Patent Application
20170309517
Embodiments of the present invention relate to a method for forming a semiconductor device, in particular to a method for forming wide band-gap semiconductor devices on wafer-level, and a semiconductor device.
In order to improve the device characteristics of semiconductor devices attempts have been made to reduce the final thickness of semiconductor material. In particular for power semiconductor devices, it is often desired that the semiconductor body of such devices has a thickness which is just sufficient for accommodating the device or circuit.
The manufacturing and handling of thin semiconductor chips and wafers is often complicated since the brittle semiconductor material such as silicon carbide (SiC), once thinned, is prone to breaking. Further, monocrystalline wide band-gap materials are comparatively expensive. To improve the mechanical stability of thin semiconductor material, carrier systems have been developed. For example, a supported wafer may be formed by bonding a monocrystalline SiC substrate to a carrier wafer and subsequent peeling the monocrystalline SiC substrate from the carrier wafer while leaving part of the single-crystal substrate on the carrier wafer. However, unless a comparatively expensive poly-SiC carrier wafer is used, the electrical contact resistance and/or the thermal contact resistance between the monocrystalline SiC substrate and the carrier wafer and/or the electrical resistance and/or the thermal contact resistance of the carrier wafer may interfere with manufacturing. In addition, the high hardness of SiC typically poses a challenge for singulating the supported wafer. Furthermore, forming a contact metallization at the supported side of the monocrystalline SiC substrate may be complicated.
Other carriers often tolerate only moderate processing conditions. For example carriers glued to e.g. SiC-wafers are often limited to temperatures below 350° C. due to the limited thermal stability of the adhesive.
For these and other reasons there is a need for the present invention.
According to an embodiment of a method of forming a semiconductor device, the method includes providing a wafer stack including a carrier wafer comprising graphite and a device wafer comprising a wide band-gap semiconductor material and having a first side and a second side opposite the first side and attached to the carrier wafer, defining device regions of the wafer stack, partly removing the carrier wafer so that openings are formed in the carrier wafer which are arranged within respective device regions and that the device wafer is supported by a residual of the carrier wafer, and further processing the device wafer while the device wafer remains supported by the residual of the carrier wafer.
According to an embodiment of a method of forming a semiconductor device, the method includes providing a wafer stack including a carrier wafer comprising graphite and a device wafer comprising a wide band-gap semiconductor material and having a first side and a second side opposite the first side and bonded to the carrier wafer, defining device regions of the wafer stack, forming openings in the carrier wafer so that areas of the second side in the device regions are uncovered and surrounded by a residual of the carrier wafer, and filling the openings in the carrier wafer with a conductive material to form respective conductive regions electrically contacting the areas of the second side.
According to an embodiment of a semiconductor device, the semiconductor device includes a semiconductor body having a first side and a second side opposite the first side and comprising a wide band-gap semiconductor material, a graphite substrate bonded to the second side of the semiconductor body and comprising an opening leaving an area of the second side of the semiconductor body uncovered by the graphite substrate, and a back-side metallization arranged in the opening of the graphite substrate, surrounded by the graphite substrate, and electrically contacting the area of the second side.
Those skilled in the art will recognize additional features and advantages upon reading the following detailed description, and upon viewing the accompanying drawings.
The components in the figures are not necessarily to scale, instead emphasis being placed upon illustrating the principles of the invention. Moreover, in the figures, like reference numerals designate corresponding parts. In the drawings:
In the following Detailed Description, reference is made to the accompanying drawings, which form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. In this regard, directional terminology, such as “top,” “bottom,” “front,” “back,” “leading,” “trailing,” etc., is used with reference to the orientation of the Figure(s) being described. Because components of embodiments can be positioned in a number of different orientations, the directional terminology is used for purposes of illustration and is in no way limiting. It is to be understood that other embodiments may be utilized and structural or logical changes may be made without departing from the scope of the present invention. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.
Reference will now be made in detail to various embodiments, one or more examples of which are illustrated in the figures. Each example is provided by way of explanation, and is not meant as a limitation of the invention. For example, features illustrated or described as part of one embodiment can be used on or in conjunction with other embodiments to yield yet a further embodiment. It is intended that the present invention includes such modifications and variations. The examples are described using specific language which should not be construed as limiting the scope of the appending claims. The drawings are not scaled and are for illustrative purposes only. For clarity, the same elements or manufacturing steps have been designated by the same references in the different drawings if not stated otherwise.
The term “horizontal” as used in this specification intends to describe an orientation substantially parallel to a first or main horizontal side of a semiconductor substrate or body. This can be for instance the surface of a wafer or a die.
The term “vertical” as used in this specification intends to describe an orientation which is substantially arranged perpendicular to the first side, i.e. parallel to the normal direction of the first side of the semiconductor substrate or body.
In this specification, n-doped is referred to as first conductivity type while p-doped is referred to as second conductivity type. Alternatively, the semiconductor devices can be formed with opposite doping relations so that the first conductivity type can be p-doped and the second conductivity type can be n-doped. Furthermore, some Figures illustrate relative doping concentrations by indicating “-” or “+” next to the doping type. For example, “n-” means a doping concentration which is less than the doping concentration of an “n”-doping region while an “n+”-doping region has a larger doping concentration than the “n”-doping region. However, indicating the relative doping concentration does not mean that doping regions of the same relative doping concentration have to have the same absolute doping concentration unless otherwise stated. For example, two different n+-doping regions can have different absolute doping concentrations. The same applies, for example, to an n+-doping and a p+-doping region.
Specific embodiments described in this specification pertain to, without being limited thereto, manufacturing wide band-gap semiconductor devices on wafer level and the manufactured semiconductor devices.
When referring to semiconductor devices, at least two-terminal devices are meant, an example is a diode. Semiconductor devices can also be three-terminal devices such as a field-effect transistors (FETs), insulated gate bipolar transistors (IGBTs), junction field effect transistors (JFETs), and thyristors to name a few. The semiconductor devices can also include more than three terminals.
The term “wide band-gap semiconductor material” as used in this specification intends to describe a semiconductor material with a band-gap above 1 eV. Wide band-gap semiconductor materials such as SiC or GaN (gallium nitride) have a high breakdown field strength and high critical avalanche field strength, respectively. Accordingly, the doping of semiconductor regions can be chosen higher compared to lower band-gap semiconductor materials which reduces the on-state resistance Ron (also referred to as on-resistance Ron). In the following, embodiments are mainly explained with regard to SiC as wide band-gap semiconductor material.
The (wide band-gap) semiconductor devices (e.g. SiC semiconductor devices or GaN semiconductor devices) to be manufactured may be power semiconductor devices.
The term “power semiconductor device” as used in this specification intends to describe a semiconductor device on a single chip with high voltage and/or high current switching capabilities. In other words, power semiconductor devices are intended for high current, typically in the ampere range. Within this specification the terms “power semiconductor device” and “power semiconductor component” are used synonymously.
The term “field-effect” as used in this specification intends to describe the electric-field mediated formation of a conductive “channel” and/or control of conductivity and/or shape of the channel in a semiconductor region using an insulated gate electrode or a Schottky-gate electrode.
In a first process, a SiC donor wafer 10 is provided. The SiC donor wafer 10 may be of the 4H—SiC polytype having a C-side (also known as C-face) 101 and Si-side (also known as Si-face) 102. Further, the SiC donor wafer 10 may be cut from a SiC ingot.
In the exemplary embodiment illustrated in
The implantation of atoms or ions (typically gas ions such as protons), respectively, may cause the formation of a delamination layer 13 which can be a micro-bubble layer or micro-porous layer along the donor wafer 10.
The implantation depth defines the position of the delamination layer 13 and thus the thickness of a split layer 1 transferred to a carrier wafer 20. For example, 80 keV protons with a dose between 5*1016 cm−2 and 8*1016 cm−2 are mainly implanted to a depth of about 0.5-2 μm in SiC. Typically, the proton implantation energy is in range from about 50 keV to about 200 keV.
The carrier wafer 20 has a lower side 201 and an upper side 202 opposite the lower side 201 and includes or is formed by a graphite wafer. As illustrated in
Graphite is one of the three known allotropes of carbon. Within this specification the terms “graphite wafer” and “carbon wafer” are used synonymously.
To protect the graphite wafer against oxidation and to prevent the release of carbon particles, the carrier wafer 20 is typically a composite wafer formed by a graphite wafer (i.e. a wafer consisting of graphite or consisting essentially of graphite) that is surrounded by thin protecting layer, typically a thin SiC-layer. The SiC-layer of the carrier wafer 20 may have a thickness in a range from about 10 to about 2000 nm, more typically in a range from about 50 to about 500 nm. For sake of clarity, the protecting layer is not shown in the cross-sections of
The vertical extension of the carrier wafer 20 is typically in a range from about 10 μm to about 2500 μm, more typically in a range from about 50 μm. to about 1500 μm.
To form a wafer stack 40, the donor wafer 10 is typically attached with its implanted side, which is the C-side 101 in the exemplary embodiment, to the carrier wafer 20. In other embodiments, particles are implanted from the Si-side 102.
Attaching of the donor wafer 10 to the carrier wafer 20 is typically done by bonding. Accordingly, the donor wafer 10 and the carrier wafer 20 are joined through a bonding layer (also not shown in
The bond connection between the donor wafer 10 and the carrier wafer 20 is typically implemented such that it can withstand temperatures of at least about 1300° C. or of at least about 1450° C., e.g. temperatures of up to about 1600° C. used for a later so-called smart-cut layer transfer and an optional subsequent epitaxial growth.
The bonding may be accomplished by adhesive bonding. For this, a ceramic-forming polymer precursor may be used as adhesion layer. For example, the bonding may be carried out with a SiC ceramic-forming polymer precursor.
Alternatively, a spin-on-glass (SoG) may be used as adhesion layer. The use of SoG facilitates an initial low temperature bond, and can withstand the thermal stresses at high temperature (800-900 ° C.) where layer splitting may occur. A SoG-adhesion layer may only be used when the delamination layers 13 is sufficiently deep to allow later device manufacturing.
By the use of an adhesive ceramic-forming precursor (e.g. an adhesive SiC precursor) as the adhesive material, thermal mismatch between the active layer and the bonding zone and the undesired forming of reaction zones between bond layer and the active layer at high temperature processes may be avoided.
The ceramic-forming polymer precursor may comprise or consist (e.g. only) of carbon, silicon, and hydrogen. When the hydrogen diffuses during the bonding process, only polycristalline silicon carbide may remain. For example, the ceramic-forming polymer precursor may be an allyl-hydrido-polycarbosilan or another polycarbosilane.
In an embodiment one or both sides of the bonding sides or surfaces 101, 201 are coated with the ceramic-forming polymer precursor, followed by tempering between 200 to 700° C. For example, the tempering may be performed at a temperature of about 530° C. for about 4 hour.
As a first part of the bonding procedure, the ceramic-forming polymer precursor may be applied either to the carrier wafer 20 or the donor wafer 10. Alternatively, the ceramic-forming polymer precursor can be applied on the surfaces 101, 201 of both, the carrier wafer 20 and the donor wafer 10. The ceramic-forming polymer precursor can be applied e.g. by spin-on or spray processes.
As indicated by the dashed-dotted arrow in
After the joining, the wafer stack 40 may be heated to form the bond. For example, temperatures ranges can be from about room temperature to about 600° C., or from 200-700° C.
Tempering the wafer stack 40 may be done in a first temperature range and subsequently in a second temperature range which is different than the first temperature range. The second temperature range can encompass temperatures higher than the first temperature range. The second temperature range can be, for example, from about 500° C. to about 1000° C. or even higher.
When using allyl-hydrido-polycarbosilane as precursor, it may be pyrolyzed to polycrystalline silicon carbide at high temperatures of e.g. 1500 ° C.-1700 ° C. (e.g. for the complete conversion of the precursor layer into polycrystalline SiC). Thus, the bonding layer between the SiC and the carrier wafer may become SiC itself during the bonding process, thus omitting effects which might arise when using other types of material and furthermore assuring electrical connectivity. For example, the bonding layer may be n-doped SiC.
Furthermore, a mechanically and thermally extremely stable bond connection may be formed by converting the bonding layer into SiC.
A strengthening of the bond connection and a separation (see below) of the SiC donor wafer 10 may occur at 700° C.-1800° C.
Thus, three tempering processes performed at different temperatures may be used. However, the tempering processes may also be combined into a single process having a given temperature profile. Further, the tempering may at least temporarily takes place under (compacting) pressure.
In an embodiment, the tempering takes place in an atmosphere including nitrogen and or a noble gas, e.g. a nitrogen atmosphere, an argon atmosphere, an atmosphere of nitrogen and argon, or an atmosphere of nitrogen and hydrogen. As nitrogen is a low donor in SiC, this may cause doping of the bond layer (e.g. polycrystalline SiC, which arises from the polymer) and an adjoining n monocrystalline SiC layer of the split layer during tempering, so that the vertical conductivity may be increased.
As illustrated in
As a result a SiC split layer 1 (also referred to as smart-cut SiC-layer, first SiC-layer and lower semiconductor layer in the following) of the donor wafer 10 remains at the carrier wafer 20. In this way the split layer 1 is transferred from the donor wafer 10 to the the carrier wafer 20 (smart-cut layer transfer).
The graphite of the carrier wafer 20 can be one of turbostratic graphite, pyrolytic graphite, isostatically pressed graphite, and mixtures thereof. Graphite has a thermal expansion coefficient similar to that of SiC. This makes graphite a most promising carrier material for SiC. Graphite, however, is also suitable for other semiconductor materials such as GaN. Furthermore, the thermal expansion coefficent of graphite can be fine-tuned by its porosity.
Thereafter, the split layer 1 may be polished, for example by using a CMP-process (chemical mechanical polishing).
The split part 10′ of the donor wafer 10 may be reused (e.g. more than 5 times or more than 10 times) as a donor, as it can be brought back to a suitable initial state for smart-cut layer transfer by polishing and/or epitaxy. This may be very cost-efficient.
Alternatively to the mentioned procedure, also other methods (e.g. oxygen implantation) may be suitable to separate and transfer the semiconductor layer.
As illustrated in
Prior to epitaxial growth, the split layer 1 may have a thickness of 5 μm, 2 μm, 1 μm, or even only 0.5 μm.
In other embodiments (not illustrated in the figures), the split layer 1 has a thickness substantially corresponding to the thickness of the semiconductor body of the device to be manufactured.
As the epitaxy is performed at the Si-side 11 of the split layer 1, a high quality crystalline pattern of the epitaxial SiC-layer 2, even a more homogeneous crystalline pattern than that of the split layer 1, may be achieved. Furthermore, doping may be better controlled during epitaxy at the Si-side 11 compared to the C-side. Further, less (thickness) of the donor wafer 10 may be required. Accordingly, the process of transferring thin SiC-layers 1 of the donor wafer 10 can be repeated more often.
Further, several epitaxial SIC-layers 2 of different doping type, different doping concentration and/or different thickness may be formed on the split layer 1. For example, a highly n-doped first epitaxial SiC-layer may be formed at the split layer 1 and a low n-doped second epitaxial SIC-layer may be formed at the first epitaxial SIC-layer. The thickness of the first epitaxial SIC-layer may be chosen in accordance with mechanical stability requirements. The thickness and doping concentration of the second epitaxial SiC-layer may be chosen in accordance with device type and voltage class. For example, the second epitaxial SiC-layer may have a thickness of about 4.5 μm and a doping concentration of about 2*1016/cm3 to form a drift region in a SiC-MOSFET of the 650V class to be manufactured.
The doping of the epitaxial SiC-layer(s) 2 may be adjusted during epitaxial growth but may also include dopant implantation(s) and a subsequent annealing.
Furthermore, prior to epitaxially depositing SiC, a mask with openings in the device regions D, for example a metal carbide mask, may be formed at the split layer 1. Accordingly, epitaxially SiC depositing is inhibited outside the device regions D. Accordingly, later singulation into several chips may be facilitated.
The processes explained with regard to
In the above explained embodiments, in which the wafer stack 40 is formed using the smart-cut layer transfer technique, the second side 22 typically corresponds to a particle radiated side 101 of a monocrystalline SiC carrier wafer 10.
Thereafter, a plurality of device regions D separated from each other by singulation regions (also referred to as scribe line) may be defined. As shown in
In embodiments in which the device wafer 1, 2 and the carrier wafer 20 have substantially the same horizontal shape, the carrier wafer 20 is typically hidden by the device wafer 1, 2 in top view.
The mask layer 7 may be made of a nitride such as silicon nitride, for example a low-temperature nitride, or an oxide such as silicon oxide, for example a low-temperature nitride. Further, the mask layer 7 may be formed by chemical vapor deposition (CVD).
Prior to forming the mask layer 7, the carrier wafer 20 may be suitably thinned at the side 202. The thinning process can include, without being limited thereto, grinding, etching and polishing of the side 202.
Depending on the devices to be manufactured, one or more front-side metallizations 3 per device region D may be formed on the first side 21.
The front-side metallizations 3 are typically separated from each other. Forming the front-side metallizations 3 may include depositing one or more metal, e.g. Cu, or metal-like materials and subsequent masked etching.
Furthermore, passivation layers (not shown in
Typically, at least one of the front-side metallizations 3 per device region D is in electric contact with, e.g. ohmic contact with the epitaxial SIC-layer 2. Alternatively, at least one of the front-side metallizations 3 per device region D forms a Schottky-contact with the epitaxial SiC-layer 2.
Even further, device manufacturing from the first side 21 may be almost or even completely finished prior to forming the mask layer 7.
Device manufacturing from the first side 21 may include processes like implanting dopants, annealing, etching trenches (also referred to as first trenches) from the first side 21 into the device wafer 1, 2, insulating sidewalls of the trenches, etching a wide trench (also referred to as second trench) into the device wafer 1, 2, forming insulated gate electrodes in the trenches, and/or forming insulated gate electrodes on and/or at the first side 21.
While the trenches are typically formed in the device areas D, the wide trench is typically formed outside the device areas D, i.e. in the singulation regions S. For example, the layout of the wide trench may, when seen from above, correspond to a lattice structure as shown by the vertical and horizontal lines in
The trenches and the wide trench are typically formed using common processes. Since an etching mask has typically an opening for the wide trench with a larger (horizontal) width compared to openings for the trenches, the wide trench typically also extends vertically deeper into the device wafer 1, 2 when common processes are used. In addition to the common processes, a further etching may be used to further deepen the wide trench while the first trenches are covered.
The singulation regions (scribe line) S between the device regions D in
In a subsequent process, the device wafer 1, 2 may be attached with the first side 21 to a reversible carrier 5, e.g. a glass substrate.
As illustrated in
Alternatively, a substantially flat glass substrate may be used as reversible carrier. In one embodiment, the flat glass substrate is arranged on a passivation layer, typically an imid layer, formed on the front-side metallizations (3). Typically, a peripheral gap formed between the flat glass substrate and the first side (21) of the device wafer (1, 2) is sealed by an adhesive. To avoid any impact of adhesive residues in the device regions (D), the glue-bond between the flat glass substrate and the device wafer (1, 2) may only be formed in the peripheral area.
In the exemplary embodiment illustrated in
In one embodiment, adhesive bonding using an adhesive is used for bonding the device wafer 1, 2 and the glass substrate 5. For example, acryl based adhesive or silicate adhesives can be used which are commercially available, for example, from Dow Corning. Depending on the nature of the glass adhesive, the adhesive bond connection (not shown in
Thereafter, the mask layer 7 may be structured to form a hard mask 7 covering the singulation regions S and having openings 27 in the device regions D.
As illustrated in
Thereafter, the exposed portion of the carrier wafer 20 may be removed.
Accordingly, openings 25 are formed in the carrier wafer above the device regions D of the device wafer 1, 2 while a residual portion 20′ of the carrier wafer 20 remains attached to the device wafer 1, 2. As shown in
When seen in top view onto the residual portion 20′ of the carrier wafer 20 and the back-side 22, respectively, each device region D may partly overlap with the residual portion 20′. Further, each device region D is typically surrounded by a corresponding part of the residual portion 20′ when seen in top view onto the residual portion 20′.
Accordingly, each device region D typically remains mechanically supported by the residual portion 20′. In the following, the residual 20′ of the carrier wafer 20 is also referred to as structured graphite carrier 20′.
Typically, the structured graphite carrier 20′ provides chip specific support allowing further processing on wafer level. Furthermore, the structured graphite carrier 20′ facilitate singulation into individual semiconductor chips (see below).
Due to the mechanical support provided by the structured graphite carrier 20′, the device wafer may be comparatively thin. For example, the silicon carbide layer 1 (including the optional epitaxial layer(s) 2) may have a thickness of less than 100 μm, less than 75 μm or even less than 50 μm, for example down to 5 μm, that can, due to the structured graphite carrier 20′, be safely further processed from both sides. This allows flexible and cost efficient manufacturing of SiC-devices on wafer-level.
The structured graphite carrier 20′ is typically contiguous. This ensures a particularly high mechanical stability.
In embodiments in which the carrier wafer 20 and the device wafer 1, 2 completely overlap when see from above prior to removing the exposed portion of the carrier wafer 20,
Removing the exposed portion of the carrier wafer 20 may be done by ion beam etching and/or plasma etching.
For example, ion beam etching may be used to remove a SiC protecting layer of the carrier wafer 20. Plasma etching (in particular plasma ashing) may be used to remove the exposed graphite.
Further, exposed portions of the bond layer 42 may be removed using chemical etching to partly expose the device wafer 1, 2.
In embodiments in which the bond-layer 42 is made of highly conductive (n-doped, typically nitrogen-doped or phosphorus-doped) SiC, the bond layer 42 is typically not removed. Further, the hard mask 7 may be removed even prior to plasma etching. This is because the remaining portions of the SiC protecting layer may also be used as mask for plasma etching.
Alternatively, or in addition removing the exposed portion of the carrier wafer 20 may include one or more processes of milling and/or grinding the carrier wafer 20. For example, a cutter may be used to remove portions of the carrier wafer 20 except for thin remainders that are removed by plasma ashing stopping at the SiC on and/or at the device wafer 1, 2.
Openings of the SiC protecting layer on the on the residual 20′ of the carrier wafer 20 (where graphite may be exposed) may be closed by depositing silicon and a thermal process.
After removing the exposed portion of the carrier wafer 20 and forming the structured graphite carrier 20′, respectively, the reversible carrier 5 may be removed.
Thereafter, the device wafer may be further processed while the device wafer 1, 2 remains mechanically supported by the structured graphite carrier 20.
The formed wafer structure 40 may be described as a wafer structure including a device wafer 1, 2 typically comprising a wide band-gap semiconductor material, e.g. a silicon carbide wafer 1, 2, and a support structure 20′ which is glue-bonded to the device wafer 1, 2, comprises or essentially consists of graphite, and leaves at least central portions of device areas D of the device wafer 1, 2 uncovered but surrounded when seen from above.
Prior to further processing, the formed wafer structure 40 may be stored and even shipped after suitable packaging.
Due to the small thermal mismatch between silicon and graphite, the structured graphite carrier 20′ is also well suited as a support structure of silicon device wafers, in particular thin (e.g. less than 100 μm or even 50 μm thick) silicon device wafers. Accordingly, a method for forming semiconductor devices includes in an embodiment, providing a wafer stack 40 including a graphite carrier wafer 20 and a device wafer having a back-side 22 glue-bonded to the graphite carrier wafer 20, defining device regions D of the wafer stack 40, partly removing the carrier wafer 20 so that openings 25 are formed in the carrier wafer 20 arranged within respective device regions D and that the device wafer 1, 2 is supported by a residual 20′ of the graphite carrier wafer 20, and further processing the device wafer 1, 2 while the device wafer 1, 2 remains supported by the residual 20′ of the carrier wafer 20.
Prior to removing the reversible carrier 5, the openings 25 are typically filled with one or more conductive materials, typically metals or alloys, to form conductive regions electrically contacting areas of the second side 22 that are not covered by the structured graphite carrier 20′. Accordingly, a back-side metallization may be formed in each device area D which is typically in electric contact (more typically in ohmic contact) with the device wafer 1, 2. Such a back-side metallization is anyways to be formed for vertical semiconductor devices. The back-side metallization(s) further stabilizes the device wafer 1, 2. In addition, the thermal contact resistance between the device wafer 1, 2 and the support structure formed by the structured graphite carrier 20′ and the back-side metallizations as well as the electrical resistance and the thermal contact resistance of the support structure is further reduced. This may facilitate further manufacturing. Furthermore, the thermal mass (also known as heat capacity and thermal capacity) is increased. Accordingly, the “inertia” against temperature fluctuations is increased.
In one (not illustrated) embodiment, the residual 20′ of the carrier wafer 20 is completely removed after forming the back-side metallization(s). In this embodiment, the typically thick back-side metallization(s) may be used as support structure during further processing.
Forming the back-side metallizations in the openings 25 of the structured graphite carrier 20′ has two further advantages. First, mechanical stress is reduced compared to forming a contiguous back-side metallization covering at least a major part of the device wafer 1, 2. Second, cutting or sawing through the back-side metallization is not required for singulating (see below).
Forming the back-side metallizations in the openings 25 may include depositing a contact layer of e.g. NiSi. The contact layer may be activated by LTA (laser thermal anneal). Thereafter, a conformal seed layer 6 may be formed. The resulting wafer structure is illustrated in
The seed layer 6 may be formed by physical vapor deposition of e.g. Cu.
Prior to forming the seed layer 6, barrier layers such as (Ti, W, Ta, TaN, TiN, TaN, TiWN, Ru, or combinations of them) may be deposited. For sake of clarity, the contact layer and the barrier layers are not shown in
As illustrated in
After forming the seed layer 6, the openings 25 may be filled with a further conductive region 8, typically a metal such as Cu or an alloy of high specific heat capacity and/or high density and/or low specific electric resistivity. The resulting wafer structure 40 is shown in
Filling the openings 25 may include electro-chemical plating, in particular pattern plating, i.e. forming a plating mask, electro-chemical plating and removing the plating mask, as well as optional processes of over-etching the seed layer 6 and a planarization (e.g. a CMP-process). Filling the openings 25 may also include printing a conductive paste, e.g. screen printing a sinterable metal paste, or inkjet printing. In these embodiments, no seed layer (6) is needed.
In embodiments referring to manufacturing of MOSFETs, the back-side metallization typically forms a drain metallization.
In embodiments referring to manufacturing of diodes, the back-side metallization typically forms a cathode metallization.
In a further process, the reversible carrier 5 may be removed. The resulting wafer structure 40 is shown in
Thereafter, the wafer structure 40 may be further processed. This may include thermal annealing.
Further processing may also include (individually) testing the functionality of the device regions D. For this purpose the formed back-side metallizations 6, 8 and the front-side metallizations 3 may be temporarily contacted.
For example, a blocking capability of a pn-diode, a barrier height of a Schottky-diode or integrity of a gate oxide formed in respective device regions D may be determined and tested, respectively.
When an electrically conductive bond layer 42 was formed, those tests may also be performed prior to forming the mask layer 7 which may also be formed after attaching the device wafer 1, 2 to the reversible carrier 5.
Testing the functionality may include processing control structures and/or electrical function tests.
With regard to
As shown in
In one embodiment, the wafer structure 40 is laminated to the foil 36.
In addition,
To protect the typically fragile wafer edge, an outer (peripheral), typically ring-shaped portion of the structured graphite carrier 20′ may be thickened. After attaching the wafer structure 40 to the foil 36 and prior to the singulation process, the outer portion of the structured graphite carrier 20′ may be removed or at least thinned, e.g. by means of mechanical or laser based technologies.
Subsequently, the device wafer 1, 2 and the semiconductor wafer structure 40, respectively, is diced along the predefined separation lines S.
For this purpose, separation trenches 12 may be formed or deepened from the front side 21.
As illustrated in
The separation trenches 12 are typically formed such that they extend from the front side at least partly into the structured graphite carrier 20′, more typically through the structured graphite carrier 20′ or even partly into the foil 36.
When the separation trenches 12 extend at least to the foil 36, the individual semiconductor chips or dies 100, 100′ may simply be picked.
Otherwise, the wafer structure 40 may be separated into individual semiconductor chips or dies 100, 100′ by breaking along the separation trenches 12.
The separation trenches 12 may completely be formed after attaching the wafer structure 40 to the foil 36.
In other embodiments, the separation trenches 12 are partly formed prior to attaching the wafer structure 40 to the foil 36. For example, the wide trenches 50 may be formed prior to attaching the wafer structure 40 to the foil 36. Alternatively or in addition, lower portions of the trenches 12 at least partially extending through the structured graphite carrier 20′ may be formed prior to attaching the wafer structure 40 to the foil 36. The structured graphite carrier 20′ may even be substantially or completely removed prior to attaching the wafer structure 40 to the foil 36.
All this ensures that the final separation does not occur through both the device wafer 1, 2 and the thick metallisation regions 8 in the openings 25.
As illustrated in
Singulating the wafer stack into individual chips 100, 100′ and forming the separation trenches 12 may include one or more of the following processes:
The methods explained above were mainly explained with regard to SiC, but may also be used for forming GaN-devices and Si-devices.
In the exemplary embodiment, the semiconductor device 200 includes a semiconductor chip 100 with a semiconductor body 10′ that may be formed by a portion of the divided device wafer 1, 2 shown in
The semiconductor body 10′ extends between a first side 21 and a second side 22. A graphite substrate 20′ is bonded to the second side 22 and has an opening 25 leaving an area of the second side 22 uncovered by the graphite substrate 20′. A back-side metallization 8 is arranged in the opening 25 and surrounded by the graphite substrate 20. The back-side metallization 8 electrically contacts the second side 22.
The semiconductor device 200 includes two terminals which are formed in this embodiment by lead structures 56 and 57. These lead structures extend laterally from a carrier substrate 58.
In the exemplary embodiment, the semiconductor body 10′ has at least one doping region 62. The doping region 62 is formed at the first surface or side 21 and in electrical contact with the front-side metallization 3 disposed on the first surface 21. The front-side metallization 3 forms in this embodiment the anode of a power diode. The doping region 62 is oppositely doped to the material of the embedding upper semiconductor layer 2 which usually has a low n-doping concentration in case of power diodes. A lower semiconductor layer 1 adjoins the upper semiconductor layer 2. The lower semiconductor layer 1 is of the same doping type and the same semiconductor material (e.g. SiC) as the upper semiconductor layer 2. However, a doping concentration of the lower semiconductor layer 1 is typically higher compared to the upper semiconductor layer 2.
The back-side metallization 8 fills the opening 25 of the graphite substrate 20′ and provides an ohmic contact between the lower semiconductor layer 1 and a pad structure 56′ of the lead structure 56. The back-side metallization 8 forms in this embodiment the cathode of the power diode. A metallic seed layer (6) is not illustrated in this embodiment but can be provided if desired.
The back-side metallization 8 is typically made of or mainly made of Cu or another metal (Sn, Au, Ag, Al, Mo,) or alloy with high electric conductivity and high thermal mass.
An extension of the back-side metallization 8 in vertical direction (i.e. a direction perpendicular to the side 21, 22) and in the openings 25 is typically in a range from about 500 nm to about 200 μm, for example in a range from about 50 μm to about 200 μm.
Accordingly, the back-side metallization 8 may also provide a heatsink of high thermal mass. This improves avalanche robustness, surge current strength and/or short-circuit strength of the semiconductor device 200.
The front-side metallization 3 may have the same or similar vertical dimensions and/or the same or a similar composition as the back-side metallization 8. Accordingly, the formation of so called hot spots during operating the semiconductor device 200 may be avoided or at least substantially reduced.
The front-side metallization 3 is electrically connected to a pad structure 57′ of the lead structure 57 by a bond wire 55 or a clip bond (not shown).
Further, the semiconductor chip 100 is encapsulated in an insulating material 65 such as a cured epoxy resin (so-called hard casting) or a silicone gel (so-called soft casting). Accordingly, the semiconductor chip 100 is protected against moisture and/or mechanical damaging.
As indicated by the dashed vertical lines in
Although various exemplary embodiments of the invention have been disclosed, it will be apparent to those skilled in the art that various changes and modifications can be made which will achieve some of the advantages of the invention without departing from the spirit and scope of the invention. It will be obvious to those reasonably skilled in the art that other components performing the same functions may be suitably substituted. It should be mentioned that features explained with reference to a specific figure may be combined with features of other figures, even in those cases in which this has not explicitly been mentioned.
Spatially relative terms such as “under”, “below”, “lower”, “over”, “upper” and the like are used for ease of description to explain the positioning of one element relative to a second element. These terms are intended to encompass different orientations of the device in addition to different orientations than those depicted in the figures. Further, terms such as “first”, “second”, and the like, are also used to describe various elements, regions, sections, etc. and are also not intended to be limiting. Like terms refer to like elements throughout the description.
As used herein, the terms “having”, “containing”, “including”, “comprising” and the like are open ended terms that indicate the presence of stated elements or features, but do not preclude additional elements or features. The articles “a”, “an” and “the” are intended to include the plural as well as the singular, unless the context clearly indicates otherwise.
With the above range of variations and applications in mind, it should be understood that the present invention is not limited by the foregoing description, nor is it limited by the accompanying drawings. Instead, the present invention is limited only by the following claims and their legal equivalents.
| Number | Date | Country | Kind |
|---|---|---|---|
| 102015112649.4 | Jul 2015 | DE | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 15224099 | Jul 2016 | US |
| Child | 15645402 | US |
