US20080284463A1

US20080284463A1 - programmable circuit having a carbon nanotube

Info

Publication number: US20080284463A1
Application number: US11/749,857
Authority: US
Inventors: Andrew Marshall; Tito Gelsomini; Harvey Edd Davis
Original assignee: Texas Instruments Inc
Current assignee: Texas Instruments Inc
Priority date: 2007-05-17
Filing date: 2007-05-17
Publication date: 2008-11-20
Also published as: WO2008144607A1; US20090315081A1; US8455305B2

Abstract

A semiconductor device comprising a programming circuit that includes an active device on or in a substrate and a programmable electronic component on the substrate. The programmable electronic component includes at least one carbon nanotube having a segment with an adjusted diameter. The programmable electronic component has a value that depends upon the adjusted diameter. The programming circuit also includes interconnects that couple the active device to the programmable electronic component. The active device is configured to control a current transmitted to the programmable electronic component.

Description

TECHNICAL FIELD

The disclosure is directed, in general, to semiconductor devices, and more specifically, to a device for programming a circuit and its method of manufacture.

BACKGROUND

The programming of application-specific semiconductor devices often relies on the use of fuses as a programming component. To program an integrated circuit device, fuses in the circuit can be selectively left intact, or opened, to create circuit paths according to a predefined design. Fuses can thereby be used to implement a variety of programming functions.
One problem with the use of conventional fuses, however, is that the size of fuses is not scaling down as rapidly as transistor sizes are. This can be problematic in devices that incorporate thousands of fuses to implement increasingly sophisticated circuit programming. That is, the size of fuses can limit the extent of miniaturization of semiconductor devices. Another problem is that only a binary signal information is obtained from a fuse (e.g., a zero or nonzero current). Consequently, to send more complex control signals, several fuses have to be used, thereby increasing the amount of space on a circuit that is occupied by fuses.
Accordingly, what is needed is a method for programming a circuit that addresses the drawbacks of the prior art methods and devices.

SUMMARY

One aspect of the disclosure is a semiconductor device. The device comprises a programming circuit that includes an active device on or in a substrate and a programmable electronic component on the substrate. The programmable electronic component includes at least one carbon nanotube having a segment with an adjusted diameter. The programmable electronic component has a value that depends upon the adjusted diameter. The programming circuit also includes interconnects that couple the active device to the programmable electronic component. The active device is configured to control a current transmitted to the programmable electronic component.
In one embodiment of the device, the programming circuit includes transistors located on or in a substrate, a fusible link on the substrate that includes at least one of the above-described carbon nanotubes, and interconnects that couple the transistors to the programmable electronic component. The transistors are configured to control a current transmitted to the fusible link such that the segment is configured to open when the current, equal a predefined level, is transmitted through the carbon nanotube. The fusible link thereby has a value that depends upon the adjusted diameter, the value configured to equal a zero or nonzero current depending on whether the segment is opened or not opened, respectively.
In another embodiment of the device, the programming circuit includes transistors located on or in a substrate, a capacitor on the substrate and interconnects that couple the transistors to the capacitor. The capacitor includes at least one the above-described carbon nanotubes having a segment with an adjusted diameter and a conductive body capacitively coupled to the carbon nanotube. A distance between the segment and the conductive body is configured to change as a function of the adjusted diameter. The transistors are configured to control a current transmitted to the capacitor, and the capacitor has a value that depends upon the adjustable diameter, the value configured to equal to a capacitance that depends on the adjusted diameter.
Still another aspect of the disclosure is a method of manufacturing a semiconductor device. The method comprises fabricating a programming circuit, including forming an active device on or in a substrate and forming a programmable electronic component. Forming the programmable electronic component including depositing the above-described carbon nanotube on the substrate. Fabricating a programming circuit also include forming interconnects that couple the active device to the programmable electronic component.

BRIEF DESCRIPTION OF DRAWINGS

The disclosure is described with reference to example embodiments and to accompanying drawings, wherein:

FIG. 1 shows a plan view (at the device level) of an example semiconductor device of the disclosure;

FIG. 2 illustrates a cross-sectional view of an example device of the disclosure along view lines 2-2 as depicted in FIG. 1;

FIG. 3 illustrates a cross-sectional view of an example device of the disclosure along view lines 3-3 as depicted in FIG. 1; and

FIG. 4 shows a circuit diagram of a example device of the disclosure;

FIGS. 5 to 12 illustrate cross-section views of selected steps in example implementation of a method of fabricating semiconductor devices of the disclosure.

DETAILED DESCRIPTION

The disclosure benefits from the realization that a programmable electronic component that comprises a carbon nanotube (CNT) provides several advantages over conventional fuses. CNTs are substantially smaller (at least an order of magnitude) than conventional fuse components. Additionally, the diameter of CNTs can be adjusted after forming the CNT in a circuit. A segment of the CNT having the adjusted diameter can be used to facilitate the CNT as a fusible link, or as a capacitor when coupled to a conductive body. Circuitry having such a programmable electronic component can be substantially smaller than conventional fuses.
One embodiment of the present invention is a semiconductor device. FIG. 1 shows a plan view of an example semiconductor device 100 of the disclosure. FIGS. 2 and 3 illustrate cross-sectional view of the device 100, along view lines 2-2 and 3-3, respectively, as depicted in FIG. 1.
The device 100 comprises a programming circuit 105 that includes an active device 110 and a programmable electronic component 115 (FIG. 1). In some embodiments, the semiconductor device 100 is or includes an integrated circuit and the active device 110 and a programmable electronic component 115 are components of the integrated circuit. The plan view of FIG. 1 shows the device 100 at the layer that the active device 110 and programmable electronic component 115 are located in.
The active device 110 is located on or in a substrate 117. Example substrates 117 include semiconductors such as silicon, silicon-on-insulator, or silicon germanium, or non-semiconductors, such as sapphire or quartz. Some embodiments of the active device 110 comprise one or more transistors 120 (FIG. 1). The transistors 120 can comprise an nMOS or pMOS transistor, or combination of such transistors. The active device 110 is configured to control the amount of current 210 (FIG. 2) transmitted to the programmable electronic component 115. As illustrated in FIG. 1 the transistor 120 can comprise a gate 122, source and drain structures, 125. Additional components of the transistor 120, including gate sidewalls 215, and a doped well 220 (FIG. 2). To isolate the active device 110 and programmable electronic component 115 the device 100 can also include insulating structures 127 (e.g., field oxide or shallow trench isolation structures) in or on the substrate 117. In some embodiments the transistor 120 can be configured as a sensor, and the transistors 120 gate 122 is connected to a resistor 225 (FIG. 2) that provides current control to the programmable electronic component 115.
The programmable electronic component 115 is located on the substrate 117 and includes at least one CNT 130 having a segment 135 with an adjusted diameter 140. The term adjusted diameter 140, as used herein, refers to the diameter after exposing the segment 135 to an electron beam to shrink its non-adjusted diameter 145, or after applying a current 210 sufficient to cause an open or short to occur in the segment 135. The term CNT, as used herein refers, to a carbon-based tubular fullerene structures having a non-adjusted diameter 145 of 1 micron or less. Both multi-wall and single-wall CNTs are within the scope of the disclosure.
The device 100 further comprises interconnects 230 (e.g., lines, vias, contacts) that couple the active device 110 to the programmable electronic component 115 (FIG. 2). The interconnects 230 can be patterned metal line (e.g., tungsten), single or dual damacence metal structures (e.g., copper), or other electrically conductive materials that are patterned or deposited on the substrate 117 (e.g., polysilicon or other CNTs). As further illustrated in FIG. 2 the device 100 can further comprise insulating layers 240, such as pre-metal dielectric (PMD) or interlayer dielectric (ILD) layers. The insulating layers 240 help to electrically isolate the active device 110 and the programmable electronic component 115 from each other, or from other active structures in the device 100.
With continuing reference to FIGS. 1-3, the active device 110 is configured to control a current transmitted to the programmable electronic component 115, and the programmable electronic component 115 has a value that depends upon the adjusted diameter 140. One skilled in the art would understand how the value of the programmable electronic component 115 could be used to perform a variety of device programming functions. Examples include programming the device 100 to allow redundant components to replace defective components, adapting the device to perform a specific operation, such as trim an oscillator of the device 100 or trim a voltage of the device 100, or to provide a unique identification code for the device 100.
In some embodiments of the device 100, the programmable electronic component 115 is configured as a fusible link. In such embodiments, the segment 305 is configured to open when a current 210 equal to a predefined level, is transmitted through the CNT 130. E.g., in some embodiments, the active device 110 has transistors 120 that are configured to control a current 210 transmitted to the fusible link. In such embodiments, the segment 135 is configured to open when a current 210, equal to a predefined level, is transmitted through the one or more CNTs 130 of the programmable circuit component 115. In such embodiments, either the segment 135 forms an open circuit when the predefined level of current 210 is transmitted through the CNT 130, or the segment 135 remains unopened (i.e., closed) when the current 210 is less than the predefined level.
The value of the programmable electronic component 115 is thereby configured to equal a zero or nonzero current, depending on whether the segment 135 is opened or not opened, respectively.
In other embodiment, however, programmable electronic component 115 is configured as a capacitor. In such embodiments, the CNT 130 is capacitively coupled to a conductive body 150. E.g., the CNT 130 and conductive body 150 serve as capacitor plates and together have a capacitance. In some embodiments, the active device 110 has transistors 120 that are configured to control a current 210 transmitted to the capacitor (e.g., to the CNT 130 or the conductive body 150).
A distance 160 between the segment 135 and the conductive body 150 is configured to change as a function of the adjusted diameter 140, and the value is configured to be equal to a capacitance. The capacitance may have any number of discrete values that can be used by the programming circuit 105 to control other circuit components. E.g., when the programming circuit 105 has an output of a capacitance that is equal to some predefined value, the programming circuit can use the capacitance in a predetermined fashion to adjust (e.g., activate or deactivate) other circuit components in the device 100.
Because the capacitance is inversely proportional to the distance 160 (FIG. 1), a larger dynamic range of discrete capacitance values can be obtained by having a large range of possible adjusted diameters 140. E.g., consider an embodiment where prior to adjusting the segments diameter, the diameter 145 equals about 16 nm and the distance 160 between the conductive body 150 and the segment 135 equals about 30 nm. After adjusting the diameter 140 of the segment 135 from 16 nm to 3 nm, the distance 160 increases from about 30 nm to about 36.5 nm. Consequently, the capacitance between the conductive body 150 and the CNT 130 decreases by about 18 percent. In some embodiments the diameter 140 can range from about 30 (prior to adjustment) to 0 nm (after adjustment), thereby providing an even larger dynamic range of capacitance values.
By configuring the conductive body 150 as a second CNT, the dynamic range of capacitance values can be nearly doubled. Consider an embodiment where the conductive body 150 also comprises a second CNT and a diameter 165 of the conductive body 150 prior to its adjustment equals about 16 nm. After adjusting the diameters 140, 165 of both the CNT 130 and conductive body 150 from about 16 nm to 3 nm, the distance 160 increases from about 30 nm to about 43 nm. Consequently, the capacitance between the conductive body 150 and the CNT 130 decreases by about 30 percent.
It is non-intuitive to use a CNT as a capacitor plate because CNTs are generally cylindrically shaped. Cylindrically-shaped plates do not present as large surface area as a planar surface, and therefore a lesser amount charge can be stored, as compared to capacitor plates having a planar surface. As part of the present disclosure, it was realized that despite these shortcoming, CNTs can still be effectively employed to generate a capacitance value that is sufficient to be used by the programming circuit 105 to control other circuit components. Importantly, because the diameter 140 of the CNT's segment 135 can be adjusted, several different control signals can be generated by the programming circuit 105. This can be an advantage over a single fuse which is limited to producing a binary control signal (e.g., zero or nonzero current flowing through the fuse).
Using the same reference numbers to show device components analogous to that depicted in FIGS. 1-3, FIG. 4 shows a circuit diagram of an example semiconductor device 100, when the programmable electronic component 115 is configured as a capacitor. In such embodiments, the programming circuit 105 further includes an inverter 410, having an output 415, and a comparator 420. The programmable electronic component 115 is connected to the output 415 of the inverter 410. The inverter 410 can comprise transistors 120, such as pMOS and nMOS transistors.
The comparator 420 has a first input 425 comprising a reference signal 430 and a second input 435 comprising an output 440 of the programmable electronic component 115. In some cases the reference signal 430 comprises a voltage (e.g., a DC voltage), while in other cases the reference signal 430 comprises a clock signal (e.g., an AC voltage). A programming output 445 of the comparator 420 depends upon the value of the programmable electronic component 115, which as noted above, can have a number of different capacitances. E.g., the programming output 445 can equal a tripping time whose value increases as the capacitance increases. In turn, the capacitance increases as the adjusted distance 140 between the segment 135 and the conductive body 150 decreases.
One skilled in the art would understand that the circuit depicted in FIG. 4 is just one of many configurations of the programming circuit 105. In other embodiments, e.g., one or both of the inverter 410 and comparator 420 can be replaced with other types of circuitry configured to accomplish analogous functions. E.g., the inverter 410 can be replaced with switched resistors or other analog circuitry, and the comparator 420 can be replaced by other type of voltage measurement circuitry.
Some embodiments of the device 100 includes additional circuitry to facilitate a more accurate measurement of the capacitance of the programmable electronic component 115. E.g., in cases where the output 430 of the inverter 410 can vary from one device 100 to another, it is desirable to further include one or more calibration circuits 450, 452, which are coupled to the programming circuit 105 to thereby determine its output 440, e.g., determine the capacitance. In some cases, the calibration circuits 450, 452 can be part of the programming circuit 105, while in other cases the calibration circuits 450, 452 are separate from the programming circuit 105.
Each calibration circuit 450, 452 can respectively include a second inverter 460, 462, whose output is coupled to the input of a known capacitance 470, 472. An output of the known capacitance 470, 472 is coupled to a second comparator 480, 482. Preferably, each of the known capacitances 470, 472, equals one of a target discrete value that the programming circuit 115 is configured to have. By comparing the programming output 445 of the comparator 420 to the analogous output of the calibration circuits 450, 452 an accurate capacitance value can be determined.
Another embodiment of the disclosure is a method of manufacturing a semiconductor device. FIGS. 5 to 10 illustrate cross-section views, analogous to that shown in FIG. 2 or 3, of selected steps in example methods of manufacturing a device of the disclosure. The same reference numbers are used to depict analogous features as shown in FIGS. 1-3.
Manufacturing the device 100 includes fabricating a programming circuit 105, aspects of which are illustrated in FIGS. 5 to 10. FIG. 5 shows a cross-section view, analogous to that depicted in FIG. 2, of the device 100 after forming an active device 110 on or in the substrate 117. Forming the active device 110 can include forming one or more transistors 120. Forming the transistors 120 can include depositing and patterning dielectric and conductive layers to form a gate structure 122, depositing gate sidewalls 215, implanting and activating dopants to form source and drain structures 125, and a doped well 220 in the substrate 117, and forming insulating structures 127 (e.g., field oxide or shallow trench isolation structures) in or on the substrate 117.
FIG. 6 shows the device 100 depicted in FIG. 5, at an intermediate step in forming a programmable electronic component 115 of the device 100. FIG. 7 shows the device 100 at the same stage of fabrication, but from a view analogous to that shown in FIG. 3 (view line B-B in FIG. 1). Forming the programmable electronic component 115 includes depositing a CNT 130 on the substrate 117. E.g., a multiwalled CONT 130 can be synthesized using an arc-discharge or pyrolysis method. The CONT 130 can then be dispersed in an organic liquid (e.g., ortho-dichlorobenzene or isopropyl alcohol). The CNT-containing liquid can be deposited at discrete locations on the substrate 117, after which the liquid is removed (e.g., evaporated) leaving the CNT 130 on the substrate 117. For further examples of forming and depositing CNTs, see U.S. Patent Application 2003/0190278 to Wang and Zettl (“Wangt”), or U.S. Patent Application 2006/0228287 to Zettle et al. (“Zettle”), which are incorporated by reference in their entirety.
A variety of methods can be used to adjust a segment of the deposited CNT 130 such that its diameter is reduced and the programmable electronic component 115 is thereby configured to have a value. E.g., FIGS. 8 and 9 show different embodiments of the device 100 after adjusting a diameter 140 of a segment 135 of the CNT 130.
FIG. 8 shows a cross-sectional view of one embodiment of the device 100 depicted in FIG. 7, after adjusting the diameter 140 by opening the segment 135. Creating an opening 810 in the segment 130 includes transmitting a predefined current 210 (FIG. 2) through the CNT 130, such that at least a portion of the segment 135 corresponding to the opening 810 has a diameter 140 of zero. The programmable electronic component 115 thereby has a value that is equal to a zero current. In other cases, when the predefined current 210 is not transmitted through the CNT 130, and an opening 810 is not created, the programmable electronic component 115 has a value that is equal to a nonzero current.
FIG. 9 shows a cross-sectional view of another embodiment of the device 100 depicted in FIG. 7, after adjusting the diameter 140 by irradiating the segment 135 with an electron beam. In some embodiments, the electron beam has an energy ranging from about 1 to 100 keV. This energy range is conducive to shrinking certain embodiments of the CNT 130 while maintaining its tubular fullerene structure. In some cases, the electron beam can comprise the electron beam from a transmission electron microscope. In some embodiments, a potential (e.g., about 2 to 3 Volts) is applied to the segment 135 during the electron beam irradiation. Applying a potential the segment 135 can generate a current flow that is sufficient to thermally anneal structural damage and reshape the segment 135. For other examples of irradiating CNTs with electron beams, see Yuzvinsky et al., Nanoletters 6:2718-22, 2006, incorporated herein in its entirety.
The energy of the electron beam, the magnitude of the applied potential, and the durations of the electron beam irradiation and the optional simultaneously applied potential, can be individually adjusted to control the shrinkage of the CNT 130 to the desired adjusted diameter 140. Because the distance 160 between the segment 135 and the conductive body 150 (FIG. 1) depends upon the adjusted diameter 140, the programmable electronic component 115 can thereby be configured to have a value equal to any number of predefined capacitances. In some cases, the process to adjust the diameter can be configured to provide one of multiple discrete diameters 140 (e.g., 3, 8, 12 and 16 nm) so as to provide discrete target capacitance values.
In some embodiments, the segment 135 is irradiated with an electron beam to adjust its diameter 140 before transmitting the predefined current 210 (FIG. 2) such as described above in the context of FIG. 8. Shrinking the segment 135 so that it has a smaller adjusted diameter 140 than other portions of the CNT 130 helps to define where along the CONT 130 the opening 810 will be formed when the predefined current 210 is transmitted.
FIG. 10 shows the device 100 depicted in FIG. 6, at an intermediate step in forming the programmable electronic component 115 configured as a capacitor, that includes forming a conductive body 150 close to (e.g., within about 100 nm) of the CNT 130. The CNT 130 and the conductive body 150 can thereby establish a capacitance. In embodiments where the programmable electronic component 115 is configured as a fusible link, a conductive body need not be formed. In some cases, the conductive body 150 is formed by depositing a layer of conductive material (e.g., a polysilicon layer or metal layer deposited by chemical vapor or physical vapor deposition techniques) and then patterning the layer using conventional photolithography processes. In other cases, such as depicted in FIG. 10, the conductive body 150 is formed by depositing a second CNT on the substrate 117. E.g. the second CNT can be deposited in substantially the same fashion and at the same time as the CNT 130.
In such embodiments, the diameters of the CNT 130 and conductive body 150 can both be adjusted via irradiation with an electron beam, such as described in the context of FIG. 8. E.g., FIG. 11 shows the device 100 depicted in FIG. 10 after adjusting the diameters 140, 165 of the CNT 130 and conductive body 150 (configured as a second CNT) with electron beam irradiation.
FIG. 12 shows the device 100 depicted in FIG. 11, after depositing insulating layers 240 (e.g., silicon oxide, or low-k dielectric material deposited as PMD or ILD layers) and after forming interconnects 230 (e.g., tungsten contacts and copper vias and lines) in or on the insulating layers 240. The interconnects 230 are configured to complete the formation of the active device 110 (e.g., by interconnecting the transistors 120 of the active device 110) and to couple the active device 110 to the programmable electronic component 115.
Those skilled in the art to which the disclosure relates will appreciate that other and further additions, deletions, substitutions, and modifications may be made to the described example embodiments, without departing from the disclosure.

Claims

1. A semiconductor device, comprising:

a programming circuit, including:

an active device located on or in a substrate;

a programmable electronic component on said substrate, wherein said programmable electronic component includes at least one carbon nanotube having a segment with an adjusted diameter; and

interconnects that couple said active device to said programmable electronic component, wherein said active device is configured to control a current transmitted to said programmable electronic component and said programmable electronic component has a value that depends upon said adjusted diameter.

2. The device of claim 1, wherein said programmable electronic component is configured as a fusible link, wherein said segment is configured to open when said current, equal to a predefined level, is transmitted through said carbon nanotube, and said value is thereby configured to equal a zero or nonzero current depending on whether said segment is opened or not opened, respectively.

3. The device of claim 1, wherein said programmable electronic component is configured as a capacitor with said carbon nanotube being capacitively coupled to a conductive body such that a distance between said segment and said conductive body is configured to change as a function of said adjusted diameter, and said value is configured to be equal to a capacitance.

4. The device of claim 3, wherein said conductive body comprises a second carbon nanotube.

5. The device of claim 3, wherein said distance ranges from about 30 to 43 nm.

6. The device of claim 3, wherein said adjusted diameter ranges from about 3 to 30 nm.

7. The device of claim 3, wherein said programming circuit further includes:

an inverter and a comparator, wherein said programmable electronic component is connected to an output of said inverter; and said comparator has a first input comprising a reference signal and a second input comprising an output of said programmable electronic component, and wherein a programming output of said comparator depends upon said value.

8. The device of claim 7, wherein said reference signal comprises a voltage or a clock signal.

9. The device of claim 7, wherein said programming output of said comparator comprises a tripping time of said comparator, said tripping time depending upon said discrete capacitance.

10. The device of claim 7, further including one or more calibration circuits, wherein each of said calibration circuits includes a second inverter whose output is coupled to the input of a known capacitance, and an output of said known capacitance is coupled to a second comparator, wherein said one or more calibration circuits are coupled to said programming circuit to thereby determine said capacitance.

11. The device of claim 7, wherein said value is configured to trim an oscillator of said device, trim a voltage of said device, or to provide a unique identification code for said device.

12. A semiconductor device, comprising:

a programming circuit, including:

transistors located on or in a substrate;

a fusible link on said substrate that includes at least one carbon nanotube having a segment with an adjusted diameter; and

interconnects that couple said transistors to said programmable electronic component,

wherein said transistors are configured to control a current transmitted to said fusible link such that said segment is configured to open when said current, equal to a predefined level, is transmitted through said at least one carbon nanotube, and wherein said fusible link thereby has a value that depends upon said adjusted diameter, said value is being configured to equal a zero or nonzero current depending on whether said segment is opened or not opened, respectively.

13. A semiconductor device, comprising:

a programming circuit, including:

transistors located on or in a substrate;

a capacitor on said substrate that includes:

at least one carbon nanotube having a segment with an adjusted diameter; and

a conductive body capacitively coupled to said carbon nanotube, a distance between said segment and said conductive body is configured to change as a function of said adjusted diameter; and

it interconnects that couple said transistors to said capacitor,

wherein said transistors are configured to control a current transmitted to said capacitor, and said capacitor has a value that depends upon said adjusted diameter, said value configured to be equal to a capacitance.

14. A method of manufacturing a semiconductor device, comprising:

fabricating a programming circuit, including:

forming an active device on or in a substrate;

forming a programmable electronic component, including depositing a carbon nanotube on said substrate, wherein said carbon nanotube has a segment with an adjustable diameter, and said programmable electronic component has a value that depends upon said adjustable diameter; and

forming interconnects that couple said active device to said programmable electronic component.

15. The method of claim 14, further includes adjusting said diameter by opening said segment, including transmitting a predefined current through said carbon nanotube, thereby adjusting said value to a zero current.

16. The method of claim 14, further including adjusting said diameter by irradiating said segment with an electron beam, thereby adjusting said value to a predefined capacitance.

17. The method of claim 15, wherein said segment is irradiated with an electron beam to adjust said diameter before transmitting said predefined current.

18. The method of claim 16, further including applying a potential to said segment during said irradiating.

19. The method of claim 14, wherein forming said programmable electronic component includes forming a conductive body close to said carbon nanotube, thereby establishing a capacitance between said carbon nanotube and said conductive body.

20. The method of claim 19, wherein forming said conductive body includes depositing a conductive layer on said substrate and patterning said conductive layer.

21. The method of claim 19, wherein depositing said conductive body includes depositing a second carbon nanotube on said substrate.