Background of the Invention Control systems and methods can essentially be divided into two groups. The first group includes feedback controls in which a control signal is generated after the parameter of the process which is to be controlled shows a deviation from the value this parameter is supposed to have. The second group includes feed forward controls in which variables of a process are sensed before they become effective on the parameter to be controlled and a control operation is taken to prevent a deviation of the parameters from the target value. Modern control systems and methods combine both feed forward and feedback controls to obtain results that are more accurate.

Some processes that utilize a flowing liquid require that the temperature of the liquid be controlled precisely. However, the various components of the system. such as valves, tubing, chambers, cause transient cooling affects which absorb heat w-hen changing from hot to cold liquids. Additionally, precise temperature control ot a flowing liquid becomes extremely difficult when the flow rate and incoming temperature of the liquid varies during the process.

U. S. Patent No. 5,1 30,920 (Gebo) discloses an adaptive process control system especially for controlling the temperature of flowing liquids. The system uses an adaptive process control system based upon the feed forward

controller. The control function is adapted by the feedback controller. That is, the system exercises adaptive control on the feed forward control function based upon the feedback loop. The feed forward controller alone or together with the feedback controller provides an output for controlling the process variable of interest.

Brief Summary of the Invention The present invention is directed to an improved method and apparatus for controlling the outgoing temperature of a flowing liquid in which the incoming temperature and flow rate of the liquid may vary.

In one embodiment, the method for controlling the temperature of a flowing liquid in a process includes identifying process parameters in a controller.

The process parameters typically comprise a set point temperature and data corresponding to a specific heat and density of the liquid. One or more sensors permit the controller to monitor the incoming temperature, outgoing temperature and flow rate of the liquid at the heater. A theoretical power required for the heater to adjust the outgoing temperature of the liquid to be substantially equal to the set point temperature is determined. A heater efficiency correction factor derived from information comprising the incoming liquid temperature and flow rate is also determined. R calculated power is derived from information comprising the theoretical power and the heater efficiency correction factor. A final power to be applied to the heater is derived from information comprising the calculated power and an error term. The error term is substantially continuously derived by a control algorithm using information comprising a proportional term, an integral term and a derivative term. The derivative term is derived from information comprising the difference between the set point temperature and the outgoing temperature of the liquid (TsET-ToL-r). The final power is applied to the heater to adjust the outgoing temperature of the liquid to be substantially equal to the set point temperature.

As used herein,"substantially equal to the set point temperature"or similar statements refer to an outgoing temperature of the liquid being within an acceptable temperature range. The acceptable temperature range can vary from process to process and liquid to liquid. As used herein,"identifying process parameters in the controller"refers to entering the heat capacity and density of the liquid into the controller or identifying the liquid and allowing the controller to correlate the identity of the liquid with heat capacity and density values previously stored. In one embodiment, the heat capacity and density for a given liquid is entered or stored as a composite number.

In one embodiment, the step of determining the theoretical power (TTHEORETICAL) comprises determining how much power is required to heat the liquid from TIN to TSET. Theoretical power is typically calculated by identifying the specific heat and density of the liquid being used and multiplying the flow rate, the density and the specific heat of the liquid by the difference between the incoming temperature and the set point temperature (TSET-TIN) using the equation PTHERORETIC.-\L = QFLONV PCP (TSET-TIN) The step of determining the heater efficiency comprises empirically deriving an equation defining heater efficiency for a given liquid, a given flow rate, a given temperature. or a combination thereof. In one embodiment, the step of determining the heater efficiency correction factor (f comprises the step of empirically deriving constants for a quadratic equation defining heater efficiency in terms of variables comprising flow rate (QFLOW) and temperature. The temperature term is typically the difference between the incoming temperature and the set point temperature (TSET-TIN), such as for example/= ao + al (TsET-TN) + a2 (TsET-Tr) 2+ a3QFLOW + a4QFLoNv'a5 (TSET-TU\ :) QFLOW The step of determining the calculated power comprises the step of evaluating the quadratic equation and modifying the theoretical power by the heater efficiency correction factor. In one embodiment, it is possible to have different heater efficiency correction factors (or different constants for the quadratic equation) for different liquids or set point temperatures. It is also

possible for the controller to automatically recalculate the heater efficiency correction factors or constants to compensate for changes in the actual efficiency of the heater.

The error term for calculating the final power is determined by using a control algorithm, such as Proportional Integral Derivative (PID) control. The proportional term is proportional to the difference between the set point temperature and the outgoing temperature of the liquid (TSET-Tou-r). The integral term is the cumulative area under the curve between the outgoing temperature and the set point temperature for a given time interval. The derivative term provides the slope or approach rate of the outgoing temperature toward the set point temperature. In one embodiment, the contribution from the integral term is not added to the cumulative area during time intervals where the incoming temperature is not within a predetermined range of the set point temperature. In one embodiment, the predetermined range is where the incoming temperature of the liquid is between 75% and 125% of the set point temperature.

The present invention is also directed to controlling the temperature of a flowing liquid to process microelectronic components, such as integrated circuits, components for flat panel displays, precursor components including silicon wafers, and various other electrical components. The microelectronic components or their precursors are located in a processing chamber or an immersion bath. The flowing liquid within the desired temperature range is applied to the microelectronic components or their precursors.

The present invention is also directed to an apparatus for processing microelectronic components with a flowing liquid. The apparatus includes a processing chamber or an immersion bath. A controller has process parameters typically comprising a set point temperature and data corresponding to a specific heat and density of the liquid. The liquid flows through a heater to the processing chamber or the immersion bath. Process sensors are positioned to monitor an incoming temperature, an outgoing temperature and a flow rate of the liquid at a

heater. The controller includes means for determining a theoretical power comprising multiplying the difference between the incoming temperature and the set point temperature by the flow rate, specific heat and density of the liquid; means for determining a heater efficiency correction factor derived from information comprising the incoming temperature and flow rate of the liquid and for adjusting the theoretical power with the heater efficiency correction factor to determine a calculated power; and means for calculating a final power derived from information comprising the calculated power and an error term. The error term is substantially continuously determined by a proportional term, an integral term and a derivative term. The derivative term is derived from information comprising the difference between the set point temperature and the outgoing temperature of the liquid (TSET- ToLT.) A power controller is provided for applying the final power to the heater. An applicator in the process chamber or the immersion bath is positioned to apply the liquid to the microelectronic components.

Brief Description of the Several Views of the Drawin Figure 1 is a schematic illustration of the flowing liquid temperature control system in accordance with the present invention.

Figure 2 is a sample graph illustrating one mode of operation for the flow-ing liquid temperature control system of Figure 1.

Figure 3 is a schematic illustration of a process utilizing the flowing liquid temperature control system of the present invention.

Figure 4 is a sample graph illustrating a specific application of the present flowing liquid temperature control system.

Detailed Description of the Invention Figure I is a schematic illustration of a flowing liquid temperature control system 20 in accordance with the present invention. A liquid flows through

the inlet pipe 22 to a heater 24 that applies heat to the liquid at or near a control point 26. After heating, the liquid continues through the outlet pipe 28 to a process 30.

The incoming temperature of the liquid is measured ahead of the control point 26 by a temperature sensor 32. The outgoing temperature of the liquid is measured by temperature sensor 34. In the illustrated embodiment, the temperature sensors 32,34 are located within about 30 centimeters of the control point 26 so that the time delay between the temperature measurements is minimal and the temperature at the sensor 32 is substantially the same as the temperature of the liquid at the control point 26.

Suitable temperature sensors are available from Omega Engineering Incorporated located in Stamford, CT under the model number W2103. In the illustrated embodiment, the temperature sensors are encapsulated in an inert material such as TeflonS) or PVDF (polyvinylidinefluoride).

A flow sensor 36 is located along the inlet pipe 22 to sense the flow rate of the liquid through the system 20. In the illustrated embodiment, the incoming flow rate of the liquid through the inlet pipe 22 is substantially the same as the outgoing flow rate through the outlet pipe 28. A suitable fiow sensor is available from Futurestar Corp. located in Edina, MN under model number Futurestar TPW.

The output from the temperature sensors 32,34 and fiow sensor 36 are maintained by controller 40. The controller 40 includes an input/output device 42, such as a touch screen or a keyboard and display. The controller is connected to a power controller 44 that controls the connection of a power supply 45 (e. g., 110 volts, 220 volts, 440 volts) to the heater 24. A power controller 44 suitable for use in the present invention is available from Control Concepts of Chanhassen, MN under model number 1029C. A controller 40 suitable for use in the present flowing liquid temperature control system 20 is the 25 MHz-CPU card available from Win Systems of Arlington, Texas under part number M486SX25BM-0479A. In the illustrated embodiment, the sensors 32,34,36 are polled by the controller 40 about every 40 milliseconds to about every 2 seconds.

A variety of devices may be used for the heater 24, such as the microwave generator described in U. S. Patent No. 5,130,920. In the embodiment illustrated in Figure 3, the heater 24 is about 25 feet of 3/8 inch OD and'4 ID perfluoroalkoxy (PFA) tubing 70 coiled around three 2 kilowatts elongated infrared lamps 72 available from General Electric under part number QH2M-T31 CL/VB. A reflective shell 74 can be wrapped around the assembly of lamps and tubing to minimize heat loss. The tubing 70 is available from Fluoroware of Chaska, Minnesota. The heater utilized in the spray acid processor available under the trade name MERCUREY from FSI International of Chaska, Minnesota is also suitable for use in the present flowing liquid temperature control system 20.

The method of the present flowing liquid temperature control system utilizes the controller 40 to provide feed forward control in response to the incoming temperature (Trx) and flow rate (QFLOW) of the liquid as measured by the sensors 32, 36 and feedback control in response to the outgoing temperature (Tour) of the liquid as measured by the sensor 34. An operator uses the input/output device 42 to enter a set point temperature (TSET) and identify the liquid. The identity of the liquid can be correlated with a heat capacity (Cp) and density (p) for that liquid. The heat capacity (Cp) and density (p) for that liquid can be stored in the controller 40 as a composite number. In an alternate embodiment, the input/output device 42 is utilized for entering a recipe that identifies a series of set point temperatures, time intervals, flow rates, liquids, etc. that are used through a segment of the process 20.

Once the set point temperature and identity of the liquid are provided to the controller 40, the controller 40 monitors the temperature sensor 32 and flow sensor 36 and performs a feed forward calculation to predict the theoretical power (PTHERORETIC. AL) required to adjust the outgoing temperature (Tour) of the liquid to be substantially equal to the set point temperature (TSET). In the illustrated embodiment, the PTHEROREnC. AL is calculated by the equation : PTHERORETICAL = QFLOW PCP(TSET-TN)

where QFLOW is the flow rate, p is the density of the liquid, and Cp is the heat capacity of the liquid. This feed forward calculation of the theoretical power required to adjust the outgoing temperature of the liquid to the set point temperature, however, excludes systemic factors such as heat loss, non-uniformity of the heating elements, etc.

The second step in the feed forward process is to determine a heater efficiency correction factor (f for the heater 24 and apply it to the PTHERORETICAL, t0 calculate the power (PCALCULATED) required. The heater efficiency correction factor can be determined empirically for a given liquid, a given flow rate, a desired temperature rise, or combinations thereof. In the illustrated embodiment, the heater efficiency correction factor (1), is a quadratic equation of flow rate (QFLOW) and the difference between the incoming temperature and the set point temperature of the liquid (TSET-T>,). The heater efficiency correction factor (/) can be calculated based on the equation: /= ao + ai (TsET-T) + a2 (TSET-TIN) + a3QFLOW + a4QFLOW +a5 (TSET-TIN) QFLOW where ao, al. a2, a3, a4, a5 are constants. The constants ao, al, a2, a3, a4, a5 are empirically derived using known techniques. For example, in one embodiment the constants ao, ai, a2, a3. a4, a5 are empirically derived by operating the system 20 at a fixed incoming temperature (TIN) and three different final power levels (PptNAJb) and flow rates (QFLOW) until the outgoing temperature (Tour) stabilizes. A least-square fit analysis is performed on the nine resulting data points to determine the constants ao, ai, a2, a3, a4, a. ;. This analysis can optionally be performed automatically by the controller to periodically update the constants ao, a,, a2, a3, a4, as to monitor and/or compensate for changes in the heater efficiency over time.

The constants ao, a1, a2, a3, a4, as preferably do not change during operation of the present flowing liquid temperature control system. However, in one embodiment, the constants ao, a1, a2, a3, a4, as may optionally be different for various liquids or set point temperatures. The controller 40 can select the appropriate

constants based on the identity of the liquid or set point temperature provided through the input/output device 42 at the beginning of the process.

The heater efficiency correction factor is intended to compensate for heat loss by the various components of the system 20, such as the heater 24, pipes 22, 28, temperature sensor 34, and various other factors. In most applications, the heater efficiency correction factor (, 0 is a number greater than 1, such as 1.1 or 1.2. In the illustrated embodiment, the calculated power is determined by the equation: PcALCt.'LATED=PTHEORET ! CAL * Although the calculated power should accurately reflect the power required to adjust the outgoing temperature of the liquid to the set point temperature, in actual application, external forces can cause the outgoing temperature to vary from the set point temperature. That is, external forces tend to cause the heater 24 to overshoot or undershoot the set point temperature. These external forces include, but are not limited to aging of the infra-red lamps 72, variations in the heater 24 or power controller 44 due to normal manufacturing tolerances and variations in the actual heater efficiency correction factorffrom the simple quadratic approximation. The calculated power (PC ALCUL. A. TED) also does not include the power required to heat the reflective shell 74 after a change in set point temperature (Tset). In order to minimize the variation in Tocr from TSET of the heater 24, a feedback loop utilizing a control algorithm to determine an error term. In the illustrated embodiment, the control algorithm is Proportional Integral Derivative (PID) control. The error term is calculated using the equation: Error = P(TSET - TOUT) +I#(TSET - TOUT) + D((TSET - TOUT)/#TIME). where (TsET'Tour)'s the difference between the outgoing temperature of the liquid and the set point temperature and P, I and D are constants. For some embodiments, one or more of the constants P, I and D may equal 1.

Application of this Error equation is illustrated in Figure 2. Figure 2 is an exemplary illustration of the operation of the Error equation in the present system

20 where the curve is Tour The proportional term P (TSET-Tour) is proportional to the distance between the outgoing temperature (Tour) and the set point temperature (TSET), such as for example the distances 50a, 50b multiplied by the constant P. The integral term is proportional to a running total or cumulative area between the curve of the outgoing temperature Tour curve and TSET for a given time interval, such as for the time interval up to 52a, 52b. The derivative term is proportional to the slope or approach rate of the outgoing temperature of the liquid to the set point temperature TSET, such as for example the slope of the tangents 54a, 54b to Tour In one embodiment of the present invention, contributions from the integral term are not added to the running total unless the outgoing temperature (Tour) is within a predetermined range of the set point temperature (TSET). In the illustrated embodiment, the integral term is not added to the running total until the outgoing temperature is between 75% and 125% of the set point temperature. In another embodiment. the integral term is set to zero until the outgoing temperature is between 85% and 115% of the set point temperature.

The final power (PFIN. L) necessary for the heater 24 to adjust the temperature of the flowing liquid to the set point temperature is calculated by the equation: PFIN.AL = PCALCULATED (l+Error) The final power requirement is communicated to the power controller 44, which in turn controls the connection of a 220-volt power supply to the heater 24.

Figure 3 illustrates application of the present flowing liquid temperature control system 20. Figure 3 schematically illustrates a spray acid processor available under the trade name MERCURYX available from FSI International for processing microelectronic components, such as integrated circuits, components for flat panel displays, precursor components including silicon wafers, and various other electrical components 62 located in a process chamber 64. The process chamber 64 includes a spray post 66 for applying various chemicals, such as

de-ionized water, HF, H202, HCI, NH4OH, H2S04, to the components 62. These chemicals flow through the heater 24 that is attached to the control 40, discussed above. In the illustrated embodiment, the various chemicals are delivered to the heater at a temperature at or below the set point temperature. Another application of the present flowing liquid temperature control system 20 is disclosed in a commonly assigned U. S. Patent application entitled A METHOD AND SYSTEM TO UNIFORMLY ETCH SUBSTRATES USING AN ETCHING COMPOSITION COMPRISING A FLUORIDE ION SOURCE AND A HYDROGEN ION SOURCE (Christianson et al., attorney docket no. 15676-217185), filed on the same date herewith.

Example Figure 4 is a sample graph illustrating the operation of the present flowing liquid control system for processing microelectronic components. This example was conducted using the heater and spray acid processor available under the trade name NIERCURYt) from FSI International of Chaska, Minnesota. The liquid was about 2° o by volume deionized water, about 2% by volume HF and about 96% by volume eth-lene glvcol. The flow rate (QFLOW) was about 4 liters/minute and the set point temperature (TSE1) was about 85 degrees C. The upper curve illustrates ToLT and the lower curve illustrates T-.

Initial Tour is about equal to TIN. During the three-minute time index from about 770 seconds to about 950 seconds, Tol, oscillates around TsEr During the two and a half minute time index from about 950 seconds to about 1100 seconds, the outgoing temperature of the liquid being is within an acceptable temperature range, even though the incoming temperature (T,,,) declines about 20 degrees C during that period.

All patents and patent applications disclosed herein, including those disclosed in the background of the invention, are hereby incorporated by reference.

While several embodiments of the present invention have now been described, it will be apparent to those of ordinary skill in the art that various changes and modifications may be made without deviating from the inventive concept set forth above. Thus, the scope of the present invention should not be limited to the structures described in this application, but only by the structures described by the language of the claims and the equivalents of those structures.