Fluid handling – Systems – Multiple inlet with multiple outlet
Reexamination Certificate
1999-09-20
2004-10-05
Bueker, Richard (Department: 1763)
Fluid handling
Systems
Multiple inlet with multiple outlet
C137S486000, C137S487500, C118S715000
Reexamination Certificate
active
06799603
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to semiconductor process equipment, and more particularly, to a method and systems for controlling gas flow to a semiconductor processing reactor.
BACKGROUND OF THE INVENTION
Semiconductor processing typically involves the formation of one or more layers on a semiconductor substrate. For example, silicon epitaxy, sometimes called epi, is a process in which one or more layers of single-crystal (monocrystalline) silicon are deposited on a monocrystalline silicon wafer.
FIG. 1
is a schematic representation of a semiconductor processing system
10
in accordance with the prior art. As shown in
FIG. 1
, system
10
included a susceptor
12
enclosed within a barrel reactor
14
. Susceptor
12
supported a plurality of substrates
16
, typically monocrystalline silicon wafers.
During processing, substrates
16
were heated with an external radiation source such as tungsten halogen lamps, resistive heating elements and/or RF heaters (not shown).
A process gas was introduced into reactor
14
through one or more injector ports
18
. The process gas typically included trichlorosilane although other process gases besides trichlorosilane sometimes were used depending upon the particular type of layer that was deposited. The process gas reacted with heated substrates
16
resulting in the deposition of layers on substrates
16
as those skilled in the art understand. The spent process gas was exhausted through a vacuum pump
20
to exhaust
23
. Alternatively, the spent process gas was directly exhausted to exhaust
23
and vacuum pump
20
was not used.
Of importance, to insure the consistency and quality of the deposited layers on substrates
16
, the composition and mass flow rate of the process gas delivered to reactor
14
was carefully controlled. For this reason, system
10
included a gas flow control system
21
coupled to injector ports
18
by a process gas line
24
. Gas flow control system
21
was generally located in a gas cabinet
22
located at a distance from reactor
14
.
Located within a gas bottle cabinet
49
were three process gas sources
26
,
28
,
30
and a carrier gas source
50
. Illustratively, process gas sources
26
,
28
,
30
and carrier gas source
50
included compressed gas cylinders containing process gases A, B, C, and carrier gas CG, respectively.
Process gas sources
26
,
28
,
30
were coupled to a gas manifold
38
of system
21
through mass flow controllers (MFCs)
32
,
34
,
36
, respectively, of system
21
. Gas manifold
38
had a plurality of input ports
38
A,
38
B,
38
C, a first output port
38
Y and a second output port
38
Z. MFCs
32
,
34
,
36
controlled and regulated the mass flow rates of flows of process gases A, B, C from process gas sources
26
,
28
,
30
, respectively, to input ports
38
A,
38
B and
38
C, respectively, of gas manifold
38
. Output port
38
Y of gas manifold
38
was coupled to process gas line
24
by valve
40
of system
21
. Output port
38
Z of gas manifold
38
was coupled to an inlet of vacuum pump
20
(generally referred to as exhaust
23
) by valve
42
of system
21
. An outlet of vacuum pump
20
was coupled to exhaust
23
. Alternatively, vacuum pump
20
was not used and output port
38
Z of gas manifold
38
was directly coupled to exhaust
23
by valve
42
.
Carrier gas source
50
was coupled to process gas line
24
through a mass flower controller (MFC)
52
of system
21
. MFC
52
controlled and regulated the mass flow rate of a flow of carrier gas CG from carrier gas source
50
to process gas line
24
.
To illustrate the operation of gas flow control system
21
, assume that a heavily doped P type silicon layer was to be deposited after which a lightly doped P type silicon layer was to be deposited on substrates
16
. In this example, process gas C was a P type dopant gas. Further, process gas B was a source of silicon, e.g., was trichlorosilane.
Initially, to form the heavily doped P type silicon layer, valve
42
was open and valve
40
was closed. Process gases B. C from process gas sources
28
,
30
flowed through MFCs
34
,
36
, respectively, to gas manifold
38
. In gas manifold
38
, process gases B, C mixed (the mixture of process gases B, C is hereinafter referred to as high dopant concentration process gas). The high dopant concentration process gas flowed from gas manifold
38
through valve
42
to exhaust
23
.
As those skilled in the art understand, gas must flow through a mass flow controller (MFC) for a certain period of time after activation of the MFC to allow the mass flow rate of the flow of gas through the MFC to stabilize and to allow the MFC to accurately control the mass flow rate of the flow of gas. Thus, the flow of the high dopant concentration process gas to exhaust
23
continued until the mass flow rates of the flows through MFCs
34
,
36
stabilized. Valve
40
was opened and valve
42
was closed thereby providing the high dopant concentration process gas through process gas line
24
and injector ports
18
into reactor
14
. The high dopant concentration process gas reacted with heated substrates
16
and formed the heavily doped P type silicon layer on each of substrates
16
.
After a predefined time period, valve
40
was closed to stop the flow of the high dopant concentration process gas into reactor
14
and to stop the deposition of the heavily doped P type silicon layer on substrates
16
.
FIG. 2
is a graph of the concentration of the high dopant concentration process gas in reactor
14
verses time after shutting-off the flow of the high dopant concentration process gas to reactor
14
by closing valve
40
.
Referring to
FIGS. 1 and 2
together, time T=0 is at the end of the predefined period when valve
40
was closed. After valve
40
was closed, the concentration of the high dopant concentration process gas gradually decreased in reactor
14
as the high dopant concentration process gas was displaced by carrier gas CG supplied from carrier gas source
50
. In particular, a length of time T=T
1
, e.g., thirty seconds to two minutes or more, after valve
40
was closed passed before the high dopant concentration process gas was fully removed from reactor
14
. Undesirably, the high dopant concentration process gas continued to react and formed a transition layer on the newly formed heavily doped P type silicon layer until the high dopant concentration process gas was fully removed from reactor
14
.
After the high dopant concentration process gas was fully removed from reactor
14
, the lightly doped P type silicon layer was deposited. Valve
42
was opened and process gas A, hereinafter referred to as low dopant concentration process gas, flowed through MFC
32
through valve
42
to exhaust
23
until the mass flow rate of the flow through MFC
32
stabilized. Valve
40
was opened and valve
42
was closed thereby providing the low dopant concentration process gas into reactor
14
. The low dopant concentration process gas reacted with heated substrates
16
and formed the lightly doped P type silicon layer on substrates
16
.
FIG. 3
is a graph of dopant concentration versus depth in a substrate
16
in accordance with the prior art process described above. Referring to
FIG. 3
, the top of the heavily doped P type silicon layer described above (hereinafter referred to as HD layer L
1
) was located at a distance D
1
from a surface of substrate
16
.
Referring to
FIGS. 1 and 3
together, after HD layer L
1
was formed with a desired thickness D
1
, valve
40
was closed to stop the flow of the high dopant concentration process gas to reactor
14
. However, after closing of valve
40
, transition layer TL was formed on HD layer L
1
.
Since the concentration of the high dopant concentration process gas diminished in reactor
14
after valve
40
was closed, the dopant concentration of transition layer TL gradually changed from heavily doped HD at the bottom of transition layer TL to lightly doped LD at the top of transition layer TL. The lightly doped P typ
Bueker Richard
Gunnison McKay & Hodgson, L.L.P.
Hodgson Serge J.
Moore Epitaxial Inc.
LandOfFree
Gas flow controller system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Gas flow controller system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Gas flow controller system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3278450