RELATED APPLICATIONS This application is a Divisional of Ser. No. 10/230,937, filed Aug. 29, 2002, which is a Continuation of Ser. No. 09/031,398 filed on Feb. 26, 1998, now U.S. Pat. No. 6,449,757, which are incorporated herein by reference.
FIELD OF THE INVENTION This invention relates generally to the design of semiconductors, and more particularly to such design that is hierarchical in nature.
BACKGROUND OF THE INVENTION Semiconductor technology pervades most electronic devices today. Computers, televisions, videocassette recorders, cameras, etc., all use semiconductor integrated circuits to varying degrees. For example, the typical computer includes microprocessors and dedicated controller integrated circuits (i.e., video controllers, audio controllers, etc.), as well as memory, such as dynamic random-access memory. The design of semiconductors, therefore, is a crucial consideration of the design of almost any electronic device.
One type of semiconductor design is the design of semiconductor test structures. A semiconductor integrated circuit, for example, must be able to operate in a variety of different conditions (varying temperatures, for example), and perform within a variety of different specifications (i.e., speed, power consumption, etc.). Semiconductor test structures are therefore utilized to ensure that various components of a given semiconductor will perform according to specification in different conditions. Test structures are not integrated circuits sold to end consumers as part of an electronic device, but rather are used internally to ascertain that the end products will perform correctly.
To aid in the design of semiconductors in general, and the design of semiconductor test structures in particular, software such as Design Framework II (DF2), available from Cadence Design Systems, Inc., has been developed. DF2, for example, includes an editor that permits a designer to place various components over a semiconductor substrate as necessary. DF2 also provides for a degree of flexibility in the design of such components. Specifically, DF2 includes parameterized cells, or pcells, that allow the designer to create customized instances of a pcell every time the pcell is placed on a layer. For example, a transistor can be created and have parameters assigned thereto to provide for control of its width, length, and number of gates. When instances of the transistor are placed on the layer, different values may be assigned to each of these parameters. According to the parameter values, each instance varies in size and composition.
The pcell approach of DF2, however, is a top-down semiconductor design approach, and thus has limitations and disadvantages associated with it. A designer may, for example, first draw a transistor, and then program that transistor to respond to parameters that will cause various parts of the design to take on those parameter values. This can be a very complex, tedious and error-prone process. For example, if the designer desires contacts to fill in the available active area space while maintaining a certain pitch and minimum separation from the active area edge, the equations to accomplish this for an arbitrarily sized active area are complex within DF2. Furthermore, these equations are specific to the transistor under development. If the designer desires to design another parameterized object—for example, a field transistor or a contact chain—he or she needs to repeat the entire process.
Therefore, there is a need for an approach to the designing of semiconductors that avoids the pitfalls of top-down design. The approach should enable a semiconductor designer to avoid having to “start from scratch” when designing a new parameterized object. Thus, the approach should be more flexible and easier to use than prior art design approaches.
BRIEF DESCRIPTION OF THE DRAWINGS FIGS.1(a),1(b), and1(c) show diagrams of a representative hierarchy of a semiconductor test structure, according to an exemplary embodiment of the invention;
FIG. 2 shows a flowchart of a method according to an exemplary embodiment of the invention;
FIG. 3 shows a diagram of a computer in conjunction with which an exemplary embodiment of the invention may be implemented;
FIG. 4 shows a diagram of a semiconductor memory in conjunction with which a semiconductor test structure hierarchically designed in accordance with an embodiment of the invention may be tested;
FIG. 5 shows a diagram of the parameters contained within a basic atom cell, according to one embodiment of the invention amenable to implementation in conjunction with Design Framework II (DF2) software available from Cadence Systems, Inc.;
FIG. 6 shows a table of basic atom cells, according to one embodiment of the invention;
FIG. 7 shows a diagram of a master cell for use in accordance with one embodiment of the invention;
FIG. 8 shows a c9—2225678 higher-order cell, according to one embodiment of the invention;
FIG. 9 shows the c9—2225678 cell ofFIG. 8 after it has been converted into a VanDerPauw resistor, according to an embodiment of the invention;
FIG. 10 shows a table of higher-order cells, according to one embodiment of the invention; and,
FIG. 11 shows a table of devices and structures, according to one embodiment of the invention.
DETAILED DESCRIPTION OF THE INVENTION In the following detailed description of exemplary embodiments of the invention, reference is made to the accompanying drawings which form a part hereof, and in which is shown by way of illustration specific exemplary embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that logical, mechanical, electrical and other changes may be made without departing from the spirit or scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.
Those of ordinary skill within the art will appreciate that the detailed description is presented in accordance with the example of designing a semiconductor test structure. However, the invention itself is not limited to the design of semiconductor test structures. Rather, the invention may be utilized in the design of any semiconductor structure, in a manner identical to that described with respect to semiconductor test structures. The example of the semiconductor test structure is only presented in the detailed description specifically as an exemplary structure, to provide for clear description of the invention.
The detailed description is divided into three sections. In the first section, an exemplary embodiment of the invention is described. In the second section, a specific embodiment of the invention that may be practiced in conjunction with Design Framework II (DF2) software available from Cadence Design Systems, Inc., is presented. Finally, in the third section, a conclusion of the detailed description is provided.
Exemplary Embodiment of the Invention A description of an exemplary embodiment of the invention is provided in this section of the detailed description. The description is provided in conjunction with reference to FIGS.1(a),1(b),1(c),2,3 and4. FIGS.1(a)-1(c) show diagrams of a representative hierarchy of a semiconductor test structure, according to an exemplary embodiment.FIG. 2 shows a flowchart of a method according to an exemplary embodiment, whileFIG. 3 shows a diagram of a computer in conjunction with which an exemplary embodiment of the invention may be implemented. Finally,FIG. 4 shows a diagram of a semiconductor memory in conjunction with which a semiconductor test structure hierarchically designed in accordance with an embodiment of the invention may be tested.
Referring first toFIG. 1(a), a diagram of three higher-order cells, each defined by relating a number of instances of basic atom cells, is shown. Higher-order cell100, defining type “1” higher-order cells, is defined by relating four instances of basic atom cells,atoms102,104,106 and108.Atoms102 and108 are instances of basic atom cells of type “1”;atom104 is an instance of basic atom cell of type “2”; and,atom106 is an instance of basic atom cell of type “3”. The type defined by higher order cell100 (i.e., “1”), and the types ofatoms102,104,106 and108 (i.e., “1”, “2”, and “3”), are for representative purposes only, and do not specifically relate to any given type of semiconductor component. Thus, the types as used inFIG. 1(a) (and as will be used inFIG. 1(b) andFIG. 1(c) as well) are for notational and descriptive purposes only.
Furthermore, each ofhigher order cell100, andatoms102,104,106 and108 have a set of parameters related to its type. For example, the parameters may be related to placement, size, etc. (i.e., different attributes of the given cell). Desirably,higher order cell100 has parameters that when changed also change the parameters ofatoms102,104,106 and108 as necessary. Thus,higher order cell100 relatesatoms102,104,106 and108 to one another. Changing a parameter incell100 that causes that cell to become larger, for example, causes corresponding changes inatoms102,104,106 and108 that make up that instance ofcell100.
Still referring toFIG. 1(a), two other higher-order cells are defined,cells110 and112.Cell110 is made up ofatoms114,116 and118.Atoms114 and116 are of type “2”, andatom118 is of type “3”;cell110 itself defines type “2” for higher-order cells. Similarly,cell112 is made up ofatoms120,122,124,126 and128, whereatoms120 and122 are of type “2”,atom124 is of type “3”, andatoms126,128 and130 are of type “3”.Cell112 itself defines type “3” for high-order cells. As withcell100 and its constituent atoms,cells110 and112 and their constituent atoms each has a set of parameters related to its type. Desirably, when a parameter of eithercell110 or112 changes, one or more parameters of one or more of the associated constituent atoms also change.
The basic hierarchical structure shown inFIG. 1(a) is a powerful tool for the design of semiconductor test structures. For example, once higher-order cells100,110 and112 have been defined as is shown inFIG. 1(a), they may be utilized to create more complex devices and structures, without forcing the designer to concern him or herself over details regarding the individual constituent atoms of the higher-order cells. For example, the designer may wish to design a transistor.Atoms102 and104 may be the two basic atoms necessary in such a design; each exists independently and has significant programming therein.Cell100, then, may be a higher-level structure, where parameters fromatoms102 and104 are inherited up tocell100. A cell called tran, for transistor, is then created by placing an instance ofcell100 and setting the parameters ofcell100 such that a transistor is formed—the cell tran can then be used by anyone by setting its parameters. Transistors of different sizes and shapes can be created.
An additional level of the hierarchical structure initially described inFIG. 1(a) is shown inFIG. 1(b), which is a diagram of two devices, each defined by relating a number of instances of the higher-order cells that have been defined inFIG. 1(a).Device132, defining type “1” devices, is defined by relating three instances of higher-order cells,cells134,136 and138.Cells134 and138 are instances of cells of type “1,” as has been defined ascell100 ofFIG. 1(a);cell136 is an instance of cells of type “3,” as has been defined ascell112 ofFIG. 1(a). As withFIG. 1(a), the type defined bydevice132 is for representative purposes only, and does not specifically relate to any given type of semiconductor component.
Each ofdevice132 andcells134,136 and138 has a set of parameters related to its type. Desirably,device132 has parameters that when changed also change the parameters ofcells134,136 and138, which in turn change the parameters of the atoms making up these cells (not shown inFIG. 1(b)). That is, changing a parameter fordevice132 may change a parameter forcell136, which as a type “3” higher-order cell has six constituent atoms, as has been shown in and described in conjunction withFIG. 1(a). Thus, the changing of the parameter forcell136 instigated by changing a parameter fordevice132 also may change one or more parameters of one or more of these six constituent atoms.
Still referring toFIG. 1(b), one other device is defined,device140.Device140 is made up of two instances of higher-order cells,cells142 and144.Cell142 is of type “2,” as has been defined ascell110 ofFIG. 1(a), andcell144 is of type “3,” as has been defined ascell112 ofFIG. 1(a).Device140 itself defines type “2” for devices. As withdevice132 and its constituent higher-order cells,device140 and its constituent higher-order cells each has a set of parameters related to its type. Desirably, when a parameter ofdevice140 changes, one or more parameters of one or more of its constituent cells changes as well, propagating a change of one or more parameters of one or more of the atoms making up these constituent cells.
Therefore, the basic hierarchical structure shown inFIG. 1(a) is expanded by the structure shown inFIG. 1(b). InFIG. 1(b), two devices are defined. A designer of a semiconductor test structure may therefore utilize these devices within the test structure, such that the designer does not need to concern him or herself with the actual cells making up these devices, or the constituent atoms making up the higher-cells. The devices may thus be viewed as a higher abstraction than the higher-order cells, just as the higher-order cells are a higher abstraction than the basic atom cells. Changing the parameter of a device, for instance, may cause many changes in the parameters of the basic atom cells. Without the invention, there would be no lower cells at all; all the programming would be done at the highest level. By comparison, under the invention, the designer only needs to change the parameter of a device, and if the device and its higher-order cells are defined correctly, appropriate changes are propagated through to and made within the basic atom cells.
The hierarchical structure shown inFIG. 1(a) and extended inFIG. 1(b) may be additionally extended as shown inFIG. 1(c), which is a diagram of a semiconductor test structure, defined by relating three instances of the devices that have been defined inFIG. 1(b).Semiconductor test structure146 is defined by relating two instances of devices of type “1,”devices148 and152, as devices of type “1” have been defined asdevice132 ofFIG. 1(b), and one instance of devices of type “2,”device150, as devices of type “2” have been defined asdevice140 ofFIG. 1(b). The semiconductor test structure ofFIG. 1(c) is for representative purposes only, and does not specifically relate to any given type of semiconductor structure.
Each ofstructure146 anddevices148,150 and152 has a set of parameters related to its type. Desirably,structure146 has parameters that when changed also change the parameters ofdevices148,150 and152, which in turn change the parameters of the higher-order cells making up these devices (not shown inFIG. 1(c)), which in turn change the parameters of the atoms making up these cells (also not shown inFIG. 1(c)). That is, changing a parameter forstructure146 may change a parameter fordevice150, which as a type “2” device has two constituent higher-order cells, as has been shown in and described in conjunction withFIG. 1(b). Further, this change in a parameter fordevice150 may cause a change in one of the parameters of one of the two constituent higher-order cells, which may then cause a change in one of the parameters of one of the basic atom cells of this higher-order cell.
Thus, the semiconductor test structure ofFIG. 1(c) (as based on the structures of FIGS.1(a) and1(b)) may be viewed as being represented by a hierarchical data structure having four layers of abstraction: a highest layer of abstraction, the test structure itself; a second highest layer of abstraction, the devices making up the test structure; a third highest layer of abstraction, the higher-order cells making up the devices; and a lowest level of abstraction, the basic atom cells making up the higher-order cells. Changing the parameters of any one layer of abstraction causes the changing of the parameters of an immediately lower layer of abstraction, which then propagates changes down to the lowest level of abstraction. The FIGS.1(a),1(b) and1(c) may also be viewed as a computerized system, such that changing one aspect (parameter) of the system during the design of a test structure causes lower aspects of the system to automatically change. Note that other layers of abstraction can be formed on top of the four shown in and described in conjunction with FIGS.1(a),1(b), and1(c), such as circuits and integrated circuit chips.
The hierarchical design of semiconductor test structures as has been shown in and described in conjunction with FIGS.1(a),1(b) and1(c) provides for advantages not found in the prior art. By abstracting each layer within a semiconductor test structure, for example, different users can be responsible for different parts of the design, without having to be skilled in all aspects of the structure's design. For instance, one designer may be responsible for designing a library of basic atom cells, or a single very flexible basic atom cells. Another designer may be responsible for designing a library of higher-order cells based on the basic atom cell or cells. Still another designer may be responsible for designing devices based on the higher-order cells. A designer who is responsible for designing the structure itself can piece together a structure based on the devices and higher-order cells that have already been created. Finally, an end user may use this structure to create different instances thereof by simply changing the parameters of the structure in accordance with current specifications. This is advantageous, because this user does not have to be skilled in manipulation of the basic atom cells, since changing the parameters thereof will be accomplished automatically by changing parameters of the structure itself. Another advantage of the invention is that the cell designer can build in optimal design characteristics into the cell structure, and be guaranteed that those characteristics are retained in a specific instance of the cell placement by a cell user who may not be fully aware of the optimal design characteristics. In this way, the cells can incorporate and pass on a high level of design experience and avoid the possibility of design errors caused by inexperienced designers.
A hierarchical semiconductor test structure design according to an exemplary embodiment of the invention has been shown and described. Those of ordinary skill within the art will appreciate that the invention is not limited to the specific embodiment shown in and described in conjunction with FIGS.1(a)-1(c), however. For instance, there may be many more higher-order cells than the three defined inFIG. 1(a), each of which may have many more constituent basic atom cells than the number shown inFIG. 1(a). For further instance, there may be many more devices than the two defined inFIG. 1(b), each of which also may have many more constituent higher-order cells than the number shown inFIG. 1(b). Finally, the structure shown inFIG. 1(c) may have many more constituent devices than the number shown inFIG. 1(c).
Referring next toFIG. 2, a flowchart of a method according to an exemplary embodiment of the invention is shown. The method may be implemented as a computerized method executed as a computer program on a suitably equipped computer (in particular, executed by a processor of the computer from a computer-readable medium of the computer, such as a memory). Such a computer program may be stored on a computer-readable medium, such as a floppy disk, a compact-disc read-only-memory (CD-ROM), or a memory such as a random-access memory (RAM) or a read-only memory (ROM). The invention is not so limited. The method may be also be utilized to create a semiconductor test structure in accordance with an embodiment of the invention.
Instep200, a basic atom cell is created. This basic cell may be such as those described in conjunction withFIG. 1(a). The basic atom cell has at least one parameter that affects attributes thereof. The basic atom cell is the lowest abstraction within the hierarchical data structure for the semiconductor test structure.
Instep202, higher-order cells, such as those ofFIG. 1(a), are created. Each higher-order cell relates a plurality of instances of the basic atom cell. Desirably, higher-order cells also have parameters that affect attributes thereof. These parameters are such that when one of the parameters changes, one or more of the parameters of one or more of the plurality of instances of the basic atom cell related by the higher-order cell also change. In this way, the higher-order cells are a higher abstraction than the basic atom cell, and enable a designer to work with higher-order cells without having to specifically work with basic atom cells.
Instep204, devices, such as those ofFIG. 1(b), are created. Each device relates a plurality of instances of higher-order cells. Desirably, devices also have parameters that affect attributes thereof. These parameters are also such that when one of the parameters changes, one or more of the parameters of the one or more of the plurality of instances of the higher-order cells also change (and thus instigating change to basic atom cells as well). The devices are a higher abstraction than the higher-order cells, permitting a designer to work with devices without having to specifically work with higher-order cells or basic atom cells.
Finally, instep206, a test structure, such as that ofFIG. 1(c), is created. A test structure relates a plurality of instances of devices. Desirably, test structures also have parameters that affect attributes thereof. These parameters are such that when one of them changes, one or more of the parameters of the one or more of the plurality of instances of the devices also change (and thus instigating change to higher-order cells and basic atom cells as well). The structures are the highest abstraction, and permit a designer to work with a test structure without having to specifically work with devices, higher-order cells, or basic atom cells.
The method ofFIG. 2 thus provides for the design of a semiconductor test structure in a hierarchical manner. Each of the basic cells ofstep200 may be used in a number of different higher-order cells, which may be used in a number of different devices, which may be used in a number of different test structures. The hierarchical approach permits specialization as well: a designer may specifically only be skilled at creating one of the levels of abstraction, saving his or her work in a library such that the designer constructing the next layer of abstraction is able to utilize the immediately lower layer without having particular skill in construction of such lower layers.
Referring next toFIG. 3, a diagram of a computer in conjunction with which an exemplary embodiment of the invention may be implemented is shown. Those of ordinary skill within the art will recognize that the invention is not limited to the computer shown inFIG. 3, however. In one embodiment, the computer is running Design Framework II (DF2) software, available from Cadence Design Systems, Inc., and in conjunction with which an embodiment of the invention may be implemented.
Computer310 ofFIG. 3 is operatively coupled to monitor312, pointingdevice314, andkeyboard316.Computer310 includes a processor (such as an Intel Pentium processor or a reduced instruction set (RISC) processor), random-access memory (RAM), read-only memory (ROM), and one or more storage devices, such as a hard disk drive, a floppy disk drive (into which a floppy disk can be inserted), an optical disk drive, and a tape cartridge drive. The memory, hard drives, floppy disks, etc., are types of computer-readable media. The invention is not particularly limited to any type ofcomputer310.Computer310 desirably is a computer running a version of the UNIX operating system. The construction and operation of such computers are well known within the art.
Furthermore,computer310 may be communicatively connected to a local-area network (LAN), a wide-area network (WAN), an Intranet, or the Internet, any particular manner by which the invention is not limited to, and which is not shown inFIG. 3. Such connectivity is well known within the art. In one embodiment, the computer includes a modem and corresponding communication drivers to connect to the Internet via what is known in the art as a “dial-up connection.” In another embodiment, the computer includes an Ethernet or similar hardware card to connect to a local-area network (LAN) or wide-area network (WAN) that itself is connected to an Intranet or the Internet via what is know in the art as a “direct connection” (e.g., T1 line, etc.).
Monitor312 permits the display of information, including computer, video and other information, for viewing by a user of the computer. The invention is not limited to anyparticular monitor312, and monitor312 is one type of display device that may be used by the invention. Such monitors include cathode ray tube (CRT) displays, as well as flat panel displays such as liquid crystal displays (LCD's).Pointing device314 permits the control of the screen pointer provided by the graphical user interface of operating systems. The invention is not limited to anyparticular pointing device314. Such pointing devices include mouses, touch pads, trackballs, remote controls and point sticks. Finally,keyboard316 permits entry of textual information intocomputer310, as known within the art, and the invention is not limited to any particular type of keyboard.
Referring finally toFIG. 4, a diagram of a semiconductor memory in conjunction with which a semiconductor test structure hierarchically designed in accordance with an embodiment of the invention may be tested is shown. That is,FIG. 4 shows a semiconductor memory for which semiconductor test structures designed in accordance with the hierarchical manner of an embodiment of the invention may be utilized—the reason why semiconductor test structures are necessary is to ensure that semiconductor circuits such as the memory ofFIG. 4 correctly perform according to specification. However, as described in the beginning of this detailed description, the invention itself is not limited to the design of a semiconductor test structure; the invention may be used in conjunction with the design of any semiconductor structure. The design of a semiconductor test structure is merely an exemplary use, and is used specifically in the detailed description only as such.
FIG. 4 is specifically a schematic/block diagram illustrating generally an architecture of one embodiment of amemory400 in conjunction with which the present invention may be utilized. In the embodiment ofFIG. 4,memory400 is a dynamic random access memory (DRAM). However, the invention can be applied to other semiconductor memory devices, such as static random access memories (SRAMs), synchronous random access memories or other types of memories that include a matrix of selectively addressable memory cells. Furthermore, as has been described in the beginning of the detailed description, the invention can be applied to any type of semiconductor device, and is not limited to memory only.
Memory400 includes amemory cell array405, having memory cells therein that include floating gate transistors.X gate decoder415 provides a plurality of gate control lines for addressing floating gate transistors inarray405. Y source/drain decoder420 provides a plurality of source/drain interconnection lines for accessing source/drain regions of the floating gate transistors inarray405. Input/output circuitry425 includes necessary sense amplifiers and input/output (I/O) circuitry for reading, writing, and erasing data to and from array105. In response to address signals that are provided onaddress lines435 during read, write, and erase operations, the operation ofdecoders415 and420 are controlled. The address signals are provided by a controller such as a microprocessor that is fabricated separately or together withmemory400, or otherwise provided by any other suitable circuits.
The description of an exemplary embodiment of the invention has been provided. Specifically, in conjunction with FIGS.1(a)-1(c), a description of a hierarchical manner by which semiconductor test structures may be designed has been presented. In conjunction withFIG. 2, a description of a method according to which such structures may be designed in accordance with the invention has been provided. In conjunction withFIG. 3, a description of a computer in which embodiments of the invention has been presented. Finally, in conjunction withFIG. 4, a description of a semiconductor memory that may be the motivation for the hierarchical design of semiconductor test structures of the invention has also been provided.
Specific Embodiment of the Invention A description of an exemplary embodiment of the invention has been described in the previous section of the detailed description. In this section of the detailed description, a description of a specific embodiment of the invention is presented. Specifically, the description relates to an embodiment of the invention implemented using Design Framework II (DF2) software available from Cadence Design Systems, Inc. The description is provided in sufficient detail to enable one of ordinary skill in the art to make and use an embodiment of the invention utilizing DF2.
Referring first toFIG. 5, a diagram of the parameters contained with a basic atom cell, according to one embodiment of the invention amenable to implementation in conjunction with DF2, is shown. Basic atom cell500 is termed a C1 cell, based upon pcell functionality available within DF2. Pcell functionality provides for the taking of an existing geometry and parameterizing it so that when instances of that geometry are placed, parameters can be specified that will customize that instance to meet specific design rule requirements.
Cell500 includesunderlayer geometry502,contacts504 that can be arrayed into a contact block and aligned overgeometry502, metal caps506 that can be placed over eachindividual contact504, andmetal pad508 that can globally cover allcontacts504. Full programming control is provided for every possible relationship of the four layers (i.e.,geometry502,contacts504, caps506 and pad508) with respect to one another.Cell510 includesgrid parameter510 to ensure that all geometries and shifts within cell500 are accomplished in units of a grid. This ensures that cell500 is always consistent with an underlying grid structure.
Geometry502 has three associated parameters,1x512,1y514, andlayer516.Lx512 and1y514 are x and y values, respectively, that control the size ofgeometry502.Layer516 specifies the type of the base layer provided by geometry502 (such as nplsaa, npoly, etc., as known within the art). An “N” type forlayer516 turns off this base layer within cell500.
Contacts504 have thirteen associated parameters, cx518,cy520,cont522,cpx524,cpy526,cmx528,cmy530,nx532,ny534,ax536,ay538,cofx540 and cofy542.Cx518 andcy520 are x and y values, respectively, that control the size of the layer ofcontact504.Cont522 specifies the type of the base layer forcontacts504. An “N” type forcont522 turns off this layer within cell500.Cpx524 andcpy526 are x and y values, respectively, that specify the pitch of the contacts within the base layer.Cmx528 andcmy530 are x and y values, respectively, that specify the minimum allowed distance of the contact block to the base layer edge provided withincontacts504.
Nx532 andny534 are x and y values, respectively, that specify the number of contacts within the allowed area of the base layer provided bycont522. The allowed area is the region of the base layer that is defined by (1x minus two times cmx) and (1y minus two times cmy) in dimension. A value of nx and ny of 0.0 fills up zero percent of the allowed area in the base layer with contacts (i.e., no contacts). A value of 1.0 for nx and ny fills 100% of the allowed area in the base layer with contacts. A negative number for nx and ny forces the absolute value of that number of contacts to be placed. For example, an nx value of negative three and an ny value of negative five creates a contact block of three by five contacts independent of the allowed region of the base layer.
Ax536 anday538 are x and y values, respectively, that align the contact block within the allowed region of the base layer. An ax value of negative one pushes the contact block to the extreme left of the allowed region. An ax value of one pushes the contact block to the extreme right of the allowed region. An ax value of zero centers the contact block within the allowed based region. Similar behavior applies to ay538. Fractional values between negative one and positive one accords proportional behavior.
Whileax536 anday538 produce shifts in the contact block relative to the base layer,cofx540 and cofy542 are x and y values, respectively, that produce absolute shifts in addition to those produced byax536 and538. For example, if contacts are to be 0.1 micron off center in the x direction and centered exactly in the y direction, the parameters are to be set as follows: ax as zero, cofx as 0.1, ay as zero, and cofy as 0.
Thus, nx and ny calculate the size of the contact array, and ax and ay align that array over the base layer. If nx and ny are negative, then the absolute value of that number is the number of contacts; for example, if nx is minus eight and ny is minus five, then an eight by five array of contacts is aligned over the allowable base layer. If nx and ny are positive, then it can take values from zero through one. If nx is one, for example, then 100% of the allowable area is filled with contacts in the x direction. If nx is 0.5, then 50% of the allowable area is filled within contacts in the x direction. (The allowable area is 1x minus two times cmx and 1y minus two times cmy.)
Once nx and ny have been used to determine the size of the contact array,ax536 anday538 are used to align that array over the allowable area. If ax is one, then the contacts are pushed to the extreme right of the allowable area. If ax is zero, then the contacts are centered in the allowable area. If ax is minus one, then the contacts are pushed to the extreme left of the allowable area.Ay538 behaves similarly in the y direction.
Ax and ay shift contacts based upon a percentage of the available space. The contact offset parameters,cofx540 and cofy542, allow the contacts to be shifted by a fixed amount from the default positions given byax536 anday538. For example, if contacts are to be shifted 0.1 micron to the right of center, ax is set to zero and cofx is set to 0.1.
Metal caps506 has five associated parameters,cap544,csx546,csy548,csofx550, andcsofy552.Cap544 specifies the type of the cap layer provided bymetal caps506. A value of “N” turns off the cap layer.Csx546 andcsy548 are x and y values, respectively, that specify the surround ofcaps506 with respect tocontacts504. Values of zero for csofx and csofy center the caps about the contacts. Any other values cause offsets of the cap layer by the specified amount.
Finally,metal pad508 has eight associated parameters,pad554,psx556,psy558,padrel560,apx562, apy564,psofx566 and psofy568.Pad554 specifies the layer type of the pad layer provided bypad508. A value of “N” turns off the pad layer.Psx556 andpsy558 are x and y values, respectively, that control the size of the pad layer that globally surrounds the contact block. The effect of psy and psx depends on the setting ofpadrel560.
Padrel560 is a boolean parameter determining the effect ofpsx556 andpsy558. If padrel is set to “Y,” then the pad layer covers the contact block by values of (csx plus psy) and (csy plus psy) in the x and y directions, respectively (that is, the pad is placed relative to the contact block). If padrel is set to “N,” then the size of the pad is provided by psx and psy independent of the size of the contact block and other parameters.
Apx562 andapy564 are x and y values, respectively, that align the pad with respect to the contact block, and behaves in a similar fashion to ax536 anday536 that have already been described.Psofx566 and psofy568 are x and y values, respectively, that offset the pads by the given amount from the alignment that results from the values of psx, psy, apx, apy, and padrel.
As has been described, C1 cell500 is a parameterized cell that has four layers: underlayer geometry (or base layer)502,contacts layer504,metal caps layer506, andmetal pad layer508. The variables (parameters) within the C1 cell allow any size rectangular base layer to be created. Contacts of any size can be put into this base layer. Caps can be placed over these contacts with any cap overlap contact dimension in the x and y direction. The number of contacts that are placed within the base layer can be specified directly (e.g., nx as minus eight, ny as minus thirteen), or can be input as a percentage of the allowable area that can hold contacts.
This allowable area is determined by subtracting two times cmx and two times cmy from the x and y dimensions of the base layer, respectively. For example, if 1x and 1y are 100 and 100 (specifying size of the base layer), and cmx and cmy are 20 and 30, then the allowable area for contacts is 60 in the x direction and 40 in the y direction. Contacts fill up this area based upon the contact size and contact pitch that is specified. Once the number of allowable contacts are placed, then the contacts can be shifted as a group anywhere within the allowable area. The caps over the contacts, and the metal pads, are completely programmable in terms of size as well as offsets in the x and the y directions.
The layer parameters,layer516,cont522,cap544, andpad554, are used to determine the layers that are used in cell500. For example, the base layer specified bylayer516 can be changed to an allowable layer. If an “N” is input in either layer, cont, cap or pad, then those layers will not be placed. That is, if cap or cont is “N” then no contact or cap layer will be present regardless of the values any variables related to those layers may have.
The basic atom cell described and shown as cell500 ofFIG. 5 is termed a C1 cell. It is the most general cell, allowing full control in the x and y directions of cont, cap and pad. The contacts align to the base layer, the caps align to the contacts, and the pad aligns to the contacts. Other basic atom cells derived from the C1 cell are also desirable, to allow for easier creation of higher order-cells, and subsequently devices and structures.FIG. 6 shows a table of such other basic atom cells (table600) according to one embodiment of the invention. Each of these other basic atom cells are derived from the C1 cell, or from another cell within the table. Those of ordinary skill within the art can appreciate that the invention can be used to design other different types of basic cells with specialized properties and features that can be used to create higher-order complex objects with a minimum of programming effort.
Referring next toFIG. 7, a diagram of a master cell for use in accordance with one embodiment of the invention is shown.Master cell700 is completely programmable to produce any desired subcell by easily eliminating undesired basic atom cells frommaster cell700. Thus, as shown inFIG. 7,master cell700 includes nine C1 cells, such asC1 cell702, nine 11 cells, such as 11cell704, and nine c1a cells, such asc1a cell706. Associated with the cells, as known to those of ordinary skill in the art of DF2 software, are a number of horizontal and vertical stretch lines, such asstretch lines708 and710, which adjust the positioning of cells such as702,704 and706. Using a master cell provides for quicker generation of higher-order cells, devices and structures because it is generally much quicker to delete elements from the master cell than it is to create them from a blank slate.
The parameters of the basic atom cells provide great flexibility in producing a base layer with contacts, caps, and pad. Any orthogonal parametric structure should be able to be decomposed into an array of C1 cells, for instance, with different relative orientations to one another. For example, a two-terminal resistor can be thought of as a C1 cell on the left with layer, contacts, and pad turned on and caps off; a C1 cell in the middle with contacts, cap, and pad turned off; and a C1 cell on the right with layer, contacts, pad turned on and caps turned off.
This two-terminal resistor can be viewed as including three C1 cells withcell2 oriented tocell1 andcell3 oriented tocell2. This is referred to as ac3—2 structure. It may be an end test structure in and of itself, or it may be a higher-order cell structure for use in other more complex devices and structures. Another type of test structure may be built from three C1 cells where bothcell2 andcell3 align tocell1. This is referred to as ac3—1 structure.
A VanDerPauw resistor, known within the art, can be built from a c9-2225678:cell1 forms the body of the resistor,cells2,3,4 and5 form the arms that align tocell1, and cells6,7,8, and9 form the pads on the arms that align tocells2,3,4 and5, respectively.FIG. 8 shows such a c9—2225678 higher-order cell,cell800, whileFIG. 9 shows such a c9—2225678 cell after it has been converted into a VanDerPauw resistor,resistor900. Higher-order cells such as c9—2225678 may be referred to as elements, and can themselves by manipulated into a vast array of parametric devices and structures.
A set of parameters referred to as shift parameters are used in higher-order cells, devices and structures to determine the relative orientation of the lower-level abstractions, such as C1 cells, with respect to one another. For example, one set of shift parameters may be shift 32x, a32x, o32x (with a similar set existing for y directions). With these parameters, the alignment ofcell3 with respect tocell2 can be controlled. For example, setting shift 32x to 2, a32x to 0, and o32x to 0centers cell3 with respect tocell2. Setting o32x to 0.1offsets cell3 by 0.1 micron from the center ofcell2. The a parameters take on values from minus one to plus one, and behave similar to the ax and ay parameters that shift the contacts within the C1 cell itself, i.e., it produces a relative shift about an axis. The shift32x parameter determines the axis about which shifting occurs. For example, shift32x set to one shiftscell3 about the left edge ofcell2; shift 32x set to twoshifts cell3 about the center ofcell2, etc. The o32x parameter provides offsets from the shift and a parameters.
Thus, a collection of higher-order cells (or elements) may be created to assist in development of even higher levels of abstractions, such as structures and devices. The invention is not particularly limited to any set of higher-order cells. However, a table of higher-order cells according to one embodiment of the invention is shown inFIG. 10. WithinFIG. 10, table1000 includes two columns:column1002, which lists the higher-order cells, andcolumn1004, which lists the basic atom cells that constitute these higher-order cells, and/or a description of the higher-order cells.
The creation of such higher-order cells is accomplished by relating together two or more basic atom cells, and attaching appropriate parameters thereto. For example, the c2 cell includes two C1 cells. Each C1 cell has its own set of parameters such as 11x, 12x, 11y, 12y, etc. In addition, there are a set of parameters that determine how the second C1 cell is aligned to the first. That is, the combination of the set of parameters determines how the second C1 cell is aligned to the first C1 cell in the x and y directions.
Specifically, there are six parameters: shift21x, shift21y, a21x, a21y, o21x, and o21y. Shift21x determines the type of shift thatlayer2 does with respect tolayer1 in the x direction. Shift21y determines the type of shift thatlayer2 does with respect tolayer1 in the y direction. A21x determines the percentage amount of shift in the x direction, while a21y determines the percentage amount of shift in the y direction. O21x determines the absolute shift in the x direction after a21x has been applied, while o21y determines the absolute shift in the y direction after a21y has been applied. The a21x and a21y parameters have possible values ranging from minus one to plus one.
For further example, a c3 cell has three C1 cells that align to one another.Cell2 aligns only tocell1 butcell3 aligns to eithercell1 or tocell2. Thus, there are two types of c3 cells:c3—1 andc3—2.C3—1 hascell3 aligning tocell1 andc3—2 hascell3 aligning tocell2. The c1 and c2 cells require no extensions. The c3 cell has local parameters such as 11x, 12x and 13x, and also global parameters to determine the relative shifts of the c1, c3 and c3 cells. Thus, inc3—2 there are parameters shift 32x, shift 32y, a32x, a32y, o32x, o32y, shift21x, shift21y, a21x, a21y, o21x and o21y. Inc3—1 there are the parameters shift 31x, shift 31y, a31x, a31y, o31x, o31y, shift21x, shift21y, a21x, a21y, o21x and o21y. As a final example, a c4 cell has four C1 cells that align to one another. There are four variations, namelyc4—11, c4—12,c4—22 and c4—23.
Once a core library of higher-order cell has been created, as has been shown inFIG. 10, a library of devices and structures can then also be created. All devices and structures eventually lead back to basic atom cells. A table of exemplary devices and structures, and their descriptions, according to one embodiment of the invention is shown as table1100 inFIG. 11. Devices are desirably built up from higher-order cells, and structures are desirably built up from devices. Further layers of abstraction are also possible, such as modules, built up from structures, and integrated circuit chips, built up from modules.
Thus, once structures have been created, higher levels of abstraction, such as the module, can also be created. A module has programmable pads, with structures in-between the pads. The size of the structures and their orientation are connected to the location, size and orientation of the interconnection pads. Higher level parameters may be used to configure the entire module, controlling all aspects of the module from pad size and pitch to inner-structure details such as transistor nibble, gate length, etc.
A description of a specific embodiment of the invention, for implementation in conjunction with DF2 software, has been described. Those of ordinary skill within the art will appreciate that while the invention has been described in relation to DF2 software, the invention is not so limited. Thus, an embodiment of the invention utilizing other software, or programmed from scratch, is within the scope of the invention.
CONCLUSION The above-mentioned shortcomings, disadvantages and problems are addressed by various embodiments of the present invention, which will be understood by reading and studying the specification. One aspect of the invention is a computerized system that includes a semiconductor structure and a basic atom. The system also includes a hierarchy of abstractions ordered from highest to lowest. Each abstraction relates a plurality of instances of an immediately lower abstraction; the highest abstraction corresponds to the structure, and the lowest abstraction corresponds to the basic atom. A plurality of sets of parameters also is included within the system, where each set of parameters corresponds to an instance of an abstraction. Changing one of the set of parameters for an instance of an abstraction changes at least one of the set of parameters for an instance of an immediately lower abstraction. Parameters desirably relate to attributes of an abstraction.
For example, in one embodiment, the hierarchy may have six abstractions: atoms, higher-order cells, devices, structures, and also circuits and integrated circuit chips, ordered from lowest to highest. Each of these abstractions has an associated set of parameters. Instances of atoms are used to create higher-order cells, instances of higher-order cells are used to create devices, and instances of devices are used to create structures. Each instance of an abstraction relates together a plurality of instances of an immediately lower-level abstraction. Thus, changing parameters associated with an instance of a higher-order cell, for example, automatically changes the parameters of the instances of atoms related by that higher-order cell.
In this manner, once appropriate atoms and higher-order cells have been designed, devices and structures can be designed easily by relating together instances of the atoms and higher-order cells. Most importantly, if the specifications governing a given structure need to be changed, a user merely has to change the parameters for the structure, which then affects the parameters of the instances of the lower level devices, higher-order cells, and atoms. That is, redesign of the structure at the atom, or even at the higher-order cell, level is not necessary. This means that semiconductor design becomes more intuitive, and enables modification of existing structures to create new structures, in a non-tedious and non-time-consuming manner.
The present invention includes computerized systems, methods, hierarchical data structures, semiconductor structures, computer-readable media, basic atom cells, and computers of varying scope. In one embodiment of the invention, the invention is implemented in conjunction with Design Framework II (DF2) software available from Cadence Design Systems, Inc. In addition to the aspects and advantages of the present invention described in this summary, further aspects and advantages of the invention will become apparent by reference to the drawings and by reading the detailed description.
Hierarchical semiconductor structure design has been described. Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that any arrangement which is calculated to achieve the same purpose may be substituted for the specific embodiments shown. This application is intended to cover any adaptations or variations of the present invention. Therefore, it is manifestly intended that this invention be limited only by the following claims and equivalents thereof.