C and around 350 ° C increased. This is presumed to be because H 0 was further occluded in the macropores in the silicon oxide silicon film by leaving it to stand.
2  2
[0140] 一方、上記のクラックが入らな 、サンプル Bでは、放置しても脱水ピークが増大する ことなぐ緻密な膜となっている。よって、埋め込み特性に乏しいという点を除けば、サ ンプル Bの酸ィ匕シリコン膜は FeRAMの層間絶縁膜に適した膜である。 [0140] On the other hand, in the case of Sample B without the above cracks, it is a dense film in which the dehydration peak does not increase even if left as it is. Therefore, the sample B silicon oxide film is suitable for the FeRAM interlayer insulation film, except that it has poor embedding characteristics.
[0141] これに対し、 HDPCVD法により酸化シリコン膜を形成したサンプル Cでは、 350°Cあ たりの脱水が図 12と比較して増大している。これは、 TEOSガスを使用して形成された サンプル Bと同様に、サンプル Cのシリコン酸ィ匕膜がポーラスな膜であることを示唆し ている。  [0141] On the other hand, in Sample C in which a silicon oxide film was formed by the HDPCVD method, dehydration per 350 ° C increased compared to FIG. This suggests that the silicon oxide film of sample C is a porous film, similar to sample B formed using TEOS gas.
[0142] また、オゾン含有雰囲気中においてァニールを行ったサンプル Dでは、図 12と比 較して脱水の増加が見られない。このことから、ァニールによって改質された酸ィ匕シリ コン膜は放置しても吸湿しないことが分かる。従って、 HDPCVD法で形成した層間絶 縁膜 40に対して上記のァニールを行った後は、第 2キャパシタ保護絶縁膜 41の形 成前に脱水を目的とするァニールを in-situで行わなくても、層間絶縁膜 40を FeRAM に使うことが可能となる。  [0142] In Sample D that was annealed in an ozone-containing atmosphere, no increase in dehydration was observed compared to FIG. From this, it can be seen that the acid-modified silicon film modified by annealing does not absorb moisture. Therefore, after the above annealing is performed on the interlayer insulating film 40 formed by the HDPCVD method, annealing for the purpose of dehydration is not performed in-situ before the formation of the second capacitor protective insulating film 41. However, the interlayer insulating film 40 can be used for FeRAM.
[0143] 次に、オゾン含有雰囲気中におけるァニールの効果を調べるために本願発明者が 行った TDS分析にっ 、て説明する。  [0143] Next, a TDS analysis conducted by the present inventor in order to investigate the effect of annealing in an ozone-containing atmosphere will be described.
[0144] 図 14は、その TDS分析の結果を示すグラフである。  [0144] FIG. 14 is a graph showing the results of the TDS analysis.
[0145] 図 14において、サンプル A、 D、 Eは図 12及び図 13で説明したのと同じである。  In FIG. 14, samples A, D, and E are the same as those described in FIGS.
[0146] 一方、サンプル Dl、 D2は、いずれも HDPCVD法により形成された酸化シリコン膜 に対して脱水のためのァニールを行ったものである。但し、そのァニール条件がサン プル D1とサンプル D2とでは異なる。 [0146] On the other hand, samples Dl and D2 were both subjected to annealing for dehydration on a silicon oxide film formed by the HDPCVD method. However, the annealing conditions are different between sample D1 and sample D2.
[0147] サンプル D1では、サンプル Dよりも高い 500°Cに基板温度を設定し、酸化シリコン 膜に対してオゾン含有雰囲気中でァニールをしたものである。なお、そのガス流量、 圧力、処理時間はサンプル Dと同じである。 [0147] In sample D1, the substrate temperature was set to 500 ° C higher than that of sample D, and the silicon oxide film was annealed in an ozone-containing atmosphere. The gas flow rate, pressure, and processing time are the same as Sample D.
[0148] また、サンプル D2は、 100%の酸素雰囲気中において、基板温度を 500°Cにして 酸ィ匕シリコン膜に対してァニールを行ったものである。 [0149] サンプル Dlでは、サンプル Dのような高温側での脱水量の大幅な増大が見られな い。高温側での脱水は、酸ィ匕シリコン膜中に少な力 ず存在する Si— OH結合の OH 基同士が脱水反応することで起こるものである。よって、上記の結果から、酸化シリコ ン膜に対するァニールの際に基板温度を高めることで、膜中の Si— OH結合が減少 すると推測される。 [0148] Sample D2 was obtained by annealing an oxide silicon film at a substrate temperature of 500 ° C in a 100% oxygen atmosphere. [0149] Sample Dl does not show a significant increase in the amount of dehydration on the high temperature side like Sample D. Dehydration on the high temperature side is caused by a dehydration reaction between OH groups of Si—OH bonds that exist in the silicon oxide film with little force. Therefore, it can be inferred from the above results that the Si—OH bonds in the film are reduced by raising the substrate temperature during annealing of the silicon oxide film.
[0150] また、そのサンプル D1の脱水プロファイルは、サンプルの加熱温度が 600°Cまで の範囲では FeRAMの製造に適しているサンプル Aと同じプロファイルを示した。  [0150] In addition, the dehydration profile of Sample D1 showed the same profile as Sample A, which is suitable for FeRAM production, when the sample heating temperature was up to 600 ° C.
[0151] ここで、酸ィ匕シリコン膜の脱水効果を高めるには、酸ィ匕シリコン膜に対するァニール の雰囲気に、本実施形態のように酸素よりも酸ィ匕力の強いオゾンを添加するのが好 ましいと考えられる。  [0151] Here, in order to enhance the dehydration effect of the silicon oxide film, ozone having a higher acid power than oxygen is added to the annealing atmosphere for the silicon oxide film as in this embodiment. Seems to be preferred.
[0152] ところが、オゾンを用いるサンプル D1と、酸素を用いるサンプル D2とを比較すると、 両者のプロファイルに大きな違いは無い。これは、サンプル D1では、ァニール時の 基板温度が 500°Cと高温であるため、酸ィ匕シリコン膜に対するァニールの際にオゾン が死活したためと考えられる。  [0152] However, comparing sample D1 using ozone with sample D2 using oxygen, there is no significant difference between the profiles. This is presumably because, in sample D1, the substrate temperature during annealing was as high as 500 ° C, so ozone was killed during annealing of the silicon oxide film.
[0153] 但し、 500°Cよりも低い基板温度でァニールをすれば、ァニール雰囲気中において 死活するオゾンが減少し、酸素のみの雰囲気中でァニールする場合よりも酸ィ匕シリコ ン膜の脱水効果が向上すると考えられる。  [0153] However, if annealing is performed at a substrate temperature lower than 500 ° C, ozone that is dead and active is reduced in the annealing atmosphere, and the dehydration effect of the acid silicon film is lower than that in annealing in an oxygen-only atmosphere. Is thought to improve.
[0154] ここで、ァニール時の基板温度を高温にすると、層間絶縁膜 40の脱水効果が向上 する一方、層間絶縁膜 40の水分が第 1キャパシタ保護絶縁膜 39を通ってキャパシタ 誘電体膜 24aまで拡散することが考えられる。よって、ァニール時にキャパシタ誘電 体膜 24aの劣化を防止するという点力 すると、ァニールの際の基板温度はなるべく 低いのが好ましい。上記のようにァニール雰囲気にオゾンを添加すると、このように基 板温度を低めても、オゾンの強 ヽ酸化力によって層間絶縁膜 40に対する脱水効果 が維持されると考えられる。  Here, when the substrate temperature during annealing is increased, the dehydration effect of the interlayer insulating film 40 is improved, while the moisture in the interlayer insulating film 40 passes through the first capacitor protective insulating film 39 and forms the capacitor dielectric film 24a. It is conceivable to diffuse to Therefore, the substrate temperature during annealing is preferably as low as possible in view of preventing deterioration of the capacitor dielectric film 24a during annealing. If ozone is added to the annealing atmosphere as described above, it is considered that the dehydration effect on the interlayer insulating film 40 is maintained by the strong oxidizing power of ozone even if the substrate temperature is lowered in this way.
[0155] ところで、 HDPCVD法で形成された酸ィ匕シリコン膜は、成膜温度によって膜中に含 まれる水分量が異なると考えられる。  [0155] By the way, it is considered that the amount of water contained in the silicon oxide film formed by the HDPCVD method varies depending on the film formation temperature.
[0156] 図 15及び図 16は、 HDPCVD法を用いて、様々な成膜温度で形成された酸化シリ コン膜の水分量を TDS分析により調査して得られたグラフである。 [0157] これらの図では、図 12〜図 14と同様に、シリコンウェハが無い状態で測定したもの をサンプル Eとした。 FIGS. 15 and 16 are graphs obtained by investigating the moisture content of the silicon oxide film formed at various film formation temperatures by TDS analysis using the HDPCVD method. [0157] In these figures, sample E was measured in the absence of a silicon wafer, as in Figs.
[0158] また、サンプル D3〜D5は、 HDPCVD法における成膜温度をそれぞれ 250°C、 30 [0158] Samples D3 to D5 had film formation temperatures of 250 ° C and 30 respectively in the HDPCVD method.
0°C、 400°Cにしてシリコンウェハ上に酸化シリコン膜を形成したものである。 A silicon oxide film is formed on a silicon wafer at 0 ° C and 400 ° C.
[0159] 図 15及び図 16から明らかなように、三つのサンプルの中で成膜温度が最も高いサ ンプル D5では、膜中から放出される水分量が最も少ない。そして、成膜温度が最も 低いサンプル D3では、その水分量が最も多い。 As is apparent from FIGS. 15 and 16, sample D5, which has the highest film formation temperature among the three samples, has the smallest amount of water released from the film. Sample D3, which has the lowest film formation temperature, has the highest water content.
[0160] 層間絶縁膜 40として使用される酸ィ匕シリコン膜は、水分含有量が少ないのが好まし いのであるが、水分含有量を減らすべく成膜温度を 400°Cと高くすると、熱により活 性ィ匕した水素によってキャパシタ誘電体膜 24aが劣化する恐れがある。 [0160] The silicon oxide film used as the interlayer insulating film 40 preferably has a low moisture content. However, if the deposition temperature is increased to 400 ° C in order to reduce the moisture content, Therefore, the capacitor dielectric film 24a may be deteriorated by the hydrogen activated.
[0161] 従って、層間絶縁膜 40の成膜時にキャパシタ誘電体膜 24aが劣化するのを防ぐと いう観点力 すると、層間絶縁膜 40の成膜温度はなるべく低ぐ例えば 350°C以下と するのが好ましい。 Therefore, from the viewpoint of preventing the capacitor dielectric film 24a from being deteriorated when the interlayer insulating film 40 is formed, the film forming temperature of the interlayer insulating film 40 is as low as possible, for example, 350 ° C. or less. Is preferred.
[0162] 一方、成膜温度を低くしすぎると、図 15及び図 16に示したように、今度は層間絶縁 膜 40の膜中に含まれる水分量が多くなる。このように水分量が多くなると、層間絶縁 膜 40をァニールする際に、膜中の水分によってキャパシタ誘電体膜 24aが劣化して しまう。  On the other hand, if the film forming temperature is too low, the amount of moisture contained in the interlayer insulating film 40 is increased as shown in FIGS. 15 and 16. When the amount of moisture increases in this way, when the interlayer insulating film 40 is annealed, the capacitor dielectric film 24a is deteriorated by moisture in the film.
[0163] よって、水分によるキャパシタ誘電体膜 24aの劣化を防止するという観点力もすると 、層間絶縁膜 40の成膜温度を低くしすぎるのは好ましくなぐ成膜温度を 250°C以上 とするのが好ましい。  [0163] Therefore, from the viewpoint of preventing the deterioration of the capacitor dielectric film 24a due to moisture, it is not preferable that the film formation temperature of the interlayer insulating film 40 is too low. The film formation temperature is 250 ° C or higher. preferable.
[0164] このように、 250°C以上 350°C以下の成膜温度で HDPCVD法により層間絶縁膜 40 を形成することで、キャパシタ誘電体膜 24aが受けるダメージを最小限にすることがで きる。  [0164] In this way, by forming the interlayer insulating film 40 by the HDPCVD method at a film forming temperature of 250 ° C or higher and 350 ° C or lower, damage to the capacitor dielectric film 24a can be minimized. .
[0165] 上記したように、 FeRAMの製造プロセスでは、 HDPCVD法によって層間絶縁膜 40 を成膜する時に、キャパシタ誘電体膜 24aが水素によって劣化しないように成膜温度 をなるベく低くし、且つその成膜後に層間絶縁膜 40をァニールして脱水することによ り、プロセス中にキャパシタ Qが受けるダメージを最も効果的に低減することができる。  [0165] As described above, in the FeRAM manufacturing process, when the interlayer insulating film 40 is formed by the HDPCVD method, the film formation temperature is made very low so that the capacitor dielectric film 24a is not deteriorated by hydrogen, and By annealing and dehydrating the interlayer insulating film 40 after the film formation, the damage to the capacitor Q during the process can be most effectively reduced.
[0166] これに対し、 FeRAM以外の通常のロジック品等では、キャパシタの劣化を気にする 必要が無いため、層間絶縁膜の成膜温度を低くする必要が無ぐ FeRAMよりも高い 成膜温度で層間絶縁膜を形成することができる。そして、このように高い成膜温度に より得られた層間絶縁膜は、既述のように膜中の水分量が少ないため、本実施形態 のような脱水を目的としたァニールを行う必要が無い。 [0166] On the other hand, in ordinary logic products other than FeRAM, we are concerned about capacitor deterioration. Since there is no need, the interlayer insulating film can be formed at a higher deposition temperature than that of FeRAM. Since the interlayer insulating film obtained at such a high deposition temperature has a small amount of moisture in the film as described above, it is not necessary to perform annealing for dehydration as in this embodiment. .
[0167] ところで、上記した本実施形態では、酸素とオゾンとの混合雰囲気中にぉ 、て、層 間絶縁膜 40に対してァニールを行 、、層間絶縁膜 40を脱水した。  In the present embodiment described above, annealing was performed on the interlayer insulating film 40 in a mixed atmosphere of oxygen and ozone, and the interlayer insulating film 40 was dehydrated.
[0168] このような脱水を目的とするなら、酸素とオゾンとの混合雰囲気の他に、他の酸化性 ガス、例えば二酸化窒素 (N 0)の含有雰囲気中で層間絶縁膜 40をァニールすること  [0168] For the purpose of such dehydration, the interlayer insulating film 40 is annealed in an atmosphere containing other oxidizing gas, for example, nitrogen dioxide (N 0), in addition to the mixed atmosphere of oxygen and ozone.
2  2
ち考免られる。  I will be relieved.
[0169] しかし、本願発明者が行った実験によると、 HDPCVD法により低温で成膜されて多 くの水分を含む酸ィ匕シリコン膜に対して、このような二酸ィ匕窒素のァニールを行って も、脱水の効果が見られな力つた。  [0169] However, according to experiments conducted by the inventors of the present application, such a diacid-nitrogen anneal was applied to an oxide silicon film containing a large amount of moisture formed at a low temperature by the HDPCVD method. Even when I went there, I couldn't see the effect of dehydration.
[0170] 従って、 FeRAMの層間絶縁膜として好適な低温の HDPCVD法で形成された酸ィ匕シ リコン膜は、本実施形態のようにオゾンと酸素との混合雰囲気中におけるァニールに よって、効果的に脱水することができる。 [0170] Therefore, the silicon oxide film formed by the low-temperature HDPCVD method suitable as an interlayer insulating film of FeRAM is effective by annealing in a mixed atmosphere of ozone and oxygen as in this embodiment. Can be dehydrated.
[0171] 但し、図 14のサンプル D2に見られるように、 100%の酸素雰囲気中でも脱水の効 果がある程度見られるので、酸素のみの雰囲気中でそのァニールを行ってもよい。 [0171] However, as seen in sample D2 in Fig. 14, the effect of dehydration is seen to some extent even in a 100% oxygen atmosphere, so annealing may be performed in an oxygen-only atmosphere.
[0172] また、ァニール雰囲気を不活性ガスで希釈することも考えられるが、雰囲気の酸ィ匕 力を高めるという観点力 すると、雰囲気に不活性ガスを添加するのは好ましくなぐ 上記のようにオゾンと酸素のみ力 なる雰囲気、又は酸素のみ力 なる雰囲気でァ- ールを行うのが好ましい。 [0172] Although it is conceivable to dilute the annealing atmosphere with an inert gas, it is not preferable to add an inert gas to the atmosphere from the viewpoint of increasing the oxidizing power of the atmosphere. It is preferable to carry out the fire in an atmosphere that only powers oxygen and oxygen.
[0173] 更に、上記では、プラズマ化していない雰囲気中において層間絶縁膜 40をァニー ルして脱水した力 プラズマ雰囲気中にお!、てこのァニールを行うようにしてもよ!、。 [0173] Further, in the above, annealing force is obtained by annealing the interlayer insulating film 40 in a non-plasma atmosphere, and the annealing may be performed in a plasma atmosphere!

Claims

請求の範囲 The scope of the claims
[1] 半導体基板の上方に下地絶縁膜を形成する工程と、  [1] forming a base insulating film above the semiconductor substrate;
前記下地絶縁膜の上に、下部電極、強誘電体材料よりなるキャパシタ誘電体膜、 及び上部電極を備えたキャパシタを形成する工程と、  Forming a capacitor including a lower electrode, a capacitor dielectric film made of a ferroelectric material, and an upper electrode on the base insulating film;
前記キャパシタを覆う第 1キャパシタ保護絶縁膜を形成する工程と、  Forming a first capacitor protective insulating film covering the capacitor;
前記第 1キャパシタ保護絶縁膜の上に、前記半導体基板側にバイアス電圧を印加 するプラズマ CVD法により、層間絶縁膜を形成する工程と、  Forming an interlayer insulating film on the first capacitor protective insulating film by a plasma CVD method for applying a bias voltage to the semiconductor substrate;
オゾン又は酸素を含む雰囲気中で前記層間絶縁膜をァニールする工程と、 を有することを特徴とする半導体装置の製造方法。  Annealing the interlayer insulating film in an atmosphere containing ozone or oxygen. A method of manufacturing a semiconductor device, comprising:
[2] 前記層間絶縁膜をァニールする工程は、オゾンと酸素のみよりなる雰囲気中で行 われることを特徴とする請求項 1に記載の半導体装置の製造方法。 [2] The method for manufacturing a semiconductor device according to [1], wherein the step of annealing the interlayer insulating film is performed in an atmosphere composed of only ozone and oxygen.
[3] 前記層間絶縁膜をァニールする工程は、酸素のみよりなる雰囲気中で行われること を特徴とする請求項 1に記載の半導体装置の製造方法。 [3] The method for manufacturing a semiconductor device according to [1], wherein the step of annealing the interlayer insulating film is performed in an atmosphere made of only oxygen.
[4] 前記層間絶縁膜をァニールする工程は、前記層間絶縁膜の成膜温度よりも高い基 板温度で行われることを特徴とする請求項 1に記載の半導体装置の製造方法。 4. The method for manufacturing a semiconductor device according to claim 1, wherein the step of annealing the interlayer insulating film is performed at a substrate temperature higher than a film forming temperature of the interlayer insulating film.
[5] 前記層間絶縁膜をァニールする工程は、基板温度を 350°C以上 600°C以下にして 行われることを特徴とする請求項 1に記載の半導体装置の製造方法。 [5] The method of manufacturing a semiconductor device according to [1], wherein the step of annealing the interlayer insulating film is performed at a substrate temperature of 350 ° C. or higher and 600 ° C. or lower.
[6] 前記第 1キャパシタ保護絶縁膜を形成する工程において、該第 1キャパシタ保護絶 縁膜として ALD(Atomic Layer Deposition)法によりアルミナ膜を形成することを特徴と する請求項 1に記載の半導体装置の製造方法。 6. The semiconductor according to claim 1, wherein, in the step of forming the first capacitor protective insulating film, an alumina film is formed by an ALD (Atomic Layer Deposition) method as the first capacitor protective insulating film. Device manufacturing method.
[7] 前記 ALD法において、有機アルミニウム化合物とオゾンとの混合ガスを反応ガスと して用いることを特徴とする請求項 6に記載の半導体装置の製造方法。 7. The method for manufacturing a semiconductor device according to claim 6, wherein in the ALD method, a mixed gas of an organoaluminum compound and ozone is used as a reaction gas.
[8] CMP(Chemical Mechanical Polishing)法により前記層間絶縁膜の表面を平坦ィ匕する 工程を更に有し、 [8] The method further comprises a step of flattening the surface of the interlayer insulating film by a CMP (Chemical Mechanical Polishing) method,
前記層間絶縁膜をァニールする工程は、前記平坦ィ匕の後に行われることを特徴と する請求項 1に記載の半導体装置の製造方法。  2. The method of manufacturing a semiconductor device according to claim 1, wherein the step of annealing the interlayer insulating film is performed after the flattening.
[9] 前記層間絶縁膜をァニールする工程の後に、前記層間絶縁膜の上に第 2キャパシ タ保護絶縁膜を形成する工程と、 前記上部電極の上の前記第 1キャパシタ保護絶縁膜、前記層間絶縁膜、及び前記 第 2キャパシタ保護絶縁膜にホールを形成する工程と、 [9] After the step of annealing the interlayer insulating film, a step of forming a second capacitor protection insulating film on the interlayer insulating film; Forming a hole in the first capacitor protective insulating film, the interlayer insulating film, and the second capacitor protective insulating film on the upper electrode;
前記ホールを形成する工程の後に、酸素含有雰囲気中にお 、て前記キャパシタ誘 電体膜をァニールする工程を更に有することを特徴とする請求項 1に記載の半導体 装置の製造方法。  2. The method of manufacturing a semiconductor device according to claim 1, further comprising a step of annealing the capacitor dielectric film in an oxygen-containing atmosphere after the step of forming the holes.
[10] 前記層間絶縁膜の成膜温度は、 250°C以上 350°C以下であることを特徴とする請 求項 1に記載の半導体装置の製造方法。  [10] The method for manufacturing a semiconductor device according to claim 1, wherein the deposition temperature of the interlayer insulating film is 250 ° C. or higher and 350 ° C. or lower.
[11] 前記キャパシタを形成する工程は、 [11] The step of forming the capacitor comprises:
前記下地絶縁膜の上に第 1導電膜を形成する工程と、  Forming a first conductive film on the base insulating film;
前記第 1導電膜の上に強誘電体膜を形成する工程と、  Forming a ferroelectric film on the first conductive film;
前記強誘電体膜の上に第 2導電膜を形成する工程と、  Forming a second conductive film on the ferroelectric film;
前記第 2導電膜の上にハードマスクを形成する工程と、  Forming a hard mask on the second conductive film;
前記ハードマスクで覆われていない領域の前記第 1導電膜、前記強誘電体膜、及 び前記第 2導電膜をエッチングして除去する工程とを含むことを特徴とする請求項 1 に記載の半導体装置の製造方法。  The method of claim 1, further comprising: etching and removing the first conductive film, the ferroelectric film, and the second conductive film in a region not covered with the hard mask. A method for manufacturing a semiconductor device.
PCT/JP2006/316423 2006-08-22 2006-08-22 Method for manufacturing semiconductor device WO2008023409A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2006/316423 WO2008023409A1 (en) 2006-08-22 2006-08-22 Method for manufacturing semiconductor device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2006/316423 WO2008023409A1 (en) 2006-08-22 2006-08-22 Method for manufacturing semiconductor device

Publications (1)

Publication Number Publication Date
WO2008023409A1 true WO2008023409A1 (en) 2008-02-28

Family

ID=39106503

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2006/316423 WO2008023409A1 (en) 2006-08-22 2006-08-22 Method for manufacturing semiconductor device

Country Status (1)

Country Link
WO (1) WO2008023409A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0613565A (en) * 1992-02-25 1994-01-21 Ramtron Internatl Corp Method for forming ferroelectric memory circuit and method for forming ferroelectric capacitor
JP2003273332A (en) * 2002-03-19 2003-09-26 Fujitsu Ltd Semiconductor device and its manufacturing method
JP2004193280A (en) * 2002-12-10 2004-07-08 Fujitsu Ltd Semiconductor device and its manufacturing method
JP2006049795A (en) * 2004-06-28 2006-02-16 Fujitsu Ltd Semiconductor device and its manufacturing method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0613565A (en) * 1992-02-25 1994-01-21 Ramtron Internatl Corp Method for forming ferroelectric memory circuit and method for forming ferroelectric capacitor
JP2003273332A (en) * 2002-03-19 2003-09-26 Fujitsu Ltd Semiconductor device and its manufacturing method
JP2004193280A (en) * 2002-12-10 2004-07-08 Fujitsu Ltd Semiconductor device and its manufacturing method
JP2006049795A (en) * 2004-06-28 2006-02-16 Fujitsu Ltd Semiconductor device and its manufacturing method

Similar Documents

Publication Publication Date Title
JP5109341B2 (en) Semiconductor device and manufacturing method thereof
US8236643B2 (en) Method of manufacturing semiconductor device including ferroelectric capacitor
JP5205741B2 (en) Manufacturing method of semiconductor device
KR100774898B1 (en) Method of manufacturing semiconductor device
JP5568845B2 (en) Manufacturing method of semiconductor device
US8183109B2 (en) Semiconductor device and method of manufacturing the same
US8349679B2 (en) Semiconductor device and method of manufacturing the same
JP2008060126A (en) Semiconductor device and manufacturing method thereof
JP5024046B2 (en) Semiconductor device and manufacturing method thereof
JP2008124330A (en) Method for fabricating semiconductor device
JP4953580B2 (en) Manufacturing method of semiconductor device
US7728370B2 (en) Semiconductor device and manufacturing method of the same
JP2004087807A (en) Semiconductor device and method for manufacturing the same
JP4946145B2 (en) Manufacturing method of ferroelectric memory
KR101262432B1 (en) Method for manufacturing semiconductor devuce
JP4809354B2 (en) Semiconductor device and manufacturing method thereof
JP5239294B2 (en) Manufacturing method of semiconductor device
WO2008023409A1 (en) Method for manufacturing semiconductor device
JP4809367B2 (en) Semiconductor device and manufacturing method thereof
JP2010087350A (en) Semiconductor device, and method of manufacturing the same
JP2010153897A (en) Method for producing semiconductor device
WO2008004297A1 (en) Semiconductor device comprising capacitor and method for manufacturing same

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 06796637

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

NENP Non-entry into the national phase

Ref country code: RU

122 Ep: pct application non-entry in european phase

Ref document number: 06796637

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP