 |
Description  |
|
|
FIELD OF THE INVENTION
The present invention relates to a method for evaluating a semiconductor
layer epitaxially grown on a substrate and, more particularly, to a method
for evaluating the thickness of that layer during the epitaxial growth
process with high precision.
The present invention also relates to a test pattern for process evaluation
which is disposed on a substrate and used for evaluating semiconductor
layers epitaxially grown on the substrate.
The present invention also relates to a method for evaluating an AlGaAs
multiple quantum well layer and, more particularly, to a method for
evaluating the thickness and Al composition of the multiple quantum well
layer with high precision.
BACKGROUND OF THE INVENTION
FIGS. 17(a) and 17(b) are perspective views for explaining a conventional
method for measuring the thickness of a semiconductor layer epitaxially
grown on a substrate. In these figures, reference numeral 101 designates a
GaAs substrate and numeral 102 designates an epitaxial layer grown on the
GaAs substrate 101.
Initially, as illustrated in FIG. 17(a), an epitaxial layer 102 is grown on
the GaAs substrate 101 having a diameter of 2 inches by MOCVD. The
thickness of the epitaxial layer 102 is controlled by the growth time
according to the growth rate which is controlled by the quantity of the
supplied source gas. Thereafter, this wafer is cleaved to make a strip
sample 103 shown in FIG. 17(b), and the section of the sample 103 is
observed with a scanning electron microscopy (SEM) and photographed,
whereby the thickness of the epitaxial layer is determined.
In the conventional evaluation method, however, the cleaving of the wafer
takes much time and labor. In addition, since the evaluation is performed
after the growth of the epitaxial layer, a feedback control cannot be
applied during the epitaxial growth, so that the controllability of the
thickness is poor, resulting in a poor production yield.
FIG. 18 is a schematic diagram illustrating a crystal growth monitor
apparatus for optically measuring thickness of an epitaxial layer grown on
a semiconductor substrate by molecular beam epitaxy (MBE). This apparatus
is disclosed in Japanese Published Patent Application No. Hei. 2-252694.
In FIG. 18, an ultra-high vacuum container 111 having an observation window
112 contains a substrate holder 113 on which a substrate 114 is disposed.
In the ultra-high vacuum container a, molecular or atomic beam of
materials, which is produced by evaporating material sources 115 and 116,
reaches the substrate 114, whereby a crystal is grown on the substrate.
Reference numeral 117 designates a light source, numeral 119 designates a
half mirror for separating reflected light from the substrate 114, numeral
120 designates a condenser lens, numeral 121 designates a diaphragm, and
numeral 123 designates an eyepiece. Light 118 emitted from the light
source 117 travels through the half mirror 119, the condenser lens 120,
the diaphragm 121, and the observation window 112 and reaches the surface
of the crystal layer growing on the substrate. The light is reflected at
the surface of the growing crystal layer and received by the half mirror
122. The half mirror 122 is observed with the eyepiece 123 to see if the
light strikes the proper position on the surface of the substrate. The
optical path is adjusted if necessary. On the other hand, the light
reflected at the surface of the growing crystal layer is reflected by the
half mirror 119. The reflected light 124 is received by a spectroscope 125
and turned into monochromatic light. This monochromatic light is guided
through a condenser lens 126 to a Rochon prism 127. In the Rochon prism
127, the monochromatic light is divided into two beams having different
polarization planes that are at right angles to each other. These two
light beams are respectively received by PIN photodiodes 128 and 129
having similar characteristics. These photodiodes 128 and 129 are
connected in series and in reverse polarity, and the difference output is
amplified by a DC amplifier 130 and recorded in a recorder 131.
Polarized light caused by reflection at the surface of the growing crystal
layer varies according to the crystal growth condition. For example, in
GaAs growth, polarized light attains a maximum when an atomic plane of Ga
is formed. Thereafter, the polarized light gradually decreases as As
molecules are accumulated on the atomic plane and, finally, becomes a
minimum. When an As atomic plane is again formed, the polarized light
becomes the maximum again. Therefore, if orthogonal polarization
components are taken out and a difference between them is measured, the
growth process at the surface of the substrate is directly observed.
FIG. 19 illustrates the result of observation of a growing GaAs crystal
using the crystal growth monitor apparatus shown in FIG. 18, in which the
abscissa shows the growth time and the ordinate shows the output from the
DC amplifier 130. This output corresponds to a difference in intensities
of the orthogonal polarization components. In addition, the output from
the DC amplifier 130 shows a damped oscillation, and a period of this
damped oscillation corresponds to the growth of one atomic layer.
Therefore, the thickness of the growing crystal layer can be determined by
counting the periods.
In the prior art crystal growth monitor apparatus, however, since the
thickness of the growing crystal layer is measured by detecting the
variation in the reflected light due to the atomic layer level unevenness
at the surface of the crystal layer, the variation in the output signal is
very small, resulting in difficulty in the measurement. In addition, in
order to increase the S/N ratio, means for polarizing the incident light
is required, whereby the monitor apparatus is complicated and rises in
price.
FIGS. 20(a) and 20(b) are perspective views for explaining a prior art
method for evaluating an AlGaAs multiple quantum well (hereinafter
referred to as MQW) layer. In these figures, reference numeral 201
designates a GaAs substrate. A first Al.sub.0.4 Ga.sub.0.6 As layer 202 is
disposed on the GaAs substrate 201. An MQW layer 203 comprising
alternating Al.sub.0.1 Ga.sub.0.9 As well layers and Al.sub.0.3 Ga.sub.0.7
As barrier layers is disposed on the first Al.sub.0.4 Ga.sub.0.6 As layer
202. A second Al.sub.0.4 Ga.sub.0.6 As layer 204 is disposed on the MQW
layer 203.
A description is given of the evaluation process.
Initially, as illustrated in FIG. 20(a), there are successively grown on
the 2-inch diameter GaAs substrate 201 the first Al.sub.0.4 Ga.sub.0.6 As
layer 202, the MQW layer 203 comprising alternating Al.sub.0.1 Ga.sub.0.9
As well layers and Al.sub.0.3 Ga.sub.0.7 As barrier layers, and the second
Al.sub.0.4 Ga.sub.0.6 As layer 204, preferably by MOCVD. The thicknesses
of these layers are controlled by the growth time according to the growth
rate which is controlled by the quantity of the supplied source gas.
Thereafter, as illustrated in FIG. 20(b), the wafer is cleaved to make a
strip sample 205, and the section of the sample 205 is observed with an
SEM and photographed, whereby the thicknesses of the respective layers are
evaluated. Further, the Al compositions of the respective layers are
evaluated by peak wavelengths obtained in a PL (photoluminescence)
evaluation at room temperature.
In the above-described evaluation process of the thickness of the AlGaAs
MQW layer, however, the cleaving of the wafer takes much time and labor.
In addition, sufficient precision to detect an error of several nanometer
cannot be achieved. In evaluation of Al composition, the PL peak
wavelength varies due to variations in the thickness and the Al
composition, and the variation in the Al composition cannot be separated
from the variation in the thickness. Therefore, an accurate evaluation of
the Al composition is impossible.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a relatively simple
method for evaluating epitaxial layers during the epitaxial growth
process.
It is another object of the present invention to provide a test pattern for
process evaluation that achieves a precise evaluation of epitaxial layers
grown on a substrate.
It is still another object of the present invention to provide a method for
evaluating thickness and Al composition of an AlGaAs MQW layer with high
precision, without cleaving a wafer.
Other objects and advantages of the present invention will become apparent
from the detailed description give hereinafter; it should be understood,
however, that the detailed description and specific embodiment are given
by way of illustration only, since various changes and modifications
within the scope of the invention will become apparent to those skilled in
the art from this detailed description.
According to a first aspect of the present invention, in a method for
evaluating a semiconductor layer epitaxially growing on a substrate, a
plurality of stripe-shaped ridges extending in a prescribed direction are
formed on the substrate, and a semiconductor layer is grown on the
substrate including the stripe-shaped ridges while irradiating the
stripe-shaped ridges with light and monitoring diffracted light from the
stripe-shaped ridges, whereby the thickness of the epitaxially grown
semiconductor layer is evaluated. Therefore, the thickness of the
epitaxial layer is easily evaluated during the epitaxial growth process,
whereby the thickness controllability is significantly improved.
According to a second aspect of the present invention, a test pattern
comprises a plurality of stripe-shaped ridges formed on a part of a
substrate and in a prescribed direction selected so that a semiconductor
layer having a triangular section and side surfaces on which the epitaxial
growth does not proceed is grown on each of the stripe-shaped ridges. Each
ridge has a width W and a height H that satisfy a relation of
d.apprxeq.0.7 W+H where d is a desired thickness of the semiconductor
layer epitaxially grown on the substrate. Since the test pattern
comprising the stripe-shaped ridges is flat when the epitaxially growing
layer reaches the desired thickness, the thickness of the epitaxially
growing layer is evaluated with high precision.
According to a third aspect of the present invention, in a method for
evaluating an epitaxially grown AlGaAs multiple quantum well layer,
initially, an insulating mask pattern having an opening is formed on a
part of a substrate on which a plurality of semiconductor layers including
the AlGaAs multiple quantum well layer are to be grown. The ratio of the
width of the opening to the width of the insulating mask pattern is varied
continuously or in steps. Then, a plurality of semiconductor layers
including the AlGaAs multiple quantum well layer are grown on the
substrate including the insulating mask pattern using a vapor phase growth
method. Thereafter, PL peak wavelengths of the AlGaAs multiple quantum
well layer grown in the opening of the insulating mask pattern are
measured at different positions with different opening ratios of the
insulating mask pattern, and the measured values are compared with
theoretical curves obtained with respect to a standard quantum well
structure, whereby the thickness and the Al composition of the AlGaAs
multiple quantum well layer are evaluated. The PL peak wavelengths vary in
response to variations in thickness and Al composition, and the variation
in the PL peak wavelengths at the different positions of the opening of
the insulating mask pattern due to the variation in the thickness is
different from that due to the variation in the Al composition ratio, so
that the thickness and the Al composition ratio of the AlGaAs multiple
quantum well layer are evaluated with high precision, without cleaving the
wafer.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a plan view of a substrate including a test pattern for process
evaluation used for evaluating an epitaxial layer growing on the
substrate, in accordance with a first embodiment of the present invention.
FIG. 2 is a sectional view taken along a line 2--2 of FIG. 1.
FIGS. 3(a)-3(c) are sectional views for explaining the evaluation method
according to the first embodiment of the present invention.
FIGS. 4(a)-4(c) and 5(a)-5(b) are sectional views for explaining
reflections of incident light from the surface of the test pattern at
different steps in the epitaxial growth process, in accordance with the
first embodiment of the present invention.
FIG. 6 is a graph illustrating growth time vs. diffracted light intensity
characteristics.
FIG. 7 is a schematic diagram illustrating an epitaxial growth apparatus
employed for the epitaxial layer evaluation method according to the first
embodiment of the present invention.
FIG. 8 is a perspective view, partially in section, illustrating an InP
semiconductor laser including a diffraction grating.
FIGS. 9(a)-9(c) are sectional views for explaining a method for evaluating
an epitaxial layer in accordance with a second embodiment of the present
invention, which is employed in the production of the semiconductor laser
shown in FIG. 8.
FIGS. 10(a)-10(c) are sectional views for explaining a method for
evaluating epitaxial layers in accordance with a third embodiment of the
present invention.
FIG. 11 is a perspective view illustrating a substrate including a test
pattern for process evaluation used for evaluating epitaxial layers
growing on the substrate, in accordance with a fourth embodiment of the
present invention.
FIG. 12 is an enlarged view of the test pattern shown in FIG. 11.
FIG. 13 is a graph illustrating insulating mask opening ratio vs. growth
rate characteristics.
FIG. 14 is a sectional view taken along a line 14--14 of FIG. 12,
illustrating epitaxial layers grown on the substrate exposed in the
opening of the insulating mask.
FIG. 15 is a graph illustrating well layer thickness vs. PL wavelength
characteristics for explaining an evaluation method according to the
fourth embodiment of the present invention.
FIG. 16 is a plan view illustrating a variation of the test pattern
employed in the evaluation method according to the fourth embodiment of
the present invention.
FIGS. 17(a) and 17(b) are perspective views for explaining a method for
evaluating the thickness of an epitaxial layer grown on a semiconductor
substrate, according to the prior art.
FIG. 18 is a schematic diagram illustrating a crystal growth monitor
apparatus for optically measuring the thickness of an epitaxial layer
grown on a semiconductor substrate by MBE according to the prior art.
FIG. 19 is a diagram illustrating the result of observation of a growing
GaAs crystal using the crystal growth monitor apparatus shown in FIG. 18.
FIGS. 20(a) and 20(b) are perspective views for explaining a method for
evaluating an AlGaAs MQW layer according to the prior art.
FIGS. 21(a) and 21(b) are a plan view and a sectional view illustrating a
modification of the first embodiment of the present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 is a plan view of a substrate including a test pattern used for
evaluating an epitaxial layer grown on the substrate, in accordance with a
first embodiment of the present invention. In FIG. 1, reference numeral 1
designates a GaAs substrate with a (100) surface orientation. The GaAs
substrate 1 includes a region 2 where semiconductor elements are to be
produced (hereinafter referred to as element region) and a test pattern 3
for process evaluation (hereinafter referred to as TEG (Test Element
Group)) disposed outside the element region 2. FIG. 2 is a sectional view
taken along a line 2--2 of FIG. 1. As shown in FIG. 2, the TEG 3 is a
periodic pattern of stripe-shaped grooves 11 extending in the [011]
direction. Reference numeral 12 designates stripe-shaped ridge portions of
the GaAs substrate 1 produced between the grooves 11.
FIGS. 3(a)-3(c) are sectional views taken along the line 2--2 of FIG. 1 for
explaining a method for evaluating an epitaxial layer growing on the
substrate, according to the first embodiment of the present invention. In
this first embodiment, a GaAs layer 20 is epitaxially grown on the GaAs
substrate 1. Reference numeral 21 designates light incident on the TEG
comprising the alternating stripe-shaped grooves 11 and ridges 12.
Reference numeral 22 designates diffracted light produced by reflection
type diffraction of the incident light 21 at a flat portion of the TEG.
Reference numeral 23 designates diffracted light produced by reflection
type diffraction of the incident light 21 at a (111)B plane that is formed
during the crystal growth on the TEG. Reference numeral 25 designates a
normal, perpendicular to the surface of the substrate 1.
A description is given of the evaluation method.
Initially, as illustrated in FIG. 3(a), a periodic pattern of stripe-shaped
grooves 11 extending in the [011] direction is formed on the (100) surface
of the GaAs substrate 1 using photolithography and dry etching, whereby a
TEG comprising these stripe-shaped grooves 11 and a plurality of
stripe-shaped ridges 12 between these grooves 11 is produced. The width
(W) and the height (H) of the ridge 12 are adjusted so that the TEG is
completely buried by an epitaxially grown layer 20 with a flat surface
when the layer 20 attains a desired thickness d.
In the step of FIG. 3(b), a GaAs epitaxial layer 20 is grown on the GaAs
substrate 1 by MOCVD. On the stripe-shaped ridge 12 extending in the [011]
direction, the crystal growth proceeds forming (111)B planes on which no
crystal grows. In this first embodiment, during the crystal growth
process, the TEG is irradiated with light 21, such as laser light, and
either or both of the diffracted lights 22 and 23 is/are monitored. The
incident light 21 is applied to the TEG in a direction perpendicular to
the stripe direction of the TEG, at an incident angle .theta.1 with
respect to the normal 25 perpendicular to the surface of the substrate 1.
The epitaxial growth is stopped when the stripe-shaped ridges 12 of the
TEG are completely buried and the surface of the grown layer 20 becomes
flat as shown in FIG. 3(c), whereby the desired thickness d is obtained.
A description is given of the principle of the above-described evaluation
method. FIGS. 4(a)-4(c) and 5(a)-5(b) are diagrams illustrating
reflections of incident light on the TEG at different steps in the
epitaxial growth process. FIG. 6 is a graph illustrating growth time (T)
vs. diffracted light intensity (R) characteristics. In FIG. 6, a curve A
shows a change of the intensity of the diffracted light 22, and a curve B
shows a change of the intensity of the diffracted light 23.
FIG. 4(a) illustrates reflection of incident light 21 from the surface of
the TEG before the epitaxial growth, i.e., at the growth time T=0 of FIG.
6. The TEG formed on the GaAs substrate 1 is a reflection type diffraction
grating. In FIG. 4(a), assuming that the width of the ridge 12 is W, the
width of the groove 11 is D, and the diffraction angle is .theta.2, the
diffraction condition of the incident light 21 at the incident angle
.theta.1 is represented by
(D+W).multidot.(sin .theta.1+sin .theta.2)=m.lambda.
where m is the order of the diffraction.
Therefore, a main peak (m=0) of the diffracted light 22 is observed in a
direction of diffraction angle .theta.2=-.theta.1. The incident angle
.theta.1 is selected to satisfy the relation
.vertline..theta.1.vertline..ltoreq.tan.sup.-1 (2H/D)
so that a high-order component due to multiple reflection at the side
surface of the stripe-shaped ridge 12 is quenched.
In addition, the diffracted light intensity R1 of the main peak is given by
R1.about.R0 W/(D+W)
where R0 is the reflected light intensity at the flat surface.
FIG. 4(b) illustrates reflection of incident light 21 on the surface of the
TEG during the epitaxial growth, at the growth time T1 of FIG. 6. Since
the GaAs epitaxial layer 20 does not grow on the (111)B plane, the growth
proceeds forming a trapezoid portion on each stripe-shaped ridge 12 of the
TEG as shown in FIG. 4(b). The incident light 21 is diffracted at the
inclined plane of the trapezoid portion, i.e., the (111)B plane
(diffracted light 23). The incident angle .phi.1 on the inclined plane and
the diffraction angle .phi.2 of the main peak have the relation of
.phi.2=-.phi.1. Therefore, assuming that the inclination of the inclined
plane of the trapezoid portion is .chi., the diffracted light 23 is
observed in a direction of .theta.3=2.chi.-.theta.1.
FIG. 4(c) illustrates reflection of incident light 21 on the TEG during the
epitaxial growth, at the growth time T2 of FIG. 6. The epitaxial growth on
the stripe-shaped ridge 12 of the TEG stops when a triangular
cross-section is formed as shown in FIG. 4(c). When the triangular
cross-section is formed, the intensity of the diffracted light 23 in the
direction of .theta.3=2.chi.-.theta.1 becomes a maximum. Since the
inclination .chi. of the inclined plane of the trapezoid portion, i.e.,
the inclination of the (111)B plane with respect to the (100) surface of
the substrate, is about 54.degree., the diffracted light intensity R2 at
this time in the same direction is given by
R2.about.R0 W(0.7/tan .theta.1+0.5)/(D+W)
In addition, the diffracted light intensity A in the .theta.2=-.theta.1
direction becomes the minimum.
FIG. 5(a) illustrates reflection of incident light 21 on the TEG during the
epitaxial growth, at the growth time T3 of FIG. 6. As the triangular
portions on the stripe-shaped ridges of the TEG are gradually embedded by
the growing GaAs layer 20 as shown in FIG. 5(a), the diffracted light
intensity A in the .theta.2 direction increases and the diffracted light
intensity B in the .theta.3 direction decreases.
FIG. 5(b) shows reflection of incident light 21 on the TEG after the
epitaxial growth is completed, at the growth time Te of FIG. 6. When the
triangular portions on the stripe-shaped ridges of the TEG are completely
embedded in the growing GaAs layer 20, i.e., when the surface of the GaAs
layer 20 growing on the TEG becomes flat, the incident light 21 is
reflected at the flat surface in the .theta.2=-.theta.1 direction. That
is, only the diffracted light 22 remains. Accordingly, the completion of
the TEG stripe embedding growth is detected by irradiating the TEG with
light at an angle .theta.1 and observing the diffracted light intensities
in the -.theta. direction and the 2.chi.-.theta. direction.
In this first embodiment of the present invention, since light diffracted
from the TEG 3 having an intensity that is determined according to a
desired thickness of an epitaxially grown layer is monitored, an increased
signal intensity is obtained compared to the prior art method shown in
FIG. 18 in which diffracted light caused by the atomic layer level
unevenness is monitored. Therefore, the growth conditions of the epitaxial
layer are detected with high precision, without polarizing the incident
light.
A description is given of the width W and the height H of the stripe-shaped
ridges 12 of the TEG 3.
As described above, the epitaxial growth on the TEG proceeds forming a
triangular portion having (111)B planes on each stripe-shaped ridge of the
TEG. Since the angle .chi. formed between the (111)B plane and the (100)
surface of the substrate is about 54.degree., the thickness of the
triangular layer grown on the ridge 12 is about 0.7 W. On the other hand,
the thickness of the epitaxial layer grown on a region of the substrate 1
other than the TEG region 3 is approximately equal to the thickness of the
epitaxial layer grown in the stripe-shaped groove 11 of the TEG. More
specifically, the thickness d of the epitaxial layer grown on the
substrate 1 other than the TEG region 3 shown in FIG. 3(c) is
approximately equal to the thickness d' of the epitaxial layer grown in
the groove 11 shown in FIG. 5(b). The thickness d' is represented by 0.7
W+H.
Accordingly, if a desired thickness of the epitaxial layer is d, the width
W and the height H of the stripe-shaped ridge 12 are selected so that the
relation of d=0.7 W+H is satisfied, whereby an epitaxial layer with the
desired thickness d is grown on the substrate other than the TEG region at
the moment when the TEG region is flatly embedded.
As described above, according to the first embodiment of the present
invention, the TEG 3 including a plurality of stripe-shaped ridges 12
extending in a prescribed direction is formed on the substrate. The TEG is
irradiated with light in a prescribed direction during the epitaxial
growth on the substrate, and light diffracted by the TEG is monitored.
Therefore, the thickness of the epitaxial layer is evaluated with high
precision, without cleaving the substrate.
While in the above-described first embodiment the TEG 3 is formed directly
on the surface of the GaAs substrate 1, it may be formed on a
semiconductor layer 1a disposed on the substrate 1, as illustrated in
FIGS. 21(a)-21(b).
While in the above-described first embodiment the epitaxial growth is
carried out by MOCVD, the present invention may be applied to epitaxial
growths employing other vapor phase growth methods, such as MBE (Molecular
Beam Epitaxy) or CBE (Chemical Beam Epitaxy).
Further, while in the above-described first embodiment the stripe-shaped
ridges 12 of the TEG 3 are formed in a rectangular shape by dry etching,
these ridges may be formed in reverse-mesa shape using wet etching. Also
in this case, the same effects as described above are achieved.
FIG. 7 is a schematic diagram illustrating an epitaxial growth apparatus
employed for the above-described evaluation process according to the first
embodiment of the present invention. In FIG. 7, reference numeral 31
designates a reaction furnace containing a substrate holder 32. The GaAs
substrate 1 including the TEG 3 shown in FIG. 1 is disposed on the
substrate holder 32. Incident light 21 is introduced into the furnace 31
through an entrance window 33. The incident light 21 is diffracted by the
TEG 3 on the substrate 1. The diffracted light 22 is taken out of the
furnace from an exit window 34. Reference numeral 35 designates a laser
light source and reference numeral 36 designates a photodetector. The TEG
3 on the substrate 1, the entrance window 33, and the exit window 34 are
arranged so that the incident angle .theta.1 of the incident light 21 and
the diffraction angle .theta.2 of the diffracted light 22 have the
relation of .theta.2=-.theta.1 and the incident light 21 is perpendicular
to the stripe direction of the TEG 3. In a case where the GaAs substrate 1
is rotated during the epitaxial growth process, the rotation of the GaAs
substrate 1 should be synchronized with the switching of the incident
light 21 so that the evaluation is carried out only at the moment when the
incident light 21 is perpendicular to the stripe direction of the TEG 3.
Further, the entrance window 33 and the exit window 34 must be flat so
that the incident light 21 and the diffracted light 22 are not broadened.
Alternatively, those windows 33 and 34 may be convex lenses that condense
the incident light 21 and the diffracted light 22, respectively. During
the epitaxial growth, a carrier gas, such as H.sub.2, flows in the furnace
to prevent the inner wall of the furnace from being clouded.
Using the above-described epitaxial growth apparatus, the diffracted light
22 from the TEG 3 is monitored during the epitaxial growth process, and
the thickness of the epitaxially growing layer is evaluated with high
precision, whereby production yield of semiconductor devices is improved.
In the apparatus shown in FIG. 7, only the diffracted light 22 in the
.theta.2 direction, i.e., diffracted light at the flat surface of the TEG
3, is monitored through the exit window 34. However, the exit window may
be positioned so that the diffracted light 23 in the .theta.3 direction,
i.e., diffracted light at the sloping surface of the TEG 3, is monitored.
Alternatively, two exit windows may be provided so that both of the
diffracted light beam 22 and 23 are monitored.
A description is given of a second embodiment of the present invention in
which the above-described evaluation method is applied to an epitaxial
growth process in a production of a semiconductor laser.
FIG. 8 is a perspective view, partially in section, illustrating an InP
semiconductor laser having a diffraction grating. In the figure, reference
numeral 40 designates a p type InP substrate having a (100) surface
orientation. A p type InP lower cladding layer 41 is disposed on the
substrate 40. An undoped InGaAsP active layer 42 is disposed on the lower
cladding layer 41. An n type InP barrier layer 43 is disposed on the
active layer 42. A plurality of stripe-shaped n type InGaAsP layers 44
producing a diffraction grating are disposed on the InP barrier layer 43.
These stripe-shaped InGaAsP layers 44 are periodically arranged parallel
to each other and perpendicular to the resonator length direction of the
laser. A first n type InP upper cladding layer 45 is disposed on the n
type InGaAsP layers 44 and on the n type InP barrier layer 43. The first
upper cladding layer 45, the n type InGaAsP layers 44, the barrier layer
43, the active layer 42, and the lower cladding layer 41 are formed in a
stripe-shaped mesa by etching. A p type InP layer 46 is disposed on part
of the substrate 40, contacting the opposite sides of the mesa. An n type
InP layer 47 is disposed on parts of the p type InP layer 46. An
additional p type InP layer 48 is disposed on the n type InP layer 47 and
part of the p type InP layer 46. A second n type InP upper cladding layer
49 is disposed on the p type InP layer 48 as well as on the first upper
cladding layer 45. An n type InGaAsP contact layer 50 is disposed on the
second upper cladding layer 49. An insulating film, such as SiO.sub.2,
including a window opposite the stripe-shaped mesa is disposed on the top
and side surfaces of the laser structure. A p side electrode 52 is
disposed on the rear surface of the substrate 40 and an n side electrode
53 is disposed on the insulating film 51, contacting the InGaAsP contact
layer 50 through the window in that insulating film 51.
FIGS. 9(a)-9(c) are sectional views illustrating process steps in a method
for evaluating an epitaxially grown layer according to the second
embodiment of the present invention.
A description is given of the evaluation method. Initially, as illustrated
in FIG. 9(a), the p type InP cladding layer 41, the undoped InGaAsP active
layer 42, the n type InP barrier layer 43, and the n type InGaAsP layer 44
are successively grown on the (100) surface of the p type InP substrate
40. Thereafter, a mask pattern for selective etching is formed on the n
type InGaAsP layer 44 using photolithographic techniques, and the n type
InGaAsP layer 44 is patterned in a plurality of stripe-shaped ridges which
are periodically arranged parallel to each other and perpendicular to what
becomes the resonator length direction of the laser, preferably by wet
etching using HBr as an etchant, whereby a diffraction grating is
produced. If the stripe-shaped ridges of the n type InGaAsP layer 44 are
formed in the [011] direction, each ridge has a trapezoid section as shown
in FIG. 9(b).
Thereafter, the n type InP cladding layer 45 is grown on the n type InP
barrier layer 43 and on the diffraction grating 44, preferably by MOCVD.
During the MOCVD growth, the diffraction grating 44 is irradiated with
light 21, such as laser light, and diffracted light 22 is monitored,
whereby the moment when the diffraction grating 44 is completely embedded
by the n type InP cladding layer 45 as shown in FIG. 9(c) is detected. The
incident light 21 is applied perpendicular to the stripe direction of the
diffraction grating 44.
In the above-described production process of the semiconductor laser, when
the first n type InP upper cladding layer 45 is grown on the diffraction
grating comprising the stripe-shaped InGaAsP layers 44 by vapor phase
deposition, the diffraction grating unfavorably loses its initial shape
due to mass transport in the InGaAsP layers, whereby the thickness and
amplitude of the diffraction grating are reduced, resulting in difficulty
in controlling the coupling coefficient that connects the intensity of
distributed feedback applied to light. The unwanted mass transport is
suppressed by decreasing the growth temperature of the first n type InP
upper cladding layer 45. However, since low temperature growth does not
provide a high-quality crystalline layer, it is not desired that the
entire first upper cladding layer 45 be formed at a low temperature.
Employing the evaluation method according to this second embodiment of the
present invention, the moment when the diffraction grating comprising the
stripe-shaped n type InGaAsP layers 44 is completely embedded is easily
detected. Therefore, the first n type InP upper cladding layer 45 is grown
at a relatively low temperature until this moment and, thereafter, it is
grown at a relatively high temperature, whereby mass transport is
suppressed and the crystal quality of the first upper cladding layer 45 is
improved.
As described above, according to the second embodiment of the present
invention, the n type InGaAsP layer 44 is patterned in a plurality of
stripe-shaped parallel ridges to form a diffraction grating, and the first
n type InP upper cladding layer 45 is epitaxially grown thereon. During
the epitaxial growth, the stripe-shaped ridges of the diffraction grating
are irradiated with light, and light diffracted by the stripe-shaped
ridges is monitored, whereby the moment when the stripe-shaped ridges are
completely embedded in the first n type InP upper cladding layer 45 is
detected. Therefore, the growth conditions can be controlled with high
precision, improving crystal quality and structure controllability.
FIGS. 10(a)-10(c) are sectional views for explaining a method for
evaluating an epitaxially growing layer in accordance with a third
embodiment of the present invention. In these figures, reference numeral
55 designates a GaAs substrate. A first TEG 56 and a second TEG 57 are
formed at prescribed positions on the surface of the GaAs substrate 55.
The first TEG 56 includes a plurality of stripe-shaped portions of the
substrate 55, each having a width W1 and a height H1. The second TEG 57
includes a plurality of stripe-shaped portions of the substrate 55, each
having a width W2 and a height H2. A first epitaxial layer 58 is grown on
the substrate 55, and a second epitaxial layer 59 is grown on the first
epitaxial layer 58.
The widths W.sub.1 and W.sub.2 have a relation of W.sub.1 <W.sub.2, and the
heights H.sub.1 and H.sub.2 have a relation of H.sub.1 =H.sub.2.
Initially, the first epitaxial layer 58 is grown on the substrate 55
including the first and second TEG's 56 and 57. In the step of FIG. 10(a),
the epitaxial growth proceeds forming (111)B planes on the stripe-shaped
ridges of the respective TEG's 56 and 57, as described in the first
embodiment. During the epitaxial growth, the TEG's 56 and 57 are
irradiated with light, and diffracted light from the respective TEG's are
monitored.
In the step of FIG. 10(b), the moment when only the first TEG 56 is flatly
embedded in the first epitaxial layer 58 is detected by the monitoring the
light diffracted from the first TEG 56, and the growth of the first
epitaxial layer 58 is stopped at this moment. Thus, a first epitaxial
layer 58 having a desired thickness d.sub.1 approximately equal to 0.7
W.sub.1 +H.sub.1 is formed on a region of the substrate 55 where the TEG's
56 and 57 are absent. Thereafter, a second epitaxial layer 59 is grown on
the first epitaxial layer 58. In the step of FIG. 10(c), the moment when
the second TEG 57 is flatly embedded is detected by the monitoring the
light diffracted from the second TEG 57, and the growth of the second
epitaxial layer 59 is stopped at this moment. Thus, a second epitaxial
layer 59 having a desired thickness d.sub.2 approximately equal to (0.7
W.sub.2 +H.sub.2)-(0.7 W.sub.1 +H.sub.1) is formed on the first epitaxial
layer 58 in the region where the TEG's 56 and 57 are absent.
Accordingly, the first and second epitaxial layers 58 and 59 are grown to
the desired thicknesses with high precision by appropriately selecting the
widths W.sub.1 and W.sub.2 and the heights H.sub.1 and H.sub.2 of the
stripe-shaped ridges of the first and second TEG's 56 and 57,
respectively.
While in the above-described third embodiment the first and second TEG's,
which are different in size, are formed on the substrate and the
thicknesses of the first and second epitaxial layers grown on the
substrate are evaluated, three or more TEG's may be formed to evaluate
thicknesses of three or more epitaxial layers.
The evaluation method of the present invention is applied to epitaxial
growth on a GaAs substrate in the above-described first and third
embodiments and to production of an InP laser in the above-described
second embodiment. However, the evaluation method of the present invention
may be applied to epitaxial growth of other materials.
FIG. 11 is a perspective view of a substrate including a TEG for evaluating
an epitaxial layer grown on the substrate, in accordance with a fourth
embodiment of the present invention. In FIG. 11, reference numeral 61
designates a GaAs substrate having a (100) surface orientation, numeral 62
designates a region where semiconductor elements are formed (hereinafter
referred to as element region), and numeral 63 designates a TEG formed on
a region of the substrate 61 other than the element region 62. FIG. 12 is
an | | |
