This application is related to a co-pending application filed on even date and entitled “Collapsible Fluid Container,” which is incorporated herein by reference.
The present invention relates generally to the field of liquid delivery systems for use in industrial process applications. In particular, the present invention relates to a liquid delivery system that minimizes the formation of gas microbubbles in chemical liquid streams.
In many industrial process applications, fluid containers are employed as a source of process liquids for liquid delivery systems. Oftentimes the fluid containers are fabricated and filled at locations remote from the end-use facility. In such situations, the end-use facility then either directly incorporates the fluid containers into a liquid delivery system or empties the liquid from the fluid containers into a reservoir connected to the liquid delivery system.
In certain industrial process applications, the presence of gas microbubbles in liquid traveling through a liquid delivery system may have harmful effects. For example, when liquids are deposited on a substrate to form a layer, the presence of microbubbles in the deposited liquids may cause defects in the deposited layer or subsequent deposited layers. In the semiconductor industry, for example, a common manufacturing step in producing integrated circuits involves depositing photoresist solution on silicon wafers. The presence of microbubbles in the photoresist solution will typically yield defect sites on the surface of the wafer in subsequent process steps. As features on integrated circuits have continued to become smaller, the presence of microbubbles has posed an increasing danger to the quality of integrated circuits. Moreover, when microbubbles are observed in industrial liquid delivery systems, the systems are often purged until the microbubbles are eliminated, which can result in the wasting of expensive chemical liquids. Thus, it is advantageous to eliminate, or at least minimize, the presence of microbubbles in liquid delivery systems.
Although it is known that the presence of microbubbles in liquids deposited on substrates in industrial process applications can cause defects in subsequent process steps, the mechanism of microbubble formation is not well understood. Given these problems associated with formation of microbubbles, there is a need for a system and a method for storing and delivering liquids that reduces microbubble formation.
The present invention is directed to a method and a system for delivering liquid from a fluid container to a downstream process while inhibiting microbubble formation. The present invention is based on the discovery that microbubble formation in a flow path of a liquid delivery system can be inhibited by preventing a pressure in the flow path from falling below a pressure under which the liquid was equilibrated with gas.
The present invention includes a method for delivering liquid from a fluid container to a downstream process. The method comprises supplying the liquid from the fluid container into a flow path. The liquid is delivered through the flow path to the downstream process while maintaining the liquid at a pressure that inhibits formation of microbubbles in the liquid.
The present invention further includes a system for delivering liquid to a downstream process that minimizes formation of microbubbles in the liquid. The system includes a fluid container for storing the liquid. A flow path communicates with the fluid container. The flow path has an inlet end communicating with the fluid container and an outlet end communicating with a downstream process. A means for increasing a pressure inside the liquid delivery system is included to generally prevent the liquid from being subjected to a pressure that induces microbubble formation.
While the above-identified drawing figures set forth several embodiments of the invention, other embodiments are also contemplated, as noted in the discussion. In all cases, this disclosure presents the invention by way of representation and not limitation. It should be understood that numerous other modifications and embodiments can be devised by those skilled in the art which fall within the scope and spirit of the principles of the invention. The figures may not be drawn to scale. Like reference numbers have been used throughout the figures to denote like parts.
The present invention is directed to both a method and a system for delivering liquid along a flow path to a downstream process while inhibiting microbubble formation.
It is well known that gas can dissolve in liquids in a physical manner, without chemical reactions or interactions. Gas that dissolves in liquid without undergoing chemical reactions or interactions may come out of solution and form microbubbles if the solubility of the gas in the liquid decreases. The total volume of gas that will dissolve in a liquid under equilibrium conditions depends upon the composition of the liquid, the composition of the gas, the partial pressure of the gas, and the temperature. If the composition of the liquid and the gas is fixed, and the temperature remains constant, the solubility of a gas in the liquid is directly proportional to the pressure of the gas above the surface of the liquid. Unless otherwise specified, the term “gas” is intended herein to include atmospheric air, as well as any other gas or combination of gases.
The liquid in fluid container 14 has a volume of gas dissolved in it proportional to an equilibrated pressure, Peq, which is the pressure under which gas is exposed to a liquid and becomes generally equilibrated with the liquid. Assuming the liquid is exposed to the gas at Peq for a sufficient period of time, the liquid becomes generally saturated with dissolved gas. In many industrial process applications, Peq will be equal to atmospheric pressure.
As shown in
A drop in the pressure of a saturated liquid flowing through a liquid delivery system results in gas microbubbles forming in the liquid. The term “microbubble” herein is intended to include both (1) gas bubbles that are perceivable to the human eye without magnification and (2) gas bubbles that are too small to be perceived without magnification or other detection means. In the liquid delivery system of
Certain embodiments of the present invention may include a pump in the flow path to meter and/or assist the flow of liquid through the flow path.
In many industrial process applications, however, it may not be practical to elevate the fluid container relative to the flow path. Thus, an additional embodiment of the present invention is a method and a system for mimicking the effects of positive hydraulic head without actually elevating the fluid container. The method involves applying pressure to the liquid inside the fluid container to increase the pressure of the liquid. The pressure may be applied in any manner that elevates the pressure of the liquid inside the fluid container.
If headspace gas is present inside the fluid container when Pi is made greater than Peq, the increased pressure will drive additional gas into solution and microbubbles may form if the pressure subsequently falls below Pi. Thus, when Pi is greater than Peq, fluid container 14 is preferably free of headspace gas. As such, in one embodiment of the present invention, the fluid container preferably has no headspace. Accordingly, the zero headspace fluid containers and filling techniques in U.S. patent application Ser. No. 10/139,185 (Publication No.2003/0205285) filed on Nov. 6, 2003 and entitled “Apparatus and Method for Minimizing the Generation of Particles in Ultrapure Liquid” are incorporated by reference. Additional means for generally achieving zero headspace are discussed later in further detail.
To test the ability of the present invention to inhibit microbubble formation, the following experiments were conducted. All of the experiments involved circulating a liquid saturated with gas through a closed liquid delivery system, while subjecting the liquid to different pressure conditions, to determine whether pressure changes influence microbubble formation in liquid flow paths.
Liquid delivery system 40 differs from a typical liquid delivery system because instead of delivering liquid through a flow path from a fluid container to a downstream process, the liquid is circulated through the delivery system and returned to the fluid container. As such, liquid delivery system 40 constitutes a “closed” system. Liquid delivery system 40 also constitutes a “closed” system because, as described later, it may be configured to prevent both the infiltration of outside gas and the escape of gas from inside the system. The closed nature of liquid delivery system 40 allowed the effects of pressure changes on microbubble formation to be studied in the context of a fixed quantity of gas.
To make it easier to observe any outgassing of dissolved gas in liquid delivery system 40 due to pressure changes, it was desirable to use a test fluid capable of dissolving a substantial amount of gas. The reason for desiring such a test fluid is that the greater the total amount of gas dissolved in the test fluid, the larger the potential driving forces for outgassing, especially given the high likelihood that non-uniform concentration gradients will form in the circulated test fluid. If such non-uniformities were not allowed to exist, the solubility of gas in the test fluid would not matter. However, since non-uniformities can exist within test apparatus, the larger the solubility of the test fluid, the greater the mass transfer driving forces, which makes it easier to observe microbubble phenomena. Thus, it was desirable to select a test fluid for use with liquid delivery system 40 that had a high gas solubility.
In many solvents, for example water and isopropyl alcohol, an inverse correlation exists between the amount of gas dissolved in the solvent at equilibrium and the surface tension of the solvent. Since published data for surface tensions and solubility constants is not readily available for solvents other than water, the equilibrium solubilities for six organic solvents were calculated. The six organic solvents selected were isopropyl alcohol (IPA) and five photoresist casting solvents used in the microchip manufacturing industry. The gas solubilities for these solvents were calculated by first determining the surface tension for each solvent and then calculating the gas solubilities based on these surface tensions.
The resulting surface tensions and equilibrium gas solubilities for the six solvents are shown below in Table 1, with water data included for comparison. Using the density of the solvents, the molecular weights of the solvents, and the Sugden parachor, the McLeod-Sugden method was employed to calculate the surface tensions shown in the second column of Table 1. The equation used for these calculation is given below:
where σ is the surface tension in dynes/cm, [P] is the Sugden parachor of the solvent atom, ρ is the solvent density in g/ml, and M is the molecular weight of the solvent in g/mole.
The surface tensions were then used to calculate the solubility of gas in each solvent. Atmospheric air was selected as the gas for study because, given its ubiquitousness, it is the gas composition most likely to be dissolved in industrial liquid delivery systems. The composition of air was treated as a weighted average of 21% oxygen and 79% nitrogen. The resulting solubilities of air in each solvent are shown in the third column of Table 1, and are expressed in the form of the Ostwald coefficient for each solvent at 20° C. and 1 atm. The Ostwald Coefficient expresses the maximum amount of air that will dissolve in each ml of solvent under the above conditions.
According to the results in Table 1, air is most soluble in isopropyl alcohol (IPA). In consideration of this finding, a recipe of 70% IPA and 30% ethyl lactate (EL) was selected for the test fluid. Similarly, this composition was also selected because it had been observed to be particularly susceptible to microbubble formulation when dispensed from liquid delivery systems. A 500 ml volume of the test fluid was prepared in an open container by mixing 150 ml of EL with 350 ml of IPA in an environment of atmospheric pressure and approximately 20° C. The test fluid was maintained under these conditions for 18 hours to reach equilibrium with the environmental conditions. As such, the Peq of the test fluid was atmospheric pressure. Under these conditions, it is reasonable to presume that the test fluid became saturated with air. Thus, from the information in Table 1, a prorated solubility of 0.2985 ml air per ml of test fluid was calculated, meaning that approximately 0.180 grams of air should have been dissolved in the 500 ml volume of test fluid.
The particular fluid container 42 used in the experiment comprised a 500 ml intravenous bag measuring ten-inches tall by five-inches wide when laid flat. An intravenous bag was selected for use in the experiment because of its ability to be filled under near-zero-headspace conditions, thereby reducing the amount of headspace air initially trapped inside liquid delivery system 40.
As shown in
The particular pump 50 used in the experiment was a two-stage Mykrolis IntelliGen pump, although an Iwaki Tube-Phragm pump or other similar type of pump could have been used. Pump 50 includes a feed pump 64, a dispense pump 66, a filter 68, and a flow path 70. Pump 50 is a two-stage pump, in which feed pump 64 and dispense pump 66 are connected by flow path 70. Filter 68 is positioned along flow path 70 and is connected to vent line 56. Dispense pump 66 is connected to return line 58 and purge line 54.
Quick connects 44 and 46 are attached to ports 60 and 62 of fluid container 42. The quick connects used in the experiment were CPC Quick Connects, which are commercially available from the Colder Products Company, St. Paul, Minn. Using the quick connects, fluid container 42 was filled with 500 ml of the test fluid. Quick connect 44 was then connected to feed line 48. Prior to connecting fluid container 42 to liquid delivery system 40, the flow path of the system was generally filled with test fluid. As a precaution, the flow path was vented before running the experiment to purge any air trapped inside the system. Special care was taken to purge any air residing inside feed pump 64, dispense pump 66, and filter 68. As part of this venting process, quick connect 46 and an adjacent portion of attached return line 58 were detached from fluid container 42 and elevated above pump 50. Pump 50 was then run until generally all the air was vented from the system via quick connect 46. At that point, quick connect 46 was connected back to inlet port 62 of fluid container 42, thereby “closing” liquid delivery system 40. Except for a small and finite volume of air introduced upon reconnecting quick connect 46 (which was allowed to settle to the top of fluid container 42) liquid delivery system 40 was essentially free of headspace air.
After achieving a closed system, liquid delivery system 40 was configured so that a subatmospheric pressure would not develop inside the system. To ensure that the pressure of the test fluid did not fall below atmospheric pressure at any location inside the flow path of liquid delivery system 40, fluid container 42 was hung from the ceiling to yield approximately 5 feet of hydraulic head advantage relative to pump 50. Due to the positive head, microbubble formation was not expected since the pressure of the test fluid would generally not fall below the atmospheric pressure at which the test fluid had been saturated with air. In other words, microbubble formation was not expected because Pf would not generally fall below Peq, As such, the test fluid was not expected to reach a super-saturated state.
Even so, gas traps 72, 74, 76, and 78 were created in the form of tube loops located in the flow path of liquid delivery system 40 to act as traps for microbubbles, thereby making microbubble observation easier. Pump 50 was set to dispense its maximum volume, 6 ml, over a 6 second span and then recharge, purge, and vent in preparation for the next 6 ml/6 sec dispense. This resulted in a 36 second pump duty cycle with 6 seconds of dispense though lines 54, 56, and 58 and 30 seconds of recharge. In so far as steady state could be achieved in a pulsing duty cycle system, particle counts were recorded by particle counter 52 after the pump had run for 50 cycles. Particle counter 52, as used in the experiment, was a LiQuilaz S05 liquid particle counter commercially available from Particle Measuring Systems, Boulder, Colo.
No microbubbles were observable to the naked eye at any of gas traps 72, 74, 76, and 78. The distribution of particles and microbubbles in the test fluid, as recorded by particle counter 52, also indicated that microbubbles were not forming.
Liquid delivery system 40 was then reconfigured so the pressure would fall below atmospheric pressure on the suction-side of pump 50 between feed pump 64 and fluid container 42. This pressure environment was achieved by placing the liquid-filled fluid container 42 at floor level, approximately three feet below the level of pump 50, thereby creating approximately three feet of negative hydraulic head. In this configuration, Pf falls below the atmospheric Peq on the suction-side of pump 50 between fluid container 42 and feed pump 64. A subatmospheric Pf forms because, given the negative head, pump 50 must establish a subatmospheric pressure to induce test fluid to flow from fluid container 42 up to pump 50.
After reestablishing flow of the test fluid through liquid delivery system 40, newly-formed microbubbles were observed by the naked eye in gas trap 72 between feed pump 64 and filter 68. Newly-formed microbubbles were also observed in downstream gas traps 74, 76, and 78. The distribution of particles and microbubbles in the test fluid, as recorded by particle counter 52, provided further indication of microbubble formation. This particle/microbubble distribution is shown in
To test whether the microbubble formation was reversible, fluid container 42 was moved from the floor back up to the ceiling, reestablishing approximately five feet of positive hydraulic head. After elevating fluid container 42, the microbubbles residing in the gas traps dissolved back into the test fluid, leaving no observable microbubbles in liquid delivery system 40. As such, the above positive and negative hydraulic head experiments indicate that a subatmospheric pressure in a liquid delivery system contributes to the formation of microbubbles. That is, the experiments indicate that a Pf below Peq contributes to microbubble formation.
In the negative head experiment, the possibility that the observed microbubbles were caused by the subatmospheric pressure on the suction-side of pump 50 sucking air into the flow path can be dismissed. The observed microbubble formation was completely reversible when the fluid container was elevated to establish a positive hydraulic head. If infiltration of air into liquid delivery system 40 had occurred, the air would not have completely dissolved back into solution since the test fluid was generally saturated with air prior to filling the fluid container. As such, the test results indicate that the total mass of air in the system did not increase, which eliminates the possibility that air infiltration induced the microbubble formation.
Therefore, the above experiments demonstrate that a region of subatmospheric pressure (which in this case is below Peq) in a liquid delivery system contributes to the formation of microbubbles. Thus, in one embodiment of the present invention, a subatmospheric pressure is prevented from occurring in a liquid delivery system by elevating the fluid container. By elevating the fluid container relative to the other parts of the liquid delivery system, a positive hydraulic head is created which acts as a buffer to absorb pressure decreases without the pressure reaching subatmospheric levels.
Since in many industrial process applications it may not be practical to elevate the liquid source, systems and methods for applying pressure to the fluid container to mimic the effects of elevation were studied.
The system and method of
The intravenous bag was filled with 500 ml of a 70% IPA and 30% EL test fluid equilibrated at 1 atm and 20° C. to become saturated with dissolved air. The filled intravenous bag was then turned upside down and the residual air in the bag was ejected out of ports 60 and 62. As in the positive and negative head experiments, liquid delivery system 40 was initially purged of any residual air trapped inside the system. The system was then closed off so air could neither exit or infiltrate the system. Flow was induced in the system using pump 50. Microbubble formation was subsequently observed in the flow path of liquid delivery system 40.
As shown in
Collapsible liner 94 is preferably filled with liquid under zero headspace conditions to inhibit subsequent microbubble formation due to pressure drops. Furthermore, collapsible liner 94, is preferably constructed from a flexible material that is impermeable to gas transfer, thus preventing air from infiltrating interior volume 70 when intermediate area 96 is pressurized. When collapsible liner 94 is filled with liquid, air from air source 106 may be supplied through air supply line 104 to pressurize intermediate area 96 and displace liquid from interior volume 100 into dispense line 102. The liquid displaced into dispense line 102 has a pressure greater than atmospheric pressure, decreasing the likelihood of downstream microbubble formation. Fluid container 90 has the additional feature that it eliminates the need for a downstream pump in certain liquid delivery systems since the fluid container is capable of inducing flow without the assistance of a downstream pump.
Collapsible liner 110 may be formed by folding over a flexible sheet material to form a top film 112 and a bottom film 114. The peripheral edges of films 112 and 114 are sealed to form an interior volume 116 for holding liquids. The sealed together portions of films 112 and 114 are represented by hatched lines in
A fitment 118 may be sealed to collapsible liner 110 to define a port communicating with interior volume 116. Such a port may be used to supply liquid into interior volume 116, and may also be used to evacuate air trapped inside interior volume 116. In addition, fitment 118 may be used to dispense liquid from interior volume 116 into a flow path, or alternatively, additional fitment(s) may be included for such purposes. Moreover, each fitment may define a plurality of ports and may be located anywhere on the fluid container capable of communicating with the interior volume. Another feature of collapsible liner 110 is that it can hold a variable amount of liquid, thereby providing a versatile package for use in industrial process applications.
Like collapsible liner 110, collapsible liner 120 has a top film 112 and a bottom film 114 that define an interior volume 116 for holding liquid, with the peripheral portions of films 112 and 114 sealed together as represented by hatched lines in
When a generally zero headspace condition is desired inside interior volume 116, interior volume 116 is first filled with a quantity of liquid sufficient to completely fill main chamber 122. Collapsible liner 120 is then preferably oriented so auxiliary chamber 124 has the highest elevation. This orientation encourages headspace gas to congregate inside auxiliary chamber 124. The inclusion of tapered walls 126 and 128 further facilitates this gas migration. As shown in
The auxiliary chamber may be sealed off after connecting the main chamber of the interior volume to the flow path, thereby removing any headspace gas introduced as a result of the connection. Collapsible liner 120, thus, provides a convenient means for obtaining a generally zero-headspace fill.
Collapsible liner 140 has an interior volume 142 defined by a top film 144 and a bottom film 146, which are sealed together as represented by the hatched lines in
Collapsible liner 140 may be formed pursuant to the methods described above for collapsible liner 120. Portions of films 144 and 146 are sealed together to form interior volume 142, with the hatched lines in
Fitments 154 and 156 are sealed to collapsible liner 140 to define ports communicating with interior volume 142. Fitment 154 is located at an end of dispensing chamber 148 opposite collection chamber 150, and fitment 156 is located at an end of collection chamber 150 opposite dispensing chamber 148. In other embodiments, any number of fitments with any number of ports may be sealed to collapsible liner 140 at any location or locations that allow access to interior volume 142.
Similar to collapsible liner 120, collapsible liner 140 may be configured to achieve a zero headspace condition. Passage 152 may be sealed off to terminate communication between dispensing chamber 148 and collection chamber 150 and isolate headspace gas within collection chamber 150. A zero-headspace condition may be obtained inside dispensing chamber 148 using the methods described above for collapsible liner 120. For example, as shown in
Fitments 154 and 156 may be mated, respectively, with an inlet end of a flow path and an outlet end of a flow path, thereby placing each fitment in communication with the flow path. In this configuration, liquid in dispensing chamber 148 may be dispensed into the flow path and liquid from the flow path may be collected in collection chamber 150.
The liquid collected in collection chamber 150 may be drained into dispensing chamber 148 by unsealing passage 152. If the liquid is to be dispensed back into the flow path, the liquid should preferably be allowed to equilibrate within collection chamber 150 before being drained back into dispensing chamber 148, thereby reducing the amount of dissolved gas in the liquid and discouraging microbubble formation.
An additional feature of the present invention is its ability to discourage cavitation of a fluid traveling through a liquid delivery system. This characteristic is important because cavitation can lead to microbubble formation in liquid delivery systems. Cavitation occurs when a liquid flows into a region where its pressure is reduced to its vapor pressure, causing the liquid to boil and form gaseous vapor pockets. For example, for water at 20° C., the vapor pressure for the water is approximately 0.023 atm, meaning that when the pressure of the water drops to approximately 0.023 atm, the water begins to boil and develop gaseous vapor pockets. Subatmospheric pressures in liquid delivery systems, for example on the suction-side of a pump, may be sufficient to induce cavitation. Thus, by ensuring that the pressure of water in a liquid delivery system remains above atmospheric pressure, cavitation-induced microbubble formation is inhibited. In general, the present invention helps prevent cavitation-induced microbubble formation in liquids that have a vapor pressure below atmospheric pressure. In addition, the present invention also helps prevent cavitation of liquids that have vapor pressures in excess of atmospheric pressure, depending upon the proximity of the particular vapor pressure to atmospheric pressure.
Although the present invention has been described with reference to preferred embodiments, workers skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention.