Field effect transistor and method for producing field effect transistor
By introducing a support structure into the fin structure of the trench MOSFET, the problem of mechanical instability of traditional fin FETs is solved, and higher equipment stability and reliability are achieved, and suitable for high-power applications.
Patent Information
- Application Number
- CN202410547366.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-03
- Filing Date
- 2024-05-06
- Publication Date
- 2024-11-05
AI Technical Summary
During the manufacturing process of traditional groove MOSFETs, the fin structure is mechanically instable due to lack of support, which affects the reliability and performance of the equipment.
The support structure is introduced into the fin structure, and the support structure is retained during the etching process by a suitable mask reflecting the fin and the support structure, ensuring the stability and integrity of the fin. The support structure may extend vertically to the source and drain layers and connect side-by-side fins by one or more support members, or attach support members on different sides of the fins to achieve stability.
Through the introduction of the support structure, the mechanical stability of the fin is significantly improved, the problem of fins being vacant before the gate electrode is placed is avoided, and the overall stability and reliability of the equipment are enhanced. It is suitable for high-power applications such as electric driving and inverter.
Smart Images

Figure CN118899338A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a field effect transistor, in particular a so-called trench MOSFET, and a method for producing such a field effect transistor. Background Art
[0002] Field effect transistors (FETs), in particular so-called MOSFETs, are used in a variety of fields. One variant is the so-called trench MOSFET or T-MOSFET, in which the channel is constructed vertically. Here, for example, an n-doped source layer and a channel layer located between the source layer and one of the n-doped drift layers are penetrated by trenches, and the gate electrode is arranged in such trenches. Summary of the invention
[0003] According to the invention, a field effect transistor and a method for producing a field effect transistor are provided. Advantageous embodiments are the subject of the technical solutions and the following description.
[0004] The present invention relates to field effect transistors and methods for manufacturing the same, in particular field effect transistors having trenches or channels. Different types of doping are used in semiconductor materials, namely n-doping and p-doping, wherein different components may be doped differently. In the following, for better understanding, field effect transistors will be described with specific doping types, n-doping being the first type of doping and p-doping being the second type of doping. However, it should be understood that n-doping and p-doping may also be interchangeable, i.e. n-doping may be the second type of doping and p-doping may be the first type of doping.
[0005] The field effect transistor has, for example, an n-doped source layer and an n-doped drain layer (especially including an n-doped drift layer, for example applied as a so-called epitaxial layer or epitaxial layer). In addition, the field effect transistor has a channel layer vertically located between the n-doped source layer and the n-doped drift layer. The channel layer can be at least partially p-doped. But in the case where the fin is less than about 300nm, it is also possible, for example, that the channel layer is only n-doped. Typically, the n-doped drain layer includes an n-doped drift layer and an n-doped extension layer, which has a higher n-doping than the n-doped drift layer. The n-doped extension layer is vertically arranged between the channel layer and the n-doped drift layer. The n-doped extension layer can be obtained, for example, by appropriately doping a portion of the n-doped drain layer. In addition, such a field effect transistor typically has a gate trench, which extends from the n-doped source layer to the n-doped drift layer in a vertical direction and is adjacent to the channel layer, and in particular also passes through the channel layer.
[0006] The field effect transistor may also have a gate electrode at least partially surrounded by a dielectric (e.g., a so-called gate oxide), specifically, in particular, surrounded in such a way that the gate electrode is insulated relative to the n-doped source layer, the channel layer, and the n-doped drain layer. Such a gate electrode may be arranged in each gate trench. The gate electrode in the gate trench may also be divided into two, so that each of the two electrodes produced by the division is adjacent to one of the two side walls of the gate trench in the trench. The division between the electrodes allows the source electrode to pass to the bottom of the trench or the bottom of the channel in order to contact the p-shielding doping that may be located below the trench or below some trenches. Between two gate trenches arranged side by side, for example, a spacer, the so-called fin, is formed or exists. In this case, it is also called a so-called fin FET or fin MOSFET.
[0007] It will be appreciated that such a field effect transistor has, in addition to the gate electrode, a source terminal and a drain terminal, which can be constructed in a conventional manner.
[0008] A particular advantage of trench MOSFETs, and thus fin FETs, is that, due to the vertical arrangement, many gate electrodes can be arranged side by side. The field effect transistor can be configured in particular as a SiC field effect transistor or a GaN field effect transistor, i.e. the substrate and / or the semiconductor material generally used can be silicon carbide (SiC) or gallium nitride (GaN), since these semiconductor materials have a wide band gap. However, semiconductor materials with an ultra-wide band gap are also considered, such as gallium oxide.
[0009] In T-MOSFETs and also in FinFETs, the fins and gate trenches are typically arranged in a cell field. The fins in the cell field usually extend parallel to one another and horizontally (ie in the extension plane of the various layers mentioned; in contrast, a vertical direction is understood to be perpendicular to the extension plane).
[0010] The cell field is usually surrounded by an edge shielding area, which is used to weaken the electric field at the edge not abruptly but in stages. In conventional T-MOSFETs, only the gate trenches (or channels) in the cell field are etched during manufacturing; the area between the two gate trenches is usually significantly wider than the fins of the fin FET. In fin FETs, in particular power fin MOSFETs, all areas except the (usually very narrow) fins in the cell field are etched (i.e., the corresponding material is removed), i.e., the gate trenches and the areas at the edges of the cell field. This results in the fin FET, where the (narrow) fins are empty at least before the gate electrode is placed and are therefore mechanically unstable.
[0011] In the present invention, a possibility of eliminating this defect is now proposed, in that a support structure for the fin is provided. Such a support structure can be produced, for example, by a suitable mask that reflects the fin and the support structure. Thus, for example, during etching, not only the fin but also the support structure remains.
[0012] In one embodiment, the support structure extends in the vertical direction from the first type doped source layer to the first type doped drain layer. Thus, the height of the support structure (the extension dimension in the vertical direction) may be equivalent to the height of the fin.
[0013] In one embodiment, the support structure comprises one or more supports (or spacers), which respectively connect two side-by-side fins. Preferably, one or more of the fins are each assigned exactly one support. In this way, fewer supports are required in total (and therefore a smaller support structure). However, it is also possible that one or more of the fins are each assigned exactly two supports, which are arranged on different sides of the corresponding fin (i.e. one reaches the side-by-side fins and one reaches the other side-by-side fin). This can achieve better stability.
[0014] In one embodiment, the support structure includes one or more supports, each of which is attached to only one of the fins. It is therefore possible to achieve a continuous gate trench from one end of the cell field to the other end. Preferably, the support structure includes a plurality of supports grouped in pairs, wherein a pair of supports is attached to a corresponding fin on both sides at the same height (viewed in the horizontal extension direction of the fin). Here, each fin can be equipped with a plurality of supports. Similarly, the support structure may also include a plurality of supports, wherein at least two of the plurality of supports are attached to a corresponding fin on both sides at different heights. This allows stability to be achieved with fewer supports in the extension direction of the fin.
[0015] Such field effect transistors can be used alone or together with other field effect transistors, for example as power switches. Preferred fields of application are, for example, in electric drive trains of vehicles, for example in current converters (DC / DC converters, inverters), in chargers for electric vehicles or in solar inverters. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Further advantages and embodiments of the invention are apparent from the description and the drawings.
[0017] The invention is illustrated schematically in the drawings using exemplary embodiments and is described below with reference to the drawings.
[0018] Figure 1a and 1b A field effect transistor is schematically shown for explaining the present invention.
[0019] Figure 2a , 2b 2 and 2c schematically show a field effect transistor according to the present invention in various preferred embodiments.
[0020] Figure 3 The sequence of the method according to the invention in a preferred embodiment is schematically shown. DETAILED DESCRIPTION
[0021] Figure 1a and 1b A field effect transistor 100 is schematically depicted for explaining the present invention, in particular a so-called trench MOSFET as a so-called fin MOSFET. In particular, silicon carbide (SiC), gallium nitride (GaN) or gallium oxide can be used as semiconductor materials, since these have a wide to very wide band gap. Figure 1a A cross-sectional view of the complete field effect transistor 100 is shown in FIG. Figure 1b A top view of an incomplete field effect transistor 100 is shown in FIG.
[0022] As in Figure 1a As can be seen in FIG. 1 , the field effect transistor 100 has an n-doped source layer 104, an n-doped drain layer 120, and a channel layer 106 vertically located between the n-doped source layer 104 and the n-doped drain layer 120. The n-doped drain layer 120 may include, for example, an n-doped drift layer.
[0023] The field effect transistor 100 has a plurality of gate electrodes 110, each of which is placed in one of a plurality of gate trenches (not labeled here). The gate trenches extend in a vertical direction (here the z direction) from an n-doped source layer 104 to an n-doped drain layer 120 and adjoin a channel layer 106. Between each two gate trenches, and thus also between each two gate electrodes 110, there is a spacer or so-called fin 126.
[0024] The gate electrode 110 comprises an insulating oxide 114 and a dielectric or gate oxide 116, which surrounds the gate semiconductor material 112. The gate electrode (and thus the associated gate trench) adjoins at least the channel layer 106 via the dielectric 116. The gate electrode serves to control the channel region in the channel layer 106. The region in the channel layer 106 adjoining the gate electrode or the gate trench forms the channel region or channel area.
[0025] The gate introduction, ie the contacting of the gate electrode 110 , takes place, for example, at the end of the cell field into the plane of the drawing by the material 112 leaving the trench there.
[0026] Furthermore, the field effect transistor 100 has an n-doped substrate 122, which adjoins the n-doped drain layer 120 from below, and a drain material 124, for example a metal, which adjoins the n-doped substrate 122 from below. Furthermore, the field effect transistor 100 has a source material 102, for example a metal, which adjoins the n-doped source layer 104 and the gate electrode 110 from above.
[0027] Figure 1b The field effect transistor 100 is shown in a top view, but without an inserted gate electrode and without source material. In addition to the gate trench 111, in particular, fins 126 can be seen, which extend parallel to each other, specifically horizontally (i.e. in the extension plane of the different layers mentioned, here the xy plane, here also in the y direction, for example). The fins 126 and the gate trench 111 are arranged in a cell field 130.
[0028] The cell field 130 is surrounded by an edge shielding region 136, which serves to reduce the electric field at the edge not abruptly but in a stepped manner. Here, for example, the grading is shown in two regions 132 and 134, in which different p-dopings can be provided, for example.
[0029] In fin FETs, in particular power fin MOSFETs, all regions in the cell field except the fin 126 are etched (i.e., the corresponding material is removed), i.e., the gate trench 111 and the regions at the cell field edges, such as region 132. This results in the fin FET having the narrow fin 126 being free at least before the gate electrode is placed and thus mechanically unstable. This can be, for example, Figure 1a As can be seen in FIG. 1 , if the gate electrode 110 does not (yet) exist in this figure.
[0030] Figure 2a , 2b 2c schematically depicts a field effect transistor according to the present invention in different preferred embodiments, in particular in Figure 1b The view in the figure is compared to the top view.
[0031] Figure 2a A field effect transistor 200a is shown with two exemplary cell fields 230.1a and 230.2a. The two cell fields are in principle arranged in the same manner as the cell fields 130 of the field effect transistor 100 (see Figure 1b) is constructed in a comparable manner. The cell fields 230.1a and 230.2a each have fins 226 extending in parallel, and a gate electrode 210 is arranged between each two fins 226. These gate electrodes 210 can be connected at both ends of the corresponding cell field (seen in the extension direction y). In addition, the gate oxide or gate dielectric is open at both ends, so that the semiconductor material 212 is visible and contactable. Each area of the gate electrode is in contact with at least one such open surface or opening. In addition, the field effect transistor 200a has a gate pad 240 for contact, at which the gate oxide or gate dielectric is also open. In a (complete) field effect transistor, the fins and gate electrodes are usually covered with metal or power metal (Power-Metall), which contacts the gate electrode on the opening area.
[0032] The field effect transistor 200a has a support structure for the fin 226. In the cell field 230.1a, the support structure exemplarily includes a plurality of support members 250.1a, each of which connects two adjacent fins 226. The support members 250.1a can be constructed in the same way as the fins, but the fins 226 extend along the y direction, while the support members 250.1 extend perpendicularly thereto in the x direction, but in the same extension plane, namely the xy plane.
[0033] In particular, each fin of the cell field 230.1a is assigned exactly one support 250.1a. Each support 250.1a is in turn assigned exactly two fins. In other words, each two fins are connected to each other via a spacer (support). This has the advantage that each two fins can be stabilized to each other via a common spacer with minimal changes. Here, the support is arranged in the middle, for example, along the extension direction of the fin, i.e., along the y direction, so that the same support effect is produced in the direction towards the two ends of the cell field.
[0034] In the cell field 230.1b, the support structure exemplarily includes a plurality of supports 250.1b, which also connect two fins 226 side by side. But here in particular, in addition to the two outer fins, each fin of the cell field 230.1b is respectively assigned exactly two supports 250.1b (only one support for each of the two outer fins). Each support 250.1b is in turn assigned to exactly two fins. In other words, a spacer (all supports together) connects all fins and stabilizes them. Therefore, the individual supports 250.2a are all arranged at the same height in the extension direction of the fins, here in the y direction.
[0035] The upper / lower half of the fin, ie at the end side of the cell field, is therefore contacted by the upper / lower gate lead, respectively. This has the advantage that all fins are mutually stabilized, rather than only two fins in each case.
[0036] In any case, care must be taken to ensure that no gate electrode regions are produced by the support structure which are no longer connected to the opening and are therefore no longer in electrical contact.
[0037] Figure 2b A field effect transistor 200b is shown with two exemplary cell fields 230.1b and 230.2b. The field effect transistor 200b itself may correspond to the field effect transistor 200a, the difference here being only in the support structure for the fin.
[0038] In the cell array 230.2a, the support structure includes, by way of example, a plurality of support elements 250.2a, which also respectively connect two adjacent fins 226. In particular, in addition to the two outer fins, each fin of the cell array 230.2a is assigned exactly two support elements 250.2a (two outer fins each have only one support element). Each support element 250.2a is in turn assigned exactly two fins. Figure 2a The difference between the cell field 250.2a in FIG. 2 is that the individual fins 250.1b are not all arranged at the same height in the extension direction of the fins, here in the y direction, but are arranged alternately at different heights (but overall, for example, only at two different heights). The two heights are, for example, approximately one third or two thirds of the length of the fin 226, always starting from the same end of the fin on one side of the cell field.
[0039] In other words, all fins are again connected to each other via a stabilizing spacer. However, the stabilizing spacer is not continuous, but each two spacers or support members are connected to each other in a staggered manner. This has the advantage that the length of the fin 226 to the next support member is shortened.
[0040] In the cell field 230.2b, the support structure exemplarily includes a plurality of support members 250.2b, which also connect two side-by-side fins 226, respectively. In particular, in addition to the two outer fins, each fin of the cell field 230.2b is respectively assigned exactly two support members 250.2b (only one support member for each of the two outer fins). Each support member 250.2b is in turn assigned to exactly two fins. As in the cell field 250.1b, the individual support members 250.2b are not all arranged at the same height in the extension direction of the fin, here in the y direction. Unlike in the cell field 250.1b, the support members 250.2b are alternately arranged at one end and the other end of the fin 226, respectively.
[0041] In other words, the support of each two fins is arranged at the ends of the two fins, so that (together with the support) a fin meander is formed. This has the advantage that the number of corners is reduced and can be completely avoided by rounding the meander. This is advantageous for processing.
[0042] Figure 2c A field effect transistor 200c is shown with two exemplary cell fields 230.1c and 230.2c. The field effect transistor 200c itself may correspond to the field effect transistor 200, the difference here being only in the support structure for the fins.
[0043] In particular, each support element 250.1c is assigned to exactly one fin 226. And each fin 226 is respectively assigned to a plurality of support elements. In this case, the support element 250.1c is larger than the fin 226 when viewed in the x direction. Figure 2a and 2b The supports in the fin are short so that they do not touch or attach the second fin. A continuous gate trench from one end of the cell field to the other can thus be achieved, ie the gate electrode does not have to be interrupted.
[0044] Here, for each fin, the individual supports 250.1c are arranged at different heights in the extension direction of the fin, here in the y direction, and are also arranged alternately on one side or the other side of the fin (the side of the fin can be understood here as in the x direction). This allows good stability.
[0045] In the cell field 230.2c, the support structure exemplarily includes a plurality of support members 250.2c. Here, each support member 250.2c is also assigned to exactly one fin 226. Each fin 226 is respectively assigned a plurality of support members. When viewed in the x direction, the support member 250.2c is smaller than the fin 226. Figure 2a and 2b The supports in the fin are short so that these supports also do not touch or attach the second fin. A continuous gate trench from one end of the cell field to the other can thus be achieved, ie the gate electrode does not have to be interrupted.
[0046] Unlike in the unit field 230.1c, pairs of supports 250.2c are formed here, wherein the two supports of a pair are arranged at the same height in the extension direction of the fin, here in the y direction. Exemplarily, each fin is provided with multiple pairs of supports, wherein the support pairs of each fin are arranged at different heights in the extension direction of the fin, here in the y direction. In addition, different numbers of support pairs for fins are provided here by way of example, for example, three pairs and two pairs are alternately provided. The support pairs of adjacent fins are again exemplarily arranged at different heights in the extension direction of the fin, here in the y direction. This also allows good stability.
[0047] Figure 3 The flow of the method according to the present invention in a preferred embodiment is schematically depicted.
[0048] In step 300 , firstly an n-doped source layer 104 , an n-doped drain layer 120 and a channel layer 106 located vertically therebetween can be produced as a layer stack. Then, in step 304 , the starting material 302 obtained in this way can be provided.
[0049] In step 306, one can then construct Figure 2a , 2b , 2c; here, for example, the gate trench material is removed by etching. However, here, together with the fin, a Figure 2a , 2b , and the support structure 308 shown in one of 2c. This can be done, for example, using a mask 310, which is applied to the starting material, for example, and reflects the fins together with the support structure.
[0050] Then, in step 312, the field effect transistor may be completed by constructing a gate oxide, a gate semiconductor material and a gate insulation (ie, inserting a gate electrode) in the channel or in the gate trench, as well as contact contacts and power metallization.
Claims
1. A field effect transistor (200a, 200b, 200c) comprising: a first type doped source layer (104), a first type doped drain layer (120), a channel layer (106) vertically located between the first type doped source layer and the first type doped drain layer (120), and a fin and a gate trench extending in particular horizontally, wherein: The fin and the gate trench extend from a first-type doped source layer (104) to a first-type doped drain layer (120) in a vertical direction, respectively, and the gate trench is constructed between two side-by-side fins. The field effect transistor (200a, 200b, 200c) further comprises a support structure (308) for the fin (226).
2. The field effect transistor (200a, 200b, 200c) according to claim 1, wherein: The support structure extends in a vertical direction from a first type doped source layer (104) to a first type doped drain layer (120).
3. The field effect transistor (200a, 200b) according to claim 1 or 2, wherein: The support structure (308) includes one or more support members (250.1a, 250.2a), each of which connects two fins arranged side by side.
4. The field effect transistor (200a) according to claim 3, wherein: One or more of the fins are each assigned exactly one support element.
5. The field effect transistor (200b, 200c) according to claim 3 or 4, wherein: One or more of the fins are each assigned exactly two support elements, which are arranged on different sides of the respective fin.
6. The field effect transistor (200c) according to any one of the preceding claims, wherein: The support structure comprises one or more supports (250.1c, 250.2c) each of which is attached to only one of the fins.
7. The field effect transistor (200c) according to claim 6, wherein: The support structure includes a plurality of support members grouped in pairs, wherein a pair of support members are respectively attached to the same height on both sides of one of the fins.
8. The field effect transistor (200c) according to claim 7, wherein: Each fin is assigned a plurality of support members.
9. The field effect transistor (200c) according to claim 6 or 7, wherein: The support structure includes a plurality of support members, wherein at least two support members among the plurality of support members are respectively attached to two sides of one of the fins at different heights.
10. The field effect transistor (200a, 200b, 200c) according to claim 1, further comprising a gate electrode (110, 210) in the gate trench (111), the gate electrodes each being at least partially surrounded by a dielectric, wherein: Each region of the gate electrode is connected to at least one opening area for contact.
11. The field effect transistor (200a, 200b, 200c) according to one of the preceding claims, which is designed as a silicon carbide field effect transistor or a gallium nitride field effect transistor or a gallium oxide field effect transistor.
12. Method for producing a field effect transistor (200a, 200b, 200c), in particular according to one of the preceding claims, comprising the following steps: Providing (304) an initial material (302) having: a first type doped source layer (104), a first type doped drain layer (120), and a channel layer (106) vertically located between the first type doped source layer and the first type doped drain layer (120), Fins are formed (306) in a starting material (302) by removing at least material of a gate trench located between two adjacent fins (226), wherein a support structure (308) is formed for at least one of the fins.
13. The method according to claim 12, wherein: The construction (306) of the fins and support structures (308) is performed using a mask (310) that reflects the fins and support structures.
