Patent Grant
7233516
This application claims priority under 35 U.S.C. §119 on Patent Application No. 2004-185045 filed in Japan on Jun. 23, 2004, the entire contents of which are hereby incorporated by reference.
(1) Field of the Invention
The present invention relates to a semiconductor device in which a CMOS logic section and a plurality of DRAM sections used for different applications are formed together on the same semiconductor substrate, and more particularly relates to a DRAM-embedded system LSI having a reduced power consumption and an increased processing speed while ensuring a sufficient signal holding characteristic.
(2) Description of Related Art
In recent years, DRAM-embedded system LSI (large scale integrated) chips have been attracting public attention. In a DRAM-embedded system LSI chip, a CMOS (complementary metal oxide semiconductor) logic section and a general-purpose DRAM (dynamic random access memory) section as a memory device, which used to be formed on separate chips, are formed together on the same chip to satisfy a demand for diversified semiconductor devices (see Japanese Unexamined Patent Publication No. 2000-232076).
For example, a DRAM-embedded system LSI chip used for an image processing application, or the like, includes a DRAM section as a memory device for storing an image information signal, and a CMOS logic section for retrieving necessary information from the DRAM section and performing an arithmetic operation based on the retrieved information, and the DRAM section and the CMOS logic section are formed together on the same chip.
A DRAM-embedded system LSI chip as described above realizes a higher communication speed than that realized by older techniques where data or information is exchanged between a CMOS logic section and a DRAM section that are formed on separate chips. A semiconductor device in which a CMOS logic section and a plurality of DRAM sections are formed together on the same chip includes, for example, a CMOS logic section formed on a silicon substrate and a DRAM section including trench capacitors or stacked capacitors. In this relation, the trench capacitors represent cell capacitors (memory cell capacitors) of a particular type formed in the silicon substrate, and the stacked capacitors represent cell capacitors of a particular type formed on the silicon substrate.
It was technically difficult to form a CMOS logic section and a DRAM section together on the same chip because they had large areas. However, with recent miniaturization techniques, a DRAM-embedded system LSI chip having a chip size less than or equal to 100 mm2 has been realized. Now, a plurality of CMOS logic sections and a plurality of application-specific DRAM sections are formed on the same chip, whereas a single chip accommodated only one CMOS logic section and one application-specific DRAM section with older techniques.
However, the DRAM sections formed on a conventional DRAM-embedded system LSI chip all have cell capacitors of the same structure. Therefore, it is difficult to provide, on the same chip, a DRAM section that needs to be accessed at a high speed and a DRAM section that requires a sufficient signal holding characteristic. This problem will be further discussed below.
The value of a signal stored in each memory cell of a DRAM is determined based on the charge stored in the capacitor of the memory cell.
However, even if the charge Qh is stored in the cell capacitor (whereby the signal stored in the memory cell is supposed to be determined to be “high”), the state of the signal being held in the cell capacitor may change over time due to, for example, a leak current through the capacitor insulating film of the cell capacitor, an OFF-state leak current through the transfer transistor, or a leak current from the substrate contact portion (a portion where the cell capacitor and the semiconductor substrate are connected to each other) to the semiconductor substrate. As a result, even if a charge that is large enough for the signal to be determined “high” is initially stored in a cell capacitor of the DRAM section, the stored charge may gradually leak out over time, whereby the signal of the cell capacitor is no longer determined to be “high”. Where Qh is the charge initially stored in a cell capacitor, t is the charge holding time, which is the amount of time elapsed since the charge Qh is initially stored, Q′ is the amount of charge stored in the cell capacitor after the charge holding time t, and Ileak is the amount of leak current, i.e., the amount of charge that leaks out of the cell capacitor, Q′ can be expressed as in Expression (1) below. For the sake of simplicity, it is assumed that the leak current Ileak is constant over time.
Q′=Qh−Ileak·t (1)
As can be seen from Expression (1), the charge stored in the cell capacitor of the DRAM section decreases over time. Therefore, where a sufficient signal holding characteristic is required (i.e., where it is required that the charge holding time t, which is the amount of time until the charge Q′ decreases to be less than or equal to Qs, is long), the charge Qh initially stored in the cell capacitor needs to be increased. Thus, the capacitance of the cell capacitor needs to be as large as about 30 fF.
On the other hand, for the first DRAM section 12 (the DRAM section from/to which signals are read/written at a high speed, i.e., the DRAM section operating at a high frequency), the voltage to be applied to the cell capacitor (operating voltage) is switched at a higher speed than for the second DRAM section 13. Therefore, the next operation is accidentally started before the voltage to be applied to the capacitor satisfactorily reaches a predetermined voltage. As a result, the amount of charge stored in the capacitor does not reach a predetermined value. To be specific, where an equivalent circuit diagram illustrating a memory cell of the first DRAM section 12 is the same as that illustrating a memory cell of the second DRAM section 13 illustrated in
More specifically, when a signal (information) stored in a memory cell is determined to be high or low, the amount of stored charge different from the amount of charge representing real information only remains in the capacitor, leading to misoperation. If not only the operating voltage of the first DRAM section 12 but also the operating voltage of the second DRAM section 13 is increased to ensure the amount of charge stored in the first DRAM section 12, this would lead to increased power consumption for not only the first DRAM section 12 but also the second DRAM section 13 intended to store charge (hold a signal sufficiently long).
In view of the above, it is an object of the present invention to provide a DRAM-embedded device in which a CMOS logic section and a plurality of DRAM sections used for different applications are formed together on the same semiconductor substrate, and which realizes both reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic.
In order to achieve the above object, a semiconductor device of the present invention comprises: a first DRAM section formed on a semiconductor substrate and composed of a plurality of first memory cells; and a second DRAM section formed on the semiconductor substrate and composed of a plurality of second memory cells, wherein the operating speed of the first DRAM section is higher than that of the second DRAM section, and the capacitance of each said first memory cell is larger than that of each said second memory cell.
According to the semiconductor device of the present invention, the first DRAM section and the second DRAM section are provided on the same semiconductor substrate, and the capacitance of each first memory cell of the high-speed operating, first DRAM section is set to be larger than that of each said second memory cell of the low-speed operating, second DRAM section. Assuming that the operating voltage of the first DRAM section is the same as that of the second DRAM section, charge can sufficiently be stored in each said first memory cell even if the voltage applied to the first memory cell has not reached a predetermined voltage due to a high operating speed of the first DRAM section. The reason for this is that each said first memory cell has a large capacitance. For example, the amount of charge stored in a capacitor of the first memory cell can be made equal to the amount of charge stored in a capacitor of the second memory cell. Therefore, even if the first DRAM section is operated at a high speed, a signal stored in the first memory cell does not become an error signal. This can prevent misoperation from occurring and the first DRAM section can be operated at a high speed. Furthermore, as in a known DRAM-embedded device that will be described later, the operating voltage of the first DRAM section that needs to operate at high speed does not have to be increased, resulting in the reduced consumed power. On the other hand, in the second DRAM section, its low operating speed allows the voltage applied to the second memory cell to be increased sufficiently. This can provide a sufficient charge holding characteristic (signal holding characterisitic). Thus, a desired DRAM operation can be expected.
As described above, according to the semiconductor device of the present invention, a plurality of DRAM sections are composed of memory cells of a plurality of kinds having different capacitances. Thus, it is possible to realize a DRAM-embedded device that, even when a plurality of DRAM sections are mounted together with a CMOS logic section on the same chip, realizes both reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic, by optimizing the capacitance of the memory cell (accurately, a capacitive element (capacitor) located in the memory cell) of each DRAM section according to the application thereof.
On the other hand, for the known DRAM-embedded device, a plurality of DRAM sections with different operating speeds are provided on the same semiconductor substrate, and respective memory cells of all the DRAM sections are set to have equivalent capacitances. In this case, a lower voltage than the voltage to be applied to a capacitive element of each memory cell of a low-speed operating DRAM section is applied to a capacitive element of each memory cell of a high-speed operating DRAM section. The reason for this is that in the high-speed operating DRAM section, the next operation is accidentally started before a sufficient voltage is applied to the capacitive element. Therefore, only a charge smaller than a reference charge is stored in the capacitive element of each memory cell of the high-speed operating DRAM section. Furthermore, if the operating voltage of the high-speed operating DRAM section is set high to make up for such shortage of charge, another problem arises in which the consumed power increases.
According to the semiconductor device of the present invention, the operating voltage of the first DRAM section may be the same as that of the second DRAM section.
According to the semiconductor device of the present invention, the size of a capacitive element in each said first memory cell is preferably larger than that of a capacitive element in each said second memory cell.
Thus, the capacitance of each first memory cell can certainly be made larger than that of each second memory cell. To be specific, a capacitor lower electrode of each first memory cell may be provided to have a shape obtained by combining the shapes of respective capacitor lower electrodes of a plurality of second memory cells. In this case, even with the formation of a plurality of memory cell regions having different capacitance values on the same semiconductor substrate, as long as only the respective capacitor lower electrodes in the memory cell regions are designed to have different layouts while maintaining their standard shapes, the respective other members, e.g., contact plugs between the substrate and each capacitor lower electrode or impurity layers, located in the memory cell regions need not be designed to have different layouts. As a result, the region where a capacitive element of each first memory cell is formed can be made larger than the region where a capacitive element of each second memory cell is formed without complicating a semiconductor device fabricating process. Therefore, a first capacitance of each first memory cell can be made larger than a second capacitance of each second memory cell. Thus, the above-mentioned effects of the semiconductor device of the present invention can certainly be obtained.
According to the semiconductor device of the present invention, a capacitive element for each said first memory cell preferably has a configuration obtained by combining two or more capacitive elements each for each said second memory cell.
This can make the capacitance of each first memory cell larger than that of each second memory cell.
In this case, a lower electrode of the capacitive element in each said second memory cell may be electrically connected through a corresponding plug to the semiconductor substrate, and a lower electrode of the capacitive element in each said first memory cell may be composed of a plurality of adjacent electrode parts each having the same configuration as the lower electrode of the second memory cell, said plurality of electrode parts being electrically connected through a common plug to the semiconductor substrate. Thus, even with the formation of a plurality of memory cell regions having different capacitance values on the same semiconductor substrate, as long as only respective plugs located in the memory cell regions are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or impurity layers, located in the memory cell regions need not be designed to have different layouts. As a result, the region where a capacitive element of each first memory cell is formed can be made larger than the region where a capacitive element of each second memory cell is formed without designing a new layout and complicating a semiconductor device fabricating process. Therefore, a first capacitance of each first memory cell can be made larger than a second capacitance of each second memory cell. Thus, the above-mentioned effects of the semiconductor device of the present invention can certainly be obtained.
In this case, a lower electrode of the capacitive element in each said second memory cell may be electrically connected through a corresponding plug to a corresponding impurity layer in the semiconductor substrate, and a lower electrode of the capacitive element in each said first memory cell may be composed of a plurality of adjacent electrode parts each having the same configuration as the lower electrode of the second memory cell, said plurality of electrode parts being electrically connected through individual plugs to a common impurity layer in the semiconductor substrate. Thus, even with the formation of a plurality of memory cell regions having different capacitance values on the same semiconductor substrate, as long as only respective impurity layers located in the memory cell regions are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or plugs, located in the memory cell regions need not be designed to have different layouts. As a result, the region where a capacitive element of each first memory cell is formed can be made larger than the region where a capacitive element of each second memory cell is formed without designing a new layout and complicating a semiconductor device fabricating process. Therefore, a first capacitance of each first memory cell can be made larger than a second capacitance of each second memory cell. Thus, the above-mentioned effects of the semiconductor device of the present invention can certainly be obtained.
A method for fabricating a semiconductor device of a first aspect of the present invention comprises the steps of: forming interlayer insulating films on parts of a semiconductor substrate located in a first memory region and a second memory region, respectively; forming a first plug in the interlayer insulating film located in the first memory region to allow electrical connection with the semiconductor substrate and forming a second plug in the interlayer insulating film located in the second memory region to allow electrical connection with the semiconductor substrate; forming a second capacitor lower electrode on the interlayer insulating film located in the second memory region to allow electrical connection with the second plug and forming a first capacitor lower electrode on the interlayer insulating film located in the first memory region to allow electrical connection with the first plug and become larger than the second capacitor lower electrode; and successively forming a capacitor insulating film and a capacitor upper electrode on each of the first and second capacitor lower electrodes, thereby forming a first capacitive element and a second capacitive element on the interlayer insulating films located on the first memory region and the second memory region, respectively.
According to the method for fabricating the semiconductor device of a first aspect, capacitive elements are formed over the same semiconductor substrate located in the first memory region and the second memory region, respectively, and the capacitive elements located in the first memory region and the second memory region are different from each other only in the configurations of their capacitor lower electrodes. To be specific, for example, each capacitor lower electrode located in the first memory region is provided to have a shape obtained by combining two or more of a plurality of capacitor lower electrodes for the second memory region. Thus, the first capacitor lower electrode is made larger than the second capacitor lower electrode. Therefore, the capacitance of each first capacitive element located in the first memory region can be made larger than that of each second capacitive element located in the second memory region. In view of the above, also when a DRAM section composed of memory cells each having a first capacitive element is formed in the first memory region and the DRAM section is operated at a high speed (in other words, voltage is applied to the DRAM section at a high speed), charge can sufficiently be stored in a memory cell of the DRAM section without increasing the operating voltage of the DRAM section. More particularly, since a signal stored in the memory cell having the first capacitive element does not become an error signal, this can prevent misoperation from occurring and the DRAM section can be operated at a high speed. In addition, since the operating voltage of the DRAM section does not have to be increased, this can reduce consumed power. On the other hand, when a DRAM section composed of memory cells each having a second capacitive element is formed in the second memory region, the low-speed operation of this DRAM section allows sufficient increase in the voltage applied to each memory cell having a second capacitive element. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
As described above, since in the method for fabricating a semiconductor device of the first aspect a plurality of DRAM sections are composed of capacitive elements of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic, by optimizing the capacitance of the capacitive element for each DRAM section according to the application thereof.
A method for fabricating a semiconductor device of a second aspect of the present invention comprises the steps of: forming interlayer insulating films on parts of a semiconductor substrate located in a first memory region and a second memory region, respectively; forming a first plug in the interlayer insulating film located in the first memory region to allow electrical connection with the semiconductor substrate and forming a second plug in the interlayer insulating film located in the second memory region to allow electrical connection with the semiconductor substrate; forming a second capacitor lower electrode on the interlayer insulating film located in the second memory region to allow electrical connection with the second plug and forming a first capacitor lower electrode on the interlayer insulating film located in the first memory region, said first capacitor lower electrode being composed of a plurality of electrode parts each having the same configuration as the second capacitor lower electrode and each allowing electrical connection with the first plug; and successively forming a capacitor insulating film and a capacitor upper electrode on each of the first and second capacitor lower electrodes, thereby forming a first capacitive element and a second capacitive element on the interlayer insulating films located on the first memory region and the second memory region, respectively, wherein the plurality of electrode parts are connected to the first plug.
According to the method for fabricating a semiconductor device of the second aspect, when capacitive elements are formed over the same semiconductor substrate located in the first memory region and the second memory region, respectively, capacitive elements each having a configuration obtained by combining two or more of capacitive elements for the second memory region are formed in the first memory region. To be specific, capacitor lower electrodes (first capacitor lower electrodes) each composed of a plurality of electrode parts each having the same configuration as each capacitor lower electrode (second capacitor lower electrode) located in the second memory region are formed in the first memory region, and the plurality of electrode parts are connected to a common plug (first plug). More particularly, while a second capacitor lower electrode is electrically connected through a corresponding plug (second plug) to the semiconductor substrate, the plurality of electrode parts of the first capacitor lower electrode are electrically connected through the first plug that is larger than each second plug to the semiconductor substrate. Thus, as long as only respective plugs located in memory cell regions are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or impurity layers, located in the memory regions need not be designed to have different layouts. As a result, the capacitance of each first capacitive element located in the first memory region can be made larger than that of each second capacitive element located in the second memory region without complicating a semiconductor device fabricating process. In view of the above, also when a DRAM section composed of memory cells each having a first capacitive element is formed in the first memory region and the DRAM section is operated at a high speed (in other words, voltage is applied to the DRAM section at a high speed), charge can sufficiently be stored in a memory cell of the DRAM section without increasing the operating voltage of the DRAM section. More particularly, since a signal stored in the memory cell having the first capacitive element does not become an error signal, this can prevent misoperation from occurring and the DRAM section can be operated at a high speed. In addition, since the operating voltage of the DRAM section does not have to be increased, this can reduce consumed power. On the other hand, when a DRAM section composed of memory cells each having a second capacitive element is formed in the second memory region, the low-speed operation of this DRAM section allows sufficient increase in the voltage applied to each memory cell having a second capacitive element. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
As described above, since in the method for fabricating a semiconductor device of the second aspect a plurality of DRAM sections are composed of capacitive elements of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic, by optimizing the capacitance of the capacitive element for each DRAM section according to the application thereof.
A method for fabricating a semiconductor device of a third aspect of the present invention comprises the steps of: forming a first impurity layer in a part of a semiconductor substrate located in a first memory region and forming a second impurity layer in a part of a semiconductor substrate located in a second memory region; forming interlayer insulating films on parts of the semiconductor substrate in which the first and second impurity layers are formed, said parts of the semiconductor substrate being located in the first and second memory regions, respectively; forming a plurality of first plugs in the interlayer insulating film located in the first memory region to allow electrical connection with the first impurity layer and forming a second plug in the interlayer insulating film located in the second memory region to allow electrical connection with the second impurity layer; forming a second capacitor lower electrode on the interlayer insulating film located in the second memory region to allow electrical connection with the second plug and forming a first capacitor lower electrode on the interlayer insulating film located in the first memory region, said first capacitor lower electrode being composed of a plurality of electrode parts each having the same configuration as the second capacitor lower electrode and allowing electrical connection with the plurality of first plugs, respectively; and successively forming a capacitor insulating film and a capacitor upper electrode on each of the first and second capacitor lower electrodes, thereby forming a first capacitive element and a second capacitive element on the interlayer insulating films located on the first memory region and the second memory region, respectively, wherein the electrode parts are electrically connected through the first plugs, respectively, to the first impurity layer.
According to the method for fabricating a semiconductor device of the third aspect, when capacitive elements are formed over the same semiconductor substrate located in the first memory region and the second memory region, respectively, capacitive elements each having a configuration obtained by combining two or more of the capacitive elements for the second memory region are formed in the first memory region. To be specific, capacitor lower electrodes (first capacitor lower electrodes) each composed of a plurality of electrode parts each having the same configuration as each capacitor lower electrode (second capacitor lower electrode) located in the second memory region are formed in the first memory region, and the plurality of electrode parts are electrically connected through individual plugs (a plurality of first plugs) to the same impurity layer (first impurity layer). More particularly, while a second capacitor lower electrode is electrically connected through a corresponding plug (second plug) to a corresponding impurity layer (second impurity layer), the plurality of electrode parts of the first capacitor lower electrode are electrically connected through individual first plugs to the first impurity layer that is larger than the second impurity layer. Thus, as long as only respective impurity layers located in memory cell regions are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or plugs, located in the memory regions need not be designed to have different layouts. As a result, the capacitance of each first capacitive element located in the first memory region can be made larger than that of each second capacitive element located in the second memory region without complicating a semiconductor device fabricating process. In view of the above, also when a DRAM section composed of memory cells each having a first capacitive element is formed in the first memory region and the DRAM section is operated at a high speed (in other words, voltage is applied to the DRAM section at a high speed), charge can sufficiently be stored in a memory cell of the DRAM section without increasing the operating voltage of the DRAM section. More particularly, since a signal stored in the memory cell having the first capacitive element does not become an error signal, this can prevent misoperation from occurring and the DRAM section can be operated at a high speed. In addition, since the operating voltage of the DRAM section does not have to be increased, this can reduce consumed power. On the other hand, when a DRAM section composed of memory cells each having a second capacitive element is formed in the second memory region, the low-speed operation of this DRAM section allows sufficient increase in the voltage applied to each memory cell having a second capacitive element. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
As described above, since in the method for fabricating a semiconductor device of the third aspect a plurality of DRAM sections are composed of capacitive elements of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic, by optimizing the capacitance of the capacitive element for each DRAM section according to the application thereof.
The present invention relates to a DRAM-embedded system LSI. When this DRAM-embedded system LSI is applied to a semiconductor device in which a plurality of DRAM sections used for different applications are mounted together with a CMOS logic section on the same semiconductor substrate, this semiconductor device can provide reduced power consumption and increased processing speed while ensuring sufficient signal holding characteristic. Therefore, the present invention is very useful.
A semiconductor device according to a first embodiment of the present invention will now be described with reference to the drawings.
As illustrated in
A feature of the present embodiment is that the capacitances of the memory cells of the first DRAM section 102 and the second DRAM section 103 are individually set at different values for their respective applications. To be specific, the capacitance of each memory cell of the first DRAM section 102 having a higher operating speed than the second DRAM section 103 is set to be larger than the capacitance of each memory cell of the second DRAM section 103.
As illustrated in
On the other hand, as illustrated in
Subsequently, the charge storing characteristics of the first capacitor 105A illustrated in
To be specific, for example, Ca=15 fF and (VDDa−VPa)=V1=0.75 V for the first capacitor 105A of the first DRAM section 102, whereas Cb=10 fF and (VDDb−VPb)=V1=0.75 V for the second capacitor 105B of the second DRAM section 103.
In this embodiment, it is assumed that, for example, the operating speed of the first DRAM section 102 is 100 or more MHz (more specifically, approximately a few hundreds of MHz) and the operating speed of the second DRAM section 103 is less than 100 MHz (more specifically, approximately a few tens of MHz). When there exists the difference of the operating speed between the DRAM sections 102 and 103 as described above, a voltage applied to the second capacitor 105B of the second DRAM section 103 having a low operating speed is sufficiently increased to reach V1. On the other hand, a voltage applied to the first capacitor 105A of the first DRAM section 102 having a high operating speed is not increased to reach V1 but reaches only a lower value than V1, for example, approximately V2=0.5 V.
The amount of charge stored in the first and second capacitors 105A and 105B can be represented as a physical quantity whose value is equal to the area of the hatched portion in
Qa=Ca·V2=15fF·0.5V=7.5fC (2)
On the other hand, the amount of charge Qb stored in the second capacitor 105B is 7.5 fC based on Expression (3) below.
Qb=Cb·V1=10fF·0.75V=7.5fC (3)
As described above, in this embodiment, the amount of charge Qa stored in the first capacitor 105A can be set to be substantially the same as that of charge Qb stored in the second capacitor 105B. This allows both the first DRAM section 102 having a high operating speed and the second DRAM section 103 having a low operating speed to store enough charge in the respective capacitors in operating the DRAMs. Therefore, for example, a signal stored in a memory cell can be determined to be high or low without misoperation. In other words, even when the first DRAM section 102 and the second DRAM section 103 are operated at the same voltage, both the first DRAM section 102 for operating at a high speed and the second DRAM section 103 for operating at a low speed with the aim of charge retention can be operated properly. Furthermore, the operating voltage of the first DRAM section 102 that needs to operate at a high speed does not have to be increased, resulting in the reduced consumed power.
Thus, it is possible to realize a DRAM-embedded device that realizes both a reduced power consumption and an increased processing speed while ensuring a sufficient signal holding characteristic, by optimizing the capacitance of the memory cell for each DRAM section according to the application thereof, as in the present embodiment.
Meanwhile, where in the present embodiment Ca=Cb=10fF as in the known semiconductor device, the amount of charge stored in the first capacitor 105A is calculated in the following manner: 10fF×0.5 V=5.0 fC. This value becomes smaller than the aforementioned Qa, i.e., 7.5 fC, by 2 or more fC.
In the first embodiment, the capacitance of each of the first and second capacitors 105A and 105B and the operating speed of each of the first and second DRAM sections 102 and 103 can arbitrarily be determined within a certain range according to the purpose of the associated DRAM section. In this relation, the capacitance Ca of the memory cell (first capacitor 105A) of the first DRAM section 102 having a higher operating speed than the second DRAM section 103 need be set to be larger than the capacitance Cb of the memory cell (second capacitor 105B) of the second DRAM section 103.
Although in the first embodiment the first DRAM section 102 and the second DRAM section 103 are set to have the same operating voltage, they may be set to have different operating voltages.
A semiconductor device and a method for fabricating the same according to the second embodiment of the present invention will now be described with reference to the drawings.
Like the first embodiment illustrated in
As illustrated in
Meanwhile, as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
A feature of the present embodiment is that the first capacitor lower electrode 209A formed in the first memory region RA has a larger area than the second capacitor lower electrode 209B formed in the second memory region RB. To be specific, as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
Next, a method for fabricating the semiconductor device according to the second embodiment, more specifically, a method for fabricating the semiconductor device illustrated in
First, as illustrated in
Next, as illustrated in
Next, as illustrated in
Thereafter, a third interlayer insulating film 215 is formed across the entire surface of the semiconductor substrate 200. Then, as illustrated in
As described above, according to the second embodiment, cell capacitors are formed on parts of the semiconductor substrate 200 located in the first memory region RA and the second memory region RB, and the cell capacitors located in the first memory region RA are different from those located in the second memory region RB in the structures of the capacitor lower electrodes 209. To be specific, for example, the first capacitor lower electrode 209A located in the first memory region RA is provided to have a shape obtained by combining the shapes of a plurality of second capacitor lower electrodes 209B located in the second memory region RB. Thus, the first capacitor lower electrode 209A is made larger than the second capacitor lower electrode 209B. Therefore, the cell capacitance of the cell capacitor located in the first memory region RA can be made larger than that of the cell capacitor located in the second memory region RB. In view of the above, also when the first DRAM section 102 composed of memory cells each having a large capacitance is formed in the first memory region RA and the first DRAM section 102 is operated at a high speed, charge can sufficiently be stored in a memory cell of the first DRAM section 102 without increasing the operating voltage of the first DRAM section 102. More particularly, since a signal stored in a memory cell of the first DRAM section 102 does not become an error signal, this can prevent misoperation from occurring and the first DRAM section 102 can be operated at a high speed. In addition, since the operating voltage of the first DRAM section 102 does not have to be increased, this can reduce consumed power. On the other hand, when the second DRAM section 103 composed of memory cells each having a small capacitance is formed in the second memory region RB, the low-speed operation of the second DRAM section 103 allows sufficient increase in the voltage applied to each memory cell thereof. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
In summary, since in the present embodiment a plurality of DRAM sections are composed of cell capacitors of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both a reduced power consumption and an increased processing speed while ensuring a sufficient signal holding characteristic, by optimizing the capacitance of the memory cell for each DRAM section according to the application thereof.
Since in the second embodiment the first capacitor lower electrode 209A is provided to have a shape obtained by combining the shapes of the plurality of second capacitor lower electrodes 209B, the following effects can be obtained. Even with the formation of a plurality of memory cell regions having different capacitance values on the same semiconductor substrate 200, as long as only the respective capacitor lower electrodes 209 in the memory cell regions (i.e., the first memory region RA and the second memory region RB) are designed to have different layouts while maintaining their standard shape, the respective other members, e.g., the contact plugs 208 or the impurity diffusion layers 206, located in the memory cell regions need not be designed to have different layouts. As a result, a capacitor formation region located in the first memory region RA can be made larger than that located in the second memory region RB without complicating a semiconductor device fabricating process. Therefore, each memory cell located in the first memory region RA can have a larger capacitance than that located in the second memory region RB. Thus, the above-mentioned effects of the present embodiment can certainly be obtained.
In the second embodiment, the cup-like capacitor lower electrodes 209 are formed in the first memory region RA and the second memory region RB, respectively, and only the bottom and inner wall surfaces of each capacitor lower electrode 209 are utilized as a cell capacitor. Instead, as illustrated in
Furthermore, although in the second embodiment phosphorus-containing silicon is used for the first capacitor lower electrode 209A, the second capacitor lower electrode 209B and the capacitor upper electrode 211, silicon containing any other impurity or any other metal may alternatively be used.
Moreover, although in the second embodiment an ON film is used as the capacitor insulating film 210, any other insulating film may alternatively be used.
A semiconductor device according to a modification of the second embodiment of the present invention will now be described with reference to the drawings.
A plane structure of the present modification is similar to that of the second embodiment illustrated in
As illustrated in
The above-described modification also provides the effects similar to those of the second embodiment.
A semiconductor device and a method for fabricating the same according to a third embodiment of the present invention will now be described with reference to the drawings.
Like the first embodiment illustrated in
As illustrated in
Meanwhile, as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
More particularly, as illustrated in
On the other hand, features of the present invention is as follows: as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
Next, a method for fabricating the semiconductor device according to the third embodiment, more specifically, a method for fabricating the semiconductor device illustrated in
First, as illustrated in
Next, contact holes are formed, by photolithography and etching, in parts of the first interlayer insulating film 207 located in the first memory region RA and the second memory region RB to reach the first impurity diffusion layer 206A and the second impurity diffusion layer 206B, respectively. Then, a conductive film is formed across the entire surface of the semiconductor substrate 200 to completely fill each contact hole. Then, an unnecessary portion of the conductive film located outside each contact hole is removed, thereby forming, as illustrated in
In this case, the size (thickness) of the contact plug 301 formed in the first memory region RA is made equivalent to the size of a first capacitor lower electrode that will be formed later, i.e., the size obtained by combining two or more of second capacitor lower electrodes 209B for the second memory region RB. More particularly, in the present embodiment, the first memory region RA and the second memory region RB are different from each other in the layouts of masks used in a process step of patterning the first interlayer insulating film 207 to form contact plugs in the memory regions RA and RB, respectively.
Next, as illustrated in
Next, as illustrated in
Thereafter, a third interlayer insulating film 215 is formed across the entire surface of the semiconductor substrate 200. Then, as illustrated in
As described above, according to the third embodiment, when cell capacitors are formed on parts of the semiconductor substrate 200 located in the first memory region RA and the second memory region RB, cell capacitors are formed in the first memory region RA to each have a configuration obtained by combining two or more of the cell capacitors for the second memory region RB. To be specific, a capacitor lower electrode (first capacitor lower electrode) composed of a plurality of electrode parts 209A and 302 is formed in the first memory region RA, and the electrode parts 209 and 302 each have the same configuration as the capacitor lower electrode (the second capacitor lower electrode 209B) located in the second memory region RB. Simultaneously, the plurality of electrode parts 209A and 302 are connected through the same plug (first plug) 301 to the semiconductor substrate 200. Thus, the cell capacitance of the cell capacitor located in the first memory region RA can be made larger than that of the cell capacitor located in the second memory region RB.
More specifically, according to the third embodiment, as long as only plugs through which the semiconductor substrate 200 is connected to the capacitor lower electrodes in the first memory region RA and the second memory region RB, respectively, are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or impurity diffusion layers, located in the memory regions RA and RB need not be designed to have different layouts. As a result, the capacitance of each cell capacitor located in the first memory region RA can be made larger than that of each cell capacitor located in the second memory region RB without complicating a semiconductor device fabricating process. Thus, also when the first DRAM section 102 composed of memory cells each having a large cell capacitance is formed in the first memory region RA and the first DRAM section 102 is operated at a high speed, charge can sufficiently be stored in the memory cells of the first DRAM section 102 without increasing the operating voltage of the first DRAM section 102. More particularly, since a signal stored in a memory cell of the first DRAM section 102 does not become an error signal, this can prevent misoperation from occurring and the first DRAM section 102 can be operated at a high speed. In addition, since the operating voltage of the first DRAM section 102 does not have to be increased, this can reduce consumed power. On the other hand, when the second DRAM section 103 composed of memory cells each having a small capacitance is formed in the second memory region RB, the low-speed operation of the second DRAM section 103 allows sufficient increase in the voltage applied to each memory cell thereof. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
As described above, since in the present embodiment a plurality of DRAM sections are composed of cell capacitors of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both a reduced power consumption and an increased processing speed while ensuring a sufficient signal holding characteristic, by optimizing the capacitance of the memory cell for each DRAM section according to the application thereof.
In the third embodiment, the cup-like electrode parts 209A and 302 and the cup-like capacitor lower electrode 209B are formed in the first memory region RA and the second memory region RB, respectively, and only the bottom and inner wall surfaces of the electrode parts 209A and 302 and the capacitor lower electrode 209B are utilized as cell capacitors. Instead, not only the bottom and inner wall surfaces of the cup-like electrode parts 209A and 302 and the cup-like capacitor lower electrode 209B but also the outer wall surfaces thereof may be utilized as cell capacitors (see
Furthermore, although in the third embodiment phosphorus-containing silicon is used for the electrode parts 209A and 302 and the capacitor lower electrode 209B or the capacitor upper electrode 211, silicon containing any other impurity or any other metal may alternatively be used.
Moreover, although in the third embodiment an ON film is used as the capacitor insulating film 210, any other insulating film may alternatively be used.
A semiconductor device and a method for fabricating the same according to a fourth embodiment of the present invention will now be described with reference to the drawings.
Like the first embodiment illustrated in
As illustrated in
Meanwhile, as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
More particularly, as illustrated in
On the other hand, features of the present invention is as follows: as illustrated in
Furthermore, as illustrated in
Furthermore, as illustrated in
Next, a method for fabricating the semiconductor device according to the fourth embodiment, more specifically, a method for fabricating the semiconductor device illustrated in
First, as illustrated in
Next, a first interlayer insulating film 207 is deposited across the entire surface of the semiconductor substrate 200. Then, a plurality of (e.g., two) contact holes are formed, by photolithography and etching, in a part of the first interlayer insulating film 207 located in the first memory region RA to reach the first impurity diffusion layer 401, and a single contact hole is formed, by the same method, in a part of the first interlayer insulating film 207 located in the second memory region RB to reach the second impurity diffusion layer 206B. Thereafter, a conductive film is formed across the entire surface of the semiconductor substrate 200 to completely fill each contact hole. Then, an unnecessary portion of the conductive film located outside each contact hole is removed, thereby forming, as illustrated in
Next, as illustrated in
Next, as illustrated in
Thereafter, a third interlayer insulating film 215 is formed across the entire surface of the semiconductor substrate 200. Then, as illustrated in
As described above, according to the fourth embodiment, when cell capacitors are formed on parts of the semiconductor substrate located in the first memory region RA and a second memory region RB, cell capacitors are formed in the first memory region RA to each have a configuration obtained by combining two or more of the cell capacitors for the second memory region RB. To be specific, a capacitor lower electrode (first capacitor lower electrode) composed of a plurality of electrode parts 209A and 403 is formed in the first memory region RA, and the electrode parts 209A and 403 each have the same configuration as the capacitor lower electrode (the second capacitor lower electrode 209B) located in the second memory region RB. Simultaneously, the plurality of electrode parts 209A and 403 are electrically connected through the respective plugs (plugs 208A and 402) to the same impurity diffusion layer (first impurity diffusion layer) 401. More particularly, while the second capacitor lower electrode 209B is electrically connected through the corresponding plug 208B to the corresponding impurity diffusion layer 206B, a plurality of electrode parts 209A and 403 of the first capacitor lower electrode are electrically connected through the corresponding plugs 208A and 402 to the first impurity diffusion layer 401 larger than the second impurity diffusion layer 206B. Thus, the cell capacitance of the cell capacitor located in the first memory region RA can be made larger than that located in the second memory region RB.
More specifically, according to the fourth embodiment, as long as only impurity diffusion layers electrically connected through the plugs to the capacitor lower electrodes in the first memory region RA and the second memory region RB, respectively, are designed to have different layouts, the respective other members, e.g., capacitor lower electrodes or plugs, located in the memory regions RA and RB need not be designed to have different layouts. As a result, the capacitance of each cell capacitor located in the first memory region RA can be made larger than that of each cell capacitor located in the second memory region RB without complicating a semiconductor device fabricating process. Thus, also when the first DRAM section 102 composed of memory cells each having a large cell capacitance is formed in the first memory region RA and the first DRAM section 102 is operated at a high speed, charge can sufficiently be stored in the memory cells of the first DRAM section 102 without increasing the operating voltage of the first DRAM section 102. More particularly, since a signal stored in a memory cell of the first DRAM section 102 does not become an error signal, this can prevent misoperation from occurring and the first DRAM section 102 can be operated at a high speed. In addition, since the operating voltage of the first DRAM section 102 does not have to be increased, this can reduce consumed power. On the other hand, when the second DRAM section 103 composed of memory cells each having a small capacitance is formed in the second memory region RB, the low-speed operation of the second DRAM section 103 allows sufficient increase in the voltage applied to each memory cell thereof. This can provide a sufficient charge holding characteristic (signal holding characteristic). Thus, a desired DRAM operation can be expected.
In view of the above, since in the present embodiment a plurality of DRAM sections are composed of cell capacitors of a plurality of kinds having different capacitances, a DRAM-embedded device in which a plurality of DRAM sections are mounted together with the CMOS logic section on the same chip can achieve both a reduced power consumption and an increased processing speed while ensuring a sufficient signal holding characteristic, by optimizing the capacitance of the memory cell for each DRAM section according to the application thereof.
In the fourth embodiment, the cup-like electrode parts 209A and 403 and the cup-like capacitor lower electrode 209B are formed in the first memory region RA and the second memory region RB, respectively, and only the bottom and inner wall surfaces of the electrode parts 209A and 403 and the capacitor lower electrode 209B are utilized as cell capacitors. Instead, not only the bottom and inner wall surfaces of the cup-like electrode parts 209A and 403 and the cup-like capacitor lower electrode 209B but also the outer wall surfaces thereof may be utilized as cell capacitors (see
Furthermore, although in the fourth embodiment phosphorus-containing silicon is used for the electrode parts 209A and 403, the capacitor lower electrode 209B and the capacitor upper electrode 211, silicon containing any other impurity or any other metal may alternatively be used.
Moreover, although in the fourth embodiment an ON film is used as the capacitor insulating film 210, any other insulating film may alternatively be used.
A semiconductor device according to a modification of the fourth embodiment of the present invention will now be described with reference to the drawings.
A planar configuration of the present modification is similar to that of the fourth embodiment illustrated in
As illustrated in
The above-described modification also provides the effects similar to those of the fourth embodiment.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2004-185045 | Jun 2004 | JP | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5814547 | Chang | Sep 1998 | A |
| 5838603 | Mori et al. | Nov 1998 | A |
| Number | Date | Country |
|---|---|---|
| 11-087636 | Mar 1999 | JP |
| 2000-232076 | Aug 2000 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20050286292 A1 | Dec 2005 | US |
