US20050151159A1 - Solid-state high power device and method - Google Patents
Solid-state high power device and method Download PDFInfo
- Publication number
- US20050151159A1 US20050151159A1 US10/993,224 US99322404A US2005151159A1 US 20050151159 A1 US20050151159 A1 US 20050151159A1 US 99322404 A US99322404 A US 99322404A US 2005151159 A1 US2005151159 A1 US 2005151159A1
- Authority
- US
- United States
- Prior art keywords
- emitter
- row
- fingers
- finger
- subcell
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
- 238000000034 method Methods 0.000 title description 25
- 229910001218 Gallium arsenide Inorganic materials 0.000 claims description 35
- 229910052751 metal Inorganic materials 0.000 claims description 10
- 239000002184 metal Substances 0.000 claims description 10
- 238000002161 passivation Methods 0.000 claims description 5
- 238000004519 manufacturing process Methods 0.000 abstract description 18
- 230000000694 effects Effects 0.000 abstract description 8
- 230000008901 benefit Effects 0.000 abstract description 7
- 230000002411 adverse Effects 0.000 abstract description 6
- 238000012938 design process Methods 0.000 abstract 1
- 230000008569 process Effects 0.000 description 18
- 229910000577 Silicon-germanium Inorganic materials 0.000 description 14
- 238000013461 design Methods 0.000 description 14
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 12
- 238000013459 approach Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 7
- 239000000758 substrate Substances 0.000 description 6
- 238000009826 distribution Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000003071 parasitic effect Effects 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 238000000206 photolithography Methods 0.000 description 3
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 238000010420 art technique Methods 0.000 description 2
- 239000010953 base metal Substances 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 238000001459 lithography Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- LEVVHYCKPQWKOP-UHFFFAOYSA-N [Si].[Ge] Chemical compound [Si].[Ge] LEVVHYCKPQWKOP-UHFFFAOYSA-N 0.000 description 1
- 230000001668 ameliorated effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 229910052681 coesite Inorganic materials 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 229910052906 cristobalite Inorganic materials 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000005137 deposition process Methods 0.000 description 1
- 238000001312 dry etching Methods 0.000 description 1
- 230000003090 exacerbative effect Effects 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229920002120 photoresistant polymer Polymers 0.000 description 1
- 238000000623 plasma-assisted chemical vapour deposition Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000000377 silicon dioxide Substances 0.000 description 1
- 235000012239 silicon dioxide Nutrition 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 229910052682 stishovite Inorganic materials 0.000 description 1
- 229910052905 tridymite Inorganic materials 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000001039 wet etching Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/41—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions
- H01L29/423—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions not carrying the current to be rectified, amplified or switched
- H01L29/42304—Base electrodes for bipolar transistors
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/02—Semiconductor bodies ; Multistep manufacturing processes therefor
- H01L29/06—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions
- H01L29/0684—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions characterised by the shape, relative sizes or dispositions of the semiconductor regions or junctions between the regions
- H01L29/0692—Surface layout
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/40—Electrodes ; Multistep manufacturing processes therefor
- H01L29/41—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions
- H01L29/417—Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions carrying the current to be rectified, amplified or switched
- H01L29/41708—Emitter or collector electrodes for bipolar transistors
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/68—Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
- H01L29/70—Bipolar devices
- H01L29/72—Transistor-type devices, i.e. able to continuously respond to applied control signals
- H01L29/73—Bipolar junction transistors
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/68—Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
- H01L29/70—Bipolar devices
- H01L29/72—Transistor-type devices, i.e. able to continuously respond to applied control signals
- H01L29/73—Bipolar junction transistors
- H01L29/7302—Bipolar junction transistors structurally associated with other devices
- H01L29/7304—Bipolar junction transistors structurally associated with other devices the device being a resistive element, e.g. ballasting resistor
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/68—Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
- H01L29/70—Bipolar devices
- H01L29/72—Transistor-type devices, i.e. able to continuously respond to applied control signals
- H01L29/73—Bipolar junction transistors
- H01L29/737—Hetero-junction transistors
- H01L29/7371—Vertical transistors
- H01L29/7378—Vertical transistors comprising lattice mismatched active layers, e.g. SiGe strained layer transistors
Definitions
- the present invention relates generally to high power solid-state devices, and more particularly to high power solid-state device structures that are capable of providing high power performance and a corresponding method of design of such structures.
- Bipolar junction transistors including heterojunction bipolar transistors (HBTs), are commonly used in such systems for amplifying small-amplitude signals and delivering amplified RF power to an antenna.
- bipolar junction transistors BJTs are used herein to describe the background of this invention, this is by way of example and does not limit the scope of the invention disclosed herein.
- High power solid-state devices for amplifying RF and microwave signals can be fabricated using a variety of materials, but silicon germanium (“SiGe”) and gallium arsenide (“GaAs”) are the most widely used materials for commercial applications at the present time.
- SiGe silicon germanium
- GaAs gallium arsenide
- Most power amplifiers for cell phones are made using GaAs, because current fabrication technologies using that material can deliver devices with relatively high power output (1-4 W) at the relatively low frequencies (800 Mhz-1.9 Ghz) used by most cell phones.
- the effective range of a wireless communication system depends on the maximum RF power that can be produced by that wireless communication system.
- the maximum RF power that can be produced by a device depends on the active device area, with the power capacity increasing as the active device area increases.
- the bandwidth, or information transmission capacity, of a wireless communication system depends on the maximum frequency range that can be amplified effectively by that wireless communication system. As the power capacity and associated active device area of a device increases, the adverse effects of heat and increased parasitics also increase, so in practice it is generally the case that increasing the power capacity of a given device will decrease the maximum frequency range that can be amplified effectively by that device.
- PAE power added efficiency
- the power capacity of an active device increases as the active device area increases.
- the maximum RF power level that can be produced by a BJT depends on the emitter area of the BJT, with the maximum RF power level increasing as the emitter area increases. Because of problems such as the emitter current crowding effect, it is known to divide the total emitter area of a BJT into multiple emitter “fingers” separated from one another. It is also known that the specific arrangement of these multiple emitter fingers can affect many aspects of the performance of such a device.
- FIG. 1 shows an exemplary prior art BJT 20 having a base 22 , emitter 24 , collector 25 , and multiple emitter fingers 26 separated by a uniform distance X 28 .
- FIG. 1 is a compact layout that saves chip area, this type of layout is known to have poor thermal stability and performance due to severe thermal coupling between emitter fingers (they are too close) and excessive parasitic collector resistance.
- FIG. 2 shows another exemplary prior art BJT 30 also having multiple emitter fingers 26 .
- the multiple emitter fingers of the device of FIG. 2 are grouped into four subcells 32 , with each of the four subcells 32 having two emitter fingers 26 .
- the subcells 32 are separated from one another by collector regions, with a uniform distance Y 34 between the centers of adjacent subcells 32 .
- the device 30 of FIG. 2 although somewhat less compact than the device 20 of FIG. 1 , is known to provide reduced collector resistance and thermal effects in exchange for the increased device area.
- both devices have uniform spacing between the multiple emitter fingers 26 (in the device of FIG. 1 ) and between the multiple subcells 32 (in the device of FIG. 2 ). It is known that devices having uniformly spaced emitter fingers 26 , or uniformly spaced emitter finger subcells 32 , are typically subject to adverse thermal effects caused by a higher and localized device temperature rise in the center of these types of devices during operation, as further explained below.
- ballast resistors 38 typically have values in the range of 20-100 ohms, and provide a negative feedback mechanism between temperature and current of the emitter fingers or subcells.
- ballast resistors 38 can provide thermal stability and improve temperature uniformity across the multiple emitter fingers 26 , or emitter finger subcells 32 , the use of ballast resistors 38 can adversely affect important measures of device performance.
- using ballast resistors will tend to reduce the maximum RF power output from the device. This is because the emitter fingers and subcells in series with the ballast resistors will be underbiased compared to a device without ballast resistors, since the available bias voltage must be shared between the ballast resistors and the remainder of the device.
- the ballasting resistors increase the RC delay and thereby adversely affect the high frequency performance of the device.
- the voltage drop across the ballast resistors ends up as heat instead of as RF output power, thereby wasting power and reducing the efficiency (“PAE”) of converting DC supply power into RF signal power for the power transistors.
- PAE efficiency
- Another prior art approach to improve the thermal stability and maintain a uniform junction across the multiple emitter fingers of a power transistor is to make the spacing between the emitter fingers non-uniform.
- This approach is discussed, for example, in U.S. Pat. No. 6,534,857.
- the emitter finger spacing is non-uniform, with more spacing between the emitter fingers in the center region and less spacing between the edge fingers.
- the spacing is arranged with the goal of providing a uniform junction temperature across the emitter fingers.
- Similar benefits can be obtained by using progressively narrower widths of emitter fingers from the periphery toward the center region of the power transistor, or by using progressively shorter emitter finger lengths from the periphery toward the center region of the power transistor, for example as shown in U.S. Pat Nos. 5,616,950 and 5,850,099.
- the lithographic processes used to manufacture the emitter fingers always have statistical variations that cause the widths and locations of the emitter fingers to vary.
- This variation can be caused, for example, by variations in the optical column or mask used to print the emitter fingers, by variations in the photoresist or developer used to image the emitter fingers, or by variations in the etch or deposition processes used to prepare the emitter fingers.
- This variation can manifest itself, for example, in finger-to-finger variation within a transistor, in transistor-to-transistor variation within a batch, or in day-to-day variation between batches.
- emitter fingers (No. 4 -No. 16 ) in the 20-finger power transistor are all center fingers.
- the available space in the center area must be shared among all the center fingers, thus the central area of a 20 finger device becomes crowded because the edge area in a 20-finger device remains the same as in the 10-finger device.
- the difference in the theoretically optimum spacing between, for example, emitter fingers 4 and 5 and the spacing between emitter fingers 5 and 6 in a 20-finger device can be very small.
- the above-described practical limitations in manufacturing and design introduce variation in the dimensions and locations of emitter fingers that can result in significant temperature non-uniformity once the device's operation reaches its steady state.
- FIG. 7 ( a ) shows an exemplary prior art 23-emitter finger GaAs HBT, with nominal 2 ⁇ m finger width and nominal 20 ⁇ m finger length.
- the emitter fingers of the device of FIG. 7 ( a ) are arranged in a non-uniform fashion to produce theoretically uniform junction temperatures, assuming that the emitter fingers are manufactured to have perfect locations and dimensions.
- FIGS. 7 ( b ) and 7 ( c ) present calculated steady-state temperature and current distributions of the device of FIG. 7 ( a ) under the assumption that process variation has increased the width of the No. 12 emitter finger 42 to 2.02 ⁇ m. That is, the No. 12 emitter finger 42 is assumed to be 1% wider than the nominal 2 ⁇ width of the other emitter fingers in the device of FIG. 7 ( a ).
- FIG. 7 ( b ) shows the calculated temperature profile across the emitter fingers resulting from this 1% variation in width of the no. 12 emitter finger 42 .
- FIG. 7 ( c ) shows the calculated current profile across the emitter fingers resulting from this 1% variation in width of the no. 12 emitter finger 42 during steady-state high power operation.
- a preferred embodiment of a high power transistor according to the invention includes a plurality of emitter fingers fabricated on a common semiconductor chip and wherein the emitter electrodes of bipolar junction transistors (or heterojunction bipolar transistors) are in the form of parallel elongated finger elements having a uniform or non-uniform spacing. These fingers are arranged in a 1-dimensional (1-D) or 2-dimensional (2-D) form such that the potential thermal instability at high power operation may be reduced ballasting resistors with substantially reduced values, in the range of less than 10 ohms, which provides power, efficiency, and high frequency performance benefits in comparison to prior-art ballast resistor approaches which used larger resistor values typically in the range of 100 ohms.
- These fingers can also be arranged in a 1-dimensional (1-D) or 2-dimensional (2-D) form such that the potential thermal instability at high power operation is completely eliminated by forming a “hollow-center” layout where one or more elongated emitter fingers or subcells are left out during design or disconnected during manufacture, with or without the use of small ballasting resistors.
- a high power transistor according to the invention includes a base that comprises a GaAs layer less than about 100 microns thick, and emitter fingers that are less than about 1 micron wide.
- FIG. 1 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of uniformly-spaced emitter fingers;
- FIG. 2 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of emitter fingers arranged in uniformly spaced subcells each having 2 emitter fingers;
- FIG. 3 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of uniformly-spaced emitter fingers each with a ballast resistor;
- FIG. 4 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of emitter fingers arranged in uniformly-spaced subcells each having 2 emitter fingers each with a ballast resistor;
- FIG. 5 is a top planar view of the layouts of four exemplary prior-art power bipolar junction transistors having 5, 10, 20, and 30 emitter fingers arranged non-uniformly;
- FIG. 6 presents calculated temperature profiles for the four exemplary power bipolar junction transistor layouts of FIG. 5 ;
- FIG. 7 ( a ) is a top planar view of the layout of a prior-art power bipolar junction transistor having 23 emitter fingers arranged non-uniformly;
- FIG. 7 ( b ) shows the calculated temperature profile during steady-state high power operation of the device of FIG. 7 ( a ) assuming that emitter finger number 12 is 1% wider than intended;
- FIG. 7 ( c ) shows the calculated current profile during steady-state high power operation of the device of FIG. 7 ( a ) assuming that emitter finger number 12 is 1% wider than intended;
- FIGS. 8 ( a )- 8 ( f ) show an exemplary fabrication process for fabricating a SiGe (a similar process can be used for GaAs) bipolar junction power transistor having two emitter fingers;
- FIG. 9 is a top planar view of a power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor;
- FIG. 10 is a top planar view of another power bipolar junction transistor according to the invention having emitter fingers arranged in non-uniformly spaced subcells each having 2 emitter fingers each with a small ballast resistor;
- FIG. 11 ( a ) is a top planar view of another power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor and arranged in two arc-shaped rows;
- FIG. 11 ( b ) shows the calculated temperature profile along one row of emitter fingers in the device of FIG. 11 ( a );
- FIG. 12 ( a ) is a top planar view of another power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor and arranged in two arc-shaped rows surrounding a third straight center row of fingers;
- FIG. 12 ( b ) is the calculated temperature profile along the center row of emitter fingers in the device of FIG. 12 ( a );
- FIG. 13 is a top planar view of another power transistor structure according to the invention with emitter fingers surrounding a “hollow center” central gap where an emitter finger is missing;
- FIG. 15 is a top planar view of another power transistor structure according to the invention with non-uniformly spaced subcells, each with two emitter fingers, surrounding a central subcell having a single emitter finger;
- FIG. 16 ( a ) is a top planar view of an exemplary prior-art bipolar power transistor layout
- FIG. 16 ( b ) is a top planar view of a power transistor in accordance with the invention having the same number of emitter fingers and subcells, and the same chip area, as the device of FIG. 16 ( a );
- FIG. 16 ( c ) shows the measured power performance data of the device of FIG. 16 ( b ) compared to the prior art device of FIG. 16 ( a ).
- bipolar junction transistors and heterojunction bipolar transistors these multiple fingers are emitter fingers.
- FIGS. 8 ( a )- 8 ( f ) show an exemplary prior art fabrication process for fabricating a SiGe bipolar junction power transistor having two emitter fingers.
- the prior art process of FIGS. 8 ( a )- 8 ( f ) can be used for fabricating X-band SiGe HBT devices.
- This prior art process is highly repeatable and reliable, and typically employs 7 mask levels and 20 steps.
- This prior art process is used herein as an example of a process for implementing SiGe or GaAs power transistors using the specific dimensions and device layouts of the present invention, but it is by no means the only such process that could be used.
- FIGS. 8 ( a )- 8 ( f ) begins with a material stack formed on an Si substrate 62 , followed by a heavily N-type doped Si subcollector layer 60 , a lightly n-type doped Si collector layer 58 , a p-type doped SiGe base layer 56 , and an n-type doped Si emitter layer 53 .
- Emitter metal contacts 52 are formed using standard photolithography and liftoff techniques.
- the emitter metal 52 for example Cr or Au, is first evaporated on top of the highly doped emitter cap layer 53 of the Si/SiGe/Si heterostructure.
- the multi-emitter metal fingers 52 typically 2 microns wide, are formed with the first-level mask, leaving the structure shown in FIG. 8 ( a ). The width of the emitter fingers may be subject to statistical variations as discussed above.
- the patterned emitter metal fingers 52 then serve as self-aligned mask for subsequent dry/wet etching of the Si emitter layer 53 to expose the boron-doped SiGe base layer 56 , leaving the structure shown in FIG. 8 ( b ).
- Another photolithography step is used to form self-aligned base metal 55 on top of the exposed SiGe base layer 56 , leaving the structure shown in FIG. 8 ( c ).
- the third mask is used to form the base mesa by RIE and to expose the highly doped subcollector layer 60 for collector contact formation, leaving the structure shown in FIG. 8 ( d ).
- collector metal 64 is deposited and formed in another lithography step (4th mask).
- the active devices are isolated by removing the subcollector material 60 around the devices and exposing the high-resistivity Si substrate 62 (5th mask), leaving the structure shown in FIG. 8 ( e ).
- a conformal PECVD SiO2 deposition is used to form a passivation layer 66 over the active device 50 and the exposed substrate 62 .
- Contact via holes 68 are opened in the passivation layer 66 by RIE (6th mask) and interconnect pad metal 67 is deposited and then patterned with photolithography (7th mask) to finish the fabrication process and form the completed device 50 shown in FIG. 8 ( f ).
- FIG. 9 is a top planar view of a power bipolar junction transistor 70 according to the invention having non-uniformly spaced emitter fingers 26 with small ballast resistors 71 .
- a non-uniform spacing of emitter fingers 26 is arranged using the finite element analysis software, based on the known heating power density. The distance between adjacent emitter fingers 26 increases from the center of the device 70 to the periphery, so that distance X 1 >X 2 >X 3 .
- the arrangement of the multiple emitter fingers 26 must adhere to the applicable design rules of the technology used to build the device. Since any design rule will have some minimum allowable increment of distance, even if the perfect placement and spacing of the multiple emitter fingers can be calculated, this perfect placement and spacing cannot be implemented in practice, so at least some non-uniformity of junction temperature should be expected.
- the emitter fingers 26 of this power transistor are each connected in series with a small ballast resistor 71 , although this is not required and less than all of the emitter fingers may include the small ballast resistor 71 .
- small ballast resistors 71 can be omitted from the emitter fingers 26 on the periphery while maintaining small ballast resistors 71 on at least some of the central emitter fingers 26 to maintain temperature stability.
- the small ballast resistor is preferably in the range 1-10 ohms.
- such a small valued ballast resistor can be formed simply by shrinking the size of a contact hole 68 to the emitter 24 , for example to the emitter metal 52 , to create the desired resistance in series with the emitter finger 26 , with no additional structure or processing step required.
- the values of the resistors in accordance to the preferred embodiment are calculated based on the known maximum statistical variation of the finger width with the goal of a thermally stable operation condition. The calculation may involve iterations using measured or calculated thermal-electric coefficients and heating power density.
- small ballast resistors 71 are connected to emitter fingers in the device 70 of FIG. 9
- similar layouts with small ballast resistors 71 connected in series with the base 22 can also be used, instead of or in addition to small ballast resistors 71 in series with the emitter 24 .
- such a small valued ballast resistor can be formed simply by shrinking the size of a contact hole to the base 22 , for example to the base metal 55 , to create the desired resistance in series with the base 22 , with no additional structure or processing step required.
- the small ballast resistors 71 in the device 70 of FIG. 9 have values in the range 1-10 ohms, preferably around 1-3 ohms, and these small ballast resistors 71 are used to prevent thermal instability caused by small variations of finger spacing and finger width.
- the small ballast resistors 71 of this approach can have significantly improved power, efficiency, and high frequency performance compared to prior approaches that use larger ballast resistors 38 .
- FIG. 10 is a top planar view of another power bipolar junction transistor 74 according to the invention having emitter fingers 26 arranged in non-uniformly spaced subcells each having 2 emitter fingers with small ballast resistors. Although two fingers are grouped a subcell in this example, more fingers can be grouped together if finger width is small.
- the non-uniform layout is arranged by adhering to the design rule. The distance between adjacent subcells 32 increases from the center of the device 74 to the periphery, so that distance Y 1 >Y 2 .
- the values of the small ballast resistors can be determined by assuming the width of all the emitter fingers in the center subcell or subcells have the maximum possible finger width and assuming the distance between the central adjacent subcells is the minimum, given the expected variation in finger width and subcell spacing in the manufacturing process used.
- the advantages of this small ballast resistor approach are reduced base-collector junction capacitance and parasitic collector resistance in comparison to the prior art which uses relatively large ballast resistors.
- FIG. 11 ( a ) is a top planar view of another power bipolar junction transistor 78 according to the invention having non-uniformly spaced emitter fingers 26 each with a small ballast resistor 71 and arranged in two arc-shaped rows.
- the dimensions and arrangement of the 2-dimensional thermally balanced structure is designed to avoid the need for larger valued ballast resistors 38 to ensure thermal stability of the device 78 arising from the statistical variation of finger width and allowable minimal increment of emitter finger spacing in the one-row non-uniform layout.
- the side-to-side distance between adjacent emitter fingers 26 increases from the center of the device 78 to the periphery, so that distance X 1 >X 2 >X 3 .
- FIG. 11 ( b ) shows the calculated temperature profile along one row of emitter fingers of the device 78 of FIG. 11 ( a ).
- the temperature non-uniformity shown in FIG. 11 ( b ) due to the limitation of minimal increment of finger spacing is expected.
- the values of small ballast resistors 71 can be selected in a similar fashion to the devices of FIGS. 9 and 10 , by assuming that the central emitter fingers 26 have the maximum finger width and minimum spacing expected in the manufacturing process used to fabricate the device 78 , to ensure worst case thermal stability.
- FIG. 11 ( a ) includes a single finger structure for illustration, subcell structures with two or more emitter fingers grouped together can also be used. Again, small ballast resistors 71 can be placed in series with the base 22 , instead of or in addition to small ballast resistors 71 in series with the emitter 24 .
- FIG. 12 ( a ) is a top planar view of another power bipolar junction transistor 80 according to the invention having non-uniformly spaced emitter fingers 26 each having a small ballast resistor 71 and arranged in two arc-shaped rows surrounding a third straight center row of fingers.
- the side-to-side distance between adjacent emitter fingers 26 increases from the center of the device 80 to the periphery, so that distance X 1 >X 2 >X 3 .
- the front to back distance between the rows of adjacent emitter fingers 26 also increases from the center of the device 80 to the periphery, so that distance Z 1 >Z 2 .
- a three-row layout is shown in FIG. 12 ( a ) as an example, layouts of more than three rows can also be used.
- all the emitter fingers 26 in device 80 include small ballast resistors 71 , this is not necessary and in an appropriate case fewer than all the emitter fingers 26 may include a small ballast resistor 71 , and the values of the small ballast resistors 71 need not all be the same.
- FIG. 12 ( b ) is the calculated temperature profile along the center row of emitter fingers 26 in the device 80 of FIG. 12 ( a ). Similar to the device of FIG. 11 ( a ), temperature nonuniformity is expected when a design rule is adhered to, because of the imprecision in emitter finger placement resulting from the finite resolution of design rules.
- Subcell structures 32 can replace the single emitter finger structures 26 and small ballast resistors 71 instead of or in addition to small emitter ballast resistors 71 can also be used in a device 80 according to the invention similar to that of FIG. 12 ( a ). In addition, all the multiple rows can be straightened up and the center row or rows can have smaller number of emitter fingers or subcells than the outer rows. The values of the small ballast resistors 71 can be selected in a similar fashion to the device of FIG. 9 .
- FIG. 13 is a top planar view of another power transistor structure 82 according to the invention with non-uniformly spaced emitter fingers surrounding a central gap (“hollow center”) 84 .
- the side-to-side distance between adjacent emitter fingers 26 increases from the center of the device to the periphery, so that distance X 1 >X 2 >X 3 .
- the side-to-side distance X 1 is much greater than the side-to-side distance between other pairs of adjacent emitter fingers, whereby X 1 is more than twice as large as the distance between any other two pairs of adjacent emitter fingers.
- the multiple emitter fingers may be arranged either in a uniform or non-uniform fashion while adhering to the design rule, and still achieve thermally stable operation.
- the hollow center 84 in the device of FIG. 13 can result in a non-uniform temperature where the center is cooler than the periphery, to overcompensate and make the device 82 even more thermally stable.
- the “hollow center” 84 can be applied to layouts of multiple rows and can also be applied to a single row at locations instead of or in addition to center locations when the number of emitter fingers is sufficiently large. Although a non-uniform arrangement of emitter fingers is preferred, it is not required for this embodiment, and the “hollow center” 84 can be used even when all the other emitter fingers are uniformly spaced. Accordingly, both uniform and non-uniform arrangements of emitter fingers using the “hollow center” 84 can be used in a device according to the invention.
- the device 82 of FIG. 13 may also include one or more small valued ballast resistors 71 connected to one or more of the emitter fingers or the base, although this is not required.
- the side-to-side distance between adjacent subcells 32 increases from the center of the device to the periphery, so that distance Y 1 >Y 2 .
- the side-to-side distance Y 1 is much greater than the side-to-side distance between other pairs of adjacent subcells, whereby Y 1 is more than twice as large as the distance between any other two pairs of adjacent subcells.
- the hollow center 88 can be formed by electrically disconnecting or removing selected subcells during manufacture or the hollow center 88 subcells can be simply left out during layout of the device.
- the device 86 of FIG. 14 may include one or more small valued ballast resistors 71 connected to one or more of the emitter fingers and/or the base, although this is not required.
- FIG. 15 is a top planar view of another power transistor structure 90 according to the invention with subcells 94 , each with two emitter fingers, surrounding a central subcell 92 having a single emitter finger.
- the side-to-side distance between adjacent subcells increases from the center of the device to the periphery, so that distance Y 1 >Y 2 .
- the central subcells has a reduced number of emitter fingers compared to the other subcells, instead of a “hollow center” subcell 88 used in the device 86 of FIG. 14 .
- the device 90 of FIG. 15 includes one single-finger central subcell 92 surrounded by other subcells 94 each containing 2 emitter fingers, the surrounding subcells 94 could have a greater number of emitter fingers. Similarly, if the surrounding subcells have a greater number of emitter fingers, the central subcell could also have a greater number of emitter fingers, as long as the number of emitter fingers in the central subcell 92 is less than the number of emitter fingers in the surrounding subcells 94 .
- the device 90 of FIG. 15 may also include one or more small ballast resistors 71 connected to one or more of the emitter fingers and/or the base.
- FIG. 16 ( a ) is a top planar view of an exemplary prior-art SiGe HBT power transistor 96 with a uniform layout.
- the 16 emitter fingers 26 are grouped in 8 subcells 32 and uniformly spaced along a single row.
- FIG. 16 ( b ) is a top planar view of a power transistor 98 in accordance with the invention having the same number of emitter fingers and subcells, and the same chip area, as the device 96 of FIG. 16 ( a ).
- the device 98 of FIG. 16 ( b ) includes 8 subcells arranged non-uniformly in a 2-dimensional form, in accordance with the invention, while keeping the same chip area and the same subcell structure as the device 96 of FIG. 16 ( a ).
- the side-to-side distance between adjacent subcells 32 increases from the center of the device 98 to the periphery, so that distance Y 1 >Y 2 .
- the front to back distance between the rows of adjacent subcells 32 also increases from the center of the device 98 to the periphery, so that distance Z 1 >Z 2 .
- the device 98 of FIG. 16 ( b ) may also include one or more small ballast resistors 71 connected to one or more of the emitter fingers and/or the base.
- FIG. 16 ( c ) shows the measured power performance data of the device of FIG. 16 ( b ) compared to the prior art device of FIG. 16 ( a ), when both HBTs are fabricated on the same chip and operated at the same bias and input signal levels.
- the results shown in FIG. 16 ( c ) demonstrate that output power, power gain and power added efficiency are simultaneously improved using the 2-dimensional layout of the device 98 of FIG. 16 ( b ) compared to the 1-dimensional prior-art device 96 of FIG. 16 ( a ).
- each individual emitter finger or subcell is preferably maintained, to ensure thermal stability of a composite structure that comprises all the emitter fingers or subcells together.
- each finger should be less than 2 microns wide, preferably around one micron wide.
- the substrate thickness should also be thinned to a certain thickness depending on the overall heat dissipation of the device and finger width, so that the substrate thickness should be less than 130 microns, and preferably around 100 microns.
- a GaAs device having an emitter finger width less than 2 microns and a substrate thickness less than 130 microns is particularly preferred.
Landscapes
- Engineering & Computer Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Ceramic Engineering (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Bipolar Transistors (AREA)
Abstract
A high-power solid-state transistor structure comprised of a plurality of emitter or gate fingers in a uniform or non-uniform manner to provide improved high power performance is disclosed. Preferably, each of the fingers is associated with a corresponding one of a plurality of sub-cells, the sub-cells being arranged in at least one row. The advantage of the invention is that the structure can be practically implemented and the absolute thermal stability can be maintained for very high power transistors with reduced adverse effects resulting from random variation in the manufacturing and design process.
Description
- This application is a continuation-in-part of U.S. patent application Ser. No. 10/718,757 that was filed Nov. 21, 2003, the disclosure of which is incorporated by reference. This application claims the benefit of provisional application No. 60/607,767 that was filed Sep. 7, 2004, the disclosure of which is incorporated by reference. This application claims the benefit of provisional application No. 60/607,762 that was filed Sep. 7, 2004, the disclosure of which is incorporated by reference.
- This invention was made with United States government support awarded by the following agencies: National Science Foundation. Electrical & Communications System Div., Award No. 0323717. The United States government has certain rights in this invention.
- The present invention relates generally to high power solid-state devices, and more particularly to high power solid-state device structures that are capable of providing high power performance and a corresponding method of design of such structures.
- High power solid-state devices able to amplify radio frequency (“RF”) and microwave signals are used today in a variety of applications, for example in cell phones and other wireless communication systems. Bipolar junction transistors, including heterojunction bipolar transistors (HBTs), are commonly used in such systems for amplifying small-amplitude signals and delivering amplified RF power to an antenna. Although bipolar junction transistors (BJTs) are used herein to describe the background of this invention, this is by way of example and does not limit the scope of the invention disclosed herein.
- High power solid-state devices for amplifying RF and microwave signals can be fabricated using a variety of materials, but silicon germanium (“SiGe”) and gallium arsenide (“GaAs”) are the most widely used materials for commercial applications at the present time. Most power amplifiers for cell phones are made using GaAs, because current fabrication technologies using that material can deliver devices with relatively high power output (1-4 W) at the relatively low frequencies (800 Mhz-1.9 Ghz) used by most cell phones. Most power amplifiers for wireless networking products use SiGe, because current fabrication technologies using that material can allow high level integration to reduce cost deliver devices that can operate at somewhat higher frequencies (2.4 Ghz-60 Ghz) but at the reduced power (10-200 mW typical) used by wireless networking products such as 802.11b (“WiFi”). Besides the electronic properties which differ between GaAs and SiGe devices, the two different materials also have different thermal conductivities which require somewhat different techniques for heat management.
- The effective range of a wireless communication system depends on the maximum RF power that can be produced by that wireless communication system. The maximum RF power that can be produced by a device depends on the active device area, with the power capacity increasing as the active device area increases.
- The bandwidth, or information transmission capacity, of a wireless communication system depends on the maximum frequency range that can be amplified effectively by that wireless communication system. As the power capacity and associated active device area of a device increases, the adverse effects of heat and increased parasitics also increase, so in practice it is generally the case that increasing the power capacity of a given device will decrease the maximum frequency range that can be amplified effectively by that device.
- Especially for wireless communication systems, such as cell phones, that use batteries for electrical power, the so called power added efficiency (“PAE”) at which input battery power is converted to usable RF power (instead of being wasted, for example, as heat) determines how long such a device can be used before it must be recharged. For the reasons discussed above, device designs that maximize power added efficiency and RF power output while maintaining adequate high frequency performance are needed.
- As discussed above, the power capacity of an active device increases as the active device area increases. For example, the maximum RF power level that can be produced by a BJT depends on the emitter area of the BJT, with the maximum RF power level increasing as the emitter area increases. Because of problems such as the emitter current crowding effect, it is known to divide the total emitter area of a BJT into multiple emitter “fingers” separated from one another. It is also known that the specific arrangement of these multiple emitter fingers can affect many aspects of the performance of such a device.
-
FIG. 1 shows an exemplary prior art BJT 20 having abase 22,emitter 24,collector 25, andmultiple emitter fingers 26 separated by auniform distance X 28. AlthoughFIG. 1 is a compact layout that saves chip area, this type of layout is known to have poor thermal stability and performance due to severe thermal coupling between emitter fingers (they are too close) and excessive parasitic collector resistance. -
FIG. 2 shows another exemplary prior art BJT 30 also havingmultiple emitter fingers 26. Unlike the device ofFIG. 1 , the multiple emitter fingers of the device ofFIG. 2 are grouped into foursubcells 32, with each of the foursubcells 32 having twoemitter fingers 26. In the device ofFIG. 2 , thesubcells 32 are separated from one another by collector regions, with auniform distance Y 34 between the centers ofadjacent subcells 32. Thedevice 30 ofFIG. 2 , although somewhat less compact than thedevice 20 ofFIG. 1 , is known to provide reduced collector resistance and thermal effects in exchange for the increased device area. - Although the
device 20 ofFIG. 1 and thedevice 30 ofFIG. 2 have somewhat different performance characteristics, both devices have uniform spacing between the multiple emitter fingers 26 (in the device ofFIG. 1 ) and between the multiple subcells 32 (in the device ofFIG. 2 ). It is known that devices having uniformly spacedemitter fingers 26, or uniformly spacedemitter finger subcells 32, are typically subject to adverse thermal effects caused by a higher and localized device temperature rise in the center of these types of devices during operation, as further explained below. - When the transistors in the devices of
FIGS. 1 and 2 are initially biased, equal current passes through eachemitter finger 26, and this equal current produces an equal amount of heat in each finger. Over time, heat dissipates more slowly from the center fingers compared to the fingers on the periphery of the device. This is because the center fingers are surrounded by other emitter fingers that are also producing heat, unlike the peripheral fingers which are adjacent to cooler inactive regions where there are no emitter fingers producing heat. For this reason, devices like those ofFIGS. 1 and 2 tend to operate with a non-uniform temperature distribution wherein the center fingers are hotter than the edge fingers. - It is also known that the non-uniform temperature distribution common to these types of devices can make them unstable at high output levels. The higher temperature of the center fingers can cause the so-called “current hogging” effect, wherein the higher temperature center fingers draw more current than the peripheral fingers subject to the same bias voltage. Current hogging occurs when the increased temperature of the center fingers causes an increase in current through those center fingers. The increased current through the center figures increases the heat produced in those center fingers, which in turn exacerbates the temperature difference between the hotter center fingers and the cooler peripheral fingers. Even if this effect does not cause the device to go completely unstable, it can nonetheless severely degrade device performance, for example by decreasing high frequency gain. In the worst case, thermal runaway and catastrophic failure occurs.
- One prior art approach to improve the thermal stability and maintain a uniform junction temperature across the
multiple emitter fingers 26, ormultiple subcells 32, of a power transistor, is to connect equally- or unequally-valuedballast resistors 38 in series with eachemitter finger 26 as shown in thedevice 36 ofFIG. 3 and thedevice 40 ofFIG. 4 . Theseballast resistors 38 typically have values in the range of 20-100 ohms, and provide a negative feedback mechanism between temperature and current of the emitter fingers or subcells. When more current is drawn by thecenter emitter finger 26 orsubcell 32 due to the rising of temperature, the voltage drop across theballast resistors 38 increases. Hence, the voltage available to theemitter fingers 26 is reduced and less heat is thus generated by these fingers. - Although
ballast resistors 38 can provide thermal stability and improve temperature uniformity across themultiple emitter fingers 26, oremitter finger subcells 32, the use ofballast resistors 38 can adversely affect important measures of device performance. First, using ballast resistors will tend to reduce the maximum RF power output from the device. This is because the emitter fingers and subcells in series with the ballast resistors will be underbiased compared to a device without ballast resistors, since the available bias voltage must be shared between the ballast resistors and the remainder of the device. In addition, the ballasting resistors increase the RC delay and thereby adversely affect the high frequency performance of the device. Finally, the voltage drop across the ballast resistors ends up as heat instead of as RF output power, thereby wasting power and reducing the efficiency (“PAE”) of converting DC supply power into RF signal power for the power transistors. - Another prior art approach to improve the thermal stability and maintain a uniform junction across the multiple emitter fingers of a power transistor is to make the spacing between the emitter fingers non-uniform. This approach is discussed, for example, in U.S. Pat. No. 6,534,857. In this type of layout, the emitter finger spacing is non-uniform, with more spacing between the emitter fingers in the center region and less spacing between the edge fingers. The spacing is arranged with the goal of providing a uniform junction temperature across the emitter fingers. Similar benefits can be obtained by using progressively narrower widths of emitter fingers from the periphery toward the center region of the power transistor, or by using progressively shorter emitter finger lengths from the periphery toward the center region of the power transistor, for example as shown in U.S. Pat Nos. 5,616,950 and 5,850,099.
- The aforementioned techniques involving non-uniform dimensioning and placement of the emitter fingers theoretically might produce thermal stability and uniform junction temperature regardless of the total number of emitter fingers in such a device. However, there are important practical limitations to this technique, especially for very large power transistors.
- First, although it may be possible to calculate to a high degree of precision the dimensions and positions for emitter fingers that will optimize thermal stability and uniformity, it is much more difficult to actually manufacture emitter fingers in accordance with those calculated optimal dimensions and locations. The lithographic processes used to manufacture the emitter fingers always have statistical variations that cause the widths and locations of the emitter fingers to vary. This variation can be caused, for example, by variations in the optical column or mask used to print the emitter fingers, by variations in the photoresist or developer used to image the emitter fingers, or by variations in the etch or deposition processes used to prepare the emitter fingers. This variation can manifest itself, for example, in finger-to-finger variation within a transistor, in transistor-to-transistor variation within a batch, or in day-to-day variation between batches.
- Second, although it may be possible to calculate to a high degree of precision the dimensions and positions for emitter fingers that will optimize thermal stability and uniformity, practical chip layout tools do not provide for arbitrarily small increments of dimensions and positions in design rules. Thus, especially when the dimensions and spacings of the emitter fingers are of the same order of magnitude as the minimum feature size available in the process being used to manufacture the transistors and/or the minimum dimension and location increments of the design rules of the software used to layout the transistor, there are important practical limits to realizing absolute uniformity of temperature and temperature stability using the prior art techniques involving non-uniform dimensioning and spacing of the emitter fingers.
- Moreover, when the number of emitter fingers arranged in a single row becomes very large, the spacing non-uniformity of the emitter fingers residing in the center region of a power transistor becomes very gradual. As shown in
FIG. 5 , whileemitter fingers emitter fingers emitter fingers - In comparison to the 10-finger power transistor where emitter fingers No. 4-7 are the center fingers, emitter fingers (No. 4-No. 16) in the 20-finger power transistor are all center fingers. The available space in the center area must be shared among all the center fingers, thus the central area of a 20 finger device becomes crowded because the edge area in a 20-finger device remains the same as in the 10-finger device. The difference in the theoretically optimum spacing between, for example,
emitter fingers 4 and 5 and the spacing betweenemitter fingers 5 and 6 in a 20-finger device can be very small. The above-described practical limitations in manufacturing and design introduce variation in the dimensions and locations of emitter fingers that can result in significant temperature non-uniformity once the device's operation reaches its steady state. - Exacerbating the problem is the fact that the larger is the total number of emitter fingers in a power transistor, the higher is the junction temperature (
FIG. 6 ). That is, without any statistical variations in emitter finger locations and dimensions, devices with a large number of emitter fingers run hotter than similarly constructed devices with fewer emitter fingers. Because devices with large numbers of emitter fingers run hotter, statistical variations in emitter fingers and locations can render these devices especially susceptible to current “hogging” due to the formation of local hot spots, or thermal runaway trigged by local non-uniformity of emitter finger width, e.g., statistical variation of the lithography feature size. - As a result, the dimensions and locations of the emitter fingers can be quite critical, and even small variations in dimensions can create large variations in temperature uniformity.
FIG. 7 (a) shows an exemplary prior art 23-emitter finger GaAs HBT, with nominal 2 μm finger width and nominal 20 μm finger length. The emitter fingers of the device ofFIG. 7 (a) are arranged in a non-uniform fashion to produce theoretically uniform junction temperatures, assuming that the emitter fingers are manufactured to have perfect locations and dimensions. - FIGS. 7(b) and 7(c) present calculated steady-state temperature and current distributions of the device of
FIG. 7 (a) under the assumption that process variation has increased the width of the No. 12emitter finger 42 to 2.02 μm. That is, the No. 12emitter finger 42 is assumed to be 1% wider than the nominal 2μ width of the other emitter fingers in the device ofFIG. 7 (a).FIG. 7 (b) shows the calculated temperature profile across the emitter fingers resulting from this 1% variation in width of the no.12emitter finger 42.FIG. 7 (c) shows the calculated current profile across the emitter fingers resulting from this 1% variation in width of the no. 12emitter finger 42 during steady-state high power operation. These results show that even slight variations in finger dimensions can substantially degrade temperature uniformity and cause current hogging. - Generally, as the number of emitter fingers increases, smaller finger width variations will produce current hogging and temperature non-uniformity. Similarly, higher operating temperatures and higher output power tend to increase these adverse effects on temperature uniformity caused by process variation.
- In summary, practical limitations in the manufacturing and design of power transistors having a very large number of emitter fingers limit the utility of non-uniform dimensioning and location of emitter fingers in achieving temperature uniformity and thermal stability in such devices. As a result, these prior art techniques are mainly useful for low and medium power transistors where not many emitter fingers are required. What is needed is a power transistor structure suitable for high power transistors using large numbers of emitter fingers and having better power performance, improved manufacturability, and more reliable thermal stability compared to prior-art power transistor structures. What is further needed are structures and dimensions which are particularly suited for SiGe applications, and other structures and dimensions which are particularly suited for GaAs applications.
- A preferred embodiment of a high power transistor according to the invention includes a plurality of emitter fingers fabricated on a common semiconductor chip and wherein the emitter electrodes of bipolar junction transistors (or heterojunction bipolar transistors) are in the form of parallel elongated finger elements having a uniform or non-uniform spacing. These fingers are arranged in a 1-dimensional (1-D) or 2-dimensional (2-D) form such that the potential thermal instability at high power operation may be reduced ballasting resistors with substantially reduced values, in the range of less than 10 ohms, which provides power, efficiency, and high frequency performance benefits in comparison to prior-art ballast resistor approaches which used larger resistor values typically in the range of 100 ohms.
- These fingers can also be arranged in a 1-dimensional (1-D) or 2-dimensional (2-D) form such that the potential thermal instability at high power operation is completely eliminated by forming a “hollow-center” layout where one or more elongated emitter fingers or subcells are left out during design or disconnected during manufacture, with or without the use of small ballasting resistors.
- In a particularly preferred embodiment, a high power transistor according to the invention includes a base that comprises a GaAs layer less than about 100 microns thick, and emitter fingers that are less than about 1 micron wide.
- Further objects, features, and advantages of the invention will be apparent from the following detailed description when taken in conjunction with the following drawings.
- In the drawings:
-
FIG. 1 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of uniformly-spaced emitter fingers; -
FIG. 2 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of emitter fingers arranged in uniformly spaced subcells each having 2 emitter fingers; -
FIG. 3 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of uniformly-spaced emitter fingers each with a ballast resistor; -
FIG. 4 is a top planar view of the layout of a prior-art power bipolar junction transistor having a plurality of emitter fingers arranged in uniformly-spaced subcells each having 2 emitter fingers each with a ballast resistor; -
FIG. 5 is a top planar view of the layouts of four exemplary prior-art power bipolar junction transistors having 5, 10, 20, and 30 emitter fingers arranged non-uniformly; -
FIG. 6 presents calculated temperature profiles for the four exemplary power bipolar junction transistor layouts ofFIG. 5 ; -
FIG. 7 (a) is a top planar view of the layout of a prior-art power bipolar junction transistor having 23 emitter fingers arranged non-uniformly; -
FIG. 7 (b) shows the calculated temperature profile during steady-state high power operation of the device ofFIG. 7 (a) assuming that emitter finger number 12 is 1% wider than intended; -
FIG. 7 (c) shows the calculated current profile during steady-state high power operation of the device ofFIG. 7 (a) assuming that emitter finger number 12 is 1% wider than intended; - FIGS. 8(a)-8(f) show an exemplary fabrication process for fabricating a SiGe (a similar process can be used for GaAs) bipolar junction power transistor having two emitter fingers;
-
FIG. 9 is a top planar view of a power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor; -
FIG. 10 is a top planar view of another power bipolar junction transistor according to the invention having emitter fingers arranged in non-uniformly spaced subcells each having 2 emitter fingers each with a small ballast resistor; -
FIG. 11 (a) is a top planar view of another power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor and arranged in two arc-shaped rows; -
FIG. 11 (b) shows the calculated temperature profile along one row of emitter fingers in the device ofFIG. 11 (a); -
FIG. 12 (a) is a top planar view of another power bipolar junction transistor according to the invention having non-uniformly spaced emitter fingers each with a small ballast resistor and arranged in two arc-shaped rows surrounding a third straight center row of fingers; -
FIG. 12 (b) is the calculated temperature profile along the center row of emitter fingers in the device ofFIG. 12 (a); -
FIG. 13 is a top planar view of another power transistor structure according to the invention with emitter fingers surrounding a “hollow center” central gap where an emitter finger is missing; -
FIG. 14 is a top planar view of another power transistor structure according to the invention with subcells each having two emitter fingers and surrounding a “hollow center” central gap where a subcell is missing; -
FIG. 15 is a top planar view of another power transistor structure according to the invention with non-uniformly spaced subcells, each with two emitter fingers, surrounding a central subcell having a single emitter finger; -
FIG. 16 (a) is a top planar view of an exemplary prior-art bipolar power transistor layout; -
FIG. 16 (b) is a top planar view of a power transistor in accordance with the invention having the same number of emitter fingers and subcells, and the same chip area, as the device ofFIG. 16 (a); and -
FIG. 16 (c) shows the measured power performance data of the device ofFIG. 16 (b) compared to the prior art device ofFIG. 16 (a). - As previously discussed, high power transistors commonly employ multiple finger structures to increase the total power output. For bipolar junction transistors and heterojunction bipolar transistors, these multiple fingers are emitter fingers. Although the following embodiments of this invention are described using bipolar junction transistors as examples, the bipolar junction transistors are used by way of example, and not as a limitation on the scope of the invention.
- By way of example, and not as a limitation, FIGS. 8(a)-8(f) show an exemplary prior art fabrication process for fabricating a SiGe bipolar junction power transistor having two emitter fingers. The prior art process of FIGS. 8(a)-8(f), sometimes called the double-mesa process, can be used for fabricating X-band SiGe HBT devices. This prior art process is highly repeatable and reliable, and typically employs 7 mask levels and 20 steps. This prior art process is used herein as an example of a process for implementing SiGe or GaAs power transistors using the specific dimensions and device layouts of the present invention, but it is by no means the only such process that could be used.
- The exemplary process of FIGS. 8(a)-8(f) begins with a material stack formed on an
Si substrate 62, followed by a heavily N-type dopedSi subcollector layer 60, a lightly n-type dopedSi collector layer 58, a p-type dopedSiGe base layer 56, and an n-type dopedSi emitter layer 53.Emitter metal contacts 52 are formed using standard photolithography and liftoff techniques. Theemitter metal 52, for example Cr or Au, is first evaporated on top of the highly dopedemitter cap layer 53 of the Si/SiGe/Si heterostructure. Themulti-emitter metal fingers 52, typically 2 microns wide, are formed with the first-level mask, leaving the structure shown inFIG. 8 (a). The width of the emitter fingers may be subject to statistical variations as discussed above. - The patterned
emitter metal fingers 52 then serve as self-aligned mask for subsequent dry/wet etching of theSi emitter layer 53 to expose the boron-dopedSiGe base layer 56, leaving the structure shown inFIG. 8 (b). Another photolithography step is used to form self-alignedbase metal 55 on top of the exposedSiGe base layer 56, leaving the structure shown inFIG. 8 (c). - The third mask is used to form the base mesa by RIE and to expose the highly
doped subcollector layer 60 for collector contact formation, leaving the structure shown inFIG. 8 (d). Next,collector metal 64 is deposited and formed in another lithography step (4th mask). The active devices are isolated by removing thesubcollector material 60 around the devices and exposing the high-resistivity Si substrate 62 (5th mask), leaving the structure shown inFIG. 8 (e). - A conformal PECVD SiO2 deposition is used to form a
passivation layer 66 over theactive device 50 and the exposedsubstrate 62. Contact viaholes 68 are opened in thepassivation layer 66 by RIE (6th mask) andinterconnect pad metal 67 is deposited and then patterned with photolithography (7th mask) to finish the fabrication process and form the completeddevice 50 shown inFIG. 8 (f). - Referring to the drawings,
FIG. 9 is a top planar view of a powerbipolar junction transistor 70 according to the invention having non-uniformly spacedemitter fingers 26 withsmall ballast resistors 71. In thedevice 70 ofFIG. 9 , a non-uniform spacing ofemitter fingers 26 is arranged using the finite element analysis software, based on the known heating power density. The distance betweenadjacent emitter fingers 26 increases from the center of thedevice 70 to the periphery, so that distance X1>X2>X3. - The arrangement of the
multiple emitter fingers 26 must adhere to the applicable design rules of the technology used to build the device. Since any design rule will have some minimum allowable increment of distance, even if the perfect placement and spacing of the multiple emitter fingers can be calculated, this perfect placement and spacing cannot be implemented in practice, so at least some non-uniformity of junction temperature should be expected. - In the
device 70 ofFIG. 9 , theemitter fingers 26 of this power transistor are each connected in series with asmall ballast resistor 71, although this is not required and less than all of the emitter fingers may include thesmall ballast resistor 71. For example, in an appropriate applicationsmall ballast resistors 71 can be omitted from theemitter fingers 26 on the periphery while maintainingsmall ballast resistors 71 on at least some of thecentral emitter fingers 26 to maintain temperature stability. The small ballast resistor is preferably in the range 1-10 ohms. - In an appropriate fabrication process, such a small valued ballast resistor can be formed simply by shrinking the size of a
contact hole 68 to theemitter 24, for example to theemitter metal 52, to create the desired resistance in series with theemitter finger 26, with no additional structure or processing step required. The values of the resistors in accordance to the preferred embodiment are calculated based on the known maximum statistical variation of the finger width with the goal of a thermally stable operation condition. The calculation may involve iterations using measured or calculated thermal-electric coefficients and heating power density. - Although the
small ballast resistors 71 are connected to emitter fingers in thedevice 70 ofFIG. 9 , similar layouts withsmall ballast resistors 71 connected in series with the base 22 can also be used, instead of or in addition tosmall ballast resistors 71 in series with theemitter 24. In an appropriate fabrication process, such a small valued ballast resistor can be formed simply by shrinking the size of a contact hole to thebase 22, for example to thebase metal 55, to create the desired resistance in series with thebase 22, with no additional structure or processing step required. - In comparison to the prior-art approach which uses
ballast resistors 38 having values in the range of 100 ohms, thesmall ballast resistors 71 in thedevice 70 ofFIG. 9 have values in the range 1-10 ohms, preferably around 1-3 ohms, and thesesmall ballast resistors 71 are used to prevent thermal instability caused by small variations of finger spacing and finger width. As a result, only a very small fraction of ballast resistor values of the prior-art ballast resistor values is needed in this invention. Hence, thesmall ballast resistors 71 of this approach can have significantly improved power, efficiency, and high frequency performance compared to prior approaches that uselarger ballast resistors 38. -
FIG. 10 is a top planar view of another powerbipolar junction transistor 74 according to the invention havingemitter fingers 26 arranged in non-uniformly spaced subcells each having 2 emitter fingers with small ballast resistors. Although two fingers are grouped a subcell in this example, more fingers can be grouped together if finger width is small. The non-uniform layout is arranged by adhering to the design rule. The distance betweenadjacent subcells 32 increases from the center of thedevice 74 to the periphery, so that distance Y1>Y2. To ensure worst-case thermal stability, the values of the small ballast resistors can be determined by assuming the width of all the emitter fingers in the center subcell or subcells have the maximum possible finger width and assuming the distance between the central adjacent subcells is the minimum, given the expected variation in finger width and subcell spacing in the manufacturing process used. The advantages of this small ballast resistor approach are reduced base-collector junction capacitance and parasitic collector resistance in comparison to the prior art which uses relatively large ballast resistors. -
FIG. 11 (a) is a top planar view of another power bipolar junction transistor 78 according to the invention having non-uniformly spacedemitter fingers 26 each with asmall ballast resistor 71 and arranged in two arc-shaped rows. The dimensions and arrangement of the 2-dimensional thermally balanced structure is designed to avoid the need for larger valuedballast resistors 38 to ensure thermal stability of the device 78 arising from the statistical variation of finger width and allowable minimal increment of emitter finger spacing in the one-row non-uniform layout. The side-to-side distance betweenadjacent emitter fingers 26 increases from the center of the device 78 to the periphery, so that distance X1>X2>X3. The front to back distance between the rows ofadjacent emitter fingers 26 also increases from the center of the device 78 to the periphery, so that distance Z1>Z2. Of course, the arrangement of the two rows of emitter fingers 76 must adhere to the design rules of the process used to manufacture the device 78. -
FIG. 11 (b) shows the calculated temperature profile along one row of emitter fingers of the device 78 ofFIG. 11 (a). The temperature non-uniformity shown inFIG. 11 (b) due to the limitation of minimal increment of finger spacing is expected. The values ofsmall ballast resistors 71 can be selected in a similar fashion to the devices ofFIGS. 9 and 10 , by assuming that thecentral emitter fingers 26 have the maximum finger width and minimum spacing expected in the manufacturing process used to fabricate the device 78, to ensure worst case thermal stability. - Although the device of
FIG. 11 (a) includes a single finger structure for illustration, subcell structures with two or more emitter fingers grouped together can also be used. Again,small ballast resistors 71 can be placed in series with thebase 22, instead of or in addition tosmall ballast resistors 71 in series with theemitter 24. - In addition, although arc-shaped rows are used for the illustration of the 2-D thermal balance structure in the device of
FIG. 11 (a), the two rows can be straightened up. In either case, the spacing between two rows affects the selection of ballast resistor values of the emitter fingers. Usingsmall ballast resistors 71 having similar values to thesmall ballast resistors 71 used in thedevice 70 ofFIG. 9 , the thermally stable power output of a device similar to that ofFIG. 11 (a) but with two straightened up rows may be substantially greater than the device ofFIG. 9 with similar device area. -
FIG. 12 (a) is a top planar view of another powerbipolar junction transistor 80 according to the invention having non-uniformly spacedemitter fingers 26 each having asmall ballast resistor 71 and arranged in two arc-shaped rows surrounding a third straight center row of fingers. The side-to-side distance betweenadjacent emitter fingers 26 increases from the center of thedevice 80 to the periphery, so that distance X1>X2>X3. The front to back distance between the rows ofadjacent emitter fingers 26 also increases from the center of thedevice 80 to the periphery, so that distance Z1>Z2. Although a three-row layout is shown inFIG. 12 (a) as an example, layouts of more than three rows can also be used. Although all theemitter fingers 26 indevice 80 includesmall ballast resistors 71, this is not necessary and in an appropriate case fewer than all theemitter fingers 26 may include asmall ballast resistor 71, and the values of thesmall ballast resistors 71 need not all be the same. -
FIG. 12 (b) is the calculated temperature profile along the center row ofemitter fingers 26 in thedevice 80 ofFIG. 12 (a). Similar to the device ofFIG. 11 (a), temperature nonuniformity is expected when a design rule is adhered to, because of the imprecision in emitter finger placement resulting from the finite resolution of design rules.Subcell structures 32 can replace the singleemitter finger structures 26 andsmall ballast resistors 71 instead of or in addition to smallemitter ballast resistors 71 can also be used in adevice 80 according to the invention similar to that ofFIG. 12 (a). In addition, all the multiple rows can be straightened up and the center row or rows can have smaller number of emitter fingers or subcells than the outer rows. The values of thesmall ballast resistors 71 can be selected in a similar fashion to the device ofFIG. 9 . -
FIG. 13 is a top planar view of anotherpower transistor structure 82 according to the invention with non-uniformly spaced emitter fingers surrounding a central gap (“hollow center”) 84. Like thedevice 70 ofFIG. 9 , the side-to-side distance betweenadjacent emitter fingers 26 increases from the center of the device to the periphery, so that distance X1>X2>X3. However, in thedevice 82 ofFIG. 13 , the side-to-side distance X1 is much greater than the side-to-side distance between other pairs of adjacent emitter fingers, whereby X1 is more than twice as large as the distance between any other two pairs of adjacent emitter fingers. - Using the “hollow center” 84 in the
device 82 ofFIG. 13 , the multiple emitter fingers may be arranged either in a uniform or non-uniform fashion while adhering to the design rule, and still achieve thermally stable operation. Instead of seeking a nearly uniform temperature distribution, thehollow center 84 in the device ofFIG. 13 can result in a non-uniform temperature where the center is cooler than the periphery, to overcompensate and make thedevice 82 even more thermally stable. - The potential thermal instability that may be triggered by the deviation from the required emitter finger spacing and/or by statistical variation of finger width is ameliorated using the
hollow center 84. Thehollow center 84 can be implemented by placing a gap at the center of thedevice 82 when designing the device, or thehollow center 84 can be implemented by electrically disconnecting or removing one or more of thecenter emitter fingers 26 after thedevice 82 has been partially fabricated. - The “hollow center” 84 can be applied to layouts of multiple rows and can also be applied to a single row at locations instead of or in addition to center locations when the number of emitter fingers is sufficiently large. Although a non-uniform arrangement of emitter fingers is preferred, it is not required for this embodiment, and the “hollow center” 84 can be used even when all the other emitter fingers are uniformly spaced. Accordingly, both uniform and non-uniform arrangements of emitter fingers using the “hollow center” 84 can be used in a device according to the invention. The
device 82 ofFIG. 13 may also include one or more small valuedballast resistors 71 connected to one or more of the emitter fingers or the base, although this is not required. -
FIG. 14 is a top planar view of anotherpower transistor structure 86 according to the invention with non-uniformly spaced subcells 32, each with twoemitter fingers 26, surrounding a “hollow center” 88 central gap. Although thedevice 86 ofFIG. 14 has subcells 32 containing twoemitter fingers 26, this is not required and subcells containing a greater number of emitter fingers could be used. - Like the
device 74 ofFIG. 10 , the side-to-side distance betweenadjacent subcells 32 increases from the center of the device to the periphery, so that distance Y1>Y2. However, in thedevice 86 ofFIG. 14 , the side-to-side distance Y1 is much greater than the side-to-side distance between other pairs of adjacent subcells, whereby Y1 is more than twice as large as the distance between any other two pairs of adjacent subcells. - The
hollow center 88 can be formed by electrically disconnecting or removing selected subcells during manufacture or thehollow center 88 subcells can be simply left out during layout of the device. Thedevice 86 ofFIG. 14 may include one or more small valuedballast resistors 71 connected to one or more of the emitter fingers and/or the base, although this is not required. -
FIG. 15 is a top planar view of anotherpower transistor structure 90 according to the invention withsubcells 94, each with two emitter fingers, surrounding acentral subcell 92 having a single emitter finger. Like thedevice 74 ofFIG. 10 and thedevice 86 ofFIG. 14 , the side-to-side distance between adjacent subcells increases from the center of the device to the periphery, so that distance Y1>Y2. However, in thedevice 90 ofFIG. 15 , the central subcells has a reduced number of emitter fingers compared to the other subcells, instead of a “hollow center” subcell 88 used in thedevice 86 ofFIG. 14 . - Although the
device 90 ofFIG. 15 includes one single-fingercentral subcell 92 surrounded byother subcells 94 each containing 2 emitter fingers, the surroundingsubcells 94 could have a greater number of emitter fingers. Similarly, if the surrounding subcells have a greater number of emitter fingers, the central subcell could also have a greater number of emitter fingers, as long as the number of emitter fingers in thecentral subcell 92 is less than the number of emitter fingers in the surroundingsubcells 94. Thedevice 90 ofFIG. 15 may also include one or moresmall ballast resistors 71 connected to one or more of the emitter fingers and/or the base. -
FIG. 16 (a) is a top planar view of an exemplary prior-art SiGeHBT power transistor 96 with a uniform layout. The 16emitter fingers 26 are grouped in 8 subcells 32 and uniformly spaced along a single row.FIG. 16 (b) is a top planar view of apower transistor 98 in accordance with the invention having the same number of emitter fingers and subcells, and the same chip area, as thedevice 96 ofFIG. 16 (a). - Unlike the uniformly-spaced single row of subcells used in the
device 96 ofFIG. 16 (a), thedevice 98 ofFIG. 16 (b) includes 8 subcells arranged non-uniformly in a 2-dimensional form, in accordance with the invention, while keeping the same chip area and the same subcell structure as thedevice 96 ofFIG. 16 (a). The side-to-side distance betweenadjacent subcells 32 increases from the center of thedevice 98 to the periphery, so that distance Y1>Y2. The front to back distance between the rows ofadjacent subcells 32 also increases from the center of thedevice 98 to the periphery, so that distance Z1>Z2. Thedevice 98 ofFIG. 16 (b) may also include one or moresmall ballast resistors 71 connected to one or more of the emitter fingers and/or the base. -
FIG. 16 (c) shows the measured power performance data of the device ofFIG. 16 (b) compared to the prior art device ofFIG. 16 (a), when both HBTs are fabricated on the same chip and operated at the same bias and input signal levels. The results shown inFIG. 16 (c) demonstrate that output power, power gain and power added efficiency are simultaneously improved using the 2-dimensional layout of thedevice 98 ofFIG. 16 (b) compared to the 1-dimensional prior-art device 96 ofFIG. 16 (a). - There are various possibilities with regard to alternative embodiments of a solid state high power device and method according to the invention.
- In any device according to the various embodiments disclosed herein, the thermal stability of each individual emitter finger or subcell is preferably maintained, to ensure thermal stability of a composite structure that comprises all the emitter fingers or subcells together. To ensure the stability of the individual emitter fingers of a GaAs device, each finger should be less than 2 microns wide, preferably around one micron wide.
- To ensure the stability of the individual emitter fingers of a GaAs device, the substrate thickness should also be thinned to a certain thickness depending on the overall heat dissipation of the device and finger width, so that the substrate thickness should be less than 130 microns, and preferably around 100 microns. A GaAs device having an emitter finger width less than 2 microns and a substrate thickness less than 130 microns is particularly preferred.
- It is understood that the invention is not confined to the embodiments set forth herein as illustrative, but embraces all such forms thereof as come within the scope of the following claims.
Claims (40)
1. A high power transistor comprising:
a base, an emitter, a collector, and a plurality of elongated emitter fingers extending from and electrically connected to the emitter;
wherein the elongated emitter fingers are arranged side by side in a row, the row extending between a first row end and a second row end and having a row center located at the midpoint between the first row end and the second row end;
wherein the plurality of elongated emitter fingers includes a first pair of emitter fingers separated by a first side-to-side distance, and a second pair of emitter fingers separated by a second side-to-side distance, wherein the midpoint between the first pair of emitter fingers is positioned closer to the row center than the midpoint between the second pair of emitter fingers, and wherein the first side-to-side distance is greater than the second side-to-side distance; and
further comprising at least one small ballast resistor having a value of less than about 10 ohms and electrically connected in series with at least one elongated emitter finger in the plurality of emitter fingers.
2. The device of claim 1 , wherein the value of the at least one small ballast resistor is less than about 5 ohms.
3. The device of claim 1 , wherein the first side-to-side distance is at least twice as big as the second side-to-side distance.
4. The device of claim 1 , wherein the at least one small ballast resistor is formed using a contact hole through a passivation layer to an emitter metal layer, wherein the contact hole has dimensions less than about 1 micron by 1 micron.
5. The device of claim 1 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is less than about 1.2 microns.
6. The device of claim 1 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is about 0.8 microns.
7. The device of claim 1 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is less than about 100 microns.
8. The device of claim 1 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is about 80 microns.
9. The device of claim 1 , wherein the base includes a layer of GaAs, wherein the majority of the elongated emitter fingers have widths that are less than about 1.2 microns.
10. The device of claim 9 , wherein the thickness of the layer of GaAs is less than about 100 microns.
11. A high power transistor comprising:
a base, an emitter, a collector, and a plurality of emitter finger subcells, each emitter finger subcell comprising one or more elongated emitter fingers extending from and electrically connected to the emitter;
wherein the emitter finger subcells are arranged side by side in a row, the row extending between a first row end and a second row end and having a row center located at the midpoint between the first row end and the second row end;
wherein the plurality of emitter finger subcells includes a first pair of emitter finger subcells separated by a first side-to-side distance, and a second pair of emitter finger subcells separated by a second side-to-side distance, wherein the midpoint between the first pair of emitter finger subcells is positioned closer to the row center than the midpoint between the second pair of emitter finger subcells, and wherein the first side-to-side distance is greater than the second side-to-side distance; and
further comprising at least one small ballast resistor having a value of less than about 10 ohms and electrically connected in series with at least one elongated emitter finger in the plurality of emitter finger subcells.
12. The device of claim 11 , wherein each emitter finger subcell includes at least two elongated emitter fingers.
13. The device of claim 11 , wherein the first side-to-side distance is at least twice as big as the second side-to-side distance.
14. The device of claim 11 , wherein the at least one small ballast resistor is formed using a contact hole through a passivation layer to an emitter metal layer, wherein the contact hole has dimensions less than about 1 micron by 1 micron.
15. The device of claim 11 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is less than about 1.2 microns.
16. The device of claim 11 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is about 0.8 microns.
17. The device of claim 11 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is less than about 100 microns.
18. The device of claim 11 , wherein the plurality of emitter finger subcells includes a first subcell and a second subcell, wherein the first subcell is positioned closer to the row center than the second subcell, and wherein the number of elongated emitter fingers in the first subcell is less than the number of elongated emitter fingers in the second subcell.
19. The device of claim 18 , wherein the number of elongated emitter fingers in the first subcell is one.
20. The device of claim 11 , wherein the wherein the plurality of emitter finger subcells includes a first subcell and a second subcell, wherein the first subcell is positioned closer to the row center than the second subcell, and wherein the area of the elongated emitter fingers in the first subcell is less than the area of the elongated emitter fingers in the second subcell.
21. A high power transistor comprising:
a base, an emitter, a collector, and a plurality of elongated emitter fingers extending from and electrically connected to the emitter;
wherein the elongated emitter fingers are arranged side by side in a plurality of rows, each row extending between a first row end and a second row end and having a row center located at the midpoint between the first row end and the second row end;
wherein the plurality of elongated emitter fingers in at least one row in the plurality of rows includes a first pair of emitter fingers separated by a first side-to-side distance, and a second pair of emitter fingers separated by a second side-to-side distance, wherein the midpoint between the first pair of emitter fingers is positioned closer to the row center than the midpoint between the second pair of emitter fingers, and wherein the first side-to-side distance is greater than the second side-to-side distance; and
further comprising at least one small ballast resistor having a value of less than about 10 ohms and electrically connected in series with at least one elongated emitter finger in the plurality of emitter fingers.
22. The device of claim 21 , wherein the value of the at least one small ballast resistor is less than about 5 ohms.
23. The device of claim 21 , wherein the first side-to-side distance is at least twice as big as the second side-to-side distance.
24. The device of claim 21 , wherein the at least one small ballast resistor is formed using a contact hole through a passivation layer to an emitter metal layer, wherein the contact hole has dimensions less than about 1 micron by 1 micron.
25. The device of claim 21 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is less than about 1.2 microns.
26. The device of claim 21 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is about 0.8 microns.
27. The device of claim 21 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is less than about 100 microns.
28. The device of claim 21 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is about 80 microns.
29. The device of claim 21 , wherein the base includes a layer of GaAs, wherein the majority of the elongated emitter fingers have widths that are less than about 1.2 microns.
30. The device of claim 29 , wherein the thickness of the layer of GaAs is less than about 100 microns.
31. A high power transistor comprising:
a base, an emitter, a collector, and a plurality of emitter finger subcells, each emitter finger subcell comprising one or more elongated emitter fingers extending from and electrically connected to the emitter;
wherein the emitter finger subcells are arranged side by side in a plurality of rows, each row extending between a first row end and a second row end and having a row center located at the midpoint between the first row end and the second row end;
wherein the plurality of emitter finger subcells in at least one row in the plurality of rows includes a first pair of emitter finger subcells separated by a first side-to-side distance, and a second pair of emitter finger subcells separated by a second side-to-side distance, wherein the midpoint between the first pair of emitter finger subcells is positioned closer to the row center than the midpoint between the second pair of emitter finger subcells, and wherein the first side-to-side distance is greater than the second side-to-side distance; and
further comprising at least one small ballast resistor having a value of less than about 10 ohms and electrically connected in series with at least one elongated emitter finger in the plurality of emitter finger subcells.
32. The device of claim 31 , wherein each emitter finger subcell includes at least two elongated emitter fingers.
33. The device of claim 31 , wherein the first side-to-side distance is at least twice as big as the second side-to-side distance.
34. The device of claim 31 , wherein the plurality of rows includes a first row and a second row;
wherein the distance between the row center of the first row and the row center of the second row is greater than the distance between the first row end of the first row and the first row end of the second row; and
wherein the distance between the row center of the first row and the row center of the second row is greater than the distance between the second row end of the first row and the second row end of the second row.
35. The device of claim 34 , wherein the plurality of rows further comprises a third row, wherein the third row is positioned between the first row and the second row, and wherein the third row is substantially straight.
36. The device of claim 31 , wherein the base includes a layer of GaAs, and wherein the width of at least one elongated emitter finger is less than about 1.2 microns.
37. The device of claim 31 , wherein the base includes a layer of GaAs, and wherein the thickness of the layer of GaAs is less than about 100 microns.
38. The device of claim 31 , wherein the plurality of emitter finger subcells in at least one row in the plurality of rows includes a first subcell and a second subcell, wherein the first subcell is positioned closer to the row center than the second subcell, and wherein the number of elongated emitter fingers in the first subcell is zero or more and less than the number of elongated emitter fingers in the second subcell.
39. The device of claim 38 , wherein the number of elongated emitter fingers in the first subcell is one or more.
40. The device of claim 31 , wherein the plurality of emitter finger subcells in at least one row includes a first subcell and a second subcell, wherein the first subcell is positioned closer to the row center than the second subcell, and wherein the area of the elongated emitter fingers in the first subcell is less than the area of the elongated emitter fingers in the second subcell.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/993,224 US20050151159A1 (en) | 2003-11-21 | 2004-11-19 | Solid-state high power device and method |
US11/498,665 US7705425B2 (en) | 2003-11-21 | 2006-08-02 | Solid-state high power device and method |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US71875703A | 2003-11-21 | 2003-11-21 | |
US60776704P | 2004-09-07 | 2004-09-07 | |
US60776204P | 2004-09-07 | 2004-09-07 | |
US10/993,224 US20050151159A1 (en) | 2003-11-21 | 2004-11-19 | Solid-state high power device and method |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US71875703A Continuation-In-Part | 2003-11-21 | 2003-11-21 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/498,665 Continuation US7705425B2 (en) | 2003-11-21 | 2006-08-02 | Solid-state high power device and method |
Publications (1)
Publication Number | Publication Date |
---|---|
US20050151159A1 true US20050151159A1 (en) | 2005-07-14 |
Family
ID=34637201
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/993,224 Abandoned US20050151159A1 (en) | 2003-11-21 | 2004-11-19 | Solid-state high power device and method |
US11/498,665 Expired - Lifetime US7705425B2 (en) | 2003-11-21 | 2006-08-02 | Solid-state high power device and method |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/498,665 Expired - Lifetime US7705425B2 (en) | 2003-11-21 | 2006-08-02 | Solid-state high power device and method |
Country Status (2)
Country | Link |
---|---|
US (2) | US20050151159A1 (en) |
WO (1) | WO2005052997A2 (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050216873A1 (en) * | 2004-03-23 | 2005-09-29 | Raminderpal Singh | Method of checking the layout versus the schematic of multi-fingered MOS transistor layouts using a sub-circuit based extraction |
US20100066454A1 (en) * | 2008-09-18 | 2010-03-18 | Zhenqiang Ma | High-power common-base amplifier employing current source output bias |
US7795047B1 (en) * | 2004-12-17 | 2010-09-14 | National Semiconductor Corporation | Current balancing in NPN BJT and BSCR snapback devices |
US7821102B2 (en) | 2006-12-20 | 2010-10-26 | Freescale Semiconductor, Inc. | Power transistor featuring a double-sided feed design and method of making the same |
US20150123246A1 (en) * | 2013-11-04 | 2015-05-07 | Taiwan Semiconductor Manufacturing Company, Ltd. | Bipolar Junction Transistor Formed on Fin Structures |
US20150171108A1 (en) * | 2013-11-12 | 2015-06-18 | Skyworks Solutions, Inc. | Radio-frequency switching devices having improved voltage handling capability |
US20160141220A1 (en) * | 2014-11-18 | 2016-05-19 | Sumitomo Electric Industries, Ltd. | Hetero-bipolar transistor and method for producing the same |
US9899502B2 (en) | 2016-04-27 | 2018-02-20 | United Microelectronics Corp. | Bipolar junction transistor layout structure |
CN111384161A (en) * | 2018-12-25 | 2020-07-07 | 株式会社村田制作所 | Bipolar transistor |
US20220013414A1 (en) * | 2013-11-12 | 2022-01-13 | Skyworks Solutions, Inc. | Methods related to radio-frequency switching devices having improved voltage handling capability |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7932582B2 (en) * | 2008-06-18 | 2011-04-26 | National Semiconductor Corporation | Compact dual direction BJT clamps |
JP5595751B2 (en) * | 2009-03-11 | 2014-09-24 | ルネサスエレクトロニクス株式会社 | ESD protection element |
US20140021603A1 (en) * | 2012-07-23 | 2014-01-23 | Rf Micro Devices, Inc. | Using an interconnect bump to traverse through a passivation layer of a semiconductor die |
US9653447B2 (en) | 2014-09-24 | 2017-05-16 | Nxp B.V. | Local interconnect layer enhanced ESD in a bipolar-CMOS-DMOS |
US10868155B2 (en) | 2014-11-27 | 2020-12-15 | Murata Manufacturing Co., Ltd. | Compound semiconductor device |
JP6071009B2 (en) | 2014-11-27 | 2017-02-01 | 株式会社村田製作所 | Compound semiconductor device |
US11508834B2 (en) | 2014-11-27 | 2022-11-22 | Murata Manufacturing Co., Ltd. | Compound semiconductor device |
US20160372396A1 (en) | 2015-06-22 | 2016-12-22 | Globalfoundries Inc. | Chip packages with reduced temperature variation |
US9728603B2 (en) | 2015-06-22 | 2017-08-08 | Globalfoundries Inc. | Bipolar junction transistors with double-tapered emitter fingers |
EP3151283A1 (en) * | 2015-09-29 | 2017-04-05 | Nexperia B.V. | Vertical dmos bjt semiconductor device |
Citations (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3582723A (en) * | 1967-09-15 | 1971-06-01 | Philips Corp | Transistor |
US3704398A (en) * | 1970-02-14 | 1972-11-28 | Nippon Electric Co | Multi-emitter power transistor having emitter region arrangement for achieving substantially uniform emitter-base junction temperatures |
US5023189A (en) * | 1990-05-04 | 1991-06-11 | Microwave Modules & Devices, Inc. | Method of thermal balancing RF power transistor array |
US5057882A (en) * | 1989-06-30 | 1991-10-15 | Texas Instruments Incorporated | Thermally optimized interdigitated transistor |
US5177583A (en) * | 1990-02-20 | 1993-01-05 | Kabushiki Kaisha Toshiba | Heterojunction bipolar transistor |
US5252841A (en) * | 1991-05-09 | 1993-10-12 | Hughes Aircraft Company | Heterojunction bipolar transistor structure having low base-collector capacitance, and method of fabricating the same |
US5374844A (en) * | 1993-03-25 | 1994-12-20 | Micrel, Inc. | Bipolar transistor structure using ballast resistor |
US5378922A (en) * | 1992-09-30 | 1995-01-03 | Rockwell International Corporation | HBT with semiconductor ballasting |
US5506764A (en) * | 1994-01-31 | 1996-04-09 | Astec International, Ltd. | Electrical power converter with step-gapped transformer |
US5616950A (en) * | 1992-05-29 | 1997-04-01 | Texas Instruments Incorporated | Thermally uniform transistor |
US5734193A (en) * | 1994-01-24 | 1998-03-31 | The United States Of America As Represented By The Secretary Of The Air Force | Termal shunt stabilization of multiple part heterojunction bipolar transistors |
US5804867A (en) * | 1996-10-02 | 1998-09-08 | Ericsson Inc. | Thermally balanced radio frequency power transistor |
US6034383A (en) * | 1997-11-13 | 2000-03-07 | Northrop Grumman Corporation | High power density microwave HBT with uniform signal distribution |
US6051871A (en) * | 1998-06-30 | 2000-04-18 | The Whitaker Corporation | Heterojunction bipolar transistor having improved heat dissipation |
US6316827B1 (en) * | 1997-09-12 | 2001-11-13 | Nec Corporation | Semiconductor device having improved temperature distribution |
US20030038341A1 (en) * | 2001-04-13 | 2003-02-27 | Burhan Bayraktaroglu | Low stress thermal and electrical interconnects for heterojunction bipolar transistors |
US6534857B1 (en) * | 2001-11-02 | 2003-03-18 | Northrop Grumman Corporation | Thermally balanced power transistor |
US6586782B1 (en) * | 1998-07-30 | 2003-07-01 | Skyworks Solutions, Inc. | Transistor layout having a heat dissipative emitter |
US6627925B1 (en) * | 1998-07-30 | 2003-09-30 | Skyworks Solutions, Inc. | Transistor having a novel layout and an emitter having more than one feed point |
US6707341B2 (en) * | 2000-11-30 | 2004-03-16 | Renesas Technology Corp. | Semiconductor device with bipolar transistor device |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4291324A (en) * | 1980-06-20 | 1981-09-22 | Rca Corporation | Semiconductor power device having second breakdown protection |
US20030038347A1 (en) * | 2001-08-22 | 2003-02-27 | Walton Advanced Electronics Ltd | Stackable-type semiconductor package |
US6815304B2 (en) * | 2002-02-22 | 2004-11-09 | Semisouth Laboratories, Llc | Silicon carbide bipolar junction transistor with overgrown base region |
-
2004
- 2004-11-19 US US10/993,224 patent/US20050151159A1/en not_active Abandoned
- 2004-11-19 WO PCT/US2004/038823 patent/WO2005052997A2/en active Application Filing
-
2006
- 2006-08-02 US US11/498,665 patent/US7705425B2/en not_active Expired - Lifetime
Patent Citations (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3582723A (en) * | 1967-09-15 | 1971-06-01 | Philips Corp | Transistor |
US3704398A (en) * | 1970-02-14 | 1972-11-28 | Nippon Electric Co | Multi-emitter power transistor having emitter region arrangement for achieving substantially uniform emitter-base junction temperatures |
US5057882A (en) * | 1989-06-30 | 1991-10-15 | Texas Instruments Incorporated | Thermally optimized interdigitated transistor |
US5177583A (en) * | 1990-02-20 | 1993-01-05 | Kabushiki Kaisha Toshiba | Heterojunction bipolar transistor |
US5023189A (en) * | 1990-05-04 | 1991-06-11 | Microwave Modules & Devices, Inc. | Method of thermal balancing RF power transistor array |
US5252841A (en) * | 1991-05-09 | 1993-10-12 | Hughes Aircraft Company | Heterojunction bipolar transistor structure having low base-collector capacitance, and method of fabricating the same |
US5850099A (en) * | 1992-05-29 | 1998-12-15 | Texas Instruments Incorporated | Thermally uniform transistor |
US5616950A (en) * | 1992-05-29 | 1997-04-01 | Texas Instruments Incorporated | Thermally uniform transistor |
US5378922A (en) * | 1992-09-30 | 1995-01-03 | Rockwell International Corporation | HBT with semiconductor ballasting |
US5374844A (en) * | 1993-03-25 | 1994-12-20 | Micrel, Inc. | Bipolar transistor structure using ballast resistor |
US5734193A (en) * | 1994-01-24 | 1998-03-31 | The United States Of America As Represented By The Secretary Of The Air Force | Termal shunt stabilization of multiple part heterojunction bipolar transistors |
US5506764A (en) * | 1994-01-31 | 1996-04-09 | Astec International, Ltd. | Electrical power converter with step-gapped transformer |
US5804867A (en) * | 1996-10-02 | 1998-09-08 | Ericsson Inc. | Thermally balanced radio frequency power transistor |
US6316827B1 (en) * | 1997-09-12 | 2001-11-13 | Nec Corporation | Semiconductor device having improved temperature distribution |
US6034383A (en) * | 1997-11-13 | 2000-03-07 | Northrop Grumman Corporation | High power density microwave HBT with uniform signal distribution |
US6051871A (en) * | 1998-06-30 | 2000-04-18 | The Whitaker Corporation | Heterojunction bipolar transistor having improved heat dissipation |
US6586782B1 (en) * | 1998-07-30 | 2003-07-01 | Skyworks Solutions, Inc. | Transistor layout having a heat dissipative emitter |
US6627925B1 (en) * | 1998-07-30 | 2003-09-30 | Skyworks Solutions, Inc. | Transistor having a novel layout and an emitter having more than one feed point |
US6707341B2 (en) * | 2000-11-30 | 2004-03-16 | Renesas Technology Corp. | Semiconductor device with bipolar transistor device |
US20030038341A1 (en) * | 2001-04-13 | 2003-02-27 | Burhan Bayraktaroglu | Low stress thermal and electrical interconnects for heterojunction bipolar transistors |
US6534857B1 (en) * | 2001-11-02 | 2003-03-18 | Northrop Grumman Corporation | Thermally balanced power transistor |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050216873A1 (en) * | 2004-03-23 | 2005-09-29 | Raminderpal Singh | Method of checking the layout versus the schematic of multi-fingered MOS transistor layouts using a sub-circuit based extraction |
US7139990B2 (en) * | 2004-03-23 | 2006-11-21 | International Business Machines Corporation | Method of checking the layout versus the schematic of multi-fingered MOS transistor layouts using a sub-circuit based extraction |
US7795047B1 (en) * | 2004-12-17 | 2010-09-14 | National Semiconductor Corporation | Current balancing in NPN BJT and BSCR snapback devices |
US7821102B2 (en) | 2006-12-20 | 2010-10-26 | Freescale Semiconductor, Inc. | Power transistor featuring a double-sided feed design and method of making the same |
US20100066454A1 (en) * | 2008-09-18 | 2010-03-18 | Zhenqiang Ma | High-power common-base amplifier employing current source output bias |
US7830208B2 (en) | 2008-09-18 | 2010-11-09 | Wisconsin Alumni Research Foundation | High-power common-base amplifier employing current source output bias |
US9780003B2 (en) * | 2013-11-04 | 2017-10-03 | Taiwan Semiconductor Manufacturing Company, Ltd. | Bipolar junction transistor formed on fin structures |
US9419087B2 (en) * | 2013-11-04 | 2016-08-16 | Taiwan Semiconductor Manufacturing Company, Ltd. | Bipolar junction transistor formed on fin structures |
US20150123246A1 (en) * | 2013-11-04 | 2015-05-07 | Taiwan Semiconductor Manufacturing Company, Ltd. | Bipolar Junction Transistor Formed on Fin Structures |
US10580705B2 (en) | 2013-11-12 | 2020-03-03 | Skyworks Solutions, Inc. | Devices and methods related to radio-frequency switches having improved on-resistance performance |
US20150171108A1 (en) * | 2013-11-12 | 2015-06-18 | Skyworks Solutions, Inc. | Radio-frequency switching devices having improved voltage handling capability |
US11043432B2 (en) * | 2013-11-12 | 2021-06-22 | Skyworks Solutions, Inc. | Radio-frequency switching devices having improved voltage handling capability |
US20220013414A1 (en) * | 2013-11-12 | 2022-01-13 | Skyworks Solutions, Inc. | Methods related to radio-frequency switching devices having improved voltage handling capability |
US20220013415A1 (en) * | 2013-11-12 | 2022-01-13 | Skyworks Solutions, Inc. | Radio-frequency switching devices having improved voltage handling capability |
US11901243B2 (en) * | 2013-11-12 | 2024-02-13 | Skyworks Solutions, Inc. | Methods related to radio-frequency switching devices having improved voltage handling capability |
US12040238B2 (en) * | 2013-11-12 | 2024-07-16 | Skyworks Solutions, Inc. | Radio-frequency switching devices having improved voltage handling capability |
US20160141220A1 (en) * | 2014-11-18 | 2016-05-19 | Sumitomo Electric Industries, Ltd. | Hetero-bipolar transistor and method for producing the same |
US10121761B2 (en) | 2014-11-18 | 2018-11-06 | Sumitomo Electric Industries, Ltd. | Hetero-bipolar transistor and method for producing the same |
US9899502B2 (en) | 2016-04-27 | 2018-02-20 | United Microelectronics Corp. | Bipolar junction transistor layout structure |
CN111384161A (en) * | 2018-12-25 | 2020-07-07 | 株式会社村田制作所 | Bipolar transistor |
US11139390B2 (en) * | 2018-12-25 | 2021-10-05 | Murata Manufacturing Co., Ltd. | Bipolar transistor |
Also Published As
Publication number | Publication date |
---|---|
WO2005052997A2 (en) | 2005-06-09 |
US7705425B2 (en) | 2010-04-27 |
US20060267148A1 (en) | 2006-11-30 |
WO2005052997A3 (en) | 2006-09-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7705425B2 (en) | Solid-state high power device and method | |
US5734193A (en) | Termal shunt stabilization of multiple part heterojunction bipolar transistors | |
US8697532B2 (en) | InP based heterojunction bipolar transistors with emitter-up and emitter-down profiles on a common wafer | |
JPH06342803A (en) | Transistor | |
US7566920B2 (en) | Bipolar transistor and power amplifier | |
US6507047B2 (en) | Power transistors for radio frequencies | |
US20030218185A1 (en) | Semiconductor device, manufacturing thereof and power amplifier module | |
JP2007242727A (en) | Heterojunction bipolar transistor and power amplifier employing it | |
JP2007027269A (en) | Bipolar transistor and power amplifier | |
US6600179B2 (en) | Power amplifier with base and collector straps | |
US6803643B2 (en) | Compact non-linear HBT array | |
US6340618B1 (en) | RF power transistor | |
KR100434659B1 (en) | Collector-up rf power transistor | |
US11158624B1 (en) | Cascode cell | |
SHIMURA et al. | High efficiency AlGaAs/GaAs power HBTs at a low supply voltage for digital cellular phones | |
US20190189580A1 (en) | Thermal shunts and thermal management in monolithic microwave integrated circuits | |
US6503809B2 (en) | Method and system for emitter partitioning for SiGe RF power transistors | |
JP2007036138A (en) | Bipolar transistor and power amplifier | |
JPH04258135A (en) | Power transistor | |
TWI685969B (en) | Bipolar transistor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: WISCONSIN ALUMNI RESEARCH FOUNDATION, WISCONSIN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:MA, ZHENQIANG;JIANG, NINGYUE;REEL/FRAME:015855/0622 Effective date: 20050324 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |