WO2004007800A1 - Thermal processing apparatus and method for evacuating a process chamber - Google Patents

Abstract

A thermal processing apparatus (100) is provided comprising a process chamber (102) and a vacuum system (104) for evacuating the process chamber. The vacuum system comprises a pump unit (124) in flow communication with the chamber and a valve assembly (126) between the pump unit and the chamber for controlling a mass flow from the chamber to the pump unit. The valve assembly comprises a main vacuum valve (130) and one or more by-pass valves (132). The one or more by-pass valves are connected to a vacuum line (128) via by-pass lines (133) and in parallel flow in relation with the main vacuum valve.

Description

THERMAL PROCESSING APPARATUS AND METHOD FOR EVACUATING A PROCESS CHAMBER

RELATED APPLICATION The present application claims the benefit of and priority from commonly assigned U.S. Provisional Patent Applications Serial Nos. 60/396,536, entitled Thermal Processing System, and filed July 15, 2002, and 60/428,526, entitled Thermal Processing System and Method for Using the Same, and filed November 22, 2002, both of which are incorporated herein by reference in their entirety.

FIELD OF THE INVENTION The present invention relates generally to the field of semiconductor equipment and processing. More specifically, the present invention relates to a thermal processing apparatus having a vacuum system and method of evacuating down a process chamber.

BACKGROUND Thermal processing apparatuses are commonly used in the manufacture of semiconductors and integrated circuits (ICs). Thermal processing of semiconductor wafers includes, for example, deposition, etching, heat treating, annealing, diffusion etc. Some processes such as etching and chemical vapor deposition (CVD) are performed under low pressure or vacuum conditions.

During the processes under low pressure or vacuum conditions, the process chamber is evacuated from an initial pressure to an operating pressure. For example, the process chamber may initially be at atmospheric pressure for loading wafers, then evacuated to an operational pressure in the milli-torr range. The initial evacuation cycle for a process is sometimes referred to as a "pump down cycle". Typically a pump down cycle is accomplished by using a vacuum pump in flow communication with the process chamber.

It is desirable to have the pump down cycle complete as quickly as possible in order to achieve minimal overall cycle times during the manufacturing process. However, if the speed of pump down gas in the chamber is too fast, turbulent flow motion is generated. Such turbulent motion tears away particles that have been deposited on the chamber wall, wafer carrier, or other parts of the thermal processing apparatus, and transports and redistributes them in areas that can be critical, for example, on the substrates where integrated circuits are being made. The conventional vacuum systems and methods for evacuating a process chamber are limited by the deleterious physical effects of particulate contamination during the pump down process.

SUMMARY OF THE INVENTION

The present invention provides a thermal processing apparatus having a vacuum system and a method that provides for fast pumping down of a process chamber without generating particulate contamination.

The thermal processing apparatus of the invention comprises a process chamber and a vacuum system for evacuating the process chamber. The vacuum system comprises a pump unit in flow communication with the chamber and a valve assembly between the pump unit and the chamber for controlling a mass flow from the chamber to the pump unit. The valve assembly comprises a main vacuum valve and one or more by-pass valves. The one or more by-pass valves are connected to the vacuum line via by-pass lines and in parallel flow relation with the main vacuum valve.

The method of evacuating a process chamber of the invention comprises determining a critical maximal mass flow rate and maintaining substantially the maximal gas flow rate during the evacuation of the chamber. The a critical maximal mass flow rate can be empirically determined based a Reynolds number of less than about 2000 to ensure laminar flow. Alternatively, the critical maximal mass flow rate can be determined based on shear force being substantially equal to gravity force for a preidentified particulate type of concern. The step of determining a critical maximal gas flow rate can be accomplished by calculating according to the following formula:

500μA m_r

wherein m_max represents critical maximal mass flow rate, μ represents gas viscosity, A represents area of cross section of the annulus flow area formed between a wafer stack and chamber wall, and h represents distance between a wafer stack and a wall of the chamber.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention becomes better understood upon reading of the following detailed description of the invention and upon reference to the following drawings, in which: FIG. 1 is a schematic showing a thermal processing apparatus having a vacuum system according to one embodiment of the present invention.

FIG. 2 is a schematic showing velocity profiles of a laminar flow and turbulent flow near a chamber surface.

FIG. 3 is a schematic showing a flow profile between a wafer stack and a process chamber wall.

FIG. 4 is a plot showing flow regimes for particles of different size at different pressures at temperature of 650°C.

FIG. 5 is a plot showing an optimized pump down pressure trajectory according to one embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION The present invention is directed to a thermal processing apparatus and method for processing a batch of semiconductor substrates or wafers that significantly reduce processing cycle times and improve process uniformity. With reference to FIG. 1, the thermal processing apparatus of the present invention will now be described. For purpose of clarity, some elements not relevant to the present invention are omitted to avoid complicating the description. In general, the thermal processing system 100 comprises a process chamber 102 and a vacuum system 104 in flow communication with the chamber 102 for evacuating the chamber to low pressure or vacuum conditions.

The process chamber 102 can be made in any size suitable for processing either a large batch or mini batch of wafers. The process chamber 102 can be a thermal processing chamber such as a chamber for chemical vapor deposition

(CVD) and etching etc. The process chamber 102 is preferably made of a material that is capable of withstanding thermal and mechanical stresses of high temperature and vacuum operation, and resistant to erosion from gases and vapor used or released during processing. Preferably the process chamber 102 is made of quartz or silicon carbide.

The process chamber 102 is preferably a vertical reaction chamber. An entrance is provided at the lower portion of the chamber 102 for conveying a carrier or boat 106 carrying a batch of wafers 108 into and out of the process chamber 102 through a movable pedestal 110. The upper portion of the chamber 102 is closed to form a reaction zone.

The vertical reaction chamber 102 preferably has an inner tube or liner 112 and an outer tube 114. The inner tube or liner 112 is opened at the lower end and at least partially opened at the upper end. The inner tube 112 directs gas flow and prevents diffusion of impurities outside of the liner on the wafers 108. The boat 106 carrying the wafers 108 is encompassed within the inner tube 112. The outer tube 114 is closed at the upper end but open at the lower end. An annular passageway 116 is formed between the inner and outer tubes 112 and 114 for exhausting a gas in a downward direction. The inner and outer tubes 112 and 114 are hermetically connected to a plenum 118 by seals, such as o-rings. The plenum 118 is made of a material having thermal durability and corrosion resistance, such as stainless steel or quartz. The plenum 118 is sized in a short cylindrical form including an upper flange, a bottom flange and sidewall. The upper flange is adapted to receive and support the outer tube 114 of the chamber 102. The bottom flange is adapted to receive and support the inner tube or liner 112 of the chamber 102. An injection assembly 120 is disposed within the plenum 118 for introducing a process or purge gas into the chamber 102. The injection assembly 120 is disposed in such a manner as to inject the process or purge gas into the reaction zone encompassed by the inner tube 112 of the chamber 102. An exhaust port 122 is provide at the sidewall of the plenum 118 for connecting the process chamber 102 and plenum 118 to the vacuum system 104. The exhaust port 122 is disposed in such a manner as to communicate with the annular passageway 116 formed between the inner and outer tubes 112 and 114. The vacuum system 104 comprises a pump unit 124 and a valve assembly

126. The pump unit 124 is connected to the exhaust port 122 via a vacuum line 128 and in flow communication with the process chamber 102. The valve assembly 126 is disposed between the pump unit 124 and process chamber 102 on the vacuum line 128. The valve assembly 126 controls a gas flow from the process chamber 102 to the pump unit 124 and is capable of completely isolating the process chamber 102 from the pump unit 124.

The pump unit 124 can comprise a dry pump, booster pump, Roots pump, or any combination of two or more of the pumps. For example, the pump unit 124 can be a dry pump system such as Model iQDP80/iQMB1200F available from BOC Edwards in Boston, Massachusetts. The peak pump speed of such dry pump system is 15.6 m³/min around 10 torr, and the ultimate vacuum is 5 mtorr.

The valve assembly 126 comprises a main vacuum valve 130 disposed on the vacuum line 128. The main vacuum valve 130 can be any type that provides complete isolation between the process chamber 102 and the pump unit 124. Preferably the main valve 130 is a gate valve available in the prior art. The gate valve can be of an on/off type that operates between a fully open and fully close position. Preferably, the gate valve is a DC motor driven valve that can be adjusted in any position between a fully open and close position.

In one embodiment, the valve assembly 126 comprises a by-pass valve 132. The by-pass valve 132 is disposed at the vacuum line 128 via a by-pass line 133 and in parallel flow relation with the main vacuum valve 130. In another embodiment, the valve assembly 126 may further comprise a second by-pass valve (not shown) that is disposed at the vacuum line 128 via a by-pass line (not shown) and in parallel flow relation with the main vacuum valve 130. The combination of a main vacuum valve and one or more by-pass valves can achieve fast pump down of the process chamber with a substantially constant mass flow rate as described below.

Preferably the by-pass valves 132 are orifice valves having a predetermined opening area. For example, the valve assembly 126 may comprise a first and a second orifice valve, each of which has an opening area. Preferably, the second orifice valve can have an opening area greater than that of the first orifice valve. For the example, the first orifice valve has an opening with a diameter ranging from about 1 to 3 mm. The second orifice valve can have an opening with a diameter ranging from about 2 to 4 mm. By using orifice valves having different opening areas, an desirable pump down process can be achieved as described below.

Alternatively, a mass flow controller (not shown) can be used to maintain a constant mass flow rate from the chamber 102 to the pump unit 124 and completely isolate the chamber 102 from the pump unit 124.

The vacuum system 104 may further include a throttle or butter fly valve 134 between the valve assembly 126 and the pump unit 124. The throttle or butter fly valve 134 has a disk 136 that changes its angle by a controlled rotary actuator (not shown). The throttle valve 134 is used to stabilize the chamber pressure to a predetermined value at a predetermined flow rate of process chemicals. The throttle valve 134 also changes the vacuum conductance of the valve assembly 126 and thus the chamber pressure. This can be done by adjusting the angle of the actuated disk 136. When the disk 136 is parallel to the flow, the throttle valve 134 provides the highest vacuum conductance and thus the lowest chamber pressure at the steady state. When the disk 136 is perpendicular to the flow, it produces the lowest conductance, resulting the highest chamber pressure at the steady state. The thermal processing apparatus 100 may further comprise one or more temperature sensing elements (not shown) disposed between the inner and outer tubes 112 and 114 for monitoring the temperature within the process chamber 102. The thermal processing apparatus 100 may also comprise one or more pressure sensors or transducers (not shown) configured to measure the pressure within the process chamber 102.

In operation, the first by-pass valve 132 is opened first from an initial condition where both the main vacuum valve 130 and the second by-pass valve (not shown) are closed to perform a first stage of the pump down process. Then the first by-pass valve 132 is closed and the second by-pass-valve is opened simultaneously at a second pump down stage when the main vacuum valve 130 is closed. At the following third pump down stage, both the first and second by-pass valves are opened while the main vacuum 130 remains closed. Finally the first and second bypass valves are closed and the main vacuum valve 130 is opened to perform a hard pump down stage. During the final pump down stage, the throttle valve 134 is adjusted to stabilize the pressure within the process chamber 102.

A method of evacuating or pumping down a process chamber without generating particulate contamination will now be described with reference to FIGS. 2 to 5.

Each semiconductor process such as chemical vapor deposition (CVD) produces a small amount of particles. Over times, the particles accumulate on the chamber walls, as shown in FIG. 2. When there exists no gas flow within the chamber, the adhesion force between the particles and the chamber wall keeps the particles on the wall and prevents them from migrating to the semiconductor wafers.

A pump down process produces a gas flow along the chamber wall. The flow creates a shear force on particles formed on the wall. If the shear force is greater than a critical value, the particles will be dislodged from the chamber wall. Once becoming loose, the particles flow with the gas and can deposit on the semiconductor wafers, causing particle contamination.

As described above, it is desirable to have the pump down process complete as quickly as possible to achieve minimal overall cycle times during the manufacturing process. However, the speed of pump-down gas in the chamber must be controlled to avoid particulate contamination. The inventor has developed flow models for the pump down process and found that if these models are followed, the pump down cycle time can be significantly reduced, and particles undesirably deposited on the chamber wall and wafer carrier or boat will not be disturbed and therefore particulate contamination during the pump down process is avoided.

In general, the method of evacuating a process chamber of the invention comprises determining a critical maximal gas flow rate which minimizes dislodging of particulate contaminates formed in the chamber, and maintaining substantially the maximal gas flow rate during the evacuation of the chamber.

The shear force on a particle on the chamber wall depends on the types of the flow near the chamber wall. A turbulent flow refers to a flow where both the flow velocity and flow pressure fluctuate randomly. A laminar flow refers to a flow where both the flow velocity and flow pressure are stable. FIG. 2 schematically shows velocity profiles of a laminar flow and turbulent flow near a chamber wall. In a turbulent flow, the mean flow velocity increases rapidly away from the wall. In a laminar flow, the flow velocity increases only slowly away from the surface. At the center position of a particle, the velocity of a turbulent flow (U_T) is much greater than that of a laminar flow (U ). AS a result, a turbulent flow produces a much greater shear force on particles than a laminar flow and should thus be avoided during the pump down process. The Reynolds number (Re) indicates whether a gas flow between a wafer stack and chamber wall is turbulent or laminar. In accordance with the general gas law, the Reynolds number can be expressed by the following equation, with reference to FIG. 3:

4pUh

Re = (1)

where p is the gas density, U is the mean flow velocity, h is the distance between the wafer stack and chamber wall, and μ is the gas viscosity. To ensure laminar flow, the Reynolds number should be kept below 2000 during the entire pump down process as empirically determined.

During the pump down process, the gas density (p) decreases as the chamber pressure drops. On the other hand, the gas viscosity (μ) is independent of the chamber pressure. As a result, as chamber pressure decreases, the maximum allowable gas flow velocity (U) increases.

The Reynolds number can also be expressed by the following equation using mass flow rate according to the gas law:

Re = — < 2000 (3) μA

where m is the mass flow rate, A is the cross section area of the annulus flow area formed between a wafer stack and chamber wall, and h and μ are defined as above. Equation (3) shows that to keep the Reynolds number below 2000 during the pump down process, the mass flow rate should be kept below a constant value. Since the gas viscosity increases with temperature, e.g., 2.7x10^" Ns/m at 300°C and 3.9x10^"5 Ns/m² at 700°C, a higher mass flow rate is allowed at a higher temperature to avoid a turbulent flow.

As described above, the maximum shear force on particles created by the gas flow should be smaller than a critical value to avoid dislodging of the particles from the chamber wall. The flow velocity (U ) at the center position of a particle can be expressed by the following equation:

U_L - (4) h

where d is the particle diameter, and U and h are defined as above.

The share force created by a gas flow on a particle varies depending on the relative size of the particle and the molecular mean free path (λ) of the flowing gas.

The molecular mean free path (λ) refers to the distance that a molecule travels between two consecutive collisions and can be expressed by the following equation:

where k is the Boltzmann constant, k=1.38xl0^"23 J/k, T is the chamber temperature in Kelvin, σ is the cross section area of molecular collisions, σ = 4.3xl0^"19 m², and P is the chamber pressure.

When the particle diameter is substantially larger than the mean free path, for example, d > 4.5λ, the gas molecules behave as a continuum medium. In this continuum flow regime, the shear force on the particle is expressed by the following equation:

„ ,_{τ τ} 4.5πd²μU

F_c = 3πdμU_L = — *- (6) h where d, μ, U, U_L and h are defined as above.

When the particle diameter is comparable or smaller than the mean free path, for example, d < 4.5λ, the gas molecules behave as individual molecules colliding the particle. In this molecular flow regime, the shear force on the particle is expressed by the following equation:

^d²P^ U

F_m = -²P- — = = --dd^jP- (7) 9 a 3 ah

where a is the mean speed of random motion of gas molecules, which is expressed by the following equation:

where R is the gas constant. For example, the gas constant (R) for nitrogen is 297 J/kg-K.

FIG. 4 is a plot showing the flow regimes for different size particles at different pressures at temperature of 650°C. As shown in FIG. 4, a relatively large particle may experience different type of gas flows during the pump down process. For example, at the initial pump down at 760 torr, the gas in the chamber is dense, and the molecular mean free path is small. The flow around a particle of two- micron diameter on the wall is continuum. As pump down continues, the chamber pressure decreases. When the chamber pressure drops to around 360 torr, the molecular mean free path becomes comparable to the particle diameter. The gas flow around the particle of two-micron diameter becomes free molecular flow.

Small particles of sub-micron are the main concerns of the current and future semiconductor processing technologies. These small particles experience free molecular flow around them throughout the entire pump down process. Equation 7 characterize the shear force produced by the free molecular flow on the particles, which can be further expressed by the gas density and mass flow rate according to the following equation:

Equation (9) shows that the shear force on a particle is proportional to the particle volume (d³) and mass flow rate (m). Accordingly, the ratio (N) of shear force to gravity force can be expressed by the following equation:

F 3.42mVRT

N = ^ (9) Ahp_g

where p_p is the mass density of the particle, g is the gravity acceleration, and R and T are defined as above.

To prevent particles from being dislodged from the chamber wall by the gas flow, the shear force must be comparable or smaller than the gravity force as expressed by the following equation:

Combining equations (3) and (10), the critical maximal mass flow rate can be determined by the following equation:

For thermal processing apparatus where h = 1.5cm, T = 700°C, and p_p =

5 5000000kkgg//mm³ ,, tthhee RReeyynnoollddss nnuummbbeerr eexxpprreesss! ed by equation (3) dictates the maximal flow rate during the pump down process:

500μA mmax = — f- (12) h

According to the general gas law, the pressure in the chamber is related to the mass of gas in the chamber according to the following equation:

PV = MRT (13) where P is the chamber pressure, V is the chamber volume, M is the total mass of gas in the chamber, and T is the chamber temperature. The chamber pressure drops as gas molecules are removed from the chamber as expressed by the following equation:

dP _ RT dM

(14) dt^~ V dt

Combining equations (12) and (14), the maximum pump down rate without producing particulate contamination is expressed by the following equation:

Accordingly, the fastest possible pressure trajectory is expressed by the following equation:

_ RT 500μA_t^a V h

where t is the time of pump down, and P_a is ambient pressure (1.013xl0⁵ Pa), and R, T, V, μ, A, and h are defined as above. In one embodiment, the method of fast pumping down a process chamber comprises determining a critical maximal mass flow rate by calculating according to equation (12), and evacuating the chamber by flowing the gas through a mass flow controller. The mass flow controller is set substantially at the critical maximal mass flow rate so that the mass flow of the gas evacuated from the chamber during the evacuation is substantially equal to the critical maximal mass flow rate throughout the evacuation.

In another embodiment, the evacuation is performed by directing the gas from the chamber through a vacuum system having a valve assembly with an adjustable effective orifice size. The gas flow is controlled by setting the valve assembly to a first effective orifice size during a first segment of the evacuation beginning at a first pressure, and maintaining substantially the maximal mass flow rate during the first segment of the evacuation. Then, the valve assembly is set to a second effective orifice size greater than the first effective orifice size during a second segment of the evacuation beginning at a second pressure, and maintaining substantially the maximal mass flow rate during the second segment of the evacuation. The first effective orifice size is a function of the first pressure and the critical maximal mass flow rate. The second effective orifice size is a function of the second pressure and the critical maximal mass flow rate.

For example, the valve assembly may comprise a main vacuum valve, and a first and second by-pass valve. The second by-pass valve has an opening area greater than that of the first by-pass valve. In operation, the first by-pass valve is opened first and maintains the maximal mass flow rate as calculated according to equation (12) at a first time period. Then the first by-pass valve is closed and the second by-pass valve is opened simultaneously and maintains the maximal gas flow rate at a second time period when the main vacuum valve is closed. Then both the first and second by-pass valves are opened and maintain the maximal gas flow rate at a third time period. Finally, both the first and second by-pass valves are closed and the main vacuum valve is opened simultaneously and maintains the maximal flow rate at a fourth time period. The first through fourth time periods are substantially equal during the pump down process. FIG. 5 is a plot showing an example according to the method described above. The process chamber is initially pumped down from about 760 torr to about 400 torr in 1.47 minutes by using a first orifice valve having an opening diameter of about 2.2mm. Then the chamber is pumped down from about 400 torr to about 120 torr in 1.48 minutes by using a second orifice valve having an opening diameter of about 3.0mm. Then the chamber is pumped down from about 120 torr to about 20 torr in 1.45 minutes by opening both orifice valves. Finally, the chamber is pumped down from to about 20 mtorr in about 1.20 minutes by closing both orifice valves and opening a gate valve.

Of advantage of the present method is the very fast pump down of the chamber without generating particulate contamination. It takes only about 5.6 minutes to pump down the process chamber from about 760 torr to 20mtorr, as compared to conventional pump down methods which typically take more than 20 minutes. The present invention further provides a method of optimizing a pump down process comprising the following steps: a. setting a safety margin (S) to prevent particle generation during the fast pump down process, S>1; b. determining the maximal mass flow rate (m₀) in the pump down process;

500μA m₀ = - (17) hS c. determining the orifice area of a first by-pass valve (A]),

where Ti is the gas temperature at the orifice, which is typically about 400 K; d. defining a pressure Pi, at which the first by-pass valve is closed and the second by-pass valve is opened; e. determining the time (ti) needed to pump down the chamber from pressure P_a to Pi,

where the time constant is

VP. ι = (20)

RTm, determining the orifice area of a second by-pass valve (A );

g. determining a pressure P₂, at which both the first and second by-pass valves are opened;

h. determining the time (t₂) needed to pump down the chamber from pressure PI to P2,

where the time constant is VP, τ₂ = - (23) RTm i. determining the pressure P₃, at which the mass flow rate is m₀ when both the by-pass valves are closed, the gate valve is fully open, and the throttle valve is at the close position. The pressure P₃ depends on the actual vacuum system setup, such as the vacuum line diameter and length, and vacuum conductance of the throttle valve at the close position. It is typically in the range from about 1 torr to about 20 torr; j. determining the time (t₃) needed to pump down the chamber from pressure

P₂ to P₃,

where the time constant is

VP, τ₃ = (23)

RTm,

k. repeat steps of d to j until ti = t₂ = t

As described above, a thermal processing apparatus having a vacuum system and a method of fast pumping down a chamber are provided. The foregoing description of specific embodiments of the invention have been presented for the purpose of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed, and obviously many modifications, embodiments, and variations are possible in light of the above teaching. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.

Claims

CLAIMS We Claim:

1. A thermal processing apparatus comprising: a process chamber and a vacuum system for evacuating the process chamber, the vacuum system comprising a pump unit in flow communication with the chamber and a valve assembly between the pump unit and the chamber for controlling a mass flow from the chamber to the pump unit, wherein said valve assembly comprises: a main vacuum valve; and a by-pass valve, said by-pass valve being connected to the vacuum line via a by-pass line and in parallel flow relation with the main vacuum valve.

2. The thermal processing apparatus of claim 1 wherein the valve assembly further comprises a second by-pass valve connected to the vacuum line via a by-pass line and in parallel flow relation with the main vacuum valve.

3. The thermal processing apparatus of claim 2 wherein said first and second by-pass valves are orifice valves, and the second orifice valve has an opening area greater than that of the first orifice valve.

4. The thermal processing apparatus of claim 1 wherein the vacuum system further comprises a throttle valve between the valve assembly and the pump unit.

5. The thermal processing apparatus of claim 1 wherein said chamber is a chemical vapor deposition chamber.

6. A method of evacuating a process chamber, comprising: determining a critical maximal mass flow rate; evacuating gas from the chamber; and controlling mass flow of the gas evacuated from the chamber during the evacuating step so as not to exceed the critical maximal gas flow rate during the evacuating step while being substantially equal to the critical maximal mass flow rate at at least plural times during the evacuating step.

7. The method of claim 6 wherein the determination of a critical maximal mass flow rate is based on an empirical determination of a Reynolds number of less than about 2000 to ensure laminar flow.

8. The method of claim 6 wherein the determination of a critical maximal mass flow rate is based on shear force being substantially equal to gravity force for a preidentified particulate type of concern.

9. The method of claim 6 wherein the controlling step comprises flowing the gas evacuated from the chamber during the evacuating step through a mass flow controller set substantially at the critical maximal mass flow rate, whereby the mass flow of the gas evacuated from the chamber during the evacuating step is substantially equal to the critical maximal mass flow rate throughout the evacuating step.

10. The method of claim 6 wherein: the evacuating step comprises directing the gas from the chamber through a vacuum system having a valve assembly; and the controlling step comprises: setting the valve assembly to a first effective orifice size during a first segment of the evacuating step beginning at a first pressure, the first effective orifice size being a function of the first pressure and the critical maximal mass flow rate; and setting the valve assembly to a second effective orifice size greater than the first effective orifice size during a second segment of the evacuating step beginning at a second pressure less than the first pressure, the second effective orifice size being a function of the second pressure and the critical maximal mass flow rate.

11. The method of claim 6 wherein the determination of a critical maximal mass flow rate is accomplished by calculating according to the following formula:

500μA max wherein m_max represents critical maximal mass rate, μ represents gas viscosity, A represents area of cross section of the annulus flow area formed between a wafer stack and chamber wall, and h represents distance between a wafer stack and a wall of the chamber.

12. A method of evacuating a process chamber, comprising: determining a critical maximal mass flow rate by calculating according to the following formula:

500μA m_r

wherein m_max represents critical maximal mass rate, μ represents gas viscosity, A represents area of cross section of the annulus flow area formed between a wafer stack and chamber wall, and h represents distance between a wafer stack and a wall of the chamber; and maintaining the maximal mass flow rate by controlling a valve assembly, said valve assembly comprising a main vacuum valve, and first and second by-pass valves, said second by-pass valve having an opening area greater than that of the first by-pass valve, wherein said maintaining of the maximal mass flow is accomplished by opening the first by-pass valve and maintaining substantially the maximal mass flow rate at a first time period; closing the first by-pass valve and opening the second by-pass valve simultaneously and maintaining substantially the maximal mass flow rate at a second time period; opening both the first and second by-pass valves and maintaining substantially the maximal mass flow rate at a third time period; and closing both the first and second by-pass valves and opening the main vacuum valve simultaneously and maintaining substantially the maximal mass flow rate at a fourth time period; wherein said first through fourth time period are substantially equal.