U.S. patent application number 13/195583 was filed with the patent office on 2013-04-18 for three dimensional structures having improved alignments between layers of microcomponents.
This patent application is currently assigned to S.O.I.TEC SILICON ON INSULATOR TECHNOLOGIES. The applicant listed for this patent is Marcel Broekaart, Arnaud Castex. Invention is credited to Marcel Broekaart, Arnaud Castex.
Application Number | 20130093033 13/195583 |
Document ID | / |
Family ID | 40361570 |
Filed Date | 2013-04-18 |
United States Patent
Application |
20130093033 |
Kind Code |
A9 |
Castex; Arnaud ; et
al. |
April 18, 2013 |
THREE DIMENSIONAL STRUCTURES HAVING IMPROVED ALIGNMENTS BETWEEN
LAYERS OF MICROCOMPONENTS
Abstract
The invention relates to a method of initiating molecular
bonding, comprising bringing one face (31) of a first wafer (30) to
face one face (21) of a second wafer (20) and initiating a point of
contact between the two facing faces. The point of contact is
initiated by application to one of the two wafers, for example
using a bearing element (51) of a tool (50), of a mechanical
pressure in the range 0.1 MPa to 33.3 MPa.
Inventors: |
Castex; Arnaud; (Grenoble,
FR) ; Broekaart; Marcel; (Theys, FR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Castex; Arnaud
Broekaart; Marcel |
Grenoble
Theys |
|
FR
FR |
|
|
Assignee: |
S.O.I.TEC SILICON ON INSULATOR
TECHNOLOGIES
Crolles Cedex
FR
|
Prior
Publication: |
|
Document Identifier |
Publication Date |
|
US 20110278691 A1 |
November 17, 2011 |
|
|
Family ID: |
40361570 |
Appl. No.: |
13/195583 |
Filed: |
August 1, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12936639 |
Nov 15, 2010 |
8163570 |
|
|
PCT/EP2009/060250 |
Aug 6, 2009 |
|
|
|
13195583 |
|
|
|
|
12229248 |
Aug 21, 2008 |
8232130 |
|
|
12936639 |
|
|
|
|
Current U.S.
Class: |
257/432 ;
257/E31.127 |
Current CPC
Class: |
H01L 21/76251
20130101 |
Class at
Publication: |
257/432 ;
257/E31.127 |
International
Class: |
H01L 31/0232 20060101
H01L031/0232 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 28, 2008 |
FR |
0855767 |
Claims
1. A composite three-dimensional structure, comprising: a first
layer; a first plurality of microcomponents formed on a surface of
the first layer; a second layer having a first surface molecularly
bonded to the surface of the first layer on which the first
plurality of microcomponents are formed such that the first
plurality of microcomponents are buried between the first layer and
the second layer; and a second plurality of microcomponents foamed
on an exposed second surface of the first layer, at least some
microcomponents of the second plurality of microcomponents
corresponding to and generally vertically aligned with at least
some microcomponents of the first plurality of microcomponents,
wherein offsets between corresponding microcomponents of the first
and second pluralities of microcomponents are limited to values
less than 200 nm.
2. The composite three-dimensional structure of claim 1, wherein
the offsets between corresponding microcomponents of the first and
second pluralities of microcomponents are limited to values less
than 100 nm.
3. The composite three-dimensional structure of claim 1, wherein
the offsets between corresponding microcomponents of the first and
second pluralities of microcomponents are limited to values less
than 200 nm in a homogenous manner over the entire composite
three-dimensional structure.
4. The composite three-dimensional structure of claim 1, wherein
the at least some microcomponents of the first plurality of
microcomponents comprise image sensors.
5. The composite three-dimensional structure of claim 4, wherein
the at least some microcomponents of the second plurality of
microcomponents comprise color filters.
6. The composite three-dimensional structure of claim 1, wherein
the at least some microcomponents of the second plurality of
microcomponents comprise at least one of contact points and
interconnections.
7. The composite three-dimensional structure of claim 1, wherein
the second layer comprises a wafer.
8. The composite three-dimensional structure of claim 7, wherein
the wafer comprises a semiconductor material.
9. The composite three-dimensional structure of claim 7, wherein
the wafer has a diameter in a range of from 100 mm to 300 mm.
10. The composite three-dimensional structure of claim 1, wherein
the first layer comprises a material selected from the group
consisting of silicon, germanium.
11. The composite three-dimensional structure of claim 10, wherein
the second layer comprises a material selected from the group
consisting of silicon, germanium, glass, quartz, and sapphire.
12. The composite three-dimensional structure of claim 1, wherein
at least one of the surface of the first layer and the first
surface of the second layer comprises an oxide material.
13. A three-dimensionally integrated structure, comprising: a first
layer; a second layer having a first surface bonded to a surface of
the first layer by direct inter-atomic forces between the surface
of the first layer and the first surface of the second layer; a
first plurality of microcomponents at a surface of the first layer
between the first layer and the second layer; and a second
plurality of microcomponents at an exposed second surface of the
first layer, the second plurality of microcomponents aligned with
the first plurality of microcomponents, wherein unintended offsets
between aligned microcomponents of the first plurality of
microcomponents and the second plurality of microcomponents are
limited to about 200 nm or less.
14. The three-dimensionally integrated structure of claim 13,
wherein the unintended offsets between aligned microcomponents of
the first plurality of microcomponents and the second plurality of
microcomponents are limited to about 100 nm or less.
15. The three-dimensionally integrated structure of claim 13,
wherein the unintended offsets between aligned microcomponents of
the first plurality of microcomponents and the second plurality of
microcomponents are limited to about 200 nm or less in a homogenous
manner across the three-dimensionally integrated structure.
16. The three-dimensionally integrated structure of claim 13,
wherein the microcomponents of the first plurality of
microcomponents comprise image sensors.
17. The three-dimensionally integrated structure of claim 16,
wherein the at least some microcomponents of the second plurality
of microcomponents comprise color filters.
18. The three-dimensionally integrated structure of claim 13,
wherein the microcomponents of the second plurality of
microcomponents comprise at least one of contact points and
interconnections.
19. The three-dimensionally integrated structure of claim 13,
wherein at least one of the first layer and the second layer
comprises a semiconductor material.
20. The three-dimensionally integrated structure of claim 13,
wherein at least one of the first layer and the second layer
comprises a material selected from the group consisting of silicon,
germanium, glass, quartz, and sapphire.
21. The three-dimensionally integrated structure of claim 13,
wherein the first layer comprises an SOI structure.
22. The three-dimensionally integrated structure of claim 13,
wherein at least one of the surface of the first layer and the
first surface of the second layer comprises an oxide material.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 12,936,639, filed Nov. 15, 2010, pending,
which is a national phase entry under 35 U.S.C. .sctn.371 of
International Patent Application PCT/EP2009/060250, filed Aug. 6,
2009, published in English as International Patent Publication WO
2010/023082 A1 on Mar. 4, 2010, which claims the benefit under
Article 8 of the Patent Cooperation Treaty to French Application
Serial No. 08 55767, filed Aug. 28, 2008, the entire disclosure of
each of which application is hereby incorporated herein by this
reference.
TECHNICAL FIELD
[0002] The present invention relates to the field of producing
multilayer semiconductor wafers or substrates produced by
transferring at least one layer formed from an initial substrate
onto a final substrate, the transferred layer corresponding to a
portion of the initial substrate. The transferred layer may further
comprise all or part of a component or a plurality of
microcomponents.
BACKGROUND
[0003] More precisely, the present invention relates to the problem
of heterogeneous deformations that appear during transfer of a
layer from a substrate termed the "donor substrate" onto a final
substrate termed the "receiving substrate." Such deformations have
been observed in particular with the three-dimensional component
integration technique (3-D integration) that requires transfer of
one or more layers of microcomponents onto a final support
substrate, and also with the transfer of circuits or with the
fabrication of back-lit imagers. The transferred layer or layers
include microcomponents (electronic, optoelectronic, etc) produced
at least partially on an initial substrate, said layers then being
stacked on a final substrate that may optionally itself include
components. Primarily because of the very small size and the large
number of microcomponents present on a single layer, each
transferred layer must be positioned on the final substrate with
high precision in order to come into very strict alignment with the
subjacent layer. Further, it may be necessary to carry out
treatments on the layer after it has been transferred, for example
to form other microcomponents, to uncover the surface of the
microcomponents, to produce interconnections, etc.
[0004] However, the Applicant has observed that after transfer,
there are circumstances when it is very difficult if not impossible
to form additional microcomponents that are aligned with the
microcomponents formed before transfer.
[0005] This misalignment phenomenon is described with reference to
FIGS. 1A to 1E that illustrate an exemplary embodiment of a
three-dimensional structure comprising transfer, onto a final
substrate, of a layer of microcomponents formed on an initial
substrate, and formation of an additional layer of microcomponents
on the exposed face of the initial substrate after bonding. FIGS.
1A and 1B illustrate an initial substrate 10 on which a first
series of microcomponents 11 is formed. The microcomponents 11 are
formed by photolithography using a mask that can define pattern
formation zones corresponding to the microcomponents 11 to be
produced.
[0006] As can be seen in FIG. 1C, the face of the initial substrate
10 comprising the microcomponents 11 is then brought into intimate
contact with one face of a final substrate 20. Bonding between the
initial substrate 10 and the final substrate 20 is generally
carried out by molecular bonding. Thus, a buried layer of
microcomponents 11 is formed at the bonding interface between
substrates 10 and 20. After bonding and as can be seen in FIG. 1D,
the initial substrate 10 is thinned in order to remove a portion of
the material present above the layer of microcomponents 11. A
composite structure 30 is thus formed from the final substrate 20
and a layer 10a corresponding to the remaining portion of the
initial substrate 10.
[0007] As can be seen in FIG. 1E, the next step in producing the
three-dimensional structure consists of forming a second layer of
microcomponents 12 at the exposed surface of the thinned initial
substrate 10 or of carrying out additional technical steps on that
exposed surface in alignment with the components included in the
layer 10a (contact points, interconnections, etc). For the purposes
of simplification, the term "microcomponents" is used in the
remainder of this text to define devices or any other patterns
resulting from technical steps carried out on or in the layers that
must be positioned with precision. Thus, they may be active or
passive components, a mere contact point, or interconnections.
[0008] In order to form the microcomponents 12 in alignment with
the buried microcomponents 11, a photolithography mask is used that
is similar to that used to form the microcomponents 11. The
transferred layers, like the layer 10a, typically include marks
both at the microcomponents and at the section forming the layer
that are in particular used by the positioning and alignment tools
during the technical treatment steps, such as those carried out
during photolithography.
[0009] However, even using positioning tools, offsets occur between
some of the microcomponents 11 and 12, such as the offsets
.DELTA..sub.11, .DELTA..sub.22, .DELTA..sub.33, .DELTA..sub.44
indicated in FIG. 1E (respectively corresponding to the offsets
observed between the pairs of microcomponents 11.sub.1/12.sub.1,
11.sub.2/12.sub.2, 11.sub.3/12.sub.3 and 11.sub.4/12.sub.4).
[0010] Such offsets are not the result of elementary
transformations (translation, rotation or combinations thereof)
that could originate in inaccurate assembly of the substrates.
These offsets result from non-homogeneous deformations that appear
in the layer derived from the initial substrate while it is being
assembled with the final substrate. In fact, such deformations
involve local and non-uniform displacements of certain
microcomponents 11. In addition, certain of the microcomponents 12
formed on the exposed surface of the substrate after transfer
exhibit positional variations with those microcomponents 11 that
may be of the order of several hundred nanometers, or even of
micrometer order.
[0011] This phenomenon of misalignment (also termed "overlay")
between the two layers of microcomponents 11 and 12 may be a source
of short circuits, distortions in the stack, or connection defects
between the microcomponents of the two layers. Thus, when the
transferred microcomponents are imagers made up of pixels, and the
post-transfer treatment steps are aimed at forming color filters on
each of those pixels, a loss of the colorizing function is observed
for certain of those pixels.
[0012] The overlay phenomenon thus results in a reduction in the
quality and the value of the fabricated multilayer semiconductor
wafers. The impact of the phenomenon is increasing because of the
ever-increasing demand for miniaturization of microcomponents and
their increased integration density per layer.
[0013] Problems with alignment during fabrication of
three-dimensional structures are well known. The document by Burns
et al, "A Wafer-Scale 3-D Circuit Integration Technology," IEEE
Transactions on Electron Devices, vol 53, No 10, October 2006,
describes a method of detecting variations in alignment between
bonded substrates. The document by Haisma et al, "Silicon-Wafer
Fabrication and (Potential) Applications of Direct-Bonded Silicon,"
Philips Journal of Research, vol 49, No 1/2, 1995, emphasize the
importance of wafer flatness, in particular during polishing steps,
in order to obtain good quality final wafers, i.e. with as few
offsets as possible between the microcomponents. However, those
documents are concerned only with the problem of positioning the
wafers while they are being assembled. As explained above, the
Applicant has observed that even when the two wafers are perfectly
mutually aligned when put into contact (using marks provided for
that purpose), non-homogeneous displacements of certain
microcomponents occur following initiation of the bonding wave.
BRIEF SUMMARY
[0014] The invention aims to provide a solution that can limit
non-homogenous deformations which appear in a substrate during
transfer thereof onto another substrate.
[0015] To this end, the present invention proposes a method of
initiating molecular bonding, comprising bringing one face of a
first wafer or substrate to face one face of a second wafer or
substrate and initiating a point of contact between the two facing
faces, the method being characterized in that the point of contact
is initiated by applying mechanical pressure to one of the two
wafers, said pressure being in the range 0.1 MPa [megapascal] to
33.3 MPa.
[0016] Thus, by limiting the pressure applied to one of the two
substrates during initiation of a point of contact, the
non-homogeneous deformations caused in the wafer are reduced, while
carrying out bonding by molecular bonding over the whole surface of
the two wafers in contact.
[0017] By minimizing thereby the deformations normally caused by
application of a point of contact to produce bonding by molecular
bonding, the risks of overlay during subsequent formation of the
additional layers of microcomponents are substantially reduced.
[0018] In accordance with a first aspect of the invention, the
mechanical pressure is applied over a surface area of 1 mm.sup.2
[square millimeter] or less.
[0019] In accordance with a particular aspect of the invention,
initiation of the point of contact is achieved by applying a tool
to one of the two substrates, the tool having a contact surface
area on the substrate in the range 0.3 mm.sup.2 to 1 mm.sup.2 and
in that the bearing force exerted by the tool on the substrate is
in the range 0.1 N [newton] to 10 N.
[0020] The present invention also provides a method of producing a
composite three-dimensional structure, comprising a step of
producing a first layer of microcomponents on one face of a first
wafer or substrate and a step of bonding the face of the first
wafer comprising the layer of microcomponents onto a second wafer
or substrate, the method being characterized in that during the
bonding step, molecular bonding is initiated in accordance with the
molecular bonding initiation method of the invention.
[0021] The use of a molecular bonding initiation method of the
present invention can, during transfer of a layer of
microcomponents, eliminate or limit the phenomenon of overlay and
produce very high quality multilayer semiconductor wafers. The
layer of microcomponents may in particular include image
sensors.
BRIEF DESCRIPTION
[0022] FIGS. 1A to 1E are diagrammatic views showing the production
of a prior art three-dimensional structure;
[0023] FIG. 2 is a diagrammatic view of a molecular bonding
initiation method in accordance with one embodiment of the
invention;
[0024] FIGS. 3A to 3D are diagrammatic views showing the production
of a three-dimensional structure using the molecular bonding
initiation method of the present invention;
[0025] FIG. 4 is an organigram of the steps carried out during
production of the three-dimensional structure shown in FIGS. 3A to
3D;
[0026] FIGS. 5 to 7 show, highly diagrammatically, implementations
of the molecular bonding initiation method of the invention.
DETAILED DESCRIPTION
[0027] The present invention is generally applicable to the
production of composite structures including at least the bonding
of a first substrate or wafer onto a second substrate or wafer by
molecular bonding.
[0028] Bonding by molecular bonding is a technique that is well
known per se. It should be recalled that the principle of bonding
by molecular bonding is based on bringing two surfaces into direct
contact, i.e. without the use of a specific material (adhesive,
wax, brazing material, etc). Such an operation requires the
surfaces for bonding to be sufficiently smooth, to be free of
particles or of contamination, and to be sufficiently close
together for contact to be initiated, typically a distance of less
than a few nanometers. The attractive forces between the two
surfaces are then sufficiently high to cause molecular bonding
(bonding induced by the set of attractive forces (Van der Waals
forces) of electronic interaction between atoms or molecules of the
two surfaces to be bonded.
[0029] Molecular bonding is carried out by initiating a point of
contact on one wafer in intimate contact with another wafer in
order to trigger the propagation of a bonding wave from that point
of contact. The term "bonding wave" used here is the bonding or
molecular bonding front that propagates from the initiation point
and that corresponds to diffusion of the attractive forces (Van der
Waals forces) from the point of contact over the whole surface area
between the two wafers in intimate contact (bonding interface). The
point of contact is initiated by applying mechanical pressure to
one of the two wafers.
[0030] The Applicant has demonstrated that the relative
displacements between certain patterns or microcomponents in one
and the same wafer appear as a result of the step of molecular
bonding of that wafer onto another. More precisely, experiments
carried out by the Applicant have demonstrated that stresses
(tensile and/or compressive) are produced at the point of contact,
i.e. the region where the mechanical pressure is applied. These
stresses are the source of the non-homogeneous deformations
appearing in the wafer and as a result of the relative and unequal
displacements of certain patterns or microcomponents relative to
each other.
[0031] The Applicant has observed that the deformations are
principally localized at and around the point of contact and that
these deformations are elastic. These deformations may extend over
a radius of up to 15 cm [centimeter] about the point of pressure
application.
[0032] As a result, the present invention proposes controlling the
mechanical pressure applied at the contact point in order to limit
the stresses in this zone while allowing initiation and propagation
of a bonding wave between the two wafers in contact. In accordance
with the invention, the pressure applied at the point of contact is
in the range 0.1 MPa to 33.3 MPa. The initiation point may be
located anywhere on the wafer. It is preferably located close to
the edge thereof. The surface area of the zone of application of
this pressure is typically less than a few mm.sup.2, for example 1
mm.sup.2. Larger application surface areas are possible but run the
risk that too large a contact surface area (more than 5 mm.sup.2,
for example) could result in aggravation of the deformation
(overlay).
[0033] The application of such a mechanical pressure is sufficient
to initiate a point of contact between two wafers and as a result
to allow the propagation of a bonding wave over the whole contact
surface between the wafers without causing stresses that are too
high. Thus, by controlling the mechanical pressure applied to
initiate the point of contact, the deformations arising in the
wafer are reduced. Preferably, the pressure applied at the point of
contact is less than 10 MPa; more preferably, this pressure is in
the range 0.1 MPa to 5 MPa.
[0034] The period during which the mechanical pressure is applied
corresponds to at least the minimum period that can activate the
phenomenon of propagation of the bonding wave. This minimum period
substantially corresponds to the period necessary for the bonding
wave to propagate over the contact surface between the wafers. The
mechanical pressure application period is generally between 1 and
10 seconds, typically 5 seconds, in order to assemble wafers with a
200 mm diameter.
[0035] The controlled application of mechanical pressure may be
carried out using a tool. In FIG. 2, a first wafer or substrate 60
is placed in a bonding machine comprising a substrate support
device 40. The substrate support device 40 comprises a support
platen 40a the planarity defects of which are preferably less than
15 microns (.mu.m). The support platen 40a holds the first wafer
60, for example by means of an electrostatic or suction system
associated with the support platen 40a or simply under gravity,
with a view to assembling it by molecular bonding, with a second
wafer or substrate 70. The associated systems for holding the wafer
(electrostatic or by suction) are used provided that it has been
ascertained that they do not deform the wafer so as not to
accentuate problems with overlay.
[0036] As explained above and in known manner, the surfaces 61 and
71 respectively of wafers 60 and 70 that are to be bonded have been
prepared (polishing, cleaning, hydrophobic/hydrophilic treatment,
etc) in order to allow molecular bonding.
[0037] The surfaces 61, 71 of wafers 60, 70 are then brought into
intimate contact with each other. Initiation of the point of
contact for molecular bonding is carried out by means of a tool 50.
As shown in a highly diagrammatic manner in FIG. 2, the tool 50
comprises a bearing element 51, such as a stylus, and a dynamometer
53. The bearing element 51 is connected to the dynamometer 53 and
comprises a free end 52 via which a mechanical pressure is exerted
on the wafer 70 in order to initiate a point of contact between the
two wafers 60 and 70. The end 52 has a contact surface area 52a
that is in the range 0.3 mm.sup.2 and 1 mm.sup.2. Knowing the area
of the contact surface 52a of the tool 50 with the wafer 70, it is
possible to apply a mechanical pressure in the range 0.1 MPa to
33.3 MPa by controlling the bearing force F exerted by the tool on
the wafer (bearing force=mechanical pressure.times.bearing surface
area). The bearing force exerted by the end 52 on the wafer 70 is
controlled by the dynamometer 53. This force is in the range 0.1 N
to 10 N.
[0038] As an example, if a mechanical pressure of 3.5 MPa (pressure
sufficient to initiate a point of contact and, as a result, a
bonding wave between the two wafers) is to be applied with a tool
the end of which has a contact surface area of 1 mm.sup.2, a
bearing force of 3.5 N is applied, for instance for about 6
seconds.
[0039] The bearing element, and more particularly its end intended
to come into contact with the wafer may be produced from or covered
with a material such as Teflon.RTM., silicone or a polymer. In
general, the end of the bearing element is produced from or coated
with a material that is sufficiently rigid to be able to apply the
pressure in a controlled manner. Too flexible a material would
deform and result in an imprecise contact surface and as a result
in a lack of precision in the applied pressure. In contrast, too
rigid a material could result in the formation of defects
(impression) on the surface of the wafer.
[0040] The molecular bonding initiation method of the invention may
be carried out automatically in a bonding machine. The machine then
comprises a bearing element connected to an actuator (for example a
cylinder or a mechanical arm). In some embodiments, the machine has
the ability to position the bearing element at any location on the
surface of the wafer, or along a diameter or a along a radius of
the stack formed of the bonded wafers. The machine also comprises a
force sensor (dynamometer, stress gage, etc) and a servocontrol
intended to drive the actuator. The servocontrol drives the
actuator in a manner that controls the mechanical pressure applied
by the bearing element. More precisely, the servocontrol receives
data from the force sensor and compares them with a predetermined
value for the bearing force that is a function of the mechanical
pressure that should be applied and of the surface area of the end
of the bearing element. The machine may also comprise a measurement
system to determine the wafers deformation (such as bow and warp
measurements). As it will be understood from the discussion below,
low pressure to initiate the bonding wave (for instance below 1
MPa), could be achieved if the initiation location is positioned at
predetermined specific positions.
[0041] Preferably, the wafers have limited bow deformation. It may
be difficult to initiate, in a repeatable manner, the development
of the bonding wave with the limited pressure of the invention (in
particular when the pressure is selected so as to be less than 10
MPa, or in the range 0.1 to 5 MPa). Tests carried out by applying a
force of 3.7 N close to the edge of the wafer, by using a stylus
with a contact surface having an area of approximately 1 mm.sup.2
have shown that the acceptable deformation for wafers or substrates
in order to ensure good bonding should be in the range -10 .mu.m to
+10 .mu.m for the final substrate or wafer (support substrate) with
a 200 mm diameter and in the range -45 .mu.m to +45 .mu.m for the
initial substrate or wafer comprising components (this wafer having
a wider range of tolerable deformation since deposition of the
oxide or anything of any other nature that is carried out on the
components in order to facilitate the molecular bonding step
introduces additional deformation). These measurements were carried
out by means of a capacitative measurement using ADE type equipment
from KLA-Tencor Corporation. These limiting deformation values to
ensure good bonding correspond to a constant bow to wafer diameter
ratio. As a result, they are also valid for wafers with a larger
diameter (for example 300 mm), with the appropriate correction for
taking into account the larger diameter value.
[0042] When one or both of the wafers carries a deformation, it may
be advantageous to select the location at which the point of
contact (place of application of the mechanical pressure) is
initiated as a function of the shape of the wafers that are brought
into contact in order to further minimize the mechanical pressure
necessary for molecular bonding. If the two wafers to be bonded are
not perfectly planar, the local mutual separation of the surfaces
of the facing wafers will not be constant. Thus, as shown in a
highly diagrammatic manner in FIG. 5, a wafer 520, for example a
circuit wafer having a concave deformation, must be bonded by
molecular bonding onto a planar wafer 510, for example a bulk wafer
that may have been oxidized. The point of contact initiated by
application of a mechanical pressure Pm is then preferably located
at point A, namely at the center of the concave deformation, rather
than at point B since the mechanical pressure that has to be
applied to initiate bonding would be higher in absolute terms at
point B than at point A and as a result would produce greater
deformations. A test has been carried out with wafers of substrates
similar to those illustrated in FIG. 5. In this test, a force of
0.3 N is applied during about 2 seconds at the center of the
generally concave wafer or substrate 520 positioned in contact with
the generally flat wafer or substrate 510. This generally flat
wafer or substrate 510 is placed itself on a flat wafer/substrate
support device of a bonding machine. In this particular
configuration, the limited force is sufficient to initiate the
bonding wave while minimizing the wafers deformation.
[0043] The point of contact could also be selected such that it
corresponds to a location where the wafer support device and the
supported wafer are in close contact or at the shortest distance
from one another, in particular when the supported wafer presents
at least a concave or convex deformation.
[0044] These last two requirements insure that the necessary
pressure that is applied to initiate the bonding will lead to a
minimal vertical displacement of the wafers and thus to minimize
wafer deformation.
[0045] Similarly, as shown in FIG. 6, when a wafer 620, for example
a circuit wafer, has a more complex deformation, i.e. with several
concave and convex portions relative to another flat wafer 610,
then preferably the point of contact and thus application of the
mechanical pressure Pm is initiated at the center of the concave
zones of the wafer 620. The center of the concave zones corresponds
to the regions of the wafers where the distance between these
regions and the flat wafer 610 is the smallest and as a result
requires the application of a lower mechanical pressure than at
other zones on the wafers. And preferably, the mechanical pressure
is applied at a location where the substrate support device is in
close proximity of the wafer 610 to avoid any vertical displacement
of the bonded stack while the bonding wave is being displaced.
[0046] In FIG. 7, the two wafers to be assembled 710 and 720 each
have their own deformation. The choice of the point of application
of mechanical pressure Pm to initiate the point of contact between
the two wafers is determined as a function of the position of the
two wafers when they are placed facing each other.
[0047] Information of the wafers shapes collected from the wafer
deformation measurement system could be used to determine the most
appropriate location.
[0048] In a variation, for example where a tool (stylus) used to
initiate the point of contact cannot be displaced relative to the
wafers, a predetermined deformation may be imposed on one or both
wafers so that the zone of the wafer present beneath the tool
corresponds to the point requiring the least mechanical pressure.
Under such circumstances, with a bonding machine in which the tool
is in a fixed position above the center of the wafers, it may, for
example, be possible to impose on the upper wafer a deformation
similar to that of FIG. 5.
[0049] The process of the invention is applicable to assembling any
type of material that is compatible with molecular bonding, in
particular semiconductor materials such as silicon, germanium,
glass, quartz, sapphire, etc. The wafers to be assembled may in
particular have a diameter of 100 mm, 150 mm, 200 mm or 300 mm. The
wafers may also include microcomponents on the majority of their
surface or only in a limited zone.
[0050] One particular but non-exclusive field for the assembly
method of the present invention is that of producing
three-dimensional structures.
[0051] One method of producing a three-dimensional structure by
transfer of a layer of microcomponents formed on an initial
substrate onto a final substrate in accordance with an embodiment
of the invention is described with reference to FIGS. 3A to 3D and
4.
[0052] Production of the three-dimensional structure commences by
forming a first series of microcomponents 110 on the surface of a
wafer or initial substrate 100 (FIG. 3A, step S1). The
microcomponents 110 may be entire components and/or only a portion
thereof. The initial substrate 100 may be a monolayer structure,
for example a layer of silicon, or a multilayer structure such as a
SOI type structure. The microcomponents 110 are formed by
photolithography using a mask that can define pattern formation
zones corresponding to the microcomponents 110 to be produced.
During the formation of microcomponents 110 by photolithography,
the initial substrate 100 is held on a substrate support device
120. The substrate support device comprises a support platen 120a
with which the initial substrate 100 lies flush, for example by
means of an electrostatic or suction system associated with the
support platen 120a.
[0053] The face of the initial substrate 100 comprising the
microcomponents 110 is then brought into contact with one face of a
final wafer or substrate 200 (step S2) with a view to bonding by
molecular bonding. A layer of oxide, for example of SiO.sub.2, may
also be formed on the face of the initial substrate 100 comprising
the microcomponents 110 and/or on the face of the final substrate
200 intended to be brought into intimate contact.
[0054] In accordance with the invention, the point of contact is
initiated between the two substrates by applying a mechanical
pressure Pm on the substrate 200, preferably close to the edge
thereof (step S3, FIG. 3B). As indicated above, the pressure Pm is
in the range 0.1 MPa to 33.3 MPa and applied to a bearing surface
of 1 mm.sup.2 or less.
[0055] Initiation of the point of contact involves propagating a
bonding wave on the interface between the initial substrate 100 and
the final substrate 200. The two substrates are then bonded
together by molecular bonding over the whole of their contact
surface (bonding interface), without or almost without deformation
in the initial substrate 100 comprising the microcomponents 110.
This thereby produces a buried layer of microcomponents 110 at the
bonding interface between the substrates 100 and 200.
[0056] After bonding and as can be seen in FIG. 3C, the initial
substrate 100 is thinned-down in order to remove a portion of the
material present above the layer of microcomponents 110 (step S4).
When the substrate 100 is a SOI type substrate, it is
advantageously possible to use the buried insulating layer to
define the thickness of the remaining layer 100a. Thus, a composite
structure 300 is produced, formed from the final substrate 200 and
a layer 100a corresponding to the remaining portion of the initial
substrate 100. The initial substrate 100 may in particular be
thinned-down by chemical-mechanical polishing (CMP), chemical
etching, or by splitting or fracture along a plane of weakness that
has been formed in the substrate by atomic implantation.
[0057] As can be seen in FIG. 3D, the next step in producing the
three-dimensional structure consists of forming a second layer of
microcomponents 140 at the exposed surface of the thinned-down
initial substrate 100 (FIG. 3D, step S5). The microcomponents 140
may correspond to complementary portions of the microcomponents 110
to form a finished component and/or to distinct components intended
to function with the microcomponents 140. In order to form the
microcomponents 140 in alignment with the buried microcomponents
110, a photolithography mask is used that is similar to that
employed to form the microcomponents 110. As during formation of
the microcomponents 110, the composite structure 300 formed by the
final substrate 200 and the layer 100a is held on a support platen
130a of a substrate carrier device 130 that is identical to the
device 120. The photolithography mask is then applied to the free
surface of the layer 100a.
[0058] In a variation, the three-dimensional structure is formed by
a stack of layers, each layer having been transferred by the
assembly method of the present invention, and each layer being in
alignment with the directly adjacent layers.
[0059] By using the molecular bonding initiation method of the
invention, the initial substrate 100 can be bonded onto the final
substrate without deformation or at least with a reduction in the
deformations in such a manner that significant offsets of the
microcomponents 110 before and after transfer of the initial
substrate 100 onto the final substrate 200 are no longer observed.
Thus, these residual offsets can be limited to values of less than
200 nanometers (nm), or even 100 nm in a homogeneous manner over
the whole surface of the wafer. The microcomponents 140, even those
with very small sizes (for example <1 .mu.m), may thus be formed
easily in alignment with the microcomponents 110, even after
transfer of the initial substrate. This, for example, means that
the microcomponents present in two layers or those on two distinct
faces of a single layer can be interconnected via metal
connections, minimizing the risks of poor interconnection.
[0060] As a result, the assembly method of the present invention
can eliminate the phenomenon of overlay during transfer of one
circuit layer onto another layer or onto a support substrate and
produce very high quality multilayer semiconductor wafers.
* * * * *