JPS61212052A - Semiconductor device with beam structure - Google Patents

Abstract

PURPOSE:To obtain a semiconductor device with a beam structure, whose natu ral frequency is set, at low cost and with high precision by a method wherein the beam structure is constituted of a movable beam, and a weight provided on the movable beam, and the weight is made of the same material as that of the electrode of the semiconductor device. CONSTITUTION:The movable beam formed by laminating an upper nitride film 19 and a lower nitride film 17 on the upper and lower sides of a high- precision P<+> type polycrystalline Si film 18 in a sandwiched form is provided on the side of the upper surface of an SiO2 film 13. The end part of this movable beam is connected to the gate region 16 of the P-type MOSFET, wherein P<+> type diffusion layers 12a and 12b are used as the source and drain of the MOSFET, through an electrode wiring (such as an Al wiring) 22, and moreover, a weight 40 made of the same material as that of the electrode wiring 22 is provided on the upper surface of the movable beam. By changing the length of the mask to be used for forming the weight forming pattern, when the natural frequency of a beam structure 100 consisting of the movable beam and the weight is ready-conformed to the knocking vibration, the knocking detection can be performed utilizing a resonance phenomenon.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a semiconductor device having a beam structure that allows the natural frequency of a movable beam to be set with high precision.

(Background of the Invention) As a conventional semiconductor device having a beam structure, there is a device as shown in FIG. 6, which was previously proposed by the present inventor in Japanese Patent Application No. 57-154826.

FIG. 6 shows a cross-sectional view of a semiconductor device having a beam structure manufactured by the process shown in FIG.

A method of manufacturing a semiconductor device having this beam structure will be described with reference to FIGS. 5A (Δ) to 5B (H>).

(A>-... First, the N-type 3i substrate 11 is formed with P-middle regions 12 for source, drain, and fixed electrodes of P-MO3, and the surface of the N-type 3i substrate 11 is thermally oxidized with 5IChI.
1'13 is formed to a thickness of, for example, 100 mm.

(B)... Next, about 62% of S!
The polysilicon layer is thermally decomposed at 0 DEG C. to form an impurity-free polysilicon layer having a thickness of, for example, 1 to 3 .mu.m, and then polysilicon spacers 14 are formed by photoetching.

(C)...Next, apply NH3 (7-Enia) and si H2C1, ? to the entire surface by, for example, the reduced pressure C, V D method.
(Dichlorosilane) is thermally decomposed at about 750°C, and about 50%
A nitride film is formed to a thickness of 0. Then, a part of the nitride film is photo-etched to prevent oxidation of the polysilicon spacer 14.
form.

(D>... The portions of the thermally oxidized S!02 film 13 corresponding to the gate and contact portions of the P-MOS are removed by photoetching, and thermally oxidized in an oxygen atmosphere at, for example, 1050° C. to form the gate oxide film 16. Then, if necessary, apply a threshold voltage vT between the gate and source.
Ion implantation for H control is performed through the gate oxide film 16.

(E)...Next, after removing the anti-oxidation film 15 with hot phosphoric acid (150°C), the lower layer nitride film 1 with a thickness of about 300 to 500 mm is coated on the entire surface by, for example, low pressure CVD.
A polysilicon film with a thickness of about 5,000 to 10,000 wafers is formed on the entire surface by, for example, a low-pressure CVD method, and a high concentration of boron is added to this polysilicon film by an impurity diffusion method using, for example, BBr3. Doped and high concentration P
+Polysilicon film 18 is formed.

Furthermore, an upper nitride film 19 is formed on the entire surface of the high concentration P+ polysilicon film 18, for example, by low pressure CVD, which is several tens of times thicker than the lower nitride film 17.

(F)...Next, a cantilever pattern 20 is formed by plasma etching using CF 4 , and the upper layer nitride film 19 corresponding to the electrode extraction portion 21 is roughly removed by photoetching.

(G)...Next, the contact portion 30 is thermally oxidized S! by photoetching. A hole is made in the 02 membrane 13, and a hole with a thickness of 1 to 1.5 μm is formed on the entire surface by, for example, a vacuum evaporation method. A film is formed, and unnecessary portions are removed by photoetching to form electrode wiring 22.

(1-1>...Next, S!hL and PH3 (phosphine) are thermally decomposed at about 400'C by atmospheric pressure CVD method to form a PSG film with a thickness of 1.2 μm, for example, on the entire surface, and photo Unwanted portions are removed by etching to form protection 1123 except on the bonding pad and movable beam area.

Finally, add a strong alkaline aqueous solution (e.g. ethylenediamine +
Etch the entire surface using a mixture of pyrocatechol and water as an etching solution.

Then, the unlit polysilicon spacer 14 of Poron is etched at a rate of about 50 μm/hour, and the sixth
A semiconductor device having the beam structure shown in the figure is completed.

However, in a semiconductor device having such a conventional beam structure, the natural frequency f of a cantilever beam, which is a beam structure, is determined by its length, t, and density of the beam material. If ρ and Young's modulus are E, then it is determined by A method of forming a beam structure was adopted to achieve this.

For this reason, variations occur particularly in the thickness t1 Young's modulus E of the cantilever beam between manufacturing rods of semiconductor devices, and as a result, the natural frequency of the beam structure tends to vary from the design value between manufacturing rods. A problem has arisen in that a semiconductor device having a beam structure whose natural frequency is set to a high precision (for example, ±2%) cannot be manufactured at a low cost.

On the other hand, in Japanese Patent Application No. 57-148874@, the present applicant provided a weight at the tip of a movable piece whose one end was supported on a semiconductor substrate, and by changing the weight of this weight, the natural frequency (resonance) of the movable piece was This paper proposes a semiconductor device that can change the frequency.

However, in this case, it is not easy to arbitrarily change the weight of the weight, resulting in the same problems as in the conventional example described above.

(Object of the Invention) An object of the present invention is to provide a semiconductor device having a beam structure that is inexpensive and capable of setting the natural frequency of a movable beam with high precision.

(Structure of the Invention) The present invention provides at least one end supported on a semiconductor substrate,
A semiconductor device having a beam structure consisting of a movable beam connected to an electrode and a weight provided on the upper surface of the movable beam; characterized in that the weight is made of the same material as the electrode.

(Description of Examples) Examples of the present invention will be described based on the drawings. FIG. 1 shows the configuration of an embodiment of a semiconductor device having a beam structure according to the present invention, in which an N-type silicon substrate 1
P-type high concentration buried regions 12A, 12B, 1 are formed by introducing P-type impurities into the surface of 1 by a method such as ion implantation.
2C is formed.

Moreover, on top of these, S! is formed by high-temperature oxidation of the surface of the silicon substrate 110. 02 film 13 is grown and formed.

Furthermore, S! (On the upper surface side of the h film 13, there is provided a movable beam formed by stacking an upper nitride film 19 and a lower nitride film 17 in a sandwich manner above and below the high concentration P/polysilicon film 18.

The end of this movable beam is connected to the source of P1 diffusion 1i12a, 12b via electrode wiring (for example A6>22).

A weight 40 connected to the gate region 16 of the P-MOSFET serving as a drain and made of the same material as the electrode wiring 22 is provided on the upper surface of the movable beam. A protective film (for example, a PSG film) 23 is provided on the upper part of the silicon substrate 11 other than on the bonding pad and the movable beam area.
covered by.

Next, the manufacturing process of this semiconductor device is shown in FIGS.
This will be explained with reference to FIG. 2B (f).

(a)--First, the state shown in FIG. 2(a) is obtained through the same steps as in the conventional apparatus shown in FIGS. 5A (A) to 5B (F).

(b)...Next, S! of the contact portion 30 is formed by photo-etching. (H) A hole is made in the film 13, an aluminum film of about 1.5 μm is formed on the entire surface by vacuum evaporation, and unnecessary parts are removed by photoetching to form the electrode wiring 22, and the cantilever pattern 20 is roughly formed. An aluminum pattern 25 for a weight is formed near the tip.

(C) Roughly apply a negative resist to these entire surfaces, expose the necessary parts, and form the resist pattern 31.
form.

(d)...Next, a positive resist is applied to the entire surface and exposed at the first exposure position 32 as shown in FIG.
Further, the mask is shifted and exposure is performed at the second exposure position 33 to form a weight forming pattern 35 so that the center is located at the weighted position 34 of the weight.

At this time, since the exposure positions 32 and 33 can be determined freely, the length of the weight forming pattern 35 shown in FIG.
can be freely changed within the range of O-l.

(e) In this step, a resist pattern 31° and a weight forming pattern 35 are used, and unnecessary portions of aluminum of the weight aluminum pattern 30 are removed by photoetching to form a weight 40.

(f)...Next, the entire surface is coated with S! at about 400°C by, for example, atmospheric pressure CVD. H4 and PH3 are thermally decomposed to form PSG IIi with a thickness of, for example, 1.2 .mu.m, and a protective film 23 is formed by photoetching on areas other than the regions where the bonding pads and beam structures will be formed.

Finally, add a strong alkaline aqueous solution (e.g. ethylenediamine +
Etch the entire surface using a mixture of pyrocatechol and water as an etching solution. Then, the polysilicon spacer 14 to which no boron is added is etched at a rate of about 50 μm/hour, and a semiconductor device having the beam structure 50 shown in FIG. 1 is completed.

In addition, in the process shown in Fig. 2B (d), various lengths of the mask for forming the peg formation pattern 35 in Fig. 3 are changed, and the mask having the optimum length is used. Then, the weight forming pattern 35 can be created with one exposure.

In this case, the resist may be a negative resist.

Next, explain the effect of gold. As shown in FIG. 1, the natural frequency f of a beam structure 100 consisting of a cantilever made mainly of a high concentration P+ polysilicon film 18 and a weight 40 provided at its tip is determined by the width of the cantilever. When b is the length, t is the thickness, E is the Young's modulus, m is the mass of the cantilever, and M is the mass of the weight, it is expressed by the following equation.

This is the mass that is changed to match the natural frequency of the beam structure to the design value.

Next, let the natural frequency when M=A be ``.'', then f
, /fo is as follows. In the above formula (3), B and f when A=0.15
, 7f o is shown in FIG.

- For example, if the width, thickness, and density of the weight 40 are the same as the width, thickness, and density of the cantilever beam, the length of the cantilever beam is 500μ.
m, the length l of the weight 40 is 75 μm, and the change Δl in the length of the weight 40 corresponding to the mass am/200 that should be changed in order to adapt the natural frequency of the beam structure 50 to the design value is 2, It becomes 58 μm.

By changing the length l of the weight 40 by ±75 μm (B = 30) in the process shown in Figure 2B (d), the design can be achieved even if the natural frequency of the beam structure 50 deviates by about ±15% from the design value. value can be approximated.

Also, by changing the mass of the standard weight, the beam structure 50
The design value of the natural frequency itself can be changed.

As an example of the application of the semiconductor device described above, the case of detecting knocking in an automobile engine will be described.

When knocking occurs, a noticeable vibration of about 7 KHz is generated from the engine, but if the natural frequency of the beam structure manufactured by the above method is matched to the knocking imaging, this knocking can be detected using the resonance phenomenon. I can do it.

At this time, since the detection output of knocking is extracted using the resonance phenomenon, if the natural frequency of the beam structure deviates by 10% from the frequency of knocking vibration, the detection output becomes extremely small. It is necessary to match the vibration frequency to the knocking frequency with high precision.

(Effects of the Invention) As explained above, in the present invention, the beam structure is composed of a movable beam and a weight provided on the movable beam, and the weight is made of the same material as the electrode of the semiconductor device. Since I configured it,
Since a weight of arbitrary area can be created at the same time when forming the electrodes of a semiconductor device, the manufacturing conditions for the movable beam part of the semiconductor device can be relaxed, and therefore the semiconductor device has a beam structure that is inexpensive and whose natural frequency is set with high precision. is obtained.

[Brief explanation of drawings]

FIG. 1 is a cross-sectional view showing the structure of an embodiment of a semiconductor device having a beam structure according to the present invention, and FIG. 2A. 2B is a diagram showing the manufacturing process of the semiconductor device shown in FIG. 1, FIG. 3 is a diagram for explaining the process shown in FIG. 5A and 5B are diagrams showing the manufacturing process of a semiconductor device having a conventional beam structure, and FIG. 5B is a cross-sectional view showing the configuration of a semiconductor device having a conventional beam structure manufactured by the manufacturing process shown in FIG. Lower layer natride film 18...high concentration P÷polysilicon film 19...upper layer T1-ride film 22...electrode wiring 23...protective film 40...weight 100...beam structure

Claims (1)

[Claims] (1) In a semiconductor device having a beam structure consisting of a movable beam having at least one end supported on a semiconductor substrate and connected to an electrode, and a weight provided on the upper surface of the movable beam; the weight being the same as the electrode. A semiconductor device having a beam structure formed of a material.