Radio frequency (RF) switches are devices used to route high frequency signals and are commonly used in wireless communication systems. For example, RF switches may be found in cell phones, WiFi routers, and so on. Compared to complementary metal-oxide semiconductor (CMOS) switches, RF switches generally have higher power handling, better linearity, and a wider frequency band of operation.
Aspects of the present disclosure are best understood from the following detailed description when read with the accompanying figures. It is noted that, in accordance with the standard practice in the industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion.
The present disclosure provides many different embodiments, or examples, for implementing different features of this disclosure. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, merely examples and are not intended to be limiting. For example, the formation of a first feature over or on a second feature in the description that follows may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features may be formed between the first and second features, such that the first and second features may not be in direct contact. In addition, the present disclosure may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed.
Further, spatially relative terms, such as “beneath,” “below,” “lower,” “above,” “upper” and the like, may be used herein for ease of description to describe one element or feature's relationship to another element(s) or feature(s) as illustrated in the figures. The spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. The apparatus may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein may likewise be interpreted accordingly.
One type of radio frequency (RF) switch is a phase change material (PCM) RF switch, which comprises a heater, a PCM structure, and a pair of electrodes. The electrodes adjoin the PCM structure respectively on opposite sides of the PCM structure. The PCM structure overlies the heater and has a uniform thickness. Further, the PCM structure has an active portion and a pair of inactive portions between which the active portion is arranged.
The inactive portions are in a crystalline phase and extend from the active portion respectively to the electrodes. Because the inactive portions are in the crystalline phase, the inactive portions have a low resistance. The active portion overlies the heater and is configured to selectively change between the crystalline phase and an amorphous phase by selective heating from the heater. While in the amorphous phase, the active portion is in a high resistance state (HRS) that electrically isolates the electrodes from each other. While in the crystalline phase, the active portion is in a low resistance state (LRS) that electrically couples the electrodes together. The HRS may correspond to an OFF state for the PCM RF switch, whereas the LRS may correspond to an ON state for the PCM RF switch.
Performance of the PCM RF switch may be assessed in terms of ON resistance and thermal endurance. The lower the ON resistance, the lower the insertion losses. Increasing a thickness of the PCM structure reduces the ON resistance. However, increasing the thickness of the PCM structure also leads to voids in the PCM structure due to thermal stress while cycling the PCM RF switch. These voids predominantly form at the active portion, where thermal stress is highest, and reduce the thermal endurance of the PCM RF switch. Therefore, there is a tradeoff between ON resistance and thermal endurance.
The present disclosure is directed to a PCM RF switch with a PCM recess. The PCM recess is in a top of the PCM structure and overlies the heater at an active portion of the PCM structure. Because of the PCM recess, a thickness of the PCM RF structure may be small at the active portion. The small thickness prevents voids from forming at the active portion while under thermal stress from cycling of the PCM RF switch. This leads to a high thermal endurance for the PCM RF switch. Note that voids predominantly form at the active portion, where thermal stress is highest, whereby voids are not an issue at the inactive portions. Further, because of the PCM recess, the thickness of the PCM RF structure may be large at the inactive portions. The large thickness reduces resistance at the inactive portions and hence leads to a low ON resistance and low insertion losses. Hence, the PCM recess allows the PCM RF switch to obtain the benefits of both a small PCM thickness and a large PCM thickness.
With reference to
The active portion 106a has a width Wa less than the width Wh of the heater 108 and is configured to selectively change between an amorphous phase and a crystalline phase by selective heating from the heater 108. For example, the active portion 106a may be set to the amorphous phase by heating the active portion 106a to a temperature in excess of a melting temperature for a first time period and quenching (e.g., rapidly cooling) the active portion 106a. As another example, the active portion 106a may be set to the crystalline phase by heating the active portion 106a to a temperature above a crystallization temperature and below the melting temperature for a second time period greater than the first time period.
While in the amorphous phase, the active portion 106a is in a HRS. While in the crystalline phase, the active portion 106a is in a LRS. The HRS may correspond to an OFF state for the PCM RF switch 102, whereas the LRS may correspond to an ON state for the PCM RF switch 102. The active portion 106a is to be contrasted with a remainder of the PCM structure 106, which corresponds to a pair of inactive portions 106ia.
The pair of inactive portions 106ia are respectively on opposite sides of the active portion 106a and extend from the active portion 106a respectively to a first electrode 110a and a second electrode 110b. As a result, the inactive portions 106ia separate the first and second electrodes 110a, 110b from the active portion 106a. Further, the inactive portions 106ia are in the crystalline phase and remain in the crystalline phase during cycling (e.g., ON/OFF cycling) of the PCM RF switch 102. Because the inactive portions 106ia are in the crystalline phase, the inactive portions 106ia have a low resistance and provide low-resistance electrical paths from the active portion 106a respectively to the first and second electrodes 110a, 110b.
Because of the PCM recess 104, an active thickness Ta at the active portion 106a and an inactive thickness Tia at the inactive portions 106ia are different. Particularly, the active and inactive thicknesses Ta, Tia are comparatively and respectively small and large.
The small thickness Ta at the active portion 106a prevents voids from forming at the active portion 106a while under thermal stress from cycling of the PCM RF switch 102. For example, the small thickness Ta may lead to faster heat dissipation that reduces thermal stress and reduces voids. By preventing voids, the small thickness Ta leads to a high thermal endurance for the PCM RF switch 102. Note that voids predominantly form at the active portion 106a, where thermal stress is highest, whereby voids are not an issue at the inactive portions 106ia. The large thickness Tia at the inactive portions 106ia leads to a low resistance at the inactive portions 106ia. The low resistance leads to a low ON resistance for the PCM RF switch 102. Hence, the PCM recess 104 allows a the PCM RF switch 102 to obtain the benefits of both a small thickness (e.g., Ta) and a large thickness (e.g., Tia).
As a ratio of the active thickness Ta to the inactive thickness Tia (e.g., Ta/Tia) decreases, an ON resistance of the PCM RF switch 102 deceases. For example, the ratio being 1 may yield an ON resistance of X, the ratio being 0.77 may yield an ON resistance of 0.84*X, and the ratio being 0.67 may yield an ON resistance of 0.77*X. In some embodiments, the ratio is greater than or equal to 0.5 and is less than 1. If the ratio is outside this range, an ON resistance of the PCM RF switch 102 may, for example, be high and/or thermal endurance of the PCM RF switch 102 may, for example, be low.
With continued reference to
The first and second electrodes 110a, 110b are level with the heater 108 and are laterally separated from the heater 108 by a first dielectric layer 112a and a separation S. In some embodiments, the separation S is 0.15 micrometer or some other suitable value or range of values. Further, the first and second electrodes 110a, 110b underlie and directly contact the PCM structure 106 respectively on opposite sides of the PCM structure 106.
A passivation layer 114 overlies the heater 108 and the first dielectric layer 112a to separate the heater 108 from the PCM structure 106. Further, the passivation layer 114 is recessed into a bottom of the PCM structure 106, such that the passivation layer 114 and the PCM structure 106 have individual bottom surfaces that are level with each other.
During use of the PCM RF switch 102, the first and second electrodes 110a, 110b serve as RF transmission lines or are otherwise electrically coupled to RF transmission lines. A signal is input into the first electrode 110a and is selectively passed to the second electrode 110b depending on an ON/OFF state of the PCM RF switch 102. Further, the heater 108 is controlled to change the PCM RF switch 102 between an ON state and an OFF state. Compared to complementary metal-oxide semiconductor (CMOS) switches, the PCM RF switch 102 has higher power handling, better linearity, and a wider frequency band of operation.
In some embodiments, while the PCM RF switch 102 is in the ON state, the PCM structure 106 is entirely crystalline. Further, in some embodiments, while the PCM RF switch 102 is in the OFF state, the active portion 106a (demarcated by dashed lines) is entirely amorphous and a remainder of the PCM structure 106 is entirely crystalline.
In some embodiments, the first and second electrodes 110a, 110b are or comprise tungsten and/or some other suitable material having low resistance and high heat resistance. In some embodiments, the heater 108 is or comprises tungsten and/or some other suitable material having low resistance and high heat resistance. In some embodiments, the PCM structure 106 is or comprises germanium telluride and/or some other suitable phase change material. In some embodiments, the first dielectric layer 112a is or comprises silicon oxide, some other suitable oxide and/or dielectric, or any combination of the foregoing. In some embodiments, the passivation layer 114 is or comprises silicon nitride and/or some other suitable dielectric.
In some embodiments, the active thickness Ta is about 500-1000 angstroms, about 500-750 angstroms, about 750-1000 angstroms, or some other suitable value or range of values. If the active thickness Ta is too large (e.g., more than 1000 angstroms), voids may form at the active portion 106a while under thermal stress from cycling of the PCM RF switch 102. Hence, thermal endurance of the PCM RF switch 102 may be low. If the active thickness Ta is too small (e.g., less than 500 angstroms), a maximum current of the PCM RF switch 102 may be low and/or an ON resistance of the PCM RF switch 102 may be high.
In some embodiments, the inactive thickness Tia is about 750-1500 angstroms, about 750-1125 angstroms, about 1125-1500 angstroms, or some other suitable value or range of values. If the inactive thickness Tia is too small (e.g., less than 750 angstroms), a maximum current of the PCM RF switch 102 may be low and/or an ON resistance of the PCM RF switch 102 may be high. If the inactive thickness Tia is too large (e.g., more than 1500), a height of the PCM RF switch 102 may be high and may lead to a large amount of topographical variation at the PCM RF switch 102. Such topographical variation may be disruptive to semiconductor process uniformity and may lead to low manufacturing yields for the PCM RF switch 102.
In some embodiments, the width Wr of the PCM recess 104 is or comprises 0.4 micrometers, 0.35-0.45 micrometers, or some other suitable value or range of values. Further, in some embodiments, the width Wa of the active portion 106a is or comprises 0.4 micrometers, 0.35-0.45 micrometers, or some other suitable value or range of values. In some embodiments, the width Wr of the PCM recess 104 and the width Wa of the active portion 106a are the same or substantially the same as each other. In other embodiments, the width Wr of the PCM recess 104 is greater than the width Wa of the active portion 106a.
In some embodiments, a ratio of the width Wa of the active portion 106a to the width Wr of the PCM recess 104 (e.g., Wa/Wr) is greater than or equal to 0.9 and less than or equal to 1. If the ratio is too small (e.g., less than 0.9), the PCM recess 104 may extend a large amount into the inactive portions 106ia and may materially degrade ON resistance. If the ratio is too large (e.g., more than 1), the PCM recess 104 may fail to cover a large amount of the active portion 106a. Hence, voids may form and materially degrade thermal endurance.
In some embodiments, the width Wh of the heater 108 is or comprises 1 micrometer, 0.875-1.125 micrometers, or some other suitable value or range of values. In some embodiments, a ratio of the width Wa of the active portion 106a to the width Wh of the heater 108 (e.g., Wa/Wh) is about 0.4 and/or some other suitable value.
With reference to
A hard mask 202 overlies the PCM structure 106 outside the PCM recess 104, which is extended through the hard mask 202. Further, the hard mask 202 comprises a first hard mask layer 202a and a second hard mask layer 202b overlying the first hard mask layer 202a. In alternative embodiments, the first hard mask layer 202a or the second hard mask layer 202b is omitted. The first hard mask layer 202a may, for example, be or comprise silicon nitride and/or some other suitable dielectric(s), whereas the second hard mask layer 202b may, for example, be or comprise silicon oxide and/or some other suitable dielectric(s), or vice versa.
A first cap layer 204 overlies the hard mask 202 outside the PCM recess 104, which is extended through the first cap layer 204. Further, the first cap layer 204 extends to the first and second electrodes 110a, 110b along a pair of first common sidewalls, which are formed by the PCM structure 106 and the hard mask 202 and which face away from each other respectively on opposite sides of the PCM structure 106. The first cap layer 204 may, for example, be or comprise silicon nitride and/or some other suitable dielectric(s). In some embodiments, the first cap layer 204 is the same material as the first hard mask layer 202a.
A second cap layer 206 overlies and lines the first cap layer 204 and further lines the PCM recess 104. Further, the first cap layer 204, the hard mask 202, and the PCM structure 106 form a pair of second common sidewalls opposing each other in the PCM recess 104 and lined by the second cap layer 206. The second cap layer 206 may, for example, be or comprise silicon nitride and/or some other suitable dielectric(s). In some embodiments, the second cap layer 206 is the same material as the first hard mask layer 202a and/or as the first cap layer 204.
The passivation layer 114 includes a trio of segments laterally spaced from each other and respectively overlying the heater 108 and the first and second electrodes 110a, 110b. Further, the segments underlie the first and second cap layers 204, 206. The passivation layer 114 may, for example, be or comprise silicon nitride and/or some other suitable dielectric(s). In some embodiments, the passivation layer 114 is the same material as at least one or more of the first hard mask layer 202a, the first cap layer 204, or the second cap layer 206.
The first dielectric layer 112a is extended under the first and second electrodes 110a, 110b and the heater 108. As such, the first dielectric layer 112a spaces the first and second electrodes 110a, 110b and the heater 108 from a first etch stop layer 208a, which underlies the first dielectric layer 112a. In alternative embodiments, the first etch stop layer 208a is omitted. The first etch stop layer 208a is a different dielectric material than the first dielectric layer 112a and may, for example, be or comprise silicon carbide and/or some other suitable dielectric(s).
A second dielectric layer 112b overlies the PCM RF switch 102. The second dielectric layer 112b may, for example, be or comprise silicon oxide, some other suitable oxide and/or dielectric, or any combination of the foregoing. In some embodiments, the second dielectric layer 112b is the same dielectric material as the first dielectric layer 112a.
A first via 210a and a second via 210b are respectively on opposite sides of the PCM RF switch 102. Further, the first and second vias 210a, 210b respectively overlie and extend from the first and second electrodes 110a, 110b to a top of the second dielectric layer 112b. The first and second vias 210a, 210b may, for example, be or comprise aluminum, copper, aluminum copper, some other suitable metal(s), or any combination of the foregoing.
With reference to
The heater 108 comprises a central portion 108c and a pair of pad portions 108p. The central portion 108c has a columnar shape elongated between the pad portions 108p, in a first dimension along which the PCM recess 104 and the PCM structure 106 are also elongated. Further, the central portion 108c is laterally between and borders the first and second electrodes 110a, 110b in a second dimension orthogonal to the first dimension. The pad portions 108p are respectively on opposite ends of the central portion 108c and have rectangular shapes. In some embodiments, the heater 108 has I shape. In alternative embodiments, at least one or more of the central portion 108c or the pad portions 108p has/have some other suitable shape.
With reference to
With reference to
In
In
In
In
In
The first and second electrodes 110a, 110b, a third cap layer 504, and the second dielectric layer 112b overlie the flat surface. The first and second electrodes 110a, 110b have a planar profile overlying the PCM structure 106, respectively on opposite sides of the PCM structure 106. The third cap layer 504 overlies and lines the first and second electrodes 110a, 110b, and the second dielectric layer 112b overlies the third cap layer 504.
As with
In some embodiments, the planarization dielectric layer 502 is or comprises silicon oxide, some other suitable oxide and/or dielectric, or any combination of the foregoing. In some embodiments, the planarization dielectric layer 502 is the same material as the first dielectric layer 112a and/or is the same material as the second dielectric layer 112b. In some embodiments, the third cap layer 504 is or comprises silicon nitride and/or some other suitable dielectric(s). In some embodiments, the third cap layer 504 is the same material as one or more of the first cap layer 204, the second cap layer 206, or the passivation layer 114.
In
With reference to
An interconnect structure 602 overlies a semiconductor substrate 604 and comprises a plurality of wires 606 and a plurality of vias 210. The plurality of wires 606 are grouped into a plurality of wire levels, and the plurality of vias 210 are grouped into a plurality of via levels alternatingly stacked with the plurality of wire levels.
The wire levels are labeled M1, M2, and so on to M9 from a bottom of the interconnect structure 602 to a top of the interconnect structure 602. Further, the wires 606 at wire level M8 and M9 have greater heights than their counterparts at lower wire levels. The via levels are labeled V0, V1, and so on to V8 from the bottom of the interconnect structure 602 to the top of the interconnect structure 602. Further, the vias 210 at via levels V0, V7, and V8 have greater heights than their counterparts at other via levels.
In alternative embodiments, the interconnect structure 602 has more or less wire levels and/or more or less via levels. Further, in alternative embodiments, the interconnect structure 602 further comprises a middle end of line (MEOL) underlying the via level V0. The MEOL comprises a pair of contact levels, which may also be referred to as CA and CB.
A dielectric structure surrounds the interconnect structure 602 and comprises a plurality of interconnect dielectric layers 112 and a plurality of etch stop layers 208. The interconnect dielectric layers 112 and the etch stop layers 208 are alternatingly stacked from a bottom of the interconnect structure 602 to a top of the interconnect structure 602. The etch stop layers 208 enhance control over etching to form the various wire and via levels.
In some embodiments, the etch stop layers 208 are or comprise silicon carbide and/or some other suitable dielectric(s). In some embodiments, the plurality of interconnect dielectric layers 112 are or comprise undoped silicate glass (USG), borosilicate glass (BSG), silicon oxide, some other suitable dielectric, or any combination of the foregoing.
The PCM RF switch 102 is in the interconnect structure 602, vertically between the wire level M7 and the wire level M8. In alternative embodiments, the PCM RF switch 102 is vertically between two other wire levels vertically neighboring in the interconnect structure 602. The interconnect structure 602 electrically couples to the PCM RF switch 102 to facilitate electrical routing of RF signals respectively to and from the first and second electrodes 110a, 110b and to further facilitate control over the heater 108.
As noted above, the PCM RF switch 102 is as in
A heat sink 608 underlies the heater 108 in the interconnect structure 602. The heat sink 608 is formed by wires at the wire level M7 and the wire level M6, as well by vias at the via level V6. The heat sink 608 facilitates dissipation of heat while cycling the PCM RF switch 102. For example, the heat sink 608 may aid in quenching the active portion 106a (see
A pair of logic devices 610 underlie and electrically couple to the interconnect structure 602, between the interconnect structure 602 and the semiconductor substrate 604. Further, the logic devices 610 are partially formed in the semiconductor substrate 604. In alternative embodiments, there may be more or less logic devices 610 and/or there may additionally or alternatively be devices other than logic devices. In some embodiments, the semiconductor substrate 604 is or comprises silicon and/or the like.
The logic devices 610 are separated from each other by an isolation structure 612. The isolation structure 612 may, for example, be or comprise a shallow trench isolation (STI) structure, a deep trench isolation (DTI) structure, a local oxidation of silicon (LOCOS) isolation structure, some other suitable type of isolation structure, or any combination of the foregoing. The logic devices 610 are illustrated as planar field-effect transistors (planar FETs). However, the logic devices 610 may alternatively be fin field-effect transistors (FinFETs), gate-all-around (GAA) field-effect transistors (GAA FETs), some other suitable type(s) of logic device and/or transistor, or any combination of the foregoing.
The logic devices 610 comprise individual gate electrodes 614, individual gate dielectric layers 616, individual pairs of source/drain regions 618, and individual wells 620. The pairs of source/drain regions 618 are inset into a top of the semiconductor substrate 604. The gate electrodes 614 respectively overlie the gate dielectric layers 616 and are respectively between the pairs of source/drain regions 618. The wells 620 correspond to doped regions of the semiconductor substrate 604 and respectively underlie the pairs of source/drain regions 618 and the gate electrodes 614. In alternative embodiments, the wells 620 are omitted.
With reference to
With reference to
As illustrated by the cross-sectional view 800 of
In some embodiments, the conductive layer 802 is or comprises tungsten and/or some other suitable material having low resistance and high heat resistance. In some embodiments, the first dielectric layer 112a is or comprises silicon oxide, some other suitable oxide and/or dielectric, or any combination of the foregoing. In some embodiments, the first etch stop layer 208a is or comprises silicon carbide and/or some other suitable dielectric(s).
As illustrated by the cross-sectional view 900 of
As illustrated by the cross-sectional view 1000 of
As illustrated by the cross-sectional view 1100 of
As illustrated by the cross-sectional view 1200 of
As illustrated by the cross-sectional view 1300 of
As illustrated by the cross-sectional view 1400 of
As illustrated by the cross-sectional view 1500 of
As illustrated by the cross-sectional view 1600 of
The patterning may, for example, be performed by or using a photolithography/etching process or some other suitable patterning process. Further, the patterning may, for example, comprise etching the PCM layer 1061, the first hard mask layer 202a, and the second hard mask layer 202b while a mask 1602 overlies these layers (e.g., the second hard mask layer 202b) and subsequently removing the mask 1602.
As illustrated by the cross-sectional view 1700 of
As illustrated by the cross-sectional view 1800 of
The active portion 106a is configured to selectively change between an amorphous phase and a crystalline phase by selective heating from the heater 108. For example, the active portion 106a may be set to the amorphous phase by heating the active portion 106a to a temperature in excess of a melting temperature for a first time period and quenching (e.g., rapidly cooling) the active portion 106a. As another example, the active portion 106a may be set to the crystalline phase by heating the active portion 106a to a temperature above a crystallization temperature and below the melting temperature for a second time period that is greater than the first time.
While in the amorphous phase, the active portion 106a is in a HRS. While in the crystalline phase, the active portion 106a is in an LRS. The HRS may correspond to an OFF state for the PCM RF switch 102, whereas the LRS may correspond to an ON state for the PCM RF switch 102. The active portion 106a is to be contrasted with a remainder of the PCM structure 106, which corresponds to a pair of inactive portions 106ia. In some embodiments, the PCM recess 104, the active portion 106a, and the inactive portions 106ia have dimensions (e.g., widths, thicknesses, etc.) as described with regard to
The pair of inactive portions 106ia are respectively on opposite sides of the active portion 106a and extend from the active portion 106a respectively to a first electrode 110a and a second electrode 110b. As a result, the inactive portions 106ia separate the first and second electrodes 110a, 110b from the active portion 106a. Further, the inactive portions 106ia are in the crystalline phase and remain in the crystalline phase during cycling (e.g., ON/OFF cycling) of the PCM RF switch 102. Because the inactive portions 106ia are in the crystalline phase, the inactive portions 106ia have a low resistance and provide low-resistance electrical paths from the active portion 106a respectively to the first and second electrodes 110a, 110b.
Because of the PCM recess 104, an active thickness Ta at the active portion 106a and an inactive thickness Tia at the inactive portions 106ia are different. Particularly, the active and inactive thicknesses Ta, Tia are comparatively and respectively small and large.
The small thickness Ta at the active portion 106a prevents voids from forming at the active portion 106a while under thermal stress from cycling of the PCM RF switch 102. For example, the small thickness Ta may lead to faster heat dissipation that reduces thermal stress and reduces voids. By preventing voids, the small thickness Ta leads to a high thermal endurance for the PCM RF switch 102. Note that voids predominantly form at the active portion 106a, where thermal stress is highest, whereby voids are not an issue at the inactive portions 106ia. The large thickness Tia at the inactive portions 106ia leads to a low resistance at the inactive portions 106ia. The low resistance leads to a low ON resistance for the PCM RF switch 102. Hence, the PCM recess 104 allows a the PCM RF switch 102 to obtain the benefits of both a small thickness (e.g., Ta) and a large thickness (e.g., Tia).
As illustrated by the cross-sectional view 1900 of
As illustrated by the cross-sectional view 2000 of
As illustrated by the cross-sectional view 2100 of
While
With reference to
At block 2202, a pair of electrodes and a heater are formed inset into a top of a first dielectric layer, wherein the heater is between and spaced from the electrodes. See, for example,
At block 2204, a passivation layer is deposited overlying the pair of electrodes, the heater, and the first dielectric layer. See, for example,
At block 2206, the passivation layer is patterned to form a pair of openings respectively exposing the electrodes and laterally offset from the heater. See, for example,
At block 2208, a PCM layer is deposited filling the openings and overlying the passivation layer. See, for example,
At block 2210, a hard mask layer is deposited overlying the PCM layer. See, for example,
At block 2212, the PCM layer and the hard mask layer are patterned respectively into a PCM structure and a hard mask atop the PCM structure, wherein the PCM structure has a pair of outer sidewalls respectively in the openings. See, for example,
At block 2214, a first cap layer is deposited overlying the hard mask and lining the outer sidewalls. See, for example,
At block 2216, the first cap layer, the hard mask, and the PCM structure are patterned to form a PCM recess overlying the heater at an active portion of the PCM structure. See, for example,
At block 2218, a second cap layer is deposited overlying the first cap layer and lining the PCM recess. See, for example,
At block 2220, a second dielectric layer is deposited covering the second cap layer. See, for example,
At block 2222, vias are formed in the second dielectric layer and extending respectively from the electrodes. See, for example,
While the block diagram 2200 of
With reference to
As illustrated by the cross-sectional view 2300 of
The patterning may, for example, be performed by or using a photolithography/etching process or some other suitable patterning process. Further, the patterning may, for example, comprise etching the first dielectric layer 112a while a mask 2304 overlies the first dielectric layer 112a and subsequently removing the mask 2304.
As illustrated by the cross-sectional view 2400 of
As illustrated by the cross-sectional view 2500 of
As illustrated by the cross-sectional view 2600 of
As illustrated by the cross-sectional view 2700 of
While
With reference to
As illustrated by the cross-sectional view 2800 of
As illustrated by the cross-sectional view 2900 of
As illustrated by the cross-sectional views 3000-3400 of
As illustrated by the cross-sectional view 3500 of
As illustrated by the cross-sectional view 3600 of
As illustrated by the cross-sectional view 3700 of
As illustrated by the cross-sectional view 3800 of
In some embodiments, the conductive layer 3802 is deposited by an anisotropic deposition process. For example, the conductive layer 3802 may be deposited by an evaporation deposition process and/or some other suitable deposition process. Further, in some embodiments, sidewalls of the mask 3702 in the openings 3704 are formed slanted, such that individual widths of the openings 3704 increase top to bottom. The slanted sidewalls may minimize lateral accumulation from the slanted sidewalls.
As illustrated by the cross-sectional view 3900 of
As illustrated by the cross-sectional view 4000 of
As illustrated by the cross-sectional view 4100 of
As illustrated by the cross-sectional view 4200 of
While
With reference to
At block 4302, a heater is formed inset into a top of a first dielectric layer. See, for example,
At block 4304, a passivation layer, a PCM layer, and a hard mask layer are formed vertically stacked overlying the heater and the first dielectric layer. See, for example,
At block 4306, the PCM layer and the hard mask layer are patterned respectively into a PCM structure and a hard mask atop the PCM structure. See, for example,
At block 4308, a first cap layer is deposited overlying the hard mask and lining sidewalls of the PCM structure. See, for example,
At block 4310, the first cap layer, the hard mask, and the PCM structure are patterned to form a PCM recess overlying the heater at an active portion of the PCM structure. See, for example,
At block 4312, a second cap layer is deposited overlying the first cap layer and lining the PCM recess. See, for example,
At block 4314, a planarization dielectric layer is deposited covering the second cap layer. See, for example,
At block 4316, a planarization is performed into the planarization dielectric layer, the first and second cap layers, and the hard mask to expose a top surface of the PCM structure. See, for example,
At block 4318, a pair of electrodes are formed overlying the PCM structure and the planarization dielectric layer, respectively on opposite sides of the PCM structure. See, for example,
At block 4320, a third cap layer is deposited overlying and lining the electrodes. See, for example,
At block 4322, a second dielectric layer is deposited covering the third cap layer. See, for example,
At block 4324, vias are formed in the second dielectric layer, extending respectively from the electrodes. See, for example,
While the block diagram 4300 of
With reference to
As illustrated by the cross-sectional view 4400 of
The logic devices 610 comprise individual gate electrodes 614, individual gate dielectric layers 616, individual pairs of source/drain regions 618, and individual wells 620. In alternative embodiments, the wells 620 are omitted.
As illustrated by the cross-sectional view 4500 of
The interconnect structure 602 is formed so wires at the wire levels M6 and M7, as well as vias at the via level V6, form a heat sink 608 at a top of the interconnect structure 602. In alternative embodiments, the heat sink 608 is omitted. The heat sink 608 facilitates dissipation of heat generated by a heater hereafter formed. Further, in alternative embodiments, the interconnect structure 602 is formed with more or less wire levels and/or more or less via levels. Further yet, in alternative embodiments, the interconnect structure 602 is formed with a MEOL between the via level V0 and the logic devices 610.
Also illustrated by the cross-sectional view 4500 of
As illustrated by the cross-sectional view 4600 of
As illustrated by the cross-sectional view 4700 of
Also illustrated by the cross-sectional view 4700 of
While
In some embodiments, the present disclosure provides a semiconductor device, comprising: a heater over a substrate; a PCM structure over the heater; and a first electrode and a second electrode respectively on opposite sides of the heater and electrically coupled to the PCM structure respectively on the opposite sides; wherein a top of the PCM structure has a recess that overlies the heater and that has a width different (e.g., less) than a width of the heater. In some embodiments, a width of the PCM structure is greater than the width of the heater. In some embodiments, the first and second electrodes underlie the PCM structure and are level with the heater. In some embodiments, the first and second electrodes overlie the PCM structure and extend along individual sidewalls of the PCM structure. In some embodiments, the first and second electrodes have individual bottommost surfaces overlying a top surface of the PCM structure. In some embodiments, the heater is laterally between the first and second electrodes in a first dimension, wherein the heater has a pair of pads laterally spaced from each other in a second dimension orthogonal to the first dimension and between which the first and second electrodes are laterally arranged. In some embodiments, the semiconductor device further comprises: a hard mask overlying the PCM structure, wherein the hard mask and the PCM structure form a first common sidewall overlying the heater in the recess and further form a second common sidewall overlying the first electrode outside the recess; a first cap layer overlying the hard mask, wherein the first cap layer further forms the first common sidewall and lines the second common sidewall; and a second cap layer overlying the first cap layer and lining the first common sidewall and the second common sidewall.
In some embodiments, the present disclosure provides another a semiconductor device, including: a heater over a substrate; a PCM structure overlying the heater; and a first electrode and a second electrode respectively on opposite sides of the heater and electrically coupled to the PCM structure respectively on the opposite sides; wherein the PCM structure has a first thickness overlying the heater, and further has a second thickness laterally offset from the heater, and wherein the first thickness is less than the second thickness. In some embodiments, the PCM structure has a central portion, which overlies the heater and has the first thickness throughout its entirety, wherein a remainder of the PCM structure is thicker than the central portion. In some embodiments, a width of the central portion is less than a width of the heater, which is less than a width of the PCM structure. In some embodiments, the remainder of the PCM structure partially overlies the heater. In some embodiments, the semiconductor device further includes a passivation layer underlying the PCM structure, between the PCM structure and the heater, to separate the PCM structure from the heater. In some embodiments, the passivation layer has a bottom surface level with a bottom surface of the PCM structure and further has a sidewall facing the PCM structure.
In some embodiments, the present disclosure provides a method for forming a semiconductor device, including: forming a heater over a substrate; forming a pair of electrodes respectively on opposite sides of the heater; depositing a passivation layer overlying the heater; forming a PCM structure overlying the passivation layer and the heater; and performing an etch into the PCM structure to form a recess overlying the heater. In some embodiments, a portion of the PCM structure underlying the recess has a first thickness and spans an entire width of the recess, wherein a remainder of the PCM structure is thicker than the portion. In some embodiments, the passivation layer is deposited overlying the pair of electrodes, wherein the method further includes: patterning the passivation layer to form a pair of openings respectively exposing the pair of electrodes and laterally offset from the heater, wherein the PCM structure is formed in the pair of openings. In some embodiments, the forming of the PCM structure includes: depositing a PCM layer overlying the passivation layer; and patterning the PCM layer into the PCM structure, such the PCM structure has a pair of individual sidewalls respectively overlying the pair of electrodes respectively in the pair of openings. In some embodiments, the method further includes: depositing a dielectric layer overlying the PCM structure and filling the recess; and performing a planarization into the dielectric layer until a top surface of the dielectric layer is level with a top surface of the PCM structure. In some embodiments, the forming of the PCM structure includes: depositing a PCM layer overlying the passivation layer; forming a hard mask layer overlying the PCM layer; and performing an additional etch into the PCM layer and the hard mask layer with a common mask in place to respectively form the PCM structure and a hard mask atop the PCM structure, wherein the planarization removes the hard mask. In some embodiments, the method further includes depositing a conductive layer over the substrate, wherein the heater and the pair of electrodes are formed from the conductive layer.
The foregoing outlines features of several embodiments so that those skilled in the art may better understand the aspects of the present disclosure. Those skilled in the art should appreciate that they may readily use the present disclosure as a basis for designing or modifying other processes and structures for carrying out the same purposes and/or achieving the same advantages of the embodiments introduced herein. Those skilled in the art should also realize that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they may make various changes, substitutions, and alterations herein without departing from the spirit and scope of the present disclosure.
This application claims the benefit of U.S. Provisional Application No. 63/584,558, filed on Sep. 22, 2023, the contents of which are incorporated by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 63584558 | Sep 2023 | US |