Schottky barrier diode

Information

  • Patent Grant
  • 6627967
  • Patent Number
    6,627,967
  • Date Filed
    Friday, July 26, 2002
    22 years ago
  • Date Issued
    Tuesday, September 30, 2003
    21 years ago
Abstract
A Schottky barrier diode has a Schottky contact region formed in an n epitaxial layer disposed on a GaAs substrate and an ohmic electrode surrounding the Schottky contact region. The ohmic electrode is disposed directly on an impurity-implanted region formed on the substrate. An insulating region is formed through the n epitaxial layer so that an anode bonding pad is isolated form other elements of the device at a cathode voltage. The planar configuration of this device does not include the conventional polyimide layer, and thus has a better high frequency characteristics than conventional devices.
Description




BACKGROUND OF THE INVENTION




1. Field of the Invention




The invention relates to a Schottky barrier diode device made of a compound semiconductor and applied in a high frequency circuit, specifically to a Schottky barrier diode having a planar configuration to achieve a smaller operation region and a smaller overall chip size.




2. Description of the Related Art




The demand for high frequency devices has been rapidly increasing due to the expanding market for portable telephones and digital satellite communication equipment. Many of such devices include field effect transistors (referred to as FET, hereinafter) employing a gallium arsenide (referred to as GaAs, hereinafter) substrate because of its excellent high frequency characteristics. Typical application in this field includes local oscillation FETs for satellite antenna and monolithic microwave integrated circuits (MMIC) in which a plurality of FETs are integrated for wireless broadband. GaAs Schottky barrier diodes are also used in a base station of cellular phone system.





FIG. 1

is a cross-sectional view including an operation region of a conventional Schottky barrier diode. An n+epitaxial layer


22


(a silicon impurity concentration of about 5×10


18


cm


−3


) having a thickness of about 6 μm is formed on an n+ GaAs substrate


21


. An n epitaxial layer


23


(a silicon impurity concentration of about 1.3×10


17


cm


−3


) having a thickness of about 350 nm is formed on the n+ epitaxial layer


22


. This n epitaxial layer provides as an operation region.




An ohmic electrode


28


makes a ohmic contact with the n+ epitaxial layer


22


and is made of a AuGe (gold-germanium alloy)/Ni (nickel)/Au (gold) metal layer disposed as a first wiring layer. A Ti (titanium)/Pt (platinum)/Au metal


32


serves as a second wiring layer, and is divided into wiring on the anode side and wiring on the cathode side. On the anode side, the Ti/Pt/Au metal layer makes a Schottky contact with the n epitaxial layer


23


, and forms a Schottky contact region


31




a


. The portion of the Ti/Pt/Au metal layer on the anode side above the Schottky contact region


31




a


is referred to as a Schottky electrode


31


. An anode electrode


34


is formed on and completely overlaps with the Schottky electrode


31


and its extension. The anode electrode


34


provides an anode bonding pad


34




a


and is formed by Au plating using the Schottky electrode


31


and its extension as a plating electrode. A bonding wire


40


is fixed to the anode bonding pad


34




a


. The Au metal layer serves as a third wiring layer and has a thickness of about 6 μm. The thick Au layer is necessary for providing stress relief during wire bonding. On the cathode side, the cathode electrode


35


provides a cathode bonding pad and is formed of the Au layer. The Ti/Pt/Au metal layer on the cathode side directly contacts the ohmic electrode


28


. The edge of the Schottky electrode


31


needs to be on a top surface of a polyimide layer


30


to satisfy photolithographic requirements. Accordingly, a portion of the Schottky electrode


31


, near the Schottky region


31




a


, overlaps by about 16 μm with the polyimide layer


30


formed on the ohmic electrode


28


on the cathode side. The entire substrate and epitaxial layers are at a cathode voltage except the Schottky contact region


31




a


. The polyimide layer


30


insulates the anode electrode


34


from the substrate


22


and the epitaxial layers. The intersection between the anode electrode


34


and the underlying structure and the intersection between the anode bonding pad


34




a


and the underlying structure are, in total, about 3900 μm


2


, which could provide a large parasitic capacitance to the device if the thickness of the polyimide layer


30


is small. Thus, to have a reasonably small parasitic capacitance, the thickness of the polyimide layer must be as large as 6-7 μm even though the polyimide film


30


has a relatively low dielectric constant.




The n epitaxial layer


23


of the lower impurity concentration (1.3×10


17


cm


3


) is necessary for assuring a Schottky contact region


31




a


with good Schottky characteristics and a high breakdown strength (10V). The ohmic electrode


28


is formed directly on the n+ epitaxial layer


22


for reducing the resistance at the contact. For this reason, a mesa etching process is necessary for exposing the top surface of the n+ epitaxial layer


22


. The n+ GaAs substrate


21


underneath the n+ epitaxial layer


22


also has a high impurity concentration, and has a backside electrode made of the AuGe/Ni/Au metal layer for an external contact from the backside.





FIG. 2

is a schematic top view of the conventional Schottky barrier diode having the operation region shown in FIG.


1


. The Schottky contact region


31




a


formed in the n epitaxial layer


23


occupies a central portion of the device. The diameter of this region


31




a


is about 10 μm. A Schottky contact hole


29


is formed in the center of the Schottky contact region


31




a


. The Ti/Pt/Au metal layer of the second wiring layer is in direct contact with the n epitaxial layer


23


through the contact hole


29


. The ohmic electrode


28


of the first wiring layer surrounds the circular Schottky contact region


31




a


, and occupies almost a half of the top surface of the device.




The Au metal layer of the third wiring layer provides the bonding pads. On the anode side, the anode bonding pad


34




a


has a minimum area allowed for one bonding wire


40


. On the cathode side, the cathode bonding pad


35




a


is large enough to provide bonding of multiple wires


40


, which is required for reducing the inductance generated at bonding wires. The cathode bonding pad


35




a


and the cathode electrode


35


are formed directly on the ohmic electrode


8


disposed on the n+ epitaxial layer without the intervening polyimide layer


30


. The area of the anode bonding pad


34




a


is about 40×60 μm


2


and the area for the cathode bonding pad is about 240×70 μm


2


.




However, the mesa etching, which is required to expose the n+ epitaxial layer


22


through the n epitaxial layer


23


for the direct contact with the ohmic electrode


28


, is not stable enough to provide accurate patterning of the device. For example, the wet etching process used in the mesa etching may remove the oxide film


25


around the contact hole


29


, leading to formation of mesa with an irregular shape. Such an irregular mesa structure may cause adverse effects on the Schottky barrier diode, especially the characteristics of the Schottky contact region


31




a.






Furthermore, the polyimide layer


30


has a thickness as large as 6-7 μm to reduce the parasitic capacitance generated between the Schottky electrode


31


and the underlying structures (the epitaxial layers


22


,


23


and the substrate


21


) at the cathode voltage. To form a step coverage of this thick polyimide layer


30


with the electrodes


31


,


34


,


35


, the edges of the polyimide layer


30


near the Schottky contact region


31




a


must have a tapered cross-section, as shown in FIG.


1


. Such a tapered structure gives rise to a variation of the tapering angle, typically between 30 and 45 degrees. To accommodate this variation, a long separation between the Schottky contact region


31




a


and the ohmic electrode


28


is required. This separation leads to a large resistance and, thus, poor high frequency characteristics. The device shown in

FIG. 1

has a separation of about 7 μm.




It should also be noted that the large area (2400 μm


2


) of the anode bonding pad


34




a


further contributes to the increase of the parasitic capacitance of the diode device. Furthermore, the polyimide layer


30


and the thick Au layer are made of expensive materials and their use inevitably increases the production cost.




SUMMARY OF THE INVENTION




The invention provides a Schottky barrier diode including a substrate made of a compound semiconductor and an epitaxially grown layer disposed on the substrate. The Schottky contact region of the device is a part of the epitaxially grown layer. An anode wiring is connected to the Schottky contact region. An insulating region penetrates through the epitaxially grown layer and reaches the substrate. An anode bonding pad is disposed on the insulating region and connected the anode wiring.




The invention also provides a Schottky barrier diode including a substrate made of a compound semiconductor and an epitaxially grown layer disposed on the substrate. The Schottky contact region of the device is a part of the epitaxially grown layer. An anode wiring is connected to the Schottky contact region. An insulating region penetrates through the epitaxially grown layer and reaches the substrate. An anode bonding pad is disposed on the epitaxially grown layer and connected to the anode wiring. The insulating region surrounds the anode bonding pad.




The invention further provides a Schottky barrier diode including a substrate made of a compound semiconductor and an epitaxially grown layer of a conduction type disposed on the substrate, which includes a Schottky contact region. An anode wiring is connected to the Schottky contact region. An insulating region penetrates through the epitaxially grown layer and reaches the substrate. A first impurity-implanted region of the conduction type penetrates through the epitaxially grown layer. An anode bonding pad is disposed on the first impurity-implanted region and connected to the anode wiring. The insulating region surrounds the anode bonding pad.




The invention further provides a Schottky barrier diode including a substrate made of a compound semiconductor and an epitaxially grown layer of a conduction type disposed on the substrate, which includes a Schottky contact region. An anode wiring is connected to the Schottky contact region. An anode bonding pad is disposed on the epitaxially grown layer and connected to the anode wiring. A first impurity-implanted region of the conduction type penetrates through the epitaxially grown layer and is disposed underneath a peripheral portion of the anode bonding pad. An insulating region penetrates through the epitaxially grown layer to reach the substrate and surrounds the first impurity-implanted region.











BRIEF DESCRIPTION OF THE DRAWINGS





FIG. 1

is a cross-sectional view of a conventional Schottky barrier diode having a polyimide layer.





FIG. 2

is a schematic top view of the conventional device of FIG.


1


.





FIG. 3

is a cross-sectional view of a Schottky barrier diode of a first embodiment of this invention.





FIG. 4

is a schematic top view of the device of the first embodiment.





FIG. 5

is a first modification of the device of the first embodiment.





FIG. 6

is a second modification of the device of the first embodiment.





FIGS. 7A-7E

show processing steps of a manufacturing method of the Schottky barrier diode of the first embodiment.





FIG. 8

is a cross-sectional view of a Schottky barrier diode of a second embodiment of this invention.





FIG. 9

is a schematic top view of the device of the second embodiment.





FIG. 10

is a first modification of the device of the second embodiment.





FIGS. 11A-11E

show processing steps of a manufacturing method of the Schottky barrier diode of the second embodiment.











DETAILED DESCRIPTION OF THE INVENTION





FIG. 3

is a cross-sectional view of a Schottky barrier diode of a first embodiment of this invention. Specifically,

FIG. 3

focuses on an operation region of the first embodiment. The compound semiconductor substrate


1


of this embodiment is an undoped GaAs substrate. An n+ epitaxial layer


2


having a thickness of about 500 nm and a silicon impurity concentration of about 5×10


18


cm


−3


is formed on the substrate


1


. An n epitaxial layer


3


having a thickness of about 250 nm and a silicon impurity concentration of about 1.3×10


17


cm


−3


is formed on the n+ epitaxial layer


2


. The n+ epitaxial layer


2


is much thinner than the n+ epitaxial layer


22


of the conventional device of

FIG. 1

so that an insulating region


6


easily reaches the substrate


1


in an impurity injection process as described later. As shown in

FIG. 3

, the device of the first embodiment has a planar configuration without any mesa structure.




An impurity-implanted region


7


is formed underneath the ohmic electrode


8


, and penetrates through the n epitaxial layer


3


to reach the n+epitaxial layer


2


. The impurity-implanted region


7


surrounds a circular Schottky contact region


11




a


, which is a part of the n epitaxial layer


3


, and completely overlaps with an ohmic electrode


8


. In the area near the Schottky contact region


11




a


, the impurity-implanted region


7


slightly extends beyond the ohmic electrode


8


and toward the Schottky contact region


11




a


. In this configuration, the separation between the Schottky contact region


11




a


and the impurity-implanted region


7


is 1 μm. The impurity-implanted region


7


, which reaches from the ohmic electrode


8


to the n+ epitaxial layer


2


through the n epitaxial layer


3


, replaces the mesa structure of conventional device and, thus achieves a planar device structure.




A nitride film


5


insulates an anode electrode


11


from the ohmic electrode


8


, the n+ epitaxial layer


2


and the n epitaxial layer


3


, which are at a cathode voltage.




The ohmic electrode


8


disposed on the impurity-implanted region


7


is a part of a first wiring layer made of a AuGe/Ni/Au metal layer. The first wiring layer is formed by depositing AuGe, Ni and Au films in this order. The ohmic electrode


8


has a rectangular shape with a hole corresponding to the Schottky contact region


11




a


(FIG.


4


). The separation between the Schottky contact region


11




a


and the ohmic electrode


8


is about 2 μm.




The portion of the anode electrode


11


above the Schottky contact region


11




a


makes a Schottky contact with the n epitaxial layer


3


through a Schottky contact hole formed in the nitride film


5


covering the surface of the n epitaxial layer


3


. The Schottky contact region


11




a


has a diameter of about 10 μm, and provides an operation region of the Schottky barrier diode. The thickness of the n epitaxial layer


3


(250 nm) is determined to assure a proper breakdown voltage.




An insulating region


6


is formed underneath an anode bonding pad


11




b


by implanting boron ions into the epitaxial layers


2


,


3


and the substrate


1


. The insulating region


6


, which penetrates into the substrate


1


, prevents the anode bonding pad


11




b


, which is at an anode voltage, from electrically contacting the epitaxial layers


2


,


3


, which are at the cathode voltage. Accordingly, the anode bonding pad


11




b


is disposed directly on the insulating region


6


without any polyimide layer or nitride film, and is free from high frequency signal leakage. For this configuration to be effective, the insulating region


6


should at least reach the insulating substrate


1


.




The cathode electrode


15


is formed of a second wiring layer made of a Ti/Pt/Au metal layer and is in contact with the ohmic electrode


8


. One end of the cathode electrode


15


faces the anode electrode


11


, and other end extends to a cathode bonding pad


15




b


, which is formed as a part of the second wiring layer. In other words, the cathode electrode


15


provides wiring between the ohmic electrode


8


and the cathode bonding pad


15




a


. The cathode bonding pad


15




a


is disposed directly on the n epitaxial layer


3


. Accordingly, the impurity-implanted region


7


, which is in contact with the ohmic electrode


8


, and the n+ epitaxial layer


2


are at the cathode voltage.




The anode electrode


11


is formed of the second wiring layer made of a Ti/Pt/Au metal layer and provides wiring between the Schottky contact region


11




a


and the anode bonding pad


11




b


. One end of the anode electrode


11


is in contact with the Schottky contact region


11




a


, and other end extends to the anode bonding pad


11




b


, which is formed as a part of the second wiring layer. In other words, the cathode electrode


15


provides wiring between the ohmic electrode


8


and the cathode bonding pad


15




a


. The cathode bonding pad


15




a


is disposed directly on the n epitaxial layer


3


. The anode bonding pad


11




b


is insulated from the n epitaxial layer


3


and the n+ epitaxial layer


2


, which are at the cathode voltage, by the insulating region


6


. Accordingly, high frequency signal leakage at the anode bonding pad


11




b


can be prevented. Furthermore, because there is no structure contributing to capacitance formation underneath the anode bonding pad


11




b


, the overall parasitic capacitance of the Schottky barrier diode of this embodiment is reduced.




Two bonding wires


40


are fixed to the anode bonding pad


11




b


. Four bonding wires


40


are fixed to the cathode bonding pad


15




b


. A stitch bonding process is applied to the fixation of the bonding wires


40


to the corresponding bonding pads


11




b


,


15




b


because the stitch bonding can fix two wires on the bonding pad in one bonding procedure, and, thus, reduce the required area for the bonding.





FIG. 4

is a schematic top view of the Schottky barrier diode of the first embodiment. The Schottky contact region


11




a


is located approximately in the center of the GaAs chip. The insulating region


6


is larger than the anode bonding pad


11




b


, which is formed on the insulating region


6


.




The area denoted by the broken line is the ohmic electrode


8


. The impurity-implanted region


7


(omitted form

FIG. 4

) is formed underneath the ohmic electrode


8


and covers approximately the same area of the substrate as the ohmic electrode


8


. The Schottky contact region


11




a


is located in the hole created in the ohmic electrode


8


. The cathode electrode


15


made of the second wiring layer is disposed on the ohmic electrode


8


to serve as wiring leading to the cathode bonding pad


15




a


. The external connection of the cathode electrode


15


has a significant influence on the high frequency characteristics, including inductance, of the Schottky barrier diode. In this embodiment, the size of the cathode bonding pad


15




a


is large enough to provide multiple wire for reducing the inductance. The area of the anode bonding pad


14




a


is about 60×70 μm, and the area of the cathode bonding pad


15




a


is about 180×70 μm.




The intersection between the anode electrode


11


and the n epitaxial layer


3


, which is at the cathode voltage, is about 100 μm


2


, which is about one thirty ninth of the intersection area of conventional device (3900 μm


2


). Because of this reduction, the polyimide layer of conventional device can be replaced by a much thinner nitride film


5


.




Due to this planar configuration without any mesa and polyimide layer, there is no need for accommodating dimensional variation arising from processing inaccuracy. Accordingly, the separation between the Schottky contact region


11




a


and the ohmic electrode


8


is reduced to about 2 μm, and the separation between the Schottky contact region


11




a


and the impurity-implanted region


7


is reduced to about 1 μm, in comparison to the conventional device, which has a separation of about 7 μm. Because the impurity-implanted region is similar to the ohmic electrode


8


in terms of carrier conduction, the device of this embodiment has a separation of about one seventh of the conventional device. This leads to an improvement of high frequency characteristics over the conventional device since a shorter separation provides a smaller resistance.




Furthermore, with the planar configuration, the chip size reduces from 0.27×0.31 mm


2


of the conventional device to 0.25×0.25 mm


2


of this embodiment. Notably, the size of the operation region, or the Schottky contact region


11




a


, is smaller than that of the conventional device by about one tenth.




The insulting region


6


is not completely electrically insulating. Thus, it may be possible that a depletion layer is formed underneath the anode bonding pad


11




b


, which is made of the second wiring layer, and that the depletion layer reaches the ohmic electrode


8


depending on the high frequency signals the anode bonding pad


11




b


receives. If this happens, significant high frequency signal leakage results. However, in this embodiment, the depletion layer is prevented from reaching the ohmic electrode


8


by the impurity-implanted region


7


and n+ epitaxial layer


2


, both of which are capable of absorbing the depletion layer. Accordingly, the anode-cathode interference is eliminated and the two electrodes are completely separated in terms of high frequency signals.




Furthermore, the anode electrode


11


is formed by evaporating a Ti film, a Pt film and a Au film in this order. Such a thin anode electrode


11


is applicable in this embodiment because the thick polyimide layer of the conventional diode device is replaced by the thin nitride film


5


. There is no need for forming a thick Au electrode for stress relief





FIG. 5

is a cross-sectional view of a Schottky barrier diode device of a first modification of the first embodiment of this invention. The configuration of this modified embodiment is the same as the configuration of the first embodiment except two features. First, the device of the first modification has a stabilized layer


4


on top of the n epitaxial layer


3


. The stabilized layer


4


is made of an undoped InGaP and has a thickness of about 20 nm. The stabilized layer


4


protects the surface of the n epitaxial layer


3


from external contamination so that a Schottky contact with excellent high frequency characteristics is formed. For this reason, the portion of the stabilized layer


4


corresponding to the Schottky contact region


11




a


is removed immediately before the formation of the Schottky contact. Second, in this modification, a Schottky electrode


10


is first formed on the n epitaxial layer


3


as a separate element different from the anode electrode


11


. The Schottky contact region


11




a


is formed underneath the Schottky electrode


10


. The anode electrode


11


simply provides wiring between the Schottky electrode


10


and the anode bonding pad


11




b


. The Schottky electrode


10


is also made of the Ti/Pt/Au metal layer. The impurity-implanted region


7


and the insulating region


6


are formed by injecting corresponding impurities into the corresponding regions through the stabilized layer


4


.





FIG. 6

is a cross-sectional view of a Schottky barrier diode device of a second modification of the first embodiment of this invention. The configuration of the second modification is the same as the configuration of the first embodiment except that the insulating region


6


is disposed around the anode bonding pad


11




b


rather than underneath the anode bonding pad


11




b


. In this modification, the insulating region


6


surrounds portions of the n epitaxial layer


3


and the n+ epitaxial layer


2


under the anode bonding pad


11




b


so that the surrounded portions are electrically insulated from other portions of the epitaxial layers


2


,


3


. As in the case of the first embodiment, the insulating region


6


reaches the insulating substrate


1


. Although high frequency signals applied to the anode bonding pad


11




b


carry to the epitaxial layers


2


,


3


underneath the anode bonding pad


11




b


, the signals cannot advance beyond the insulating region


6


.





FIGS. 7A-7E

show process steps of a manufacturing method of the Schottky barrier diode of the first embodiment. First, a device intermediate shown in

FIG. 7A

is prepared. An n+ epitaxial layer


2


is formed on an undoped GaAs substrate


1


by a MOCVD process. The n+ epitaxial layer


2


has a thickness of about 500 nm and a silicon impurity concentration of about 5×10


18


cm


−3


. An n epitaxial layer


3


having a thickness of about 250 nm and a silicon impurity concentration of about 1.3×10


11


cm


−3


is formed on the n+ epitaxial layer


2


.




In the next step shown in

FIG. 7B

, a nitride film


5


is, first, formed on the entire top surface of the device intermediate. A resist layer is, then, formed on the nitride film


5


and is patterned to open a window for impurity injection to an insulating region


6


by a photolithographic process. Boron ions are injected through the window created in the resist film to form the insulating region


6


having a boron concentration of about 3×10


17


cm


−3


, which reaches the undoped GaAs substrate


1


. For creating the insulating region


6


of the second modification of the first embodiment, which surrounds the anode bonding pad


11




b


, the resist layer is patterned accordingly.




Then, another resist layer is formed after removal of the first resist layer, and is patterned to open a window for impurity injection to an impurity-implanted region


7


by the photolithographic process. Silicon impurities are injected through the window to form the impurity-implanted region


7


having a silicon concentration of about 1.0×10


18


cm


−3


. The impurity-implanted region


7


penetrates the n epitaxial layer


3


and reaches the n+ epitaxial layer


2


. It is better to inject the silicon impurities in a manner to assure an equal distribution of the impurities along the depth of the impurity-implanted region


7


. One such method is to inject a predetermined dose of the impurities in a plurality of separate injection steps, each of which may have a different injection condition. After removal of the second resist layer, another nitride film


5


is formed on the device intermediate for anneal protection, and the device intermediate is annealed to activate the impurity-implanted region


7


and the insulating region


6


.




The process step following the step of

FIG. 7B

is shown in

FIG. 7C. A

new resist layer is formed on the entire top surface of the device intermediate, and is patterned to open a window at a portion corresponding to a ohmic electrode formation by the photolithographic process. The nitride film


5


exposed in the window is removed to expose the top surface of the impurity-implanted region


7


. A AuGe film, a Ni film and a Au film are vacuum evaporated on the exposed impurity-implanted region


7


in this order. After a removal of the resist film with the lift-off process, the device intermediate undergoes a heat treatment so that the deposited metal films turn into an ohmic electrode


8


making a ohmic contact with the impurity-implanted region


7


.




An additional nitride film


5


having a thickness of about 500 nm is formed on the entire top surface of the device intermediate. A resist layer is, then, formed on the nitride film


5


, and is patterned to open windows at portions corresponding to a Schottky contact region


11




a


, an anode bonding pad


11




b


, a cathode electrode


15


and the cathode bonding pad


15




b


. The nitride film


5


exposed in the windows is removed by a dry etching process using the resist layer as a mask, which is also removed after the etching process. After this step, a Schottky electrode


10


may be formed after the second formation of the nitride film


5


.




In the next step shown in

FIG. 7D

, a resist layer is formed on the nitride film


5


, and is patterned to open windows at portions corresponding to an anode electrode


14


as well as the Schottky contact region


11




a


, the cathode electrode


15


, the anode bonding pad


11




b


and the cathode bonding pad


15




b


. Using the resist layer as a mask, a metal wiring pattern is formed on the top surface of the device intermediate. This second wiring layer is formed by successively vacuum depositing a Ti film, a Pt film and a Au film in this order. Accordingly, the Ti/Pt/Au metal layer directly contacts the n epitaxial layer


3


to form the Schottky contact region


11




a


, and also directly contacts the ohmic electrode


8


on its cathode side. The anode bonding pad


11




b


, which is formed directly on the insulating layer


3


, and the anode electrode


11


, which provides wiring between the Schottky contact region


11




a


and the anode bonding pad


11




b


, are a part of the second wiring layer. The cathode electrode


15


, which is formed directly on the impurity-implanted region


7


, and the cathode bonding pad


15




b


are also formed as a part of the second wiring layer. The portion of the ohmic electrode


8


under the anode electrode


11


is insulated from the Ti/Pt/Au metal layer by the nitride film


5


remaining on the top surface of the n epitaxial layer


3


. The resist layer is removed by the lift-off process and a backside rapping is performed on the device intermediate.




Because of this planar configuration of this embodiment, there is no need for forming a polyimide layer and a plated layer on top of the polyimide layer, as in the case of a manufacturing method of the conventional device of FIG.


1


. Accordingly, the manufacturing method of the device of the first embodiment is much simpler and easier to control than the conventional method.




In the next step shown in

FIG. 7E

, the Schottky barrier diode is assembled as individual devices. First, the compound semiconductor substrate


1


having individual diode elements thereon is diced and separated into individual chips. The individual chips are, then, mounted on frames (not shown in the figure) and undergo a wire bonding process to connect the bonding pads


11




b


,


15




b


and corresponding leads of the frame. In this embodiment, gold wires are bonded to the bonding pads


11




b


,


15




b


using a well known stitch bonding process. Finally, the individual chips are molded by a transfer molding process.




A stitch bonding process fixes two bonding wires in one bonding procedure and requires a smaller bonding area than a ball bonding process. In the manufacturing method of the conventional Schottky barrier diode of

FIG. 1

, however, use of a stitch bonding process results in large amount of defects including pealing and crack formation of the anode electrode during the bonding process due to the presence of the polyimide layer. The planar configuration without the polyimide layer of the diode device of this embodiment significantly reduces such defects.





FIG. 8

is a cross-sectional view of a Schottky barrier diode of a second embodiment of this invention.

FIG. 9

is a top view of the diode device of the second embodiment. This diode device has the same configuration as the diode device of the second modification of the first embodiment shown in

FIG. 6

except that an additional impurity-implanted region


16


is formed underneath the anode bonding pad


11




b.






The additional impurity-implanted region


16


underneath the anode bonding pad


11




b


is an n+ impurity region (


29


Si


+


, 1×10


18


cm


−3


), which is formed simultaneously as the impurity-implanted region


7


underneath the ohmic electrode


8


. The size of the additional impurity-implanted region


16


is larger than the size of the anode bonding pad


11




b


, as shown in FIG.


9


. The insulating region


6


, which is formed by injecting boron impurities, surrounds the additional impurity-implanted region


16


, as shown in FIG.


9


. As in the case of the first embodiment, the insulating region


6


reaches the insulating substrate


1


.




In the configuration of

FIG. 6

, a depletion layer may be formed in the insulating region


6


, and extend toward the ohmic electrode


8


when high frequency signals are applied to the anode bonding pad


11




b


. Depending on the high frequency signals the anode bonding pad


11




b


receives, the depletion layer may eventually reach the ohmic electrode


8


, resulting in significant signal leakage. In the configuration of this embodiment, however, the additional impurity-implanted region


16


, which covers a slightly larger area than the anode bonding pad


11




a


, prevents the extension of the depletion layer. Although the ohmic electrode


8


is electrically insulated from the anode bonding bad


11




b


by the insulating region


6


, the presence of the additional impurity-implanted region


16


further reduces high frequency signal leakage.





FIG. 10

is a cross-sectional view of a Schottky barrier diode of a first modification of the second embodiment of this invention. This diode device has the same configuration as the diode device of the second embodiment except that the additional impurity-implanted region


16


is formed only in the peripheral portion of the anode bonding pad


11




b


. As long as the edges of the anode bonding pad


11




b


are covered by the additional impurity-implanted region


16


, the portions of the epitaxial layers


2


,


3


under the anode bonding pad


11




b


is effectively insulated from other portions of the epitaxial layers


2


,


3


. For this configuration to be effective, the additional impurity-implanted region


16


should extend outside the anode bonding pad


11




b.







FIGS. 11A-11E

show process steps of a manufacturing method of the Schottky barrier diode of the second embodiment. A device intermediate shown in

FIG. 11A

is prepared in the same manner as the first step of the manufacturing method of the first embodiment shown in FIG.


7


A. First, an n+ epitaxial layer


2


is formed on an undoped GaAs substrate


1


by a MOCVD process. The n+ epitaxial layer


2


has a thickness of about 500 nm and a silicon impurity concentration of about 5×10


18


cm


−3


. An n epitaxial layer


3


having a thickness of about 250 nm and a silicon impurity concentration of about 1.3×10


17


cm


−3


is formed on the n+ epitaxial layer


2


.




In the next step shown in

FIG. 11B

, a nitride film


5


is, first, formed on the entire top surface of the device intermediate. A resist layer is, then, formed on the nitride film


5


and is patterned to open a window for impurity injection to an insulating region


6


by the photolithographic process. Boron ions are injected through the window created in the resist film to form the insulating region


6


having a boron concentration of about 3×10


17


cm


−3


, which reaches the undoped GaAs substrate


1


. In this embodiment, the insulating region


6


is disposed around the additional impurity-implanted region


16


.




Then, another resist layer is formed after removal of the first resist layer, and is patterned to open windows for impurity injection to the impurity-implanted region


7


and the additional impurity-implanted region


16


by the photolithographic process. Silicon impurities are injected through the window to form the impurity-implanted regions


7


,


16


having a silicon concentration of about 1.0×10


18


cm


−3


. The impurity-implanted regions


7


,


16


penetrate the n epitaxial layer


3


and reaches the n+ epitaxial layer


2


. It is better to inject the silicon impurities in a manner to assure an equal distribution of the impurities along the depth of the impurity-implanted regions


7


,


16


. One such method is to inject a predetermined dose of the impurities in a plurality of separate injection steps, each of which may have a different injection condition. After removal of the second resist layer, another nitride film


5


is formed on the device intermediate for anneal protection, and the device intermediate is annealed to activate the impurity-implanted regions


7


,


16


and the insulating region


6


. In this embodiment, the additional impurity-implanted region


16


is formed so that the anode bonding pad


11




b


is within the additional impurity-implanted region


16


when the pad is formed. As a modification, the resist layer may be patterned to form the additional impurity-implanted region


16


covering only the peripheral portion of the anode bonding pas


11




b


. Either way, the formation of the additional impurity-implanted region


16


does not require any additional processing step because it is formed simultaneously with the impurity-implanted region


7


.




The subsequent processing steps shown in

FIGS. 11C-11E

are essentially the same as the corresponding steps of the manufacturing method of the first embodiment shown in

FIGS. 7C-7E

, although the insulating region


6


occupies a location different from the first embodiment and the additional impurity-implanted region


16


occupies the location of the insulating region


6


of the first embodiment.




The above is a detailed description of a particular embodiments of the invention which are not intended to limit the invention to the embodiments described. It is recognized that modifications within the scope of the invention will occur to a person skilled in the art. Such modifications and equivalents of the invention are intended for inclusion within the scope of this invention.



Claims
  • 1. A Schottky barrier diode comprising:a substrate made of a compound semiconductor; an epitaxially grown layer disposed on the substrate; a Schottky contact region being a part of the epitaxially grown layer; an anode wiring connected to the Schottky contact region; an insulating region penetrating through the epitaxially grown layer and reaching the substrate; and an anode bonding pad disposed on the insulating region and connected the anode wiring.
  • 2. A Schottky barrier diode comprising:a substrate made of a compound semiconductor; an epitaxially grown layer disposed on the substrate; a Schottky contact region being a part of the epitaxially grown layer; an anode wiring connected to the Schottky contact region; an insulating region penetrating through the epitaxially grown layer and reaching the substrate; and an anode bonding pad disposed on the epitaxially grown layer and connected to the anode wiring, the insulating region surrounding the anode bonding pad.
  • 3. The Schottky barrier diode of claims 1 or 2, further comprising an ohmic electrode disposed adjacent the Schottky contact region and an insulating film insulating the ohmic electrode from the anode wiring.
  • 4. The Schottky barrier diode of claims 1 or 2, wherein the substrate comprises an undoped GaAs substrate.
  • 5. The Schottky barrier diode of claims 1 or 2, further comprising a bonding wire connected to the anode bonding pad.
  • 6. The Schottky barrier diode of claim 1, wherein the insulating region is larger than the anode bonding pad with respect to a plane parallel to a primary plane of the substrate and covers the entire anode bonding pad.
  • 7. The Schottky barrier diode of claims 1 or 2, further comprising an undoped InGaP layer disposed on the epitaxially grown layer.
  • 8. The Schottky barrier diode of claims 1 or 2, wherein the anode bonding pad comprises a metal film formed by evaporation.
  • 9. The Schottky barrier diode of claim 3, further comprising an impurity-implanted region disposed underneath the ohmic electrode.
  • 10. A Schottky barrier diode comprising:a substrate made of a compound semiconductor; an epitaxially grown layer of a conduction type disposed on the substrate and including a Schottky contact region; an anode wiring connected to the Schottky contact region; an insulating region penetrating through the epitaxially grown layer and reaching the substrate; a first impurity-implanted region of the conduction type penetrating through the epitaxially grown layer; and an anode bonding pad disposed on the first impurity-implanted region and connected to the anode wiring, the insulating region surrounding the anode bonding pad.
  • 11. A Schottky barrier diode comprising:a substrate made of a compound semiconductor; an epitaxially grown layer of a conduction type disposed on the substrate and including a Schottky contact region; an anode wiring connected to the Schottky contact region; an anode bonding pad disposed on the epitaxially grown layer and connected to the anode wiring, a first impurity-implanted region of the conduction type penetrating through the epitaxially grown layer and disposed underneath a peripheral portion of the anode bonding pad; and an insulating region penetrating through the epitaxially grown layer to reach the substrate and surrounding the first impurity-implanted region.
  • 12. The Schottky barrier diode of claims 10 or 11, further comprising an ohmic electrode disposed adjacent the Schottky contact region and an insulating film insulating the ohmic electrode from the anode wiring.
  • 13. The Schottky barrier diode of claims 10 or 11, wherein the substrate comprises an undoped GaAs substrate.
  • 14. The Schottky barrier diode of claim 10, wherein the first impurity-implanted region is larger than the anode bonding pad with respect to a plane parallel to a primary plane of the substrate and covers the entire anode bonding pad.
  • 15. The Schottky barrier diode of claims 10 or 11, further comprising an undoped InGaP layer disposed on the epitaxially grown layer.
  • 16. The Schottky barrier diode of claims 10 or 11, wherein the anode bonding pad comprises a metal film formed by evaporation.
  • 17. The Schottky barrier diode of claims 10 or 11, further comprising a second impurity-implanted region disposed underneath the ohmic electrode.
Priority Claims (1)
Number Date Country Kind
2001-290756 Sep 2001 JP
US Referenced Citations (4)
Number Name Date Kind
4636833 Nishioka et al. Jan 1987 A
6380552 Schmitz et al. Apr 2002 B2
6525389 Ahmed Feb 2003 B1
20020109200 Bartsch et al. Aug 2002 A1
Non-Patent Literature Citations (2)
Entry
Yasoo Harada et al. “Conversion Loss of Novel Beam-Lead GaAs Schottky-Barrier Diodes: A Comparison between Calculation and Measurement” Research Center, Sanyo Electric Co., Ltd. Oct. 1983, pp. 55-61.
Yasoo Harada “Research on High Reliability Beam Lead GaAs Schottky Diode”.