 |
Description  |
|
|
STATEMENT AS TO RIGHTS TO INVENTIONS MADE UNDER FEDERALLY SPONSORED
RESEARCH AND DEVELOPMENT
A portion of this work was done under ARPA grant F19628-94-C-0023.
BACKGROUND
The present invention relates to semiconductor lasers. More specifically,
the present invention relates to semiconductor lasers having associated
electronic components integrally formed therewith.
Semiconductor lasers are important devices used in a variety of
applications including printing, scanning, communications, etc.
Semiconductor lasers generally fall into two categories: edge-emitting and
vertical cavity surface emitting (VCSEL). Each of these types of devices
are well known. In an edge emitting semiconductor laser structure, a
number of layers are deposited onto a substrate. Following deposition, the
edges of the structure are cleaved to form partially transmissive mirrors.
One or more of the deposited layers forms an optical cavity, bound at its
edges by the mirrors. Lasing occurs within the cavity between the mirrors,
and the laser beam exits at one or both of the edges of the laser
structure in a direction parallel to the plane of the layers.
Surface emitting lasers are similar in concept, but differ in that the
laser beam is emitted orthogonal to the plane of the active layer(s). The
mirrors are above and below the optical cavity, as opposed to at each edge
of the cavity. For certain applications, a surface emitting laser provides
advantages over an edge emitting laser. For example, 2-dimensional arrays
of vertical cavity lasers may be produced in wafer form, whereas edge
emitting lasers typically must be mechanically jointed to form such
arrays. Also, surface emitting lasers typically emit circularly symmetric
Gaussian beams, as compared to highly eccentric elliptical beams of edge
emitting lasers. Accordingly, today there is much interest and development
centered around surface emitting lasers.
Associated with any semiconductor laser are numerous electronic components.
For example, the power of a laser is typically controlled by the drive
current applied to its electrodes; and one or more electronic components
such as transistors, capacitors, diodes, etc. forming drive circuitry may
be employed to control the drive current. As another example, components
such as transistors are often employed in addressing circuitry for
addressing individual lasers in arrays of such devices.
More specifically, in arrays of lasers, it is desirable to be able to
independently address each laser. This becomes problematic when dealing
with large arrays of such lasers. In such large arrays, the small size and
large density of the electrodes to which connection must be made increase
the complexity of making connections. Furthermore, the need to produce
small size arrays limits the surface area which addressing connections and
circuitry are permitted to occupy. To produce such arrays within a
practical cost structure, the addressing circuitry and scheme must be
relatively simple. Finally, the addressing circuitry and scheme must
support rapid addressing of each laser.
While there are many examples of addressing circuitry and schemes in the
art, efforts to date have not been successful in producing VCSELs having
integrated (i.e., formed either as part of the process of forming the
VCSEL or formed above the VCSEL as part of subsequent processing)
associated electronic component structures. Rather, addressing circuitry,
as well as driver circuitry and other related components, have been built
external to the laser structure itself, then interconnected for operation.
For example, FIG. 1 is an illustration of a laser structure and separately
connected voltage source 10. Laser 12 is formed on a substrate 14,
typically GaAs. A number of thin layers are first deposited to form a
lower mirror region 16, an n-type layer 18 is formed on lower mirror 16,
an intrinsic active layer 20 is formed on n-type layer 18, a p-type layer
22 is formed on intrinsic active layer 20, and an upper mirror layer 24 is
formed on p-type layer 22. Typically, a metal, n-material electrode 26 is
formed below the substrate 14, and a metal, p-material electrode 28 is
formed above the upper mirror region 18. Electrode 28 is typically annular
in planform, so as to maximize surface contact yet minimize interference
with the laser beam B.sub.1 generated by the structure.
A voltage is then applied to electrode 28 from external voltage supply 30
and addressing circuitry 32. Electrode 26 is typically connected to ground
potential. The current through the laser 12 results in the generation of
laser beam B.sub.1. The requirement of external voltage supply, addressing
circuitry, and potentially other electronic components associated with
laser 12 limits the ability to reduce size, cost, component complexity,
etc., and increase speed, efficiency, etc.
In addition, in many laser systems it is necessary to measure and control
the power of the beam emitted by the laser. For example, it is necessary
in many applications to provide a constant, predetermined beam power,
which requires compensation for the laser's temperature, aging, etc. Beam
power detection generally involves interposing a detector in a laser beam
path. In the case of certain edge emitting lasers, this may be
accomplished by detecting one of two beams. That is, where an edge
emitting laser is of the type having two beam emissions, one from each
edge (facets), one beam is referred to as a forward emission beam and the
other as a rear emission beam. The forward emission beam will generally be
of a higher power than the rear emission beam. Hence, the forward emission
beam is generally the operable beam performing the desired function, such
as writing to a photoreceptor, pulsing encoded signals to a transmission
line, cutting material, etc., while the rear emission beam is often not
used. However, the ratio of the power of the forward emission beam to the
power of the rear emission beam can be measured. Thus, by placing a
detector in the path of the rear emission beam, and by employing the
aforementioned power ratio, the power of the forward emission beam may be
determined.
This approach has limited utility for surface emitting laser structures,
for several reasons. Typically, surface emitting laser structures include
a gallium arsenide (GaAs) substrate, which is opaque for wavelengths
shorter than 870 nm. Thus, for most applications, the substrate will be
opaque, and the laser structure will be capable of producing only a
single, surface emitted laser beam. Second, in general it is desirable to
provide as high a beam power as possible, so it is a design goal to
produce single beam laser structures.
An existing approach to incorporating a detector into a single beam surface
emitting laser (assumed to be single beam herein, unless otherwise stated)
is to form the detector in the laser structure. That is, additional layers
would be epitaxially grown above the laser structure, but of the same
material as the laser structure, which would be appropriately patterned
and/or doped, and interconnected to form a detector. The detector is
generally coaxial with the laser beam, and relies on partial absorption of
the beam to create electron-hole pairs which are detected by methods
otherwise known in the art.
The upper detector electrode will have an inside diameter d, which is at
its smallest equal to the diameter of the laser beam generated by the
underlying laser. In operation, the detector converts photon energy from
the laser beam to electron-hole pairs, which migrate to respective
electrodes. Beam power is thus measured by measuring the extent of
electron-hole pair generation (i.e., current generated in the detector).
The speed of the detector is measured by the speed at which the electrons
or holes travel to their respective electrodes. Those electrons or holes
generated at the center of the annular detector electrode must travel a
distance equal to at least d/2. This is a relatively large distance, and
results in relatively slow detection.
Third, the detector is essentially a p-i-n photodiode. The detector's p-
and n-type layers are formed of GaAs, a material opaque to the laser beam
when the combined thickness of the layers is 1000 .ANG. or greater (again,
assuming a laser wavelength shorter than 879 nm). Typically, a design
point for absorption of the beam by the detector is around 5% of the
beam's energy. However, if the layers are too thin, they will be too
transparent, and not absorb sufficient photon energy to effectively serve
as a detector. Thus, to obtain the desired performance from the detector,
it is necessary to very precisely control the thicknesses of the p- and
n-type layers, which adds to the cost and complexity of manufacture.
Whether circuitry or detector, it should be possible to tailor the
thickness of the layers formed above the VCSEL to "tune" the interference
of the reflected portions of the laser beam. This would provide improved
efficiency and performance of the laser. However, to accomplish this, some
flexibility to adjust layer thicknesses is required, which the constraints
of the GaAs layers cannot provide.
In this regard, it becomes necessary to effect modifications to the laser
itself to account for the additional layers at the VCSEL's surface. For
example, the only way to compensate for destructive reflections from such
additional layers is to adjust the thicknesses of the various laser
structure layers, such as the thickness of the lower mirror layer, part of
which serves as the lower mirror below the optical cavity. This constrains
optimizing the laser structure design for peak laser performance.
Furthermore, there is no teaching in the art of an integrated VCSEL,
sensor, and associated electronic components. According to all heretofore
known techniques, such integration would significantly increase the
complexity and cost of manufacturing the laser structure. In addition, the
size and position limitations for the various contacts, due to their being
opaque, mean that an extensive network of interconnections would be
required, limiting the compaction of laser structures desired of arrays of
such structures.
Consequently, the currently accepted approach for connecting laser
structures with associated electronic components is to connect such
structures to external circuitry. This implies a sacrifice of speed,
compactness, and system complexity which the present invention strives to
overcome.
In fact conventional electronics used in conjunction with VCSELs are
virtually exclusively formed of single crystal silicon. However, single
crystal electronics must be formed over a single crystal substrate, and
thus the forming such electronics on or together with a VCSEL is very
difficult if not impossible.
SUMMARY OF THE INVENTION
The present invention is a process and structure which address the
shortcomings of the prior techniques of integrating a laser structure and
associated electronic components. As used herein, an "electronic
component" is any device capable of carrying, modifying, storing,
generating or otherwise interacting with an electrical signal. Also as
used herein, an electronic component is "associated" with the laser if it
is either directly or indirectly in electrical, optical or thermal
communication with the laser. According to one embodiment of the present
invention, a VCSEL structure is provided with matrix addressing circuitry,
in the form of transistor elements, allowing addressing of that laser
structure. The VCSEL structure may be a stand-alone laser, or may be one
of a number of such laser structures forming an array of such devices,
with the associated circuitry for addressing individual ones of the lasers
forming the array.
According to one embodiment, the transistor structures are provided in
layers deposited over the layers in which said laser structure is formed.
Also according to one embodiment of the present invention, the channel
layer(s) of such transistor structures are formed of hydrogenated
amorphous silicon (a-Si:H). According to still another embodiment, the
channel layer(s) of one or more of such transistor structures are formed
of polycrystalline silicon (polysilicon).
In addition to the formation of transistor structures, other circuitry such
as capacitors, resistors, diodes, etc. may be integrally formed over the
laser structures to accomplish desired functionality. For example, the
matrix addressing scheme of one embodiment of the present invention
benefits from the integral formation of one or more capacitors with the
formation of the transistor structures.
Furthermore, in cases where an integrated detector structure is desirable,
the laser structure is provided at its output surface with a p-i-n
photodiode sandwiched between transparent electrode layers, for example
formed of indium tin oxide (ITO). The p-material electrode of the detector
may be common with the p-material electrode of the surface emitting laser,
or may be a separate electrode.
Formed at least in part in the same layers as the detector structure is one
or more associated electronic components, such as transistors. Such
integrated transistors provide, for example, on-chip matrix addressing of
either or both the laser structure and sensor structure. However, provided
the ability to integrally form arbitrary circuitry with the VCSEL
structure, virtually any application of such circuitry in conjunction with
the laser and/or sensor structures may be enabled by the present
invention.
According to a process comprising one embodiment of the present invention,
a VCSEL structure is formed by processes otherwise well known in the art.
One or more associated electronic components, such as transistors, are
formed directly thereupon at low temperatures, and in a manner that
otherwise does not adversely affect the performance of the laser
structure.
According to another embodiment of the present invention, a detector
structure is deposited directly upon the VCSEL structure together with
transistors, at low temperatures, and in a manner that otherwise does not
adversely affect the performance of the VCSEL. Alternatively, the detector
structure may be formed on a glass, quartz or similarly transparent
substrate, then subsequently bonded to the VCSEUintegral transistor
structure. The bonding process and materials selected are benign to the
structure and operation of the laser, transistor, and detector structures.
The shortcomings of the prior techniques are addressed since no additional
photolithography of the GaAs structure is required. Any deposition,
exposure, etching, etc., required to fabricate the associated electronic
components and/or detector structure takes place subsequent to and
separate from the processing of the laser structure. And, as mentioned,
the processes involved in fabricating the transistor and/or detector
structure may be benign to the underlying surface emitting laser
structure.
Finally, the characteristics of a-Si:H and polysilicon circuitry arrays
have been studied for some time, and the benefits of these studies may be
employed in the integrated structures according to the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
A more complete appreciation of the invention and many of the attendant
advantages thereof will be readily obtained and understood by referring to
the following detailed description and the accompanying drawings in which
like reference numerals denote like elements as between the various
drawings. The drawings, briefly described below, are not to scale.
FIG. 1 is a perspective view of a vertical cavity surface emitting laser
structure with external circuitry according to the prior art.
FIG. 2 is a cross-sectional view of an improved vertical cavity surface
emitting laser structure and integrated detector structure according to
the present invention.
FIG. 3 is a schematic representation of one embodiment of an integrated
laser and transistor structure.
FIG. 4 is a plan view of one cell of an integrated laser and transistor
structure used for addressing the laser corresponding to the schematic
illustration of FIG. 3.
FIG. 5(a) and FIG. 5(b) show generic embodiments of the present invention.
FIG. 6 is a schematic representation of another embodiment of an integrated
laser and transistor structure.
FIG. 7 is a plan view of one cell of an integrated laser and transistor
structure used for addressing the laser, incorporating a capacitor
structure, corresponding to the schematic illustration of FIG. 6.
FIG. 8 is a schematic representation of one embodiment of an integrated
laser and detector structure with a further integrated detector transistor
structure.
FIG. 9 is a plan view of one cell of an integrated laser and detector
structure, including an integrated transistor structure used for
addressing the sensor, corresponding to the schematic illustration of FIG.
8.
FIG. 10 is a schematic representation of another embodiment of an
integrated laser and detector structure with integrated laser and detector
transistor structures.
FIG. 11 is a plan view of one cell of an integrated laser and detector
structure including integrated transistor structures used for separately
addressing the laser and detector corresponding to the schematic
illustration of FIG. 10.
FIG. 12 is a timing diagram showing various voltages for the cell of an
integrated laser and detector structure including integrated transistor
structures used for separately addressing the laser and detector
corresponding to the schematic illustration of FIG. 10 and plan view of
FIG. 11.
FIG. 13 is a cross section of one embodiment of the present invention in
which a detector and transistor structure are formed on a glass substrate
and bonded to an integrated laser and transistor structure.
FIG. 14 is a cross section of one embodiment of the present invention in
which a detector and transistor structure are formed on a glass substrate
and bonded to an integrated laser and transistor structure, the laser and
detector structures being in electrical communication with one another.
FIG. 15 is a cross-sectional view of another embodiment of the present
invention employing a photoconductive layer as a component of an
integrated detector.
DETAILED DESCRIPTION
An integrated surface emitting laser and associated electronic components
50 according to one embodiment of the present invention shall now be
described with reference to FIG. 2. According to the present embodiment,
the associated electronic components are thin-film transistors. This,
however, is only one example illustrating the concepts of the present
invention, and shall not limit the scope of the invention disclosed
herein.
A fundamental component of the present invention is the vertical cavity
surface emitting laser (VCSEL) structure 52. As illustrated in FIG. 2,
VCSEL 52 includes a substrate 54, for example n-type gallium arsenide
(GaAs). The doping of the GaAs substrate 54 may typically be in the range
of 5.times.10.sup.18 cm.sup.-3. Formed on or below a first side of
substrate 54 is first electrode layer 56. Typically, this first electrode
layer 56 is an n-type material, such as an alloy of germanium and gold
(GeAu).
Formed on or over a second side of substrate 54 is mirror structure 58.
Mirror structure 58 (as well as other mirror structures described herein)
may be a super lattice of multiple thin layers, or other type structure
forming a distributed Bragg reflector ("DBR"). A buffer layer of about 200
to 800 nm, which is not shown in FIG. 2, may be first formed on substrate
54 so that it lies between substrate 54 and mirror structure 58. This
lower mirror structure 58 provides part of the necessary internal
reflection required to form the VCSEL.
Lower mirror structure 58 typically comprises multiple high and low
aluminum content aluminum gallium arsenide (AlGaAs) or aluminum arsenide
(AlAs) alloy layer pairs. Each of these layers have an aluminum content,
in the range of 0% to 15% for the low aluminum layers, and in the range of
85% to 100% for the high aluminum layers, controlled so that they achieve
a desired refractive index in each layer. A typical target refractive
index for the high aluminum content layer is around 3.1 to 3.2, and the
target refractive index for the low aluminum content layer is around 3.5
to 3.6. The aluminum content of these layers must also be controlled to
render them non-absorptive at the lasing wavelength. The uppermost layer
of the lower mirror structure is a typically a high aluminum content
layer.
The target total thickness of each of the layer pairs is one half of the
optical wavelength at the desired laser operation wavelength. The target
thickness of the uppermost layer of the lower mirror structure 58 is one
quarter of the optical wavelength at the desired laser operation
wavelength. A typical laser operation wavelength is 820 nm.
Since the laser output is outcoupled through only one surface (the upper
surface) of the VCSEL, high internal reflection will maximize the laser
output. High internal reflection also generally reduces the required
threshold current of the laser. Thus, the reflectivity of the lower mirror
structure 58 should be as close to 100% as possible. It is well-known that
the reflectivity of the lower mirror structure 58 is, in part, a function
of the difference in the refractive indices between the high and low
aluminum content layers. The reflectivity is also a function of the number
of layer pairs in the structure. The greater the difference in the
refractive indices, the fewer number of layer pairs are required to reach
a given reflectivity. Typically, 20 to 40 pairs of high and low layers are
used to form the lower mirror structure 58. Lower mirror structure 58 will
generally be doped so as to be of n-type material. Silicon is a common
material with which lower mirror structure 58 may be doped to render it
n-type.
Formed on or over mirror structure 58 is first spacer layer 60. Spacer
layer 60 is typically formed of AlGaAs with an aluminum content of about
40%, and a thickness of about 100 nm. First spacer layer 60 may be doped
with silicon, selenium, tellurium, etc. to a concentration of about 1 to
3.times.10.sup.18 cm.sup.-3. Generally, the uppermost 20 nm of first
spacer layer 60 remains undoped.
On or over first spacer layer 60 is active layer 62. Active layer 62 may be
a single quantum well structure formed of GaAs or p- or n-doped GaAs or
Al.sub.z Ga.sub.1-z As, where z is very small, (Al.sub.z
Ga.sub.1-z).sub.0.5 ln.sub.0.5 P. Alternatively, active layer 62 may be a
relatively thin conventional double heterostructure (DH) structure, a
multiple quantum well structure, such as alternating layers of materials
such as GaAs and Al.sub.z Ga.sub.1-z As, or one of many other known
structures for active layers of semiconductor lasers. It is the material
comprising active layer 62 which is defined herein as the material of
which VCSEL 52 is predominantly comprised.
On or over active layer 62 is second spacer layer 64. As with first spacer
layer 60, second spacer layer 64 is typically formed of AlGaAs with an
aluminum content of about 40%, and a thickness of about 100 nm. Second
spacer layer 64 may be doped with magnesium, carbon, zinc, etc. to a
concentration of about 1 to 3.times.10.sup.18 cm.sup.-3. Generally, the
lowermost 20 nm of first spacer layer 64 remains undoped. First spacer
layer 60, active layer 62, and second spacer layer 64 together form the
optical cavity 66 in which the desired optical gain is achieved. The
combined optical thickness of the layers comprising this optical cavity is
optimally equal to an integral multiple of half of the optical wavelength
in the cavity.
On or above second spacer layer 64 is upper mirror structure 68. Upper
mirror structure 68 forms another DBR mirror, and is structurally similar
to lower mirror structure 58, except that it is doped to have opposite
polarity. That is, if lower mirror structure 58 is doped n-type, upper
mirror structure 68 is then doped p-type. In addition, upper mirror
structure 68 is formed to have a slightly reduced reflectivity as compared
to lower mirror structure 58 for the reason that the optical emission is
to be outcoupled therethrough. Typically, upper mirror structure 68 will
have a reflectivity on the order of 98% to 99%.
On or above upper mirror structure layer 68 is second electrode 70. Second
electrode 70 is typically formed of a chrome and gold bilayer, titanium,
platinum, and gold trilayer, or other similar metal system. As previously
discussed, since the material from which second electrode 70 is formed is
typically opaque, second electrode 70 must provide a region through which
the laser beam may exit the VCSEL structure. Second electrode 70 must have
a large surface area contact with its contiguous underlying layer to
provide minimal electrical resistance, yet be small enough to permit
formation of high density arrays of VCSELs. Thus, second electrode 70
often has an annular, hollow elliptical or similar planform shape.
Finally, since second electrode 70 is in contact with what has previously
(and arbitrarily) been referred to as the p-type mirror, it is also
referred to as the p-type electrode. Thus is presented an exemplary VCSEL
according to one embodiment of the present invention.
VCSELs are ideally suited to integration into large numbers of nearly
identical devices (arrays). Lithographic techniques and ion beam
processing have yielded VCSELs with diameters below 1 micrometer. The
accompanying reduction in the active volume of the VCSEL decreases driving
threshold currents, and permits formation of large, complex structures of
coherently coupled arrays. However, formation of such devices is processes
dependent. One process for forming such VCSELs is ion implantation. Such a
process is well known in the art (see, e.g., Morgan et al., IEEE Photonics
Technology Letters, vol. 7, no. 5, 1995, and the references cited
therein).
Formed atop (from the perspective of FIG. 2) VCSEL 52 is electronic
circuitry layer 100. As will be appreciated, there are many variations of
the topography, strata, and materials used to form layer 100. Furthermore,
the number and type of components formed in layer 100 are a matter of
design choice, and constrained principally by the area available in which
they may be fabricated and the performance of the materials from which
they are formed in light of the required component or system performance.
Thus, it is particularly important that it be understood that the
description contained herein is merely illustrative, and not in any way
limiting as to the scope of the invention.
In addition, it is important to appreciate that VCSEL 52 and the
component(s) formed in layer 100 may be formed of different materials. The
material of the active layer 62 of VCSEL 52 has been defined as the
predominant material from which VCSEL 52 is formed. Certain electronic
components, such as transistors, also have active layers, and the material
of such active layers is herein defined as the predominant material from
which the component is formed. However, certain components do not have
active layers as such. For such a component, the material comprising the
greatest volume of the component, excluding the electrodes of such
component, is defined herein as the predominant material from which such
component is formed. Indeed, the difference in these materials may be a
difference in alloys, difference in compositions, difference in crystal
structures (e.g., single crystal, polycrystalline, and amorphous), etc.
In the embodiment shown in FIG. 2, a passivation layer 102 is deposited
over the completed laser structure 52. Passivation layer 102 may either be
transparent to light emitted by VCSEL 52, or have an opening formed
therein (not shown) to allow the light emitted by VCSEL 52 to exit the
structure. A conductive via 104 is formed in passivation layer 102 for
subsequent electrical communication between the drain of a transistor
structure and electrode 70. Additionally, a conductive region 106 is
formed in passivation layer 102 for subsequent electrical communication
between a gate of a first transistor structure and the drain of a second
transistor structure, which will be described further below.
In the embodiment shown in FIG. 2, first and second transistor structures
108 and 110, respectively, are formed. Each transistor structure is
provided with a gate electrode, 112 and 114, respectively, formed of a
metal such as aluminum. Over each gate electrode is insulating material
116 and 118. As illustrated, insulating material 118 extends to a width
greater than the width of insulating material 116 in order to provide
isolation between drain and source electrodes of the two transistor
structures 108 and 110. Insulating material 116, 118 may typically be a
nitride such as silicon nitride (SiN).
Channel material 120 and 122, comprised of a-Si:H, polysilicon or the like,
is next deposited onto the insulating material 116, 118 respectively. If
the channel material is comprised of a-Si:H, gate 114 must be relatively
large to accommodate the large laser driving current. Passivation material
124, 126 is next deposited onto channel material 120, 122. Passivation
material 124, 126 may typically be SiN, polymide, or other material well
known to passivate and insulate the surface of channel material 120, 122.
The sources and drains of transistor structures 108 and 110 are next
formed. Deposition and masking techniques otherwise well known in the art
are employed to deposit isolated source 128 and drain 130 of first
transistor structure 108, and source 132 and drain 134 of second
transistor structure 110. Importantly, drain 130 is i | | |
