Chemical mixing apparatus, system and method

Abstract

There is disclosed a system and method of formulating a batch comprising at least two ingredients. An embodiment is disclosed wherein a feedforward algorithm can be used to control the target blend. Subsequently a feedback closed loop control loop algorithm is provided for a multivariant blend. Use of this approach allows for a continuously autoreplenished and controlled blend. Another enhancement disclosed is the ability to control via feedforward and feedback algorithms a fast responding control mode which allows for the elimination of a tank. This approach will allow for “one pass” blending with control. Feedback control can be provided via process indicators or through analytical or parametric methods. The controller implements an automated fault detection and correction system, thereby identifying necessary maintenance prior to failure. If failure does occur the signature recognition allows rapid analysis and correction thus maximizing tool availability.

Description

PRIORITY CLAIM

This application hereby claims priority to, and incorporates herein by reference in its entirety, U.S. Provisional Patent Application Ser. No. 60/586,189 entitled “Chemical Mixing Apparatus, System and Method”, filed Jul. 8, 2004; U.S. Provisional Patent Application Ser. No. 60/586,968 entitled “Point of Use Chemistry Blending Control”; U.S. patent application Ser. No. 10/877,705 entitled “Chemical Mixing Apparatus, System and Method”, filed Jul. 9, 2004; and PCT Patent Application No. PCT/US04/41053 filed Dec. 9, 2004.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates in general to an apparatus, system and method for mixing chemicals. It more particularly relates to such an apparatus, system and method for mixing ingredients in a precise manner in accordance with a given recipe.

2. Description of the Relevant Art

This section describes the background of the disclosed embodiment of the present invention. There is no intention, either express or implied, that the background art discussed in this section legally constitutes prior art.

There have been a variety of different types and kinds of apparatus, system and methods for mixing ingredients. For example, reference may be made to the following U.S. patents and patent application, each of which is incorporated herein by reference in its entirety:

	Patent No.	Inventor	Issue Date
	4,363,742	Stone, Milton	Dec. 14, 1982
	5,340,210	Patel, et al.	Aug. 23, 1994
	5,348,389	Lennart Jönsson, et al.	Sep. 20, 1994
	5,522,660	O'Dougherty, et al.	Jun. 04, 1996
	5,632,960	Ferri, J. R., et al.	May 27, 1997
	5,874,049	Ferri, J. R. et al.	Feb. 23, 1999
	5,924,794	O'Dougherty, et al.	Jul. 20, 1999
	6,120,175	Tewell, Stanley	Sep. 19, 2000
	6,290,384	Pozniak, et al.	Sep. 18, 2001
	2004/0100860	Wilmer, et al.	May 27, 2004

Currently, many manufacturing processes require the use of blended chemical compositions to treat parts during different steps of the process. Historically, these blended compositions have depended upon the input chemical control devices to achieve the desired mixture, then the mixture is tested in line for acceptable use. In some cases, an external analytical instrument or laboratory is used to confirm the blended mixture. In some other cases, an in-line test on the product is used.

While these methods may be successful for some applications to assure quality of process, they each may employ unwanted and undesirable delays. If the test fails, draining and refilling the chemistry subsequent to the test results may be required. This may result in unacceptable delays, additional costs and additional cycle time to the manufacturing process in certain applications.

SUMMARY

An apparatus, systems and methods for mixing chemicals is disclosed herein. The apparatus and systems disclosed herein may be applicable to both batch processing of chemical compositions and point of use or single pass compositions.

In one embodiment, a method of formulating a batch, includes: admitting at least two ingredients to a given size container to a fraction of the full container volume for a desired batch; determining the quantities of each ingredient in the container; calculating the ratio of the target quantity to the determined current quantity for at least one of the ingredients; calculating the next quantity of the at least one ingredient by multiplying the target quantity of the ingredient by said ratio to determine a corrected quantity; admitting the corrected quantity of the ingredient to the admixture in the container; admitting a quantity of another ingredient to adjust the proportion of ingredients to the target formulation; repeating steps B through F until the container is filled to the desired quantity of the batch; and repeating steps B through F if the container is depleted by use. The above described method may also be embodied on a computer readable medium and in a controller of a chemical mixing system.

In another embodiment, a method of producing a composition, the composition including two or more components, includes: obtaining a total flow rate for the composition; determining the flow rate for each component, wherein the sum of the flow rates for each component equals the total flow rate, and wherein the flow rate of each component is determined based on a predetermined composition formulation; initiating flow of each component into a mixing area, wherein each component is flowed at the determined flow rate, and wherein the components are combined in the mixing area to produce the composition; transferring the composition into an analyzer, wherein the analyzer is configured to measure the concentration of one or more components of the composition; and transferring the composition to a process tool if one or more components of the composition is within a predetermined concentration range.

In one embodiment, a method is disclosed for determining the homogeneity of a blend by use of a statistical approach such as Mahalanobis distances or some other appropriate data analysis. An analytical method such as RAMAN or NIR provides spectral information on a chemical system. This information is converted to a format that can be analyzed using techniques like Root mean square groupsizes computed from Mahalanobis distances. This results in a measure of the homogeneity of the blend. The approach is a statistical analysis of the analytical data and therefore can be applied to all analytical approaches to determine whether a system is well blended. This could be applied to liquid, solid or gas systems.

BRIEF DESCRIPTION OF THE DRAWINGS

Other objects, features and advantages of the invention will appear on reading the following description, given as a non-limiting example, and made with reference to the appended drawings in which:

FIG. 1 is a diagrammatic view of a chemical mixing system;

FIG. 2 is a diagrammatic front elevational view of a tank being filled using a fractional fill method;

FIG. 3 is a flow chart of a fractional fill mixing method;

FIGS. 4 and 5 are flow charts of another fractional fill mixing method;

FIG. 6 is a listing of some compositions that may be prepared using the chemical mixing systems described herein;

FIG. 7 depicts an embodiment of ingredient control devices;

FIG. 8 depicts an alternate embodiment of ingredient control devices;

FIG. 9 depicts a flow chart of a pre-weigh algorithm;

FIG. 10 is a block diagram of a controller;

FIG. 11. is a diagrammatic view of a point of use chemical mixing system;

FIG. 12 is a flow chart of a general control method for mixing chemicals; and

FIG. 13 is a flow chart for a general control method of mixing chemicals for both batch and continuous applications.

FIG. 14 is a flow chart for automatic fault detection and control for the chemical mixing system.

FIGS. 15-16 depicting methods of performing homogeneity measurements and equipment design for homogeneity measurements.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the drawing and detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention as defined by the appended claims.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

According to certain embodiments of the invention, there is provided a system and method of formulating a composition comprising at least two ingredients. The ingredients are admitted to a container to partially fill it. The quantities of the ingredient in the container are determined, and a ratio of a target quantity to the determined current quantity for at least one ingredient is calculated. The next quantity of that ingredient to be admitted to the admixture is calculated by multiplying the target quantity by the calculated ratio to determine a corrected quantity. The corrected quantity of the ingredient is admitted to the admixture, and a quantity of another ingredient is admitted to the admixture to adjust the proportion of ingredients to the target formulation. These steps may be repeated until a desired quantity of a predetermined composition is produced.

According to certain embodiments of the invention, there is provided a fractional fill mixing apparatus, system and method for mixing ingredients. In one disclosed embodiment, the fractional fill apparatus, system and method includes a container for holding ingredients, an in-line analytical instrument for measuring the concentration or quantity of ingredients disposed within the container, and an ingredient supply control device for dispensing ingredients into the container. A controller is operatively connected to one or more ingredient supply control devices and the analytical instrument. The controller further employs a fractional fill algorithm for admitting at least two ingredients to the container to a fraction of the full volume for a desired batch.

According to certain embodiments of the invention, a controller executes the fractional fill mixing algorithm to cause an initial fraction of the total volume of the container to be filled in the filling sequence. This fractional volume is recirculated to assure a homogeneous mixture, and the in-line analytical instrument determines the constituent parts of the mixture and communicates the information regarding the current mixture to the controller. The controller executing a fractional fill mixing algorithm, adjusts the ingredient supply control device in a manner that corrects errors between the actual values and the desired values of the mixture in subsequent fractions or portions of the total volume of the mixture. The resulting blend is the desired mixture and no additional testing is required for many applications. Referring now to the drawings and, more particularly, to FIG. 1, there is shown a fractional fill mixing apparatus or system 10, which is constructed in accordance with an embodiment of the present invention, and which is used to mix two or more ingredients in a tank or container 12. An analyzer or analytical instrument 14 is adapted to measure the quantities of each ingredient in the container 12. An ingredient supply control device shown generally at 16, controllably dispenses two or more ingredients into the tank or container 12. The ingredient supply control device 16 dispenses ingredients through a plurality of ingredient supply inlets, such as first ingredient supply inlet 18, second ingredient supply inlet 20, and third ingredient supply inlet 22. Each ingredient supply inlet 18, 20, and 22, is connected in fluid communication with a plurality of ingredient supplies (not shown). The manifold 24 receives the plurality of ingredients from ingredient supply control device 16. The ingredients then flow from the manifold 24 to the container 12.

As shown in FIG. 2, in use, according to a fractional fill mixing algorithm, the tank or container 12 initially may contain a residual volume of one of the plurality of ingredients to be mixed, as indicated by volLowLev 200. The low level of the tank is, therefore, indicated generally at 210 when a residual volume of one of the ingredients is present in the tank 12.

According to an embodiment of the invention, the tank 12 is then fractionally filled seriatim through two or more fractional or partial filling sequences, the volume of each are indicated at 202, 204, 206, and 208, respectively. As indicated in FIG. 2, for example, a fractional filling sequence generally may include four fractional filling sequences volFrac1, volFrac2, volFrac3, and volFrac4. It should be noted that the tank or container 12 may have additional volume capacity above the high level point 212 (not shown). Thus, the high level point 212 indicates the level that will be achieved when the fractional fill sequence is complete but not necessarily indicate the maximum capacity of the tank 12.

As shown in FIG. 3, the fractional fill mixing method begins in block 27. The fractional fill mixing method admits at least two ingredients to the container 12 to a fraction of the full container 12 volume for a desired composition (28). The method then determines the quantities of each ingredient in the container (30). The quantities of each ingredient measured in the container 12 may be in percent by weight or in percent by volume. The method then calculates the ratio of the determined current quantity for at least one of the ingredients to the target quantity for the desired composition (32). The method then calculates the next quantity of at least one ingredient by multiplying the target quantity of the ingredient by the ratio calculated in block 32 to determine a corrected quantity (34). The method then directs the ingredient supply control device 16 to admit the corrected quantity of the ingredient to the admixture in the container 12 (36). The method then admits a quantity of another ingredient to adjust the proportion of the ingredients to the target formulation (38). The process as shown in blocks 30, 32, 34, 36, and 38 is repeated until the container is filled (42) to the desired quantity of the composition. When the container 12 is filled to the desired quantity of the composition, the process terminates (44).

Considering now the method as just described in greater detail, and with reference to. FIG. 2, the method includes determining a desired fractional filling sequence of quantities of fractional fills to be performed. For example, FIG. 2 shows a tank 12 that will contain the admixture and ultimately the final desired batch to be created from the method. FIG. 2 shows a plurality of volume levels for subsequent fractional fill sequences. In the present example, four fractional filling sequences are to be performed. The first fractional filling sequence fills the container 12 to approximately 50% of its volume as shown by area 202 and this volume is indicated as volFrac 1. The partial fill volume is equal to 50% in this example including the residual volume as indicated by volLowLev 200. The residual volume is the volume of a residual ingredient already present in the tank 12 before the fractional fill method is commenced. There may or may not be a residual volume, as it depends on the user requirements. The residual volume of the ingredient in tank 12 is normally the same ingredient as one of the ingredients that will form part of the current batch. The second fractional fill fills the container an additional 25% of volume as indicated by the area 204 where the volume for this fractional fill is represented by volFrac 2. The third and fourth fractional volumes, volFrac 3 and volFrac 4 indicated by 206 and 208, respectively, each fill the container an additional 12.5% until the container is approximately full as indicated by arrow 212.

The fractional volumes and percentages just recited are for example purposes only and could be modified as desired to achieve various filling sequences as will become apparent to those skilled in the art. More or less fractional filling sequences may be used to achieve a desired volume of a predetermined composition. For example, instead of four fractional filling sequences, three fractional filling sequences could be used where each fractional volume sequence could include 33% or one-third of the approximate container volume. For sake of example only, subsequent discussions of the fractional filling method will utilize four fractional filling sequences. The first fractional filling sequence, volFrac 1, in combination with the residual volume, volLowLev, will be equal to 50% of the total composition volume, the second fractional filling sequence, volFrac 2, will contain 25% of the total composition volume, and the third and fourth fractional filling sequences, volFrac 3 and volFrac 4, will each contain 12.5% each of the total composition volume as described previously.

The total volume of the composition in container 12 is represented by the variable totalVol which equals (VolLowLev+volFrac 1+volFrac 2+volFrac 3+volFrac 4). totalVol may also be represented by (chem1TotalVol+chem2TotalVol+diwAddedVol). chem1TotalVol represents the total volume of the first ingredient in the composition. chem2TotalVol represents the total volume of the second ingredient in the composition. DiwAddedVol represents the volume of the third ingredient, typically deionized water, in the composition. It should be noted that diwAddedVol represents the third ingredient and normally is deionized water but may be any other ingredient that is desired to be part of the batch. For the sake of clarity for subsequent examples, the residual volume of the admixture in container 12 is assumed to be the same ingredient as diwAddedVol, the third ingredient of a desired batch, so that when diwAddedVol and VolLowLev are combined, the total volume of the third ingredient results.

The fractional fill mixing method then begins by filling the container to the first fractional fill percentage in the sequence. In our example, this is 50% as represented by VolFrac1 202, as best shown in FIG. 2. The actual volume of the first ingredient to meet the requirements for the current fractional fill sequence is then calculated. This volume is represented by chem1FracVol. chem1FracVol is equal to chem1TotalVol·pourUp1Frac where pourUp1Frac is a fractional fill percentage of the first fill sequence, in the present example, 50%. chem2FracVol is calculated using a similar formula.

Calculation of the total volume of the first ingredient must then be calculated as represented by chem1TotalVol. chem1TotalVol is defined as chem1Ratio·x where x is an intermediate variable. x is defined as TotalVol÷(chem1Ratio+chem2Ratio+diwRatio). chem1Ratio and chem2Ratio are defined as the ratio of the volume to be filled for the first and second ingredients, respectively. diwRatio is a ratio of the volume to be filled for the third ingredients.

The volume of the third ingredient added to VolLowLev to obtain totalVol is defined as diwAddedVol which equals (diwRatio·x)−VolLowLev.

The fractional fill mixing method next includes calculating the target quantity of one ingredient based on the target volumetric blending ratio and the supply concentration of the ingredient. The target quantity of one ingredient is referred to as concChem1, which is defined as (chem1Ratio·bulkChem1)÷(chem1Ratio+chem2Ratio+diwRatio). Where chem1Ratio and chem2Ratio and diwRatio represent the ratios of the volume to be filled for the first, second, and third ingredient, respectively, for the current fractional fill sequence. BulkChem1 represents the supply concentration of the first ingredient. The target quantity of the other ingredients are calculated using similar formulas where the numerator of the above equation is replaced with the ratio and concentration of the bulk ingredient supply from the respective ingredient being calculated. Now that chem1FracVol has been calculated, chem2FracVol and diwFracVol are also calculated as just described.

At this point in the method for fractional fill mixing according to one embodiment of the invention, the first fraction is poured by controller 26 sending a signal to ingredient supply control device 16 to dispense the volume of ingredient represented by chem1FracVol then to dispense the volume of ingredient represented by chem2FracVol and finally to dispense the volume of chemical as represented by diwFracVol.

Now that the first fractional fill has been admitted to container 12, subsequent fractional fill sequences must be calculated and admitted to container 12. To perform the remaining fractional fill sequences, an ideal chemical fraction, such as idealChem1Frac, may be calculated. An ideal chemical fraction may be calculated for each ingredient to be admitted to container 12. By way of example, idealChem1Frac is defined as (chem1TotalVol·pourUp2Frac) where chem1TotalVol represents the total volume of the first ingredient to meet the requirements for the current fractional fill sequence and pourUp2Frac is the subsequent fractional fill percentage in the sequence. For example, since this is the second correction fill sequence, pourUp2Frac in this example would now be equal to 25%. Other ideal chemical fractions may also be calculated for each ingredient by using a similar formula where chem1TotalVol is replaced with the total volume of the other ingredient being evaluated.

Next, the actual volume of each ingredient to meet the requirements for the current fractional fill sequence must be calculated. By way of example, the actual volume of the first ingredient to meet the requirements for the current fractional fill sequence is represented by chem1FracVol which is defined as (idealChem1Frac·concChem1)÷chem1Val where chem1Val is the measured quantity or concentration of the first ingredient in the batch. A similar formula may be used to calculate the actual volumes of the other ingredients to be added to the admixture during this fractal fill sequence where the theoretical quantity/concentration of the other ingredients, ideal chemical fractions, and measured quantities/concentrations may be replaced in the appropriate portions of the above formula.

The method further includes calculating the difference between the ideal and actual volume of the first ingredient. This is calculated by subtracting chem1FracVol from idealChem1Frac. The same formula is used for the second ingredient to calculate chem2FracDelta using its actual volume to meet the requirements of the current fractional fill sequence and ideal chemical fraction.

The actual volume of the third ingredient to meet the requirements for the current fractional fill sequence may use a different formula. diwFracVol is equal to (diwAddedVol·pourUp2Frac)+chem1FracDelta+chem2FracDelta where diwAddedVol is the volume of the third ingredient at its VolLowLev to obtain total volume for the third ingredient. This, as discussed above, assumes that VolLowLev, which represents the residual volume in the container, is the same ingredient as the third ingredient. chem1FracDelta is defined as the difference between the ideal and actual volume of the first ingredient and chem2FracDelta is defined as the difference between the ideal and actual volume of the second ingredient. Thus, diwFracVol serves to volumetrically fill the remaining volume for the current fractional fill sequence.

As previously stated, diwAddedVol represents the volume of the third ingredient added to VolLowLev to obtain total volume. diwAddedVol is defined as diwRatio·x−VolLowLev where x is defined as (TotalVol÷(chem1Ratio+chem2Ratio+diwRatio)). If it is determined that diwFracVol is negative, diwFracVol is then reduced by multiplying the first ingredient volume to be admitted to the admixture for the current fractional fill sequence by ((totalVol−VolLowLev)·pourUp2Frac)'(chem1FracVol+chem2FracVol). The volume of the second ingredient is also reduced by multiplying it by the same formula.

It should be noted that the target quantity of one ingredient represented in percent by weight may be modified as a function of specific gravity of each ingredient in the batch. For example, concCheml, by example, may be modified as a function of specific gravity by employing the following replacement formula (chem1Ratio·bulkChem1·sGravChem1)+((chem1Ratio·sGravChem1)+(chem2Ratio·sGravChem2))+(diwRatio·sGravChem3) where concChem1 is the target concentration of the first ingredient, chem1Ratio is a ratio of the volume to be filled for the first ingredient. chem2Ratio is a ratio of the volume to be filled for the second ingredient. diwRatio is the ratio of the volume to be filled for the third ingredient. BulkChem1 is the supply concentration of the first ingredient. sGravChem1, sGravChem2, sGravChem3 represent the specific gravity for the first, second, and third ingredients, respectively.

It should be noted that the above method may be used with concentrated bulk chemicals normally having the concentration measured in percent by weight. Therefore, in the foregoing examples, the formulas listed hereinabove in conjunction with the method for performing fractional fill mixing may use percent by weight concentration as the measure for quantity of the contemplated ingredient in the admixture or from the chemical supply. Alternatively, in other contemplated examples of embodiments of the invention not disclosed herein, percent by volume concentration or other concentration measurement values may be used in some circumstances depending on the type of analytical instrument 14 in use.

Subsequent fractional fill sequences are then calculated and added to the admixture in container 12 using the same formulas and methods stated hereinabove for the foregoing examples.

In one embodiment, the fractional fill mixing apparatus, system, and method may be used for chemical blending or mixing concentrated chemicals for use in the manufacture of semiconductor wafers. Ingredients that may be used to prepare compositions for use in the manufacture of semiconductor wafers includes, but are not limited to an oxidizing agent (e.g., H₂O₂), a base (e.g., NH₄OH) or an acid (e.g., HCl, H₂SO₄, HF, HNO₃, H₃COOH). Additionally, compositions for use in the manufacture of semiconductor wafers includes, but is not limited to water or IPA or another primary chemical constituent. The composition may be useful for echtant, selectivity, accelerants, suppressants, and dilute components of interest as examples. Processes that employ these requirements include as examples cleans, etches, slurries, polymer removal and electroplating. Some examples of compositions are provided in FIG. 6. The column labeled “Label” refers to common industry nomenclature for the composition. The column “Alt Label” refers to an alternate designation for the listed composition. The listing “TEMP” refers to temperature of composition. Analytical technologies such as those supplied by spectral absorption or conductance technologies may be employed in making realtime insitu measurements of these specific compositions.

By way of example, the above mentioned equations may be used to demonstrate how the fractional fill mixing method is employed. For this example, assume that it is desired to create a batch that contains three ingredients. The first two ingredients are named the (“first ingredient”) and the (“second ingredient”). The third ingredient will be deionized water, abbreviated (“diw”). For example, it will be assumed that each ingredient has a specific gravity equal to one. Also for the purposes of this example, it is desired that the ingredients be blended together so that a volumetric ratio of 1:1:100 be achieved where the first ingredient forms one part represented by the variable chem1Ratio, the second ingredient forms one part represented by the variable chem2Ratio and the diw forms 100 parts of the batch represented by the variable diwRatio.

For this example, a 10,000 mL tank 12 will be completely filled with the ingredients. In this example assume, for sake of clarity, that there is no residual volume of diw present in the container. Therefore, the variable VolLowLev will be equal to zero in all of the equations. The total volume of the batch to be created is represented by the variable totalVol is equal to (chem1TotalVol+chem2TotalVol+diwAddedVol) where chem1TotalVol is the total volume of the first ingredient for the batch. chem2TotalVol is the total volume of the second ingredient to meet the requirements for the batch and diwAddedVol is the volume of diw to be added to VolLowLev to meet the requirements for the batch.

Thus, the equation to calculate chem1TotalVol=chem1Ratio·(totalVol÷(chem1Ratio+chem2Ratio+diwRatio)). Plugging in the numbers from our example, chem1TotalVol=1·(10,000÷(1+1+100))=98 mL. Using similar formulas, chem2TotalVol=chem2Ratio·(totalVol+(chem1Ratio+chem2Ratio+diwRatio)). Inserting the numbers from the present example, chem2TotalVol=1·(10,000÷(1+1+100))=98 mL.

diwAddedVol which represents the volume of diw to be added to VolLowLev has a slightly different formula to account for the residual volume of diw in the tank 12. diwAddedVol=diwRatio·(totalVol÷(chem1Ratio+chem2Ratio+diwRatio))−volLowLev. Inserting the numbers from the present example, diwAddedVol=100·(10,000÷(1+1+100))−0=9804 ml.

Therefore, the volume of the batch which equals totalVol also equals (chem1TotalVol+chem2TotalVol+diwAddedVol). Inserting the numbers from the present example, totalVol=(98 mL+98 mL+9804 ml)=10,000. 10,000 mL is also the size of the container 12 that will be completely filled to verify that the calculations are correct.

The desired number of fractional filling sequence is then determined to be performed and the relative fill percentages to accompany each fill sequence. The number of fractional filling sequences and their relative percentages of fill are chosen by the operator. It has been found that this method works well for some applications with four filling sequences where the first sequence fills the container 12 with 50% of the target volume of the completed mixture. This value is assigned to pourUp1Frac. The second sequence fills the container 12 with 25% of the target volume of the completed mixture. This is assigned to variable pourUp2Frac. The third and fourth sequences fill the container 12 each with 12.5% of the target volume of the completed mixture. These values are assigned to pourUp3Frac and pourUp4Frac, respectively. Other quantities of filling sequences and their percentages may be chosen by the operator and may be modified to obtain improved results through experimentation.

In the next step in the method, the concentrations of the bulk supply for each of the ingredients are determined and will be added to the admixture. For this example, assume that the bulk supply of the first ingredient has a concentration of 29% by weight and the bulk supply of the second ingredient has a concentration of 30% by weight. diw, being pure water, in this example, is assumed to be 100% pure. These bulk concentrations may be printed on the material data sheets for the chemicals or ingredients.

The target concentration of the first two ingredients is then calculated. The fractional fill method of this example will attempt to formulate the batch to achieve the target concentrations of the first and second ingredients. These target concentrations are represented by the variables concChem1, concChem2 where concChem1 represents the target concentration of the first ingredient and concChem2 represents the target concentration of the second ingredient. The target concentration of diw is not normally calculated as diw is generally used fill the remainder of volume for a fractional fill when the first two ingredients are added to the admixture. Note that concentration may be measured as a quantity or in percent by weight or volume where either may be used in the formulas.

The variable concChem1 is then calculated by the following formula concChem1=(chem1Ratio·bulkChem1)÷(chem1Ratio+chem2Ratio+chem3Ratio). The variable concChem2 is then calculated by the following formula conceChem2=(chem2Ratio·bulkChem2)÷(chem1Ratio+chem2Ratio+chem3Ratio). Thus, plugging in the numbers from our example, concChem1=(1·29%)÷(1+1+100)=0.284% and concChem2=(1·30%)÷(1+1+100)=0.294%. Note that the specific gravity of each ingredient was not factored into this equation and was assumed to be equal to one for each ingredient.

In the present example, the next step in the method is to calculate the theoretical volumes of each ingredient to be added to the tank 12 for the first fractional fill sequence where, in this step, chem1FracVol represents the actual volume of the first ingredient to meet the requirements for the current or first fractional fill sequence. Chem2FracVol represents the actual volume of the second ingredient to meet the requirements for the current or first fractional fill sequence. diwFracVol represents the actual volume of diw to meet the requirements for the current or first fractional fill sequence.

To calculate chem1FracVol, the following equation is used: chem1lFracVol=chem1TotalVol·pourUp1Frac. Plugging in the numbers from the present example, chem1FracVol=98 mL·50%=49 ml. Chem2FracVol=chem2TotalVol·pourUp1Frac. Inserting the numbers from the present example, chem2FracVol=98 mL·50%=49 ml. Finally, diwFracVol=diwAddedVol·pourUpIFrac. Inserting the numbers from the present example, diwFracVol=9804 mL·50%=4902 ml

The method of this embodiment, as now best shown in FIG. 3. The ingredients are admitted to the container 12 to a fraction of the full container volume for the first fractional fill sequence (28). In this example, the container 12 is then filled with 49 mL of the first ingredient, 49 mL of the second ingredient, and 4902 mL of diw. The first fractional fill sequence is now complete.

Depending on what type of ingredient supply control device 16 is employed, the controller 26 may drive the supply control device 16 to dispense the required amount of ingredients using suitable equipment, such as pumps or gravity feed dispensing devices for flow controllers or others. For pumps, for example, the number of strokes of the pump may be conventionally calculated by the controller 12 and for gravity fed dispensing devices, the dispensing time may be conventionally calculated by the controller 12.

The next step in the method 30 requires that the quantities/concentration of each ingredient in the admixture be determined. An analytical instrument 14 may be utilized for this purpose. For-this example, assume that the analytical instrument 14 can measure the quantities of each ingredient in the admixture in percent by weight which is why the target quantities/concentration for each ingredient is calculated in percent by weight. For the present example, assume that the measured quantity/concentration of the first ingredient is measured at 0.210% by weight which is assigned to variable chem1Val and the measured quantity/concentration of the second ingredient is measured at 0.294% by weight which is assigned to variable chem2Val.

The second and all subsequent fractional fill sequences are then prepared and, in the present example, the ratio of the target quantity/concentration to the measured quantity/concentration of each ingredient in the admixture is calculated (32). The next quantity of each of the ingredients is then calculated by multiplying the target quantity by the ratio calculated for each respective ingredient to determine a corrected quantity (34). That corrected quantity for each ingredient is then added to the admixture.

The method of the present example for accomplishing this involves calculating a series of variables, idealChem1Frac, idealChem2Frac which represent intermediate variables to ultimately obtain chem1FracVol, chem2FracVol, and diwFracVol which represent the corrected volumes of ingredient that shall be added to the admixture to correct the quantities/concentrations of the ingredients in the admixture for the current fractal fill sequences. Thus, variable idealChem1Frac is defined as being equal to chem1TotalVol·pourUp2Frac. Using the numbers from the present example, idealChem1Frac=98 mL·25%=24.5 mL. The variable idealChem2Frac=chem2TotalVol·pourUp2Frac. Using the numbers from the present example, idealChem2Frac=98 mL·25%=24.5 mL.

Now that idealChem1Frac and idealChem2Frac have been calculated, chem1FracVol and chem2FracVol are then calculated. chem1FracVol is equal to (idealChem1Frac·concChem1)÷chem1Val. Thus, using the numbers from the present example, chem1FracVol=(24.5 mL·0.284%)÷0.210%=33.1 mL. chem2FracVol is equal to (idealChem2Frac·concChem2)·chem2Val. Thus, using the numbers from the present example, chem2FracVol=(24.5 mL·0.294%)÷0.294%=24.5 mL.

Now that chem1FracVol and chem2FracVol are calculated, chem1FracDelta and chem2FracDelta are then calculated and represent the difference between the ideal and actual volumes of the first and second ingredients, respectively. chem1FracDelta equals idealChem1Frac−chem1FracVol and chem2FracDelta equals idealChem2Frac−chem2FracVol. Thus, using the numbers in the present example, chem1FracDelta=24.5 mL−33.1 mL=−8.6 mL and chem2FracDelta=24.5 mL−24.5 mL=0 mL.

In the present example, the variable diwFracVol may now be calculated. diwFracVol is equal to (diwAddedVol·pourUp2Frac)+chem1FracDelta+chem2FracDelta. Thus, using the numbers in the equation, diwFracVol=(9804 mL·25%)+−8.6 mL+0 mL=2442.4 mL.

After the corrected fractional volumes of each ingredient for the current fractional fill sequence have been calculated, they are admitted into the admixture (36, 38). For example, 33.1 mL of the first ingredient is added to the admixture, 24.5 mL of the second ingredient is added to the admixture, and 2442.4 mL of diw is also added to the admixture for the current fractal fill sequence.

It should be noted that if diwFracVol was less than zero, then chem1FracVol and chem2FracVol exceed the volume for the current fractional fill sequence. In this situation, chem1FracVol and chem2FracVol are reduced to provide the correct volume for the fraction. Each variable is reduced by multiplying itself by the following fraction ((totalVol−volLowLev) pourUp2Frac)÷(chem1FracVol+chem2FracVol).

The container is then checked to see if the container is filled with the desired quantity of the composition (42). In the present example, this would occur when all of the fractional fill sequences are completed. If not, then the next fractional fill sequence is begun (30). If all of the fractional fill sequences are completed, the method terminates (44).

With reference to FIGS. 4 and 5, there is shown another embodiment of the present invention which includes a fractional fill method incorporating self diagnostics. The method of this embodiment begins at 46 as best shown in FIG. 4. Stored user-defined parameter values are gathered by the controller 12 for subsequent use within the fractional fill method (48). These user-defined parameter values may include the number of fractional fill sequences to be performed, and the relative fill volume percentages. The user-defined parameter values may also include information such as concentration information regarding the bulk ingredients to be added to the admixture.

The method calculates the proper volumes of ingredients to be added to the admixture for the first fractional fill sequence (50). Those ingredients are then added to the admixture. Feedback from an analytical instrument such as the analytical instrument 14 provides the quantity, expressed in a percent by weight, or percent by volume concentration or other, of each of the ingredients in the admixture stored in the tank 12 for the first fractional fill sequence (52). A decision is then made whether the method is within the first fractional fill sequence or the second fractional fill sequence (54). If this is true, self-diagnostics are then performed (56).

As best seen in FIG. 5, self-diagnostics begin at 58. The method of the example then evaluates whether or not the first fractional fill sequence was complete (60). If it was complete, the determination is made whether or not the first fractional sequence delta values are already stored (62). The first fractional fill sequence delta values represent the difference between the theoretical volumes of the ingredients that should be dispensed into the admixture compared to a revised volume for an ingredient that may be admitted to the admixture due to a variance detected by the analytical instrument 14.

If those fractional filled delta values are not already stored, the controller 26 stores those fractional delta values (64). The method as executed by controller 26 then determines whether the second fractional fill sequence is complete (66). If not, the self-diagnostics method is terminated (74) and the method then returns to the method as shown in FIG. 4 at 76. If the second fractional fill sequence has been completed, then the second fractional fill delta values are captured and the differences between the first fractional fill delta values and the second fractional filled delta values are then calculated (68).

According to this embodiment, if any of the second fractional fill delta values are greater than or equal to the first fractional filled delta values (70), then the filling sequence is stopped and an error message is displayed (72). This result occurs when the fractional fill method is unable to correct any deviation in ingredient concentration or quantity between the first fractional fill sequence and the second fractional fill sequence. In other words, if a deviation or delta is discovered in any of the ingredients for the first fractional fill and then a corrective partial fill of ingredients is added in the second fractional fill sequence, assume that it is discovered that the deviation or delta of any of the ingredients did not decrease between the first fractional fill sequence and the second fractional fill sequence. In that case, the fractional fill method is then deemed to be unable to complete the creation of the desired batch.

Referring back to decision box 70 on FIG. 5, if any of the second fractional fill delta values are not greater than or equal to the first fractional fill delta values then the self-diagnostics method terminates (74) and returns to the fractional fill method as shown on FIG. 4 at 76.

Referring now to FIG. 4, the method evaluates whether the blended constituents are on target (78). In other words, the analytical instrument 14 analyzes the quantity, percent by weight, percent by volume concentration or other, of the chemical constituents depending on the example in the admixture. If they are not on target, an error correction is then calculated for the subsequent fractional fill sequence as described previously (80). The error correction is then in the calculation of the amount of each ingredient used for the subsequent fractional fill (82). If the blended constituents are on target, the volumes for each ingredient are then calculated for the subsequent fractional fill sequence without having any error correction applied (82).

The method of the present embodiment, as shown in FIG. 4, then determines if the fourth fraction is complete (84). It should be understood that if the stored user-defined parameter values call for less or more than four fractional fill sequences, the method evaluates whether all of the desired fractional fill sequences have been completed.

If the fourth or final fractional fill has been completed, then the method of the present embodiment terminates (86) where closed loop control of the admixture in tank 12 may begin. The closed loop control approach may be feedforward or feedback. The blending may occur via liquid flow controllers, metering pump, critical orifice or gravimetric such as preweigh or a continuous drip method as examples.

In another embodiment, collected analytical data may be used to control the amount of each ingredient added in a feedback loop. For each component of the composition, a target concentration is predetermined by the user. As the composition is prepared, the analyzer determines the actual concentration of each ingredient of the composition. The error (E) is represented as the difference between the target concentration (R) and the measured concentration (B). If the error is larger than a predetermined set point, the controller takes appropriate action to modify the metering of the ingredients. In an embodiment, the controller determines how much to vary the amount of each ingredient to bring the batch composition within the targeted range. The change in quantity, can be determined based on the following equation:
quant=gain*error
The allowable error range and the gain parameter are preset for each chemical.

Considering now the fractional fill mixing apparatus of the disclosed embodiment in greater detail with reference to FIG. 1, an air operated process pump 88 may be used to re-circulate the ingredients in the tank 12 to achieve homogeneity of the mixture. The pump 88 is operatively connected through a solenoid valve 94 to a source of air under pressure. Process pump 88 may be air operated to minimize the risk of any explosions or fires since flammable compounds and ingredients may be flowing through pump 88. Process pump 88 is connected in fluid communication with tank 12 via a conduit 90. A maintenance drain 92 may be in the form of a manual valve for manually performing draining operations from the conduit 90.

A filter 96 is disposed in-line with the pump 88 within the recirculation line of the fractal fill mixing apparatus 10, and a conduit 98 connects the pump 88 to the filter 96. An air operated 3-way valve 102 is connected in the re-circulation line between the pump 88 and the filter 96 via the conduit 98, to permit the re-ionized water from a source of de-ionized water under pressure to enter the conduit 98 for the purpose of flushing out the fractional fill mixing apparatus 10.

A 3-way valve 100 is disposed in line with the valve 102 to permit draining between batches. A valve 104 is also connected in line with the valve 102 for permitting nitrogen gas under pressure to enter the fractional fill mixing apparatus 10. A 3-way valve 106 connected in fluid communication down stream of the filter 96 to selectively permit ingredients stored in tank 12 to be delivered via a conduit 124 to a process chamber (not shown) for utilization of the batch.

A conduit 108 connects the filter 96 in fluid communication with the valve 106, and an analytical pump 112. A valve 110 may be a solenoid valve which permits air under pressure to drive the analytical pump 112. A conduit 114 is connected in fluid communication between the conduit 108 and the pump 112 to re-circulate the mixture from the tank 12.

The analyzer or analytical instrument 14 is connected in fluid communication with the output of the pump 112 via a conduit 116. The analyzer 14 may be a high precision chemical concentration monitor. An example of such a device is the SC-1 monitor manufactured by HORIBA and marketed as model no. CS-131. Other monitors include, but are not limited to, HORIBA model No. CM-210 for DHf compositions, HORIBA model No. CS-131 for SC I compositions, HORIBA model No. CS-137 for BHF compositions, HORIBA model No. CS-150 for SPM compositions, and HORIBA model No. CS-152 for SC2 compositions. The analytical instrument or analyzer 14 is connected in fluid communications with a by-pass re-circulation conduit 120 via a conduit 118 to the valve 106, so that the mixture is re-circulated through both the analyzer 14 and the by-pass conduit 120 until the delivery valve 106 is actuated to deliver the batch via the conduit 124, the mixture is re-circulated to the manifold 24.

Manifold 24 is connected in fluid communication to the ingredient supply control device generally indicated at 16 via three conduits 132, 134 and 136. Ingredient supply control device 16 includes three independent ingredient control devices 126, 128 and 130. Each control device is capable of accurately dispensing ingredients from a bulk supply (not shown) into the manifold 24. Ingredient control devices 126, 128 and 130 are each independently fed from the ingredient supply tubes 18, 20, and 22, respectively. Manifold 24 is connected in fluid communication with the tank 12 via a conduit 122.

The ingredient control devices 126, 128, 130 may be any number of control devices such as pumps, gravity feed systems, flow controllers, or other. Examples of metering pumps that could be used include pumps that are driven by dry reed contact closures and are characterized by CC (volume) per stroke such as contact closure cycle. The number of strokes per minute with the number of stokes being calculated based on the desired quantity of chemical. The pumps are activated for that number of stokes at the desired speed. Another pumping approach would be to measure the flow with some measuring device such as an ultrasonic or delta pressure technique and feed back to a pump the desired rpm and to a pressure regulator the required pressure to achieve the required volume to be delivered.

FIG. 7 depicts a detailed view of an embodiment of ingredient control devices 126, 128 and 130. In this embodiment, the ingredient control devices include two liquid flow controllers (126 and 128) for ingredients 1 and 2. Ingredient 3, in this embodiment, is deionized water, but could also be any other ingredient suitable for formulating the desired composition. Ingredient 3 is dispensed gravimetrically from tank 130. Tank 130 is kept filled with at least some water. Controller 26 is coupled to tank 130 and monitors the level of water in the tank and the temperature of water in the tank. A pre-weigh algorithm, discussed below, is used to keep sufficient water in the tank. A heater is coupled to the tank to maintain the water at a temperature consistent with the desired formulation.

FIG. 8 depicts a detailed view of an alternate embodiment of ingredient control devices 126, 128 and 130. In this embodiment, the ingredient control devices include three dispensing tanks (126, 128, and 130) for ingredients 1, 2 and 3. Ingredient 3, in this embodiment, is deionized water, but could also be any other ingredient suitable for formulating the desired composition. Each of the ingredients are dispensed gravimetrically from corresponding tank. Controller 26 is coupled to each of the tanks and monitors the level of the chemical in the tank and, if appropriate the temperature of the chemical in the tank. A pre-weigh algorithm, discussed below, is used to maintain a sufficient amount of chemical in the tank. A heater may be coupled to one or more of the tanks to maintain the chemical at a temperature consistent with the desired formulation.

FIG. 9 depicts a flow chart of a pre-weigh algorithm. In this method, the amount of fluid in a storage tank is determined. If the amount of fluid in the tank is above a lower limit, there are no further process steps required and the method terminates by going back to a tank monitoring state. If it is determined that the tank has reached a lower limit, the method begins by opening valves from the ingredient bulk supply to allow the ingredient to flow into the tank. A timer is set to prevent overflow of the tank. This time period is calculated to dispense the appropriate amount. Liquid is fed to the main tank by gravity. As the ingredient is introduced into the tank, the controller monitors the fluid level in the tank. If the fluid level reaches the high limit, the introduction of the ingredient is halted. If the fluid level is still below the high level, the timer is checked. If the time has not reached the set limit, the introduction of the sample is continued. This process is continually cycled until either the high limit is reached or the time for introducing the sample has expired. In either case, the introduction of the ingredient from the bulk ingredient supply is halted. Before completing the method, an error check is performed to determine if the tank was properly filled, an indication that the tank is below the low limit would indicate that the tank is not properly filled and corrective actions would be required. The low level sensor is set such as to have the minimum amount of needed liquid in the tank.

A heater 150 heats the ingredients within the tank 12. A bath temperature controller 170 regulates the heater 150 to control the temperature of the admixture in tank 12. The bath temperature controller 170 measures the temperature of the admixture in the tank 12 via a temperature probe 146.

Ingredients supply control device 16 and its individual ingredient control devices 126, 128 and 130 are controlled by the digital outputs of the controller 26 via a cable 188. The controller 26 may be placed in a communicating relationship to a host computer 168 via a cable 186, or indirectly via a master controller (not shown) when a distributed network is desired.

In operation and with reference to FIG. 1, the controller 26 receives a series of recipe parameters from the host computer 168 that describe the desired quantities of each ingredient to be blended together in tank 12. The controller 26 then performs a first fractional fill sequence as previously described. The controller 26 sends commands to the ingredient supply control device 16 to dispense the proper amount of ingredients for the first fractional fill. When this occurs, the ingredient control devices 126, 128 and 130 begin accurately dispensing ingredients from their respective bulk ingredient supplies (not shown) via the conduits 18, 20 and 22, respectively. Each ingredient is then dispensed into the manifold 24 through the conduits 132, 134, and 136. The ingredients are partially mixed in manifold 24 and then supplied to the tank 12 through conduit 122. After the first fractional fill sequence is complete, the analyzer 14 is enabled to measure the quantity/concentration of each of the chemical constituents in the admixture stored in tank 12.

To accomplish this, the pump 88 is activated to re-circulate the mixture from the tank 12 by means of the air valve 94 which causes the admixture stored in tank 12 to flow through the conduits 90 and 98 through the filter 96 and through the conduit 108. During this operation, the maintenance drain 92 is closed as well as the drain valve 100, the valve 102 and the valve 104. The valve 106 is also closed. The admixture from tank 12 then continues to flow through the by-pass conduit 120 through the manifold 24 and back into the tank 12. The re-circulation flow of the admixture is generally shown by curved arrow 144. In this regard, the admixture stored in tank 12 is circulated through the various conduits to mix the admixture to create a more homogeneous admixture before the analytical instrument 14 measures its concentration. The analytical pump 112 is then enabled through air valve 110 which pumps some of the admixture from the conduit 108 to flow through the conduit 114 through the pump 112 and through the analytical instrument 14 where the concentration of the mixture may be measured. The admixture then exits the analytical instrument 14 via the conduit 118 to flow through the manifold 24 and into the tank 12 via the conduit 122.

For subsequent fractional fill sequences, the same general method as just described is performed again. In the present example, before subsequent fractional fill sequences are performed, the process pump 88 and analytical pump 112 are both disabled through their respective valves 94 and 110, although for other applications they may not be disabled. Subsequent to the completion of all the fractional fill sequences or at other times, the bath temperature controller 170 may be enabled to control the heater 150 to heat the admixture to a predetermined temperature. This may be required for some admixtures for subsequent use in a manufacturing process or other process or purpose.

After all of the fractional fill sequences are complete, it may be desired for some applications to transfer the admixture stored in the tank 12 to a process chamber (not shown). That may be accomplished by first ensuring that the maintenance drain 92 is closed. The drain 100 is closed, the DI flush valve 102 is closed, and the nitrogen valve 104 is also closed. In this step, however, the valve 106 is now open. Process pump 88 is then enabled through valve 94 which pumps the admixture from the tank 12 through the conduit 90, the pump 88, the conduit 98, the filter 96 and to the conduit 108. Because valve 106 is now open, the admixture then flows through valve 106 and through the conduit 124 where it is delivered to the process chamber or other destination.

A reclaim drain 3-way valve 140 is disposed between conduits, 138 and 142, so that when reclaimed drain valve 140 is open, a recycled admixture may be reclaimed into the tank 12 through conduits 138 and 142 through valve 140. It should be noted that in all other operations of the fractional fill mixing system 10, the reclaim drain valve 140 is normally closed. A Flush Drain cycle may be used which includes: Opening the drain valve 100 and opening the DI Flush valve 102 and running the process pump 88. This will result in simultaneously draining the tank and flushing the entire system including the manifold 24 and analyzer 14. This state remains for a designated parameter time. Turning off DI Flush valve 102 completes the Flush Drain cycle, and then nitrogen or some other inert gas (e.g., helium, argon, etc.) is introduced to force the tank and associated plumbing to drain. The drain tank continues until digital input Lo_Level becomes TRUE and a SYSTEM ERROR occurs if the Maximum Drain Time is exceeded. The drain valve 100 and N2 valve are closed. This would complete a Flush Drain Cycle, followed by Fill Cycles. Several Fill cycles can be used to fill a tank. The intent of separating the fill cycles is to achieve a full tank with the desired blended chemistry as soon as possible. The fractions are variable that can be adjusted to optimize the time to blend ready.

In operation the controller 26 communicates to the bath temperature controller 70 through a serial communications line 160 under the RS-485 protocol. Likewise, the controller 26 may also communicate to the ingredient supply control device 16 and its individual ingredient control devices 126, 128 and 130 through the digital serial line 188, or through an analog signal source, if desired. The controller 26 may communicate to the host computer 168 through another serial connection 186.

Considering now the controller 26 in greater detail with reference to FIG. 10, the controller 164 includes a controller package 180, which includes a plurality of digital inputs, digital outputs, serial ports, A/D channels, and a PLC BUS. One example of such a controller is a Z-World controller under the model No. PK 2600. Such a controller from Z-World contains a BL 1700 controller 183 and an OP 7100 display and touch screen 182. Controller package 180 has a first serial port 182, which provides RS 232 communications between the controller 180 and an analytical instrument, such as analytical instrument 14. A second serial port 186 provides communications between the controller 180 and the host computer 168, or to a master controller (not shown). A third serial port 158 is also provided on the controller package 180 and provides RS-485 communications to the bath temperature controller 170 as best shown on FIG. 1. Controller package 180 also includes 16 digital outputs shown generally as the cable 188 that are operatively connected to various pumps and valves of the fractional fill mixing apparatus and system 10, including the ingredient supply control device 16. The controller package 180 also contains 16 digital inputs shown generally as 190 which provide digital input to the controller package 180 for various level sensors, leak detectors and other. Such a level sensor is shown on FIG. 1 as level sensor 154 connected through digital input line 156 to the controller 170.

A PLC bus is also included with the controller package 180 and shown generally as 192. The PLC bus emanates from the controller package 180 as a ribbon cable and is attached to a plurality of extension devices, such as an expansion 10 device 194, auxiliary serial output device 208, a D/A channel device 199. The PLC bus provides digital input and output control of these accessory devices from the controller package 180.

Expansion 10 device 194 provides additional digital outputs, which may be used to control additional components in the fractional fill mixing system 10.

The auxiliary serial output accessory 208 is also connected to the PLC bus 192 and provides an additional RS 232 communications port used for data logging and chit-chat used primarily for monitoring and software development. This RS 232 port shown generally at 210 may be also connected to a recorder 212 for recording and monitoring operations on the controller package 180. Software for the controller package 180 may also be loaded, if desired, through this RS 232 communications port 210.

The D/A accessory 199 is additionally connected to the PLC bus 192 and provides analog outputs to control various components on the fractional fill mixing apparatus and system 10 shown generally on FIG. 1. One such component that may be controlled by the D/A accessory 199 may be the ingredient supply control devices 126, 128, or 130 as well as the pumps 88 and 114. Optionally, a TAKVTOI accessory may be operatively coupled to the D/A accessory to convert the analog voltage outputs from the accessory 199 to a plurality of current signals. These current signals created by the TAKVTOI accessory 201 may be used to drive various metering pumps as part of a fractional fill mixing apparatus and system 10.

The controller package 180 also includes eight 12-bit A/D channels to monitor a variety of information from the fractional fill mixing system 10. For example, the thermalcouple such as the thermalcouple 146 (FIG. 1) may be coupled to one of the A/D channels 204 so that the controller package 180 may monitor the temperature of the admixture. In addition, the A/D channels may also monitor various flow controllers or metering pumps, which may be part of a typical fractional, fill mixing system 10.

A fractional fill algorithm or method may be loaded in the form of software to the controller package 180 through a suitable storage media such as a compact disk 206 which contains the fractional fill algorithm or method thereon, or loaded through the RS 232 communications port 210.

FIG. 11 depicts a schematic diagram of a system for point of use chemical mixing of chemicals. Referring now to the drawings and, more particularly, to FIG. 11, there is shown a point of use mixing apparatus or system 11, which is constructed in accordance with an embodiment of the present invention, and which is used to mix two or more ingredients in a mixing area 13. An analyzer or analytical instrument 15 is adapted to measure the quantities of each ingredient flowing from the mixing area. An ingredient supply control device shown generally at 17, controllably dispenses two or more ingredients into the mixing area 13. The ingredient supply control device 17 dispenses ingredients through a plurality of ingredient supply inlets, as has been previously discussed. Each ingredient supply inlet is connected in fluid communication with a plurality of ingredient supplies (not shown). The manifold receives the plurality of ingredients from ingredient supply control devices. The ingredients then flow from the manifold to the mixing area 13

Considering now the fractional fill mixing apparatus of the disclosed embodiment in greater detail with reference to FIG. 1, a process pump may be used to conduct fluid from the mixing area to the analyzer. Process pump is connected in fluid communication with mixing area 13. A maintenance drain may be in the form of a manual valve for manually performing draining operations from the conduit.

A filter may be disposed in-line with the pump. A 3-way valve may be connected in the re-circulation line between the pump and the filter to permit the re-ionized water from a source of de-ionized water under pressure to enter the conduit for the purpose of flushing out the mixing apparatus 10. A 3-way valve is disposed downstream from the analyzer to allow the fluid exiting the analyzer to be directed to either a drain (if the composition is not correct) or the fluid may be passed to a tool or instrument for use in a process.

The analyzer or analytical instrument 14 is connected in fluid communication with the output of the pump. The analyzer 14 may be a high precision chemical concentration monitor. Examples of chemical monitors have been previously described. Controller 27 is at least coupled to the ingredient supply control device 17 and the analyzer 15. Based on the data collected by the analyzer the controller is configured to make decisions regarding the flow rate of the various ingredients.

Referring now to FIG. 12, a single control algorithm may be applied by user selectable methods for either batch or point of use processing. Since the control action is the same, the method employed is situational and user dependent. The control algorithm provides the user with gain and dead band control so that the user can optimize the control action appropriate to the users' situation. The algorithm also contains a variable, which holds the error correction value for each chemical control device to be used for each blend sequence. This variable is updated, as required, based on the analytical feedback delta to desired blend, either in the continuous mode, or the periodic testing mode. A standard SISO control is applied to each component and a user-defined delay is applied between actions to allow time for the analytical to provide feedback and see the effect of each action upon the multi-variant components. As the tank becomes vanishing small the control and blending shrink to a small volume for mixing the desired blend. This allows a faster response time to provide process flexibility and tool versatility because blends can be provided at Point of Use with the decision to make blend changes while the composition is in use.

Turning to FIG. 12, the process begins at 700. User defined parameters are initially collected by the controller (701). These parameters include the chemical ratios for each component of the composition. Alternatively, parameters for commonly used compositions may be preset into the controller, in which case the user selects which of the preset compositions that are to be made. Also input initially is information collected by one or more analytical instruments (702). Data collected by analytical instruments during previous chemical blends may be used to modify the control parameters for subsequent runs.

After the parameters have been obtained, the controller determines the appropriate dispensing method of each ingredient of the composition that will produce the desired composition. For batch processing the controller determines the amount of each ingredient that is to be dispensed into a tank, as discussed previously. For point of use processing, the flow rate of each ingredient is calculated. The flow rate of each ingredient may be calculated using a fee-forward algorithm based on the total flow rate desired, the supply composition and the target composition. Analytical measurements received during previous use of the mixing system may be used to provide error correction values.

In a feed forward algorithm for a point of use mixing process the total flow of the process stream is typically a user selected parameter. The total flow of the process stream is the sum of the flow of each ingredient being added to the mixing area, as represented by the equation:
TotalFlow=chem1Flow+chem2Flow+diwFlow
It should be noted that while the above equation is directed to a composition that includes two chemical ingredients and deionized water, the same general equation would apply for more or less than three components.

The mixing method next includes calculating the target quantity of one ingredient based on the user supplied blending ratio and the supply concentration of the ingredient (703). The target quantity of one ingredient is referred to as chem1Targ, which is defined as (chem1Ratio·bulkChem1)÷(chem1Ratio+chem2Ratio+diwRatio). Where chem1Ratio and chem2Ratio and diwRatio represent the ratios of the volume to be filled for the first, second, and third ingredient, respectively, for the current sequence. These values are either preset into the controller or are defined by the user at start-up of the process. BulkChem1 represents the supply concentration of the first ingredient. The target quantity of the other ingredients are calculated using similar formulas where the numerator of the above equation is replaced with the ratio and concentration of the bulk ingredient supply from the respective ingredient being calculated. chem2Targ and diwTarg are also calculated as just described.

After determining the target quantity of each ingredient, the flow rate of each ingredient is determined using the following equations:
Chem1Flow=(chem1Targ/bulkChem1)*totalFlow
Chem2Flow=(chem2Targ/bulkChem2)*totalFlow
The above equations assume that the specific gravity of each ingredient and the final composition is 1. If the specific gravity is not equal to one, the following equations are used:
Chem1Flow=(chem1Targ*totalFlow*sGravPosite)/(bulkChem1*sGravChem1)
Chem2Flow=(chem2Targ*totalFlow*sGravPosite)/(bulkChem1*sGravChem2)
where sGravPosite is the specific gravity of the blended composition stream. SgravChem1 and sGravChem2 are the specific gravity of ingredient 1 and ingredient 2, respectively. The total flow of the third component, in this case deionized water, is determined from the following equation.
DiwFlow=totalFlow−chem1Flow−chem2Flow

The flow rate may also be modified based on one or more correction values that have been stored based on previous use of the mixing system. The stored error correction values (704) may be used to modify the flow rates according to the following equations:
Chem1FlowCorrected=Chem1Flow*ErrorCorr1
Where Error Corr1 is an error correction value determined from the analytical results of the process. This value would be 1 if no error correction is needed. The value can be calculated from the equation:
ErrorCorr1=Chem1Targ/Chem1Measured
Where Chem1 Measured is the amount of ingredient 1 measured in the produced composition. A similar correction value can be determined and used for ingredient 2.

In one embodiment, flow of each component of the composition is controlled by metering pumps. The metering pumps may be operated using voltage control to control the flow rate at which the ingredient is supplied. The following equations may be used to determine a voltage applied to the metering pump to achieve the calculated flow rate.
chem1Dac=(chem1Flow/chem1RangeFS)*DAC_—SIG_—FS
chem2Dac=(chem2Flow/chem2RangeFS)*DAC_—SIG_—FS
diwDac=(diwFlow/diwRangeFS)*DAC_—SIG_—FS
where chem1RangeFS, chem2RangeFS and diwRangeFS represent the maximum flow rate of the pump coupled to the respective ingredient supply sources and DAC_SIG_FS represents the voltage applied to achieve the maximum flow rate.

After determining the process parameters, the ingredient supply control devices are operated to begin producing a process stream. The system may be operated in either a continuous mode or a test mode (705). The composition stream passes into an analyzer where the amount of each ingredient is determined. The analyzer provides analytical feedback (707) to the controller regarding the concentration of the ingredients. The controller analyzes the collected data to determine if the ingredients are within a target range (706). In one embodiment, if the components are not within the target range and error correction value, calculated as described above, is determined and stored. The error correction value may be used to alter the flow rate of each of the ingredients to bring the blended composition to the target formulation.

In another embodiment, the collected analytical data may be used to control the pumping parameters in a feedback loop. For each component of the composition, a target concentration is predetermined by the user. As the composition is prepared, the analyzer determines the actual concentration of each ingredient of the composition. The error (E) is represented as the difference between the target concentration (R) and the measured concentration (B). If the error is larger than a predetermined set point, the controller takes appropriate action to modify the metering of the ingredients. In an embodiment, metering of the ingredients may be controlled by modifying the voltage applied to one or more metering pumps. The change in voltage, can be determined based on the following equation:
ΔM=(100/PB)*(((Δt/T_R)*E_N)+ΔE)
where ΔM is the change (increase or decrease) in the DAC voltage (scaled by chemRange/10 v); PB is the Proportional Band (classical definition); T_R is the reset time (seconds); Δt is the scan time (seconds); E_N is the current error; and ΔE is the change in error since the last scan. The parameters PB and T_R are preset for each chemical.

If the blended constituents of the composition within a selected target range then the composition is checked for homogeneity (711). If the composition is not homogeneous, then further error correction is performed and applied to the process. If the composition is determined to be homogenous, then the composition may be passed onto the tool or process that requires the composition. The cycle is then looped through until the process is complete (709). Upon completion of the process, the cycle is terminated (710).

In some embodiments, a recycle loop may be used to recycle the process stream back to the mixing area if the process stream does not meet the predetermined composition formulation. Alternatively, the composition may be directed to a drain where the composition is disposed of or reused in another process, if the composition formulation is out of a predetermined range. One advantage of using a combination of a feed forward and feedback control system is that the amount of composition that is produced is minimized. This is due in part to the quick response time of the system. Additionally, the storage or error correction values helps the system create a composition that is close to the desired formulation when the process is initiated. By minimizing the amount of non-usable composition, a recirculation system may not be required. By eliminating recirculation of the composition, the system may be more responsive to user needs to modify the composition formulation during use of the system.

In an embodiment, a controller may be configured to control both point of use and batch processing methods. The system may include a combination of mixing areas and batch storage tanks to allow multiple uses of the system. A general control method is presented in FIG. 13. Initiation of the process begins (800) and data regarding the specific parameters of the process are collected from the controller memory and the user. Information that is collected includes information regarding the type of device used to dispense each of the components of the composition (802). Examples of dispensing methods include, but are not limited to liquid flow controllers, metering pumps, gravimetric dispensers, critical orifice dispensing and continuous flow.

After gathering the information regarding the system and user parameters, the type of process is chosen (801). The process choices, in some embodiments include bulk chemical dispense (803), point of use small tank (804) and point of use—single pass (805). The bulk chemical dispense may be represented by a non-resident large volume tank. A point of use small tank may be a tank that is resident on a process tool.

If either of the tank options (803 or 804) are chosen, a fractional fill controlling scheme may be used (806). In some embodiments, a replenishment cycle is followed in which the tank is replenished as the composition is used. This replenishment cycle is followed while chemistry is recirculated (807) and determination is made as to whether the tank is full (808) and continues under the initial fill (817) until the fractional filling is complete.

Once the tank is filled, confirmation of the blended target (809) is achieved and confirmed. If the target is not achieved, the blend may be discarded into a drain (816). In the case of a tank or container approach, recirculation may continue (807) until the feedback closed loop control scheme (815) achieves the desired target concentration.

In addition to blending to target (809) an additional requirement may be the homogeneity of the blend (810). If this is confirmed (810) then the blend may be released to process (811). If the mixture does not meet the homogeneity requirements, a point of use single pass blend is sent to a drain, or a tank/container batch would be recirculated until homogeneity is achieved.

Process indicators (812) include, but are not limited to, endpoint detection, metrology or parametric values determined during the fabrication process. This feedback may be used to provide stored data for feed forward control (814) as well as for adjustment of feedback closed loop control (813).

In an embodiment, a controller includes an automatic detection and correction of fault system. A flow chart of an automatic detection and correction of fault system is shown in FIG. 14. Prior to any blend all inputs 902 are analyzed to ensure that the normal “state” exists for any component or subassembly 901. These conditions are then analyzed for signature failures 903 that either prevent blending 904 or allow an automatic correction strategy to be implemented 907. If the error is not correctable 905 or the error cannot be eliminated 908, a flag is provided to require maintenance 906. In any case an internal data log is generated to allow fast response for maintenance repair and the tool and fab automation are informed 910. After the initial blend cycle, subsequent blending also receives information via feedback from process which can come from the tool or APC (Automated Process Control) or from the various sensors present in the blending module 911.

In order to provide a definitive feedback to a chemical mixing apparatus, system and method a homogeneity monitor may be used. This method is applicable to spectroscopy and other analytical approaches such as conductivity. In semiconductor and other manufacturing processes, it is important to have a uniform blend of chemistry applied. In order to avoid non-homogeneous chemistry from being introduced, this monitor method will provide an indication as to the status of a blend. This approach can be applied to all blending methods and analytical measurement techniques disclosed herein. General flow charts depicting methods of performing homogeneity measurements and equipment design for homogeneity measurements are depicted in FIGS. 15 and 16.

A method of forming a composition which includes determining the homogeneity of the composition may include:

• • Collecting data provided from an analytical instrument;
  • Converting the collected data to a format that may be analyzed statistically;
  • Determining the homogeneity of a blend for multivariant components using a method such as Root mean square groupsizes computed from Mahalanobis distance; and
  • Supplying an output as a discrete or continuous indication of the blend. This output may be used by a blending apparatus. The output may also be supplied forward to a tool that requires information as to a well blended system.

The first software routine is responsible for taking data (usually spectral data) and converting the information into a format that can be analyzed by the use of Mahalanbois Distances or some appropriate statistical approach.

In some embodiments, the method includes two products. One will be a development tool which may be applied to determine the components of interest. The other may be a process tool which will be optimized for time response. In this manner a POU (point of Use) method that provides feedforward or feedback control capabilities for use in a Closed Loop Control or optimization of blend system may be achieved.

As embodied in software, the method may be configured into other controllers or imbedded into an analytical hardware to provide a discrete or continuous indication for “goodness of blend”.

All analyses disclosed herein are not limited to any one analytical approach. Absorption Spectroscopy which includes these classifications UV/VIS, NIR, MidIR, RAMAN as well as analytical instruments that provide output such as conductivity, refractive index, ultrasonic form a subset of the analytical approaches that could be used to analyze compositions.

In this patent, certain U.S. patents, U.S. patent applications, and other materials (e.g., articles) have been incorporated by reference. The text of such U.S. patents, U.S. patent applications, and other materials is, however, only incorporated by reference to the extent that no conflict exists between such text and the other statements and drawings set forth herein. In the event of such conflict, then any such conflicting text in such incorporated by reference U.S. patents, U.S. patent applications, and other materials is specifically not incorporated by reference in this patent.

Further modifications and alternative embodiments of various aspects of the invention may be apparent to those skilled in the art in view of this description. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It is to be understood that the forms of the invention shown and described herein are to be taken as the presently preferred embodiments. Elements and materials may be substituted for those illustrated and described herein, parts and processes may be reversed, and certain features of the invention may be utilized independently, all as would be apparent to one skilled in the art after having the benefit of this description to the invention. Changes may be made in the elements described herein without departing from the spirit and scope of the invention as described in the following claims. In addition, it is to be understood that features described herein independently may, in certain embodiments, be combined.

Claims

1-29. (canceled)

30. A method of producing a composition, the composition comprising two or more components, the method comprising:

obtaining a total flow rate for the composition;

determining the flow rate for each component, wherein the sum of the flow rates for each component equals the total flow rate, and wherein the flow rate of each component is determined based on a predetermined composition formulation;

initiating flow of each component into a mixing area, wherein each component is flowed at the determined flow rate, and wherein the components are combined in the mixing area to produce the composition;

transferring the composition into an analyzer, wherein the analyzer is configured to measure the concentration of one or more components of the composition;

transferring the composition to a process tool if one or more components of the composition is within a predetermined concentration range.

31. The method of claim 30, wherein the total flow rate is a total flow rate of the composition required by a process tool.

32. The method of claim 31, wherein the process tool is a semiconductor fabrication tool.

33. Method of claim 30, wherein the composition is produced as a continuously flowing composition.

34. Method of claim 30, wherein the composition comprises a solution of two or more components.

35. The method of claim 30, wherein composition formulation comprises a weight percentage of each component.

36. The method of claim 30, wherein the composition comprises an aqueous solution of two or more components.

37. The method of claim 30, wherein the components comprise one or more inorganic bases.

38. The method of claim 30, wherein the components comprise one or more mineral acids.

39. The method of claim 30, wherein the components comprise hydrogen peroxide.

40. The method of claim 30, wherein the analyzer comprises a spectral absorption analytical device.

41. The method of claim 30, wherein the analyzer comprises a conductance measurement device.

42. The method of claim 30, wherein the composition, after passing through the analyzer is passed to a waste drain if one or more of the components are not within a predetermined concentration range.

43. The method of claim 30, further comprising determining error correction values if one or more components are not within a predetermined concentration range.

44. The method of claim 30, wherein determining the flow rate for each component comprises using one or more error correction values to determine the flow rate, wherein the one or more error correction values are obtained from previous use of the system.

45. The method of claim 30, further comprising adjusting the flow rate of one or more components based on the measured concentration of the one or more components.

46. The method of claim 30, further comprising adjusting the flow rate of each of the components if one or more components of the composition are not within a predetermined concentration range.

47. A computer readable medium having stored thereon computer executable instructions, readable by a controller for a chemical mixing system, wherein the computer executable instructions are for performing a method comprising:

obtaining a total flow rate for the composition;

determining the flow rate for each component, wherein the sum of the flow rates for each component equals the total flow rate, and wherein the flow rate of each component is determined based on a predetermined composition formulation;

initiating flow of each component into a mixing area, wherein each component is flowed at the determined flow rate, and wherein the components are combined in the mixing area to produce the composition;

transferring the composition into an analyzer, wherein the analyzer is configured to measure the concentration of one or more components of the composition;

transferring the composition to a process tool if one or more components of the composition is within a predetermined concentration range.

48-63. (canceled)

64. An apparatus for preparing a composition, comprising:

a mixing area;

at least two chemical dispensing devices, each chemical dispensing device having an input and an output, each input coupled to a chemical supply and each output coupled to the mixing area;

an analytical device configured to measure the quantities of one or more ingredients in a composition, wherein the analytical device is coupled to the mixing area, and wherein the composition produced in the mixing area flows into the analytic device from the mixing area; and

a controller coupled to the chemical dispensing devices and the analytical instrument, wherein the controller is configured to perform the method comprising: obtaining a total flow rate for the composition; determining the flow rate for each component, wherein the sum of the flow rates for each component equals the total flow rate, and wherein the flow rate of each component is determined based on a predetermined composition formulation; initiating flow of each component into a mixing area, wherein each component is flowed at the determined flow rate, and wherein the components are combined in the mixing area to produce the composition; transferring the composition into an analyzer, wherein the analyzer is configured to measure the concentration of one or more components of the composition; transferring the composition to a process tool if one or more components of the composition is within a predetermined concentration range.

65-80. (canceled)