WO2023241931A1

WO2023241931A1 - Method for producing electronic assemblies, and wafer-level electronic assembly

Info

Publication number: WO2023241931A1
Application number: PCT/EP2023/064677
Authority: WO
Inventors: Markus Schindler; Robert Hofmockel; Severin Ringelstetter
Original assignee: Delo Industrie Klebstoffe Gmbh & Co. Kgaa
Priority date: 2022-06-14
Filing date: 2023-06-01
Publication date: 2023-12-21
Also published as: DE102022114911A1

Abstract

Disclosed is a method for producing wafer-level electronic assemblies (32), comprising the following steps: providing a radiation-penetrable base carrier (10) having a first side (10a) and a second side (10b); generating a temporary adhesive layer (12) by applying a first compound to the first side (10a) of the radiation-penetrable base carrier (10) so as to form a carrier substrate (14), wherein the first compound comprises a (meth)acrylate and a photoinitiator; placing at least one functional unit (16) onto the temporary adhesive layer (12) with an electrically contactable side of the functional unit (16); forming a temporary composite (24) consisting of the carrier substrate (14) and the at least one functional unit (16) by curing the first compound by irradiation with actinic radiation (22), wherein the irradiation is performed through the radiation-penetrable base carrier (10) from the second side (10b) of the base carrier (10); applying a potting layer (26) to the temporary composite (24) by metering a second compound onto the at least one functional unit (16) and onto the temporary adhesive layer (12), wherein the second compound is a compound that is curable by irradiation; forming a precursor (30) by curing the second compound by irradiation with actinic radiation (28); and detaching the carrier substrate (14) by introducing heat into the temporary adhesive layer (12) so as to form the wafer-level electronic assembly (32). Also disclosed are a wafer-level electronic assembly (32) and an electronic assembly.

Description

Method for manufacturing electronic assemblies and wafer-level electronic assembly

FIELD OF THE INVENTION

The invention relates to a method for producing wafer-level electronic assemblies, in particular wafer-level microelectronic assemblies, and a product manufactured using the method.

TECHNICAL BACKGROUND

Electronic assemblies are, for example, arrangements of processors and memory units, which can be manufactured at wafer level for cost reasons. A single electronic assembly may include multiple subassemblies that function together to perform a specific function.

The term “wafer level” here and below means that at least two electronic assemblies with their respective electronic functional units are structurally connected to one another and are separated in a downstream process step in order to obtain the individual electronic assemblies.

This manufacturing method, also known as “wafer-level packaging,” optimizes the manufacturing process in such a way that a large number of electronic assemblies can be arranged at small distances from one another and functional tests can be carried out on the entire wafer before the electronic assemblies are separated.

This is particularly advantageous if the electronic assemblies are microelectronic assemblies, the handling of which is made additionally difficult by the further reduced size of the respective microelectronic assembly. Both so-called “fan-out” and so-called “fan-in” wafer-level packaging are known.

In the “fan-out” package, conductor tracks and electrical connections of the electronic assembly are located not only below the base area of the respective functional unit, but also outside this base area. The conductor tracks are therefore “fanned out” compared to the base area of the functional unit.

In the “fan-in” package, however, the conductor tracks and electrically conductive connections are located exclusively below the base area of the functional unit.

To produce (micro)electronic assemblies at wafer level, in a conventional process a base carrier is provided on a first side with a temporary adhesive layer, which can be thermally removed. Functional units such as chips are attached to this temporary adhesive layer with their active, electrically contactable side, whereupon the temporary adhesive layer is hardened and the functional units are then encapsulated with a potting layer. With the application of heat, the temporary adhesive layer between the first side and the encapsulated functional units is removed again, so that the electrically contactable side is accessible for later contacting.

In principle, two technical problems arise in this process, which lead to faulty and/or poor-quality (micro)electronic assemblies at the wafer level or which negatively influence the subsequent process steps, namely the so-called “warpage” and the so-called “die shift “.

“Warpage” describes the warping of the wafer that occurs when the functional units are encapsulated with a potting compound. Particularly at high curing temperatures, due to the different thermal expansion coefficients of the base support, which can be a glass plate, for example, and the organic casting compounds, which usually have significantly higher thermal expansion coefficients, high stresses occur at the latest when cooling. This effect is all the more pronounced the larger the potted area is and can lead to severe warping of the wafer, which is already visible to the naked eye. However, for later process steps it is absolutely necessary that the wafer is flat. Further processing is usually carried out using a vacuum suction plate, which cannot pick up a warped wafer and transport it to the next station.

Further problems arise when later applying a redistribution layer (also known as a “redistribution layer” or “RDL”). The redistribution layer is used to reroute electrical connections to desired locations to provide better access when needed. For this purpose, for example, copper conductor tracks are applied using photolithographic process steps. For example, exposure via a mask to develop a photoresist or vapor deposition with copper on a curved surface is not possible evenly and precisely.

In addition to problems in downstream process steps, the “warpage” effect also has a negative influence on the reliability of individual (micro)electronic assemblies over their service life.

The term “die shift” refers to the shifting of the functional units of the (micro)electronic assembly on the carrier substrate, i.e. the composite of base carrier and temporary adhesive layer. Several factors promote the displacement of the functional units on the carrier substrate.

The warping of the wafer can be counteracted by using highly filled, thermosetting materials. Such compounds usually have a high viscosity, with the highly viscous casting compounds exerting a so-called drag force on the functional units during metering. This material property causes the functional units placed on the temporary adhesive layer to shift from their original position.

Furthermore, due to their mechanical properties, highly viscous potting compounds can only be applied and cured at very high temperatures and high pressures in the potting step. The so-called LCM process (also known as “liquid compression molding”) is known for this. Such a procedure also leads to the functional units moving away from their original position.

An existing “die shift” ensures, analogous to the “warpage” effect, that the redistribution layer can only be applied with insufficient precision or not at all.

To reduce the “warpage” effect, heat-curing resins, in particular epoxy resins, are primarily used, the coefficient of thermal expansion of which is adapted to the carrier substrate by adding fillers.

US 2015 0 152 260 A1 discloses epoxy resin matrices that contain at least 85% by weight of fillers. The fillers are predominantly of inorganic nature, such as colloidal SiC>2 nanoparticles. In conjunction with a high curing temperature between 120 °C and 150 °C, this composition can be used in the LCM process to significantly minimize the “warpage” effect. The disadvantage of this approach is that the high filling level results in unfavorable dosing properties of the mass and the freedom of formulation is severely limited. The liquid masses all have very high viscosities, so that they can only be dosed at points and not over the area. The problem here is that the pressing of the dosed mass at certain points creates so-called “flow marks”, i.e. the flow limits of the pressed mass points themselves become visually recognizable. At these limits, the hardened mass has different mechanical properties and chemical compositions, which do not correspond to the required quality. At the same time, the hardened encapsulation material has less flexibility, is stiff, brittle and prone to cracking. Due to the high pressure that must be applied in the LCM process, there is an additional risk that the functional units will shift.

In US 20200055979 A1, plate-shaped “prepregs” are used as the encapsulation material. Prepregs are fibers pre-impregnated with reaction resins (also known as “preimpregnated fibers”). In this case, a cellulose fleece is soaked with, for example, an SiO2-filled epoxy resin. The encapsulation is carried out by laminating the plate-shaped prepreg onto the functional units with pressure, for example with a pressure of 0.1 to 5 MPa. The warping of the wafer can be reduced with this approach, but the use of fibers still has the disadvantage of low flexibility. In order to achieve a complete positive connection between the functional units to be protected and the prepregs in the lamination process, high pressure is essential, which in turn can lead to the functional units being displaced.

US 9,853,000 B2 shows another approach. In the process described there, a “warpage” leveling layer, for example made of SiN or SiC, is applied to the package layers using a removable adhesive. By determining the curvature, it is determined where the adhesive needs to be removed so that the “warpage” effect can be compensated for and sufficiently reduced. Detachment can be done using laser ablation or UV light. The disadvantage of this approach is the high effort involved in treating and removing the compensation layer.

US 10211 072 B2 describes a method that prevents the functional units from being moved. After the functional units have been placed on a releasable adhesive layer on the base support, before applying a casting compound, a UV-curable immobilization compound is applied via a process gas, which comprises monomers and at least one initiator precursor. The initiator precursors are thermally converted into reactive initiators in order to polymerize the monomers to form the immobilization mass. This serves to fix the functional units on the removable adhesive layer and can be applied either completely or only selectively on the sides. A deviation from the original position can thus be significantly reduced or largely prevented. However, the process complexity increases due to the need to apply the additional immobilization mass.

In EP 3 940 764 A1, a base material layer, for example consisting of polypropylene (PP) or polyethylene terephthalate (PET), is provided on a first side with an adhesive layer made of (meth)acrylate, which serves to temporarily fix functional units. A second, thermally releasable adhesive layer containing heat-expandable particles is applied to a second side of the base material layer. On the second A glass carrier is attached to the adhesive layer. In a first step, the composite of functional units, three-layer structure and glass carrier is pre-hardened at temperatures below the release temperature of the heat-expandable particles (also referred to as the “trigger temperature”). This pre-curing step increases the adhesion force of the first adhesive layer with respect to the functional unit. In a subsequent casting step using the LCM process, the aim is to prevent the additionally fixed functional units from shifting.

If the shifting of the functional units in the process cannot be avoided or can only be solved inadequately, it is possible in a later, complex process step to compensate for the deviations during the application of the redistribution layer. For this purpose, the actual position of the functional units is determined and the template for the copper conductor tracks is aligned to the measured position. The copper conductor tracks are then applied dynamically adapted to the measured position of the functional units. This process is also called “Adaptive Patterning™”, but it involves a lot of effort.

What all of these approaches have in common is that each only addresses one of the technical problems. The solutions proposed in the prior art involve complex process steps and/or only reduce the problem of “die shifts” or the “warpage” effect to an insufficient extent.

SUMMARY OF THE INVENTION

The invention is based on the object of at least partially overcoming the disadvantages of the prior art and of providing a method through which cost-effective (micro-)electronic assemblies, in particular at the wafer level, can be produced easily and quickly. In particular, the invention is intended to achieve a simplification of the manufacturing process while simultaneously achieving high yields and good quality as well as reduced manufacturing costs.

The object of the invention is achieved by a method for producing electronic assemblies at wafer level according to claim 1. Further embodiments of the invention are specified in the subclaims, which can optionally be combined with one another.

The method according to the invention for producing electronic assemblies at the wafer level comprises the following steps: a) providing a irradiable base carrier with a first side and a second side, b) generating a temporary adhesive layer by applying a first mass to the first side of the irradiable base carrier with formation a carrier substrate, wherein the first mass comprises a (meth)acrylate and a photoinitiator, c) placing at least one functional unit with an electrically contactable side of the functional unit on the temporary adhesive layer, d) forming a temporary composite of the carrier substrate and the at least one functional unit by means of Hardening of the first mass by irradiation with actinic radiation, the irradiation taking place from the second side of the irradiable base support through the base support, e) applying a casting layer to the temporary composite by metering a second mass onto the at least one functional unit and onto the temporary Adhesive layer, wherein the second mass is a radiation-curable mass, f) forming a precursor by hardening the second mass by irradiating it with actinic radiation, and g) detaching the carrier substrate with heat input into the temporary adhesive layer to form the electronic assembly at the wafer level.

It was recognized that a particularly simple method for producing electronic assemblies at wafer level can be implemented by using both a temporary adhesive layer for the interim Fixing the at least one functional unit as well as in a mass for producing the potting layer, a radiation-curable mass is used. This mechanism for hardening the materials used is available for both materials according to the invention in that a base support that can be irradiated is used. At the same time, the use of a radiation-curable potting compound counteracts warpage and die shift effects. In this way, wafer-level electronic assemblies of high quality are obtained with at least a reduced error rate, thereby providing a more efficient and cost-effective process.

It is understood that in a downstream process step, individual electronic assemblies can be produced by separating the electronic assembly at wafer level.

The term “irradiable” base carrier refers to a base carrier that is transparent to the actinic radiation used to harden or harden the first mass and the second mass to such an extent that the first and/or the second mass can be hardened over the entire layer thickness of the masses can be achieved. The base support preferably has a transmittance for the actinic radiation of at least 75%.

The term “actinic radiation” refers to electromagnetic radiation that has a photochemical activity towards at least one component that is exposed to the electromagnetic radiation.

According to the invention, the second mass is metered not only onto the at least one functional unit, but also onto the temporary adhesive layer, so that surface areas of the temporary adhesive layer, which are arranged between the functional units, are also at least partially, in particular completely, coated with the second mass. In this way, a uniform casting layer is obtained in a single and easy-to-implement process step, which counteracts “die shift” particularly effectively.

The electronic assembly is preferably a microelectronic assembly. The method according to the invention is therefore used to produce an electronic assembly at the wafer level, preferably a microelectronic assembly at the wafer level.

The second side of the irradiable base carrier is arranged in particular opposite to the first side of the base carrier. In this way, it is particularly easy to harden the first mass as evenly as possible in order to form the temporary bond. In particular, the structure of a device for carrying out the method according to the invention can be further simplified.

The base support can be made of glass or plastic. The choice of the material of the base support depends in particular on the required mechanical stability that the base support provides during the manufacturing process, as well as the actinic radiation used to harden the first mass and the second mass, the base support being sufficiently affected by the actinic radiation used in each case is irradiable.

The plastic of the base support can be selected, for example, from the group of polysulfones, polyethersulfones, cycloolefin polymers, polyamides, polyimides, polyamide-imides, polyesters and polycarbonates as well as blends and/or copolymers of these polymers.

The thickness of the base support is in particular in a range between 100 pm and 3000 pm. Base supports with a thickness of less than 100 μm are difficult to handle, so that the effort required to carry out the method according to the invention with such base supports would be excessively increased.

The first mass is applied to create the temporary adhesive layer, for example by means of spin coating or stencil printing.

The layer thickness of the temporary adhesive layer is in particular in a range between 5 pm and 500 pm, in particular in a range between 10 pm and 300 pm, preferably in a range between 30 pm and 150 pm, taking into account the specified range limits. A thinner adhesive layer can result in insufficient adhesion of the functional unit(s). have, while greater layer thicknesses result in excessive use of material and make it difficult to harden the first mass evenly.

The functional unit(s) can be placed on the temporary adhesive layer using a so-called “pick & place” device.

The at least one functional unit can be selected from the group of active functional units and passive functional units.

Active functional units are controllable and able to amplify an incoming signal.

Examples of active functional units are transistors, diodes, rectifiers, processors, ICs (“integrated circuits”) and LEDs (“light emitting diodes”).

In contrast to active functional units, passive functional units have no reinforcing effect.

Examples of passive functional units are capacitors, potentiometers, oscillators, coils and resistors.

In the event that the method according to the invention is a “fan out” wafer level packing method, the functional units only have exposed contact points on their electrically contactable side.

If a “fan in” wafer-level packaging process is carried out instead, the functional units on their electrically contactable side have, in addition to the exposed contact points, electrically conductive connections that are assigned to the contact points.

The potting layer can be applied using a printing, slot die coating or UV stamping process. Such methods are known in the art and can be precisely controlled, so that a uniform dosage of the second mass for applying the potting layer can be ensured.

The hardening of the first mass and/or the second mass can be carried out by irradiation with actinic radiation of a wavelength in a range from 200 nm to 1000 nm, in particular in a range from 320 nm to 480 nm. It is understood that the first mass and the second mass can be hardened with actinic radiation of the same or different wavelength.

The second mass can be hardened over a large area and/or at specific points by scanning. Surface hardening, for example using a surface emitter, results in a particularly simple process control, while hardening by scanning, for example using a laser point source, enables particularly precise control of the hardening process.

The second mass is hardened in particular when forming the precursor.

“Curing” is defined as a polymerization or addition reaction beyond the gel point of the respective mass. The gel point is the point at which the storage modulus G' becomes equal to the loss modulus G'.

The first mass, hardened by irradiation with actinic radiation, can be removed by heating. By heating the hardened temporary adhesive layer, the bond between the carrier substrate and the electronic assembly is broken at the wafer level.

The carrier substrate can be removed by heating the hardened first mass to a release temperature between 50 ° C and 300 ° C, in particular to a release temperature between 100 ° C and 250 ° C, preferably to a release temperature between 180 ° C and 220 ° C, where the specified range limits are included. A release temperature below 50 °C increases the likelihood that the carrier substrate will detach unintentionally, while release temperatures above 300 °C result in increased energy consumption and thermal damage to the components of the precursor can occur.

The detachment temperature can be generated by irradiation with IR radiation of a wavelength in the range from 780 nm to 1000 nm, the IR radiation being generated in particular by an oven or a hot plate. This enables particularly simple process management and thus further reduces the costs of the process. In principle, a mechanical force can also be exerted on the carrier substrate in order to further accelerate the detachment.

The temporary adhesive layer is preferably removed without leaving any residue, so that additional grinding, washing or cleaning steps can be dispensed with.

In one variant, the temporary composite is tempered after application of the second mass at a temperature below the release temperature of the hardened first mass. The annealing step allows the mechanical properties of the first mass and the second mass to be adjusted in a targeted manner, for example in the case that further processing steps are provided between the formation of the precursor and the detachment of the carrier substrate, for which the precursor must be handled, and / or that the precursor should be stored for a specified time.

The materials used and the layers created using these materials are described in more detail below.

First mass

The temporary adhesive layer is provided using the first composition, the first composition comprising a (meth)acrylate and a photoinitiator, in particular a photoinitiator for radical polymerization.

The term “(meth)acrylate” in the sense of the invention includes both acrylates and the analogous methacrylates.

According to the invention, the first mass is a radiation-hardening mass.

The components of the first mass for producing the temporary adhesive layer are described in detail below.

Component (A) of the first mass: (meth)acrylate

The first mass contains at least one polymerizable component (A) based on a (meth)acrylate, in particular a free-radically polymerizable component based on a (meth)acrylate. The (meth)acrylates of component (A) are not further structurally restricted and include, for example, linear, cyclic, branched, aliphatic, aromatic and heterocyclic (meth)acrylates and combinations thereof.

(Meth)acrylic resins within the meaning of the invention are also monomeric, oligomeric or polymeric compounds, as long as they contain at least one radically crosslinkable (meth)acrylate group.

In one embodiment, the free-radically polymerizable compound (A) may comprise one or more monofunctional (meth)acrylates (A1).

Examples of monofunctional aliphatic (meth) acrylates are isobutyl (meth) acrylate, isooctyl (meth) acrylate, isodecyl (meth) acrylate,

Lauryl (meth) acrylate, isononyl (meth) acrylate, ethylhexyl (meth) acrylate, 2-propylheptyl (meth) acrylate, 4-methyl-2-propylhexyl (meth) acrylate,

Pentadecyl (meth)acrylate, cetyl (meth)acrylate, 2-hydroxyethyl (meth)acrylate,

4-hydroxybutyl (meth)acrylate, 2-hydroxy-3-phenoxypropyl (meth)acrylate,

2-Hydroxypropyl (meth)acrylate, Pal-mitolyl (meth)acrylate, Heptadecyl (meth)acrylate and Stearyl (meth)acrylate.

Examples of monofunctional cycloaliphatic (meth)acrylates are cyclohexyl (meth)acrylate, isobornyl (meth)acrylate, cyclic trimethylol propane formyl (meth)acrylate, dicyclopentenyl (meth)acrylate,

Dicyclopentenyloxyethyl (meth)acrylate, dicyclopentanyl (meth)acrylate,

4-tert-Butylcyclohexyl (meth)acrylate, 3,3,5.Trimethylcyclohexyl (meth)acrylate, tert-Butylcyclohexanol methacrylate and Octahydro-4,7-methano-1H-indenylmethyl (meth)acrylate.

Examples of monofunctional aromatic (meth)acrylates are 2-(o-phenylphenoxy)-ethyl (meth)acrylate, 2-(o-phenoxy)ethyl (meth)acrylate, ortho-phenylbenzyl (meth)acrylate, ethoxylated nonylphenol (meth) acrylate and ethoxy-phenyl acrylate.

Examples of heterocyclic, ethoxylated and other monofunctional

Meth(acrylates) are tetrahydrofurfuryl(meth)acrylate, (2-ethyl-2-methyl-1,3-

Dioxolat-4-yl)methyl acrylate, 5-ethyl-1,3-dioxan-5-yl)methyl acrylate, Caprolactone (meth)acrylate, 2-(2-ethoxyethoxy)ethyl (meth)acrylate and alkoxylated lauryl acrylate.

In addition to the monofunctional (meth)acrylates (A1), component (A) can also preferably comprise di- or higher-functional crosslinkers (A2), in particular based on (meth)acrylates.

Examples of di- or higher-functional (meth) acrylates are

Hexanediol di(meth)acrylate, di(trimethylpropane) tetraacrylate,

4-Butanediol di(meth)acrylate, tripropylene glycol di(meth)acrylate

T riethylene glycol di (meth) acrylate, polyethylene glycol di (meth) acrylate 1, 10-decanediol di (meth) acrylate, tricyclodecanedimethanol di (meth) acrylate, cyclohexane dimethylol di (meth) acrylate, nonanediol di (meth) acrylate,

Trimethylolpropane tri(meth)acrylate, tris-(2-hydroxyethyl)iso-cyanurate Tri(meth)acrylate, pentaerythritol hexa(meth)acrylate, dipentaerythritol hexa-(meth)acrylate, BPA diepoxypropane diacrylate and ethoxylated bisphenol A-di( meth)acrylate.

The (meth)acrylates mentioned are commercially available, for example, from the companies Arkema Sartomer, BASF, IMG Resins, Sigma Aldrich or TCI.

Urethane (meth)acrylates based on polyesters, polyacrylates, polyisoprenes, polyethers, polycarbonate diols and/or (hydrogenated) polybutadiene diols can be used as higher molecular weight, free-radically polymerizable compounds. These are typically di- or higher-functional.

In addition to a monofunctional (meth)acrylate (A1), component (A) preferably comprises an at least difunctional crosslinker (A2) based on an aliphatic and/or aromatic urethane (meth)acrylate.

Examples of suitable commercially available urethane (meth)acrylates are Visiomer HEMA-TMDI, available from Evonik, SUO-1020 NI (polycarbonate base) or SUO-H8628 (polybutadiene base), available from Shin-A T&C, CN9014NS, available from Sartomer, UV-3200B (polyester base), available from Nippon Goshei, or the XMAP types (polyacrylate base), available from Kaneka. Further free-radically polymerizable compounds (A) that can be used in the context of the invention are acrylic acid and methacrylic acid, acrylamides, acryloylmorpholines, bismaleimides, N-vinyl compounds such as vinylmethyloxazolidinone (VMOX), N-vinylcaprolactam, N-vinylpyrrolidone and N-vinylimidazole, as well as compounds with allyl groups, such as 1,3,5-triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-trione, which is commercially available as TAICROS®.

Unhydrogenated polybutadienes with free double bonds, such as the Poly BD® types, can also be used as free-radically polymerizable compounds.

The above list of suitable substance classes is intended as an example and not in a restrictive sense.

The component (A) is preferably present in the first compositions according to the invention in a proportion of 5 to 98% by weight, preferably from 10 to 90% by weight or from 15 to 85% by weight, in each case based on the total weight of the first mass.

The proportion of monofunctional (meth)acrylates (A1) in component (A) is preferably from 1 to 100%, more preferably from 1 to 95%, 1 to 80% or from 1 to 60%, in each case based on the total weight of the Component (A).

The proportion of at least difunctional crosslinkers (A2) in component (A) is preferably from 0 to 99%, more preferably 5 to 99%, 20 to 99% or 40 to 99%, in each case based on the total weight of component (A) .

In one variant, component (A) consists of the monofunctional (meth)acrylate (A1) and the at least difunctional crosslinker (A2) based on a (meth)acrylate and/or urethane (meth)acrylate.

Component (B) of the first mass: photoinitiator

In addition to component (A), the first compositions according to the invention comprise at least one photoinitiator (B), in particular a photoinitiator for radical polymerization. The usual, commercially available compounds can be used as radical photoinitiators (B), such as a-hydroxyketones, benzophenone, a,a'-diethoxyacetophenone, 2-benzyl-2-(dimethylamino)-4'-morpholinobutyrophenone, 4-isopropylphenyl -2-hydroxy-2-propyl ketone, 4,4-bis(diethylamino)benzophenone, 2-ethylhexyl-4-(dimethylamino)benzoate, ethyl 4-(dimethylamino)benzoate, 2-butoxyethyl-4-(dimethylamino)benzoate, 1-Hydroxycyclohexylphenyl ketone, isoamyl-p-dimethylaminobenzoate, methyl 4-dimethylaminobenzoate, methyl-o-benzoyl benzoate, benzoin, benzoin ethyl ether, benzoin isopropyl ether, benzoin isobutyl ether, 2-hydroxy-2-methyl-1-phenylpropan-1-one, 2- isopropylthioxanthone,

Dibenzosuberone, ethyl (3-benzoyl-2,4,6-trimethylbenzoyl)(phenyl) phosphinate, methyl benzoyl formate, oxime ester, 2,4,6-trimethylbenzoyldiphenylphosphine oxide, ethyl (2,4,6-trimethylbenzoyl) phenyl phosphinate and bisacylphosphine oxide.

Radical photoinitiators that can be activated by exposure to actinic radiation in the UV range are referred to as UV photoinitiators.

The IRGACURE™ types from BASF SE, for example, can be used as UV photoinitiators, for example the types IRGACURE 184, IRGACURE 500, IRGACURE 1179, IRGACURE 2959, IRGACURE 745, IRGACURE 651, IRGACURE 369, IRGACURE 907, IRGACURE 1300, IRGACURE 819 , Irgacure 819dw, Irgacure 2022, Irgacure 2100, Irgacure 784, Irgacure 250, Irgacure TPO, IRGACURE TPO -L Further can also be used by BASF SE, for example the ^types Darocur MBF, Darocur TPO and Darocur 4265 .

The radical photoinitiator used as component (B) in the compositions according to the invention can preferably be activated by exposure to actinic radiation with a wavelength of 200 to 600 nm, particularly preferably with actinic radiation with a wavelength of 320 to 480 nm.

If necessary, the radical photoinitiator can be combined with a suitable sensitizer. The radical photoinitiator (B) is present in the compositions according to the invention in a proportion of 0.01 to 5% by weight, based on the total weight of the first composition.

Component (C) of the first mass: Latent release agent

In addition to components (A) and (B), the first mass preferably comprises at least one latent release agent (C).

The latent release agent (C) serves to support the detachment of the temporary adhesive layer from the carrier substrate and/or the electronic assembly at the wafer level due to heat input. In other words, latent release agents (C) are components of the first mass which, by introducing thermal energy, weaken the cohesive properties of the first mass in such a way that detachment from the carrier substrate and/or the electronic assembly at the wafer level is facilitated or enabled.

The first mass particularly preferably contains thermo-expandable microcapsules (C1), which are a component of the latent release agent (C) or form the latent release agent (C).

Thermo-expandable microcapsules (C1) release a substance in a gaseous state at elevated temperatures and thus lead to a loss of cohesion in the hardened first mass. This allows easier removal of the carrier substrate as well as particularly precise control when detaching the carrier substrate due to the heat input.

The specific composition and production of suitable thermo-expandable microcapsules (C1) can be found, for example, in US 3,615,972 A, US 9,902,829 B2 or US 8,486,531 B2.

Commercial products are available, for example, under the trade name Expancel from Nouryon or under the name Matsumoto Microsphere® from Matsumoto Yushi Seiyaku Co. Japan Ltd. available.

By choosing the thermo-expandable microcapsules (C1) used, it is possible to specifically adjust the thermal energy required to detach the carrier substrate and to adjust the detachment temperature. The latent release agent (C) can alternatively or additionally contain or be a blowing agent based on a nitrogen compound (C2).

Blowing agents based on a nitrogen compound (C2) decompose at elevated temperatures with a high gas yield to form gaseous decomposition products, for example nitrogen, carbon monoxide, carbon dioxide, ammonia, hydrogen and combinations thereof.

The blowing agent based on a nitrogen compound (C2) can be selected from the group consisting of diazo compounds, sulfonyl hydrazides, tetrazoles, nitroso compounds, acylsulfonyl hydrazides, hydrazones, thiatriazoles, azides, sulfonyl azides, thiatriazine dioxides and combinations thereof.

For example, the blowing agent based on a nitrogen compound (C2) is 1, 1 '-azodicarboxamide, p-toluenesulfonylhydrazine, 5-phenyltetrazole, isatoic anhydride or a combination of these compounds.

Furthermore, the latent release agent (C) can contain or be a peroxide (C3), an isosorbide resin (C4) and/or an acetal resin (C5).

The latent release agent (C) is present in the first mass in particular in a proportion of 1 to 50% by weight, preferably from 5 to 45% by weight, more preferably from 10 to 40 or 15 to 25% by weight, each based on the total weight of the first mass.

Component (D): Additives

The first masses described can contain further optional components as additives (D) in addition to the previously described components (A) to (C).

Fillers (D1) can be used as additive (D). In addition to adjusting the layer thickness of the joint, these also allow the chemical resistance, media absorption, mechanical properties and the thermal expansion coefficient of the hardened first mass to be influenced. In order to achieve a low thermal expansion coefficient of the first mass, quartz and/or quartz glass can be used as filler (D1).

Materials with negative thermal expansion coefficients, such as zirconium tungstate, can also be used as filler (D1).

To achieve higher thermal conductivity, fillers (D1) such as aluminum oxide, aluminum nitride, boron nitride, graphite (including expanded graphite or graphite-based nanotechnology products), graphene, carbon nanotubes and/or metallic fillers can be used.

To achieve isotropic or anisotropic electrical conductivity, metallic fillers or non-metallic fillers coated with electrically conductive layers can be used as fillers (D1).

To achieve flame retardant properties, for example aluminum salts, phosphate esters and/or alkyl phosphonate salts can be added to the mass as fillers (D1).

To achieve defined adhesive layer thicknesses, so-called “spacer particles” with specified particle shapes and particle size distributions can be used as filler (D1).

The selection of fillers (D1) is in no way limited with regard to particle shapes (such as angular, spherical, platelet or needle-shaped, hollow shapes) and particle sizes (macroscopic, microscopic, nanoscale).

As is known, different particle shapes or particle sizes or particle size distributions can also be used in combination in order to achieve, for example, a low viscosity, a higher maximum filling level and/or a high electrical and thermal conductivity.

An effect enhancer (D2) for the latent release agent (C) can also be provided as an additive (D).

In this sense, an effect enhancer (D2) is understood to mean a substance that does not alone enable the temporary adhesive layer to be thermally resolvable, but rather enhances the effect of an existing latent release agent (C). The effect enhancer (D2) is preferably a plasticizer which increases the detachability of the first mass when thermal energy is applied.

In addition, further additives (D3) can be provided as additives (D).

The further additives (D3) are preferably selected from the group consisting of dyes, pigments, anti-aging agents, fluorescent agents, sensitizers, accelerators, stabilizers, adhesion promoters, drying agents, crosslinkers, flow improvers, wetting agents, thixotropic agents, non-reactive

Flexibilizers, non-reactive polymeric thickeners, corrosion inhibitors, plasticizers and combinations thereof.

In particular, the first mass for forming the temporary adhesive layer can contain compounds (D4) which improve the thermal absorption behavior of the masses, for example when irradiated with infrared light. This leads to a faster and more efficient heat input and thus to a simplified removal of the carrier substrate in the method according to the invention.

In particular, IR absorbers, as described in EP 3 943 534 A1, can advantageously be used as compound (D4) in additive (D).

According to the invention, the additives (D) are present in the first mass in particular in a proportion of up to 70% by weight, based on the total weight of the first mass.

In particular, the first mass comprises or consists of 5 to 98% by weight of the (meth)acrylate (A), 0.01 to 5% by weight of the photoinitiator (B), 1 to 50% by weight of the latent release agent (C) and 0 to 70% by weight of the additive (D), each based on the total weight of the first mass.

Second mass

The potting layer is provided using a second mass based on actinically radiation-curable compounds. Dual-curing materials are also considered radiation-curable materials, which also contain one Can be subjected to hot curing. For the purposes of the invention, however, exclusively heat-curing materials are not used as the second material.

The use of radiation-curable casting compounds enables a reduction in cycle times during the manufacturing process, as radiation curing can be carried out more quickly than thermal curing processes. In addition, by foregoing pure hot hardening, the thermal expansion of the second mass is reduced during the hardening process and thus the introduction of stresses into the components involved is reduced.

Furthermore, the use of radiation-curable materials as a second material allows the casting compound to harden at specific points in the sense of a preferred embodiment of the method according to the invention.

According to the invention, compositions which are dual-curing, i.e. are both actinically radiation-curable and additionally heat-curing, can also be used as a second composition.

In this variant, a post-hardening step can be provided, that is, additional hardening or hardening by means of hot hardening after the precursor has been formed.

Furthermore, heat hardening of the casting layer can occur as a result of the heat input when the temporary adhesive layer is removed.

The second mass for producing the potting layer is not subject to any further chemical restrictions as long as it is actinically radiation-curable.

In a first embodiment, the second mass comprises at least one (meth)acrylate and a photoinitiator for the radical polymerization. The second mass can therefore be formulated from components analogous to the first mass, although the second mass has no thermal detachability.

In other words, the second mass does not separate from the encapsulated functional units at the release temperature of the hardened first mass. In particular, in this first embodiment, component (C) is not present in the second mass. In other words, the second mass is in particular free of the latent release agent (C).

The second mass can in particular be composed of components (A), (B) and (D) of the first mass, with the same or different compositions being able to be used. With regard to suitable compounds of components (A), (B) and (D) for the second mass, reference is made to the examples of the corresponding components (A), (B) and (D) described previously for the first mass.

In this variant, the second mass comprises in particular 5 to 98% by weight of the (meth)acrylate (A), 0.01 to 5% by weight of the photoinitiator (B) and 0 to 90% by weight of the additive (D ).

More preferably, the second mass in this embodiment contains as additive (D) proportions of inorganic fillers in a proportion of up to 85% by weight, based on the total weight of the second mass. This leads to an additional reduction in shrinkage when the second mass is hardened, further minimizing the introduction of stresses into the components involved.

In a second embodiment, the radiation-curable second mass, and thus the casting compound, is cationically polymerizable. In other words, the second mass in the second embodiment comprises at least one cationically polymerizable component, which is described in more detail below.

Component (a) of the second mass: cationically polymerizable component

The cationically polymerizable component (a) is not further restricted in terms of its chemical basis.

The cationically polymerizable component is preferably selected from the group consisting of epoxide-containing compounds (a1), oxetane-containing compounds (a2), vinyl ethers (a3) and combinations thereof.

Optionally, the cationically polymerizable component can additionally contain one or more alcohols (a4) as chain transfer agents and/or cationically polymerizable hybrid compounds (a5). It is also possible to use cyclic lactones or carbonates as cationically polymerizable component (a).

The epoxide-containing compound (a1) can include aliphatic, aromatic and/or cycloaliphatic epoxy compounds.

The epoxide-containing compound (a1) in the second compositions according to the invention preferably comprises one or more at least difunctional epoxide-containing compounds.

At least “difunctional” means that the epoxy-containing compound contains at least two epoxy groups.

The cationically polymerizable component (a) preferably comprises at least one aromatic epoxy compound.

The group of aromatic epoxy compounds includes, for example, bisphenol A epoxy resins, bisphenol F epoxy resins, phenol novolak epoxy resins, cresol novolak epoxy resins, biphenyl epoxy resins, 4,4'-biphenyl epoxy resins, divinyl benzene dioxide, glycidyl phenyl ether, naphthalene diol diglycidyl ether, glycidyl ether of Tris( hydroxyphenyl)methane, p-tert-butylphenol glycidyl ether and glycidyl ether of tris(hydroxyphenyl)ethane, and mixtures thereof.

Furthermore, all fully or partially hydrogenated analogues of aromatic epoxy compounds can also be used as epoxide-containing compound (a1).

The group of cycloaliphatic epoxy compounds includes, for example, cyclohexenylmethyl-3-cyclohexylcarboxylate diepoxide, 3,4-epoxycyclohexyl-alkyl-3',4'-epoxycyclohexanecarboxylate, 3,4-epoxy-6-methylcyclohexylmethyl-3',4'-epoxy-6-methylcyclohexanecarboxylate , vinylcyclohexene dioxide,

Bis(3,4-epoxycyclohexylmethyl) adipate, dicyclopentadiene dioxide,

Diycyclopentadienyloxy-ethyl glycidyl ether, limonene dioxide and 1,2-epoxy-6-(2,3-epoxypropoxy)hexahydro-4,7-methanindane, and mixtures thereof.

Isocyanurates and other heterocyclic compounds substituted with epoxide-containing groups can also be used in the second ones according to the invention Masses can be used as component (a1). Triglycidyl isocyanurate and monoallyl diglycidyl isocyanurate may be mentioned as examples.

In addition, polyfunctional epoxy resins of all of the compound classes mentioned, tough-elasticated epoxy resins and mixtures of various epoxy compounds can be used in the second compositions according to the invention.

Also within the meaning of the invention is a combination of several epoxide-containing compounds, of which preferably at least one is di- or higher-functional.

In addition to the at least difunctional epoxide-containing compounds, monofunctional epoxides can also be used as reactive diluents.

Examples of commercially available monofunctional epoxides are products available under the trade name Glycirol ED 509-S from Adeka, D.E.R. 727 from Olin, Heloxy Modifier AQ from Hexion, Cardolite Ultra Lite 513 from Cardolite or iPox RD 17 from iPox Chemicals GmbH.

Examples of other commercially available di- or higher-functional aliphatic epoxy compounds are products available under the trade names iPox RD21, iPox CL60, iPox CL9 from ipox Chemicals GmbH or YED-216D from Mitsubishi Chemical, Japan or Heloxy Modifier HD from the company Hexion or Araldite DY 3601 from Huntsman.

Examples of commercially available cycloaliphatic epoxy compounds are products sold under the trade names CELLOXIDETM 2021 P, CELLOXIDETM 8000 from Daicel Corporation, Japan or Omnilane 1005, Omnilane 2005, Omnilane OC 3005 from IGM Resins B.V. or TTA21, TTA26 and TTA60 from Jiangsu Tetra New Material Technology Co. Ltd. or Syna Epoxy 21 from Synasia I nc. , to be expelled.

Examples of commercially available aromatic epoxy compounds are products sold under the trade names Epikote Resin 828 LVEL, Epikote Resin 166, Epikote Resin 169 from Hexion Specialty Chemicals BV, Netherlands or as EPICLONTM 840, 840-S, 850, 850-S, EXA850CRP, 850-LC by the company DIC KK, Japan.

Epoxy compounds containing basic groups that can inhibit cationic curing are not preferred. The second mass is preferably free of glycidylamines.

Alternatively or in addition to the epoxide-containing compound (a1), oxetane-containing compounds (a2) can also be used in the second mass as a component of the cationically polymerizable component (a). Processes for producing oxetanes are known in particular from US 2017/0198093 A1.

The cationically polymerizable component (a) preferably comprises at least one oxetane-containing compound (a2). In particular, low-viscosity oxetanes enable the advantageous use of the second mass according to the invention as a casting material.

Examples of commercially available oxetanes (a2) are bis(1-ethyl-3-oxetanylmethyl) ether (DOX), 3-allyloxymethyl-3-ethyloxetanes (AQX), 3-ethyl-3-[(phenoxy)-methyloxetanes ( POX), 3-Ethyl-3-hydroxymethyl-oxetane (OXA), 1,4-Bis[(3-ethyl-3-oxetanylmethoxy)methyl]benzene (XDO), 3-Ethyl-3-[(2-ethylhexyloxy) methyl]oxetane (EHOX). The oxetanes mentioned are from TOAGOSEI CO., Ltd. commercially available.

Likewise, in addition to component (a1) and optionally to component (a2), vinyl ethers (a3) can also be used as cationically polymerizable components in the second mass. Suitable vinyl ethers are trimethylolpropane trivinyl ether, ethylene glycol divinyl ether and cyclic vinyl ethers and mixtures thereof. Vinyl ethers of polyfunctional alcohols can also be used.

In addition, the cationically polymerizable component can also comprise one or more alcohols (a4), which are used as chain transfer agents. Higher molecular weight polyols in particular can be used to make cationic masses more flexible. Suitable polyols are available, for example, based on polyethers, polyesters, polycaprolactones, polycarbonates, polybutadiene diols or hydrogenated polybutadiene diols. Examples of commercially available higher molecular weight polyols are products sold under the trade names ETERNACOLL UM-90 (1/1), ETERNACOLL UHC50-200 from UBE Industries Ltd., as CapaTM 2200, CapaTM 3091 from Perstorp, as Liquiflex H from Petroflex, as Merginol 901 from HOBIIM Oleochemicals, as Placcel 305, Placcel CD 205 PL from Daicel Corporation, as Priplast 3172, Priplast 3196 from Croda, as Kuraray Polyol F-3010, Kuraray Polyol P-6010 from Kuraray Co., Ltd. , as Krasol LBH-2000, Krasol HLBH-P3000 from Cray Valley or as Hoopol S-1015-35 or Hoopol S-1063-35 from Synthesia Internacional SLU.

Finally, in addition to components (a1) to (a4), hybrid compounds (a5) can also be used. In addition to at least one of the cationically polymerizable groups mentioned, these also contain radically radiation-curable groups. In particular, epoxy (meth)acrylate hybrid compounds are within the meaning of the invention. Examples of commercially available epoxy (meth)acrylates are CYCLOMER M100 from Daicel, Epoxy Acrylat Solmer SE 1605, UVACURE 1561 from UCB, Miramer PE210HA from Miwon Europe GmbH and Solmer PSE 1924 from Soltech Ltd. Oxetane (meth)acrylates such as ETERNACOLL OXMA from UBE Industries LTD can also be used as a hybrid compound (A5).

The list of cationically polymerizable components (a) is to be seen as an example and not exhaustive.

A mixture of the cationically polymerizable components (a1) to (a5) mentioned is also within the meaning of the invention.

According to the invention, the cationically polymerizable component (a) is present in the second embodiment of the second mass, based on the total weight of the second mass, in particular in a proportion of 10 to 95% by weight, preferably in a proportion of 10 to 80% by weight. % or 15 to 75% by weight, each based on the total weight of the second mass.

Component (b) of the second mass: initiators for the cationic polymerization

In the second embodiment, the second mass according to the invention contains, in addition to component (a), at least one initiator (b) for the cationic Polymerization. This includes at least one acid generator, which can be activated by actinic radiation. In addition, the second mass can also contain a thermally activated acid generator as initiator (b).

The initiators (b) are also described below as “photolatent acids”, if they can be activated by actinic radiation, or “heat-latent acids”, if they can be activated thermally.

Suitable initiators (b) are, for example, acid formers based on metallocenium and/or onium compounds, which preferably serve as photolatent acids.

An overview of various metallocenium salts is disclosed in EP 0 542 716 B1. Examples of different anions of the metallocenium salts are HSO, PFe', SbFß', AsFß', Ck, Br, I; CIO4', PO , SOsCF , OTs' (tosylate), aluminates and borate anions such as BF and B(CeF5)4 ^_ .

Preferred metallocenium compounds are selected from the group of ferrocenium salts.

Preferred onium compounds are selected from the group of arylsulfonium salts and aryliodonium salts and combinations thereof and are described in the prior art.

Triarylsulfonium-based photoinitiators (b1) commercially available as photolatent acids are available under the brand names Chivacure 1176, Chivacure 1190 from Chitech, Irgacure 290, Irgacure 270, Irgacure GSID 26-1 from BASF, Speedcure 976 and Speedcure 992 from Lambson, TTA UV-692, TTA UV-694 from Jiangsu Tetra New Material Technology Co., Ltd. or UVI-6976 and UVI-6974 available from Dow Chemical Co.

Photoinitiators (b1) based on diaryliodonium that are commercially available as photolatent acids are available, among others, under the brand names LIV1240, LIV1242 or LIV2257 from Deuteron and Bluesil 2074 from Bluestar.

The photoinitiator (b1) used in the second mass according to the invention is preferably irradiated with actinic radiation of a wavelength in Can be activated in the range from 200 to 480 nm, more preferably at a wavelength of 250 to 365 nm.

According to the invention, a mixture of photoinitiators (b1) which can be activated at different excitation wavelengths can also be used.

A mixture of metallocenium-based photoinitiators, preferably ferrocenium-based, and onium compounds, preferably sulfonium compounds, is preferred.

The metallocenium-based photoinitiators may have an excitation wavelength in the range of 400 to 700 nm, preferably in the range of 430 to 500 nm.

The excitation wavelength of the onium compounds used as photoinitiators is in particular in the range from 200 to 380 nm, preferably from 300 to 380 nm.

If necessary, the photoinitiator (b1) can be combined with a suitable sensitizing agent.

In addition to the at least one acid generator that can be activated by actinic radiation, component (b) of the second mass can comprise a thermally activated acid generator (b2).

Thermally activated acid generators (b2) are known in the prior art. In particular, heat-latent acids based on an aromatic sulfonium salt can be used, which can be activated by heating and releases an acid suitable for the cationic polymerization of component (a).

In addition to at least one sulfonium cation, the thermally activatable acid generator (b2) can comprise a large number of anions. For example, antimonates, fluorophosphates, aluminates, titanates and borate anions, such as BF ₄ ' and B(C6F ₅ )4 ^_ are within the meaning of the invention.

Corresponding products are known as SAN-AI D SI-B2A, SAN-AI D SI-B3A, SAN-AI D SI-B7, SAN-AI D SI-45, SAN-AI D SI-60, SAN-AI D Sl- 80 and SAN-AID SI-100 from San-Shin Chemical Industry Co. Ltd.

Furthermore, quaternary N-benzylpyridinium salts and N-benzylammonium salts are suitable as thermally activatable acid generators (b2), as disclosed in EP 0 343 690 A2 or WO 2005/097883 A2.

Commercially available products are available from King Industries Inc. under the names K-PLIRE CXC-1614, K-PURE CXC-1821 or K-PURE CXC-1733.

The initiator (b) for the cationic polymerization is present in the second embodiment of the second mass, based on the total weight of the second mass, according to the invention in particular in a proportion of 0.01 to 5% by weight, based on the total weight of the second mass .

Component (d) of the second mass: additive

In the second embodiment, the second mass according to the invention contains, in addition to components (a) and (b), optionally an additive (d).

The additive (d) is further preferably an inorganic filler and is in particular present in a proportion of up to 85% by weight, based on the total weight of the second mass.

Structurally, the additive (d) can also be selected in the second embodiment of the second mass analogously to the component (D) of the first mass.

However, the use of latent release agents (C) is also dispensed with in the second compositions of the second embodiment. In other words, no “component (c)” analogous to component (C) of the first mass exists in the second embodiment of the second mass.

In both the first embodiment and the second embodiment of the second mass, the second mass may contain a thermochromic filler. In this way, the color of the potting layer can be used to determine its temperature, which enables even more precise control of the process flow in the production of the electronic assembly. In particular, due to the thermochromic filler, the casting layer becomes opaque under the influence of IR rays as soon as a predetermined process temperature is reached, in particular as soon as the release temperature of the hardened first mass is reached.

“Opaque” means that the UV-VIS transmittance of the hardened casting layer at 450 nm and a layer thickness of 200 pm is at least 25% lower after carrying out the method according to the invention than before irradiation.

The object of the invention is further achieved by an electronic assembly at the wafer level, obtainable according to a method as described above, comprising at least one functional unit encapsulated by a potting layer in such a way that the electrically contactable side of the functional unit is exposed.

The features and properties of the method described above apply accordingly to the electronic assembly at the wafer level and vice versa.

The wafer-level electronic assembly is in particular obtained according to a method as described above.

In particular, the wafer-level electronic assembly is a wafer-level microelectronic assembly.

In addition, the object of the invention is achieved by an electronic assembly, obtained by separating the electronic assembly at wafer level according to claim 14.

All methods known in the prior art can be used for separation.

BRIEF DESCRIPTION OF DRAWINGS

Further advantages of the features and properties of the invention result from the following description of exemplary embodiments and from the drawings to which reference is made. In the drawings show: - Fig. 1a - h a sequence of steps of a first embodiment of a method according to the invention for producing an electronic assembly on wafer level for the “fan-out” wafer level packaging process;

- Fig. 2a - h a sequence of steps of a second embodiment of the method according to the invention for producing an electronic assembly on wafer level for the "fan-in" wafer level packaging process; and

- Fig. 3 is a block diagram of the method according to the invention from Figs. 1a - h and Figs. 2a - h.

DETAILED DESCRIPTION

A method according to the invention is described below with reference to FIGS. 1a to 1h or 2a to 2h, with FIGS. 1a to 1h showing a first embodiment of the method according to the invention for producing an electronic assembly at wafer level in the form of a “fan-out”. Method and Figures 2a to 2h show a method according to the invention for producing an electronic assembly at wafer level in the form of a “fan-in” method.

(a) Providing a transilluminable base support

1a and 2a each show the provision of a irradiable base carrier 10 (cf. step S1 in FIG. 3), with FIG. 1a showing a cross-sectional view through the base carrier 10.

The base support 10 consists of a rigid, irradiable material such as glass or plastic and has a first side 10a and a second side 10b, which is arranged opposite to the first side 10a.

The irradiable base support 10 can have any geometric shapes, for example that of a rectangle or a circle, when viewed in the direction of the first side 10a.

Typical circular base supports 10 for wafer-level processes are 12 inches (30.48 cm) in diameter. Rectangular base supports 10 usually have dimensions of up to 610 x 457 mm.

The thickness of the base support is in a range between 100 pm and 3000 pm.

It is understood that other dimensions of the base carrier 10 are possible depending on the requirements for the electronic assemblies to be produced at the wafer level.

(b) Creating a temporary adhesive layer

In FIGS. 1b and 2b, the irradiable base support 10 is coated with a first mass, forming a temporary adhesive layer 12 (cf. step S2 in FIG. 3).

The first mass is dosed onto the first side 10a of the irradiable base carrier 10. Suitable dosing methods include spin coating or stencil printing.

The first mass, and thus also the temporary adhesive layer 12 produced from the first mass, can be hardened by actinic radiation and can be removed from the base carrier 10 by heat.

The layer thickness of the temporary adhesive layer 12 is in a range between 5 pm and 500 pm, in particular in a range between 10 pm and 300 pm, preferably in a range between 50 pm and 200 pm.

The irradiable base carrier 10 and the temporary adhesive layer 12 together form a carrier substrate 14.

(c) Placing functional units

As can be seen in Fig. 1c, functional units 16 are placed or placed at regular intervals on the still liquid temporary adhesive layer 12 (cf. step S3 in Fig. 3), for example via a so-called “Pick & Place” (not shown). “Device as is known from the prior art. The distances between the functional units 16 are in a range between 10 pm and 1000 pm, in particular in a range between 50 pm and 500 pm.

It goes without saying that there can also be fewer or more functional units 16 than shown in FIG. 1c and also that other distances between the functional units 16 can be selected as required.

The functional units 16 are active functional units, for example transistors, diodes, rectifiers, processors, ICs and/or LEDs, or passive functional units, for example capacitors, potentiometers, oscillators, coils and/or resistors.

In addition, the functional units 16 have contact points 18 on their base area facing the temporary adhesive layer 12, which serve to electrically contact the respective functional unit 16.

2c shows the corresponding illustration for the “fan-in” process, in which the functional unit 16 additionally has electrically conductive connections 20, each of which is assigned to one of the contact points 18.

The electrically conductive connections 20 are arranged within the still liquid adhesive mass, namely within the first mass applied to the base carrier 10, the temporary adhesive layer 12.

The electrically conductive connections 20 consist, for example, of a solder alloy, a combination of copper and a solder alloy or pure copper.

In this case, the “fan-in” process differs from the “fan-out” process in that the functional units 16 are equipped not only with the contact points 18, but also with the electrically conductive connections 20. (d) Forming a temporary composite of carrier substrate and functional units

In order to temporarily fix the at least one functional unit 16 on the carrier substrate 14, the temporary adhesive layer 12 is, as shown in FIGS. 1d and 2d, passed through the underside, that is, starting from the second side 10b of the irradiable base carrier 10, by irradiation actinic radiation 22 hardened (cf. step S4 in FIG. 3), in particular hardened.

The actinic radiation 22 has a wavelength in the range from 200 nm to 1000 nm, in particular in a range from 320 to 480 nm.

For example, LED lamps from DELO industrial adhesives can be used, such as the DELOLUX 20, DELOLUX 202, DELOLUX 203 or DELOLUX 820.

The carrier substrate 14, i.e. the base carrier 10 and the hardened, in particular hardened, temporary adhesive layer 12, as well as the at least one functional unit 16 together form a temporary composite 24 in this way.

(e) Applying a potting layer

1e and 2e show that a second mass is applied to the temporary composite 24 to create a casting layer 26 (cf. step S5 in FIG. 3), that is to say both via the at least one functional unit 16 and on the Areas of the hardened, in particular hardened, temporary adhesive layer 12 surrounding the functional units 16.

The layer thickness of the casting layer 26 is in a range between 1 pm and 2000 pm, in particular in the range between 50 pm and 1000 pm and preferably in the range between 100 and 700 pm, measured starting from the top of the temporary adhesive layer 12 pointing in the direction of the functional units 16 .

The potting layer 26 in particular has a thickness that corresponds at least to the height of the functional units 16. Depending on the process and target geometry, suitable application methods for the second mass are, for example, screen printing, stencil printing or UV molding.

(f) Forming a precursor

The second mass of the casting layer 26 is, analogous to the hardening of the temporary adhesive layer 12 in FIGS. 1d and 2d, hardened, in particular hardened, by irradiation with actinic radiation 28 to form a precursor 30 (cf. step S6 in FIG. 3), as shown in Figures 1f and 2f.

In other words, the precursor 30 includes the base carrier 10, the (hardened) temporary adhesive layer 12, the functional units 16 and the (hardened) potting layer 26.

The actinic radiation 28 can be in a wavelength range from 200 nm to 1000 nm, in particular in a range from 320 to 480 nm.

For example, lamps from DELO industrial adhesives can be used, such as the DELOLUX 20, DELOLUX 202, DELOLUX 203 or DELOLUX 820.

In one embodiment, the entire potting surface, that is, the entire side of the precursor 24 pointing upwards in FIGS. 1e and 2e, is simultaneously irradiated with a surface emitter. The entire casting layer 26 is therefore hardened simultaneously and in one step.

A further embodiment provides for selective, scanning irradiation. Here, the potting layer 26 is hardened point by point one after the other using a laser point source (not shown) or a narrowly focused LED lamp. This enables the hardening to be decoupled in terms of time and location, meaning that hardening shrinkage only occurs locally and at certain points. Using suitable optics (e.g. a MEMS mirror system), rapid scanning of large areas can be achieved despite the selective hardening. (g) Detachment of the carrier substrate

In a final method step, as can be seen in FIGS. 1g and 1h and 2g and 2h, the carrier substrate 14 is detached from the precursor 30 to form an electronic assembly at wafer level 32 (cf. step S7 in FIG. 3).

For this purpose, the precursor 30 is heated, for example by means of an IR radiator (not shown), an oven or a heating plate, as indicated by the arrows 34 in FIGS. 1g and 2g, which schematically indicate the acting IR radiation.

On the part of the irradiable base carrier 10, the precursor 30 is irradiated, for example, with IR radiation with a wavelength of 780 to 1000 nm and brought to a release temperature between 50 ° C and 300 ° C, in particular to a temperature between 100 ° C and 250 ° C , preferably to a temperature between 180 °C and 220 °C.

The effect of heat causes so-called “thermal debonding” to take place, so that the hardened, in particular hardened, temporary adhesive layer 12 is separated from the remaining components of the precursor 30 and the electronic assembly is obtained at wafer level 32.

The electronic assembly at wafer level 32 thus includes the (hardened) potting layer 26 and the functional units 16, whose electrical contact points 18 are exposed again after the carrier substrate 14 has been removed. In the case of the “fan-in” process, the electrically conductive connections 20 also remain part of the electronic assembly at wafer level 32.

Removal can also be assisted by mechanical forces.

The electronic assembly at wafer level 32 is now available for further processing steps such as grinding and dicing.

The hardened, in particular hardened, temporary adhesive layer 12 is preferably removed without leaving any residue, so that additional grinding, washing or cleaning steps can be dispensed with. The microelectronic assembly at wafer level 26 obtained after the temporary carrier substrate 14 has been removed can be seen in FIG. 1h for the “fan-out” process. In this case, the at least one functional unit 16 is embedded in the potting layer 20 from three sides, while the contact points 18 are exposed for the application of a redistribution layer.

For the “fan-in” process, FIG. 2h shows the microelectronic assembly at wafer level 32, the functional units 16 of which are exposed with their active side and with the electrically conductive connections 20.

It is understood that further processing steps can also take place between the individual method steps shown in FIGS. In particular, necessary handling steps, rest phases and/or additional hardening processes can be provided.

Examples of the behavior of the compositions used according to the invention are described below in order to further clarify the properties and features of the method according to the invention.

example 1

A first mass was applied via spin coating to a transmissible, round base support 10 made of glass, hereinafter also referred to as a glass support, with a diameter of 12 inches (30.48 cm) and a height of 1 mm, in order to create a temporary adhesive layer 12 with a To create a layer thickness of 200 pm on the top of the glass slide.

The first mass used for the temporary adhesive layer 12 consisted of the following components:

(A) 15 to 85% by weight (meth)acrylate, with

(A1) 1 to 100%, based on the total weight of the component

(A), at least one monofunctional (meth)acrylate, (A2) 0 to 99%, based on the total weight of component (A), of an at least difunctional crosslinker, in particular an at least difunctional (meth)acrylate and/or urethane (meth)acrylate,

(B) 0.01 to 5% by weight of a photoinitiator for radical polymerization,

(C) 5 to 45% by weight of thermo-expandable microcapsules (C1) as a latent release agent,

(D) 0 to 70% by weight of additives, where the proportions of components (A) to (D) each relate to the total weight of the first mass and all components add up to 100% by weight.

Using a “Pick & Place” device, functional units 16 with an area of 1 mm ² were placed one after the other on the still liquid first mass for the temporary adhesive layer 12 at a distance of 300 pm from one another, with a total of 25 functional units in a grid of 5 x 5 units were placed in the x and y directions.

The functional units 16 had a height of less than 700 pm.

The first mass for the temporary adhesive layer 12 was cured over the underside of the glass carrier under an oxygen-reduced atmosphere (proportion O2 <10% by volume) for 30 s with a wavelength of 400 nm and an intensity of 200 mW/cm ² .

A second mass for the potting layer 26 was then dosed onto the functional units 16 and distributed with a stamp to a uniform layer thickness of 700 μm.

The second mass used for the casting layer 26 consisted of the following components:

(A) 5 to 98% by weight (meth)acrylate, with

(A1) 1 to 100%, based on the total weight of the component

(B) 0.01 to 5% by weight of a photoinitiator for radical polymerization,

(D) up to 90% by weight of additives, whereby the details of components (A), (B) and (D) each relate to the total weight of the first mass and all components add up to 100% by weight.

In other words, the second mass is formulated analogously to the first mass with the components (A), (B) and (D), but free of latent release agent (C) with correspondingly adjusted contents of the remaining components. The same or different compounds of components (A), (B) and (D) can be used for the first and second masses, depending on the requirements for mechanical and chemical properties of the hardened masses.

The second mass for the casting layer 26 was cured over a large area and over a period of 60 s at a wavelength of 400 nm and an intensity of 200 mW/cm ² .

Example 2

A first mass for the temporary adhesive layer 12 with a layer thickness of 200 pm applied to the top of the glass slide, the first mass being formulated analogously to the first mass in Example 1.

Using a “Pick & Place” device, functional units with an area of 1 mm ² were placed one after the other on the still liquid mass for the temporary adhesive layer 12 at a distance of 300 pm from one another, whereby A total of 25 functional units were placed in a grid of 5 x 5 units in the x and y directions.

The functional units 16 have a height of less than 700 pm.

(a) 10 to 95% by weight of at least one cationically polymerizable compound selected from the group of epoxide-containing compounds (a1), oxetane-containing compounds (a2), vinyl ethers (a3), alcohols (a4), cationically polymerizable hybrid compounds (a5) and mixtures of that,

(b) 0.01 to 5% by weight of an initiator for the cationic polymerization,

(d) up to 85% by weight of at least one inorganic filler and optionally further additives (d), the details of components (a), (b) and (d) each referring to the total weight of the second mass and all Add components to 100% by weight.

In this variant too, the second mass is formulated to be free of latent release agent (C).

The second mass for the casting layer 26 was hardened over a large area and over a period of 30 s at a wavelength of 365 nm and an intensity of 200 mW/cm ² . Determination of the die shift

The shift that occurred according to the preparation as described for Example 1 and Example 2 was determined as follows.

Using a microscope, the distances between the functional units 16 were measured both after the formation of the temporary composite 24 (process step d), cf. step S4 in FIG. 3) and after the formation of the precursor 30 (process step f), cf. step S6 in Fig. 3) determined.

Due to the arrangement of the functional units 16 in a 5x5 grid according to Examples 1 and 2, a total of 20 measured values in the x direction and 20 measured values in the y direction were obtained, with the x direction and y direction perpendicular to each other and parallel to the top of the temporary network 24 or the precursor 30.

The measured values were then arithmetically averaged, using the following definition to assess the die shift:

Deviation of <= 15 pm: no die shift

Deviation of > 15 pm: The shift

As a comparative example, a precursor was produced analogously to Examples 1 and 2, but using a free-radically polymerizable mass as the mass for producing a temporary adhesive layer and a thermosetting mass based on an epoxy and an amine as the mass for producing a casting layer.

The results of the measurements are given in Table 1.

From Table 1 it can be seen that when producing electronic assemblies at wafer level according to the method according to the invention, a significantly lower extent of “die shift” can be observed than occurs when using thermosetting materials, as is known in the prior art. In addition, only very short times are sufficient for the respective mass to harden, so that the overall duration of the manufacturing process according to the invention is optimized.

Furthermore, the process according to the invention offers a high degree of freedom in formulation, so that both a free-radically polymerizable material and a cationically polymerizable material can be used for the second material, as long as the materials are radiation-curable.

When using known thermosetting materials, however, a much more pronounced “die shift” is observed, which requires further processing steps of the resulting electronic assembly at wafer level or even makes it unusable.

Determination of the warpage (curvature)

To estimate the expected warpage in the manufacture of electronic assemblies at the wafer level, the second masses for the potting layer 26 described for Examples 1 and 2 are placed on a round base carrier 10 made of glass with a diameter of 12 inches (30.48 cm). and a height of 1 mm.

The casting compound is distributed with a stamp to a uniform layer thickness of 700 μm on the base support 10, with an outer edge of 2 cm in length in the radial direction of the base support 10 not being coated for the measurement.

The respective second mass is then hardened by surface irradiation over a period of 30 s at a wavelength of 400 nm for the second mass according to Example 1 and of 365 nm for the second mass according to Example 2, each at an intensity of 200 mW / cm ² .

The curvature (warpage) of the substrate caused by the hardening, i.e. the base support 10 with the hardened second mass applied thereon, is determined using a laser profile sensor from the LJ-V7060 model series from Keyence. Laser profile sensors are laser displacement sensors that collect elevation data across a laser line rather than just a single point. The width of the laser line generated by the laser profile sensor is approximately 1.5 cm and provides height data points over this entire width, whereby the curvature of the substrate compared to a flat reference surface can be determined.

The base support 10 provided with the casting layer 26 is therefore positioned on a free-floating, flat glass plate in such a way that the casting layer 26 points upwards, i.e. on the side of the base support 10 opposite the flat glass plate.

Four measuring lines are used, each of which runs over the uncoated outer area of the base support 10 as well as over the free-floating glass plate underneath.

The four measuring lines lie on two orthogonal straight lines, the intersection of which lies at the center of gravity of the circular base support 10 (in the case of a rectangular base support 10, at the center of gravity of the rectangular base support 10). In particular, the straight lines are oriented so that they intersect the boundary of the base support 10 perpendicularly.

The width of the measuring beam is such that half of the height data of the base support provided with the casting layer 26 and the free-floating flat glass plate are determined.

In this context, a curvature caused by the hardening of the second mass also affects the uncoated outer area of the base support 10, so that the curvature caused by the hardening can be determined using the measuring lines positioned as described above.

Five measurements are carried out per measuring line and their average is calculated.

The height difference between the flat glass plate and the upper uncoated edge of the base support 10 is measured. In order to determine the actual curvature of the base support 10, the thickness of the base support must be subtracted. These measurements are taken once directly, i.e. within 15 minutes, once after 2 hours and once after 24 hours, each based on the time after curing with light.

The measurement after 24 hours takes into account any post-hardening and/or post-crosslinking of the second mass. In this way, the behavior in typical manufacturing processes for (micro)electronic assemblies is simulated, in which analog precursors are usually only processed further after a waiting period has been observed.

For a wafer with a diameter of 12 inches (30.48 cm), a warp of <= 200 pm after 24 hours is considered acceptable (OK) in the sense of the method according to the invention.

For a wafer with a diameter of 12 inches (30.48 cm), a warp of > 200 pm after 24 hours is considered not to be OK (not OK) in the sense of the method according to the invention.

The results of the measurements are shown in Table 2.

It can be seen that both a radiation-curable second mass based on a radically polymerizable system and a radiation-curable second mass based on a cationically polymerizable system have a curvature 24 hours after the respective mass has hardened, with which further processing of a corresponding wafer would be possible without any problems .

In the case of a free-radically polymerizable mass, a correspondingly slight distortion is achieved after less than 2 hours.

However, conventional heat-curing compounds lead to significantly greater curvature, which causes the glass carrier used to break.

It can therefore be seen that the radiation-curable compositions provided in the method according to the invention result in both the occurrence of a “die shift” being minimized and the warping of the substrate used in each case being suppressed. Table 1: Measurement results for determining the die shift

: not according to the invention, literature values

Table 2: Measurement results for determining the warpage (curvature)

: not according to the invention

Claims

Patent claims

1. Method for producing electronic assemblies at wafer level (32) comprising the following steps: a) providing a transilluminable base carrier (10) with a first side (10a) and a second side (10b), b) producing a temporary adhesive layer ( 12) by applying a first mass to the first side (10a) of the irradiable base carrier (10) to form a carrier substrate (14), the first mass comprising a (meth)acrylate and a photoinitiator, c) placing at least one functional unit (16) with an electrically contactable side of the functional unit (16) on the temporary adhesive layer (12), d) forming a temporary composite (24) from the carrier substrate (10) and the at least one functional unit (16) by hardening the first mass Irradiation with actinic radiation, the irradiation taking place from the second side of the irradiable base support (10) through the base support (10), e) applying a casting layer (26) to the temporary composite (24) by dosing a second mass onto it at least one functional unit (16) and on the temporary adhesive layer (12), the second mass being a radiation-curable mass, f) forming a precursor (30) by hardening the second mass by irradiating it with actinic radiation, and g) detaching the carrier substrate ( 14) with heat input into the temporary adhesive layer (12) to form the electronic assembly at wafer level (32).

2. The method according to claim 1, characterized in that the base support (10) consists of glass or plastic.

3. The method according to claim 1 or 2, wherein the at least one functional unit (16) is selected from the group of active functional units and passive functional units.

4. The method according to any one of the preceding claims, wherein the application of the potting layer (26) is carried out by a printing, slot die coating or UV stamping process

5. The method according to any one of the preceding claims, wherein the hardening of the first mass and / or the second mass is carried out by irradiation with actinic radiation (22, 28) of a wavelength in a range from 200 nm to 1000 nm, in particular in a range of 320 nm to 480 nm.

6. The method according to any one of the preceding claims, wherein the second mass is hardened over a large area and/or at specific points by scanning.

7. The method according to any one of the preceding claims, wherein the detachment of the carrier substrate (14) is carried out by heating to a detachment temperature between 50 ° C and 300 ° C, in particular to a detachment temperature between 100 ° C and 250 ° C, preferably to a detachment temperature between 180°C and 220°C.

8. The method according to any one of the preceding claims, wherein the detachment temperature is generated by irradiation with IR radiation of a wavelength in the range from 780 nm to 1000 nm, the IR radiation being generated in particular by an oven or a heating plate.

9. The method according to any one of the preceding claims, wherein the temporary composite (30) is tempered after application of the second mass at a temperature below the release temperature.

10. The method according to any one of the preceding claims, wherein the first mass contains thermo-expandable microcapsules.

11. Method according to one of the preceding claims, wherein the first mass 5 to 98% by weight of the (meth)acrylate, 0.01 to 5% by weight of the photoinitiator, 1 to 50% by weight of a latent release agent and 0 to 70% by weight of an additive, based on the total weight of the first mass.

12. The method according to any one of the preceding claims, characterized in that the second mass comprises a radically polymerizable component, the radically polymerizable component being present in particular in a proportion of 5 to 98% by weight, based on the total weight of the second mass, or that the second mass comprises a cationically polymerizable component, the cationically polymerizable component being present in particular in a proportion of 10 to 95% by weight, based on the total weight of the second mass.

13. The method according to any one of the preceding claims, characterized in that the second mass contains a thermochromic filler.

14. Electronic assembly at wafer level obtainable according to a method according to one of the preceding claims, comprising at least one functional unit (16) encapsulated by a potting layer (26) in such a way that the electrically contactable side of the functional unit (16) is exposed.

15. Electronic assembly obtained by separating the electronic

Wafer-level assembly (32) according to claim 14.