U.S. patent application number 09/747820 was filed with the patent office on 2001-05-31 for methods of producing doped semiconductors.
This patent application is currently assigned to SEH America, Inc.. Invention is credited to Anderson, Douglas G..
Application Number | 20010001943 09/747820 |
Document ID | / |
Family ID | 22591885 |
Filed Date | 2001-05-31 |
United States Patent
Application |
20010001943 |
Kind Code |
A1 |
Anderson, Douglas G. |
May 31, 2001 |
Methods of producing doped semiconductors
Abstract
Methods for producing doped polycrystalline semiconductors and
for producing doped monocrystalline semiconductors from predoped
monocrystalline and polycrystalline semiconductors. The methods for
producing doped polycrystalline semiconductors may include (1)
providing a reactor for chemical vapor deposition, (2) creating a
vapor within the reactor that includes a silicon compound and a
preselected dopant, and (3) providing a substrate, exposed to the
vapor, onto which the silicon and the dopant in the vapor are
deposited to form doped polycrystalline silicon. The methods for
producing doped monocrystalline semiconductors may include (1)
selecting a first amount of a first semiconductor, the first
semiconductor having a first concentration of the dopant, (2)
selecting a second amount of a second semiconductor, and (3) using
the first and second amounts to grow a monocrystalline
semiconductor having a preselected concentration of the dopant.
Inventors: |
Anderson, Douglas G.;
(Vancouver, WA) |
Correspondence
Address: |
KOLISCH, HARTWELL, DICKINSON
McCORMACK & HEUSER
200 Pacific Building
520 S.W. Yamhill Street
Portland
OR
97204
US
|
Assignee: |
SEH America, Inc.
|
Family ID: |
22591885 |
Appl. No.: |
09/747820 |
Filed: |
December 22, 2000 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09747820 |
Dec 22, 2000 |
|
|
|
09163858 |
Sep 30, 1998 |
|
|
|
6171389 |
|
|
|
|
Current U.S.
Class: |
117/13 ; 117/2;
117/200; 257/E21.106 |
Current CPC
Class: |
C30B 15/04 20130101;
C30B 11/04 20130101; H01L 21/02576 20130101; Y10T 117/10 20150115;
H01L 21/02381 20130101; H01L 21/02579 20130101; C30B 13/10
20130101; H01L 21/02532 20130101; H01L 21/0262 20130101 |
Class at
Publication: |
117/13 ; 117/2;
117/200 |
International
Class: |
C30B 015/00; C30B
021/06 |
Claims
I claim:
1. A method of forming a monocrystalline semiconductor, the method
comprising: selecting a desired final concentration of a dopant for
the monocrystalline semiconductor, where the dopant is selected
from the group consisting of n-type dopants and p-type dopants;
selecting a first amount of a first semiconductor, the first
semiconductor having a first concentration of the dopant, wherein
the first concentration is higher than the desired final
concentration of the dopant; selecting a second amount of a second
semiconductor; and using the first and second amounts to grow the
monocrystalline semiconductor; wherein the first amount of the
first semiconductor and the second amount of the second
semiconductor are selected so that the monocrystalline
semiconductor has the desired final concentration of the
dopant.
2. The method of claim 1, the second semiconductor having a second
concentration of the dopant, wherein the second concentration is
lower than the desired final concentration of the dopant.
3. The method of claim 1, wherein the first and second
semiconductors are polycrystalline.
4. The method of claim 1, wherein the monocrystalline semiconductor
includes silicon.
5. The method of claim 1, wherein the monocrystalline semiconductor
includes gallium arsenate.
6. The method of claim 1, wherein the desired final concentration
of the dopant is chosen to correspond to a preselected
resistivity.
7. The method of claim 1, wherein the step of using the first and
second amounts includes employing the Czochralski method.
8. The method of claim 1, wherein the dopant is selected from the
group consisting of diborane (B.sub.2H.sub.2), phosphine
(PH.sub.3), and arsine (AsH.sub.3).
9. The method of claim 1, wherein the monocrystalline semiconductor
is a p-type semiconductor.
10. The method of claim 1, wherein the monocrystalline
semiconductor is an n-type semiconductor.
11. The method of claim 1, wherein the dopant in the first and
second semiconductors is a first dopant, further comprising:
selecting a third amount of a second dopant; and using the third
amount together with the first and second amounts to grow the
monocrystalline semiconductor; wherein the first, second, and third
amounts are selected so that the new semiconductor has the desired
final concentration of the first dopant and a desired final
concentration of the second dopant.
12. The method of claim 1, wherein the dopant in the first and
second semiconductors is a first dopant, further comprising:
selecting a third amount of a third predoped semiconductor having a
third concentration of a second dopant; and using the third amount
together with the first and second amounts to grow the
monocrystalline semiconductor; wherein the first, second, and third
amounts are selected so that the new semiconductor has the desired
final concentration of the first dopant and a desired final
concentration of the second dopant.
13. A method of forming a monocrystalline semiconductor, the method
comprising: selecting a desired final concentration of a dopant for
the monocrystalline semiconductor, where the dopant is selected
from the group consisting of n-type dopants and p-type dopants;
selecting a first amount of a predoped semiconductor having a first
concentration of the dopant, wherein the first concentration is
lower than the desired final concentration of the dopant; selecting
a second amount of the dopant; and using the first and second
amounts to grow the monocrystalline semiconductor; wherein the
first amount of the predoped semiconductor and the second amount of
the dopant are selected so that the monocrystalline semiconductor
has the desired final concentration of the dopant.
14. The method of claim 13, wherein the predoped semiconductor is
polycrystalline.
15. The method of claim 13, wherein the step of using the first and
second amounts includes employing the floating zone method.
16. A method of forming a monocrystalline semiconductor, the method
comprising: selecting a polycrystalline semiconductor that includes
a dopant, where the dopant is selected from the group consisting of
n-type dopants and p-type dopants; and growing the monocrystalline
semiconductor from the polycrystalline semiconductor by the
Czochralski method or the floating zone method.
17. The method of claim 16, further comprising forming the
polycrystalline semiconductor that includes the dopant by a
chemical vapor deposition method.
18. A method of forming doped polycrystalline silicon, the method
comprising: providing a reactor for chemical vapor deposition;
creating a vapor within the reactor that includes a silicon
compound and a dopant, where the dopant is selected from the group
consisting of antimony, arsenic, boron, and phosphorous; and
providing a substrate, exposed to the vapor, onto which the silicon
and the dopant in the vapor are deposited to form doped
polycrystalline silicon.
19. The method of claim 18, wherein the silicon compound includes
at least one of the following: monochlorosilane, dichlorosilane,
trichlorosilane, and tetrachlorosilane.
20. The method of claim 18, wherein the dopant is selected from the
group consisting of diborane (B.sub.2H.sub.2), phosphine
(PH.sub.3), and arsine (AsH.sub.3).
21. The method of claim 18, wherein the vapor also includes
molecular hydrogen.
22. The method of claim 18, further comprising heating the
substrate.
Description
CROSS-REFERENCES
[0001] This application is a continuation of U.S. patent
application Ser. No. 09/163,858, filed Sep. 30, 1998, which is
incorporated herein by reference.
[0002] The following references are incorporated herein by
reference: (1) "Standard Practice for Conversion Between
Resistivity and Dopant Density for Boron-Doped and Phosphorus-Doped
Silicon," ASTM Designation F 723-82 (1987); and (2) "Crystal
Fabrication," by Takao Abe, in VLSI Electronics: Microstructure
Science, volume 12, pages 3-61 (1985).
TECHNICAL FIELD
[0003] The invention relates to semiconductors. More particularly,
the invention relates to methods for producing doped
polycrystalline semiconductors and to methods for producing doped
monocrystalline semiconductors from predoped monocrystalline and
polycrystalline semiconductors,
BACKGROUND OF THE INVENTION
[0004] In recent years, electronics has come to be dominated by
semiconductor devices, which are found in the discrete devices and
integrated circuits of computers, calculators, televisions, VCRs,
radios, telephones, answering machines, wristwatches, cameras, and
cars, among others. Semiconductor devices are formed from
semiconductors, which are compounds having conductivities
intermediate between those of the high-conductivity conductors and
the low-conductivity insulators. Here, conductivity refers to a
compound's ability to conduct electricity; compounds with greater
conductivities are able to conduct greater amounts of
electricity.
[0005] Semiconductors are important in part because their
conductivity readily may be altered by the addition of certain
foreign compounds. These foreign compounds are known as dopants,
and the addition of these foreign compounds to semiconductors is
known as doping.
[0006] Doping may be used to create two types of semiconductors:
n-type semiconductors and p-type semiconductors. In n-type
semiconductors, the dopant adds negative charge carriers, which
typically comprise extra electrons. Examples of n-type dopants for
silicon-based semiconductors include phosphorus (P), arsenic (As),
and antimony (Sb). In p-type semiconductors, the dopant adds
positive charge carriers, which typically comprise holes (or
missing electrons). Examples of p-type dopants for silicon-based
semiconductors include boron (B).
[0007] Although doping is essential to semiconductor technology,
current doping methods suffer from a number of shortcomings. In
particular, current doping methods involve doping monocrystalline
semiconductors as they are produced from polycrystalline
precursors. Doping monocrystalline semiconductors may involve
frequent storing, weighing, and handling of dopant. This processing
requires special equipment, which may be bulky and expensive. This
processing also requires an operator, which may expose the operator
to extremely toxic dopants, such as arsenic. Doping monocrystalline
semiconductors also may involve loss or uneven distribution of
dopant.
SUMMARY OF THE INVENTION
[0008] The present invention addresses these and other shortcomings
by providing methods for producing doped polycrystalline
semiconductors and methods for producing doped monocrystalline
semiconductors from predoped monocrystalline and polycrystalline
semiconductors. These methods may reduce or eliminate the need to
store, weigh, and handle dopant during the production of doped
monocrystalline semiconductors. These methods also may enhance the
uniformity of dopant distribution.
[0009] In a first set of embodiments, the invention provides
methods of forming doped polycrystalline silicon. One such method
involves (1) providing a reactor for chemical vapor deposition, (2)
creating a vapor within the reactor that includes a silicon
compound and a preselected dopant, and (3) providing a substrate,
exposed to the vapor, onto which the silicon and the dopant in the
vapor are deposited to form doped polycrystalline silicon.
Additional, related methods are described in the detailed
description and claims.
[0010] In a second set of embodiments, the invention provides
methods of forming a monocrystalline semiconductor having a
preselected concentration of a dopant. One such method involves (1)
selecting a first amount of a first semiconductor, the first
semiconductor having a first concentration of the dopant, wherein
the first concentration is higher than the preselected
concentration, (2) selecting a second amount of a second
semiconductor, and (3) using the first and second amounts to grow
the monocrystalline semiconductor. The first and second amounts are
selected so that the monocrystalline semiconductor has the
preselected concentration of the dopant. Additional, related
methods are described in the detailed description and claims.
[0011] The nature of the invention will be understood more readily
after consideration of the drawings and the detailed description of
the invention that follow.
BRIEF DESCRIPTION OF THE DRAWINGS
[0012] FIG. 1 is a flowchart showing a method for forming doped
polycrystalline semiconductors.
[0013] FIG. 2 is a schematic view of an apparatus for forming doped
polycrystalline semiconductors.
[0014] FIG. 3 is a flowchart showing a method for forming doped
monocrystalline semiconductors.
[0015] FIG. 4 is a schematic partial view of a floating zone (FZ)
apparatus for forming doped monocrystalline semiconductors.
[0016] FIG. 5 is a schematic view of a Czochralski (CZ) apparatus
for forming doped monocrystalline semiconductors.
DETAILED DESCRIPTION OF THE INVENTION
[0017] The present invention provides (1) methods for producing
doped polycrystalline semiconductors and (2) methods for producing
doped monocrystalline semiconductors from predoped monocrystalline
and polycrystalline semiconductors.
[0018] FIG. 1 is a flowchart showing a method for forming doped
polycrystalline silicon in accordance with the invention.
Generally, doped polycrystalline silicon is formed by (1) providing
a chemical vapor deposition (CVD) reactor 10, (2) creating a vapor
12 within the CVD reactor that includes a silicon compound and a
preselected dopant, and (3) providing a substrate 14, exposed to
the vapor, onto which the silicon and the dopant in the vapor are
deposited 16 to form the doped polycrystalline silicon.
[0019] FIG. 2 is a schematic view of an apparatus for forming doped
polycrystalline silicon in accordance with the invention. The
apparatus includes a CVD reactor 40, which in turn includes a
reaction chamber 42 for enclosing a vapor 44 and a substrate 46
exposed to the vapor. Reaction chamber 42 may take various forms,
including a quartz bell jar. Substrate 46 also may take various
forms, including silicon. In CVD reactor 40, substrate 46 is
connected to a power source 48, which may be used to create an
electrical current within the substrate. This current is used to
heat substrate 46 and vapor 44, which catalyzes a reaction in the
vapor that leads to the deposition of doped polycrystalline silicon
50 onto substrate 46.
[0020] Vapor 44 is created within CVD reactor 40 from a silicon
compound and a preselected dopant. The silicon compound may be
monochlorosilane (SiH.sub.3Cl), dichlorosilane (SiH.sub.2Cl.sub.2),
trichlorosilane (SiHCl.sub.3), and/or tetrachlorosilane
(SiCl.sub.4), among others. The dopant may be phosphine (PH.sub.3),
diborane (B.sub.2H.sub.6), and/or arsine (AsH.sub.3), among others.
A reductant, such as molecular hydrogen (H.sub.2), also may be
present.
[0021] In use, substrate 46 is heated to approximately 1100.degree.
C. The heat causes vapor 44 to undergo a reaction, which in turn
causes doped polycrystalline silicon to deposit on substrate 46. If
the reaction is maintained for 200-300 hours, deposits of doped
polycrystalline silicon 150-200 mm in diameter may be formed. These
deposits may be epitaxial, if substrate 46 is such that the doped
polycrystalline silicon is deposited over the surface of another
crystal of different chemical composition but similar
structure.
[0022] FIG. 2 also shows a three-step method for producing
trichlorosilane for use in the above method. In this three-step
method, trichlorosilane is produced by (1) separating
metallurgical-grade silicon (MG-Si) from quartzite (SiO.sub.2) by a
carbon reduction reaction in an arc furnace, (2) chlorinating MG-Si
with HCl to produce trichlorosilane (SiHCl.sub.3) in a fluidized
bed reactor 52, and (3) purifying SiHCl.sub.3 by distillation in a
distillation tower 54.
[0023] FIG. 3 is a flowchart of a method for forming doped
monocrystalline semiconductors from predoped monocrystalline and/or
polycrystalline semiconductors in accordance with the invention.
The FZ method is the preferred method for producing semiconductor
discrete devices. The CZ method is the preferred method for
producing semiconductor integrated circuits. Generally, a doped
monocrystalline semiconductor having a preselected concentration of
a dopant may be formed by (1) selecting a first amount 80 of a
predoped semiconductor having a first concentration of the dopant,
wherein the first concentration is lower than the preselected
concentration, (2) selecting a second amount 82 of the dopant, and
(3) using the first and second amounts 84 to grow the
monocrystalline semiconductor. The FZ method 86 is the preferred
method of using the first and second amounts to produce
semiconductor discrete devices. The CZ method 88 is the preferred
method to produce semiconductor integrated circuits. The first and
second amounts are selected so that the monocrystalline
semiconductor has the preselected concentration of the dopant. The
first and second semiconductors may be polycrystalline
semiconductors, or they may be monocrystalline semiconductors.
[0024] Alternatively, a doped monocrystalline semiconductor having
a preselected concentration of a dopant may be formed by (1)
selecting a first amount of a first semiconductor, the first
semiconductor having a first concentration of the dopant, wherein
the first concentration is higher than the preselected
concentration, (2) selecting a second amount of a second
semiconductor, and (3) using the first and second amounts to grow
the monocrystalline semiconductor.
[0025] Alternatively, a monocrystalline semiconductor that includes
a preselected dopant may be formed by (1) selecting a
polycrystalline semiconductor that includes the dopant, and (2)
growing the monocrystalline semiconductor from the polycrystalline
semiconductor. The polycrystalline semiconductor that includes a
preselected dopant may be formed as described above. Typically, the
concentration of dopant in the polycrystalline semiconductor should
equal the desired concentration of dopant in the monocrystalline
semiconductor, although additional dopant may be added during the
FZ method.
[0026] FIG. 4 shows a schematic view of an FZ apparatus for forming
doped monocrystalline semiconductors from predoped monocrystalline
and/or polycrystalline semiconductors. As shown in Panel (a), the
FZ apparatus includes a radio-frequency coil 100 and a conductive
carbon ring 102. Radio-frequency coil 100 heats conductive carbon
ring 102, which in turn heats a tip 104 of an oriented piece of
predoped polycrystalline semiconductor 106. As shown in Panel (b),
tip 104 melts when it reaches a sufficiently high temperature
(e.g., 1427.degree. C. for silicon), forming a conductive melt drop
108, which may directly be heated by radio-frequency coil 100. Melt
drop 108 is brought into contact and fused with a monocrystalline
seed crystal 110. As shown in Panel (c), seed crystal 110 may be
rotated, and a neck 112 may be formed between seed crystal 110 and
predoped polycrystalline semiconductor 106. As shown in Panel (d),
seed crystal 110 then may be lowered slowly, so that predoped
polycrystalline semiconductor 106 is pulled through radio-frequency
coil 100, melted, and recrystallized as a monocrystalline
semiconductor 114 continuous with neck 112 and seed crystal 110. As
shown in Panel (e), a rotating support structure 116 may be
employed to support the monocrystalline semiconductor 114 as it
grows. As shown in Panel (f), the FZ method may be terminated by
turning down radio-frequency coil 100 to reduce the quantity of the
melt, and separating predoped polycrystalline semiconductor 106
from monocrystalline semiconductor 114.
[0027] Predoped polycrystalline semiconductors used in the FZ
method preferably are free of cracks and other major
discontinuities, and may be used to produce monocrystalline
semiconductors of various diameters and resistivities. The pulling
conditions can vary the diameter of the final monocrystalline
semiconductor independent of the diameter of the starting
polycrystalline semiconductor. For example, a 150-mm doped
monocrystalline semiconductor may be grown from a 125-mm diameter
predoped polycrystalline semiconductor.
[0028] FIG. 5 shows a schematic view of a (CZ) apparatus for
forming doped monocrystalline semiconductors from predoped
monocrystalline and/or polycrystalline semiconductors. The
apparatus includes a crucible 150 for holding the predoped
semiconductor 152 and a heater 154 for heating the crucible and
semiconductor. The apparatus also includes a seed holder 156 for
holding a monocrystalline seed crystal 158 and a pulling wire 160
attached to the seed holder for pulling the seed crystal away from
crucible 150.
[0029] The CZ apparatus is used as follows. Heater 154 heats
crucible 150 and predoped semiconductor 152 until predoped
semiconductor 152 melts. Pulling wire 160 then lowers seed crystal
158 until it is immediately adjacent predoped semiconductor 152.
When temperature conditions at the liquid/solid interface 161 are
optimal, seed crystal 158 is touched to the surface and then slowly
pulled away. Both the seed crystal and the crucible are in
rotation, with the direction and speed of the rotation determined
by the required crystal parameter. As seed crystal 158 slowly is
pulled away, a conical/cylindrical monocrystalline semiconductor
162 forms as molten predoped semiconductor 152 solidifies. The
crystal orientation of monocrystalline semiconductor 162 will be
the same as that of seed crystal 158. The process is terminated
when the desired charge, or length of crystal, is grown.
Semiconductor 162 is then cooled and extracted from wire puller
156.
[0030] Typically, the crucible in the CZ method is made out of
quartz (SiO.sub.2), and a certain amount of the crucible dissolves
into the molten semiconductor during crystal formation. In
particular, oxygen from the crucible will be doped into
monocrystalline semiconductor 162. Such oxygen has certain
advantages, such as intrinsic gettering capabilities, but also
certain disadvantages, such as an ability to migrate and form
inhomogeneities.
[0031] The preselected concentration of dopant in both the FZ and
CZ methods typically will be chosen so that the monocrystalline
semiconductor has a preselected conductivity. The relationship
between dopant concentration and conductivity depends on the dopant
and on the semiconductor into which the dopant is added. Empirical
relationships have been derived for various pairings; examples are
given below both for a p-type semiconductor and an n-type
semiconductor. These relationships are described in terms of
resistivity, which is the multiplicative inverse of the
conductivity.
[0032] Equation 1 describes the relationship between resistivity
and concentration in boron-doped silicon, which is a p-type
semiconductor. 1 n = 1.330 .times. 10 16 + 1.082 .times. 10 17 [ 1
+ ( 54.56 ) 1.105 ] ( 1 )
[0033] Here, resistivity is measured in Ohm-centimeters
(.OMEGA.-cm), and concentration is measured in atoms per cubic
centimeter (atoms/cm.sup.3). Table 1 shows concentrations derived
from Eq. 1 for seven different resistivities.
1TABLE 1 Example of groupings for a p-type semiconductor. Resis-
tivity Dopant Concentration Dopant Concentration Mass Group
[.OMEGA.-cm] [atoms/cm.sup.3] [atoms/g] [g] A 0 n.sub.s,1 = 0
i.sub.s,1 = 0 W.sub.1 B 0.005 n.sub.s,2 = 2.01396 .times. 10.sup.19
i.sub.s,2 = 8.64359 .times. 10.sup.18 W.sub.2 C 0.01 n.sub.s,3 =
8.48622 .times. 10.sup.18 i.sub.s,3 = 3.64215 .times. 10.sup.18
W.sub.3 D 0.06 n.sub.s,4 = 6.04728 .times. 10.sup.17 i.sub.s,4 =
2.59540 .times. 10.sup.17 W.sub.4 E 0.3 n.sub.s,5 = 6.00476 .times.
10.sup.16 i.sub.s,5 = 2.57715 .times. 10.sup.16 W.sub.5 F 2.0
n.sub.s,6 = 6.95122 .times. 10.sup.15 i.sub.s,6 = 2.98336 .times.
10.sup.15 W.sub.6 G 20.0 n.sub.s,7 = 6.67378 .times. 10.sup.14
i.sub.s,7 = 2.86428 .times. 10.sup.14 W.sub.7
[0034] Equation 2 describes the relationship between resistivity
and concentration in phosphorous-doped silicon, which is an n-type
semiconductor. 2 n = 6.242 .times. 10 18 .times. 10 z ( 2 ) Z = A 0
+ A 1 x + A 2 x 2 + A 3 x 3 1 + B 1 x + B 2 x 2 + B 3 x 3 ( 3 )
[0035] Here
.chi.=log.sub.10.rho. (4)
A.sub.0=-3.1083 (5a)
A.sub.1=-3.2626 (5b)
A.sub.2=-1.2196 (5c)
A.sub.3=-0.13923 (5d)
B.sub.1=1.0265 (5e)
B.sub.2=0.38755 (5f)
B.sub.3=0.041833 (5g)
[0036] Again, resistivity is measured in .OMEGA.-cm, and
concentration is measured in atoms/cm.sup.3. Table 2 shows
concentrations derived from Eqs. 2-5 for seven different
resistivities.
2TABLE 2 Example of groupings for an n-type semiconductor. Resis-
tivity Dopant Concentration Dopant Concentration Mass Group
[.OMEGA.-cm] [atoms/cm.sup.3] [atoms/g] [g] A 0 n.sub.s,1 = 0
i.sub.s,1 = 0 W.sub.1 B 0.005 n.sub.s,2 = 1.22426 .times. 10.sup.19
i.sub.s,2 = 5.25434 .times. 10.sup.18 W.sub.2 C 0.01 n.sub.s,3 =
4.53266 .times. 10.sup.18 i.sub.s,3 = 1.94535 .times. 10.sup.18
W.sub.3 D 0.06 n.sub.s,4 = 1.74061 .times. 10.sup.17 i.sub.s,4 =
7.47045 .times. 10.sup.16 W.sub.4 E 0.3 n.sub.s,5 = 1.86957 .times.
10.sup.16 i.sub.s,5 = 8.02392 .times. 10.sup.15 W.sub.5 F 2.0
n.sub.s,6 = 2.33724 .times. 10.sup.15 i.sub.s,6 = 1.00311 .times.
10.sup.15 W.sub.6 G 20.0 n.sub.s,7 = 2.19207 .times. 10.sup.14
i.sub.s,7 = 9.40801 .times. 10.sup.13 W.sub.7
[0037] Monocrystalline semiconductors having a preselected
concentration of a dopant may be formed from predoped
monocrystalline or polycrystalline semiconductors from the groups
in Tables 1 and 2, or from any other groups of predoped
semiconductors, so long as the groups have dopant concentrations
that bracket the preselected concentration. Generally, the dopant
concentration in any monocrystalline semiconductor formed from any
combination of predoped semiconductors is given by Eq. 6. 3 i T = (
j i s , j W J ) W T ( 6 )
[0038] Here, i.sub.T is the dopant concentration in the
monocrystalline semiconductor, i.sub.sj is the dopant concentration
in the jth predoped semiconductor, W.sub.T is the mass or amount of
the monocrystalline semiconductor, and W.sub.j is the mass or
amount of the jth semiconductor. Eq. 6 may be used to derive
expressions for the amount of predoped semiconductors that must be
used to form a monocrystalline semiconductor having a preselected
dopant concentration, which obtains whenever i.sub.T is specified.
For example, if there are two predoped semiconductors, the
following equations will apply:
[0039] i.sub.T=(i.sub.s,uW.sub.u+i.sub.s,1W.sub.1)/W.sub.T (7)
W.sub.T=W.sub.u+W.sub.1 (8)
[0040] Here, subscript u is for a first semiconductor, which has a
dopant concentration higher than the preselected dopant
concentration, subscript 1 is for a second semiconductor, which has
a dopant concentration lower than the preselected dopant
concentration, and subscript T is for the composite semiconductor.
These equations may be solved for the first and second amounts
W.sub.u and W.sub.1, wherein the monocrystalline semiconductor has
the preselected concentration, yielding: 4 W u = ( i T - i s , 1 i
s , u - i s , 1 ) W T ( 9 ) W 1 = ( i T - i s , u i s , 1 - i s , u
) W T ( 10 )
EXAMPLE 1
[0041] This example shows how the invention may be used to form a
p-type monocrystalline boron/silicon semiconductor having a mass of
150 kilograms (kg) and a resistivity of 0.03 .OMEGA.-cm. Eq. 1
shows that the associated preselected dopant concentration is
n=1.77044.times.10.sup.18 atoms/cm.sup.3, or
7.59846.times.10.sup.17 atoms/g.
[0042] Predoped semiconductors may be chosen from any group of
p-type predoped semiconductors having dopant concentrations that
bracket the preselected concentration, such as those in Table 1.
For example, groups "C" and "D" may be chosen because they most
closely bracket the preselected concentration. Eqs. 9 and 10 yield
a charge of 83.13 kg for group "D" and 66.87 kg for group "C,"
respectively. If these amounts are rounded to the nearest kg, such
as 83 kg of group "D" and 67 kg of group "C," the error in the
final resistivity would be about +1.2%. Alternatively, groups "B"
and undoped "A" also may be chosen, because they also bracket the
preselected concentration, although not as closely as groups "C"
and "D." Eqs. 9 and 10 yield a charge of 30.673 kg of group "B" and
119.327 kg of undoped group "A." If these amounts are rounded to
the nearest half kg, such as 30.5 kg of group "B" and 119.5 kg of
group "A", the error in the final resistivity would be about
1.7%.
EXAMPLE 2
[0043] This example shows how the invention may be used to form an
n-type monocrystalline phosphorous/silicon semiconductor having a
mass of 100 kg and a resistivity of 5 .OMEGA.-cm. Eqs. 2-4 show
that the associated dopant concentration is
n=9.04553.times.10.sup.14 atoms/cm.sup.3, or
3.88220.times.10.sup.14 atoms/g. Predoped semiconductors may be
chosen from any group of n-type predoped semiconductors having
dopant concentrations that bracket the preselected concentration,
such as those in Table 2. For example, groups "F" and "G" may be
chosen. Eqs. 9 and 10 yield of charge of 10.842 kg of group "G" and
89.158 kg of group "F." If these amounts are rounded to the nearest
kg, the error in the final resistivity would be about -1.1%.
[0044] These examples are intended to be illustrative and do not
exhaust the flexibility of the associated methods. For example, the
methods also may be used with a first amount of a predoped
semiconductor and a second amount of a straight dopant, and with
other combinations described in the claims. For critical
specifications or epitaxial substrate applications, semiconductors
may be predoped to the exact concentration preselected for the
monocrystalline semiconductor.
[0045] Accordingly, while the invention has been disclosed in
preferred forms, the specific embodiments thereof as disclosed and
illustrated herein are not to be considered in a limiting sense,
because numerous variations are possible and no single feature,
function, or property of the preferred embodiments are essential.
The invention is to be defined only by the scope of the issued
claims.
* * * * *