Patent Application
20130269732
The present invention relates to a carbon dioxide supply system and method based on a novel process sequence for removal of contaminants from a substrate surface. Specifically, the process involves introduction into a process chamber a combination of different carbon dioxide phases, including supercritical carbon dioxide, to create a specific cleaning sequence designed to remove byproduct contaminants without damaging the device features formed on the substrate.
Dynamic Random Access Memory (DRAM) manufacturers continue to investigate and develop device scaling with high Aspect Ratio (AR) stacked microelectronic device features, such as, for example, cylindrical capacitors. The International Technology Roadmap for Semiconductors (ITRS) has indicated that AR's greater than, for example, 50:1 at 32 nm node and below will be required to maintain sufficient capacitance for next-generation capacitors. Such AR's for microelectronic device features continue to increase to meet the ever increasing need for processing speed and memory density of integrated circuits.
The fabrication of high AR microelectronic features can include several processing steps, such as, for example, patterning, etching and deposition of materials to produce the device features. Conductive features can be formed within a sacrificial layer which is thereafter removed by an etchant solution. The etchant solution and byproduct are typically rinsed and dried with deionized water and/or organic solvents. However, the conductive features are prone to collapse during etching, cleaning and drying by virtue of the surface tension of deionized water and organic solvents. The occurrence of such feature collapse is becoming more frequent and problematic, as the width dimensions of structures continue to decrease and their AR's continue to increase.
One method for reducing feature collapse is the use of supercritical carbon dioxide as the solvent for etching, cleaning, and drying of such features. Supercritical carbon dioxide does not have any surface tension. As a result, device structures would not collapse when in contact with supercritical carbon dioxide. Nonetheless, drawbacks exist with the use of supercritical carbon dioxide. For instance, during formation of supercritical carbon dioxide, the liquid carbon dioxide is pressurized and heated to at least the supercritical phase of 1072 psi and 31° C., during which time impurities contained in the liquid carbon dioxide, such as Non Volatile Organic Residues (NVOR's) and metals, can dissolve into the supercritical carbon dioxide. These impurities manifest themselves as particulate defects on the wafer surface at the end of process. The net effect is that the microelectronic features are unusable. To compound the problem, the etch byproducts generated during etching with supercritical carbon dioxide tend to have relatively low solubility in the supercritical carbon dioxide and, as a result, will tend to precipitate onto the wafer surface. In some cases, the precipitates of the etch byproducts may adversely alter the functionality of the resultant microelectronic device. As a consequence, the precipitant material needs to be removed by a wet rinse. However, as mentioned, utilizing wet rinse processes having high AR device structures tends to cause feature collapse by virtue of the surface tension of the solvents.
Accordingly, the need for eliminating residual byproducts during etching, cleaning and drying of a substrate would be advantageous.
The invention relates, in part, to a carbon dioxide supply method and system for removing contaminants from a substrate, in particular a semiconductor wafer. The timing and sequence for delivering a combination of various carbon dioxide phases, together with the processing conditions for such delivery, have been found to affect the ability to remove contaminants from a substrate surface, resulting in an improved substrate treatment process that is particularly advantageous for semiconductor processing applications.
It has been found that during processing of a semiconductor substrate with supercritical carbon dioxide, other phases of carbon dioxide can facilitate and enhance the removal of contaminants from a substrate surface. The introduction into a process chamber of a combination of carbon dioxide phases including supercritical carbon dioxide creates a specific cleaning sequence designed to remove byproduct contaminants while maintaining the structural integrity of the high AR microelectronic devices. The carbon dioxide supply method and system is capable of removing contaminants from progressively smaller device features without causing damage to such features. The process is conducive for removing contaminants within high Aspect Ratio (AR) stacked microelectronic device features, such as, for example, cylindrical DRAM capacitors or Shallow Trench Isolations, and others.
In one aspect of the invention, a method for delivering supercritical and non-supercritical phases of carbon dioxide to create a customized cleaning sequence for the removal of contaminants from a surface of a substrate is provided. The method includes introducing a solvent fluid comprising carbon dioxide mixed with co-solvent additives knows as co-solvents in a supercritical phase into a chamber containing the substrate; transferring contaminants from the substrate surface into the supercritical carbon dioxide phase to form at least a partially spent supercritical phase of carbon dioxide; removing the at least partially spent supercritical carbon dioxide phase from the chamber and simultaneously introducing fresh carbon dioxide, optionally with or without co-solvents dissolved therein, in the supercritical phase into the chamber so as to dilute the spent supercritical carbon dioxide and substantially inhibit the contaminants from precipitating onto the substrate surface; subsequently introducing carbon dioxide in a liquid phase into a chamber; and flowing the carbon dioxide liquid phase over the substrate surface to flush and rinse the substrate surface and thereby remove the contaminants and any co-solvents and additional co-solvents that may have remained on the substrate surface after the cleaning sequence.
In another aspect of the invention, a method for delivering different phases of carbon dioxide to create a customized cleaning sequence for the removal of contaminants from a surface of a substrate is provided. The method includes introducing pure gas phase carbon dioxide to pressurize the chamber to a first pressure below a saturated vapor pressure; removing the pure gas phase and subsequently introducing carbon dioxide in a supercritical phase to increase the chamber pressure from the first pressure to a second pressure higher than the first pressure; introducing a solvent fluid at the second pressure comprising carbon dioxide in a supercritical phase mixed with co-solvents into a chamber containing the substrate; transferring contaminants from the substrate surface into the supercritical carbon dioxide phase to form at least a partially spent supercritical phase of carbon dioxide; removing the at least partially spent supercritical carbon dioxide phase from the chamber and simultaneously introducing fresh carbon dioxide in the supercritical phase without the co-solvents into the chamber so as to dilute the spent supercritical carbon dioxide and substantially inhibit the contaminants from precipitating onto the substrate surface; subsequently introducing pure carbon dioxide in a liquid phase into a chamber; and flowing the carbon dioxide liquid phase over the substrate surface to flush and rinse the substrate surface and thereby remove the contaminants and any co-solvents residually remaining on the substrate surface.
In another aspect of the invention, a supply system for purifying and delivering multiple phases of carbon dioxide to a downstream chamber is provided. The supply system includes a first accumulator positioned between a purification unit and the chamber, the first accumulator comprising saturated liquid phase carbon dioxide and saturated vapor phase carbon dioxide; a second accumulator positioned between the purification unit and the chamber, the second accumulator comprising supercritical phase carbon dioxide; a purification unit positioned upstream of the first and the second accumulators to produce purified carbon dioxide from a bulk tank containing crude carbon dioxide; and a flow network positioned at the outlet of the first and the second accumulators and having a first leg, a second leg, a third leg, a first control valve a second control valve and a third control valve.
Advantageously, the carbon dioxide supply system can be constructed utilizing system components that are commercially available, thus enabling and simplifying the overall assembly of the system and method of use thereof. Aspects of purified carbon dioxide delivery to a process tool can be carried out using standard techniques or equipment.
The objects and advantages of the invention will be better understood from the following detailed description of the preferred embodiments thereof in connection with the accompanying figures wherein:
The above and other features of the invention including various details of construction and combinations of parts, and other advantages, will now be more particularly described with reference to the accompanying drawings and pointed out in the claims. It will be understood that the particular carbon dioxide supply system and method of delivery embodying the invention are shown by way of illustration and not as a limitation of the invention. The principles and features of this invention may be employed in various and numerous embodiments without departing from the scope of the invention.
As used herein and in the claims, all concentrations are expressed as volumetric or mole percentages. As used herein and in the claims, the term “contaminants” refers to solid particles, non-volatile residue (“NVR”), and non-volatile organic residue (“NVOR”), metals and any other byproducts produced as a result of cleaning, etching and rinsing various microelectronic device features, and any remaining co-solvents. Solid particles, also known as inhomogeneous contaminants, typically refers to small (e.g., microscopic) pieces of metal shed by machinery used in the carbon dioxide application and any contaminants generated from the process which is not soluble in carbon dioxide at a given pressure and temperature. Generally, solid contaminants do not dissolve in high pressure or subcritical carbon dioxide. NVR refers to that portion of contaminants that remains following sublimation or evaporation of the carbon dioxide at room temperature and pressure. A portion of this NVR will typically consists of solid particles, such as may have shed from metal surfaces during processing as described above. A further portion of the NVR typically consists of NVOR which is that portion of the NVR that is soluble in carbon dioxide at some pressure and temperature. Examples include aliphatic hydrocarbon-based heavy oils, halocarbons, and particulate matter that is soluble in carbon dioxide under certain conditions but can precipitate out at other pressure and temperature conditions, e.g., during lowering of the chamber pressure at the end of the process. Sources of NVOR include compressor oils and elastomeric materials that have some solubility in liquid carbon dioxide and are commonly found in in the original carbon dioxide or could come from the chamber components such as gaskets and valve seat materials. In clean distribution systems, the majority of the NVR present is commonly in the form of NVOR.
The delivery of purified saturated liquid or vapor carbon dioxide 104 from the first accumulator 101 and the delivery of purified supercritical carbon dioxide 105 from the second accumulator 102 is enabled by a flow network 185 that connects the first and the second accumulators 101 and 102 to the process chamber 111 containing a substrate 110 therewithin. The flow network 185 is positioned at the outlet of the first and the second accumulators 101 and 102. The network 105 includes a first leg 106, a second leg 107 and a third leg 108, a first control valve 109, a second control valve 113 and a third control valve 112.
After vapor phase carbon dioxide flows through heater 131, it may enter the catox reactor 132 at a temperature ranging from about 175° C. to about 400° C. The catox reactor 132 includes a suitable catalyst, such as a precious and/or nonprecious metal. Hydrocarbon molecules, which form the NVOR's in carbon dioxide, react with O2 in the reactor in the presence of heat and the catalyst to form carbon dioxide and water. The catox reactor 132 also helps in breaking down the halocarbons in carbon dioxide in the presence of moisture. Details for removing contaminants from the carbon dioxide by employing the catox reactor 132 are disclosed in U.S. Pat. No. 6,962,629, which is incorporated herein by reference in its entirety.
The purified carbon dioxide vapor is heated in the catox reactor 132 and then flows through a particulate filter 133 (e.g., 0.003 μm pore size) to remove any inorganic and metallic particulates which were not oxidized in the catox reactor 132. The purified carbon dioxide vapor exits the particle filter 133 and then flows through a chiller or a heat exchanger 134 to reduce its temperature by a predetermined amount. In one embodiment, a heat exchanger 134 is employed and comprised of two coiled tube heat exchangers. The cooling can be provided in these heat exchangers 134 through the use of cold water flowing in the outer coil of the heat exchanger tubes.
The carbon dioxide is cooled, condensed and flows with sufficient pressure from the outlet of the heat exchanger 134 into one of two accumulators 101 or 102 for subsequent delivery to a process tool 111. Supercritical carbon dioxide 105 can be achieved in corresponding accumulator 102 by pressurizing and heating accumulator 102 as required. To generate the necessary higher pressures at a given temperature to attain supercritical carbon dioxide in the second accumulator 102, pump 135 can be employed. The independent control of the pressure and temperature of the stored supercritical carbon dioxide phase 105 allows the user to have process flexibility for a variety of different processing conditions. For example, different wafer cleaning processes in semiconductor applications may require different densities of supercritical carbon dioxide, which can be tuned by its pressure and temperature. One batch of wafers might require processing at a supercritical phase of carbon dioxide at 3000 psig and 35 C, while another batch might require the supercritical phase of carbon dioxide to be at 1500 psig and 35 C.
Stored carbon dioxide 104 consists of saturated liquid-vapor carbon dioxide, which can be maintained and stored in accumulator 101 (i.e., a subcritical phase of carbon dioxide) by heating accumulator 101 independent from accumulator 102. The carbon dioxide fluid in the liquid/vapor accumulator 101 will equilibrate to its vapor pressure at that temperature corresponding to the vapor pressure curve 600 of carbon dioxide, as shown in
Each of the accumulators 101 and 102 preferably will have a port to sample the concentration of the stored carbon dioxide 104 and 105 and detect for impurities (e.g., metals and NVOR contaminats). The analysis of the carbon dioxide can be performed by any method known in the art, such as, for example, GC-MS (Gas Chromatogram with Mass Spectrometer) and ICP-MS (Inductively Coupled Plasma Mass Spectrometer). Details of one exemplary sampling method are disclosed in U.S. Pat. No. 7,064,834, which is incorporated herein by reference in its entirety. In-situ sampling for contaminants contained in the stored carbon dioxide 104 and 105 allows an operator to check the quality of the stored carbon dioxide 104 and 105 in accumulators 101 and 102, respectively, before it is delivered into the downstream processing tool 111 and contacts the surface of a substrate 110. In the semiconductor industry, the ability to conduct such in-situ sampling allows detection of impure stored carbon dioxide 104 and 105 before contamination of a batch of semiconductor wafers.
The above purification process of crude carbon dioxide 170 contained in the bulk tank 130 is merely an illustrative embodiment. It should be understood that other means for purifying the carbon dioxide into accumulators 101 and 102 is contemplated. For instance, a filtration system may be employed for purifying other gases besides carbon dioxide. In one embodiment, the purification unit could incorporate a filtration system for purifying gases such as helium, nitrogen, argon and other gases which do not mix with carbon dioxide and which remain in the gas phase at the working pressure of the process tool 111. Such gases could be used as a pusher gas that is introduced into the process tool 111 during the de-pressurization steps 304 and 405 shown in
By separately maintaining the control of temperature and pressure in accumulator 101 independent from the control of temperature and pressure in accumulator 102, the various phases of vapor, liquid and supercritical carbon dioxide are simultaneously available on-demand to create novel cleaning sequences of a substrate 110, as will now explained.
In step 202, the chamber 111 is maintained at a working pressure of at least 1072 psi and a temperature of 31.1° C. During this time, other chemical reagents known as co-solvents can be introduced into the chamber 111 for cleaning or etching of the substrate 110. For purposes of clarity, the co-solvent injection path has not been shown in
When the contaminants have been determined to be sufficiently extracted and removed, and/or when the supercritical carbon dioxide has attained its solubility limit, depressurization of the chamber 111 occurs at Step 203. Typically, the end of step 202 and start of step 203 will be determined experimentally by varying the duration of step 202, as well as pressure, temperature, co-solvents, etc. Once it is determined that a certain set of parameters work for a given process, then when running in production, the time for Step 202 is set in the process “recipe” and the wafer cleaning/etching tool automatically goes to the next step after the time elapses. The spent supercritical carbon dioxide, along with any co-solvent and additives introduced during step 202, is removed in a depressurization procedure. Step 203 shows a linear de-pressurization to a certain pressure level at which point liquid CO2 is flowed in a rinse mode in step 204). Depressurization continues until the pressure in the chamber is below the supercritical point. In one example, the pressure in the chamber is reduced to about 850 psi.
Following the controlled depressurization, step 204 of the cleaning sequence 200 can begin. The supply of fresh supercritical carbon dioxide into chamber 111 stops by closing valve 112, which connects the second accumulator 102 containing supercritical carbon dioxide to the process chamber 111. Valve 113 is opened to allow access to accumulator 101, which contains stored carbon dioxide 104 in the form of saturated liquid carbon dioxide in equilibrium with its vapor phase carbon dioxide. During step 204, saturated liquid carbon dioxide is drawn from the bottom of the accumulator 101 and then directed through leg 107 of flow network 185 into the chamber 111. The liquid carbon dioxide is discharged from the first accumulator 101 at a pressure greater than the pressure in the chamber 111. The liquid carbon dioxide enters the chamber 111 at a relatively low flow rate dictated by the pressure differential between chamber 111 and the accumulator 101 so as to not damage any of the high AR features contained on the substrate 110. The liquid carbon dioxide facilitates removal of contaminants which might have re-deposited onto the surface of the substrate 110 during the de-pressurization step of 203. The liquid carbon dioxide continuously flows over the surface of the substrate 110. The drag force of the liquid carbon dioxide enables the contaminants along the substrate 111 to be moved, thereby allowing the flushing and rinsing of contaminants across the surface of the substrate 110. Additionally, the higher density of the liquid carbon dioxide facilitates the solubility of any re-deposited contaminants into the liquid carbon dioxide. Discharge valve 115 is opened to allow liquid carbon dioxide to flow in a flow-through mode between the accumulator 101 and the chamber 111. In this example, Preferably, the pressure of the liquid carbon dioxide in the first accumulator 101 is sufficiently high to avoid pumping. d into the chamber 111. When liquid carbon dioxide cleaning is completed, valve 113 is closed.
At step 205, the chamber 111 can be vented until the pressure in the chamber reduces to atmospheric pressure. Upon the chamber 111 being vented to atmospheric pressure, the cleaned substrate 110 can be removed from the chamber 111. This embodiment for removal of contaminants contained on a substrate 110 illustrates how successively delivering a combination of supercritical carbon dioxide and liquid carbon dioxide at predetermined times during a process cleaning sequence can provide improved removal of contaminants from the substrate 110 compared to conventional carbon dioxide cleaning processes. Supercritical carbon dioxide is initially introduced to extract contaminants from the high AR features without causing such features to collapse or buckle, while liquid carbon dioxide is subsequently flowed along the substrate at low flow rates to remove any re-deposited contaminants on the surface of the substrate 110 by virtue of the liquid carbon dioxide drag force and higher solubility of the contaminants in the liquid carbon dioxide. Such a synergistic combination of supercritical carbon dioxide followed by liquid carbon dioxide cleaning can improve removal of the contaminants.
When the desired working pressure in the process chamber 111 is achieved, valve 112 is closed. The cleaning sequence at step 302 can now begin utilizing the supercritical carbon dioxide that has been introduced from the previous step 301. The cleaning with the supercritical carbon dioxide occurs in the same manner as described in the cleaning sequence 200 of
When the extraction of contaminants by supercritical carbon dioxide is completed, a bleed and feed dilution procedure occurs at step 303. Valve 112 is open to allow fresh supercritical carbon dioxide to be introduced into the chamber 111, and valve 113 is open to allow spent supercritical carbon dioxide to be removed from the chamber 111. The fresh supercritical carbon dioxide is introduced at a pressure approaching the working pressure, while spent supercritical carbon dioxide is removed from the chamber 111 at approximately the same flow rate at which the fresh supercritical carbon dioxide is introduced. Such processing conditions for the “bleed and feed” allow the chamber 111 to be maintained at a pressure substantially close to the process or working pressure (i.e., the pressure at which supercritical carbon dioxide removed and extracted contaminants in step 302) while successively diluting portions of the spent supercritical carbon dioxide. The resultant pressure profile is shown to be slightly saw-tooth as fresh supercritical carbon dioxide enters the chamber 111 and spent supercritical carbon dioxide exits the chamber 111. The pressure spikes are a result of the pressure in the second accumulator 102 being higher in comparison to the chamber 111. The slightly lower series of drops in pressure are attributed to the removal of spent supercritical carbon dioxide from the chamber 111. This slight predetermined pressure difference that is intentionally maintained between the second accumulator 102 and the chamber 111 is sufficient for the fresh supercritical carbon dioxide to flow into the chamber 111. The average pressure in the chamber remains relatively constant so that the overall “bleed and feed” is conducted at about a constant pressure that is equal to the working pressure in previous step 302.
When the constant pressure bleed and feed step 303 is completed, valve 112 is closed to cease the supply of supercritical carbon dioxide into the chamber 111. Depressurization step 304 can now occur. The process chamber 111 can be vented below the supercritical pressure of carbon dioxide, to a pressure in the range of about 300 psi to about 1000 psi, as shown in step 304, keeping the pressure in chamber 111 below the pressure in the accumulator 101 to allow liquid carbon di-oxide to flow into the chamber. When the pressure in the chamber 111 has been sufficiently depressurized, valve 112 is closed and valve 113 is opened to allow liquid carbon dioxide to flow from the bottom of the first accumulator 101 into the process chamber 111, as shown in the flush and rinse step 305. Step 305 is a flow through rinse in which liquid carbon dioxide flows along the surface of the substrate 110. The liquid carbon dioxide ensures removal of any residual contaminants that may still be present on the surface of the substrate 110 or might have re-deposited during the depressurization step 304.
The density of the liquid carbon dioxide is sufficient to allow the contaminants on the substrate 111 to be dissolved into the liquid carbon dioxide. Additionally, the liquid carbon dioxide has a sufficient drag force that can remove the contaminants disposed along the substrate 111. As a result, submicron contaminants may be removed in the flush and rinse step 305.
Following the flush and rinse step 305, valve 113 is closed to stop the supply of liquid carbon dioxide through leg 107 and into the process chamber 111. The chamber 111 is depressurized to atmospheric pressure as shown in step 306 to vent the remaining liquid and/or supercritical carbon dioxide from chamber 111.
As can be seen, the carbon dioxide supply system 100 with accumulators 101 and 102 and corresponding flow network 185 with valves 109, 112 and 113 allow the pressure profile of each of the steps in the process sequence 300 to be controlled so as to create improved removal of contaminants without damaging the high AR features along the surface of the substrate 110. Supercritical carbon dioxide is first introduced to extract contaminants from small AR features without buckling such features. Liquid carbon dioxide is subsequently introduced to dissolve residual contaminants and also push the contaminants contained along the substrate surface by virtue of the liquid carbon dioxide's drag force. The ability to deliver supercritical and liquid carbon dioxide in a specific process sequence and on-demand during the cleaning sequence 300 is made possible by the supply system 100 of
The gas-phase carbon dioxide is not saturated at the temperature within chamber 111. Accordingly, the elimination of liquid condensation during pressurization of the chamber 111 at step 401 circumvents the surface tension effects inherent in liquid carbon dioxide, which can potentially damage the high AR pattern features on the substrate 110.
The gas-phase pressurization of step 401 can be performed relatively fast, compared to the flush and rinse of step 406, as the pressure in the chamber 111 is maintained below saturation vapor pressure. When delivery of the gas-phase carbon dioxide into the chamber 111 causes the pressure to rise to a predetermined pressure that is below the saturated vapor pressure, valve 109 is closed to stop the flow of the vapor phase carbon dioxide from the top of first accumulator 101, and valve 112 is opened to begin flow of supercritical phase carbon dioxide 104 from the second accumulator 102 into the chamber 111. In one example, the predetermined pressure reaches about 800 psig at temperature of about 31° C. prior to switching from gas-phase carbon dioxide to supercritical phase carbon dioxide to ensure that the pressure in chamber 111 is below the saturation vapor pressure of carbon dioxide.
With valve 109 closed and valve 112 opened, supercritical carbon dioxide 105 pressurizes the chamber 111 from a pressure below saturation pressure to the final working pressure. Supercritical carbon dioxide is introduced through leg 108 of flow network 185 and into the chamber 111. The corresponding pressure rise for this step is shown at step 402.
The balance of the steps for the cleaning sequence 400 is identical to that described in
Accordingly, the process sequence 400 of
Still referring to
It should be understood that other variations to
It should be further understood that the principles of the present invention can be extended to multiple purification and process systems to serve multiple process tools.
While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.
The present application claims priority from U.S. Provisional Application Ser. No. 61/625,265, filed Apr. 17, 2012, which is incorporated by reference herein in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61625265 | Apr 2012 | US |
