U.S. patent application number 15/374875 was filed with the patent office on 2020-01-02 for lateral semiconductor power devices.
This patent application is currently assigned to MaxPower Semiconductor, Inc.. The applicant listed for this patent is MaxPower Semiconductor Inc.. Invention is credited to Richard A. Blanchard, Mohamed N. Darwish, Hamza Yilmaz, Jun Zeng.
Application Number | 20200006499 15/374875 |
Document ID | / |
Family ID | 59013424 |
Filed Date | 2020-01-02 |
![](/patent/app/20200006499/US20200006499A1-20200102-D00000.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00001.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00002.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00003.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00004.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00005.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00006.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00007.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00008.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00009.png)
![](/patent/app/20200006499/US20200006499A1-20200102-D00010.png)
View All Diagrams
United States Patent
Application |
20200006499 |
Kind Code |
A1 |
Darwish; Mohamed N. ; et
al. |
January 2, 2020 |
Lateral Semiconductor Power Devices
Abstract
Methods and systems for lateral power devices, and methods for
operating them, in which charge balancing is implemented in a new
way. In a first inventive teaching, the lateral conduction path is
laterally flanked by regions of opposite conductivity type which
are self-aligned to isolation trenches which define the surface
geometry of the channel. In a second inventive teaching, which can
be used separately or in synergistic combination with the first
teaching, the drain regions are self-isolated. In a third inventive
teaching, which can be used in synergistic combination with the
first and/or second teachings, the source regions are also isolated
from each other. In a fourth inventive teaching, the lateral
conduction path is also overlain by an additional region of
opposite conductivity type.
Inventors: |
Darwish; Mohamed N.;
(Campbell, CA) ; Zeng; Jun; (Torrance, CA)
; Yilmaz; Hamza; (Gilroy, CA) ; Blanchard; Richard
A.; (Los Altos, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
MaxPower Semiconductor Inc. |
San Jose |
CA |
US |
|
|
Assignee: |
MaxPower Semiconductor,
Inc.
San Jose
CA
|
Family ID: |
59013424 |
Appl. No.: |
15/374875 |
Filed: |
December 9, 2016 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62266536 |
Dec 11, 2015 |
|
|
|
62267784 |
Dec 15, 2015 |
|
|
|
62416645 |
Nov 2, 2016 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H01L 29/0634 20130101;
H01L 21/26586 20130101; H01L 29/7835 20130101; H01L 29/408
20130101; H01L 29/0619 20130101; H01L 29/404 20130101; H01L 29/0623
20130101; H01L 29/7823 20130101; H01L 29/1095 20130101; H01L
29/66681 20130101; H01L 29/402 20130101; H01L 29/1083 20130101;
H01L 21/2236 20130101; H01L 29/0847 20130101; H01L 29/0653
20130101; H01L 29/41758 20130101; H01L 29/42368 20130101; H01L
29/78624 20130101; H01L 29/407 20130101 |
International
Class: |
H01L 29/40 20060101
H01L029/40; H01L 29/66 20060101 H01L029/66; H01L 29/78 20060101
H01L029/78; H01L 29/10 20060101 H01L029/10 |
Claims
1. A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region which is laterally adjacent to the
source region; a first-conductivity-type drain region; a
first-conductivity-type drift region, which is laterally interposed
between the drain region and the body region in an electrical
relation such that, when the gate electrode has a voltage which
inverts part of the body region to form a channel therein, majority
carriers can flow from the source region, through the channel,
through the drift region, to the drain region; wherein the source,
body, drift, and drain regions are all located near a single
surface of a second-conductivity-type semiconductor substrate, to
permit predominantly lateral flow of carriers when a channel is
present; and second-conductivity-type charge balancing regions
which laterally confine a portion of the drift region, and which
are laterally flanked by insulating trenches; wherein, when the
gate is not inverting the body, and reverse bias is present between
the source and drain regions, the drift region, and the charge
balancing regions adjacent to the drift region, will be
substantially depleted before breakdown occurs.
2. The lateral power semiconductor device of claim 1, wherein the
first conductivity type is n-type, and the second conductivity type
is p-type.
3. The lateral power semiconductor device of claim 1, wherein the
semiconductor substrate is monolithic silicon.
4. The lateral power semiconductor device of claim 1, wherein the
semiconductor substrate is a semiconductor-on-insulator
structure.
5. The lateral power semiconductor device of claim 1, wherein the
gate is insulated from the body region by a thin dielectric
layer.
6. The lateral power semiconductor device of claim 1, wherein the
drift region overlaps the entire width of the gate electrode at an
interface between the body region and the drift region.
7. A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located within a
mass of second-conductivity-type semiconductor material;
second-conductivity-type charge balancing regions which laterally
flank the drift region, and which are laterally flanked by
insulating trenches; and first-conductivity-type bottom regions
below the trenches; wherein, when reverse bias is present between
the source and drain regions, the space charge of depleted portions
of the second-conductivity-type charge balancing regions will at
least partially balance the space charge of depleted portions of
the drift region, and the space charge of depleted portions of the
first-conductivity-type bottom regions will at least partially
balance the space charge of depleted portions of the
second-conductivity-type semiconductor mass.
8. The lateral power semiconductor device of claim 7, wherein the
first conductivity type is n-type, and the second conductivity type
is p-type.
9. The lateral power semiconductor device of claim 7, wherein the
semiconductor substrate is monolithic silicon.
10. The lateral power semiconductor device of claim 7, wherein the
charge balancing regions are laterally flanked by insulating
trenches.
11. The lateral power semiconductor device of claim 7, wherein the
semiconductor substrate is a semiconductor-on-insulator
structure.
12. The lateral power semiconductor device of claim 7, wherein the
gate is insulated from the body region by a thin dielectric
layer.
13. The lateral power semiconductor device of claim 7, wherein the
drift region overlaps a total width of the gate electrode at an
interface between the body region and the drift region.
14. A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, laterally through the drift region, to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of second-conductivity-type semiconductor
material; second-conductivity-type charge balancing regions which
laterally flank the drift region, and which are laterally flanked
by insulating trenches; and a second-conductivity-type upper region
which overlies the drift region; wherein, when reverse bias is
present between the source and drain regions, the space charge of
depleted portions of the second-conductivity-type charge balancing
regions and of the upper region will at least partially balance the
space charge of depleted portions of the drift region.
15. The lateral power semiconductor device of claim 14, wherein the
first conductivity type is n-type, and the second conductivity type
is p-type.
16. The lateral power semiconductor device of claim 14, wherein the
semiconductor substrate is monolithic silicon.
17. The lateral power semiconductor device of claim 14, wherein the
charge balancing regions are laterally flanked by insulating
trenches.
18. The lateral power semiconductor device of claim 14, wherein the
semiconductor substrate is a semiconductor-on-insulator
structure.
19. The lateral power semiconductor device of claim 14, wherein the
gate is insulated from the body region by a thin dielectric
layer.
20. The lateral power semiconductor device of claim 14, wherein the
drift region overlaps a total width of the gate electrode at an
interface between the body region and the drift region.
21-51. (canceled)
Description
CROSS-REFERENCE
[0001] Priority is claimed from U.S. provisional applications
62/266,536 filed Dec. 11, 2015, 62/267,784 filed Dec. 15, 2015, and
62/416,645 filed Nov. 2, 2016, all of which are hereby incorporated
by reference.
BACKGROUND
[0002] The present application relates to lateral power
devices.
[0003] Note that the points discussed below may reflect the
hindsight gained from the disclosed inventions, and are not
necessarily admitted to be prior art.
[0004] Power MOSFETs are widely used as switching devices in many
electronic applications. In order to minimize conduction power loss
it is desirable that power MOSFETs for a given breakdown voltage
have low specific on-resistance. Specific on-resistance (Rsp) is
defined as the on-resistance area product (Ron*A). There is a need
for new structures in order to meet the increasing requirement of
reduced Rsp for many new applications. Lateral power MOSFET
structures are required in many applications that require the
monolithic integration of one or more MOSFET in addition to other
circuit components.
Lateral Semiconductor Power Devices
[0005] The present application discloses a number of lateral power
semiconductor structures and methods of fabrication utilizing
charge balance techniques to achieve high breakdown voltage. The
new structures have several advantages over the state of the art
devices in particular having low specific on-resistance Rsp, more
cost effective and compatible with conventional termination
structures such as Field Plates, Guard Rings or Junction
Termination Extension (JTE).
[0006] The present application teaches, among other innovations,
lateral power devices, and methods for operating them, in which
charge balancing is implemented in new ways. In a first inventive
teaching, the lateral conduction path is laterally flanked by
regions of opposite conductivity type which are self-aligned to
isolation trenches. In a second inventive teaching, which can be
used separately or in synergistic combination with the first
teaching, the drain regions are self-isolated. In a third inventive
teaching, which can be used separately or in synergistic
combination with the first and/or second teachings, the source
regions are also isolated from each other. In a fourth inventive
teaching, the lateral conduction path is also overlain by an
additional region of opposite conductivity type.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] The disclosed inventions will be described with reference to
the accompanying drawings, which show important sample embodiments
and which are incorporated in the specification hereof by
reference, wherein:
[0008] FIG. 1A shows one embodiment of a lateral n-channel MOSFET
transistor with the source, gate and drain terminals are accessible
at the device surface. FIGS. 1B, 1C, and 1D show cross-sections of
this device.
[0009] FIG. 1E shows a cross section of an alternative device.
[0010] FIG. 1F shows a cross section of another alternative
device.
[0011] FIGS. 2A and 2B shows another innovative structure.
[0012] FIGS. 3A and 3B show another innovative structure.
[0013] FIG. 4 shows a cross section of another alternative
device.
[0014] FIG. 5 shows integration of multiple power devices on a
single die.
[0015] FIG. 6 shows a cross section of another alternative
device.
[0016] FIGS. 7A and 7B show another innovative structure.
[0017] FIGS. 8A and 8B show another innovative structure, with
isolated source terminals from the substrate terminal.
[0018] FIGS. 9A and 9B show another innovative structure, with
isolated source terminals.
[0019] FIGS. 10A and 10B show another innovative structure, with
isolated source terminals.
[0020] FIGS. 11A and 11B show yet another innovative structure.
[0021] FIGS. 12A-12D show an example of process steps in a sample
implementation.
[0022] FIGS. 12E-12H show an example of process steps in another
sample implementation.
[0023] FIG. 13A shows another innovative structure.
[0024] FIG. 13B shows another innovative structure.
[0025] FIG. 13C shows a cross section of FIGS. 13A and 13B.
[0026] FIG. 14A shows another innovative structure, and FIG. 14B
shows a cross section of it.
[0027] FIG. 14C shows another innovative structure.
[0028] FIG. 15A shows another innovative structure using SOI.
[0029] FIG. 15B shows another innovative structure using a partial
SOI structure.
[0030] FIG. 16A shows another innovative structure, and FIGS. 16B,
16C, and 16D show cross sections of it.
[0031] FIGS. 17A-17D show another innovative structure.
DETAILED DESCRIPTION OF SAMPLE EMBODIMENTS
[0032] The numerous innovative teachings of the present application
will be described with particular reference to presently preferred
embodiments (by way of example, and not of limitation). The present
application describes several inventions, and none of the
statements below should be taken as limiting the claims
generally.
[0033] This application discloses a number of lateral power
semiconductor structures and methods of fabrication utilizing
charge balance (PN Superjunction) and/or permanent charge
techniques to achieve high breakdown voltage. The new structures
have several advantages over the state of the art devices in
particular having low specific on-resistance Rsp, are more cost
effective and are compatible with conventional termination
structures such as Field Plates, Guard Rings or Junction
Termination Extension (JTE).
[0034] FIG. 1A shows one embodiment of the invention of the top
view of a lateral n-channel MOSFET transistor with the source metal
102, gate 106, and drain metal 104, all accessible at the device
surface. Also visible are a field plate 108, the location of the
drift region 122, and two trenches 110 which laterally flank the
drift region.
[0035] FIG. 1B shows the cross section along the line A-A of the
device shown in FIG. 1A. As shown, the device has a P-type
substrate 121, the N-type drift region 122, trenches 110 filled
with a dielectric material such as silicon dioxide SiO.sub.2
(oxide), and a P-type charge-balancing region 112 adjacent the
sidewalls of trenches 110. An overlying dielectric layer 131, e.g.
of oxide, is also shown.
[0036] The N-type drift region 122 is implanted and driven to the
required junction depth before etching the trench. Alternatively,
the drift region 122 can be implanted after trench etching using
tilt angle implant or using techniques such as plasma immersion ion
implantation through the trench sidewalls (and subsequently
diffused). The P-type charge-balancing region 112, in this example,
is implanted (after etching the trench) through the trench
sidewalls (and subsequently diffused). The depth of trenches 110
can be for example 5 to 20 microns. The N-type drift region 122 and
the P-type charge-balancing region 112 are designed, in relation to
their doping concentration, width and thickness, such that these
two layers are both fully depleted at breakdown voltage. This helps
to provide a uniform electric field distribution and higher
breakdown voltage. Having both the N-type drift region 122 and
P-type region 112 implanted results in a more precise charge
balance control which is required for any charge balanced devices.
Using a P- substrate without the need for an epitaxial layer
results in a lower cost process and is compatible with conventional
CMOS processes. Specific examples will be given below.
[0037] The P-substrate doping is chosen to support the required
breakdown voltage, for example 700V.
[0038] FIG. 1B also shows representative depletion boundaries in
the P, N-drift and P-substrate regions. For clarity, this is shown
under less than full voltage in the OFF state; however, it is
preferred that the P-type charge-balancing region 112 and N-type
drift region 122 both be fully depleted at breakdown voltage. In
N-type drift region 122, the depleted portion 122' will have a
positive space charge. Since the mobile electrons have been
expelled, the immobile ionized dopant atoms, e.g. donor species
such as arsenic or phosphorus, will have a net positive charge.
Similarly, in the P-type charge-balancing regions 112, and in the
P-type substrate 121, the depleted portions 112' and 121' will have
a negative space charge. (Since the mobile holes have been
expelled, the immobile ionized acceptor dopant atoms, e.g. boron,
will have a net negative charge.)
[0039] FIG. 1C shows a cross section along the line B-B of the
device shown in FIG. 1A. This Figure shows a planar gate structure,
including an n+ source region 142, P-type body 146, and p+ body
contact region 144. The p+ body contact 144 can be planar or can be
made using a shallow trench contact. The trench 110 is filled with
a dielectric material such as silicon dioxide. A polysilicon gate
electrode 106 with stepped oxide thickness is used, and, in
combination with the source metal 102, forms a field plate. An
important feature of this structure is that the N-Drift region 122
overlaps the total or entire width of the gate electrode 106 at the
P-body 146 and N-drift 122 interface. This provides an efficient
spreading path for the electron current flowing from the channel
into the N-drift layer 122; the current spreads both laterally and
vertically which results in a low Rsp. A deep N+ sinker 151 at the
drain is used to collect the electron current to lower the on
resistance. At the drain contact a polysilicon layer, which is
connected to the drain metal, is used as a field plate to reduce
the electric field.
[0040] FIG. 1D shows a cross section along the line C-C of the
device shown in FIG. 1A. This Figure shows the n+ source 142,
P-type body region 146 (in which a channel is formed by inversion
when the device is turned on), polysilicon gate electrode 106,
N-type drift region 122, and deep N+ sinker region 151 which
connects to drain metal 104. Note that in this embodiment the
charge-balancing region 112 extends below the trench 110, as well
as laterally adjacent to it.
[0041] For the specific example of a 700V voltage rating, the above
relations can be implemented, for example, with the following
dimensions and dopings: [0042] Substrate 121: doped to e.g. about
3E14 (i.e. 3.times.10.sup.14) cm.sup.-3 with boron [0043] Body 146:
doped with boron to a peak concentration of e.g. about
1.times.10.sup.17 cm.sup.-3. [0044] Drift 122: e.g. 7 microns deep,
implanted at e.g. 3-6.times.10.sup.12 cm.sup.-2 with Phosphorus.
[0045] Charge-balancing regions 112: e.g. about 1-2 microns thick,
implanted at e.g. 1-2.times.10.sup.12 cm.sup.-2 with boron.
[0046] FIG. 1E shows a cross section of an alternative structure
along the line C-C of the device shown in FIG. 1A. The structure
shown FIG. 1E is similar to that shown in FIG. 1D except that a
Local Oxidation (LOCOS) process is used to form a partly recessed
field oxide 131' which extends below the silicon surface.
[0047] FIG. 1F shows a cross section of an alternative structure
along the line A-A of the device shown in FIG. 1A. The structure
shown in FIG. 1F is generally somewhat similar to that shown in
FIG. 1D, except that the trenches 110 are not completely filled
with dielectric material and include voids or air gaps 111.
[0048] FIGS. 2A and 2B show another innovative structure, which is
generally somewhat similar to that shown in FIGS. 1B and 1C, but
which also has an n-type bottom region 123 generally underlying the
bottom of the trench. The process window to achieve charge balance
in superjunction devices in production is tight. This effect is
more challenging in lateral superjunction devices due to the charge
sharing effect of the P-substrate since its depletion charge varies
laterally along the N-drift region. Having the n-type trench bottom
layer 123 provides a way to alleviate this three-dimensional
problem by separating the charge at trench sidewalls and trench
bottom. The doping of n-type region 123 is adjusted such that at
breakdown voltage it is fully depleted with uniform electric field
along its length, to maximize the charge balance process control
window.
[0049] FIGS. 3A and 3B show another innovative structure, which is
generally somewhat similar to that shown in FIGS. 1B and 1C, but
which also has a p-type region 113 at the bottom of the trench. The
doping of P-type balancing region 112 is preferably less than that
of the p-type trench bottom region 113. Similarly the trench bottom
p-type 113 is adjusted such that at breakdown voltage it is fully
depleted with a uniform electric field along its length to maximize
the charge balance process control window.
[0050] FIG. 4 shows another embodiment where permanent charge 181
(positive in this example, and created for example at the
silicon-silicon oxide interface, for example by implanting cesium
ions) creates an electron inversion layer in the silicon along the
trench sidewalls, which provides another path to conduct current in
the on-state, and therefore results in further reduction in
R.sub.sp.
[0051] Note that, in this case, the Cs+ ions will increase the
charge imbalance, so the total doping of the p-type charge
balancing region 112 is preferably increased to e.g. about 2E12
cm.sup.-2.
[0052] The devices shown in FIGS. 1-4 have the advantages of
providing low Rsp by using charge balance techniques to permit
increasing the doping concentration of the N-drift. Furthermore,
the need for an expensive epitaxial layer is avoided by the
innovative combination of diffused and implanted N-drift, P-layer,
and N+ sinker. This provides a very cost effective structure and
process.
[0053] In addition, the devices shown in FIGS. 1-4 have
self-isolated drains, so that multiple MOSFET devices can be
integrated on the same substrate with different drain terminals
without the need for additional isolation between them. An example
of this is shown in FIG. 5. This Figure (not to scale) shows (in a
sectional view) integration of two different power MOSFETs on a
single die. The two MOSFETs each have their own separate drain
contacts (Drain1 and Drain2), which can be at different potentials.
The drain junctions are isolated as a result of the junction
between the N+ sinker (N-drift) and the P-type substrate. This
significantly simplifies the integrated structure and reduces its
cost. This is not limited to just two power MOSFETs, but can
include any number of additional power MOSFETs. Preferably each of
the power MOSFETs includes charge balancing regions adjacent to the
lateral channel, as described herein.
[0054] FIG. 6 shows another innovative structure. It is generally
somewhat similar to that shown in FIG. 1D, except that this
structure includes a shield electrode 208 (which is preferably
connected to the source electrode). This shield electrode reduces
the gate-drain capacitance C.sub.gd (or equivalently the gate-drain
charge Q.sub.gd), which is an important requirement for high
switching speed applications.
[0055] FIGS. 7A and 7B show another innovative structure, which is
generally somewhat similar to that shown in FIGS. 1D and 1F, but
which also has a P-type surface region 310. This region too
provides a space charge (under reverse bias) which balances the
space charge of the drift region. In this innovation, a
cross-section of the n-type drift region is surrounded on all four
sides by p-type charge balancing material. This provides additional
charge in the drift region while preserving the charge balance at
breakdown voltage. This allows for increased N-drift doping
concentration without lowering the breakdown voltage. This results
in a further reduction in Rsp. The p-surface layer is connected to
the P-body/p+ source contact at certain locations of the device
that is not shown in the figure. The doping in the region 310 is,
in this example, about 1-2.times.10.sup.12 cm.sup.-2.
[0056] FIGS. 8A and 8B show another innovative structure, which is
generally somewhat similar to that shown in FIGS. 1D and 2B, but
which also includes an additional back N layer 125 below the P-
layer. This structure provides isolation between the source region
and the substrate. One option is to form this back N layer 125 by
diffusion or implantation of the wafer back side.
[0057] The structure of FIGS. 8A-8B permits power MOSFETs to have
separate respective source regions which are isolated from the
bottom backside contact. This is an important advantage.
[0058] FIGS. 9A and 9B show another device structure in which the
N-drift layer 122 also surrounds the p-body region 146. This
structure permits power MOSFETs, with respective source regions
which are isolated from each other, to be combined on a single die.
This is a major advantage.
[0059] FIGS. 10A and 10B show an alternative device structure which
is generally somewhat similar to that shown in FIGS. 8A and 8B, but
with the addition of an N-type buried layer 1002 between the p-type
substrate 121, the p-type well region 120 and p-type trench bottom
layer 113 (like that of FIGS. 3A-3B). This structure advantageously
provides isolated source terminals as discussed above.
[0060] Note that these isolated-source structures can (and still
preferably do) incorporate the self-isolated drains of FIG. 5. This
class of implementations provides further synergistic advantages:
by having multiple fully isolated power channels on a single die,
it is possible to integrate multiple power devices on a single die.
This allows system designers to integrate more functionality into a
one-chip solution.
[0061] FIGS. 11A and 11B show another innovative structure, which
is generally somewhat similar to that shown in FIGS. 1C and 1D, but
which is implemented in Silicon-On-Insulator (SOI) material. In
such structures an SOI barrier dielectric 1102 separates the active
device areas from the SOI substrate 1121. In this example the
substrate 1121 is p-type silicon, as in the embodiments described
above, but alternatives are also possible. Dielectric-filled
trenches 1111 provide full dielectric isolation if desired.
[0062] FIGS. 12A-12D show the one example implementation of the
steps of forming the N-drift region (using adequate drive-in time
and temperature after implantation or other dopant introduction, to
achieve the desired diffusion length root-Dt), and then etching the
trench, forming the P-layer, n-type trench bottom region using zero
angle tilt implant and filling the trench.
[0063] FIGS. 12E-12H show an alternative method where the N-drift
layer is formed by tilted angle implantation or using techniques
such as plasma immersion ion implantation through the trench
sidewalls (and subsequently diffused) after etching the trench. The
latter method allows the use of deep trenches which results in
further Rsp reduction.
[0064] FIG. 13A shows another embodiment which is generally
somewhat similar to that shown in FIG. 1A, but with trench
sidewalls lined by a dielectric material such as oxide and filled
with a Field Plate 1310. The field plate, in this example, is made
of polysilicon which is heavily doped with n-type doping such as
phosphorus. The field plate 1310 is preferably biased to the source
or gate potential, or alternatively can be floating.
[0065] The oxide layer width along the trench sidewalls can be
thinner than that at the drain end of the trench. This can be
achieved, for example, by filling the trench with a thick oxide
layer, and then etching the oxide layer in certain regions using
photoresist, followed by growing or depositing a thinner oxide
layer. This step is preferably followed by polysilicon layer
deposition.
[0066] FIG. 13B shows another innovative structure, which is
generally somewhat similar to that shown in FIG. 13A, but which has
multiple polysilicon regions separated by oxide layers. This
provides, in effect, a segmented field plate structure. The
polysilicon regions can be biased or floating.
[0067] FIG. 13C shows a cross section along line A-A of both FIGS.
13A and 13B.
[0068] FIG. 14A shows an example of implementation of
metallization. This is a top view of a lateral n-channel transistor
with the drain fingertip, source, gate and drain terminals. The
cross section along the line Y-Y can be for example that shown in
any of the FIGS. 1D, 6, 7A, 8A and 11B.
[0069] FIG. 14B is a cross section along the line X-X of the device
shown in FIG. 14A showing one implementation using a field plate
1402 extending toward the edge over increasing dielectric thickness
in a stair step fashion; this permits the field plate to smooth the
potential gradients in the semiconductor material nearest the
junction, while avoiding any sharp change in potential at the outer
edge of the field plate.
[0070] It should be noted that the surface of the device can be
covered by an additional suitable dielectric material such as
silicon nitride or similar passivation layers. Such passivation
layer(s) were not shown in the previous figures, for
simplicity.
[0071] FIG. 14C shows another innovative structure, in which
floating N-type guard rings 1412 are used in the termination. It
should be noted that the surface of the device can be covered by an
additional suitable dielectric material such as silicon nitride or
similar passivation layers. Such passivation layer was not shown in
the previous figures for simplicity, but would commonly be
included.
[0072] FIG. 15A shows another example, where the device termination
uses an SOI structure.
[0073] FIG. 15B shows another embodiment of a device termination
using a partial SOI structure for improved thermal dissipation.
[0074] FIG. 16A shows another innovative structure. This Figure
shows a top view of a lateral IGBT, in which the emitter, gate and
collector terminals (1602, 1604, and 1606 respectively) are all
accessible at the device surface.
[0075] FIG. 16B shows a cross section along the line A-A of the
device shown in FIG. 16A.
[0076] FIG. 16C shows a cross section along the line B-B of the
device shown in FIG. 16A. The structure of FIG. 16C is generally
somewhat similar to that shown in FIG. 1C, except that the deep N+
sinker at the drain is replaced by an N-buffer layer and a P+ layer
at the collector contact.
[0077] FIG. 16D shows a cross section along the line C-C of the
device shown in FIG. 10A. The structure in this Figure is generally
somewhat similar to that shown in FIG. 1D, except that the deep N+
sinker at the drain is replaced by an N-buffer layer 1632 and a P+
layer 1634 at the collector contact.
[0078] FIGS. 17A-D show another embodiment of a lateral IGBT, which
is generally somewhat similar to that shown in FIGS. 16A-D, except
for the use of partial SOI. FIG. 17A shows a cross section along
line A-A of the top view shown in FIG. 16A, and FIG. 17B shows a
cross section along line B-B.
[0079] FIG. 17C shows a cross section of an alternative embodiment
along the line B-B.
[0080] FIG. 17D shows a cross section along line C-C of the top
view shown in FIG. 16A.
Advantages
[0081] The disclosed innovations, in various embodiments, provide
one or more of at least the following advantages. However, not all
of these advantages result from every one of the innovations
disclosed, and this list of advantages does not limit the various
claimed inventions. [0082] Power semiconductor devices with lower
on-resistance: [0083] Power semiconductor devices with lower cost;
[0084] Power semiconductor devices which do not require epitaxial
material; [0085] Power superjunction semiconductor devices with
better charge balance control. [0086] Integrable power
semiconductor devices; [0087] Power semiconductor devices with more
ruggedness; [0088] Power semiconductor devices with higher
breakdown voltage; [0089] Dice with multiple independent power
drivers; and [0090] Dice with multiple independent power drivers,
plus small-signal circuits, all integrated together.
[0091] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region which is laterally adjacent to the
source region; a first-conductivity-type drain region; a
first-conductivity-type drift region, which is laterally interposed
between the drain region and the body region in an electrical
relation such that, when the gate electrode has a voltage which
inverts part of the body region to form a channel therein, majority
carriers can flow from the source region, through the channel,
through the drift region, to the drain region; wherein the source,
body, drift, and drain regions are all located near a single
surface of a second-conductivity-type semiconductor substrate, to
permit predominantly lateral flow of carriers when a channel is
present; and second-conductivity-type charge balancing regions
which laterally confine a portion of the drift region and which are
laterally flanked by insulating trenches; wherein, when the gate is
not inverting the body, and reverse bias is present between the
source and drain regions, the drift region, and the charge
balancing regions adjacent to the drift region, will be
substantially depleted before breakdown occurs.
[0092] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located within a
mass of second-conductivity-type semiconductor material;
second-conductivity-type charge balancing regions which laterally
flank the drift region, and which are laterally flanked by
insulating trenches; and first-conductivity-type bottom regions
below the trenches; wherein, when reverse bias is present between
the source and drain regions, the space charge of depleted portions
of the second-conductivity-type charge balancing regions will at
least partially balance the space charge of depleted portions of
the drift region, and the space charge of depleted portions of the
first-conductivity-type bottom regions will at least partially
balance the space charge of depleted portions of the
second-conductivity-type semiconductor mass.
[0093] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, laterally through the drift region, to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of second-conductivity-type semiconductor
material; second-conductivity-type charge balancing regions which
laterally flank the drift region, and which are laterally flanked
by insulating trenches; and a second-conductivity-type upper region
which overlies the drift region; wherein, when reverse bias is
present between the source and drain regions, the space charge of
depleted portions of the second-conductivity-type charge balancing
regions and of the upper region will at least partially balance the
space charge of depleted portions of the drift region.
[0094] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device, comprising: an
n-type source region; a p-type body region, and a gate electrode
which is capacitively coupled to a portion of the body region; an
n-type drain region; an n-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
positive voltage which inverts part of the body region to form a
channel therein, electrons can flow from the source region, through
the channel, laterally through the drift region, and to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of p-type semiconductor material; and p-type
charge balancing regions which laterally flank the drift region,
and which are laterally flanked by insulating trenches which
include immobile positive electrostatic charge; and wherein, when
reverse bias is present between the source and drain regions, the
negative space charge of depleted portions of the p-type charge
balancing regions, will at least partially balance the positive
space charge of depleted portions of the drift region in
combination with the fixed charge in the trenches
[0095] According to some but not necessarily all embodiments, there
is provided: A power semiconductor device, comprising, on a single
die, multiple lateral power transistors which each include: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, laterally through the drift region, to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of second-conductivity-type semiconductor
material; and second-conductivity-type charge balancing regions
which laterally flank the drift region, and which are laterally
flanked by insulating trenches; wherein, when reverse bias is
present between the source and drain regions, the space charge of
depleted portions of the second-conductivity-type charge balancing
regions and of the upper region will at least partially balance the
space charge of depleted portions of the drift region; and wherein
the respective drain regions of the multiple transistors are
isolated from each other by intervening portions of the
second-conductivity-type mass of semiconductor material.
[0096] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device, comprising: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located within a
second-conductivity type well, which overlies a
first-conductivity-type buried layer, which in turn overlies a mass
of second-conductivity-type semiconductor material;
second-conductivity-type charge balancing regions which laterally
flank the drift region, and which are laterally flanked by
insulating trenches; and wherein, when reverse bias is present
between the source and drain regions, the space charge of depleted
portions of the second-conductivity-type charge balancing regions
will at least partially balance the space charge of depleted
portions of the drift region.
[0097] According to some but not necessarily all embodiments, there
is provided: A lateral power semiconductor device with bipolar
conduction, comprising: a first-conductivity-type emitter region; a
second-conductivity-type body region, and a gate electrode which is
capacitively coupled to a portion of the body region; a
second-conductivity-type collector region; a
first-conductivity-type drift region, which is laterally interposed
between the collector region and the body region in an electrical
relation such that, when the gate electrode has a voltage which
inverts part of the body region to form a channel therein, majority
carriers can flow from the emitter region, through the channel,
through the drift region, to the collector region; and minority
carriers are injected from the collector region through the drift
region to the emitter region; and second-conductivity-type charge
balancing regions which laterally flank the drift region, and which
are laterally flanked by insulating trenches; wherein, when reverse
bias is present between the emitter and collector regions, the
space charge of depleted portions of the second-conductivity-type
charge balancing regions will at least partially balance the space
charge of depleted portions of the drift region.
[0098] According to some but not necessarily all embodiments, there
is provided: A method for switching electrical power, comprising:
driving the gate of a lateral power semiconductor device,
comprising: a first-conductivity-type source region; a
second-conductivity-type body region, and a gate electrode which is
capacitively coupled to a portion of the body region which is
laterally adjacent to the source region; a first-conductivity-type
drain region; a first-conductivity-type drift region, which is
laterally interposed between the drain region and the body region
in an electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located near a
single surface of a second-conductivity-type semiconductor
substrate, to permit predominantly lateral flow of carriers when a
channel is present; and second-conductivity-type charge balancing
regions which laterally confine a portion of the drift region;
wherein, when the gate is not inverting the body, and reverse bias
is present between the source and drain regions, the drift region,
and the charge balancing regions adjacent to the drift region, will
be substantially depleted before breakdown occurs.
[0099] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device which comprises: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located within a
mass of second-conductivity-type semiconductor material;
second-conductivity-type charge balancing regions which laterally
flank the drift region, and which are laterally flanked by
insulating trenches; and first-conductivity-type bottom regions
below the trenches; wherein, when reverse bias is present between
the source and drain regions, the space charge of depleted portions
of the second-conductivity-type charge balancing regions will at
least partially balance the space charge of depleted portions of
the drift region, and the space charge of depleted portions of the
first-conductivity-type bottom regions will at least partially
balance the space charge of depleted portions of the
second-conductivity-type semiconductor mass.
[0100] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device which comprises: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, laterally through the drift region, to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of second-conductivity-type semiconductor
material; second-conductivity-type charge balancing regions which
laterally flank the drift region, and which are laterally flanked
by insulating trenches; and a second-conductivity-type upper region
which overlies the drift region; wherein, when reverse bias is
present between the source and drain regions, the space charge of
depleted portions of the second-conductivity-type charge balancing
regions and of the upper region will at least partially balance the
space charge of depleted portions of the drift region.
[0101] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device which comprises: an
n-type source region; a p-type body region, and a gate electrode
which is capacitively coupled to a portion of the body region; an
n-type drain region; an n-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
positive voltage which inverts part of the body region to form a
channel therein, electrons can flow from the source region, through
the channel, laterally through the drift region, and to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of p-type semiconductor material; and p-type
charge balancing regions which laterally flank the drift region,
and which are laterally flanked by insulating trenches which
include immobile positive electrostatic charge; and wherein, when
reverse bias is present between the source and drain regions, the
negative space charge of depleted portions of the p-type charge
balancing regions, will at least partially balance the positive
space charge of depleted portions of the drift region in
combination with the fixed charge in the trenches
[0102] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device which comprises, on a
single die, multiple lateral power transistors which each include:
a first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, laterally through the drift region, to the drain
region; wherein the source, body, drift, and drain regions are all
located within a mass of second-conductivity-type semiconductor
material; and second-conductivity-type charge balancing regions
which laterally flank the drift region, and which are laterally
flanked by insulating trenches; wherein, when reverse bias is
present between the source and drain regions, the space charge of
depleted portions of the second-conductivity-type charge balancing
regions and of the upper region will at least partially balance the
space charge of depleted portions of the drift region; and wherein
the respective drain regions of the multiple transistors are
isolated from each other by intervening portions of the
second-conductivity-type mass of semiconductor material.
[0103] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device which comprises: a
first-conductivity-type source region; a second-conductivity-type
body region, and a gate electrode which is capacitively coupled to
a portion of the body region; a first-conductivity-type drain
region; a first-conductivity-type drift region, which is laterally
interposed between the drain region and the body region in an
electrical relation such that, when the gate electrode has a
voltage which inverts part of the body region to form a channel
therein, majority carriers can flow from the source region, through
the channel, through the drift region, to the drain region; wherein
the source, body, drift, and drain regions are all located within a
second-conductivity type well, which overlies a
first-conductivity-type buried layer, which in turn overlies a mass
of second-conductivity-type semiconductor material;
second-conductivity-type charge balancing regions which laterally
flank the drift region, and which are laterally flanked by
insulating trenches; and wherein, when reverse bias is present
between the source and drain regions, the space charge of depleted
portions of the second-conductivity-type charge balancing regions
will at least partially balance the space charge of depleted
portions of the drift region.
[0104] According to some but not necessarily all embodiments, there
is provided a method for switching electrical power by driving the
gate of a lateral power semiconductor device with bipolar
conduction which comprises: a first-conductivity-type emitter
region; a second-conductivity-type body region, and a gate
electrode which is capacitively coupled to a portion of the body
region; a second-conductivity-type collector region; a
first-conductivity-type drift region, which is laterally interposed
between the collector region and the body region in an electrical
relation such that, when the gate electrode has a voltage which
inverts part of the body region to form a channel therein, majority
carriers can flow from the emitter region, through the channel,
through the drift region, to the collector region; and minority
carriers are injected from the collector region through the drift
region to the emitter region; and second-conductivity-type charge
balancing regions which laterally flank the drift region, and which
are laterally flanked by insulating trenches; wherein, when reverse
bias is present between the emitter and collector regions, the
space charge of depleted portions of the second-conductivity-type
charge balancing regions will at least partially balance the space
charge of depleted portions of the drift region.
[0105] According to some but not necessarily all embodiments, there
is provided methods and systems for lateral power devices, and
methods for operating them, in which charge balancing is
implemented in a new way. In a first inventive teaching, the
lateral conduction path is laterally flanked by regions of opposite
conductivity type which are self-aligned to isolation trenches
which define the surface geometry of the drift region. In a second
inventive teaching, which can be used separately or in synergistic
combination with the first teaching, the drain regions are
self-isolated. In a third inventive teaching, which can be used in
synergistic combination with the first and/or second teachings, the
source regions are also isolated from each other. In a fourth
inventive teaching, the lateral conduction path is also overlain by
an additional region of opposite conductivity type.
Modifications and Variations
[0106] As will be recognized by those skilled in the art, the
innovative concepts described in the present application can be
modified and varied over a tremendous range of applications, and
accordingly the scope of patented subject matter is not limited by
any of the specific exemplary teachings given. It is intended to
embrace all such alternatives, modifications and variations that
fall within the spirit and broad scope of the appended claims.
[0107] For example, it should be noted that the deep N+ sinker can
be deeper or shallower than N-Drift layer. Furthermore, the N+
sinker can be formed by using a deep trench filled with conducting
material such as tungsten and surrounded by a n+ layer formed by an
phosphorus or arsenic implant.
[0108] For example, it should also be noted that conventional
techniques such as Local Field Oxidation (LOCOS) can be used to
form the thick field oxide.
[0109] The doping levels needed to achieve high breakdown and
low-resistance are governed by the well known charge balance
condition. The specific electrical characteristics of devices
fabricated using the methods described in this disclosure depend on
a number of factors including the thickness of the layers, their
doping levels, the materials being used, the geometry of the
layout, etc. One of ordinary skill in the art will realize that
simulation, experimentation, or a combination thereof can be used
to determine the design parameters needed to operate as
intended.
[0110] While the figures shown in this disclosure are qualitatively
correct, the geometries used in practice may differ and should not
be considered a limitation in anyway. It is understood by those of
ordinary skill in the art that the actual cell layout will vary
depending on the specifics of the implementation and any depictions
illustrated herein should not be considered a limitation in any
way.
[0111] While only n-channel MOSFETs are shown here, p-channel
MOSFETs are realizable with this invention simply by changing the
polarity of the permanent charge (if any) and swapping n-type and
p-type regions in any of the figures. This is well known by those
of ordinary skill in the art.
[0112] Additionally while only MOSFETs and IGBTs are shown, many
other device structures are implementable using this invention
including diodes, thyristors, JFETs, BJTs, and the like.
[0113] It should be noted that the deep N+ sinker can be deeper or
shallower than N-Drift layer. Furthermore, the N+ sinker can
alternatively be formed by using a deep trench filled with
conducting material such as tungsten and surrounded by a n+ layer
formed by an phosphorus or arsenic implant.
[0114] In some embodiments, the fraction of the width used for the
balancing regions 112 depends on the ratio of doping concentration
between balancing regions 112 and the drift region 122.
[0115] It should also be understood that numerous combinations of
the above embodiments can be realized.
[0116] Those of ordinary skill in the relevant fields of art will
recognize that other inventive concepts may also be directly or
inferentially disclosed in the foregoing. NO inventions are
disclaimed.
[0117] None of the description in the present application should be
read as implying that any particular element, step, or function is
an essential element which must be included in the claim scope: THE
SCOPE OF PATENTED SUBJECT MATTER IS DEFINED ONLY BY THE ALLOWED
CLAIMS. Moreover, none of these claims are intended to invoke
paragraph six of 35 USC section 112 unless the exact words "means
for" are followed by a participle.
[0118] The claims as filed are intended to be as comprehensive as
possible, and NO subject matter is intentionally relinquished,
dedicated, or abandoned.
* * * * *