JP6512985B2 - Silicon substrate processing method - Google Patents

Description

  The present invention relates to a method of processing a silicon substrate in which a hole (referred to as a through hole) penetrating a substrate surface of a silicon substrate is formed, and a liquid discharge head using the silicon substrate obtained by using the processing method.

  By forming through holes in a silicon substrate, many MEMS (Micro Electro Mechanical Systems) devices are manufactured. One example is a liquid discharge head that discharges a liquid.

  Examples of the liquid discharge head include an ink jet recording head. In the ink jet recording head, an energy generating element for giving energy for discharging ink is formed on the substrate surface. Further, a discharge port forming member is formed on the surface of the substrate, and an opening (discharge port) for discharging the ink is formed on the upper surface of the energy generating element. Furthermore, a through hole is formed in the substrate, and the ink is supplied from the back side to the front side of the substrate through the through hole.

  In recent years, in order to arrange the discharge port and the energy generating element at high density, a liquid supply path consisting of two stages of through holes has been used. In the two-step through hole, a single groove (back hole: common liquid chamber) is formed halfway of the substrate from the back surface of the substrate, and a plurality of small holes (surface holes: individual supply) on the surface of the substrate A through hole is formed, and the front surface hole and the back surface hole are in communication in the silicon substrate.

  In an inkjet recording head having a two-step through hole, a wiring (typically formed by photolithography) for electrically connecting to an energy generating element on the substrate surface is laid between a plurality of through holes. Thus, it is possible to lead to an electrode pad at the end of the chip or a circuit area. Therefore, it is possible to arrange more energy generating elements and discharge ports.

  As a method of forming such a two-step through hole, in Patent Document 1 and Patent Document 2, the surface hole portion and the back surface hole portion are individually formed using dry etching from the front surface side and the back surface side of the substrate. There is disclosed a method of communicating inside.

JP 2004-237734 A U.S. Pat. No. 7,334,875

  FIG. 1 shows a cross-sectional view of a liquid discharge head manufactured using the manufacturing method described in Patent Document 1 and Patent Document 2. As shown in FIG. In the silicon substrate 101, a groove is etched as a back surface hole portion 106 from the back surface side, and a plurality of small holes are etched as a front surface hole portion 105 from the front surface side to form a two-step through hole. . This method is advantageous in that through holes can be formed with high accuracy by dry etching. However, when a large number of chips are formed on a wafer, the etching rate distribution of dry etching becomes large as the wafer size increases, and the control of the etching depth of each hole becomes a problem.

The etching rate distribution α is defined by the following equation, where dmin is the depth of the slowest portion of the etched region, and dmax is the depth of the fastest etching rate.
α = (dmax−dmin) / {(dmax + dmin) / 2} (Expression 1)

  Here, the etching depth is defined by (dmax + dmin) / 2, and the etching depth distribution is defined by (dmax−dmin). The etching rate distribution α is determined by the rate distribution over the entire silicon substrate and the rate distribution inside a single etching hole, and can be regarded as constant during etching. Further, according to Equation 1, the etching depth distribution is a value obtained by multiplying the etching depth by the etching rate distribution α, and the etching depth distribution becomes larger as the etching depth becomes deeper.

  As shown in FIG. 1, when the depth of the groove to be the back surface hole portion 106 varies, the depth of the front surface hole portion 105 also varies. In addition, when the etching depth distribution (dmax−dmin) of the groove becomes too large, a part of the groove penetrates to the surface side, which causes a problem that two-step through holes can not be formed.

  As another viewpoint, in the case of densifying the discharge ports 104, it is necessary to reduce the cross-sectional area of the surface hole 105. When the opening cross-sectional area of the surface hole portion 105 decreases, the flow resistance to which the ink is subjected increases, and therefore, the surface hole portion 105 needs to be shallow in order to reduce the flow resistance. For this purpose, it is necessary to make the groove to be the back surface hole 106 deeper. However, as described above, since the etching depth distribution increases in proportion to the etching depth, there arises a problem that two-step through holes can not be formed by penetrating the silicon substrate 101 at a part of the grooves. Can not do it. As a result, it is difficult to reduce the opening cross-sectional area of the surface hole portion 105 and to increase the density of the energy generating element 107 and the discharge port 104.

  The present invention has been made to solve the above-mentioned problems that occur when forming through holes having steps in a silicon substrate, and to provide a method for reducing the depth distribution of surface holes. To aim.

One embodiment of the present invention is a method of processing a silicon substrate in which a through hole is formed in a silicon substrate,
(A) preparing a set of a first silicon substrate, a second silicon substrate, and a stopper layer provided between the first silicon substrate and the second silicon substrate;
(B) In the second silicon substrate, the stopper layer is exposed by the first etching for etching the second silicon substrate from the surface opposite to the interface between the stopper layer and the second silicon substrate. Forming the first hole portion;
(C) removing at least a part of the exposed stopper layer to expose the first silicon substrate;
(D) forming a second hole by performing a second etching to etch the exposed first silicon substrate, and stopping the second etching in the first silicon substrate;
(E) In the first silicon substrate, a third etching for etching the first silicon substrate is performed from the surface opposite to the surface on the side of the second hole, in the first silicon substrate Forming a third hole by stopping the third etching;
Including
After performing either the second etching or the third etching, the second hole and the third hole are communicated with each other to form the through hole.

  As a result of the improvement in the depth distribution of the surface holes, it becomes possible to form through holes having steps in the silicon substrate with good yield, and the productivity is improved. Further, in the liquid discharge head, it is possible to form the surface holes shallowly by reducing the depth distribution of the surface holes, and to make the cross-sectional area of the surface holes finer, and It is possible to arrange the outlets more densely.

It is sectional drawing of the liquid discharge head produced using the conventional method. FIG. 7 is a cross-sectional view showing an example of a method of manufacturing a liquid discharge head according to the first embodiment of the present invention. It is a figure which shows the relationship between the thin process thickness of a 1st silicon substrate, and a breaking load. It is a figure which shows the relationship between the depth of a surface hole, and depth distribution. FIG. 7 is a cross-sectional view showing an example of a method of manufacturing a liquid discharge head according to the first embodiment of the present invention. FIG. 14 is a cross-sectional view showing an example of a method of manufacturing a liquid discharge head according to a second embodiment of the present invention. It is a figure which shows the relationship of the aperture ratio and etching rate distribution in etching. It is sectional drawing which shows an example of the silicon substrate processed by the 2nd Embodiment of this invention. It is (A) top view and (B) A-A 'sectional drawing which show an example of the silicon substrate processed by the 2nd Embodiment of this invention. FIG. 14 is a cross-sectional view showing an example of a method of manufacturing a liquid discharge head according to a second embodiment of the present invention.

  The method of processing a silicon substrate according to the present invention is applicable to the manufacture of a micromachine such as an acceleration sensor, in addition to the method of manufacturing a substrate for a liquid discharge head. Hereinafter, the present invention will be described in detail while describing a method of manufacturing a liquid discharge head substrate according to an embodiment of the present invention.

First Embodiment
FIG. 2 is a cross-sectional view showing an example of a method of manufacturing a liquid discharge head according to the first embodiment of the present invention. Hereinafter, with reference to FIG. 2, the formation method of the two-step through-hole based on this embodiment is demonstrated.

  First, a first silicon substrate 131 having an energy generating element (heater) 107 formed on its surface is prepared (FIG. 2A). On the surface (first surface) of the first silicon substrate 131, a surface membrane layer 103 composed of wiring, an interlayer insulating film, and the like is formed. Next, the second silicon substrate 132 is prepared, and the first silicon substrate 131 and the second silicon substrate 132 are bonded via the stopper layer 111. At this time, it is preferable to process (thin) the first silicon substrate 131 before bonding the first silicon substrate 131 and the second silicon substrate 132. Further, for the purpose of protecting the substrate surface in the thin processing process and supporting the thinned wafer, it is preferable to bond the support member 110 to the surface of the first silicon substrate 131 using an adhesive (FIG. B)).

  The support member 110 may, for example, be a rigid substrate made of an inorganic material. Specifically, a silicon substrate or a glass substrate such as quartz may be mentioned. The thickness of the rigid substrate is preferably 300 to 730 μm. Examples of the adhesive for bonding the support member 110 to the silicon substrate include a thermoplastic resin which is softened by heat to change its adhesive strength, and an ultraviolet curable resin whose adhesive strength is reduced by irradiation of ultraviolet light. The adhesive may have a thickness sufficient to fill the irregularities on the surface of the first silicon substrate 131, and is preferably 2 to 100 μm.

  In the case where a silicon substrate or a glass substrate is used as the support member 110, there is a problem that the cost of consumable members such as a substrate is high and the manufacturing cost is long due to the long and complicated manufacturing process. As a means for solving the problem, there is a method of using an inexpensive resin material for the support member 110. Examples of the support member 110 made of a resin material include a tape with an adhesive layer. Specifically, a thermal peeling tape in which the adhesive force is lowered by heat, and an ultraviolet ray curable tape in which the adhesive force of the adhesive is lowered by irradiating the ultraviolet ray can be mentioned. The thickness of the tape is preferably 20 to 700 μm.

  After bonding the support member 110 to the first silicon substrate 131 as necessary, thin processing is performed from the side opposite to the side on which the support member 110 is provided. Examples of the method of thin processing include mechanical processing using a grinding apparatus and chemical processing in which silicon is wet etched using a mixed solution of hydrofluoric acid and nitric acid. Here, when the first silicon substrate 131 is thin-processed, it is necessary to make the thickness sufficient to maintain sufficient mechanical strength after the thin-processing. This is because it is necessary to carry out many steps such as polishing, cleaning, bonding, etc. where mechanical load is applied to the thinned substrate after thinning, and if the mechanical strength of the substrate is insufficient, the substrate will crack, This is because the above steps can not be performed, and the yield is reduced. In particular, when a resin is used as the support member 110, the rigidity of the support member 110 is low, so that the thinned wafer is likely to be mechanically stressed.

  FIG. 3 is an experimental result showing the relationship between the thin processing thickness of the silicon substrate and the breaking load. In this experiment, thin processing was performed by a grinding apparatus using a silicon substrate in which an ultraviolet curing tape (support member) having a thickness of 120 μm was bonded to a substrate surface opposite to the grinding surface. A load was measured at the same time by pressing one point of the silicon substrate surface (grind surface) with a pressure jig whose contact portion with the substrate was a circular shape having a diameter of 1 mm. The load was increased, and the load when the substrate was broken was defined as the breaking load. In addition, it evaluated by making the back surface of a substrate float at the time of pressurization. The breaking load can be regarded as the mechanical strength of the substrate against pressure from the top surface of the substrate, and the larger the breaking load, the higher the mechanical strength of the substrate.

  As shown in FIG. 3, the breaking load of the silicon substrate (thickness: 730 μm) before thin processing is 95.1 N (9.7 kgf). As the thin processed thickness of the silicon substrate becomes thinner, the breaking load becomes smaller, and the breaking load at a thin processed thickness of 150 μm becomes 10.8 N (1.1 kgf). In addition, when the thin processing thickness is thinner than 150 μm, the breaking load decreases rapidly, and at a thin processing thickness of 50 μm, the breaking load decreases to 1.37 N (0.14 kgf).

  When the thin processing thickness of the silicon substrate is smaller than 150 μm, the breaking load on the silicon substrate before thin processing becomes 1/10 or less. When the breaking load, that is, the mechanical strength of the substrate is lowered by one digit, a crack or the like due to the insufficient strength may occur.

  In fact, also in the experiments of the inventors, when the silicon substrate thickness is processed to be thinner than 150 μm using the same tape as the experiment shown in FIG. 3 as the support member 110, polishing, cleaning, bonding, etc. after thin processing It has been confirmed that wafer cracking rapidly increases in each process of Furthermore, as the mechanical strength decreases, the substrate itself may be bent by its own weight, which may make it difficult to transport the apparatus. Therefore, when using resin for the support member, it is preferable to set the thin processing thickness of the first silicon substrate 131 to 150 μm or more.

  When a rigid substrate made of an inorganic material is used as the support member 110, mechanical strength is maintained to some extent because the silicon substrate is firmly supported by the rigid substrate even if the silicon substrate is processed to be thinner than 150 μm. . Even in such a case, however, the breaking load is significantly reduced when the thickness of the silicon substrate becomes very thin. As a result, as in the case of using a resin as the support member, the mechanical strength can not be maintained, and the substrate may be broken in the process after the thin processing. In the experiments of the inventors, when quartz glass which is an inorganic material is used as the support member 110, when the thin processed thickness of the silicon substrate becomes thinner than 30 μm, the wafer edge portion having relatively weak strength is broken. The problem was seen. Therefore, when using an inorganic material for the support member 110, it is preferable to set the thin processing thickness of a silicon substrate to 30 micrometers or more.

  In addition, when the first silicon substrate is thin-processed, it is preferable to reduce the thickness distribution (maximum thickness-minimum thickness) of the silicon substrate, specifically, to about several μm. Is preferred. For example, in the case of performing thin processing using a grinding apparatus, it is preferable because the substrate thickness distribution can be reduced by improving the parallelism of each member of the apparatus regardless of the thickness of scraping the substrate.

  After thinning the first silicon substrate 131, the back surface of the substrate is smoothed. The reason for smoothing is to bond to the second silicon substrate 132 in the next step. As a method of smoothing, CMP (Chemical Mechanical Polishing) or a method of dissolving silicon by the wet etching may be mentioned.

  Next, the first silicon substrate 131 and the second silicon substrate 132 are bonded via the stopper layer 111 (FIG. 2C). The thickness of the second silicon substrate 132 may be set in consideration of the substrate thickness after bonding to the first silicon substrate, and is, for example, 200 to 700 μm. The stopper layer 111 is formed before bonding to the bonding surface of either the first silicon substrate 131 or the second silicon substrate 132. The stopper layer 111 is a layer responsible for stopping the etching to be performed later.

  Examples of the material of the stopper layer 111 include an organic film. The organic film also has a function as an adhesive, and is preferable because the first silicon substrate and the second silicon substrate can be bonded as they are after film formation. Examples of the material of the organic film include acrylic resins, polyimide resins, silicon resins, fluorine resins, epoxy resins, and polyether amide resins.

  Specific examples of epoxy resins include SU-8 (trade name) manufactured by Kayaku Microchem. Examples of polyether amide resins include HIMAL (trade name) manufactured by Hitachi Chemical Co., Ltd., BCB (Benzocyclobutene), HSQ (Hydrogen Silsesquioxane), and the like. These materials are preferable because they can be bonded at a low temperature of 300 ° C. or less and, for example, they can be bonded without breaking the wiring layer present on the surface of the first silicon substrate 131.

  Examples of film formation means of the organic film (stopper layer) include spin coating and slit coating. The organic film is formed on the bonding surface of either the first silicon substrate 131 or the second silicon substrate 132. After the organic film is formed on the bonding surface, baking is performed, and the first silicon substrate 131 and the second silicon substrate 132 are bonded by means such as thermocompression bonding. Further, the bonding is completed by heat treatment to cure the organic film.

  As a material of the stopper layer 111, an inorganic film may be used. Specifically, a ceramic such as a silicon oxide film, a silicon nitride film, or a silicon carbide film, or a metal film such as tantalum, gold, nickel or the like can be mentioned. As a method of forming the inorganic film, a sputtering method, a CVD (Chemical Vapor Deposition) method, a vapor deposition method, a plating method and the like can be mentioned. When these methods are used, it is preferable because a film can be formed at a relatively low temperature of about 100 to 300 ° C.

  Alternatively, as the inorganic film, a thermal oxide film formed on the front and back surfaces of the wafer may be used by heating the second silicon substrate 132 to a high temperature (800 to 1100 ° C.) in an oxidizing atmosphere. The thermal oxide film is preferable because the surface is smooth and the film is dense, so it is excellent in bonding property and chemically stable.

  The first silicon substrate 131 and the second silicon substrate 132 are bonded via the inorganic film (stopper layer) formed by the above means. As a bonding method via an inorganic film, a bonding method which can be carried out at low temperature and enables strong bonding is suitable. Specifically, plasma activated bonding may be mentioned, in which the bonding surface is exposed to plasma to form a hydroxy group, and a detachment reaction of water from the hydroxy group is caused to form a covalent bond. In addition, it is also preferable to perform normal temperature bonding in which the surface is etched by argon plasma or the like in a high vacuum to expose atomic bond hands for bonding.

  When bonding is performed by plasma activation bonding or normal temperature bonding, it is necessary to sufficiently smooth the bonding surface (typically, the surface roughness is 1 nm or less) before bonding. In addition, it is necessary to remove the foreign matter by cleaning so that the foreign matter that causes the void does not remain on the bonding surface.

  The stopper layer 111 may have a thickness sufficient to stop the etching, and is designed in consideration of the etching selectivity to silicon. Approximately 0.01 to 10 μm is preferred.

  In addition to bonding the first silicon substrate 131 and the second silicon substrate 132 via the stopper layer 111, an SOI substrate in which an oxide film is embedded in advance in the silicon substrate may be used. In that case, the oxide film functions as the stopper layer 111. Then, the energy generating element 107 and the circuit are manufactured on the surface of the SOI substrate.

  After bonding the first silicon substrate 131 and the second silicon substrate 132, the support member 110 is removed. Next, a surface protection member 117 for protecting the surface of the first silicon substrate 131 is attached during etching. The surface protective member 117 may, for example, be a tape or a glass substrate. Although it is not necessary to use the surface protection member 117, it is preferable to etch after bonding the surface protection member 117 from the viewpoint of protecting the surface of the first silicon substrate. Also, the support member 110 can be used as it is as the surface protection member 117 without peeling it off.

After the surface protection member 117 is bonded, a back surface etching mask 116 is formed on the side (back side) of the second silicon substrate 132 opposite to the interface with the stopper layer 111. Then, the second silicon substrate 132 is etched from the back surface side provided with the back surface etching mask to form a first hole (first supply port 112) (FIG. 2D). As an etching method, dry etching is preferable. In particular, reactive ion etching is preferable, and among them, the Bosch process excellent in deep etching of silicon is preferable. The Bosch process is a method in which silicon is anisotropically etched by alternately repeating the formation of a carbon-based deposited film and the etching using SF ₆ gas or the like.

  The second silicon substrate 132 is etched to form the first supply port 112, and when the etching reaches the stopper layer 111, the etching is stopped. By stopping etching once at the stopper layer 111, the etching depth distribution is made uniform. Usually, the first supply port 112 needs to be etched to the deepest, but eventually it is possible to stop the etching at the stopper layer 111, so it does not contribute to the depth distribution. Since the present invention has such an advantage, it is possible to improve productivity by using etching conditions with a large etching rate, for example, at the expense of the etching rate distribution.

  When the formation of the first supply port 112 is completed at all places in the wafer, the etching is stopped. At this point, the stopper layer 111 is exposed at the bottom of all the first supply ports 112. Then, by removing at least a part of the exposed stopper layer 111, the first silicon substrate 131 is exposed. In the present embodiment, all of the stopper layer 111 exposed at the bottom of the first supply port 112 is removed. The stopper layer 111 can be removed by means such as dry etching or wet etching.

  After the first silicon substrate 131 is exposed, etching (second etching) is performed again from the same side to form a second hole (second supply port 113) (FIG. 2E). As an etching method, dry etching is preferable. In particular, reactive ion etching is preferable, and among them, the Bosch process excellent in deep etching of silicon is preferable. The opening cross-sectional area of the second supply port 113 is the same as the opening cross-sectional area of the first supply port 112 because all the stopper layers 111 exposed at the bottom of the first supply port 112 are removed. The opening cross-sectional area is the opening cross-sectional area of the opening in the plane parallel to the substrate surface of the silicon substrate. Since the second supply port 113 is required to have depth accuracy, it is preferable to perform etching under the condition that the etching rate is small and the etching rate distribution is small. After the second supply port 113 is formed to a desired depth, the etching is stopped. Thereafter, the deposit inside the formed groove is removed using means such as oxygen ashing or a chemical solution, and the back surface etching mask 116 is further removed with a peeling solution or the like, and then the back surface side is cleaned with a chemical solution.

  Next, the back surface protection member 120 for protecting the back surface is attached to the back surface of the second silicon substrate 132, and the surface protection member 117 is removed (FIG. 2F). Although the back surface protection member 120 may not be provided, the back surface protection member 120 is used to protect the back surface side during etching, like the surface protection member 117, and the back surface protection member is provided. Is preferred.

  Subsequently, an etching mask is formed on the side (surface) of the first silicon substrate 131 opposite to the surface on the side of the second supply port 113. Then, the first silicon substrate 131 is etched (third etching) from the surface side on which the etching mask is formed, thereby forming a third hole (third supply port 114). As an etching method, dry etching is preferable. Among them, the Bosch process which is superior in deep etching of silicon is preferable. The third supply port 114 is composed of a plurality of minute holes, and the opening cross-sectional area is determined in consideration of the supply port depth. When the third supply port 114 and the second supply port 113 communicate with each other, the etching is stopped. Thus, the silicon substrate having the two-step through holes is obtained.

  After stopping the etching, the deposit inside the third supply port 114 is removed, and the etching mask on the surface is removed. Next, the back surface protection member 120 is peeled off, and the front and back surfaces of the silicon substrate and the inner wall of the supply port are cleaned. Thereafter, the discharge port forming member 102 is formed on the silicon substrate in which the two-step through holes are formed. First, the wall 118 of the discharge port forming member is formed. A dry film resist in which a photosensitive resin is coated on a film substrate is bonded onto a silicon substrate, and then exposed and developed to pattern the wall 118 of the discharge port forming member. Similarly, the top plate 119 of the discharge port forming member is formed by bonding a dry film resist, performing exposure and development, and patterning. Thus, the liquid discharge head shown in FIG. 2 (G) is completed.

  FIG. 4 is a view showing the depth of the surface hole (surface hole depth) and the distribution thereof. Here, the surface hole depth on the horizontal axis of the graph is a design value (design surface hole depth), and FIG. 4 shows that the surface hole depth becomes a design value based on the etching rate distribution α. The surface pore depth distribution at the time of etching is shown. The maximum value and the minimum value of the surface hole depth are the maximum value and the minimum value of the depth centered on the design surface hole depth. Further, a value obtained by dividing the difference between the maximum value and the minimum value of the surface hole depth by the design surface hole depth by a percentage is shown as the surface hole depth distribution on the vertical axis of the graph.

  In FIG. 4, a silicon substrate (comparative example, sample 1) in which through holes are formed by the conventional method shown in FIG. 1 and a silicon substrate (samples 2 to 4) in which through holes are formed by the method of this embodiment Compared. In each case, the etching rate distribution α from the back surface was 40%, and in the comparative example, the thickness of the substrate was 730 μm, and etching was performed until the design surface hole depth was reached. Since the etching distribution of the second silicon substrate 132 and the stopper layer 111 can be neglected with respect to the substrate of the present invention, the etching was evaluated by the etching to the design surface hole depth with respect to the first silicon substrate 131. In addition, since the thin processing thickness of the first silicon substrate 131 is 300 μm (sample 4), 150 μm (sample 3), and 60 μm (sample 2), the designed surface hole depth larger than the thin processing thickness is taken into consideration. Absent.

  It can be seen from FIG. 4 that the surface hole depth distribution of the through holes formed using the method of the present invention is extremely small compared to the conventional example. For example, in sample 1 of the comparative example, when the design surface hole depth is 150 μm, the surface hole depth is widely distributed from 34 μm to 266 μm, and as a result, the surface hole depth distribution is 155%. On the other hand, when the design surface pore depth is 150 μm in sample 4 of the present invention (substrate thickness 300 μm), the surface pore depth is distributed from 120 μm to 180 μm and the surface pore depth distribution is 40% It's getting lower.

  Further, in the sample 1 of the comparative example, when the design surface hole depth is set to 120 μm or less, the depth (minimum value of the depth) of the shallowest part of the surface hole becomes zero. That is, since the silicon substrate is penetrated by the etching from the back surface, such surface holes can not be formed. On the other hand, in the sample of the present invention, the design surface hole depth is 50 μm for sample 4, 25 μm for sample 3, and 10 μm for sample 2 when part of the silicon substrate penetrates by etching from the back surface. is there. In practice, the depth of the surface hole (third supply port) can be made shallower by etching the second supply port 113 under the condition that the etching rate is small and the etching rate distribution is small. it can. Thus, by using the method of the present invention, it is possible to make the surface hole depth very shallow. As a result, the flow resistance of the through hole can be significantly reduced, which is advantageous in miniaturizing the surface hole portion (third supply port) and densifying the discharge port.

  As described above, if it is possible to suppress the influence of the reduction in substrate strength due to the thin processing of the first silicon substrate 131, the first silicon substrate 131 is made thinner to reduce the surface hole depth distribution. And the surface hole depth can be made shallow.

  Further, in the present embodiment, an example of dry etching is described as the etching method of the first supply port 112, but a method other than dry etching capable of deep etching of silicon, for example, anisotropic wet etching using a strong alkali It is also possible to use means such as etching by sand blasting and etching by laser processing. Even in the case of using such an etching method, it is necessary to form the first supply port 112 and select the material of the stopper layer 111 so that the etching can be stopped by the stopper layer 111. By stopping the etching once, it is possible to solve the depth distribution of the first supply port 112 in proportion to the etching depth as in the case of the dry etching.

  The first supply port 112 (first hole) and the second supply port 113 (second hole) are, as shown in FIG. 2, a plurality of third supply ports 114 (third Need not be in the form of a single groove with a large open cross-sectional area in communication with the For example, as shown in FIG. 5, the first supply port and the second supply port are composed of a plurality of individual holes, and each of the holes independently communicates with one third supply port. It may be a configuration.

  In the method of manufacturing the liquid discharge head shown in FIG. 5, as shown in FIG. 5A, the first supply port 112 is formed as a plurality of holes, and the etching is stopped at the stopper layer 111. Next, the second supply port 113 is formed in the first silicon substrate 131 with the same opening shape as the first supply port 112, and the etching is stopped at a predetermined depth (FIG. 5B). Next, the third supply port 114 is independently communicated with the second supply port 113 to form a two-stage through hole (FIG. 5C). Finally, by forming the wall 118 of the discharge port forming member and the top plate 119 (discharge port forming member) of the discharge port forming member, the liquid discharge head is completed (FIG. 5D).

  Further, in the present embodiment, although the example in which the third supply port 114 is formed after the second supply port 113 is formed, the stopper layer 111 is removed after the first supply port 112 is formed. Instead, the third supply port 114 may be formed and finally the second supply port 113 may be formed. Therefore, after performing either the second etching or the third etching, the second supply port and the third supply port communicate with each other to form a two-step through hole.

Second Embodiment
As shown in FIG. 6, the silicon substrate obtained by the present embodiment has a second supply port (second hole) 113 more than the opening cross-sectional area of the first supply port (first hole) 112. It is characterized in that the opening cross-sectional area of is small. Hereinafter, with reference to FIG. 6, the formation method of the through-hole which concerns on this embodiment is demonstrated.

  6A and 6B, the step of bonding the support member 110 to the surface of the first silicon substrate 131 and thinning the first silicon substrate 131 (FIGS. 6A and 6B) corresponds to the first embodiment. It is similar. The present embodiment is characterized in that the stopper layer 111 is patterned in the step of forming the stopper layer 111 shown in FIG. 6 (C). That is, the patterning portion 121 is formed on the stopper layer 111 by half etching so that the stopper layer 111 becomes an etching mask for forming the second supply port 113. The material and formation method of the stopper layer 111 are the same as in the first embodiment.

  The first silicon substrate 131 and the second silicon substrate 132 are bonded via the patterned stopper layer 111. After bonding, a back surface etching mask 116 is formed on the back surface side of the second silicon substrate 132, and etching is performed from the back surface side of the second silicon substrate 132 to form a first supply port 112 (first etching). This etching is stopped by the stopper layer 111. The etching is stopped when the formation of the first supply port 112 is completed at all places in the wafer (FIG. 6D), and the stopper layer exposed at the bottom of the first supply port 112 by dry etching or wet etching. Etch the entire surface of 111.

  The stopper layer 111 is etched, and the etching is stopped when the first silicon substrate 131 is exposed at the bottom of the half-etched portion. Thus, part of the stopper layer 111, that is, the patterned stopper layer is selectively removed. Next, the first silicon substrate 131 is etched (second etching) through the stopper layer 111 to form a second supply port 113. After the second supply port 113 is formed to the desired depth, the etching is stopped (FIG. 6E). The thickness of the stopper layer 111 and the depth of the half etching are appropriately set to withstand the etching of the second supply port 113.

  When the opening shape of the second supply port 113 becomes smaller, the aperture ratio (area to be etched / wafer area) decreases. In particular, in the case of etching having a large aperture ratio, the etching rate distribution is improved by the reduction of the aperture ratio. This is because the substance having an etching action (etchant) is consumed more and becomes more deficient as the opening ratio increases, but the shortage of etchant is eliminated by the decrease of the opening ratio, and the etchant between the central portion of the wafer and the peripheral portion of the wafer The amount of is equalized. FIG. 7 shows the relationship between the aperture ratio and the etching rate distribution. For example, when the opening ratio of the second supply port 113 is 35%, the etching rate distribution is 40%. On the other hand, when the opening cross-sectional area of the second supply port 113 is halved to reduce the aperture ratio to 17%, the etching rate distribution is reduced to 20%. As a result, the depth distribution of the second supply port 113 can also be improved to half.

  The steps after stopping the etching for forming the second supply port 113 are the same as in the first embodiment. That is, the back surface protection member 120 which protects a back surface is bonded together, and the surface protection member 117 is removed. Then, an etching mask is formed on the surface side of the first silicon substrate 131, a third supply port 114 (third hole) is formed, and when it is in communication with the second supply port 113, etching (third The etching) is stopped (FIG. 6F). Thus, a silicon substrate in which three stages of through holes are formed can be obtained. Thereafter, the wall 118 of the discharge port forming member and the top plate 119 (discharge port forming member) of the discharge port forming member are formed on the surface of the first silicon substrate 131 to complete the liquid discharge head (FIG. 6G) ).

  With regard to the opening shape of the second supply port 113, it is preferable to reduce the aspect ratio in the substrate in-plane direction of the opening in order to improve the etching rate distribution. In plasma dry etching, a supply port with a large aspect ratio usually lowers the etching rate at both ends in the longitudinal direction. The reason why the etching rate decreases is that the plasma sheath at the time of etching is distorted at the end of the groove having a large aspect ratio, and the etchant is less likely to enter the bottom. Here, the aspect ratio is the maximum value of the value obtained by dividing the longer width by the shorter width when the width in each direction is measured in two orthogonal axial directions for an arbitrary two-dimensional shape. It is.

  In order to reduce the aspect ratio of the opening, as shown in FIG. 8, the plurality of second supply ports 113 may be separately communicated to each of the plurality of third supply ports 114. Further, as shown in FIG. 9, the first supply port and the second supply port are provided with a plurality of independent third supply ports such that the aspect ratio of the second supply port 113 is close to 1. The first supply port may be in communication with the plurality of second holes independent of each other. With such a configuration, the etching depth distribution of the second supply port 113 can be improved. Further, in the present embodiment, it is preferable that the aspect ratio of the opening shape in the substrate in-plane direction is smaller in the second supply port than in the first supply port.

  Further, as a method of forming the second supply port 113, as shown in FIG. 10, the stopper layer 111 may be formed of two or more kinds of materials. For example, the stopper layer 111 can also be formed from the first stopper layer 122 and the second stopper layer 123. First, a recess 124 penetrating the first stopper layer 122 is formed (FIG. 10 (B)), and a second stopper layer 123 is formed to fill the recess 124 (FIG. 10 (C)), The second stopper layer 123 is embedded in the depression 124 by planarization by CMP, whole surface etching, or the like (FIG. 10D). Here, it is not necessary to completely fill the recess 124 with the second stopper layer 123, as long as the recess 124 is closed with a sufficient thickness.

  Subsequently, as shown in FIG. 10E, after bonding the first silicon substrate 131 and the second silicon substrate 132, the first supply port 112 is etched from the back surface side of the second silicon substrate 132. And stop at the above two stopper layers. Then, the first silicon substrate 131 is exposed by removing only the second stopper layer 123, and the second supply port 113 is formed by etching using the first stopper layer 122 as a mask (FIG. 10). (F)).

  The materials of the first stopper layer 122 and the second stopper layer 123 can be selected from those exemplified as the material of the stopper layer 111 of the first embodiment. However, since the second stopper layer 123 needs to be able to be selectively removed with respect to the first stopper layer 122, it is necessary to use a material different from the first stopper layer 122. That is, the patterned portion of the stopper layer is preferably composed of two or more types of materials. In addition, it is necessary to use a material sufficiently resistant to etching for the stopper layer.

  Membrane layer 103 such as aluminum wiring, interlayer insulating film of silicon oxide film and silicon nitride film, heater thin film of tantalum nitride on an 8-inch silicon substrate (thickness: 730 μm) as a first silicon substrate 131 by photolithography process A contact pad electrically connected to an energy generating element 107 formed of a pattern and an external control unit was formed (FIG. 2A).

  Next, an ultraviolet curing tape with a thickness of 120 μm was attached as the support member 110 to the surface of the first silicon substrate 131, and the back surface of the silicon substrate was thin-processed to a thickness of 300 μm by a grinding device (FIG. 2 (B)). Thereafter, the surface was polished by a CMP apparatus in order to smooth the ground surface. After smoothing the surface roughness to 0.3 nm using a slurry composed mainly of colloidal silica and a non-woven polishing pad by a CMP apparatus, 8 mass% of ammonia, 8 mass% of hydrogen peroxide, 84 mass of pure water The polished surface was cleaned using a cleaning solution consisting of a mixture of 1% and 10%.

  On the other hand, a silicon substrate with a thermal oxide film (substrate thickness 430 μm, second silicon substrate 132) having a thermal oxide film with a thickness of 1 μm formed on the surface as the stopper layer 111 is prepared. After cleaning with a hydrofluoric acid solution, bonding was performed with the thinned first silicon substrate 131 (FIG. 2C). Bonding was performed by exposing the bonding surface to nitrogen plasma, aligning the wafer with a bonding alignment device and fixing the wafer, and then bonding directly at room temperature using a bonding machine.

  After bonding, the ultraviolet curing tape (support member 110) was removed, and heat treatment was performed in an oven at 300 ° C. for 1 hour. Thereafter, a protective tape as a protective member 117 is attached to the surface of the bonded substrate, and a back surface etching mask 116 of the first supply port is formed of a positive resist on the back surface side of the substrate. Thereafter, a first supply port 112 (first hole) was formed by etching using a silicon dry etching apparatus using a Bosch process. The etching was stopped when the 1 μm thermal oxide film as the stopper layer 111 was exposed (FIG. 2 (D)).

Thereafter, the stopper layer 111 was removed by anisotropic etching using an etching gas containing CF ₄ as a main component by an oxide film dry etching apparatus. Then, using a silicon dry etching apparatus, the second supply port 113 (second hole) was etched by Bosch process. The etching was stopped at the point where the second supply port was formed to a depth of 150 μm. After that, the etching deposit in the back surface etching mask 116 and the supply port was removed by combining cleaning with a stripping solution and oxygen plasma ashing (FIG. 2E).

Subsequently, a protective tape was attached as the back surface protective member 120 to the back surface side of the substrate, and then the surface protective member 117 on the front surface side of the substrate was peeled off. Thereafter, a resist mask was formed on the substrate surface side, and the membrane layer 103 (silicon oxide film and silicon nitride film) on the surface of the silicon substrate was processed by an oxide film dry etching apparatus. Furthermore, the first silicon substrate 131 therebelow was processed by a silicon dry etching apparatus to form a third supply port 114 (third hole) having a square shape of 50 × 50 μm ² (see FIG. 2 (F)). The etching is stopped when the third supply port 114 communicates with the second supply port 113, the back surface protection member 120 is peeled off, and the entire substrate is cleaned with a peeling solution, whereby a substrate having two stages of through holes is obtained. It formed.

  Next, a negative dry film resist (trade name TMMF (trade name) manufactured by Tokyo Ohka Kogyo Co., Ltd.) having a thickness of 20 μm was attached to the surface of the substrate having the two-step through holes by a tape laminator. Next, exposure and development were performed using an exposure apparatus, and the wall 118 of the discharge port forming member was patterned. Further, the dry film resist was laminated on the wall 118 of the discharge port forming member, exposed and developed to form the top plate 119 of the discharge port forming member provided with the discharge port. After that, baking was performed in an oven under conditions of 200 ° C. for 1 hour to produce a liquid discharge head shown in FIG.

DESCRIPTION OF SYMBOLS 101 Silicon substrate 102 Discharge opening formation member 103 Membrane layer 104 Discharge opening 105 Surface hole part 106 Back surface hole part 107 Energy generating element 110 Support member 111 Stopper layer 112 First supply port (first hole)
113 Second supply port (second hole)
114 Third supply port (third hole)
116 back surface etching mask 117 surface protecting member 118 wall of discharge port forming member 119 top plate 120 of discharge port forming member back surface protecting member 121 patterning portion 122 first stopper layer 123 second stopper layer 124 recess 131 first silicon substrate 132 Second silicon substrate

Claims (13)

A method of processing a silicon substrate, wherein a through hole is formed in the silicon substrate,
(A) preparing a set of a first silicon substrate, a second silicon substrate, and a stopper layer provided between the first silicon substrate and the second silicon substrate;
(B) In the second silicon substrate, the stopper layer is exposed by the first etching for etching the second silicon substrate from the surface opposite to the interface between the stopper layer and the second silicon substrate. Forming the first hole portion;
(C) removing at least a part of the exposed stopper layer to expose the first silicon substrate;
(D) forming a second hole by performing a second etching to etch the exposed first silicon substrate, and stopping the second etching in the first silicon substrate;
(E) In the first silicon substrate, a third etching for etching the first silicon substrate is performed from the surface opposite to the surface on the side of the second hole, in the first silicon substrate Forming a third hole by stopping the third etching;
Including
After the second etching or the third etching is performed, the second hole and the third hole are communicated with each other to form the through hole. Method.   The method of processing a silicon substrate according to claim 1, wherein the first, second and third etchings are dry etchings.   The step (A) includes the step of bonding the first silicon substrate to the second silicon substrate through the stopper layer after the step of thinning the first silicon substrate. A method of processing a silicon substrate according to item 1 or 2.   The method of processing a silicon substrate according to claim 3, wherein in the step of thin processing the first silicon substrate, the thickness of the first silicon substrate after thin processing is 150 μm or more.   5. The method for processing a silicon substrate according to claim 3, wherein, in the step of processing the first silicon substrate, the support member containing a resin is attached to the first silicon substrate and then processing is performed.   The method for processing a silicon substrate according to any one of claims 1 to 5, wherein in the step (C), the stopper layer exposed to the bottom of the first hole is completely removed.   In the step (C), part of the stopper layer exposed to the bottom of the first hole is removed, and the opening cross-sectional area of the first hole is formed by the second etching in the step (D). The method of processing a silicon substrate according to any one of claims 1 to 5, wherein the second hole having a smaller opening cross sectional area is formed.   8. The method according to claim 7, wherein the stopper layer is patterned in the step (A), and the step of removing a portion of the stopper layer selectively removes the patterned stopper layer. The processing method of the silicon substrate as described in.   9. The method of processing a silicon substrate according to claim 8, wherein the patterned portion of the stopper layer is composed of two or more kinds of materials.   The method for processing a silicon substrate according to any one of claims 7 to 9, wherein an aspect ratio regarding an opening shape in a substrate in-plane direction is smaller in the second hole than in the first hole.   The method for processing a silicon substrate according to any one of claims 1 to 10, wherein the first and second holes communicate with a plurality of independent third holes.   The method of processing a silicon substrate according to claim 11, wherein the first hole communicates with the plurality of second holes independent of each other.   The method for processing a silicon substrate according to any one of claims 1 to 10, wherein the first and second holes communicate with one of the third holes.

Images