U.S. patent application number 11/388755 was filed with the patent office on 2006-09-28 for flip chip interconnection having narrow interconnection sites on the substrate.
This patent application is currently assigned to STATS ChipPAC, Ltd.. Invention is credited to Rajendra D. Pendse.
Application Number | 20060216860 11/388755 |
Document ID | / |
Family ID | 37053986 |
Filed Date | 2006-09-28 |
United States Patent
Application |
20060216860 |
Kind Code |
A1 |
Pendse; Rajendra D. |
September 28, 2006 |
Flip chip interconnection having narrow interconnection sites on
the substrate
Abstract
A flip chip interconnect of a die on a substrate is made by
mating the interconnect bump onto a narrow interconnect pad on a
lead or trace, rather than onto a capture pad. The width of the
narrow interconnect pad is less than a base diameter of bumps on
the die to be attached. Also, a flip chip package includes a die
having solder bumps attached to interconnect pads in an active
surface, and a substrate having narrow interconnect pads on
electrically conductive traces in a die attach surface, in which
the bumps are mated onto the narrow pads on the traces.
Inventors: |
Pendse; Rajendra D.;
(Fremont, CA) |
Correspondence
Address: |
HAYNES BEFFEL & WOLFELD LLP
P O BOX 366
HALF MOON BAY
CA
94019
US
|
Assignee: |
STATS ChipPAC, Ltd.
Singapore
SG
569059
|
Family ID: |
37053986 |
Appl. No.: |
11/388755 |
Filed: |
March 24, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60665208 |
Mar 25, 2005 |
|
|
|
Current U.S.
Class: |
438/108 ;
257/678; 257/E21.503; 257/E23.068; 257/E23.07; 438/106 |
Current CPC
Class: |
H01L 2924/3011 20130101;
H01L 2224/13111 20130101; H01L 2224/13144 20130101; H01L 2924/15787
20130101; Y02P 70/611 20151101; H01L 21/563 20130101; H01L
2224/81136 20130101; H01L 2924/01322 20130101; H01L 2224/16225
20130101; H01L 2924/00013 20130101; H01L 2224/73204 20130101; H01L
2224/81204 20130101; H01L 2924/01082 20130101; H05K 2201/09427
20130101; H01L 2924/01079 20130101; H01L 2224/0554 20130101; H01L
2924/01006 20130101; H01L 2224/05571 20130101; H01L 2224/32225
20130101; H01L 24/81 20130101; H01L 2924/14 20130101; H01L 23/49838
20130101; H01L 2224/83856 20130101; H01L 2224/92125 20130101; H01L
2924/00014 20130101; H01L 2224/73203 20130101; H01L 2224/81815
20130101; H01L 2224/83192 20130101; H01L 2924/01078 20130101; H05K
2201/0989 20130101; H01L 2924/01033 20130101; H01L 2924/014
20130101; H05K 2201/10674 20130101; H05K 3/3452 20130101; Y02P
70/50 20151101; H01L 2924/01005 20130101; H01L 2924/01075 20130101;
H01L 2224/0557 20130101; H01L 2224/16237 20130101; H01L 24/29
20130101; H01L 2224/83102 20130101; H01L 2224/13013 20130101; H05K
1/111 20130101; H01L 2224/1607 20130101; H01L 2224/81385 20130101;
H01L 23/49811 20130101; H01L 2224/16238 20130101; H01L 2224/81193
20130101; H05K 2201/09727 20130101; H01L 2224/05573 20130101; H01L
2224/8121 20130101; H01L 2224/81211 20130101; H01L 2924/01046
20130101; H01L 2924/01074 20130101; H01L 2224/13144 20130101; H01L
2924/00014 20130101; H01L 2924/00013 20130101; H01L 2224/13099
20130101; H01L 2224/16225 20130101; H01L 2224/13144 20130101; H01L
2924/00 20130101; H01L 2224/16225 20130101; H01L 2224/13111
20130101; H01L 2924/00 20130101; H01L 2224/73204 20130101; H01L
2224/16225 20130101; H01L 2224/32225 20130101; H01L 2924/00
20130101; H01L 2224/83192 20130101; H01L 2224/32225 20130101; H01L
2924/00 20130101; H01L 2224/92125 20130101; H01L 2224/73204
20130101; H01L 2224/16225 20130101; H01L 2224/32225 20130101; H01L
2924/00 20130101; H01L 2224/83192 20130101; H01L 2224/73204
20130101; H01L 2224/16225 20130101; H01L 2224/32225 20130101; H01L
2924/00 20130101; H01L 2924/15787 20130101; H01L 2924/00 20130101;
H01L 2924/00014 20130101; H01L 2224/05599 20130101; H01L 2924/00014
20130101; H01L 2224/0555 20130101; H01L 2924/00014 20130101; H01L
2224/0556 20130101 |
Class at
Publication: |
438/108 ;
438/106; 257/678 |
International
Class: |
H01L 21/00 20060101
H01L021/00; H01L 23/02 20060101 H01L023/02 |
Claims
1. A flip chip interconnection, comprising a solder bump attached
to an interconnect pad on a die and mated onto a pad at an
interconnect site on a substrate, wherein a width of the pad at the
interconnect site is less than a nominal width of the contact
interface between the bump and the die pad.
2. The flip chip interconnection of claim 1 wherein the width of
the pad at the interconnect site is at least about 20% as great as
the nominal width of the contact interface between the bump and the
die pad.
3. The flip chip interconnection of claim 1 wherein the width of
the pad at the interconnect site is at least about 25% as great as
the nominal width of the contact interface between the bump and the
die pad.
4. The flip chip interconnection of claim 2 wherein the width of
the pad at the interconnect site is less than about 80% as great as
the nominal width of the contact interface between the bump and the
die pad.
5. The flip chip interconnection of claim 2 wherein the width of
the pad at the interconnect site is less than about 60% as great as
the nominal width of the contact interface between the bump and the
die pad.
6. The flip chip interconnection of claim 1 wherein the width of
the pad at the interconnect site is greater than about 120% of a
trace design rule width.
7. A flip chip interconnection, comprising a solder bump attached
to an interconnect pad on a die and mated onto a pad at an
interconnect site on a substrate, wherein a width of the pad at the
interconnect site is less than a design width of the contact
interface between the bump and the die pad.
8. The flip chip interconnection of claim 7 wherein the width of
the pad at the interconnect site is at least about 20% as great as
the design width of the contact interface between the bump and the
die pad.
9. The flip chip interconnection of claim 7 wherein the width of
the pad at the interconnect site is at least about 25% as great as
the design width of the contact interface between the bump and the
die pad.
10. The flip chip interconnection of claim 8 wherein the width of
the pad at the interconnect site is less than about 80% as great as
the design width of the contact interface between the bump and the
die pad.
11. The flip chip interconnection of claim 8 wherein the width of
the pad at the interconnect site is less than about 60% as great as
the design width of the contact interface between the bump and the
die pad.
12. The flip chip interconnection of claim 1 wherein the width of
the pad at the interconnect site is greater than about 120% of a
nominal trace width.
13. A flip chip semiconductor package comprising a plurality of
solder bumps each attached to an interconnect pad on a die and
mated onto a corresponding a pad at an interconnect site on a
substrate, wherein a width of the pad at the interconnect site is
less than a nominal width of the contact interface between the bump
and the die pad.
14. The flip chip semiconductor package of claim 13 wherein the
width of the pad at the interconnect site is at least about 20% as
great as the nominal width of the contact interface between the
bump and the die pad.
15. The flip chip semiconductor package of claim 13 wherein the
width of the pad at the interconnect site is at least about 25% as
great as the nominal width of the contact interface between the
bump and the die pad.
16. The flip chip semiconductor package of claim 14 wherein the
width of the pad at the interconnect site is less than about 80% as
great as the nominal width of the contact interface between the
bump and the die pad.
17. The flip chip semiconductor package of claim 14 wherein the
width of the pad at the interconnect site is less than about 60% as
great as the nominal width of the contact interface between the
bump and the die pad.
18. The flip chip semiconductor package of claim 14 wherein the
width of the pad at the interconnect site is greater than about
120% of a nominal trace width.
19. A flip chip semiconductor package comprising a plurality of
solder bumps each attached to an interconnect pad on a die and
mated onto a corresponding a pad at an interconnect site on a
substrate, wherein a width of the pad at the interconnect site is
less than a design width of the contact interface between the bump
and the die pad.
20. The flip chip semiconductor package of claim 19 wherein the
width of the pad at the interconnect site is at least about 20% as
great as the design width of the contact interface between the bump
and the die pad.
21. The flip chip semiconductor package of claim 19 wherein the
width of the pad at the interconnect site is at least about 25% as
great as the design width of the contact interface between the bump
and the die pad.
22. The flip chip semiconductor package of claim 20 wherein the
width of the pad at the interconnect site is less than about 80% as
great as the design width of the contact interface between the bump
and the die pad.
23. The flip chip semiconductor package of claim 20 wherein the
width of the pad at the interconnect site is less than about 60% as
great as the design width of the contact interface between the bump
and the die pad.
24. The flip chip semiconductor package of claim 20 wherein the
width of the pad at the interconnect site is greater than about
120% of a nominal trace width.
25. The flip chip interconnection of claim 1, the interconnection
being non-solder mask defined.
26. The flip chip interconnection of claim 1, the interconnection
being solder mask defined.
27. A substrate for flip chip interconnection of a die, the die
having a bumps each attached on a die pad, the substrate comprising
a patterned metal layer having pads at interconnect sites, wherein
a width of a said pad at a said interconnect site is less than a
nominal width of a contact interface between a said bump and the
die pad.
28. The substrate of claim 27 wherein the width of the pad at the
interconnect site is at least about 20% as great as the nominal
width of the contact interface between the bump and the die
pad.
29. The substrate of claim 27 wherein the width of the pad at the
interconnect site is at least about 25% as great as the nominal
width of the contact interface between the bump and the die
pad.
30. The substrate of claim 28 wherein the width of the pad at the
interconnect site is less than about 80% as great as the nominal
width of the contact interface between the bump and the die
pad.
31. The substrate of claim 28 wherein the width of the pad at the
interconnect site is less than about 60% as great as the nominal
width of the contact interface between the bump and the die
pad.
32. The substrate of claim 28 wherein the width of the pad at the
interconnect site is greater than about 120% of a trace design rule
width.
33. A method for forming a flip chip interconnection, comprising
providing a substrate as recited in claim 27; supporting the
substrate and the die; dispensing a quantity of a curable adhesive
on the substrate; positioning the die with the active side of the
die toward the die attach surface of the substrate, and aligning
the die and substrate and moving one toward the other so that the
bumps contact the corresponding narrow interconnection pads on the
substrate; applying a force to press the bumps onto the mating
narrow pads, sufficient to displace the adhesive from between the
bump and the mating narrow pad; at least partially curing the
adhesive; melting and then re-solidifying the solder, forming a
metallurgical interconnection between the bump and the pad.
34. The method of claim 33 wherein dispensing a quantity of a
curable adhesive on the substrate covers at least the
interconnection pads on the substrate.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority from U.S. Provisional
Application No. 60/665,208, filed Mar. 25, 2005, titled "Flip chip
interconnection having narrow interconnection sites on the
substrate", which is hereby incorporated herein by reference.
[0002] This application is related to U.S. application Ser. No.
10/985,654, filed Nov. 10, 2004, titled "Bump-on-lead flip chip
interconnection".
BACKGROUND
[0003] This invention relates to semiconductor packaging and,
particularly, to flip chip interconnection.
[0004] Flip chip packages include a semiconductor die mounted onto
a package substrate with the active side of the die facing the
substrate. The substrate is made up of a dielectric layer and at
least one metal layer, patterned to provide substrate circuitry,
which includes among other features traces ("leads") leading to
interconnect pads. The metal layer may be patterned by, for
example, a mask-and etch process. Conventionally, interconnection
of the circuitry in the die with circuitry in the substrate is made
by way of bumps which are attached to an array of interconnect pads
on the die, and bonded to a corresponding (complementary) array of
interconnect pads (often referred to as "capture pads") on the
substrate. The capture pads are typically much wider than the
leads, and can be as wide as, for example, about 2 to 4 times the
nominal or design width of the leads.
[0005] The areal density of electronic features on integrated
circuits has increased enormously, and chips having a greater
density of circuit features also may have a greater density of
sites ("die pads") for interconnection with the circuitry on a
package substrate.
[0006] The package is connected to underlying circuitry, such as a
printed circuit board (e.g., a "motherboard"), in the device in
which the package is employed, by way of second level interconnects
(e.g., pins, secondary interconnect solder balls) between the
package and the underlying circuit. The second level interconnects
have a greater pitch than the flip chip interconnects, and so the
routing on the substrate conventionally "fans out". Significant
technological advances in patterning the metal layer on the
substrate have enabled construction of fine lines and spaces; but
in the conventional arrangement space between adjacent pads limits
the number of traces than can escape from the more inward capture
pads in the array, and the fan out routing between the capture pads
beneath the die and the external pins of the package is
conventionally formed on multiple metal layers within the package
substrate. For a complex interconnect array, substrates having
multiple layers may be required to achieve routing between the die
pads and the second level interconnects on the package.
[0007] Multiple layer substrates are expensive, and in conventional
flip chip constructs the substrate alone typically accounts for
more than half the package cost (about 60% in some typical
instances). The high cost of multilayer substrates has been a
factor in limiting proliferation of flip chip technology in
mainstream products.
[0008] In conventional flip chip constructs the escape routing
pattern typically introduces additional electrical parasitics,
because the routing includes short runs of unshielded wiring and
vias between wiring layers in the signal transmission path.
Electrical parasitics can significantly limit package
performance.
SUMMARY
[0009] According to the invention flip chip interconnect is
accomplished by connecting the interconnect bump directly onto a
narrow interconnection pad, or narrow pad, rather than onto a
conventional capture pad. The width of the narrow pad according to
the invention is selected according to the base diameter of the
interconnect bump that is to be connected onto the narrow pad.
Particularly, the width of the narrow pad is less (such as in a
range about 20% to about 80%) than the base diameter of the
interconnect bump. The invention provides more efficient routing of
traces on the substrate. Particularly, the signal routing can be
formed entirely in a single metal layer of the substrate. This
reduces the number of layers in the substrate, and forming the
signal traces in a single layer also permits relaxation of some of
the via, line and space design rules that the substrate must meet.
This simplification of the substrate greatly reduces the overall
cost of the flip chip package. The bump-on-narrow-pad architecture
also helps eliminate such features as vias and "stubs" from the
substrate design, and enables a microstrip controlled impedance
electrical environment for signal transmission, thereby greatly
improving performance.
[0010] In one general aspect the invention features a flip chip
interconnection having solder bumps attached to interconnect pads
on a die and mated onto corresponding narrow interconnection pads
on a substrate.
[0011] In another general aspect the invention features a flip chip
package including a die having solder bumps attached to
interconnect pads in an active surface, and a substrate having
narrow interconnection pads in a die attach surface, in which the
bumps are mated onto the narrow pads.
[0012] In some embodiments the bump-on-narrow-pad interconnection
is formed according to methods of the invention without use of a
solder mask to confine the molten solder during a re-melt stage in
the process. Avoiding the need for a solder mask allows for finer
interconnection geometry.
[0013] In some embodiments the substrate is further provided with a
solder mask having openings over the narrow interconnection pads.
In some embodiments the substrate is further provided with solder
paste on the narrow interconnection pads.
[0014] In another general aspect the invention features a method
for forming flip chip interconnection, by providing a substrate
having narrow interconnection pads formed in a die attach surface
and a die having bumps attached to interconnect pads in an active
surface; supporting the substrate and the die; dispensing a
quantity of a curable adhesive on the substrate (covering at least
the narrow interconnection pads) or on the active side of the die
(covering at least the bumps); positioning the die with the active
side of the die toward the die attach surface of the substrate, and
aligning the die and substrate and moving one toward the other so
that the bumps contact the corresponding narrow interconnection
pads on the substrate; applying a force to press the bumps onto the
mating narrow pads, sufficient to displace the adhesive from
between the bump and the mating narrow pad; at least partially
curing the adhesive; melting and then re-solidifying the solder,
forming a metallurgical interconnection between the bump and the
narrow pad.
[0015] In another general aspect the invention features a method
for forming flip chip interconnection, by providing a substrate
having narrow interconnection pads formed in a die attach surface
and having a solder mask having openings over the narrow pads, and
a die having bumps attached to interconnect pads in an active
surface; supporting the substrate and the die; positioning the die
with the active side of the die toward the die attach surface of
the substrate, and aligning the die and substrate and moving one
toward the other so that the bumps contact the corresponding narrow
pads on the substrate; melting and then re-solidifying to form the
interconnection between the bump and the narrow pad.
[0016] In some embodiments the solder bump includes a collapsible
solder portion, and the melt and solidifying step melts the bump to
form the interconnection on the narrow pad. In some embodiments the
substrate is provided with a solder paste on the narrow pads, and
the step of moving the die and the substrate toward one another
effects a contact between the bumps and the solder on the narrow
pads, and the melt and solidifying step melts the solder on the
narrow pad to form the interconnection.
[0017] In another general aspect the invention features a method
for forming flip chip interconnection, by providing a substrate
having narrow interconnection pads formed in a die attach surface
and having a solder mask having openings over the narrow pads and
having solder paste on the narrow pads, and a die having bumps
attached to interconnect pads in an active surface; supporting the
substrate and the die; positioning the die with the active side of
the die toward the die attach surface of the substrate, and
aligning the die and substrate and moving one toward the other so
that the bumps contact the solder paste on the corresponding narrow
pads on the substrate; melting and then re-solidifying the solder
paste, forming a metallurgical interconnection between the bump and
the narrow pad.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] FIG. 1 is a diagrammatic sketch of a portion of a
conventional bump-on-capture pad flip chip interconnection, in a
sectional view parallel to the plane of the package substrate
surface, as indicated by the arrows 1-1' in FIG. 2.
[0019] FIG. 2 is a diagrammatic sketch showing a portion of a
conventional bump-on-capture pad flip chip interconnection, in a
sectional view perpendicular to the plane of the package substrate
surface, as indicated by the arrows 2-2' in FIG. 1.
[0020] FIG. 3 is a diagrammatic sketch showing a portion of another
conventional bump-on-capture pad flip chip interconnection, in a
sectional view perpendicular to the plane of the package substrate
surface.
[0021] FIG. 4 is a diagrammatic sketch of a portion of an
embodiment of a bump-on-narrow-pad flip chip interconnection
according to an embodiment of the invention, in a sectional view
parallel to the plane of the package substrate surface.
[0022] FIG. 5 is a diagrammatic sketch showing a portion of an
embodiment of a bump-on-narrow-pad flip chip interconnection
according to an embodiment of the invention as in FIG. 4, in a
sectional view perpendicular to the plane of the package substrate
surface, as indicated by the arrows 5-5' in FIG. 4.
[0023] FIG. 6 is a diagrammatic sketch of a portion of another
embodiment of a bump-on-narrow-pad flip chip interconnection
according to an embodiment of the invention, in a sectional view
parallel to the plane of the package substrate surface.
[0024] FIG. 7 is a diagrammatic sketch showing a portion of an
embodiment of a bump-on-narrow-pad flip chip interconnection
according to an embodiment of the invention as in FIG. 6, in a
sectional view perpendicular to the plane of the package substrate
surface, as indicated by the arrows 7-7' in FIG. 6.
[0025] FIGS. 8 and 9 are diagrammatic sketches, each of a portion
of another embodiment of a bump-on-narrow-pad flip chip
interconnection according to an embodiment of the invention, in a
sectional view parallel to the plane of the package substrate
surface.
[0026] FIGS. 10A-10C are diagrammatic sketches in a sectional view
illustrating steps in a process for making a flip chip
interconnection according to an embodiment of the invention.
[0027] FIGS. 11A-11D are diagrammatic sketches in a sectional view
illustrating steps in a process for making a flip chip
interconnection according to an embodiment of the invention.
[0028] FIG. 12 is a diagrammatic sketch showing a force and
temperature schedule for a process for making a flip chip
interconnection according to an embodiment of the invention.
[0029] FIG. 13 is a diagrammatic sketch in a sectional view showing
a bump-on-narrow-pad flip chip interconnection according to an
embodiment of the invention.
[0030] FIGS. 14A-14E are diagrammatic sketches in plan view
illustrating various interconnect pad shapes according to
embodiments of the invention.
[0031] FIGS. 15A-15C are diagrammatic sketches in plan view
illustrating various interconnect pad configurations according to
embodiments of the invention.
[0032] FIGS. 16A and 16B are diagrammatic sketches in plan view
illustrating solder mask openings according to embodiments of the
invention.
[0033] FIG. 17 is a diagrammatic sketch in plan view illustrating
details of various interconnect pad configurations in relation to a
solder mask opening according to embodiments of the invention.
[0034] FIG. 18 is a diagrammatic sketch in plan view illustrating
details of various solder mask configurations in relation to an
interconnect pad according to embodiments of the invention.
DETAILED DESCRIPTION
[0035] The invention will now be described in further detail by
reference to the drawings, which illustrate alternative embodiments
of the invention. The drawings are diagrammatic, showing features
of the invention and their relation to other features and
structures, and are not made to scale. For improved clarity of
presentation, in the FIGs. illustrating embodiments of the
invention, elements corresponding to elements shown in other
drawings are not all particularly renumbered, although they are all
readily identifiable in all the FIGs.
[0036] The conventional flip chip interconnection is made by using
a melting process to join the bumps (conventionally, solder bumps)
onto the mating surfaces of the corresponding capture pads and,
accordingly, this is known as a "bump-on-capture pad" ("BOC")
interconnect. Two features are evident in the BOC design: first, a
comparatively large capture pad is required to mate with the bump
on the die; second, an insulating material, typically known as a
"solder mask" is required to confine the flow of solder during the
interconnection process. The solder mask opening may define the
contour of the melted solder at the capture pad ("solder mask
defined"), or the solder contour may not be defined by the mask
opening ("non-solder mask defined"); in the latter case--as in the
example of FIG. 1, described in more detail below--the solder mask
opening may be significantly larger than the capture pad. The
techniques for defining solder mask openings have wide tolerance
ranges. Consequently, for a solder mask defined bump configuration,
the capture pad must be large (typically considerably larger than
the design size for the mask opening), to ensure that the mask
opening will be located on the mating surface of the pad; and for a
non-solder mask defined bump configuration, the solder mask opening
must be larger than the capture pad. The width of capture pads (or
diameter, for circular pads) is typically about the same as the
ball (or bump) diameter, and can be as much as two to four times
wider than the trace width. This results in considerable loss of
routing space on the top substrate layer. In particular, for
example, the "escape routing pitch" is much bigger than the finest
trace pitch that the substrate technology can offer. This means
that a significant number of pads must be routed on lower substrate
layers by means of short stubs and vias, often beneath the
footprint of the die, emanating from the pads in question.
[0037] FIGS. 1 and 2 show portions 10, 20 of a conventional flip
chip package, in diagrammatic sectional views; the partial
sectional view in FIG. 1 is taken in a plane parallel to the
package substrate surface, along the line 1-1' in FIG. 2; and the
partial sectional view in FIG. 2 is taken in a plane perpendicular
to the package substrate surface, along the line 2-2' in FIG. 1.
Certain features are shown as if transparent, but many of the
features in FIG. 1 are shown at least partly obscured by overlying
features. Referring now to both FIG. 1 and FIG. 2, a die attach
surface of the package substrate includes a metal layer formed on a
dielectric layer 12. The metal layer is patterned to form leads 13
and capture pads 14. An insulating layer 16, typically termed a
"solder mask", covers the die attach surface of the substrate; the
solder mask is usually constructed of a photodefinable material,
and is patterned by conventional photoresist patterning techniques
to leave the mating surfaces of the capture pads 14 exposed.
Interconnect bumps 15 attached to pads on the active side of the
die 18 are joined to the mating surfaces of corresponding capture
pads 14 on the substrate to form appropriate electrical
interconnection between the circuitry on the die and the leads on
the substrate. After the reflowed solder is cooled to establish the
electrical connection, an underfill material 17 is introduced into
the space between the die 18 and the substrate 12, mechanically
stabilizing the interconnects and protecting the features between
the die and the substrate.
[0038] As FIG. 1 shows by way of example, signal escape traces in
the upper metal layer of the substrate (leads 13), lead from their
respective capture pads 14 across the die edge location, indicated
by the broken line 11, and away from the die footprint. In a
typical example the signal traces may have an escape pitch P.sub.E
about 112 um. A 30 um/30 um design rule is typical for the traces
themselves in a configuration as shown in FIG. 1; that is, the
traces are nominally 30 um wide, and they can be spaced as close
together as 30 um. The capture pads are typically three times
greater than the trace width and, accordingly in this example the
capture pads have a width (or diameter, as they are roughly
circular in this example) nominally 90 um. And, in this example,
the openings in the solder mask are larger than the pads, having a
nominal width (diameter) of 135 um.
[0039] FIGS. 1 and 2 show a non-solder mask defined solder contour.
As the fusible material of the bumps on the die melt, the molten
solder tends to "wet" the metal of the leads and capture pads, and
the solder tends to "run out" over any contiguous metal surfaces
that are not masked. The solder tends to flow along the contiguous
lead 13, and here the solder flow is limited by the solder mask,
for example at 19 in FIG. 1. A non-solder mask defined solder
contour at the pad is apparent in FIG. 2, in which the material of
the bumps 15 is shown as having flowed, 29, over the sides of the
capture pads 14 and down to the surface of the dielectric layer of
the substrate 12. This is referred to as a non-solder mask defined
contour because the solder mask does not limit the flow of solder
over the surface and down over the sides of the capture pads,
and--unless there is a substantial excess of solder at the pad--the
flow of solder is limited by the fact that the dielectric surface
of the substrate is typically not wettable by the molten solder. A
lower limit on the density of the capture pads in a conventional
arrangement, as in FIG. 1, is determined by, among other factors,
limits on the capacity of the mask forming technology to make
reliable narrow mask structures, and the need to provide mask
structures between adjacent mask openings. A lower limit on the
escape density is additionally determined by, among other factors,
the need for escape lines from more centrally located capture pads
to be routed between more peripherally located capture pads.
[0040] FIG. 3 shows a conventional solder mask defined solder
contour, in a sectional view similar to that in FIG. 2. A die 38 is
shown affixed by way of bumps 35 onto the mating surfaces of
capture pads 34 formed along with traces (leads 33) by patterning a
metal layer on the die attach side of a dielectric layer of the
substrate 32. After the reflowed solder is cooled to establish the
electrical connection, an underfill material 37 is introduced into
the space between the die 38 and the substrate 32, mechanically
stabilizing the interconnects and protecting the features between
the die and the substrate. Here the capture pads 34 are wider than
in the example of FIGS. 1 and 2, and the solder mask openings are
smaller than the capture pads, so that the solder mask material
covers the sides and part of the mating surface of each capture
pad, as shown at 39, as well as the leads 33. When the bumps 35 are
brought into contact with the mating surfaces of the respective
capture pads 34, and then melted, the solder mask material 39
restricts the flow of the molten solder, so that the shapes of the
solder contours are defined by the shapes and dimensions of the
mask openings over the capture pads 34. In a typical example of a
conventional solder mask defined bump-on-capture pad
interconnection, the capture pad has a diameter about 140 um, and
the solder mask opening has a diameter about 90 um, and the routing
traces are about 25-30 um wide. The diameter of the mating surface
for attachment of the bump to the die pad (not shown in FIG. 2 or
3), that is, the place of interface between the bump and the die
pad, is defined by the solder mask opening as having a diameter
about 90 um, in this example.
[0041] FIGS. 4 and 6 each show a portion of a bump-on-narrow-pad
("BONP") flip chip interconnection according to an embodiment of
the invention, in a diagrammatic partial sectional view taken in a
plane parallel to the substrate surface, along the lines 4-4' and
6-6' in FIGS. 5 and 7, respectively. Certain features are shown as
if transparent. According to the invention the interconnection is
achieved by mating the bumps onto respective narrow interconnection
pads on the substrate and, accordingly, this is referred to herein
as a "bump-on-narrow-pad" ("BONP") interconnect. Solder mask
materials typically cannot be resolved at such fine geometries and,
according to these embodiments of the invention, no solder mask is
used. Instead the function of confining molten solder flow is
accomplished without a solder mask in the course of the assembly
process (as described below). FIG. 5 shows a partial sectional view
of a package as in FIG. 4, taken in a plane perpendicular to the
plane of the package substrate surface, along the line 5-5' in FIG.
4; and FIG. 7 shows a partial sectional view of a package as in
FIG. 6, taken in a plane perpendicular to the plane of the package
substrate surface, along the line 7-7' in FIG. 6.
[0042] Escape routing patterns for bump-on-narrow-pad ("BONP")
substrates according to the invention are shown by way of example
in FIGS. 4 and 6: in FIG. 4, arranged for a die on which the die
attach pads for the interconnect balls are in a row near the die
perimeter, the bumps 45 are mated onto corresponding narrow
interconnection pads on the escape traces 43 in a row near the edge
of the die footprint, indicated by the broken line 41; in FIG. 6,
arranged for a die on which the die attach pads are in an array of
parallel rows near the die perimeter, the bumps 65 are mated onto
corresponding narrow interconnection pads on the escape traces 63
in a complementary array near the edge of the die footprint,
indicated by the broken line 61.
[0043] As FIGS. 4 and 6 illustrate, the routing density achievable
using bump-on-narrow-pad interconnect according to the invention
can equal the finest trace pitch offered by the substrate
technology. In the specific case illustrated, this constitutes a
routing density which is approximately 90% higher than is achieved
in a conventional bump-on-capture pad arrangement. In the perimeter
row embodiments of BONP (e.g., FIG. 4), the bumps are placed at a
fine pitch, which can equal the finest trace pitch of the
substrate. This arrangement poses a challenge for the assembly
process, because the bumping and bonding pitch must be very fine.
In the perimeter array version of BONP (e.g., FIG. 6), the bumps
are arranged on an area array, providing greater space for a larger
bumping and bonding pitch, and relieving the technological
challenges for the assembly process. Even in the array embodiments,
the routing traces on the substrate are at the same effective pitch
as in the perimeter row arrangement, and an arrangement as in FIG.
6 relieves the burden of fine pitch bumping and bonding without
sacrificing the fine escape routing pitch advantage.
[0044] Referring particularly now to FIGS. 4 and 5, leads 43 and
narrow interconnection pads 46 are formed by patterning a metal
layer on a die attach surface of a substrate dielectric layer 42.
The narrow pads 46 are formed as a widening of the traces 43 at the
interconnection sites. The "width" of an interconnection pad
(W.sub.b in FIG. 5) is the nominal or design dimension across the
widened part of the trace at the interconnection site. According to
the invention, the width of the narrow interconnection pad on a
substrate is established according to the bump base width ("base
diameter") of the bumps on the die that is to be connected to the
substrate. The "bump base width" (W.sub.p in FIG. 5) is the nominal
or design diameter of the generally round (approximately circular)
contact interface between the bump 45 and the die pad 49. (As may
be appreciated, the diameter of the bump, taken in a plane parallel
to the bump-pad interface, may be greater than the bump base width,
as illustrated diagrammatically in FIGS. 2, 3, 5 and 7, for
example.) Particularly according to the invention, the
interconnection pad width W.sub.b is smaller than the bump base
width W.sub.p, and the narrow interconnection pad width may be as
small as 20% of the bump base width. In many embodiments the narrow
pad width is in a range about 20% to about 80% of the bump base
width. In some embodiments the narrow interconnection pad width is
less than the bump base width and greater than about 25% of the
bump base width. In some embodiments the narrow pad width is less
than about 60% of the bump base width.
[0045] According to the invention, electrical interconnection of
the die 48 is made by joining the bumps 45 on the die onto the
narrow interconnection pads 46 on the leads 43. The conventional
comparatively wide capture pad is unnecessary according to the
invention and, in embodiments as in FIGS. 4 and 5, no solder mask
is required; the process is described in detail below.
[0046] Conventional capture pads typically are about the same width
(diameter) as the bumps, and are typically two to four times as
wide as the trace or lead width. As will be appreciated, some
variation in the width of leads is expected. As used herein, a
narrow interconnection pad has a nominal or design width at least
about 120% of the nominal or trace design rule width, and
bump-on-narrow-lead interconnection according to the invention
includes bumps connected to widened parts of traces that are
greater than about 120% of the nominal or trace design rule width,
and less than the bump base diameter. An interconnection site that
has a width less than about 120% does not constitute a narrow
interconnect pad, and interconnection made by connecting bumps onto
portions of leads that are less than about 120% of the nominal or
trace design rule width is referred to as a "bump-on-lead"
interconnection.
[0047] Similarly, referring to FIGS. 6 and 7, leads 63 and narrow
interconnection pads 66 are formed by patterning a metal layer on a
die attach surface of a substrate dielectric layer 62. The signal
escape traces lead across the die edge location, indicated by the
broken line 61, and away from the die footprint. The narrow pads 66
are formed as a widening of the traces 63 at the interconnection
sites. The "width" of an interconnection pad (W.sub.b in FIG. 7) is
the nominal or design dimension across the widened part of the
trace at the interconnection site. In this example, as in the
example shown in FIGS. 4 and 5, according to the invention, the
width of the narrow interconnection pad on a substrate is
established according to the bump base width of the bumps on the
die that is to be connected to the substrate. The "bump base width"
(W.sub.p in FIG. 7) is the nominal or design diameter of the
generally round (approximately circular) contact interface between
the bump 65 and the die pad 69. Particularly according to the
invention, the interconnection pad width W.sub.b is smaller than
the bump base width W.sub.p, and the narrow interconnection pad
width may be as small as 20% of the bump base width. In many
embodiments the narrow pad width is in a range about 20% to about
80% of the bump base width. In some embodiments the narrow
interconnection pad width is less than the bump base width and
greater than about 25% of the bump base width. In some embodiments
the narrow pad width is less than about 60% of the bump base
width.
[0048] According to the invention, electrical interconnection of
the die 68 is made by joining the bumps 65 on the die narrow
interconnection pads 66 on the leads 63. Certain of the escape
traces, e.g. 66, leading across the die edge location from
interconnect sites in rows toward the interior of the die
footprint, pass between the bumps 65 on more peripheral rows of
interconnect sites. No capture pads are required according to the
invention and, in embodiments as in FIGS. 6 and 7, no solder mask
is required; the process is described in detail below.
[0049] According to the invention, as the techniques for forming
the traces improves, it is possible to reliably form traces having
nominal or design rule widths less than about 25 um. The reduced
trace widths can provide for increased routing density. However,
the mechanical reliability of a "bump-on-lead" flip chip
interconnect on leads less than about 25 um may be unsatisfactory,
because the dimensions of the interface between the bump and the
lead are small, and may not provide sufficient bonding strength to
provide a good electrical interconnection. The invention provides
for reliable mechanical connection (and good electrical
interconnection) by forming a narrow interconnect pad by widening
the lead to an extent dimensionally related to the bump base
diameter, and limited to less than the bump base diameter.
[0050] The narrow interconnect pad according to the invention may
be shaped in any of a variety of ways. Some such shapes may be more
readily manufacturable, and some may provide other process
advantages. For example, the narrow pad may be generally
rectangular, either square or elongated, as shown for example in
FIGS. 14A and 14B; or, it may be generally round, either circular
or elliptical, as shown for example in FIGS. 14C and 14D. Other
shapes may be employed; one particularly useful shape is shown by
way of example in FIG. 14E, having semicircular portions separated
lengthwise the lead or trace by a square or rectangular portion.
Also, the narrow pad can be formed as a symmetrical or an
asymmetrical widening in the lead or trace, as shown in FIGS. 15A
and 15B (showing a generally rectangular pad as an example). Also,
the narrow pad need not be situated at, or near, the end of the
lead or trace, but may be formed at any point where interconnection
is specified, as illustrated in FIG. 15C (showing a generally
rectangular pad as an example). Forming the pad longer than wide
increases the wettable mating surface of the narrow pad (planar
surface plus the exposed parts of the sides), and can improve the
mechanical strength of the interconnection. Also, where the pad is
longer than wide, the tolerance for misalignment of solder mask
openings (or bump) is increased; particularly where the pad is at
the end of the trace, an elongated pad can reduce the likelihood
that a solder mask opening (or bump) will be situated off the end
of the pad.
[0051] The solder mask openings shown by way of example in FIGS. 4,
6, 8 and 9 are generally round (circular or elliptical), but
according to the invention the solder mask opening may be shaped in
any of a variety of ways. It may be useful for example, to provide
a generally rectangular solder mask opening, either square or
elongated, as shown in FIGS. 16A, 16B, (showing a generally
rectangular pad as an example). A square or rectangle of a given
width has a greater area than a circle or ellipse having the same
width (diameter, short axis). For this reason a square or
rectangular mask opening has a capacity to hold a greater quantity
of solder paste (or other fusible material), and accordingly this
may provide an advantage where a fusible material such as a solder
paste is to be applied to the mating surfaces on the narrow pads
prior to mating with the bumps (described in more detail below).
Also, it may be easier to print a fusible material into a square or
rectangular mask opening than into a circular or elliptical mask
opening, because there is greater tolerance for misalignment in the
printing process. Also, given a width limitation for the mask
opening, a square or rectangular mask opening provides a greater
open area for mounting a large bump on the pad during the
interconnection process.
[0052] Various narrow pad configurations according to embodiments
of the invention are shown in FIG. 17 by way of example in relation
to a circular mask opening 174 in a solder mask 176. The mask
opening in each example has a width (diameter) Wm, which may be,
for example, about 90 um. A bump-on-lead configuration is shown at
173. The lead or trace 172 has a nominal (design) width WL, which
may be, for example, about 30 um. A narrow pad having a rectangular
shape is shown at 175. In this example the lead or trace at which
the narrow pad is formed has a nominal (design) width WL', which
may be, for example, about 30 um. The rectangular narrow pad has a
width WP', which may be, for example, about 45 um. A narrow pad
having an oval shape is shown at 177 formed at a wider lead or
trace, having a nominal (design) width WL'', which may be, for
example, about 50 um. A narrow pad having a rectangular shape
expanded with an oval shape is shown at 179. In this example the
narrower lead or trace at which the narrow pad is formed has a
nominal (design) width WL''', which may be, for example, about 30
um. The rectangular portion of the narrow pad 179 has a width WP'',
which may be, for example, about 45 um; and the oval expanded
portion has a width WPE, which may be for example, about 50 um.
[0053] Various solder mask opening configurations according to
embodiments of the invention are shown in FIG. 18 by way of example
in relation to a lead (or trace) or narrow pad 182. In these
examples the lead or narrow pad at the interconnect site has a
width WL, which may be, for example, about 40 um. In a first
example, a circular solder mask opening 185 having a width
(diameter) Wm, which may be, for example, about 90 um, exposes an
interconnect site portion 183. In a second example a rectangular
solder mask opening 187 having a width (across the lead or narrow
pad) Wm', which may be, for example, about 80 um, and a length Lm',
which may be, for example, about 120 um., exposes an interconnect
site portion 183'. In a third example an elliptical solder mask
opening 189 having a width (across the lead or narrow pad) Wm'',
which may be, for example, about 80 um, and a length Lm'', which
may be, for example, about 120 um., exposes an interconnect site
portion 183''. Both the rectangular opening 187 and the oval
opening 189 expose a greater length (hence, area) of the lead or
pad at the site 183'', 183'' than does the circular solder mask
opening 185, even though the circular opening in this example has a
greater diameter. This provides a greater area for solder reflow
during the interconnect process, and can result in a more robust
interconnection. The area exposed by the rectangular opening 187 is
slightly greater than that provided by the elliptical opening 189
having the same width and length; and moreover, the area would be
reduced if there were a slight misalignment of the elliptical
opening, but not by a slight misalignment of the rectangular
opening. As a practical matter, however, an design rectangular
opening may have more or less rounded corners because of resolution
limitations in processes for patterning openings in the solder mask
dielectric.
[0054] In some illustrative examples according to the invention,
the diameter of the bump base on the die to be mounted may be about
90 um, and the narrow interconnect pad is formed on the substrate
to a width in a range about 25 um (where the trace width is less
than about 25 um) to about 50 um. This provides a significant
improvement in routing density, as compared with a substrate having
a conventional capture pad having a much larger diameter, which may
be typically two to four times as great as the trace width.
[0055] As FIGS. 4 and 6 illustrate, bump-on-narrow-pad interconnect
according to the invention can provide a significantly higher
signal trace escape routing density. Also, as FIGS. 4 and 6
illustrate, the BONP interconnect according to this aspect of the
invention does not require use of a solder mask to define the
solder contour at the interconnect site.
[0056] The BONP interconnection structure of embodiments such as
are shown by way of example in FIGS. 4, 5, 6 and 7 can be produced
according to the invention by any of several methods, not requiring
a solder mask. In general, interconnect bumps (typically solder
bumps) are affixed onto interconnect pads on the active side of the
die. A die attach surface of the substrate (termed the "upper"
surface) has an upper metal layer patterned to provide the traces
and narrow pads at interconnect sites as appropriate for
interconnection with the arrangement of bumps on the particular
die. In a preferred method of the invention, an encapsulating resin
adhesive is employed to confine the solder flow during a melt phase
of the interconnection process.
[0057] FIGS. 8 and 9 show two examples of a portion of a
bump-on-narrow-pad flip chip interconnection according to other
embodiments of the invention, in a diagrammatic sectional view
taken in a plane parallel to the substrate surface. Certain
features are shown as if transparent. According to this aspect of
the invention a solder mask is provided, which may have a nominal
mask opening diameter in the range about 80 um to 90 um. Solder
mask materials can be resolved at such pitches and, particularly,
substrates can be made comparatively inexpensively with solder
masks having 90 um openings and having alignment tolerances plus or
minus 25 um. In some embodiments laminate substrates (such as 4
metal layer laminates), made according to standard design rules,
are used. In the embodiments of FIGS. 8 and 9, for example, the
traces may be at .about.90 um pitch and the narrow pads may be in a
270 um area array, providing an effective escape pitch .about.90 um
across the edge of the die footprint, indicated by the broken line
81.
[0058] In embodiments as in FIGS. 8 and 9 a no-flow underfill is
not required; a conventional capillary underfill can be
employed.
[0059] In embodiments as in FIG. 8 the interconnection is achieved
by mating the bumps directly onto an narrow interconnect pad 84 on
a narrow lead or trace 83 patterned on a dielectric layer on the
die attach surface of the substrate 82; the solder mask 86 serves
to limit flow of solder within the bounds of the mask openings 88,
preventing solder flow away from the interconnect site along the
solder-wettable lead. The solder mask may additionally confine flow
of molten solder between leads, or this may be accomplished in the
course of the assembly process.
[0060] In embodiments as in FIG. 9, as in FIG. 8, narrow pads on
traces 93 are patterned on a dielectric layer on the die attach
surface of the substrate 92. Solder paste is provided at the
interconnect sites (narrow pads) 94 on the leads 93, to provide a
fusible medium for the interconnect. The openings 98 in the solder
mask 96 serve to define the paste. The paste is dispensed, for
example by a standard printing process, then is reflowed, and then
may be coined if necessary to provide uniform surfaces to meet the
balls. The solder paste can be applied in the course of assembly
using a substrate as described above with reference to FIG. 8; or,
a substrate may be provided with paste suitably patterned prior to
assembly. Other approaches to applying solder selectively to the
interconnect sites may be employed in the solder-on-narrow-pad
embodiments of the invention, including electroless plating and
electroplating techniques. The solder-on-narrow-pad configuration
provides additional solder volume for the interconnect, and can
accordingly provide higher product yield, and can also provide a
higher die standoff.
[0061] Accordingly, in some embodiments the solder-on-narrow-pad
configuration according to the invention is employed for
interconnection of a die having high-melting temperature solder
bumps (such as a high-lead [high Pb] solder, conventionally used
for interconnection with ceramic substrates) onto an organic
substrate. The solder paste can be selected to have a melting
temperature low enough that the organic substrate is not damaged
during reflow. To form the interconnect in such embodiments the
high-melting interconnect bumps are contacted with the
solder-on-narrow-pad sites, and the remelt fuses the
solder-on-narrow-pad to the bumps. Where a noncollapsible bump is
used, together with a solder-on-narrow-pad process, no preapplied
adhesive is required, as the displacement or flow of the solder is
limited by the fact that only a small quantity of solder is present
at each interconnect, and the noncollapsible bump prevents collapse
of the assembly.
[0062] In other embodiments the solder-on-narrow-pad configuration
according to the invention is employed for interconnection of a die
having eutectic solder bumps.
[0063] One embodiment of a preferred method for making a
bump-on-narrow-pad interconnection is shown diagrammatically in
FIGS. 10A-10C.
[0064] Referring to the FIGs., a substrate 112 is provided, having
at least one dielectric layer and having a metal layer on a die
attach surface 113, the metal layer being patterned to provide
circuitry, particularly narrow interconnection pads 114 on traces
or leads, on the die attach surface. The substrate 112 is
supported, for example on a carrier or stage 116, with a substrate
surface opposite the die attach surface 113 facing the support. A
quantity of an encapsulating resin 122 is dispensed over the die
attach surface 113 of the substrate, covering at least the narrow
interconnection pads 114 on the leads. A die 102 is provided,
having bumps 104 attached to die pads (not shown in the FIG.) on
the active side 103. The bumps include a fusible material which
contacts the mating surfaces of the narrow pads. A pick-and-place
tool 108 including a chuck 106 picks up the die by contact of the
chuck 106 with the backside 101 of the die. Using the
pick-and-place tool, the die is positioned facing the substrate
with the active side of the die toward the die attach surface of
the substrate, as shown in FIG. 10A; and the die and substrate are
aligned and moved one toward the other (arrow M) so that the bumps
104 contact the corresponding narrow interconnection pads 114 on
the traces (leads) on the substrate. Then a force is applied (arrow
F) to press the bumps 105 onto the mating surfaces 134 at the
narrow pads 115 on the leads, as shown in FIG. 10B. The force must
be sufficient at least to displace the adhesive 122 from between
the bumps and the mating surfaces 134 at the narrow interconnection
pads 115. The bumps may be deformed by the force, breaking the
oxide film on the contacting surface of the bumps and/or on the
mating surface of narrow pads. The deformation of the bumps may
result in the fusible material of the bumps being pressed onto the
top and over the edges of the narrow pads. The adhesive is caused
to cure at least partially, as shown at 132, as for example by
heating to a selected temperature. At this stage the adhesive need
only be partially cured, that is, only to an extent sufficient
subsequently to prevent flow of molten solder along an interface
between the adhesive and the conductive traces. Then the fusible
material of the bumps 105 is melted and then is re-solidified,
forming a metallurgical interconnection between the bump 105 and
narrow pad 115, and the adhesive curing is completed, to complete
the die mount and to secure the electrical interconnection at the
mating surface (now an interconnect interface) 144, as shown
generally at 140 in FIG. 10C. In the plane of the sectional view
shown in FIG. 1C, interconnection is formed between certain of the
bumps 145 and corresponding narrow interconnection pads 155 on
certain of the leads, as for example in a configuration as in FIG.
6. Other leads 156 are interconnected on narrow interconnection
pads at other localities, which would be visible in other sectional
views. A comparatively high trace density is shown. The curing of
the adhesive may be completed prior to, or concurrently with, or
following melting the solder. Typically, the adhesive is a
thermally curable adhesive, and the extent of curing at any phase
in the process is controlled by regulating the temperature. The
components can be heated and cured by raising the temperature of
the chuck on the pick and place tool, or by raising the temperature
of the substrate support, for example.
[0065] The process is shown in further detail in FIGS. 11A-11D. In
FIG. 1A, a substrate 212 is provided on a die attach surface with
conductive (metal) traces, and narrow interconnection pads 214 at
interconnect sites on the traces are covered with an adhesive 222.
The die 202 is positioned in relation to the substrate 212 such
that the active side of the die faces the die attach side of the
substrate, and is aligned (arrows A) such that bumps 204 on the die
are aligned with corresponding mating surfaces on narrow pads 214.
The die and the substrate are moved toward one another so that the
bumps contact the respective mating surfaces on the narrow pads.
Then as shown in FIG. 11B a force is applied to move the bumps 205
and narrow pads 215 against one another, displacing the adhesive as
shown at 232 in FIG. 1I B, and deforming the bumps onto the mating
surfaces 234 and over the edges of the narrow pads. Deformation of
the bumps on the narrow pads breaks the oxide film on the contact
surfaces of the bumps and the mating surfaces of the narrow pads,
establishing a good electrical connection, and deformation of the
bumps over the edges of the narrow pads helps establish a good
temporary mechanical connection. As in the example of FIG. 10A-10C,
the narrow interconnection pads of certain of the traces 216 are
out of the plane of FIG. 11B. Heat is applied to partially cure the
adhesive as shown at 236 in FIG. 11C. Then heat is applied to raise
the temperature of the bumps sufficiently to cause the fusible
material of the bumps to melt, as shown in FIG. 11D. This
substantially (though not necessarily fully) completes the cure of
the adhesive 246 and completes the metallurgical interconnection of
the bumps 245 onto the mating surfaces 244 at the narrow
interconnection pads 215. The cured adhesive stabilizes the die
mount.
[0066] In an alternative embodiment of a preferred method, the
adhesive can be pre-applied to the die surface, or at least to the
bumps on the die surface, rather than to the substrate. The
adhesive can, for example, be pooled in a reservoir, and the active
side of the die can be dipped in the pool and removed, so that a
quantity of the adhesive is carried on the bumps; then, using a
pick-and-place tool, the die is positioned facing a supported
substrate with the active side of the die toward the die attach
surface of the substrate, and the die and substrate are aligned and
moved one toward the other so that the bumps contact the
corresponding traces (leads) on the substrate. Such a method is
described in U.S. Pat. No. 6,780,682, Aug. 24, 2004, which is
hereby incorporated by reference. Then forcing, curing, and melting
are carried out as described above.
[0067] A force and temperature schedule for a process according to
the invention is shown diagrammatically by way of example in FIG.
12. In this chart, time runs from left to right on the horizontal
axis; a force profile 310 is shown as a thick solid line, and a
temperature profile 320 is shown as a dotted line. The temperature
profile begins at a temperature in the range about 80.degree.
C.-about 90.degree. C. The force profile begins at essentially zero
force. Beginning at an initial time t.sub.i the force is rapidly
(nearly instantaneously) raised 312 from F.sub.i to a
displacement/deformation force F.sub.d and held 314 at that force
for a time, as discussed below. F.sub.d is a force sufficiently
great to displace the adhesive away from between the bumps and the
mating surfaces of the narrow interconnection pads; and,
preferably, sufficient to deform the fusible (narrow
pad-contacting) portion of the bumps onto the mating surface,
breaking the oxide films and forming a good metal-to-metal
(metallurgical) contact, and, in some embodiments, over the edges
of the narrow pads to establish a mechanical interlock of the bumps
and the narrow pads ("creep" deformation). The total amount of
force required will depend upon the bump material and dimensions
and upon the number of bumps, and can be determined without undue
experimentation. As the force is raised, the temperature is also
rapidly raised 322 from an initial temperature T.sub.i to a gel
temperature Tg. The gel temperature Tg is a temperature sufficient
to partially cure the adhesive (to a "gel"). Preferably, the force
and temperature ramps are set so that there is a short lag time
t.sub.def, following the moment when F.sub.d is reached and before
T.sub.g is reached, at least long enough to permit the elevated
force to displace the adhesive and to deform the bumps before the
partial cure of the adhesive commences. The assembly is held 314,
324 at the displacement/deformation pressure F.sub.d and at the gel
temperature T.sub.g for a time t.sub.gel sufficient to effect the
partial cure of the adhesive. The adhesive should become
sufficiently firm that it can subsequently maintain a good bump
profile during the solder remelt phase--that is, sufficiently firm
to prevent undesirable displacement of the molten fusible material
of the bump, or flow of the molten fusible material along the
narrow pads and leads. Once the adhesive has partially cured to a
sufficient extent, the pressure may be ramped down rapidly 318 to
substantially no force (weight of the components). The temperature
is then rapidly raised further 323 to a temperature T.sub.m
sufficient to remelt the fusible portions (solder) of the bumps,
and the assembly is held 325 at the remelt temperature T.sub.m for
a time t.sub.melt/cure at least sufficient to fully form the solder
remelt on the narrow pads, and preferably sufficient to
substantially (though not necessarily fully) cure the adhesive.
Then the temperature is ramped down 328 to the initial temperature
T.sub.i, and eventually to ambient. The process outlined in FIG. 12
can run its course over a time period of 5-10 seconds.
[0068] The adhesive in embodiments as in FIG. 12 may be referred to
as a "no-flow underfill". In some approaches to flip chip
interconnection, the metallurgical interconnection is formed first,
and then an underfill material is flowed into the space between the
die and the substrate. The "no-flow underfill" according to the
invention is applied before the die and the substrate are brought
together, and the no-flow underfill is displaced by the approach of
the bumps onto the narrow pads, and by the opposed surfaces of the
die and the substrate. The adhesive for the no-flow underfill
adhesive according to the invention is preferably a fast-gelling
adhesive--that is, a material that gels sufficiently at the gel
temperature in a time period in the order of 1-2 seconds. Preferred
materials for the no-flow underfill adhesive include, for example,
so-called non-conductive pastes, such as those marketed by Toshiba
Chemicals and by Loktite-Henkel, for example.
[0069] Alternative bump structures may be employed in the
bump-on-narrow-pad interconnects according to the invention.
Particularly, for example, so-called composite solder bumps may be
used. Composite solder bumps have at least two bump portions, made
of different bump materials, including one which is collapsible
under reflow conditions, and one which is substantially
non-collapsible under reflow conditions. The non-collapsible
portion is attached to the interconnect site on the die; typical
conventional materials for the non-collapsible portion include
various solders having a high lead (Pd) content, for example; and
gold (Au), for example. The collapsible portion is joined to the
non-collapsible portion, and it is the collapsible portion that
makes the connection with the narrow interconnect pad according to
the invention. Typical conventional materials for the collapsible
portion of the composite bump include eutectic solders, for
example.
[0070] An example of a bump-on-narrow-pad interconnect employing a
composite bump is shown in a diagrammatic sectional view in FIG.
13. Referring now to FIG. 13, die 302 is provided on die pads in
the active side of the die with composite bumps that include a
noncollapsible portion 345 and a collapsible portion 347. The
collapsible portion may be, for example, a eutectic solder or a
relatively low temperature melt solder). The collapsible portion
contacts the mating surface of the narrow pad and, where
deformation of the fusible portion of the bump over the narrow pad
is desired, the collapsible portion of the bump is deformable under
the conditions of force employed. The noncollapsible portion may
be, for example, a solder having a high lead (Pb) content. The
noncollapsible portion does not deform when the die is moved under
pressure against the substrate during processing, and does not melt
during the reflow phase of the process. Accordingly the
noncollapsible portion can be dimensioned to provide a standoff
distance between the active surface of the die and the die attach
surface of the substrate.
[0071] As may be appreciated, the bumps in embodiments as shown in,
for example, FIGS. 4, 5, 6 and 7 need not necessarily be fully
collapsible bumps. The structures shown in those FIGs. may
alternatively be made using composite bumps, or using
non-collapsible bumps (high-Pb, or Au) in a solder-on-narrow-pad
method, as described above.
[0072] And, as may be appreciated in view of the foregoing, an
interconnect as appears for example in FIG. 13 can be formed by
bringing a non-composite non-collapsible bump (high-Pb, Au) into
contact with a narrow interconnect pad provided on the mating
surface with a fusible material (such as, for example, a eutectic
solder or a relatively low temperature melt solder, which may be
provided as a solder paste). Or, the narrow interconnect pad may be
provided on the mating surface with a fusible material and the
bumps may be composite bumps, also provided with a collapsible
(fusible) portion. Where the narrow interconnect pads are provided
on the mating surface with a fusible material, it may be preferred
to employ a solder mask, followed by a capillary underfill, in the
process.
[0073] Other embodiments are within the following claims.
* * * * *