WO2000075968A1

WO2000075968A1 - Method for transferring elements and device enabling said transfer

Info

Publication number: WO2000075968A1
Application number: PCT/FR2000/001507
Authority: WO
Inventors: Michel Bruel
Original assignee: Commissariat A L'energie Atomique
Priority date: 1999-06-02
Filing date: 2000-05-31
Publication date: 2000-12-14
Also published as: FR2794443B1; FR2794443A1; JP2003501827A; EP1183714A1

Abstract

A handle (20) comprising receiving surfaces (7) and expandable or elastic means (4) whereby the receiving surfaces (7) can be moved away or moved closer in a uniform manner. The handle (20) is used to transfer elements such as pads (1) or chips (1, 10, 9, 42) from a transmitting substrate (2) to a receiving substrate (3) with a modification of the pitch.

Description

METHOD FOR TRANSFERRING ELEMENTS AND DEVICE FOR PROVIDING SAME

DESCRIPTION

Field of the invention

The invention relates to the field of transfer of elements from an emitting substrate to a receiving substrate. It relates to a transfer handle, to a transfer method using the handle according to the invention and to a method of manufacturing multifunction chips using handles according to the invention.

Prior art

It is known to transfer chips from a first substrate to a second substrate by means of what is commonly called a handle and which is in fact an intermediate substrate which can be reused or not. It is also known to transfer chips from a first substrate directly onto chips from a second substrate in order to carry out chip-to-chip hybridization between the chips of the first substrate and the chips of the second substrate. Collective transfers are always made between chips at the same steps in the direction in X and in the direction in Y respectively.

Brief description of the invention

The subject of the invention is a device and a method for the collective transfer of elements from an emitting substrate on which they are spaced apart. a first step in a first direction and a second step in a second direction, towards a receiving substrate, the elements being distributed on the receiving substrate with a second step in the first direction and a second step in the second direction.

By element is meant both a wafer, a chip or even a sub-chip, depending on the intended applications.

The term wafer includes any monolayer or multilayer structure capable of comprising functional entities.

A wafer may be subdivided into identical or different chips, each chip comprising one or more functional entities which can be grouped into identical or different sub-chips; the sub-chips can be separated.

In biochemical and chemical applications, subchips are generally called studs. The invention allows, in particular, the production of chips comprising pads of different functions from chips comprising only single-function pads. Thus, the invention allows the transfer of both platelets from an emitting support to a receiving support, the transfer of chips from a wafer onto one or more wafers which may themselves include chips, or even the transfer of sub- chips, a chip on one or more chips. The invention applies in particular to collective hybridization, for example chip on chip. According to the invention, the elements of a substrate known as an emitting substrate, on which the chips or pads are organized in a network of steps di in a first axial direction X, and d ₂ in a second axial direction Y are transferred collectively from the substrate emitter towards an intermediate substrate or handle comprising means of bringing together and / or of moving away.

When this first transfer is carried out, the approximation and / or distance means are actuated in the direction X and / or in the direction Y, until the new steps Di in the direction X and D _{2 are} obtained in the direction Y .

When this approximation and / or distance is achieved, the terminal transfer is made to the receiving substrate according to known methods.

Depending on the means of approximation and / or distance employed, it may happen that the setting to the new steps is carried out in several times and for example twice; in this case, the step Di is carried out for example by a first transfer from the emitting substrate to a first intermediate substrate comprising approximation and / or distance means, then when the approximation or the distance has been carried out for setting step Di, a new transfer is made to the second step D ₂ , to a second intermediate substrate comprising means for bringing together and / or moving away. Thus, the invention relates to a handle for the transfer of elements arranged in line at a spacing di, or in a matrix network along the spacing di in a first direction X and a spacing d ₂ in a second direction Y, the handle being provided with at least as many receiving surfaces as elements to be transferred arranged at a pitch di in the first direction and optionally a step d ₂ in the second direction, handle characterized in that it comprises:

- Means for approaching and / or distancing the receiving surfaces from one another, and able to move the receiving surfaces to give them a variable pitch at least in the first direction.

It should be noted that the directions X and Y are generally perpendicular to each other. This provision is however not compulsory. In a first embodiment, the handle comprises an extensible membrane and the simple act of tensioning the membrane causes uniform spacing of the elements with respect to each other in the direction or directions of traction of the membrane. In a parent embodiment of this first embodiment, the handle includes an elastic membrane so that traction on the membrane allows the deposited elements to be separated. It is also possible using this membrane to bring the elements together uniformly. For that, it is enough to pretend the membrane before depositing the elements then to release the tension to obtain the approach.

In a second embodiment, the receiving surfaces of the elements comprise supports each having an upper surface. The means of removal and / or reconciliation include elastic means connecting at least, in the first direction, two consecutive supports and means for applying traction on external supports, at least in the first direction. The traction means may comprise traction rods integral with each first and last support of a line of supports, movable in translation parallel to the columns.

They may further comprise traction rods integral with each first and last support of a column of supports, movable in translation parallel to the lines. In a variant of this embodiment, the receiving surfaces are capable of receiving a liquid. They are for example hollow or porous.

Such handles can be used in particular to make multifunctional chips from single-function chips. It is assumed, for example, that the pads of a multifunction chip are distributed according to the same step di in a direction along an axis X and according to the same step d ₂ along an axis Y.

If we want to make N multifunction chips of n = neither n ₂ pads each having neither columns, therefore a dimension ni.di in the direction X, and n ₂ lines therefore a dimension n ₂ .d ₂ in the direction Y proceed as follows. Single-function chips are produced having a number of pads at least as large as the number N of multifunction chips that one wishes to produce. For example, if we want N multifunction chips each having n pads, each pad having a specific function, we will make n chips monofunction each having N similar studs. The single-function chips will for example have n ₃ columns and n ₄ rows with n ₃ .n ₄ = N.

We will thus make N multifunction chips for example arranged in a matrix according to n ₃ columns and n ₄ lines, each chip having n _x columns and n ₂ lines.

To obtain these multifunction chips, we proceed as follows:

- We will first transfer the N pads of the first single-function chip to a handle provided with receiving surfaces and means of approximation and / or distance;

- We will bring or move the receiving surfaces in the direction X, so that two consecutive columns are separated from each other by a distance ni di possibly increased by the width of a cutting path C;

- We will then bring the receiving surfaces closer or further away in the Y direction to separate two consecutive lines by a distance n ₂ d ₂ possibly increased by the width of a cutting path C;

- we will transfer the N plots to the final substrate.

The two steps of approaching or moving away in the directions X and Y respectively can be carried out successively or simultaneously.

This gives, for example, the studs in the upper left corner of each multifunction chip. We have ₃ columns separated from each other by n _x .dι + C (C being the width of the cutting path) and n ₄ rows separated from each other by n ₂ . d ₂ + C. It is enough to repeat the same operations for the Nl remaining single-function chips, the insertion of the pads of rank i (i <ni) in the direction X and of rank j (j <n ₂ ) in the direction Y by shifting the receiving substrate of (il) dι in the x direction and (jl) d ₂ in the y direction.

Naturally if the studs do not all have the same dimension di in the direction x but dimensions of, d _i2 .... dim and in the direction y of the dimensions d ₂ ι .... d ₂ --2, then it should not be replaced or. di by di, (il) dι by T. dli and iin ϋ - 1 replace n ₂ . d ₂ by ^ d _2] and (j -l) d ₂ by ^ d _2] . ii

If we use the variant with receiving surfaces capable of receiving a liquid, and if it is the same handle that is used for the different transfers, we can plan after each transfer a cleaning step and then a collective deposit on all the receiving surfaces of a specific liquid medium of the next pads to be transferred.

Brief description of the drawings

The invention will now be described using the accompanying drawings, in which: - Figure 1 illustrates the purpose of

The invention;

- Figures 2, 3, 4, 5 and 6 illustrate a handle according to the invention and stages of its use for carrying out a transfer elements arranged on a first substrate in a first step towards a second substrate in a second step;

- Figure 7 shows a detail of the handle as shown in Figure

3;

- Figures 8 and 9 illustrate a second embodiment of a handle according to the invention;

- Figure 10 illustrates a first use of the handle and the method according to the invention; and

- Figure 11 illustrates a second use of the handle and the method according to the invention. It includes parts A and B.

Figure 1 illustrates the object of the invention. It is a question of ordering elements such as chips, sub-chips or pads 1 of a wafer which are ordered according to a first network in which a first step between two consecutive chips, (or if it is of pads of a chip, two consecutive pads) is di in a first direction and d ₂ in a second direction, as shown in part A of the figure, according to a second pitch D _x in the first direction and D ₂ in the second direction as shown in part B. In FIG. 1, the elements to be transferred are materialized by squares which bear the reference 1. In FIG. 1 the first steps di and d ₂ are smaller than the second steps Di and D ₂ but the first steps could be larger than the second or one larger and the other smaller. Referring now to FIG. 2, there is shown in section along a plane perpendicular to a substrate 2 and parallel to a row, chips 10 spaced from each other by a pitch d. The substrate 2 is an emitting substrate from which the chips 10 will be transferred, by means of a handle 20, to a receiving substrate 3 shown in FIGS. 5 and 6. According to the method of the invention, the chips will all 'first be transferred to a handle 20 shown in section of Figure 3. The handle 20 comprises on a support plate 6 an extensible or elastic membrane 4 whose ends are equipped with traction means 5. The support plate 6 can be circular or rectangular. The edges of the plate have a shape without sharp edges, for example rounded, so as not to injure the membrane 4. The traction means 5 may have an annular shape allowing simultaneous traction in all directions or be made up of four parts, two parts arranged symmetrically to provide traction in the X direction and two parts arranged symmetrically to provide traction in the Y direction. In FIG. 3 is shown the emitter substrate 2 and the chips 10 of the emitter substrate 2, in the position ready for transfer from the substrate 2 to the extensible or elastic membrane 4 of the handle 20. In known manner the chips 10 are transferred from the emitting substrate 2 to the receiving membrane 4, where they are received on the receiving surfaces 7 of the membrane 4, providing between them intermediate surfaces 8. As shown in FIG. 4, the membrane 4 is then towed, so that it is agra ndie in the direction or directions of traction. Due to the homogeneity of the membrane 4, it increases uniformly so that the chips 10 are distributed uniformly according to a new pitch D. If the membrane 4 is simply extensible without being elastic the dimension D is necessarily greater than the dimension d, on the other hand if the membrane 4 is elastic, it suffices to pretend it and then to let it contract after the transfer of the chips 10 onto the handle 20 to obtain a pitch D smaller than the initial pitch d. Obviously if the membrane 4 is elastic, it is also possible to obtain a greater pitch than the initial pitch by stretching the membrane 4 as in the case of the simply extensible membrane. When the chips 10 have been set to the desired pitch, they are transferred from the handle 20 to the receiving substrate 3, as shown in FIG. 5. When the transfer is complete, the chips 10 are found on the receiving substrate 3 at the desired pitch D as shown in the figure. 6. FIG. 7 represents an enlarged view of the traction means 5 shown diagrammatically in FIGS. 3 to 5. This means is constituted for example by jaws 11, 12 between which the membrane 4 is inserted. A clamping means 13 jaws 11, 12, for example a screw collar allows the jaws 11, 12 to be brought together and tightened on the membrane 4 over the entire width of the membrane. Tightening over the entire width in a direction perpendicular to the direction of traction allows a uniform extension in the direction of traction. For a bidirectional extension another jaw is provided perpendicular to the second direction of traction. The traction means 5 can also have an annular shape allowing extension in different directions by exerting a single traction. FIG. 8 schematically illustrates a second embodiment of a handle 30 according to the invention. According to this mode, the receiving surfaces 7 are individualized and of predetermined size. The receiving surfaces 7 are constituted by the upper surfaces of supports 15. In a particular embodiment, the upper surface 7 of the support is able to contain a liquid, it is for example concave or comprises a porous medium. The supports 15 are formed in a matrix network in lines in the first direction and in columns in the second direction. Each line of supports comprises a first support, a last support and intermediate supports. In the first direction a support is separated from the support immediately following by an intermediate space 19. All the intermediate supports have in the first direction a face which is opposite the face of another support. The first and last supports of each line have a face perpendicular to the first direction which is not opposite another face of a support. Likewise, each support column comprises a first support, a last support and intermediate supports. In the second direction a support is separated from the support immediately following by an intermediate space 21. All the intermediate supports have, in the second direction, a face which is opposite the face of another support. The first and last supports of each column has a face perpendicular to the second direction which is not opposite another face of a support. The external supports 15 of the handle 30 are mounted connected by traction rods 25, 24, oriented in the first direction, to first traction application means 27, 23 and 26, 22 respectively and by traction rods 35 , 34, oriented in the second direction to second traction application means 32, 36 and 33, 37 respectively. In each of the free spaces 19, 21 between two opposite faces of consecutive supports of the springs 18, 17 are connected to the supports 15 so as to exert a force on the support 15 in the first, (springs 18), and in the second directions , (springs 17), respectively. The first traction application means comprise traction bars 22, 23 perpendicular to the first direction placed on either side of all of the receiving surfaces. The pull rods 25, 24 link the external supports of each line to rings 27, 26 respectively slidably mounted on the pull bars 23, 22 respectively.

Similarly, the second traction application means comprise traction bars 32, 33 perpendicular to the second direction placed on either side of the set of receiving surfaces. Rods 35, 34 link the external supports of each column to rings 37, 36 respectively slidably mounted on the draw bars 33, 32 respectively. '' The operation of this handle 30 is illustrated in Figure 9. This figure shows the same handle 30 as that of Figure 8, in which the supports 15 have been spaced from each other. To simplify the representation, the supports have been shown smaller than in FIG. 8 and on a different scale than the other elements. When a traction F is exerted on the traction bars 22, 23 parallel to the first direction, the column rods 34, 35 integral with the rings 36, 37 respectively slide on the rods 32, 33 and move away uniformly one on the other under the action of the springs 18 inserted in the spaces 19 between supports 15. Likewise if a traction force F 'is applied to the traction bars 32, 33 the line rods 24, 25 inserted in the through holes 28 of the supports 15 slide on the rods 22, 23 to move away uniformly from each other, under the action of the springs 17 inserted in the spaces 21 inter supports 15. Locking means not shown inserted in the rings 27, 37 located in the corner, for example locking screws allow the supports 15 to be fixed in a determined configuration. Naturally, if one wishes to obtain a bringing together of chips or studs between the emitting substrate and the receiving substrate, it will suffice to move the supports 15 away from each other beforehand, to unlock the locking means, corner, and finally to maneuver the rods 22, 23, 32, 33 to the desired point. FIG. 10 illustrates a first possible use of a handle 20, 30 according to the invention for hybridize chips 9 of a donor matrix 14 arranged in a step di, d ₂ in the first and in the second direction respectively on chips 38 of a receiving matrix 39, arranged in a step Di, D ₂ in the first and in the second direction respectively. The chips 9 are transferred to a handle 20 or 30 according to the invention. The handle is not shown in FIG. 10. The receiving surfaces are put in step Di, D ₂ in the first and in the second direction respectively. The transfer onto the receiving matrix 39 is then carried out in order to obtain the hybridized matrix 40. It has thus been possible to hybridize AsGa chips of 1 mm * l mm ensuring radiofrequency functions (amplification, frequency change) on silicon chips of 8 mm * 10 mm providing all the logic and analog processing functions for the low and medium frequency radio signal.

Another use of the method and of handles 20, 30 according to the invention is illustrated in FIG. 11. This involves producing biochips 41. Each biochip 41 is composed of studs 42 that are functionalized. For example, each pad can carry one or a plurality of identical molecules (chemical or bio-chemical), called probes capable of selective hybridization reactions with target molecules brought into contact with the probes. These probes can for example be strands of DNA. In FIG. 11, with the aim of simplifying the figure, part A shows a single-function chip 43 comprising 12 pads 42 distributed in matrix fashion in 3 columns and 4 lines. For the purpose of simplification also the studs 42 are assumed to be squares of dimension d. It will be explained below how with twelve of these single-function chips 43, each carrying pads 42 of different functionalities, it is possible to produce twelve multi-function chips 41 distributed matrically according to 3 columns and 4 lines, each chip comprising 6 columns and 2 lines.

The pads 42 of the first of the single-function chips 43 are firstly transferred to a handle 20 or 30 according to the invention. The pads 42 are spaced from each other in the first direction by a distance equal to six times the distance d optionally increased by the width of a cutting path 44 provided on the terminal receiving substrate. The pads 42 are also spaced from each other in the second direction by a distance equal to twice the distance d optionally increased by the width of the cutting path 44 provided on the terminal receiving substrate. The pads 42 thus arranged are then transferred to a terminal receiving substrate 45 where they are in the configuration of the pads 42 shown in Figure 11B. It is sufficient to repeat the transfer and the spacings described above for five other single-function matrices 43 and to carry out the transfer on the receiving substrate 45 by shifting before each transfer the substrate 45 by the distance d in the first direction to obtain the first line of each of the twelve multifunction chips 41. The same operations are then repeated six times with the six remaining monofunction chips 43 having previously shifted the substrate 45 by the distance d in the second direction to obtain the twelve multifunction chips terminated as represented by the chip 41 placed in the upper left corner of the substrate 45. Naturally it is not compulsory for the chips 41 to comprise a number of pads which is an integer multiple of the number of columns or lines of the multifunction matrix 41. In this case one or more last lines will be incomplete.

The chips 41 can then be cut if necessary along the cutting path 44.

As already indicated above, if necessary, and if the handle 30 is used in its variant where the upper surfaces 7 of the supports 15 are able to contain a liquid, insert a cleaning step after the transfer step to the receiving substrate. This cleaning step can be followed by a collective depositing step on all of the receiving surfaces 7, of a specific liquid corresponding to the function of the pads which will be deposited immediately after.

With the method described above, 900 bio chips each of size 3 mm * 3 mm were produced, for example, each bio chip comprising 225 pads distributed in 15 rows and 15 columns, from 225 single-function matrices each comprising 900 identical pads of 200 * 200 μm distributed in 30 rows and 30 columns. The spacing to be made at the transfer handle is 3.1 mm in each direction, that is 15 * 200 μm + 100 μm for the cutting path.

Claims

1. Handle (20,30) for the transfer of elements (1) arranged in line according to a step di, or in a matrix network in lines and columns according to a step di in a first direction X and a step d ₂ in a second direction Y, the handle being provided with at least as many receiving surfaces (7) as elements (1) to be transferred, arranged in a step di in the first direction and possibly a step d ₂ in the second direction, handle characterized in that it includes:

- Means (5,4,7,8,17,18,22-27,32-37) for bringing the receiving surfaces (7) closer and / or further apart and able to move the receiving surfaces (7) to give them a variable pitch at least in the first direction.

2. Handle (20) according to claim 1, characterized in that the means for distance and / or approximation (4,5) of the receiving surfaces (7) comprise a membrane (4) extendable at least in the first direction.

3. Handle (20) according to claim 1, characterized in that the means (4,5) for approaching and / or moving away from the receiving surfaces comprise an elastic membrane (4) at least in the first direction, and the means approximation (4,5) of the receiving surfaces (7) being obtained by loosening the elastic membrane (4) which has been previously tensioned. ''

4. Handle (30) according to claim 1, characterized in that the receiving surfaces (7) comprise supports (15) each having an upper surface (7), the means for moving away and / or bringing together comprise elastic means (18,17) connecting at least, in the first direction, two consecutive supports (15) and traction application means (22-27; 32-37) on external supports (15), at least in the first direction.

5. Handle (30) according to claim 4, characterized in that the traction means (22-27; 32-37) comprise rods (24,25) of traction integral with each first and last support (15) of a line of supports (15), movable in translation parallel to the columns.

6. handle (30) according to claim 4, characterized in that the traction means (22-27; 32-37) further comprise traction rods (34,35) integral with each first and last support (15) a column of supports (15), movable in translation parallel to the lines.

7. Handle (30) according to one of claims 5 or 6, characterized in that the traction rods (24.25; 34.35) are integral with rings (26.27; 36.37) sliding in draw bars (32,33; 22,23) parallel to the pull direction of the draw rods (24,25 34,35).

8. Handle (30) according to one of claims 4 to 7, characterized in that the receiving surfaces (7) of the supports (15) are capable of receiving a liquid.

9. Method for transferring elements (1) arranged in line according to a step d _x in a first direction or in a matrix network according to a step di in the first direction and according to a step d ₂ in a second direction, the method consisting in first transfer the elements from the first substrate (2) to a handle (20,30) comprising at least as many receiving surfaces (7) as elements (1) to be transferred, then from the handle to a second substrate ( 3), process characterized in that:

- a handle is used, the receiving surfaces (7) of which are movable relative to one another by means of approximation and / or distance in the first and possibly in the second direction;

- the elements (1) of the first substrate (2) are transferred to the receiving surfaces of the handle, -

- the receiving surfaces (7) are moved away or closer to one another according to the first and possibly in the second direction;

- The elements (1) of the handle are transferred to the second substrate (3).

10. Method for manufacturing N chips (41) each having n pads (42) (Pi, P ₂ , .... P _n ) of functions (Fi, F ₂ , • - •• F _n ), each of the N chips (41) having n _x columns and n ₂ rows (nι.n ₂ = n) from n single-function chips each having N identical pads distributed according to n ₃ columns and n ₄ rows (n ₃ .n ₄ = N), the pads being distributed with a pitch di in a first direction and a pitch d ₂ in a second direction, the first chip having N pads P _x of function Fi, and the last chip having N pads P _n with function F _n , process characterized in that: a) the pads of the first monofunction chip are transferred to a handle comprising at least N receiving surfaces (1) arranged in a matrix according to n ₃ columns and n ₄ lines, the receiving surfaces (7,15) being movable relative to each other by means of approximation and / or distance; b) the approximation and / or distance means are actuated in the direction of the lines so as to space the columns between them by a distance at least equal to n _x times the step di, possibly increased by the width of a cutting path (44), - c) the means of bringing together and / or moving in the direction of the columns are actuated so as to space the lines between them by a distance at least equal to n ₂ times the step d ₂ , possibly increased by the width of a cutting path (44); d) the pads of the first chip of the handle are then transferred to the substrate for receiving the N chips; e) the operations a) to d) above are repeated (nι-1) times, with (nι-1) other single-function chips, each time moving the receiving substrate in the direction of the lines by a distance d _x so as to produce the first lines of the (n ₃ xn ₄ = N) multifunction chips (41); f) the operations a) to e) above are repeated (n ₂ -l) times, with nι (n ₂ -l) other single-function chips, for the (nl) other lines by each time moving the receiving substrate (45) in the direction of the columns by a distance d ₂ so as to produce the (n ₂ -l) other lines of the (n ₃ xn ₄ = N ) multifunction chips (41); g) optionally, the receiving substrate is cut according to the planned cutting paths (44).

11. The manufacturing method according to claim 10 characterized in that a transfer handle (30) is used according to claim 8, and in that prior to the transfer of the pads from at least one of the single-function chips, a collective deposit on all of the receiving surfaces 7, of a specific liquid corresponding to the function of the pads which will be deposited immediately after.