JPH0132659B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a structure of a capacitor used as a component of a semiconductor device, particularly a MOS type semiconductor device.

Conventionally, when providing a capacitor element together with a MOS transistor on a semiconductor substrate, for example, as shown in FIG.
After forming a conductive layer 2 made of a (P-well) diffusion layer on the surface layer of a shaped silicon substrate 1, a conductive layer 4 made of a polycrystalline silicon film is formed via an insulating layer 3, and then After covering the entire surface with an insulator layer 5, a contact window is provided on the conductor layer 2 and the conductor layer 4, and a pair of electrodes 6 and 7 are attached to the conductor layer 2 and the conductor layer 4 through this window. was. At this time, as shown in the plan view of FIG. 1B, the conductive layer 2 and the conductive layer 4 were connected to the electrodes 6 and 7 at the electrode extraction portions 8 and 9, respectively, to obtain a potential. FIG. 1A corresponds to the cross section taken along the line -' in FIG.

However, the capacitance value of the capacitor is proportional to the overlapping area of the conductor layer 4 and the conductor layer 2,
Since it is inversely proportional to the thickness of the insulating layer 3, in such a configuration, the area of the overlapping portion 10 (portion shown by cross hatching) changes every time misalignment occurs between the conductive layer 4 and the conductive layer 2. The capacitance value will change. As an example, the dimension x ₁ in the X direction of the rectangular part of the conductive layer 4 shown in Figure 1b is 30 μm, and the dimension in the Y direction
When y ₁ is 30 μm, the dimension in the X direction of the overlapping portion 10 between the protrusion for the potential extraction portion of the conductor layer 4 and the conductor layer 2 is x ₂ 8 μm, and the dimension y ₂ in the Y direction is 10 μm, the mask Due to misalignment of the glass dry plate and mask, the relative position of the conductor layer 4 to the conductor layer 2 may be
If there is a deviation of ±1 μm in the direction, the area of the overlapping portion 10 changes by 1.02%, and therefore the capacitance value as a capacitor also changes by 1.02%.

Next, we will discuss a switched capacitor filter (hereinafter abbreviated as SCF), whose circuit is constructed using a MOSFET and a built-in capacitor.

In the SCF, an equivalent RC active filter is constructed using a resistor that is equivalently simulated with a small capacitance that is switched at high speed.

An example of an integrator, which is the basic configuration of an SCF, is shown in Figure 2a. In FIG. 2a, the potential of node 112 is
V _io , the capacitance value of the capacitor 114 is C ₁ , and the switching period between nodes 112 and 113 of the switch 111 is 1/ _c , then on average, i=C ₁ between the nodes 112 and 113.・A current of V _io・_c flows. Therefore, between node 112 (V _io ) and node 113 (input of operational amplifier 17), it is equivalent to having a resistance of R _eq =V _io /i=1/C ₁ · _c , and the clock When the frequency _c is sufficiently higher than the passing frequency, the circuit of FIG. 2a is equivalent to the circuit of FIG. 2b, where the resistance value of the resistor 118 is Req.

In FIG. 2a, when the capacitance value of the capacitor 115 is C ₂ and the potential of the node 116 is V _OUT , the transfer function is V _OUT /V _io = _c /S·C ₁ /C ₂ .
This switched-capacitor integrator in Figure 2a has a gain equal to _c ·(C ₁ /C ₂ ),
The factor determines the parameters such as bandwidth or peaking frequency in the active filter circuit. So, SCF
When the switching frequency _c of the switch is constant, the important characteristics of the circuit are determined only by the capacitance ratio of capacitors C ₁ and C ₂ , and this gives the SCF its greatest advantage of high accuracy and stability. It had become a sexual thing.

However, the fluctuation in the capacitance values of capacitors C ₁ and C ₂ , which is a factor that determines the characteristics of this filter, is
This not only impairs the superiority of SCF, but also changes the characteristics of SCF itself. As an example,
In a band pass filter using an integrator with the configuration shown in Figure 2a, the center frequency _p of the passband is
_p = _c / 2π・C ₁ /C ₂ , but if the capacitance value of either capacitor C ₁ or C ₂ , which is a constituent element of the integrator, changes by, for example, 1%, the band pass The center frequency _p of the filter will also vary by 1%.

Therefore, when capacitors with the configurations shown in Figure 1 a and b are used as constituent elements of the SCF,
For example, in band pass filters, problems occur when the center frequency deviates by, say, 1% from the design value, and SCFs continue to have problems such as decreased reliability and yield, as well as increased costs. It was not possible to use such a capacitor.

The present invention aims to eliminate these drawbacks and provide a high-precision capacitor, for example, to construct a high-precision SCF.

Embodiments of the present invention will be described in detail below with reference to the drawings. FIGS. 3a and 3b show an embodiment to which the present invention is applied. In FIG. 3, the same parts as in FIG. 1 are given the same numbers. 2 and 4 are conductive layers, and a capacitor is formed at the overlapping portion of these two conductive layers. 8 and 9 are potential extraction parts of the conductive layers 2 and 4, respectively, and 19 is a protrusion of the electrode 4 provided by applying the present invention, which is formed on the opposite side to the protrusion where the extraction part 9 is provided. In other words, the protrusion and the width are formed at opposing positions with the same dimension. Therefore, according to this configuration, the overlapping portion 1
It has an overlapping part 20 of 19 and 2 opposite to 0.

In a capacitor with this structure, if the relative positions of 2 and 4 shift in the X direction due to a shift in the mask glass dry plate or mask alignment, for example, 10 and 20
Although the area changes, the amount of change has the same absolute value and different polarity, so the total amount of change in the sum of areas of 10 and 20 is zero, and the sum of areas is always constant. Therefore, the capacitance value is also constant.

Furthermore, in the configuration of the present invention, the capacitance value remains unchanged not only due to deviations in the X and Y directions but also due to rotational deviations between the two conductive layers. Therefore, by creating a structure in which a protruding portion having the same width as the potential extraction portion is provided at an opposing position, it is possible to form a constant capacitance according to the designed value, which is not affected by the misalignment of the mask-aligned dry plate or mask alignment. .

FIG. 4 shows another embodiment of the present invention, in which a capacitor with a capacitance value n times that of the unit capacitor is constructed by regularly arranging unit capacitors of the same area and connecting them in parallel. The protrusion 19x (x=a to d) having the same width as the protrusion in which the take-out part 9x (x=a to d) is formed,
Each is provided at a position facing 9x.

The capacity of such a configuration of the present invention is, for example,
When applied to SCF, it is possible to maintain the characteristics as designed.
It is possible to realize SCF, and it is possible to stably provide high-precision, highly reliable SCF at low cost without being affected by misalignment of mask-aligned dry plates or mask alignment. Here, in the embodiment, a capacitor element in which the conductor layer 2 is formed of a P-well diffusion layer formed in an n-type semiconductor substrate or the conductor layer 4 is formed of polycrystalline silicon has been described with reference to the drawings. The electrode may be formed on an insulating film on a semiconductor substrate. In other words, the structure of the electrode is not limited to a P-well diffusion layer and a polycrystalline silicon film layer, but also a polycrystalline silicon film layer and an aluminum film layer, or a first polycrystalline silicon film layer and a second polycrystalline silicon film layer.
Two conductive layers, such as polycrystalline silicon film layers,
Any structure may be used as long as it has a conductive layer-insulating layer-conductive layer structure. In any of these configurations, the effects of the present invention can be achieved by forming a protruding portion having the same width opposite the potential extraction portion.

As explained above, by adopting the configuration shown in the present invention, a constant capacitance as designed can be formed in a semiconductor device without being affected by mechanical misalignment of glass mask dry plates, mask alignment, etc. . Also,
When considering the etching effect when forming the electrode extraction portion and the protrusion, it is clear that a constant capacitance can be formed best when the shape, area, and peripheral length of the electrode extraction portion and the protrusion portion are made the same. If the constant capacitance configured according to the present invention is used in a semiconductor integrated circuit that requires a highly accurate and highly stable capacitance, such as a switched capacitor filter, the characteristics of the SCF will always be stable as designed values. As a result, the accuracy, reliability, and yield of SCF can be improved.
It has the advantage of being able to provide SCF at low cost, and has great industrial value.

[Brief explanation of drawings]

Figures 1a and b are plane and cross-sectional views of conventional capacitive elements, a is a cross-sectional view taken along the line -' of b, Figure 2a is a block diagram of an integrator using a switched capacitor,
Figure b is a block diagram of an RC integrator, Figures 3a and b are schematic plan views of a capacitive element according to one embodiment of the present invention, and Figure 4 is a schematic plan view of another embodiment of a capacitive element according to the present invention. be. 1...Silicon substrate, 2...Conductor layer (P well diffusion layer), 3...Insulator layer (silicon dioxide film), 4...Conductor layer (polycrystalline silicon film), 5...
... Insulator layer (silicon dioxide film), 6, 7 ... Aluminum electrode, 8 ... Potential extraction part of 2, 9
. . . potential extraction _portion of 4, 10, 20 _. . . overlapping portion, 20 . . . protruding portion, 114 .

Claims (1)

[Claims] 1 a first conductor layer formed on a semiconductor substrate surface;
It is composed of this first conductor layer and a second conductor layer installed through a dielectric layer, and the second conductor layer has a smaller area than the first conductor layer. a first rectangular protrusion formed on one side of the main body portion and extending outward beyond the first conductor layer; and a first rectangular protrusion formed on the opposite side of the first rectangular protrusion. a second rectangular protrusion extending outward from the first conductor layer, and the first rectangular protrusion extends beyond the first conductor layer.
A semiconductor device, characterized in that the first and second rectangular protrusions include a capacitive element connected to an electrode on the outside of a conductor layer, and the first and second rectangular protrusions have the same width.