BACKGROUND
Silicon is the basic building block material of most microelectronic devices. Other micro-scale devices such as microelectro-mechanical devices (MEMs), and micro-optic devices, are also generally made of silicon. These devices are used in virtually all modern electronic products. The raw silicon material used in making these types of microscopic devices is ordinarily provided in the form of thin flat polished wafers.
Porous silicon is a form of silicon having tiny openings or pores. These pores can absorb and emit light. This allows porous silicon devices to interact with light and electronic devices in many useful ways. Porous silicon also has a very large surface area and acts as a strong adsorbent. These properties make porous silicon useful in mass spectrometry, micro-fluidic devices, sensors, fuel cell electrodes, optical, chemical and mechanical filters, biochips and biosensors, fuses for airbags, and various other products.
The porous silicon material itself may also be used as a porous and/or solvable substrate, for example in diagnostic or therapeutic products. Accordingly, porous silicon is increasingly becoming an important material in a wide range of products and technologies.
Porous silicon is generally manufactured in an electro-chemical etching process. A silicon wafer is typically exposed to an electrolyte including concentrated hydrofluoric acid (HF). The electrolyte on one side of the wafer is sealed off from the electrolyte on the other side of the wafer. Electrical current is passed through the electrolyte on each side, making one side the cathode and the other side the anode. The silicon wafer may optionally be exposed to light during this process. The process etches pores in the wafer. The pores are microscopic. A 150 mm diameter wafer may have more than 1 billion pores after electro-chemical processing.
Although various types of porous silicon machines or processors have been used, disadvantages remain in performance, reliability, speed, and other design parameters. HF is highly corrosive and toxic. Accordingly, it must be carefully contained within the processor. Since HF will react with virtually all metals, metals cannot effectively be used in areas of the processor that may come into contact with HF. Moreover, even the smallest of amount of interaction between the HF in the electrolyte and metal can contaminate the wafer. The uniform processing required to consistently produce high quality porous silicon also requires uniform electrical current flow through the electrolyte. Achieving uniform current flow is affected by the design of the processor and may be challenging to achieve. Existing processors have offered only varying results in the face of these engineering design challenges. In view of these factors, improved methods, processors and systems for making porous silicon are needed.
SUMMARY
A novel processor has now been invented providing various improvements in making porous silicon or in similar electro-chemical processing. This new processor provides highly uniform processing. Potential for contamination of wafers before, during, and after processing is significantly reduced. Potential for corrosion of processor components is similarly largely avoided, offering long term reliability and performance, and reduced maintenance requirements. The processor is also adaptable for use in an automated processing system, providing relatively rapid processing. These advantages are achieved via a new processor having a first seal in housing, and a first electrode in the housing associated with the first seal. A second seal in the housing may be moved relative to the first seal. A second electrode is associated with the second seal. The housing may be set up to pivot from a horizontal position to a vertical position. This allows a wafer to be loaded and unloaded in a horizontal position, and processed in a substantially vertical position.
The invention resides as well in methods for electro-chemical processing, and in sub-combinations of the elements and steps described.
BRIEF DESCRIPTION OF THE DRAWINGS
In the drawings, wherein the same reference number indicates the same element, in each of the views:
FIG. 1 is a top and front perspective view of a porous silicon processor.
FIG. 2 is a bottom and back perspective view of the processor shown in FIG. 1 with the mounting plates omitted for purpose of illustration.
FIG. 3 is an exploded top and front perspective view of the processor shown in FIGS. 1 and 2.
FIG. 4 is a plan view of the processor shown in FIGS. 1-3.
FIG. 5 is a partial section view taken along line 5-5 of FIG. 4.
FIG. 6 is a partial section view taken along line 6-6 of FIG. 4.
FIG. 7 is a partial section view taken along line 7-7 of FIG. 4.
FIG. 8 is a partial section view taken along line 8-8 of FIG. 4.
FIG. 9 is a section view taken along line 9-9 of FIG. 4 and showing the head of FIGS. 1-8 alone.
FIG. 10 is a top and front perspective view of the head shown in FIG. 9 with the head cover removed.
FIG. 11 is a top and front perspective view of the base shown in FIGS. 1-8, with FIG. 11 showing the base separated from the head.
FIG. 12 is a plan view of the base shown in FIG. 11.
FIG. 13 is a section view taken along line 13-13 of FIG. 12.
FIG. 14 is a section view taken along line 7-7 of FIG. 4 and showing the processor in a closed or processing position, with FIGS. 5-8 showing the processor in an open position, for loading and unloading a wafer.
FIG. 15 is an enlarged section view of the seals and ejector tab shown in FIG. 5.
FIG. 16 is a perspective view of the lower seal retainer shown in FIGS. 13-15.
FIG. 17 is a perspective view of the upper seal retainer shown in FIGS. 14-15.
FIG. 18 is a section view taken along line 18-18 of FIG. 17.
FIG. 19 is a perspective view of an automated processing system including the processor shown in FIGS. 1-15.
FIG. 20 is a top view of the system shown in FIG. 19.
FIG. 21 is a front view of the system shown in FIG. 19.
FIG. 22 is a side view of the system shown in FIG. 19.
DETAILED DESCRIPTION OF THE DRAWINGS
Turning now in detail to the drawings, as shown in FIGS. 1-3, a first processor assembly or head 34 is attached to a second processor assembly or base 32, to form an electro-chemical processor 30. A motor or other actuator, such as a rotate motor 38, can move the processor 30 from the horizontal position shown in FIGS. 1-3, one-quarter turn, into a vertical position.
A retainer generally designated 48 is provided on the head and/or the base for holding them together. Various forms of retainer 48 may be used. In a basic form, the retainer 48 may simply be bolts or other fasteners holding the head onto the base. FIGS. 1-4 show another form of retainer 48 having four spaced apart cam handles 50 pivotably attached to the base 32 via pivot bolts 54, and with a cam latch 52 pivotably attached onto each cam handle 50. When engaged or locked, the cam handles 50 securely seal the head 34 to the base 32, as shown in FIG. 1. The cam handles 50 may be quickly released (by pulling radially outwardly), to allow the head 34 to be separated from the base 32, for system set up, inspection, or maintenance.
Turning in addition now to FIG. 5, when secured together, the head 34 and the base 32 may form a containment chamber 60, with process chambers 146 and 240 within the containment chamber. Referring momentarily again to FIGS. 1-3, a load/unload workpiece opening or slot 56 extends through the base 32, to allow a wafer to be moved through the containment chamber 60 to the process chambers. A containment drain or opening and gas/vapor exhaust 58 may be provided in the base 32, generally opposite from the load slot 56. A frame 62 may surround the load slot 56 at the front of the processor 30.
For electro-chemical processing, the processor 30 is provided with two electrodes and two process chamber seals. At least one process chamber seal is moveable. An electrode may move with the moveable process seal. The moveable seal may be in the head 34 or in the base 32. The other electrode and process chamber seal, may be fixed or moving, and typically are fixed in place within the processor 30. The drawings show an example of the processor 30 where the moving electrode and seal is in a head, and a fixed electrode and seal is in a base, positioned vertically on top of the base. However, these positions may be reversed, as they are not essential to the invention. Except for the two electrodes and the two process chamber seals, the other specific components described below, including those forming the containment chamber 60, are not necessarily essential and may be omitted, or substituted out in place of an equivalent functional element.
The specific mechanism or force selected to move the moveable seal is also not essential. This movement may be provided by hydraulic, pneumatic, electric, gas or steam pressure, or mechanical forces. In the design shown, hydraulic force is used, with water as the hydraulic fluid. In this example of a hydraulically driven processor, as shown in FIG. 5, a piston cap 78 is attached to the head cover 70. A piston 74 is fixed in position relative to the cover 70 by a piston nut 76 and a piston plate 80. A cylinder 82 is supported around the piston 74, with the lower end of the piston 74 sealed against the interior cylinder walls by a piston seal 94. A piston ring 88 is attached to the upper end of the cylinder 82. An inner piston ring seal 90 seals the piston ring 88 against the piston 74. An outer piston ring seal 92 seals the piston ring 88 against the cylinder 82. For clarity of illustration, fluid and electrical lines and cables are generally omitted from the drawings.
A cylinder ring 86 is attached to an annular flange 84 of the cylinder 82. An upper or first electrode ring 106 is in turn attached to an annular flange of the cylinder ring 86 via cap screws 112. An upper or first electrode 96 (in this case, the cathode) is held in place between the electrode ring 106 and the cylinder ring 86. The electrode 96 is sealed against the electrode ring 106 by first and second seals 110 and 108 at the front surface and cylindrical side of the electrode 96. A third seal 104 and a fourth seal 105 seal the back surface of the electrode 96 against the cylinder ring 86. An annular groove 102 is positioned between the third seal 104 and the fourth seal 105 for improved leak detection, as described below. An electrode lead or wire 95 runs through a electrical fitting 155 on the upper fittings plate 180 and through a fitting 160 on the cylinder ring 86 and is attached to a buss plate 98 via a cap screw 100. Metal cap screws may be used to secure the buss plate onto the back surface of the electrode 96. Typically, multiple cap screws are used to secure the buss plate to the electrode, in a geometric pattern, since the number and location of the screws may affect the uniformity of current flow through the electrode, and ultimately affect current uniformity at the wafer.
Referring to FIGS. 5 and 15, a head seal 128 is attached to the bottom surface of the electrode ring 106 via an upper seal clamp ring 132. The cylindrical open space shown in FIG. 5 as CC between the electrode 96 and the plane of the head seal forms an upper process chamber 146 when the seal 128 is in contact with a wafer. The upper end of a bellows 120 is attached to the underside of an annular cam ring 72 on the cover 70 by an upper bellows retainer ring 122. The lower end of the bellows 120 is attached to a lip on the electrode ring 106 via a lower bellows retainer ring 124. As a result, upon actuation of the cylinder 82, the entire moveable electrode assembly 152, including the piston ring 88, cylinder 82, cylinder ring 86, electrode 96, electrode ring 106, and the seal 128 can move vertically relative to the cover 70, as well as to the base 32, with the bellows 120 maintaining a seal between the electrode assembly 152 and the base 32.
Referring now also to FIGS. 7 and 10, an electrolyte liquid port or outlet 148 connects from an outlet or recirculation fitting 150 to a duct 148 in the electrode ring 106 that opens into the chamber 146. An electrolyte liquid inlet 142, which may be located opposite from the outlet 148, leads to an electrolyte recirculation or inlet fitting 140, with an connection line 153, shown in dotted lines in FIG. 10, connecting the fitting 140 to a fitting 154 on the upper fitting bracket 180. A diffuser plate 126 having multiple small openings is positioned over the inlet 148 in the upper process chamber 146. Windows or openings 71 and 73 are provided in the side walls of the cover 70, to provide clearance for the up and down vertical movement of the fluid and electrical fittings, e.g., the electrical connector 160 and the liquid process chemical recirculation or outlet fitting 150, as the electrode assembly 152 moves up and down. Referring still to FIG. 7, an optical liquid detector 170 extends through a clamp nut 172 and seal 174 in the cylinder ring 86, with the tip of the detector 170 positioned within the groove 102. The detector 170 can be connected to a processor controller (such as the controller 304 described below) via fiber optic lines passing through a strain relief feature on the head. The fluid and electrical or optical lines connecting to the head may be made through adapters on connectors on the upper fitting bracket to provide strain relief as the processor 30 moves between horizontal and vertical positions.
As also shown in FIG. 7, an upper seal vent 116 is provided between the second seal 108 and the third seal 104. The vent 116 is designed to reduce wicking of electrolyte inwardly between the back surface of the electrode 96 and the bottom surface of the cylinder ring 86. Similarly, a lower seal vent 118 is located to reduce wicking of electrolyte between the back surface of the lower or second electrode 208 and the electrode cover 206, in the event of a leak of electrolyte past the first seal 110 and the second seal 108.
Turning to FIG. 8, an optical flag plate 168 extends up from the cylinder ring 86. Upper and lower optical sensors 162 and 164 attached to the cover 70, and are also connected via the fiber optic leads 163 and 165 to the processor controller. Strain relief fittings such as the clamp plate 169 shown in FIG. 5, are typically provided on these leads, so that they may better accommodate the movement of the processor. The sensors 162 and 164 detect the position of the flag plate 168, which corresponds to the position of the moveable electrode assembly 152 relative to the base 32. Also as shown in FIG. 8, the electrode ring 106 includes bores 114. Guide pins 214 in the base 32 extend into the bores 114, to maintain the moveable electrode assembly 152 in alignment with the base 32.
FIG. 9 shows the head 34 separated from the base 32. With the head retainer 48 released, in this case by pulling the cam handles 50 outwardly, the head 34 may be separated from the base 32. The electrical and fluid lines connecting to the head may be flexible, so that the head may be removed from the base without the need to break these connections.
FIG. 10 shows the head 34 separated from the base 32, and with the head cover 70 removed, for purpose of illustration. As shown in FIG. 10, the upper fitting bracket 180 and the optical flag plate 168 are attached to the cylinder ring 86. The cap screws 79 which ordinarily attach the piston cap 78 and the piston plate 80 to the cover 70 are shown in their assembled positions, but without the cover 70 in place. The upper optical sensor 162 and the lower optical sensor 164 are attached to the side wall of the cover 70. However, they are shown in FIG. 10 for purpose of illustration only. An alignment pin 182 in the electrode ring 106 may extend into a vertical slot on an inside surface of the head cover 70, to keep the head cover angularly aligned with the movable electrode assembly 152. As shown in FIGS. 1, 2 and 9, cylinder water supply and return lines extend from fittings 176 and 178 on the upper fittings bracket 180 to upper and lower cylinder ports 156 and 158 extending through the walls of the cylinder 82. Alternately supplying water under pressure through the upper and lower cylinder ports 156 and 158 hydraulically moves the moveable electrode assembly 152 between up or open and down or closed positions.
FIGS. 11, 12, and 13 show the base 32 separated from the head 34. Referring to FIGS. 11, 12, and 8, three guide pins 214 project upwardly from a base electrode ring 204 into the bores 114 in the head electrode ring 106. A pin seal 215 seals the base of the pin against the base electrode ring 204, although in ordinary use, the guide pins 214 are not extensively exposed to the corrosive electrolyte liquid.
As shown in FIGS. 11 and 12, the two front guide pins 214, closest to the load slot 56, are spaced apart by a dimension DD which is nominally larger than the wafer diameter. The third guide pin 214 is located towards the back of the processor 30, closer to the containment drain 58. Referring to FIGS. 8 and 12, the guide pins 214 are located on a diameter concentric with and slightly greater than, the diameter of the seal 128. The lower section of each guide pin 214 has a shoulder 218 which may act as a hard stop to set the spacing between the seals when the processor is closed, thereby also setting a predefined amount of seal compression on the wafer.
Referring to FIG. 13, the base 32 has a base ring 200. A seal seat 202 at the upper end of the base ring 200 holds a containment seal 36, which is shown in FIG. 3. The containment seal 36 seals the head to the base to form the containment chamber. Similar to the head 34, the lower or second electrode 208 (in this case the anode) is secured in place in the base by a base electrode ring 204 and a base electrode cover 206. As in the head 34, the electrode 208 is sealed against the base electrode ring 204 by first and second seals 110 and 108. The base electrode ring 204 is similarly sealed against the electrode 208 by third and fourth seals 104 and 105 positioned on opposite sides of a groove 102. As in the head 34, an optical liquid detector 170 extends through the base electrode cover 206 to the groove 102. A buss plate 98 is similarly attached to the electrode 208, as described above with reference to the head 34 in FIG. 5. Referring to FIGS. 13 and 15, a base seal 210 is attached to the base electrode ring 204 by a base seal retainer 212 attached to the base electrode ring 204.
Wafer guides or protrusions 213 extend up slightly from the base seal retainer 212 as shown in FIGS. 5-8 and 16 and help with wafer alignment or positioning, when a wafer is placed onto the base seal 210, as described below. The upper or head seal 128 is generally the same diameter as the base seal 210. Indeed, the head and base seals may be the same. As shown in FIG. 3, the processor 30 may have seals 128 and 210 and other components adapted for processing a wafer having a flat edge. In this case the seals, seal retainers, and the electrodes may be generally D-shaped. For round wafers, these components may be round.
Referring to FIGS. 15, 17 and 18, the upper seal clamp ring 132 has resilient ejector tabs 130 that press down slightly on the outer edge of the wafer, when the head seal 128 is engaged against the wafer 250. FIG. 15 shows in dotted lines the nominal position where the ejector tabs 130 would be with no wafer present. The seal 128 and the upper seal clamp ring 132 are dimensioned so that as the movable electrode assembly 152 moves up away from the wafer 250, the seal 128 separates from the wafer first, while the ejector tabs 130 continue to hold the wafer down onto the base seal 210. This prevents any potential for having a wafer stick to the head seal 128 as the seal is lifted away from the wafer after processing.
As shown in FIG. 13, an electrolyte inlet fitting 220 leads into an inlet 224 in the base electrode ring 204. As in the head 34, a diffuser plate or similar liquid diffusing element 126 may be attached to the base electrode ring 204 over the inlet 224. With the base seal 210 in contact with a wafer, a base or lower process chamber 240 is formed between the electrode 208 and the wafer, with the lower process chamber 240 surrounded by the base electrode ring 204. An electrolyte outlet 226 runs from the lower process chamber 240 to an electrolyte outlet fitting 222. A diffuser plate 126 may also be provided over the electrolyte outlet 226.
Referring back to FIG. 5, a lower fitting bracket 228 is attached to the base electrode cover 206. The optical liquid detector 170 in the base 32 connects to a fitting 232 on the bracket 228, along with the electrolyte line connections, to provide strain relief. A lower electrode wire 234 extends through an electrical feed through fitting 230 in the electrode cover 206 and connects to the buss plate 98 on the electrode 208.
In the processor 30 shown in FIGS. 1-8, the electrode 96 in the head 34 is typically made the cathode, while the electrode 208 in the base 32 is typically made the anode, by selecting the polarity of the current source attached to each electrode. The electrodes 96 and 208 may otherwise be the same. Although various materials may be used, the design shown uses electrodes made from boron-doped silicon. Each electrode is about 25 mm thick. The diameter of the electrodes may be substantially the same as the diameter of the wafer. The diameter of the head and base seals 128 and 210 is typically 2-10 mm less than the wafer diameter, providing, for example, an edge exclusion zone (the outer annular area of the wafer protruding beyond the seal) from about 1-5 mm. The electrode surface may be diamond coated. If used, the diamond coating is doped to make it electrically conductive.
As the electrolyte generally will include concentrated hydrofluoric acid, the components of the processor 30 coming in contact with the electrolyte are made of materials, such as Teflon (fluorine resins) or PVDF, which are resistant to corrosion by HF or other reactive electrolyte chemicals. The cap screws or other fasteners in the processor 30 generally may be made of similar plastic or non-metal materials. Referring to FIG. 5, the buss plates 98, electrode wire leads, and wire lead attaching screws are metal, as these components require high electrical conductivity. However, these may be the only metal components in the processor 30 (excluding the motor 38 or other external components). In addition, these metal components are sealed off from the electrolyte introduced into the process chambers 146 and 240 by the first seal 110, second seal 108, third seal 104, and the fourth seal 105.
In the event of any leakage around the electrode, electrolyte would first collect in the groove 102, and be detected by the optical liquid detector 170. In addition, the seal vents 116 and 118 will tend to divert any leaking electrolyte away from the back of the electrode. Upon detection of a leak, the controller shuts down the processor 30, before any electrolyte can move past the fourth seal 105. In this way, the electrolyte is entirely isolated from any metal in the processor 30. Metal contamination of the electrolyte or wafer, or inadvertent release of electrolyte into the head or base, is accordingly avoided.
The processor 30 provides highly uniform current flow through the process chambers 146 and 240, yet within a relatively small space. The clearance space around the processor 30, to allow it to rotate between horizontal and vertical positions, is also relatively small. Referring again to FIG. 5, the electrodes shown have a diameter of about 150 mm and a thickness of about 25 mm. For this type of design, an electrode (or wafer) diameter to electrode thickness ratio of about 4-10:1 may be used. The height of the chambers (dimension CC in FIG. 5) is also generally about the same as the electrode thickness in this example, with ratios of chamber height to electrode thickness ranging from about 2.5 to 1 to about 1 to 1 may be useful.
The height of the chambers, and/or the electrode thickness, can of course also exceed these ranges, although this may tend to make the entire processor larger, with no improvement in current uniformity. The diameter and height of the containment chamber 60 are not critical and may be selected to accommodate the size and/or shape of other internal processor components, within a compact space. While the drawings show the chambers 146 and 240 as having substantially the same height, one chamber may have a larger height than the other. The chambers 128 and 240 as described above have minimal diameter and height. In addition to providing for a compact processor, this also speeds up processing, since process liquids can be quickly filled and drained from the chambers. Typically, the containment chamber 60 may have a diameter of about 1.1 to 2 or 1.1 to 3 times the diameter of the seals 128 or 210 or the workpiece. The height of the containment chamber 60 may be from about %5-%50 of the diameter of the seals 128 or 210.
In use, the processor 30 is initially loaded with a wafer 250. For loading (and unloading a wafer), the processor 30 is in the horizontal position as shown in FIG. 1. The moveable electrode assembly 152 in the head 34 is in the up position, as shown in FIGS. 5-9. A wafer is moved into the processor 30 through the load slot 56, typically by a robot. As the wafer moves into the processor, the edges of the wafer may contact the wafer guides 213 which protrude up from the base seal retainer 212. This properly centers and locates the wafer 250 relative to the seals 128 and 210. The wafer 250 is then released, with the wafer resting on the lower or base seal 210. The robot (or other wafer mover used) is withdrawn.
The controller opens valves causing water to be supplied under pressure to the lower cylinder port 158. This drives the cylinder 82 and the entire moveable electrode assembly 152 downwardly. Referring to FIGS. 9 and 14, water within the cylinder 82 above the piston seal 94 flows out of the cylinder via the upper cylinder port 156 and out the processor 30 via a return line. The water is used only as a hydraulic fluid and does not come into contact with the electrolyte or wafer 250. Accordingly, water purity is not critical, so that standard tap water, under standard plumbing pressures, may be used. The moveable electrode assembly 152 continues to move down towards the wafer until it bottoms out on the hard stop provided by the shoulders 218 of guide pins 214. The lower position sensor 164 provides a signal to the controller confirming that the moveable electrode assembly is in the process position. The water pressure provided to the cylinder may optionally be regulated, although regulation is not necessary. The upper or head seal 128 is pressed into sealing contact with the top surface of the wafer 250. The base seal 210 is also in sealing contact with the wafer 250. The processor 30 is then in a closed position, as shown in FIG. 14. Water pressure may be maintained during processing.
Referring momentarily to FIGS. 1-4, the controller actuates the rotate motor 38, pivoting the entire processor 30 about 90 degrees, so that the wafer 250 is moved into a vertical orientation, with the slot 56 facing up and with the drain 58 facing down. The rotate motor 38, or an equivalent driving element, may be supported on a deck 42 or other supporting surface. The processor 30 can be supported on a pair of pivot blocks 332 attached to the base 32. In FIG. 4, the pivot blocks 332 are attached to flange bearings 40 with the rotate motor attached to the right side pivot block 332, optionally through a gear drive reducer 39.
The controller then opens valves supplying electrolyte to the processor 30. Electrolyte flows into the upper and lower process chambers 146 and 240 through the inlets 148 and 224, and through the diffuser plates 126, as shown in FIGS. 7 and 14. The process chambers are filled from the bottom up. Valves controlling flow through the return lines 142 and 226 may be opened to allow the chambers to vent while filling with electrolyte, or during other times, or at all times during the processing.
Electrical current is applied to the electrodes 96 and 208. Current flows from the cathode or first electrode, through the electrolyte in the chamber 240, through the wafer, and through the electrolyte in the chamber 146 to the anode, or other electrode. The wafer is sufficiently conductive to provide a bi-polar electrode function. Electrolyte may be continuously provided at a low flow rate, so that the electrolyte in the chambers 146 and 240 is constantly refreshed, although without substantial fluid turbulence.
During processing, the chambers 146 and 240 are virtually entirely filled with electrolyte to provide more uniform processing. Gasses generated during processing may be carried off via the circulation of electrolyte through the chambers 146 and 240. Alternatively, separate gas exhaust ports may optionally be used in the chambers 146 and 240. The motor 38 may be controlled to oscillate the processor 30 about a near vertical position, to assist with gas removal, either while the chamber is being filled with electrolyte, or during processing, or both. The process described produces amorphous porous silicon.
The electrolyte parameters, such as chemical composition, temperature, pressure, flow rate, concentration, etc., may be varied to achieve desired process results. Current flow may also be selected as desired. The current may be increased to a high enough level to transition from a porous silicon process to a wafer polishing process. The processor 30 may therefore be used for wafer polishing. The electrolyte may include water, relatively concentrated HF, and an alcohol, such as isopropyl alcohol. Processing continues, for example, for about 2-10 minutes, until the wafer surface 250 is sufficiently etched and becomes porous silicon. Electrical current is turned off. The electrolyte is drained from the chambers. The chambers and workpiece may then be rinsed by filling the chambers with a rinse liquid, such as de-ionized water, and then draining the rinse liquid. The rotate motor 38 is actuated in the reverse direction, to pivot the processor 30 back into the horizontal position shown in FIG. 1.
The controller then supplies water pressure to the cylinder 82 in the reverse direction, to lift the moveable electrode assembly 152 up and away from the wafer 250. As shown in FIG. 15, the ejector tabs 130 on the upper seal clamp ring 132 hold the wafer 250 down onto the base seal 210 until the head seal 128 is separated from the wafer 250. This prevents the wafer 250 from inadvertently sticking to the head seal 128. The wafer is then removed from the processor 30, again, typically via a robot grasping the edges of the wafer 250 from above, and withdrawing the wafer out of the processor 30 through the load slot 56. The processor 30 is then ready to process a subsequent wafer 250. The rotate motor 38 may optionally gently or even rapidly rock or oscillate the chamber 30 at various times, to help to agitate the electrolyte, displace gas bubbles, provide mixing of other chemicals that may be used, or to help distribute rinsing liquid. A gas, such as heated nitrogen, may also optionally be provided into the processor for drying the wafer.
In some applications, the processor may operate with the chambers filled with electrolyte or other process liquid, but with no electrical current flowing. Since the processor is well designed to operate with highly reactive or corrosive electrolyte, it can also operate with other reactive or corrosive process liquids, including HF, without use of electricity. This provides a purely chemical process, rather than an electro-chemical process. Since the chambers 128 and 240 are sealed off from each other, different process liquids may be provided into each chamber, simultaneously or sequentially. Consequently, the front or device side of the wafer and the back side of the wafer may simultaneously be processed using different process liquids and/or gases. With this type of processing, the process liquids may optionally be introduced into the chambers 128 and 240 with the wafer in a horizontal orientation, or in a vertical orientation. If the processor 30 is intended for non-electrical processing, the electrodes may be removed and the electrode rings simply replaced with plates to form the upper and lower process chambers. In addition, the seals 128 and 210 may be designed to seal directly against each other, without contacting the wafer at all, and with the wafer supported within, rather than on, the lower seal 210.
Some wafers may be provided with a mask to determine which areas of the wafer are made porous. After electro-chemical processing, electrical current may be turned off, additional chemical processing steps may be performed, with or without changes to the electrolyte, to etch off the mask, or another layer or film on the wafer.
Referring to FIGS. 5, 6, and 7, during processing the electrolyte is sealed within the upper and lower process chambers 146 and 240. These process chambers are surrounded by the containment chamber 60, which is open at one side at the load slot 56, and at the opposite side at the containment drain 58. However, there is no other fluid pathway out of the containment chamber 60. The bellows 120 seals the upper electrode ring 106 to the cover 70. Accordingly, a rinse liquid, such as water, may be provided into the containment chamber 60 via the load slot 56, to rinse all exposed surfaces within the containment chamber 60. The containment chamber itself may also be provided with rinse nozzles 252 connected to a rinse liquid source as shown in FIG. 5, for rinsing the chamber.
The rinse liquid may be provided between wafer processing, while the processor is open and the seals 128 and 210 are completely exposed. This allows virtually all surfaces of the seals to be rinsed, removing any trapped or adhering electrolyte. Rinsing can advantageously be performed with the processor 30 once again rotated into the vertical orientation, with the rinse liquid flowing via gravity through the containment chamber 60 and draining out of the containment drain 58. Rinsing with the chamber open allows the processor to maintain uniform process start up conditions, since a complete rinse of all surfaces contacted by electrolyte (or other process chemicals) may be achieved between each process cycle.
The substantially non-conductive rinse liquid may also optionally be flowed through the containment chamber 60 while the processor is closed, during actual processing of a wafer. Since the electrolyte is sealed within the process chambers 146 and 240, the rinse liquid does not come into contact with the electrolyte, and the rinse liquid only contacts the outer seal surfaces and the annular edge of the wafer extending radially outwardly beyond the seals 128 and 210 (typically by about 2-6 mm). Since the electrolyte is an electrical conductor, any leaking electrolyte may alter the otherwise uniform conduction path provided by the processor 30. This can cause non-uniform processing. Running rinse liquid through the containment chamber during processing will remove any leaking electrolyte, thereby maintaining the uniform conduction path necessary for providing high quality porous silicon.
The rinse liquid may also be provided into or through the containment chamber upon detection of an electrolyte leak or other fault condition, to carry away any leaking or exposed electrolyte. One example of a containment chamber rinse system is described below in connection with an automated processing system.
As shown in FIGS. 19-22, the processor 30 may be used in an automated processing system 300. Various types of automated systems may be used, including varying numbers of processors 30, arranged in whichever way (e.g., linear array, arcuate array, vertically stacked, etc.) may be preferred. In the automated system 300 shown in FIGS. 19-22, two processors 30 and two spin rinse dryers 322 are provided within an enclosure 302. A load window 308 is provided at a load/unload station 306, at the front of the enclosure 302. Air inlets 310 are located on the top of the enclosure 302. A robot 316 moves along a lateral rail, to transfer wafers 250 from the load station 306 into or out of the processors 30 or spin rinse dryers 320. An isolation wall 318 may be provided between the processors 30 and the robot 316, to reduce exposure of the robot 316 to process chemicals used in the processors 30. An opening is provided in the isolation wall 318, to allow an end effector of the robot 316 to reach into the processors 30. A wafer transfer zone or area 340 may be provided between the isolation wall 318 and the processors 30. When a wafer is moved between processors 30 and 320, the wafer may remain within the transfer zone 340 as the robot 316 carries the wafer parallel to the isolation wall 318. Any residual process liquids that may be on the wafer may therefore come into contact only with the end effector, but not with the rest of the robot 316. A controller 304 controls movement and operation of the robot 316, processors 30, spin rinse dryers 320, as well as various other components within the system 300 (e.g., pumps, valves, actuators, displays, interlocks, communications, etc.), as is well known in the semiconductor field. The deck 324, isolation wall 318, enclosure 302, and other components of the system 300 may advantageously be made of plastic materials, to better avoid contamination and corrosion.
In use, wafers 250 are delivered to the load station 306, typically within a cassette, box, or carrier 314. The load window 308 opens. The robot 316 picks up a wafer 250 at the load station 306 and moves the wafer 250 into one of the processors 30. The wafer may optionally be first moved into a pre-aligner 312, or other chamber for a pre-processing step. The wafer 250 is processed within the processor 30, as described above. In the interim, the robot 316 may return to the load station 306 to repeat the load sequence and load another wafer into the second processor 30. Upon completion of processing, each wafer 250 may be moved by the robot 316 into one of the spin rinse dryers 320. The spin rinse dryers 320 shown have lift/rotate apparatus 322 used to lift and rotate the head of the spin rinse dryer 322 into a load/unload position. Various types of spin rinse dryers (or other types of additional chambers or processors, e.g., metrology, anneal, etc.), with or without lift/rotate apparatus, may be equivalently used. After each wafer 250 is rinsed and dried, the robot 316 moves the wafer back to the load station 306, where the wafer is typically placed back into the same cassette 314, or into a different cassette.
The processor 30 itself may also perform rinsing and drying, as a stand alone unit or within a processing system. After an optional line purging step, rinsing may be performed by flowing a rinse liquid, such as de-ionized water, through the process chambers 146 and 240, typically with the wafer and the chamber in the vertical position. The rinse liquid may then be relatively slowly drained out, to perform a slow extraction type of drying process. In many applications, this process will leave the wafer sufficiently dry for subsequent handling or processing, even if some droplets of rinse liquid remain on the wafer. In an alternative drying process, a surface tension/meniscus drying step may be used after rinsing the wafer. In this alternative process, a drying fluid, such as isopropyl alcohol, can be provided into the process chambers during the drying step.
Referring to FIG. 21, the system 300 may optionally include an isolation chamber rinse system 190. If used, the rinse system 190 may include a rinse water channel or array of spray nozzles 192 aligned with and adjacent to the load slot frame 62 of the processor 30, when the processor is rotated into the vertical position. Similarly, an isolation drain collection pipe 194 may be provided in the system 300, adjacent to and aligned with the isolation drain 58, when the processor 30 is in the vertical position. Adequate clearance is provided between the processor 30 and the channel 192 and collection pipe 194, so that the processor 30 can freely pivot between the vertical and horizontal positions. The isolation drain 58 may optionally be connected to a drain/exhaust line, such as the collection pipe 194, via a flexible tube 334 shown in FIG. 14, which can accommodate the vertical to horizontal movement of the processor 30, to help insure that no liquid or vapors draining from the processor 30 are released into the system 300.
The isolation chamber which is generally shown at 60 in FIGS. 5, 6 and 14, may be rinsed by moving the processor 30 into the vertical position, and then providing rinse water from the overhead array of spray nozzles 192 through the load slot 56 into the processor 30. The rinse water drains out of the processor 30 through the isolation drain 58, and into the collection pipe 194. The collection pipe 194 connects to a system or facility drain. The isolation chamber rinse system 190 may be used to provide routine rinsing of the isolation chamber 60, during wafer processing, between wafer processing, or periodically, after processing a preselected number of wafers. The isolation chamber rinse system 190 may also be used if a fault condition is detected indicating a potential leak of electrolyte. In this condition, the isolation chamber 60 can be rapidly flooded with rinse water, to reduce any leakage of electrolyte (liquid or vapors) into the system 300. The processor may optionally be made and used without any isolation chamber 60. The processor 30 may also be used as a chemical process chamber, rather than as an electro-chemical processor, by not providing any current to the electrodes, or by omitting the electrodes entirely, e.g., by replacing the rings 106 and 204 with plates with a solid and continuous center section. In this design, the various chemical process liquids may be used, instead of electrolyte.
Although wafer loading/unloading with the wafer in a horizontal position is more commonly used in many types of existing wafer handling equipment, the processor 30, or the automated system 300, may also be adapted to operate with wafer loading/unloading in a vertical orientation. Terms used here, including in the claims, such as upper and lower, above and below, etc. are intended for explanation and not requirements that one element be above or below another element. Indeed, the processor 30 may be operated upside down. While porous silicon has been described above, the processor 30 may also be used for processing similar materials, including gallium compounds. The terms vertical and horizontal here include positions within 5, 10 or 15 degrees of vertical or horizontal, respectively. The processor 30 may also be used in a fixed position. For example, the processor 30 may be used without any rotate motor 38 and mounting plates 330. In this design, the processor 30 may be supported in a fixed horizontal position, or in a fixed vertical position, or at an angle between horizontal and vertical.
Various changes and substitutions may of course be made without departing from the spirit and scope of the invention. The invention, therefore, should not be limited, except by the following claims and their equivalents.
Patent Application
20100078334
