US20030075789A1 - Semiconductor storage device having memory chips in a stacked structure - Google Patents

Semiconductor storage device having memory chips in a stacked structure Download PDF

Info

Publication number
US20030075789A1
US20030075789A1 US10/197,901 US19790102A US2003075789A1 US 20030075789 A1 US20030075789 A1 US 20030075789A1 US 19790102 A US19790102 A US 19790102A US 2003075789 A1 US2003075789 A1 US 2003075789A1
Authority
US
United States
Prior art keywords
memory
semiconductor storage
bit units
storage device
memory chips
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US10/197,901
Inventor
Masayasu Kawamura
Atsushi Nakamura
Yoshihiro Sakaguchi
Yoshitaka Kinoshita
Yasushi Takahashi
Yoshihiko Inoue
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to US10/197,901 priority Critical patent/US20030075789A1/en
Publication of US20030075789A1 publication Critical patent/US20030075789A1/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L25/00—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof
    • H01L25/03—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes
    • H01L25/04—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes the devices not having separate containers
    • H01L25/065—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes the devices not having separate containers the devices being of a type provided for in group H01L27/00
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L23/00—Details of semiconductor or other solid state devices
    • H01L23/48—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor
    • H01L23/488—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor consisting of soldered or bonded constructions
    • H01L23/495—Lead-frames or other flat leads
    • H01L23/49537—Plurality of lead frames mounted in one device
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C5/00—Details of stores covered by group G11C11/00
    • G11C5/02—Disposition of storage elements, e.g. in the form of a matrix array
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C5/00—Details of stores covered by group G11C11/00
    • G11C5/02—Disposition of storage elements, e.g. in the form of a matrix array
    • G11C5/025—Geometric lay-out considerations of storage- and peripheral-blocks in a semiconductor storage device
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C5/00—Details of stores covered by group G11C11/00
    • G11C5/02—Disposition of storage elements, e.g. in the form of a matrix array
    • G11C5/04—Supports for storage elements, e.g. memory modules; Mounting or fixing of storage elements on such supports
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C5/00—Details of stores covered by group G11C11/00
    • G11C5/06—Arrangements for interconnecting storage elements electrically, e.g. by wiring
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C8/00—Arrangements for selecting an address in a digital store
    • G11C8/12—Group selection circuits, e.g. for memory block selection, chip selection, array selection
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L23/00—Details of semiconductor or other solid state devices
    • H01L23/48—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor
    • H01L23/488—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor consisting of soldered or bonded constructions
    • H01L23/495—Lead-frames or other flat leads
    • H01L23/49503—Lead-frames or other flat leads characterised by the die pad
    • H01L23/4951—Chip-on-leads or leads-on-chip techniques, i.e. inner lead fingers being used as die pad
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L23/00—Details of semiconductor or other solid state devices
    • H01L23/48—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor
    • H01L23/488—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor consisting of soldered or bonded constructions
    • H01L23/495—Lead-frames or other flat leads
    • H01L23/49575—Assemblies of semiconductor devices on lead frames
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L23/00—Details of semiconductor or other solid state devices
    • H01L23/48—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor
    • H01L23/50—Arrangements for conducting electric current to or from the solid state body in operation, e.g. leads, terminal arrangements ; Selection of materials therefor for integrated circuit devices, e.g. power bus, number of leads
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L24/00—Arrangements for connecting or disconnecting semiconductor or solid-state bodies; Methods or apparatus related thereto
    • H01L24/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L24/02—Bonding areas ; Manufacturing methods related thereto
    • H01L24/04—Structure, shape, material or disposition of the bonding areas prior to the connecting process
    • H01L24/06—Structure, shape, material or disposition of the bonding areas prior to the connecting process of a plurality of bonding areas
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L25/00—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof
    • H01L25/03—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes
    • H01L25/10—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes the devices having separate containers
    • H01L25/105—Assemblies consisting of a plurality of individual semiconductor or other solid state devices ; Multistep manufacturing processes thereof all the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/00, or in a single subclass of H10K, H10N, e.g. assemblies of rectifier diodes the devices having separate containers the devices being of a type provided for in group H01L27/00
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/02—Bonding areas; Manufacturing methods related thereto
    • H01L2224/04—Structure, shape, material or disposition of the bonding areas prior to the connecting process
    • H01L2224/04042—Bonding areas specifically adapted for wire connectors, e.g. wirebond pads
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/02—Bonding areas; Manufacturing methods related thereto
    • H01L2224/04—Structure, shape, material or disposition of the bonding areas prior to the connecting process
    • H01L2224/05—Structure, shape, material or disposition of the bonding areas prior to the connecting process of an individual bonding area
    • H01L2224/0554—External layer
    • H01L2224/0555—Shape
    • H01L2224/05552—Shape in top view
    • H01L2224/05554—Shape in top view being square
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/02—Bonding areas; Manufacturing methods related thereto
    • H01L2224/04—Structure, shape, material or disposition of the bonding areas prior to the connecting process
    • H01L2224/05—Structure, shape, material or disposition of the bonding areas prior to the connecting process of an individual bonding area
    • H01L2224/0554—External layer
    • H01L2224/05599—Material
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/02—Bonding areas; Manufacturing methods related thereto
    • H01L2224/04—Structure, shape, material or disposition of the bonding areas prior to the connecting process
    • H01L2224/06—Structure, shape, material or disposition of the bonding areas prior to the connecting process of a plurality of bonding areas
    • H01L2224/061—Disposition
    • H01L2224/0612—Layout
    • H01L2224/0613—Square or rectangular array
    • H01L2224/06134—Square or rectangular array covering only portions of the surface to be connected
    • H01L2224/06136—Covering only the central area of the surface to be connected, i.e. central arrangements
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/26—Layer connectors, e.g. plate connectors, solder or adhesive layers; Manufacturing methods related thereto
    • H01L2224/31—Structure, shape, material or disposition of the layer connectors after the connecting process
    • H01L2224/32—Structure, shape, material or disposition of the layer connectors after the connecting process of an individual layer connector
    • H01L2224/321—Disposition
    • H01L2224/32151—Disposition the layer connector connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive
    • H01L2224/32221—Disposition the layer connector connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked
    • H01L2224/32245—Disposition the layer connector connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked the item being metallic
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/42—Wire connectors; Manufacturing methods related thereto
    • H01L2224/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L2224/48—Structure, shape, material or disposition of the wire connectors after the connecting process of an individual wire connector
    • H01L2224/4805—Shape
    • H01L2224/4809—Loop shape
    • H01L2224/48091—Arched
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/42—Wire connectors; Manufacturing methods related thereto
    • H01L2224/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L2224/48—Structure, shape, material or disposition of the wire connectors after the connecting process of an individual wire connector
    • H01L2224/481—Disposition
    • H01L2224/48151—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive
    • H01L2224/48221—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked
    • H01L2224/48245—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked the item being metallic
    • H01L2224/48247—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked the item being metallic connecting the wire to a bond pad of the item
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/42—Wire connectors; Manufacturing methods related thereto
    • H01L2224/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L2224/48—Structure, shape, material or disposition of the wire connectors after the connecting process of an individual wire connector
    • H01L2224/481—Disposition
    • H01L2224/48151—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive
    • H01L2224/48221—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked
    • H01L2224/48245—Connecting between a semiconductor or solid-state body and an item not being a semiconductor or solid-state body, e.g. chip-to-substrate, chip-to-passive the body and the item being stacked the item being metallic
    • H01L2224/4826—Connecting between the body and an opposite side of the item with respect to the body
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L2224/42—Wire connectors; Manufacturing methods related thereto
    • H01L2224/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L2224/49—Structure, shape, material or disposition of the wire connectors after the connecting process of a plurality of wire connectors
    • H01L2224/491—Disposition
    • H01L2224/4912—Layout
    • H01L2224/49171—Fan-out arrangements
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/73—Means for bonding being of different types provided for in two or more of groups H01L2224/10, H01L2224/18, H01L2224/26, H01L2224/34, H01L2224/42, H01L2224/50, H01L2224/63, H01L2224/71
    • H01L2224/732—Location after the connecting process
    • H01L2224/73201—Location after the connecting process on the same surface
    • H01L2224/73215—Layer and wire connectors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2224/00—Indexing scheme for arrangements for connecting or disconnecting semiconductor or solid-state bodies and methods related thereto as covered by H01L24/00
    • H01L2224/80—Methods for connecting semiconductor or other solid state bodies using means for bonding being attached to, or being formed on, the surface to be connected
    • H01L2224/85—Methods for connecting semiconductor or other solid state bodies using means for bonding being attached to, or being formed on, the surface to be connected using a wire connector
    • H01L2224/8538—Bonding interfaces outside the semiconductor or solid-state body
    • H01L2224/85399—Material
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2225/00—Details relating to assemblies covered by the group H01L25/00 but not provided for in its subgroups
    • H01L2225/03—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00
    • H01L2225/10—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers
    • H01L2225/1005—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2225/00—Details relating to assemblies covered by the group H01L25/00 but not provided for in its subgroups
    • H01L2225/03—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00
    • H01L2225/10—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers
    • H01L2225/1005—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00
    • H01L2225/1011—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00 the containers being in a stacked arrangement
    • H01L2225/1017—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00 the containers being in a stacked arrangement the lowermost container comprising a device support
    • H01L2225/1029—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00 the containers being in a stacked arrangement the lowermost container comprising a device support the support being a lead frame
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2225/00—Details relating to assemblies covered by the group H01L25/00 but not provided for in its subgroups
    • H01L2225/03—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00
    • H01L2225/10—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers
    • H01L2225/1005—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00
    • H01L2225/1011—All the devices being of a type provided for in the same subgroup of groups H01L27/00 - H01L33/648 and H10K99/00 the devices having separate containers the devices being of a type provided for in group H01L27/00 the containers being in a stacked arrangement
    • H01L2225/1047—Details of electrical connections between containers
    • H01L2225/107—Indirect electrical connections, e.g. via an interposer, a flexible substrate, using TAB
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L24/00—Arrangements for connecting or disconnecting semiconductor or solid-state bodies; Methods or apparatus related thereto
    • H01L24/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L24/42—Wire connectors; Manufacturing methods related thereto
    • H01L24/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L24/48—Structure, shape, material or disposition of the wire connectors after the connecting process of an individual wire connector
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L24/00—Arrangements for connecting or disconnecting semiconductor or solid-state bodies; Methods or apparatus related thereto
    • H01L24/01—Means for bonding being attached to, or being formed on, the surface to be connected, e.g. chip-to-package, die-attach, "first-level" interconnects; Manufacturing methods related thereto
    • H01L24/42—Wire connectors; Manufacturing methods related thereto
    • H01L24/47—Structure, shape, material or disposition of the wire connectors after the connecting process
    • H01L24/49—Structure, shape, material or disposition of the wire connectors after the connecting process of a plurality of wire connectors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/0001—Technical content checked by a classifier
    • H01L2924/00014—Technical content checked by a classifier the subject-matter covered by the group, the symbol of which is combined with the symbol of this group, being disclosed without further technical details
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01004—Beryllium [Be]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01005—Boron [B]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01006—Carbon [C]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01014—Silicon [Si]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01019—Potassium [K]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01023—Vanadium [V]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01033—Arsenic [As]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01039—Yttrium [Y]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01041—Niobium [Nb]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01042—Molybdenum [Mo]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01055—Cesium [Cs]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01059—Praseodymium [Pr]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01075—Rhenium [Re]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01076—Osmium [Os]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01082—Lead [Pb]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/01—Chemical elements
    • H01L2924/01087—Francium [Fr]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/013—Alloys
    • H01L2924/014—Solder alloys
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/10—Details of semiconductor or other solid state devices to be connected
    • H01L2924/11—Device type
    • H01L2924/13—Discrete devices, e.g. 3 terminal devices
    • H01L2924/1304—Transistor
    • H01L2924/1306—Field-effect transistor [FET]
    • H01L2924/13091—Metal-Oxide-Semiconductor Field-Effect Transistor [MOSFET]
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/10—Details of semiconductor or other solid state devices to be connected
    • H01L2924/11—Device type
    • H01L2924/14—Integrated circuits
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/15—Details of package parts other than the semiconductor or other solid state devices to be connected
    • H01L2924/181—Encapsulation
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/19—Details of hybrid assemblies other than the semiconductor or other solid state devices to be connected
    • H01L2924/1901—Structure
    • H01L2924/1904—Component type
    • H01L2924/19041—Component type being a capacitor
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L2924/00—Indexing scheme for arrangements or methods for connecting or disconnecting semiconductor or solid-state bodies as covered by H01L24/00
    • H01L2924/30—Technical effects
    • H01L2924/301—Electrical effects
    • H01L2924/3011—Impedance

Definitions

  • This invention relates to a semiconductor storage device and a memory module, and to techniques which are effective when utilized chiefly in a universal semiconductor storage device such as dynamic RAM (Random Access Memory), and a memory module employing the semiconductor storage device.
  • a universal semiconductor storage device such as dynamic RAM (Random Access Memory)
  • a memory module employing the semiconductor storage device.
  • the thickness (in the direction of the height) of the semiconductor integrated circuit device viewed as one semiconductor storage device is made equal to or less than the thickness of a universal semiconductor storage device having an equal storage capacity.
  • the resulting device merely sealed becomes a unique semiconductor storage device, the uses of which are limited.
  • the semiconductor integrated circuit device in the aforementioned official gazette, the memory chips are held in a stacked structure so that the front surfaces (main surfaces) thereof may be opposite to each other, and a height for wire bonding which executes connections with leads is required over each of the front surfaces of the two chips, so that the package thickens inevitably.
  • each of semiconductor storage devices is often used in the way that memory chips are mounted on both surfaces of a mounting circuit board, thereby to construct a memory module, and that a plurality of such memory modules are plugged into a plurality of connectors which are juxtaposed on a mother board.
  • the intervals of the individual connectors are set as near as possible. Therefore, the memory modules need to be formed by matching the overall thickness with, at least, the pitches of the connectors. Considering the radiation of heat during a memory operation, however, the memory modules should desirably be spaced from one another even if slightly.
  • the inventors of the present application have found it very important that, when two memory chips are mounted in a single package so as to realize a double storage capacity, the thickness of such a package is made equal to or less than the thickness of an existing semiconductor storage device in which a single memory chip is mounted.
  • An object of this invention is to provide a semiconductor storage device which can be formed thin while employing two memory chips. Another object of this invention is to provide a semiconductor storage device which has an excellent universality while employing two memory chips. Still another object of this invention is to provide a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging. A further object of this invention is to provide a memory module which is well adapted for use.
  • a typical one of inventions disclosed in the present application is briefly summarized as follows: Two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units.
  • a memory module is so constructed that a plurality of semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, are mounted on a mounting circuit board which is square and formed with electrodes (connectors) along one latus thereof.
  • FIG. 1 is a schematic perspective view of essential portions showing an embodiment of a semiconductor storage device according to this invention
  • FIGS. 2A and 2B schematic pattern diagram showing an embodiment of two memory chips which are made into the stacked structure of the semiconductor storage device according to this invention
  • FIG. 3 is a top plan view showing an embodiment of the semiconductor storage device according to this invention.
  • FIG. 4 is an output system diagram showing an embodiment of the semiconductor storage device according to this invention.
  • FIG. 5 is a sectional view of essential portions showing an embodiment of a memory module which employs semiconductor storage devices according to this invention
  • FIG. 6 is a front surface view showing an embodiment of the memory module according to this invention.
  • FIG. 7 is a back surface view showing an embodiment of the memory module according to this invention.
  • FIG. 8 is a block diagram showing an embodiment of the memory module shown in FIGS. 6 and 7;
  • FIG. 9 is a schematic constructional view for explaining a memory system which employs memory modules according to this invention.
  • FIG. 10 is a sectional view of essential portions showing another embodiment of the memory module employing semiconductor storage devices according to this invention.
  • FIG. 11 is a sectional view of essential portions showing another embodiment of the memory module employing semiconductor storage devices according to this invention.
  • FIG. 12 is a schematic layout diagram showing an embodiment of a dynamic RAM to which this invention is applied.
  • FIG. 13 is a circuit diagram showing an embodiment of a dynamic RAM according to this invention.
  • FIG. 14 is a whole block diagram showing an embodiment of a synchronous DRAM to which this invention is applied;
  • FIG. 15 is a sectional view of essential portions showing a further embodiment of the memory module employing semiconductor storage devices according to this invention.
  • FIG. 16 is a plan view for explaining the relationship between one of the memory chips and a lead frame in the semiconductor storage device according to this invention.
  • FIG. 1 Shown in FIG. 1 is a perspective view of the essential portions of one embodiment of a semiconductor storage device according to this invention.
  • two memory chips are held in a structure in which they are stacked back to back within an identical package. That is, the memory chips are placed one over the other so that the back sides of these memory chips may be opposite to each other, and that the front surfaces thereof, on each of which bonding pads, etc. are formed, may face outwards.
  • some relevant to data terminals are representatively exemplified in order to facilitate understanding the invention.
  • a method for easily enlarging a semiconductor storage device is a technique in which two memory chips are mounted in a single package as stated in the official gazette mentioned before.
  • 128 Mbits have been realizable when two 64-Mbit chips, in each of which ⁇ 4, ⁇ 8 and ⁇ 16 bits are existent, are stacked and assembled into one package.
  • 512 Mbits have been realizable when two 256-Mbit chips are assembled into one package.
  • Such a stacking scheme has come into the limelight for inexpensive uses because it can shorten the development period than in the new development of a single chip of 128 Mbits or 512 Mbits.
  • the whole package thickens more as compared with the package of a semiconductor storage device which has a storage capacity equal to the total storage capacity of the two memory chips.
  • the semiconductor storage device having the two memory chips becomes incompatible with a semiconductor storage device in which one memory chip is mounted in one package.
  • two memory chips each of which can be set into a ⁇ 2-bit configuration are made into a stacked structure, thereby to construct a semiconductor storage device of ⁇ 4-bit configuration.
  • the stacked structure is so contrived as to join the back surfaces of the two memory chips, in order that the total thickness of the memory chips to be assembled into the single package stated above may be equalized to the thickness of the package in which one chip is mounted.
  • the back surfaces of the memory chips are prevented to the utmost from coming into touch with a sealing resin which constitutes the single package.
  • the invention attains the thinning of the semiconductor storage device while adopting the stacked structure.
  • the memory chips are disposed back to back as described above, whereby the sealing resin at the back surface parts as is contained in the prior-art semiconductor integrated circuit device can be decreased to thin the package to that extent.
  • the thickness of the package in this embodiment can be equalized to that of the package of the existing semiconductor storage device having the equal storage capacity, even when the two memory chips are accommodated in the single package by arranging leads of LOC (Lead On Chip) structure on both the upper and lower surfaces of the memory chip assembly as shown in the figure, and connecting these leads and the bonding pads of the memory chips by pieces of bonding wire.
  • LOC Lead On Chip
  • address terminals and control terminals which are disposed back to back in the respective memory chips need to have the corresponding leads connected in common outside.
  • the relations between the leads and the bonding pads are set to be inverse laterally.
  • data items each being of ⁇ 2 bits collide at an external terminal.
  • the address terminals etc. are held in mirror inversion between in the upper memory chip and lower memory chip of the stacked structure, whereupon the leads and the bonding pads are connected.
  • the two data terminals are connected with the two leads on the same side as viewed from the main surface of each memory chip. More specifically, in the illustrated example, each of leads for data formed as external terminals DQ 0 and DQ 1 extends toward both the upper memory chip and the lower memory chip, and the upper leads are respectively connected to the pads DQ(a) and DQ(c) of the upper memory chip through the wire pieces, while the lower leads are not connected to any of the pads of the lower memory chip.
  • Each of leads for data formed as external terminals DQ 3 and DQ 2 extends toward both the upper memory chip and the lower memory chip, and the lower leads are respectively connected to the pads DQ(a) and DQ(c) of the lower memory chip through the wire pieces, while the upper leads are not connected to any of the pads of the lower memory chip.
  • the pad arrangement of the lower memory chip and that of the upper memory chip are similar, and pads DQ(b) and DQ(d) representatively shown on the upper memory chip are used together with the above pads DQ(a) and DQ(c) in the case where data are outputted in 4-bit units from one of the memory chips.
  • the pads DQ(b) and DQ(d) are not used in the embodiment of FIG. 1.
  • the lead of A-signal is connected to pads A-PAD and B-PAD which are formed on the respective memory chips at the distal ends of both its leads extending toward the upper memory chip and the lower memory chip, so that the identical A-signal is supplied in common to both the memory chips.
  • the leads on the side on which they are not connected may be cut away in order to lower input/output capacitances as will be explained later.
  • FIGS. 2A and 2B are schematic pattern diagrams of one embodiment of the two memory chips which are made into the stacked structure of the semiconductor storage device according to this invention.
  • FIG. 2A shows the memory chip on the upper side (UPPER) and the leads and bonding wire pieces corresponding thereto
  • FIG. 2B shows the memory chip on the lower side (LOWER) and the leads and bonding wire pieces corresponding thereto.
  • the ⁇ 2-bit configuration, the ⁇ 4-bit configuration, the ⁇ 8-bit configuration and the ⁇ 16-bit configuration can be selected by a metal option, a bonding option or the combination thereof as will be explained later.
  • the leads are shown in the maximum number so that they can cope with a plurality of these sorts of bit configurations.
  • This embodiment exemplifies the leads and the bonding wire pieces in the case where each memory chip has a storage capacity of 256 Mbits in the ⁇ 2-bit configuration. It is accordingly to be understood that the leads with no signal names affixed thereto are the ones which are not existent in the above memory of the ⁇ 2-bit configuration.
  • the memory chip has the bonding pads arranged substantially rectilinearly, substantially on the center line thereof in the longitudinal direction thereof. Such a construction in which the bonding pads are aligned, becomes a very useful array of the bonding pads in the case to be explained later where the two memory chips are made into the stacked structure back to back as stated above, where the leads such as address terminals and control terminals are made common outside, and where the resulting structure is mounted in the single package.
  • Those lead terminals of the upper memory chip and lower memory chip which have signal names are arranged in lateral symmetry on the drawing, except for the leads for data, DQ 0 through DQ 3 enclosed with rectangles.
  • the respective leads arranged on the right side of the upper memory chip UPPER agree with the leads which are arranged on the left side of the lower memory chip LOWER, and which are obtained by the mirror inversion with respect to the center line of the memory chip in the longitudinal direction thereof.
  • the respective leads arranged on the left side of the upper memory chip UPPER agree with the leads which are arranged on the right side of the lower memory chip LOWER, and which are obtained by the mirror inversion with respect to the center line of the memory chip in the longitudinal direction thereof.
  • those leads of the upper memory chip UPPER and lower memory chip LOWER to which the same signals are fed are arranged so as to be laterally inverse.
  • the bonding wire pieces may be arranged for the leads so as to be laterally inverse between the upper side and the lower side.
  • the leads and bonding wire pieces of the memory chip LOWER are arrayed in agreement with those of the memory chip UPPER reflected in a mirror which is put on the right side of the memory chip UPPER.
  • the bonding wire pieces are connected to the leads DQ 3 and DQ 2 on the right side in the upper memory chip UPPER, while the bonding wire pieces are connected to the leads DQ 0 and DQ 1 on the right side in the lower memory chip LOWER. That is, the bonding pads of the memory chips and the leads to be connected thereto lie at the same positions, but the lead names are made different as stated above.
  • a pair of leads are extended in the longitudinal direction of each memory chip and are respectively connected to leads for the voltage VDD of a power supply and the ground potential VSS of circuitry at the upper and lower ends of the chip.
  • the pair of leads are formed as bus bars, and are respectively used for appropriately feeding the power supply voltage VDD and circuital ground potential VSS to the memory chip through low power supply impedances.
  • the bus bars VDD and VSS are also arranged in the mirror inversion form between in the upper memory chip UPPER and in the lower memory chip LOWER.
  • the address terminals and the control terminals have the respectively corresponding leads placed one over the other vertically and connected in common.
  • each of the data terminals has the two leads laterally separated from each other and electrically isolated from each other in the stacked structure, whereby the data terminals DQ 0 through DQ 3 of ⁇ 4 bits can be formed.
  • FIG. 3 Shown in FIG. 3 is a top plan view of one embodiment of the semiconductor storage device according to this invention.
  • the memory chips each being of 256 Mbits are made into the stacked structure as explained above and are mounted in the single package, so as to have the storage capacity of 512 Mbits in total.
  • This embodiment illustrates pin arrays in the cases where the two memory chips are set into the ⁇ 2-, ⁇ 4 - and ⁇ 8-bit configurations, thereby to realize the three sorts of bit configurations of ⁇ 4 bits, ⁇ 8 bits and ⁇ 16 bits, respectively.
  • a power supply terminal VCC corresponds to the power supply leads VDD mentioned before, and power supply terminals VCCQ and VSSQ are used for feeding the voltage VCC of a power supply and the ground potential VSS of circuitry therefrom without being connected to the bus bars.
  • the power supply voltages VCCQ and VSSQ serve to feed an operating voltage for an output buffer.
  • power supply noise having developed in the output buffer is prevented from propagating to any other internal circuit through the bus bars inside the semiconductor storage device. It is consequently possible to stabilize the operations of internal circuits.
  • the two memory chips each being of 256 Mbits are mounted back to back in the standard package of 54 pins.
  • the package has a thickness specified as a standard.
  • FIG. 4 Shown in FIG. 4 is an output system diagram of one embodiment of the semiconductor storage device according to this invention.
  • this embodiment illustrates an example which employs memory chips each having a storage capacity of 64 Mbits, and it shows the relationships between memory array positions and DQ Nos. in a normal mode.
  • one can be selected from among the four sorts of bit configurations of a ⁇ 2-bit product, a ⁇ 4-bit product, a ⁇ 8-bit product and a ⁇ 16-bit product.
  • a memory array corresponding to one memory bank is divided into 16 subarrays, two pairs of main input/output lines (MIO) are laid from each of the subarrays, and 32 main amplifiers are disposed in correspondence with the lines MIO.
  • the 16 subarrays are subdivided into left and right (L/R) groups each consisting of 8 subarrays.
  • the left and right halves of the memory array correspond to the data terminals DQ 0 and DQ 1 as indicated by the symbols L and R, respectively.
  • a selection signal YS is generated in accordance with address signals A 0 through A 7 , and signals are delivered from one memory bank to the 32 pairs of MIO (main input/output lines).
  • the 32 pairs of MIO are separated into the two sets of the left L and right R sets each consisting of 16 pairs.
  • buffer circuits-BC disposed in the number of 32 in correspondence with the main amplifiers MA are separated into two sets which correspond to the sets L and R, and which are respectively connected in common to data terminals DQ 3 (0) and DQ 15 (1).
  • the output lines of the buffer circuits CB are separated into four sets.
  • the address signal A 11 for selecting the main amplifiers MA is correspondingly invalidated (degenerated), whereby data are read out in 4-bit units.
  • the output lines of the buffer circuits CB are separated into eight sets, and the address signals A 11 and A 10 for selecting the main amplifiers MA are invalidated (degenerated), whereby data are read out in 8-bit units.
  • the output lines of the buffer circuits CB are separated into 16 sets, and the address signals A 11 , A 10 and A 9 for selecting the main amplifiers MA are invalidated (degenerated), whereby data are read out in 8-bit units.
  • FIG. 5 Shown in FIG. 5 is a sectional view of the essential portions of one embodiment of a memory module employing semiconductor storage devices according to this invention.
  • the semiconductor storage devices each of which has memory chips held back to back, are mounted on both surfaces of a module substrate.
  • Leads are bonded onto the memory chips through adhesive layers.
  • Bus bars which extend orthogonally to the extending direction of the leads, have adhesive layers therefor formed thin so as to be smaller in height than the leads.
  • a margin in a height direction can be enlarged in order to prevent the bus bars from being touched by pieces of wire which connect the selected ones of the leads and bonding pads provided on the front surfaces of the memory chips.
  • Each of the semiconductor storage devices mounted on both surfaces of the module substrate is constructed as the stacked structure of the two memory chips as explained before.
  • the thickness of a sealing resin can be set small. That is, the semiconductor storage device of this embodiment can be assembled into a package having a thickness equal to that of the ordinary semiconductor storage device in which only one memory chip is mounted and which has a storage capacity being half of that of the device of this embodiment, and it can be replaced with a memory module which employs the ordinary semiconductor storage device. By such replacement with the memory module, the storage capacity can be doubled for the same packaging volume, and a packaging area can be reduced to half for the same storage capacity.
  • the semiconductor storage device is sealed in the state where the back surfaces of the first and second memory chips (that is, the upper and lower memory chips) are placed one over the other. It includes the first lead DQ 0 and second lead DQ 1 which extend from the first latus of the sealed structure to the inside and outside this sealed structure, and the third lead DQ 2 and fourth lead DQ 3 which extend from the second latus of the sealed structure opposite to the first latus to the inside and outside this sealed structure.
  • Each of the first and second memory chips includes the first terminal DQ(a), second terminal DQ(b), third terminal DQ(c) and fourth terminal DQ(d) for outputting data, which are arranged along a straight line being parallel to the longer latera of each memory chip and passing through the central parts of the shorter latera thereof, on the circuit forming surface of each memory chip.
  • the second terminal DQ(b) is arranged at a position nearer to the first terminal DQ(a) between this first terminal DQ(a) and the third terminal DQ(c), while the third terminal DQ(c) is arranged at a position nearer to the fourth terminal DQ(d) between the second terminal DQ(b) and this fourth terminal DQ(d).
  • the first lead DQ 0 and fourth lead DQ 3 are arranged at positions which oppose with the straight line held therebetween, while the second lead DQ 1 and third lead DQ 2 are arranged at positions which oppose with the straight line held therebetween.
  • Each of the first and second memory chips outputs data in 2-bit units by employing either of the first and second terminals, for example, the first terminal DQ(a), and either of the third and fourth terminals, for example, the third terminal DQ(c), whereby data are outputted from the first, second, third and fourth leads in 4-bit units in total.
  • terminals corresponding to the second terminal DQ(b) and the fourth terminal DQ(d) are not depicted in the lower memory chip, and they have been omitted from illustration in order to avoid the complicacy of the drawing.
  • the upper chip and the lower chip have similar constructions, and the respective terminals exist in correspondence.
  • the data outputting in 2-bit units and the data outputting in 4-bit units can be selected for the first and second memory chips.
  • the first terminal DQ(a), second terminal DQ(b), third terminal DQ(c) and fourth terminal DQ(d) are employed as explained before.
  • the positional relations of these terminals as stated above are determined in accordance with the normalized positions of the corresponding leads. That is, the individual terminals are arranged in such positional relations that the wire pieces can be shortened when the leads and terminals corresponding to each other are to be connected by these wire pieces.
  • first terminal DQ(a) and second terminal DQ(b) of each memory chip are arranged near the first lead DQ 0 and fourth lead DQ 3
  • third terminal DQ(c) and fourth terminal DQ(d) of each memory chip are arranged near the second lead DQ 1 and third lead DQ 2 .
  • each memory chip can employ any two of the four data terminals in any desired combination.
  • the first terminal DQ(a) and third terminal DQ(c) of the first memory chip can be easily connected to the first lead DQ 0 and second lead DQ 1 extending near these terminals, by using the wire pieces, respectively.
  • the first terminal DQ(a) and third terminal DQ(c) of the first memory chip can be easily connected to the fourth lead DQ 3 and third lead DQ 2 extending near these terminals, by using the wire pieces, respectively.
  • the leads for data which extend toward the memory chips but which are not connected, are shortened as in an embodiment illustrated in FIG. 15, whereby the parasitic capacitances of the leads can be lowered.
  • the leads for data not to be connected are cut at the end parts of the memory chips. It is also possible, however, to omit such non-connected data leads themselves.
  • FIG. 16 Shown in FIG. 16 is a plan view for explaining the relationship between the memory chip and a lead frame. The memory chip and leads illustrated are those of one of the two memory chips. As indicated by a lead enclosed with a broken line, each of the leads to be cut among a plurality of leads is formed in the shape of the alphabetic letter T facing toward the memory chip, so as not to fall off the sealing resin.
  • the first lead DQ 0 should preferably be cut at or ahead of a position where it bifurcates toward the upper memory chip and the lower memory chip.
  • leads having shapes after the cutting may be formed at the beginning of manufacture. That is, the asymmetric construction of each data lead concerning the lengths thereof extending toward the first and second memory chips can improve the signal transmission speed on the basis of the lowering of the parasitic capacitance, over the symmetric construction concerning the extending lengths.
  • FIG. 6 Shown in FIG. 6 is a front surface view of one embodiment of the memory module according to this invention, and shown in FIG. 7 is a back surface view of this memory module.
  • Each of the semiconductor storage devices numbering 36 in total mounted on both surfaces of the memory module substrate is constructed as the stacked structure of the two memory chips as shown in FIG. 1 or FIG. 5.
  • a register, a semiconductor integrated circuit device for signal conversion, capacitors and various resistors for a stabilized power supply, etc. are mounted between the semiconductor storage devices which are arrayed in the shape of a matrix such as of 9 ⁇ 2 and connectors (connecting electrodes) which are disposed on one end side of the memory module in the longitudinal direction thereof.
  • FIG. 8 Shown in FIG. 8 is a block diagram showing one embodiment of the memory module in FIGS. 6 and 7.
  • the memory module of this embodiment is such that a plurality of semiconductor storage devices, in each of which memory chips of ⁇ 2 bits are made into a stacked structure so as to equivalently form a ⁇ 4-bit configuration, are mounted on a module substrate, thereby to construct a memory system of 64 bits. That is, a memory system is constructed in which the semiconductor storage devices are disposed on both sides with a module substrate held therebetween and which has memory banks (BANK) “0” and “1”.
  • BANK memory banks
  • the memory banks “0” and “1” correspond to the nine semiconductor storage devices shown in FIGS. 6 and 7, respectively. More specifically, the nine semiconductor storage devices which take charge of the data terminals DQ 0 through DQ 31 and parity bits CB 0 through CB 3 of the memory banks “0” and “1” correspond to the 9 ⁇ 2 devices shown in FIG. 6, while the nine semiconductor storage devices which take charge of the data terminals DQ 32 through DQ 64 and parity bits CB 4 through CB 7 of the memory banks “0” and “1” correspond to the 9 ⁇ 2 devices shown in FIG. 7. In this manner, the memory module is constructed into a 64-bit configuration having the data terminals DQ 0 through DQ 63 .
  • the eight bits; CB 0 through CB 7 are provided as the parity bits. Therefore, in a mother board on which the memory module according to this embodiment is mounted, data buses to be connected to sockets into which the connectors of the memory module are plugged are brought into a 72-bit configuration.
  • FIG. 9 Shown in FIG. 9 is a schematic constructional view for explaining a memory system which employs the memory module according to this invention.
  • the figure illustrates a situation where connectors (connecting electrodes) which are juxtaposed along one latus of each of the mounting circuit boards constituting a plurality of memory modules are arrayed in a state in which they are plugged in sockets provided on a mother board not shown.
  • a plurality of memory modules need to be arrayed at narrow intervals in correspondence with the sockets, and it is important for this purpose that semiconductor storage devices to be mounted in the memory modules are formed thin.
  • This embodiment indicates an example in which memory chip SDRAMs are aligned on a module substrate, unlike in the memory module of the embodiment shown in FIGS. 6 and 7.
  • FIG. 10 Shown in FIG. 10 is a sectional view of the essential portions of another embodiment of the memory module employing semiconductor storage devices according to this invention.
  • the stacked structure in order to make two memory chips into a stacked structure and to diminish the thickness of a sealing resin, the stacked structure is formed so that the back surfaces of the memory chips may face outwards, reversely to the directions thereof in the embodiment of FIG. 5.
  • the back surface of each memory chip is not formed with any molding resin for the purpose of sealing.
  • This does not preclude a thin sealing-resin layer from being formed on a part or all of the back surface of the memory chip in the course of the manufacture of the semiconductor storage device.
  • the sealing resin In bestowing the function of sealing, the sealing resin must not be formed thick even on the back surfaces as in the prior art.
  • the back surfaces of the memory chips it is not indispensable that the back surfaces of the memory chips must be held in completely exposed states in order to thin the semiconductor storage device.
  • the sealing resin is not precluded from being formed thin even on the back surfaces in the course of the fabrication of the stacked structure, and the back surfaces may be coated with an appropriate insulating coating material or the like.
  • the back surface of each memory chip may be formed with an oxide film so as to afford an insulating property. It incurs no problem in the semiconductor storage device itself to leave the back surface of a semiconductor substrate exposed as it is. However, when the semiconductor storage device has been installed in or is assembled into an electron device, an undesirable electrical contact might occur. It is therefore desirable to make the back surface electrically insulated as stated above.
  • the semiconductor storage device of this embodiment has the stacked structure in the aspect in which the back surfaces of the memory chips face outwards and in which the front sides formed with circuit elements are opposite to each other. Even when the directions of the memory chips are respectively reversed to those in the embodiment of FIG. 5 in this manner, the semiconductor storage device can be, in effect, thinned as in the embodiment of FIG. 5 for the reason that the sealing resin on the back surfaces can be dispensed with.
  • FIG. 11 Shown in FIG. 11 is a sectional view of the essential portions of another embodiment of the semiconductor storage device according to this invention.
  • leads are printed on a tape (a flexible substrate), the tape is stuck on the front surface of each memory chip, and the leads are connected to bonding pads by wire bumps.
  • a sealing resin is deposited extending from the front surface of each memory chip to the side surfaces thereof, and the back surface of each memory chip is substantially exposed without being sealed with a resin, in the sense as explained in conjunction with FIG. 10, whereby the whole semiconductor storage device is formed thin. Therefore, two such semiconductor storage devices are made into a stacked structure on a module substrate as they are held independent.
  • the ⁇ 2 configuration is added in the memory chip, thereby to realize the bit configurations of ⁇ 4 bits etc. in the stacked products.
  • the ⁇ 2, ⁇ 4, ⁇ 8 and ⁇ 16 configurations are all set at bond options, whereby the chips fabricated by an identical wafer pre-process are permitted to expand merely by changing the setup of some pads at the stage of assemblage. Owing to such bond options, the memory chips to be formed by the wafer pre-process can be mass-produced, and reduction in cost can be promoted.
  • Table 1 exemplifies the relationships between pad inputs for making the bonding options and functions in the case where the semiconductor storage device according to the invention of the present application has a storage capacity of, e.g., about 64 Mbits and where it is applied to EDO/FP products.
  • the four sorts of bit configurations are switched using two pads BOP0B and BOP1B among four option pads, the EDO and first page (FP) are switched using one pad BOP2B, and 4K refresh (4KR) and 8K refresh (8KR) are switched using one remaining pad BOP3B. (TABLE 2)
  • Table 2 shows the relationships of the roles of DQ pins.
  • the memory module includes the DQ pins in the number of 16.
  • the pins which are not used increase in number in the order of the ⁇ 8, ⁇ 4 and ⁇ 2 configurations, and they may be made “open” in these cases.
  • the pins DQ 1 , DQ 3 , DQ 13 and DQ 15 are employed in the ⁇ 4 configuration, and are also employed for the input/output pins of a parallel test which uses four DQ pins.
  • the pin DQ 15 is used in all cases. That is, it is used in both the ⁇ 2-bit configuration and an EIAJ test.
  • the two BOP0B, BOP1B and, if necessary, the two BOP2B, BOP3B may be connected to external terminals.
  • the external terminals have been mounted in the memory module, voltages as indicated in Table 1 are fed to these external terminals, whereby the bit configurations can be set as in the foregoing.
  • a user mounts the external terminals in the memory module, he/she can select the bit configuration of the data terminals.
  • FIG. 12 Shown in FIG. 12 is a schematic layout diagram of one embodiment of a dynamic RAM to which this invention is applied. Individual circuit blocks in the figure are formed on a single semiconductor substrate, such as of single-crystal silicon, by known techniques for manufacturing semiconductor integrated circuits. Individual circuits in the figure are depicted substantially in agreement with the geometrical arrangement thereof on the semiconductor substrate. In this embodiment, memory arrays are separated into four as a whole, thereby to construct memory banks “0” through “3”.
  • the memory banks “0” through “3” correspond to the memory arrays which are separated in two vertically and in two laterally in the longitudinal direction of a semiconductor chip.
  • a peripheral circuit PERI which includes an address input circuit, a data input/output circuit and a bonding pad array, is disposed at the central part of the chip extending in the longitudinal direction thereof.
  • an example of the peripheral circuit PERI includes a step-up voltage generator circuit and a control circuit for controlling the operation thereof, a voltage divider circuit in the case of an interface of SSTL as divides an external power supply voltage VDDQ into 1 ⁇ 2 and thus forms the reference voltage of an input circuit constructed of a differential circuit, an input/output circuit and a clock control circuit therefor, a Y predecoder and a read/write buffer, a step-down circuit for forming the operating voltage of a peripheral circuit, a VPP sensor for detecting whether or not a voltage VPP has a desired value, an X address latch circuit, a Y clock circuit, a mode decoder/clock buffer and a command circuit, a Y counter and a control circuit therefor, a refresh control circuit, a bonding option circuit, a power supply turn-ON detector circuit, etc.
  • an X-coordinate predecoder circuit ROWPDC and remedy circuit ROWRED and a Y-coordinate predecoder circuit COLPDC and remedy circuit COLRED are collectively arranged at the lateral middle part of the semiconductor chip in the longitudinal direction thereof. That is, two sets each consisting of the X-coordinate predecoder circuit ROWPDC and remedy circuit ROWRED and the Y-coordinate predecoder circuit COLPDC and remedy circuit COLRED are distributed in correspondence with the two left and right memory arrays among the four memory arrays.
  • Main word driver areas MWD are formed at the intermediate part of the memory arrays likewise to the above, and they drive main word lines laid extending downwards and upwards in correspondence with the corresponding memory arrays, respectively.
  • the main word lines are extended so as to penetrate through 16 subarrays.
  • Y decoders YDC are disposed on the peripheral sides of the chip remote from the central part thereof in the memory arrays. That is, the memory arrays separated in four are so arranged as to be held between main amplifiers MA located on the central side and the Y decoders YDC located on the peripheral sides, respectively.
  • the four memory arrays correspond to the four memory banks.
  • one memory array corresponding to the memory bank has the 16 subarrays in the direction of the word lines, and the two pairs of main input/output lines from each of the subarrays are extended in the direction of bit lines between the subarrays, thereby to be led to the input terminals of the 32 main amplifiers MA as stated before.
  • the bit configurations of ⁇ 2, ⁇ 4, ⁇ 8 and ⁇ 16 as explained before can be switched.
  • each memory array is divided into a plurality of subarrays 15 .
  • Each of such subarrays 15 is so formed as to be surrounded with sense amplifier areas 16 and sub word driver areas 17 which are respectively located holding the subarray 15 therebetween.
  • the intersection parts between the sense amplifier areas 16 and the sub word driver areas 17 are determined as cross areas 18 .
  • Sense amplifiers to be provided in the sense amplifier areas 16 are constructed in conformity with a shared sense scheme. Except for the sense amplifiers which are arranged at both ends of each memory cell array, complementary bit lines are laid at the right and left of the sense amplifiers and are selectively connected to either of the right and left complementary bit lines of the memory cell array.
  • Each of the subarrays 15 includes sub word lines in the number of, for example, 256 and 256 pairs of complementary bit lines (or data lines) being orthogonal to the sub word lines, though not shown.
  • the subarray is also provided with spare word lines and spare complementary bit lines in order to remedy defective word lines and defective bit lines, respectively.
  • the subarrays are disposed in the number of 16 in the direction of arraying the word lines, so that the sub word lines are laid for about 4 kbits as a whole, while the subarrays are disposed in the number of 16 in the direction of arraying the bit lines, so that the complementary bit lines are laid for about 4 kbits as a whole.
  • each of the subarrays 15 may well include sub word lines in the number of, for example, 512 and 512 pairs of complementary bit lines (or data lines) being orthogonal to the sub word lines, though not shown.
  • the subarray is also provided with spare word lines and spare complementary bit lines in order to remedy defective word lines and defective bit lines, respectively.
  • the subarrays are disposed in the number of 16 in the direction of arraying the word lines, so that the sub word lines are laid for about 8 kbits as a whole, while the subarrays are disposed in the number of 16 in the direction of arraying the bit lines, so that the complementary bit lines are laid for about 8 kbits as a whole.
  • the lengths of the complementary bit lines are divided into ⁇ fraction (1/16) ⁇ in correspondence with the 16 subarrays in any of the above cases.
  • the lengths of the sub word lines are divided into ⁇ fraction (1/16) ⁇ in correspondence with the 16 subarrays.
  • the sub word driver (sub word line driving circuit) 17 is provided every divided subarray 15 of each of the memory arrays.
  • the sub word driver 17 forms a selection signal for the sub word line which is divided to the length of ⁇ fraction (1/16) ⁇ relative to the main word line as stated above, and which is extended in parallel with the main word line.
  • the sub word lines in the number of four are laid in the direction of the complementary bit lines in correspondence with one main word line though not especially restricted.
  • a sub word selection driver is arranged in the main word driver MWD in order to select one sub word line from among the sub word lines which are divided into eight in the direction of the main word line and which are allotted in the direction of the complementary bit lines in the number of four, as stated above.
  • the sub word selection driver forms a selection signal for selecting one of four sub word selection lines which are extended in the direction of arraying the sub word drivers.
  • the main amplifier MA and the Y decoder YDC are arranged on both sides of the memory array. Therefore, the sum between a signal transmission path for selecting the complementary bit lines of the subarray and a signal transmission path extending from the selected complementary bit lines to the input end of the main amplifier MA through the input/output lines becomes a signal transmission path merely for traversing the memory array whichever complementary bit lines are selected, and this signal transmission path can be shortened to half of a signal transmission path for one reciprocation as stated above. Thus, it is permitted to quicken memory accesses.
  • FIG. 13 Shown in FIG. 13 is a circuit diagram of one simplified embodiment from address inputting to data outputting as centers round the sense amplifier portion of a dynamic RAM according to this invention.
  • a sense amplifier 16 vertically held between two subarrays 15 , and circuits disposed in a cross area 18 are exemplified, and the others are shown in blocks.
  • a dynamic memory cell is representatively exemplified by one which is provided between a sub word line SWL laid in one of the subarrays 15 and one bit line BL of complementary bit lines BL, BLB.
  • the dynamic memory cell includes an address selecting MOSFET Qm and a storage capacitor Cs.
  • the gate of the address selecting MOSFET Qm is connected to the sub word line SWL, the drain of this MOSFET Qm is connected to the bit line BL, and the source thereof has the storage capacitor Cs connected thereto.
  • the electrodes of such storage capacitors Cs remote from the sources of the MOSFETs Qm are made common, and are fed with a plate voltage VPLT.
  • a back bias voltage VBB being minus is applied to the substrate (channels) of the MOSFETs Qm.
  • the back bias voltage VBB is set at a voltage of, e.g., ⁇ 1 V.
  • the selection level of the sub word line SWL is set at a high voltage VPP which is higher than the high level of the bit line by the threshold voltage of the address selecting MOSFET Qm.
  • the sense amplifier In the case where the sense amplifier is operated by an internal voltage drop VDL, the high level which is obtained by the amplification of the sense amplifier and which is applied to the bit line is set at the level of the internal voltage drop VDL. Accordingly, the high voltage VPP which corresponds to the selection level of the word line becomes VDL+Vth+ ⁇ .
  • One pair of complementary bit lines BL and BLB of the subarray laid on the left side of the sense amplifier are arranged in parallel as shown in the figure. Such complementary bit lines BL and BLB are respectively connected with the input/output nodes of the unit circuit of the sense amplifiers by shared switch MOSFETs Q 1 and Q 2 .
  • the unit circuit of the sense amplifiers includes a CMOS latch circuit configured of amplifying MOSFETs Q 5 , Q 6 of N-channel type and amplifying MOSFETs Q 7 , Q 8 of P-channel type, the gates and drains are cross-connected into latch forms.
  • the sources of the N-channel type MOSFETs Q 5 and Q 6 are connected to a common source line CSN.
  • the sources of the P-channel type MOSFETs Q 7 and Q 8 are connected to a common source line CSP.
  • Power switch MOSFETs are respectively connected to the common source lines CSN and CSP.
  • an operating voltage correspondent to the ground potential is applied to the common source line CSN to which the sources of the amplifying MOSFETs Q 5 and Q 6 of N-channel type are connected, by the power switch MOSFET Q 14 of N-channel type which is provided in the cross area 18 though not especially restricted.
  • the common source line CSP to which the sources of the amplifying MOSFETs Q 7 and Q 8 of P-channel type are connected is provided with the power MOSFET Q 15 of N-channel type which feeds the internal voltage VDL.
  • the power switch MOSFETs may be dispersedly disposed in the respective unit circuits.
  • Activation signals SAN and SAP for the sense amplifier which are fed to the gates of the power MOSFETs Q 14 and Q 15 of N-channel type, are inphase signals which are brought to high levels in activating the sense amplifier.
  • the high level of the signal SAP is set at the level of the stepped-up voltage VPP. Since the stepped-up voltage VPP is brought to about 3.6 V for the voltage VDL of 1.8 V, it can hold the common source line CSP at the level of the internal voltage VDL while satisfactorily keeping the N-channel type MOSFET Q 15 in its ON state.
  • a precharging (equalizing) circuit which includes an equalizing MOSFET Q 11 for short-circuiting the complementary bit lines, and switch MOSFETs Q 9 and Q 10 for feeding a half precharge voltage VBLR to the complementary bit lines.
  • the gates of the MOSFETs Q 9 through Q 11 are fed with a precharge signal PCB in common.
  • a driver circuit for forming the precharge signal PCB is such that an inverter circuit is provided in the cross area so as to quicken the rise and fall of the precharge signal. That is, in starting memory accesses, the MOSFETs Q 9 through Q 11 constituting the precharging circuits are switched at high speed through the inverter circuits dispersedly disposed in the respective cross areas, preceding the selection timing of a word line.
  • An IO switch circuit IOSW switch MOSFETs Q 19 , Q 20 for connecting local IO and main IO
  • IOSW switch MOSFETs Q 19 , Q 20 for connecting local IO and main IO
  • a half precharge circuit for the common source lines CSP and CSN of the sense amplifier a half precharge circuit for local input/output lines LIO, a VDL precharge circuit for main input/output lines, dispersed driver circuits for shared selection signal lines SHR and SHL, etc. are disposed in addition to the circuits shown in FIG. 3 as required.
  • the unit circuit of the sense amplifiers is connected to the similar complementary bit lines BL, BLB of the subarray 15 on the lower side of the drawing, through shared switch MOSFETs Q 3 and Q 4 .
  • the upper shared switch MOSFET Q 1 and Q 2 of the sense amplifier are brought into ON states, and the lower shared switch MOSPET Q 3 and Q 4 are brought into OFF states.
  • Switch MOSFETS Q 12 and Q 13 constitute a column (Y) switch circuit.
  • the switch MOSFETs Q 12 and Q 13 are brought into ON states, thereby to connect the input/output nodes of the unit circuit of the sense amplifiers with the local input/output lines LIO, LIO 1 B, LIO 2 , LIO 2 B, etc.
  • the input/output nodes of the sense amplifier are connected to the upper complementary bit lines BL, BLB, the minute signal of the memory cell connected to the selected sub word line SWL is amplified, and the amplified signal is conveyed to the local input/output lines LIO 1 , LIO 1 B through the column switch circuit (Q 12 and Q 13 ).
  • the local input/output lines LIO 1 , LIO 1 B are extended along the array of the sense amplifiers, that is, in the lateral direction in the figure.
  • the local input/output lines LIO 1 , LIO 1 B are connected to the main input/output lines MIO, MIOB to which the input terminals of a main amplifier 61 are respectively connected, through the IO switch circuit which includes the N-channel type MOSFETs Q 19 and Q 20 and which is provided in the cross area 18 .
  • the IO switch circuit is switch-controlled by a selection signal which is formed by decoding an address signal of X coordinates.
  • the IO switch circuit may have a CMOS switch construction in which P-channel type MOSFETs are connected in parallel with the N-channel type MOSFETs Q 19 and Q 20 , respectively.
  • the column selection signal YS is switched by a counter operation, and the connections between the local input/output lines LIO 1 , LIO 1 B and LIO 2 , LIO 2 B and every two pairs of the complementary bit lines BL, BLB of the subarray are switched in succession.
  • An address signal Ai is fed to an address buffer 51 .
  • This address buffer operates in time-division fashion so as to accept an X address signal and a Y address signal.
  • the X address signal is fed to a predecoder 52 , and a selection signal for a main word line MWL is formed through a main row decoder 11 and a main word driver 12 .
  • the address buffer 51 receives the address signal Ai fed from an external terminal and is operated by a power supply voltage VDDQ fed from an external terminal, the predecoder is operated by a stepped-down voltage VPERI (VDD) obtained by lowering the power supply voltage, and the main word driver 12 is operated by the stepped-up voltage VPP.
  • VDDQ power supply voltage
  • VPERI stepped-down voltage
  • a column decoder (driver) 53 includes a driver circuit whose operating voltage is formed by a MOSFET Q 23 constituting the VCLP generator circuit, and it forms the selection signal YS upon receiving the Y address signal which is fed by the time-divisional operation of the address buffer 51 .
  • the main amplifier 61 is operated by the stepped-down voltage VPERI (VDD), and it delivers an output from an external terminal Dout through an output buffer 62 which is operated by the power supply voltage VDDQ fed from the external terminal.
  • VDD stepped-down voltage
  • a write signal received from an external terminal Din is accepted through an input buffer 63 , and it is fed to the main input/output lines MIO and MIOB through a write amplifier (write driver) which is included in the main amplifier 61 in the figure.
  • the input portion of the output buffer 62 is provided with a level conversion circuit, and a logic portion for delivering the output signal of the level conversion circuit in synchronism with a timing signal correspondent to the clock signal.
  • the power supply voltage VDDQ which is fed from the external terminal is set at 3.3 V in a first aspect
  • the stepped-down voltage VPERI (VDD) which is fed to the internal circuits is set at 2.5 V
  • the operating voltage VDL of the sense amplifier is set at 1.8 V.
  • the selection signal (stepped-up voltage) of the word line is set at 3.6 V.
  • the precharge voltage VBLR of the bit line is set at 0.9 V corresponding to VDL/2
  • the plate voltage VPLT is also set at 0.9 V.
  • the substrate voltage VBB is set at ⁇ 1.0 V.
  • the power supply voltage VDDQ which is fed from the external terminal may be set at a low voltage of, e.g., 2.5 V in a second aspect.
  • the stepped-down voltage VPERI (VDD) and the stepped-down voltage VDL may be equally set at about 1.8 V.
  • the power supply voltage VDDQ which is fed from the external terminal is set at 3.3 V
  • the stepped-down voltage VPERI (VDD) which is fed to the internal circuits, and the operating voltage VDD of the sense amplifier are equally set at 2.0 V or 1.8 V.
  • the internal voltages can adopt various aspects of performance relative to the external power supply voltage VDDQ.
  • FIG. 14 Shown in FIG. 14 is a whole block diagram of one embodiment of a synchronous DRAM (hereinbelow, simply termed “SDRAM”) of about 256 Mbits to which this invention is applied.
  • SDRAM synchronous DRAM
  • a memory array 200 A constructing the memory bank “0” and a memory array 200 D constructing the memory bank “3” are exemplified though no special restriction is made.
  • memory arrays 200 B and 200 C corresponding to two memory banks “1” and “2” among the four memory banks are omitted.
  • the selection terminals of the memory cells arranged in an identical column are coupled to a word line (not shown) every column, and the data input/output terminals of the memory cells arranged in a identical row are coupled to complementary data lines (not shown) every row.
  • One of the unshown word lines of the memory array 200 A is driven to a selection level in accordance with the decoded result of a row address signal by a row decoder 201 A.
  • the unshown complementary data lines of the memory array 200 A are coupled to an I/O line 202 A which includes a sense amplifier and a column selector circuit.
  • the sense amplifier in the I/O line 202 A which includes the sense amplifier and the column selector circuit, is an amplifier circuit for detecting and amplifying a minute potential difference which appears between the corresponding complementary data lines by reading data out of the memory cell.
  • the column switch circuit in the I/O line 202 A is a switch circuit for selecting the individual pairs of complementary data lines and conducting them to complementary I/O lines.
  • the column switch circuit performs the selection operation in accordance with the decoded result of a column address signal by a column decoder 203 A.
  • Each of the memory arrays 200 B through 200 D is similarly provided with a row decoder 201 D, an I/O line 202 D which includes a sense amplifier and a column selector circuit, and a column decoder 203 D, as exemplified in the memory array 200 D.
  • the complementary I/O lines are connected to the output terminals of write buffers 214 A,B and the input terminals of main amplifiers 212 A,D.
  • the output signal of the main amplifiers 212 A,D is conveyed to the input terminal of a latch/register 213 , the output signal of which is outputted from an external terminal through an output buffer 211 .
  • a write signal inputted from an external terminal is conveyed to the input terminal of the write buffer 214 A, D through an input buffer 210 .
  • the external terminal is made a data input/output terminal which delivers data D 0 -D 15 composed of 16 bits.
  • Main amplifiers and write buffers similar to the above are respectively disposed in correspondence with the omitted memory arrays 200 B and C.
  • Address signals A 0 through A 13 fed from address input terminals are accepted in an address multiplexing form by a column address buffer 205 and a row address buffer 206 .
  • an address terminal for inputting an address signal A 14 is provided.
  • the address signals up to A 11 are validated in the ⁇ 4-bit configuration
  • the address signals up to A 10 are validated in the ⁇ 8-bit configuration
  • the address signals up to A 9 are validated in the ⁇ 16-bit configuration.
  • the address signals up to A 10 are validated in the ⁇ 4-bit configuration
  • the address signals up to A 9 are validated in the ⁇ 8-bit configuration
  • the address signals up to A 8 are validated in the ⁇ 16-bit configuration as shown in the figure.
  • the address signals fed from the address input terminals are held in the respectively corresponding buffers.
  • the row address buffer 206 accepts a refresh address signal outputted from a refresh counter 208 , as the row address signal in a refresh operation mode.
  • the output of the column address buffer 205 is fed as the preset data of a column address counter 207 , which outputs the column address signal being the preset data or values obtained by incrementing the column address signal in succession, toward the column decoders 203 A through 203 D in accordance with an operation mode designated by commands to be explained later.
  • a controller 209 indicated by a broken line in the figure is fed with external control signals such as a clock signal CLK, a clock enable signal CKE, a chip select signal /CS, a column address strobe signal /CAS (symbol “/” signifies that a signal bearing this symbol is a low enable signal), a row address strobe signal /RAS and a write enable signal /WE; and control data from the address input terminals A 0 through A 11 ; and it forms internal timing signals for controlling the operation mode of the SDRAM and the operations of the circuit blocks on the basis of the level changes and timings of the signals, etc. It includes a mode register 10 , a command decoder 20 , a timing generator circuit 30 and a clock buffer 40 , etc.
  • the clock signal CLK is inputted through the clock buffer 40 to the clock synchronizer circuit 50 as explained before, whereby an internal clock is generated.
  • the internal clock is employed as a timing signal for activating the output buffer 211 and the input buffer 210 , and it is also fed to the timing generator circuit 30 , in which a timing signal to be fed to the column address buffer 205 , row address buffer 206 and column address counter 207 is formed on the basis of such a clock signal.
  • the other external input signals are asserted in synchronism with the rising edge of the internal clock signal.
  • the chip select signal /CS commands the start of a command input cycle by its low level.
  • the chip select signal /CS is at its high level (in the non-selected state of the chip), any other input is insignificant.
  • the selected state of a memory bank and the internal operations of a burst operation etc. as will be explained later are not affected by the change of the SDRAM into the chip non-selection state.
  • the signals /RAS, /CAS and /WE differ in functions from respectively corresponding signals in an ordinary DRAM, and they are asserted when a command cycle to be explained later is defined.
  • the clock enable signal CKE is a signal for commanding the validity of the next pulse of the clock signal.
  • the rising edge of the next pulse of the clock signal CLK is validated for the high level of the signal CKE, and is invalidated for the low level thereof.
  • an external control signal /OE which performs an output enabling control for the output buffer 211 in a read mode
  • such a signal /OE is also fed to the controller 209 , and the output buffer 211 is brought into a high output impedance state when the signal is at a high level by way of example.
  • the row address signal is defined by the levels of the signals A 0 through A 11 in a row address strobe/bank active command cycle to be explained later, which is synchronized with the rising edge of the clock signal CLK (internal clock signal).
  • the address signals A 12 and A 13 are regarded as a bank selection signal in the row address strobe/bank active command cycle. That is, one of the four memory banks “0” through “3” is selected by the combination of the signals A 12 and A 13 .
  • a control for selecting the memory bank can be performed by such a processing as activating only the row decoder on the selected memory bank side, turning OFF all the column switch circuits on the non-selected memory bank sides, or connecting only the selected memory bank side to the input buffer 210 and output buffer 211 .
  • the column address signal is defined by the levels of the signals A 0 through A 9 in a read or write command (column address read command or column address write command to be explained later) cycle, which is synchronized with the rising edge of the clock signal CLK (internal clock signal). Besides, a column address thus defined is used as the start address of a burst access.
  • the register set data contain a burst length, a CAS latency, a write mode, etc.
  • settable burst lengths are “1”, “2”, “4”, “8” and “full page”; settable CAS latency values are “1”, “2”, and “3”; and settable write modes are “burst write” and “single write”.
  • the CAS latency indicates that number of cycles of the internal clock signal which are expended on a time period since the fall of the signal /CAS till the output operation of the output buffer 211 , in a read operation which is instructed by the column address read command to be explained later.
  • An internal operation time period for reading data is required before the settlement of read data, and it is set by this command in accordance with the working frequency of the internal clock signal.
  • the CAS latency is set at a relatively large value in the case of employing an internal clock signal of high frequency, whereas the CAS latency is set at a relatively small value in the case of employing an internal clock signal of low frequency.
  • an address fed to the terminals A 0 through A 9 are accepted as a row address signal, while signals fed to the terminals A 12 and A 13 are accepted as a selection signal for the memory bank.
  • the accepting operation is done in synchronism with the rising edge of the internal clock signal as stated above. For example, when the command is designated, a word line in the memory bank instructed by this command is selected, and memory cells connected to the word line are respectively conducted to corresponding complementary data lines.
  • a column address fed to the terminals A 0 through A 9 (in the case of the ⁇ 16-bit configuration) are accepted as a column address signal.
  • the column address signal thus accepted is fed to the column address counter 207 as a burst start address.
  • the memory bank and the word line included therein have been selected in the row address strobe/bank active command cycle beforehand, and the memory cells of the selected word line are sequentially selected and have data consecutively read therefrom in accordance with address signals which are outputted from the column address counter 207 in synchronism with the internal clock signal.
  • the number of the data items which are consecutively read is the number indicated by the burst length.
  • the data read from the output buffer 211 is started after that number of the cycles of the internal clock signal which is specified by the CAS latency.
  • the column address signal thus accepted is fed to the column address counter 207 as a burst start address in the burst write.
  • the steps of the burst write operation thus instructed are executed similarly to those of the burst read operation.
  • the CAS latency is not involved in the write operation, and the acceptance of write data is started from the column address/write command cycle.
  • This is a command which instructs the inhibition of any substantial operation, and which is designated by the signal /CS low level and the high levels of the signals /RAS, /CAS, /WE.
  • the SDRAM in the case where the burst operation is being executed in one memory bank and where the row address strobe/bank active command is issued midway by indicating another memory bank, the operation of a row address system in the other memory bank is permitted without exerting any influence on the operation in one memory bank under execution.
  • the SDRAM includes means for holding therein data, addresses and control signals fed from outside, and the held contents, particularly the addresses and the control signals, are retained every memory bank though not especially restricted.
  • data items which correspond to one word line in the memory block selected by the row address strobe/bank active command cycle are retained in the latch/register 213 for the read operation in advance of a column-system operation.
  • the precharge command and the row address strobe/bank active command can be issued during the execution of a command whose processing has not been ended, and to any memory bank different from the memory bank which is processed by the command under execution, whereby the internal operations of the different memory bank are started beforehand.
  • the SDRAM of this embodiment is subjected to memory accesses in 16-bit units, it has about 4M addresses based on the addresses of the signals A 0 through A 11 , and it is constructed of the four memory banks. In total, therefore, it has the storage capacity of, e.g., about 256 Mbits (4M ⁇ 4 banks ⁇ 16 bits).
  • Each of the memory chips is divided into a plurality of memory banks, and when the memory chips have been placed one over the other so as to hold their back surfaces in touch, the positions of the memory banks to which the same address is assigned are made different on the back surfaces of the memory chips, thereby to bring forth the effect that places to generate heat can be dispersed.
  • the two memory chips are placed one over the other so that their front surfaces may be opposite to each other, and they are shaped so that the sealing resin may be in touch with, at least, parts of the front surfaces held between the two memory chips, thereby to bring forth the effect that the thinning of a package can be realized.
  • Each of the two memory chips is provided with leads of tape state on its front surface thereof, and at least, the front surface part is in touch with the sealing resin except the back surface, thereby to bring forth the effect that the structure can be thinned in spite of stacking the two memory chips.
  • the thickness of the semiconductor storage device which includes the two memory chips held in the stacked structure is made equal to or less than the thickness of a semiconductor storage device whose storage capacity is half of or equal to the storage capacity of the former semiconductor storage device, and which has a single memory chip, thereby to bring forth the effect that the former semiconductor storage device is permitted to replace the existing (universal) semiconductor storage device.
  • the memory chips have a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by employing wiring means or bonding wire and by setting signal transmission paths and/or signal levels of the memory chips, whereby many kinds of memory chips can be formed by an identical process, and hence, the effect is brought forth that mass production can be achieved.
  • the plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units, thereby to bring forth the effects that semiconductor storage devices of 4-, 8-, 16- and 32-bit configurations which are generally and extensively employed and which include the stacked structure of the two chips can be realized on the basis of substantially one kind of memory chip, and that a storage capacity which is, at its maximum, double the storage capacity of one memory chip can be achieved.
  • the semiconductor storage device is formed with leads so as to extend from both side surfaces of a square resin-sealing package stretching in the longitudinal direction thereof; the 4 bits are formed by data terminals which consist of a first terminal through a fourth terminal; a pair of the first and second terminals, and a pair of the third and fourth terminals are separately disposed at positions symmetric with respect to a center line parallel to the longitudinal direction, on both sides of the square resin-sealing package; and electrodes corresponding respectively to the data terminals of 2 bits are electrically connected to the first terminal and the second terminal in each of the two memory chips; thereby to bring forth the effect that the data terminals of each of the two memory chips can be held in correspondence with the four data terminals in isolated fashion.
  • a plurality of semiconductor storage devices in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure a state where a sealing resin is not in touch with the back surfaces of the two memory chips, so as to make memory accesses in 4-bit units, are disposed on a square mounting circuit board which is formed with electrodes along one latus thereof, thereby to bring forth the effect that a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging can be obtained.
  • a plurality of such memory modules are used in a state where the electrodes are respectively plugged into a plurality of connectors which are arranged in parallel on a mother board, thereby to bring forth the effect that an increase in a storage capacity per unit volume and high-density packaging can be realized.
  • the two memory chips are placed one over the other so as to hold their back surfaces in touch with each other, and at least their front surfaces are in touch with the sealing resin, thereby to bring forth the effects that replacement with a universal semiconductor storage device can be realized, and that an increase in a storage capacity per unit volume and high-density packaging can be realized.
  • the two memory chips are placed one over the other so as to hold their front surfaces in touch with each other, and the sealing resin is formed so as to be in touch with, at least the front surfaces of the two memory chips, thereby to bring forth the effects that replacement with a universal semiconductor storage device can be realized, and that an increase in a storage capacity per unit volume and high-density packaging can be realized.
  • the thickness of the two memory chips held in the stacked structure is made equal to or less than the thickness of a single universal semiconductor storage device whose storage capacity is half of or equal to the total storage capacity of the two memory chips, thereby to bring forth the effects that replacement with a memory module employing the universal semiconductor storage device can be realized, and that an increase in the storage capacity and high-density packaging can be realized.
  • the semiconductor storage devices are provided with a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by setting voltages that are applied to external terminals, thereby to bring forth the effect that the bit configurations of data terminals can be selected in mounting the semiconductor storage devices on a memory module.
  • the plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units, thereby to bring forth the effect that memory modules of 4-, 8-, 16- and 32-bit configurations which are generally and extensively employed and which include the stacked structure of the two chips can be obtained.
  • the constructions of the memory array, subarray and sub word driver can adopt various aspects of performance
  • the input/output interface of the dynamic RAM can adopt various aspects of performance such as ones suited to LAN bus specifications etc. other than synchronous specifications.
  • the word lines may conform to a word shunt scheme other than the hierarchical word line scheme as stated before.
  • the two memory chips may be constructed from the combination of so-called “partial chips” in each of which a half storage area is validated. More specifically, the two memory chips, in each of which defects are partly existent and in each of which memory accesses can be made only to the half storage area, may be combined into the stacked structure as stated before, thereby to construct a single semiconductor storage device. In this case, when a non-defective semiconductor storage device constructed of a single memory chip is combined with the two memory chips in each of which the memory accesses can be made only to the half storage area, a semiconductor storage device which is the same as the single memory chip as externally viewed can be constructed.
  • the facility of making the memory accesses in 2-bit units can be effectively utilized even in the case stated above where the same semiconductor storage device as the memory chip when externally viewed is constructed by combining the two memory chips in each of which the memory accesses can be made only to the half storage area.
  • the reason therefor is that the two memory chips whose storage areas having the same addresses assigned thereto are valid in the valid semiconductor storage device can be combined to operate as the semiconductor storage device in which memory accesses in 4-bit units can be made.
  • each of the two memory chips may have a 4-bit configuration or 8-bit configuration.
  • memory accesses are made to either of the two memory chips by address signals, so that memory accesses may be made in plural-bit units such as 4-bit, 8-bit or 16-bit units corresponding to the bus width of a memory module or the like to which the semiconductor storage device is connected, unlike the memory accesses in the 2-bit units.
  • the semiconductor storage device may be a static RAM or a read-only memory such as EPROM or EEPROM.
  • This invention can be extensively utilized for various semiconductor storage devices and memory modules each of which has a stacked structure.
  • a plurality of semiconductor storage devices in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, are installed on a mounting circuit board which is square and which is formed with connector electrodes along one latus thereof, thereby to attain the effect that a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging can be obtained.

Abstract

Two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units. A memory module is so constructed that a plurality of such semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, are mounted on a mounting circuit board which is square and which is formed with electrodes along one latus thereof.

Description

    BACKGROUND OF THE INVENTION
  • This invention relates to a semiconductor storage device and a memory module, and to techniques which are effective when utilized chiefly in a universal semiconductor storage device such as dynamic RAM (Random Access Memory), and a memory module employing the semiconductor storage device. [0001]
  • An investigation conducted after the completion of the present invention has revealed that there is the official gazette of Japanese Patent Application Laid-open No. 222989/1992 as a document which is thought to be relevant to the present invention to be described later. The technique of the official gazette consists in the contrivance that stress-strain states which appear within a plurality of chips are uniformalized among the chips, the states being ascribable to heat generation during the operation of a semiconductor integrated circuit device in which the chips are sealed in a single package, and an example which realizes a double storage capacity by employing two memory chips is disclosed in paragraph 0031 in the official gazette. The official gazette does not contain any statement to the effect that, as in the invention of the present application to be described later, the thickness (in the direction of the height) of the semiconductor integrated circuit device viewed as one semiconductor storage device is made equal to or less than the thickness of a universal semiconductor storage device having an equal storage capacity. [0002]
  • By the way, memories each having output terminals of 2-bit configurations are stated in the official gazettes of Japanese Patent Applications Laid-open no. 20387/1985, No. 130139/1985, No. 171687/1985, No.146495/1987 and No. 262291/1987. [0003]
    • SUMMARY OF THE INVENTION
  • In the case where a plurality of chips are sealed into one package in this manner, the resulting device merely sealed becomes a unique semiconductor storage device, the uses of which are limited. For the purpose of permitting a semiconductor storage device to be widely utilized, it is important to provide this device with compatibility with the existing semiconductor storage devices. According to the semiconductor integrated circuit device in the aforementioned official gazette, the memory chips are held in a stacked structure so that the front surfaces (main surfaces) thereof may be opposite to each other, and a height for wire bonding which executes connections with leads is required over each of the front surfaces of the two chips, so that the package thickens inevitably. [0004]
  • Since packages have a tendency to be more thinned in universal semiconductor storage devices, it is difficult to realize a thickness equal to the thickness of the universal semiconductor storage device having an equal storage capacity by the technique stated in the aforementioned official gazette. In order to realize a large storage capacity within a small volume, each of semiconductor storage devices is often used in the way that memory chips are mounted on both surfaces of a mounting circuit board, thereby to construct a memory module, and that a plurality of such memory modules are plugged into a plurality of connectors which are juxtaposed on a mother board. In this case, the intervals of the individual connectors are set as near as possible. Therefore, the memory modules need to be formed by matching the overall thickness with, at least, the pitches of the connectors. Considering the radiation of heat during a memory operation, however, the memory modules should desirably be spaced from one another even if slightly. [0005]
  • As a result of the study stated above, the inventors of the present application have found it very important that, when two memory chips are mounted in a single package so as to realize a double storage capacity, the thickness of such a package is made equal to or less than the thickness of an existing semiconductor storage device in which a single memory chip is mounted. [0006]
  • An object of this invention is to provide a semiconductor storage device which can be formed thin while employing two memory chips. Another object of this invention is to provide a semiconductor storage device which has an excellent universality while employing two memory chips. Still another object of this invention is to provide a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging. A further object of this invention is to provide a memory module which is well adapted for use. The above and other objects and novel features of this invention will become apparent from the description of this specification and the accompanying drawings. [0007]
  • A typical one of inventions disclosed in the present application is briefly summarized as follows: Two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units. [0008]
  • Another typical one of the inventions disclosed in the present application is briefly summarized as follows: A memory module is so constructed that a plurality of semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, are mounted on a mounting circuit board which is square and formed with electrodes (connectors) along one latus thereof.[0009]
    • BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 is a schematic perspective view of essential portions showing an embodiment of a semiconductor storage device according to this invention; [0010]
  • FIGS. 2A and 2B schematic pattern diagram showing an embodiment of two memory chips which are made into the stacked structure of the semiconductor storage device according to this invention; [0011]
  • FIG. 3 is a top plan view showing an embodiment of the semiconductor storage device according to this invention; [0012]
  • FIG. 4 is an output system diagram showing an embodiment of the semiconductor storage device according to this invention; [0013]
  • FIG. 5 is a sectional view of essential portions showing an embodiment of a memory module which employs semiconductor storage devices according to this invention; [0014]
  • FIG. 6 is a front surface view showing an embodiment of the memory module according to this invention; [0015]
  • FIG. 7 is a back surface view showing an embodiment of the memory module according to this invention; [0016]
  • FIG. 8 is a block diagram showing an embodiment of the memory module shown in FIGS. 6 and 7; [0017]
  • FIG. 9 is a schematic constructional view for explaining a memory system which employs memory modules according to this invention; [0018]
  • FIG. 10 is a sectional view of essential portions showing another embodiment of the memory module employing semiconductor storage devices according to this invention; [0019]
  • FIG. 11 is a sectional view of essential portions showing another embodiment of the memory module employing semiconductor storage devices according to this invention; [0020]
  • FIG. 12 is a schematic layout diagram showing an embodiment of a dynamic RAM to which this invention is applied, [0021]
  • FIG. 13 is a circuit diagram showing an embodiment of a dynamic RAM according to this invention; [0022]
  • FIG. 14 is a whole block diagram showing an embodiment of a synchronous DRAM to which this invention is applied; [0023]
  • FIG. 15 is a sectional view of essential portions showing a further embodiment of the memory module employing semiconductor storage devices according to this invention; [0024]
  • FIG. 16 is a plan view for explaining the relationship between one of the memory chips and a lead frame in the semiconductor storage device according to this invention; and[0025]
    • DETAILED DESCRIPTION OF THE INVENTION
  • Shown in FIG. 1 is a perspective view of the essential portions of one embodiment of a semiconductor storage device according to this invention. In this embodiment, two memory chips are held in a structure in which they are stacked back to back within an identical package. That is, the memory chips are placed one over the other so that the back sides of these memory chips may be opposite to each other, and that the front surfaces thereof, on each of which bonding pads, etc. are formed, may face outwards. In the figure, among a large number of leads necessary for the package and the memory chips, some relevant to data terminals are representatively exemplified in order to facilitate understanding the invention. [0026]
  • A method for easily enlarging a semiconductor storage device is a technique in which two memory chips are mounted in a single package as stated in the official gazette mentioned before. Heretofore, in a 64-Mbit DRAM, 128 Mbits have been realizable when two 64-Mbit chips, in each of which ×4, ×8 and ×16 bits are existent, are stacked and assembled into one package. Likewise, 512 Mbits have been realizable when two 256-Mbit chips are assembled into one package. Such a stacking scheme has come into the limelight for inexpensive uses because it can shorten the development period than in the new development of a single chip of 128 Mbits or 512 Mbits. With the bit configuration of a prior-art single chip of ×4 bits or more, however, only ×8 bits or more have been realizable as the number of bits even when such chips are stacked. With the stacking scheme as stated above, therefore, it has been impossible to fabricate any ×4-bit product with 128 Mbits or 512 Mbits. [0027]
  • Moreover, when the two memory chips are assembled into one package as stated in the official gazette, the whole package thickens more as compared with the package of a semiconductor storage device which has a storage capacity equal to the total storage capacity of the two memory chips. Even when external terminals are equally arrayed, there occurs the problem that they cannot be matched with the pitches of connectors assembled at a high density in the case of mounting the memory chips in a memory module as explained later. In other words, the semiconductor storage device having the two memory chips becomes incompatible with a semiconductor storage device in which one memory chip is mounted in one package. [0028]
  • In the invention of the present application, two memory chips each of which can be set into a ×2-bit configuration are made into a stacked structure, thereby to construct a semiconductor storage device of ×4-bit configuration. In this case, the stacked structure is so contrived as to join the back surfaces of the two memory chips, in order that the total thickness of the memory chips to be assembled into the single package stated above may be equalized to the thickness of the package in which one chip is mounted. In this structure, the back surfaces of the memory chips are prevented to the utmost from coming into touch with a sealing resin which constitutes the single package. This is considerably different from the prior-art semiconductor integrated circuit device which is so constructed that the whole semiconductor chips including the back surfaces and front surfaces thereof are in touch with the sealing resin for the purpose of sealing. Owing to such a contrivance, the invention attains the thinning of the semiconductor storage device while adopting the stacked structure. [0029]
  • More specifically, in this embodiment, the memory chips are disposed back to back as described above, whereby the sealing resin at the back surface parts as is contained in the prior-art semiconductor integrated circuit device can be decreased to thin the package to that extent. As a result, the thickness of the package in this embodiment can be equalized to that of the package of the existing semiconductor storage device having the equal storage capacity, even when the two memory chips are accommodated in the single package by arranging leads of LOC (Lead On Chip) structure on both the upper and lower surfaces of the memory chip assembly as shown in the figure, and connecting these leads and the bonding pads of the memory chips by pieces of bonding wire. [0030]
  • In such a stacked structure, address terminals and control terminals which are disposed back to back in the respective memory chips need to have the corresponding leads connected in common outside. In the two memory chips, therefore, the relations between the leads and the bonding pads are set to be inverse laterally. In contrast, when the corresponding ones of the data terminals are connected in common outside likewise to the address terminals as stated above, data items each being of ×2 bits collide at an external terminal. [0031]
  • In this embodiment, accordingly, the address terminals etc. are held in mirror inversion between in the upper memory chip and lower memory chip of the stacked structure, whereupon the leads and the bonding pads are connected. In contrast, the two data terminals are connected with the two leads on the same side as viewed from the main surface of each memory chip. More specifically, in the illustrated example, each of leads for data formed as external terminals DQ[0032] 0 and DQ1 extends toward both the upper memory chip and the lower memory chip, and the upper leads are respectively connected to the pads DQ(a) and DQ(c) of the upper memory chip through the wire pieces, while the lower leads are not connected to any of the pads of the lower memory chip.
  • Each of leads for data formed as external terminals DQ[0033] 3 and DQ2 extends toward both the upper memory chip and the lower memory chip, and the lower leads are respectively connected to the pads DQ(a) and DQ(c) of the lower memory chip through the wire pieces, while the upper leads are not connected to any of the pads of the lower memory chip. The pad arrangement of the lower memory chip and that of the upper memory chip are similar, and pads DQ(b) and DQ(d) representatively shown on the upper memory chip are used together with the above pads DQ(a) and DQ(c) in the case where data are outputted in 4-bit units from one of the memory chips.
  • Accordingly, the pads DQ(b) and DQ(d) are not used in the embodiment of FIG. 1. The lead of A-signal is connected to pads A-PAD and B-PAD which are formed on the respective memory chips at the distal ends of both its leads extending toward the upper memory chip and the lower memory chip, so that the identical A-signal is supplied in common to both the memory chips. Regarding the leads for data, the leads on the side on which they are not connected may be cut away in order to lower input/output capacitances as will be explained later. [0034]
  • Shown in FIGS. 2A and 2B are schematic pattern diagrams of one embodiment of the two memory chips which are made into the stacked structure of the semiconductor storage device according to this invention. FIG. 2A shows the memory chip on the upper side (UPPER) and the leads and bonding wire pieces corresponding thereto, while FIG. 2B shows the memory chip on the lower side (LOWER) and the leads and bonding wire pieces corresponding thereto. [0035]
  • In the memory chips of this embodiment, the ×2-bit configuration, the ×4-bit configuration, the ×8-bit configuration and the ×16-bit configuration can be selected by a metal option, a bonding option or the combination thereof as will be explained later. The leads are shown in the maximum number so that they can cope with a plurality of these sorts of bit configurations. This embodiment exemplifies the leads and the bonding wire pieces in the case where each memory chip has a storage capacity of 256 Mbits in the ×2-bit configuration. It is accordingly to be understood that the leads with no signal names affixed thereto are the ones which are not existent in the above memory of the ×2-bit configuration. [0036]
  • The memory chip has the bonding pads arranged substantially rectilinearly, substantially on the center line thereof in the longitudinal direction thereof. Such a construction in which the bonding pads are aligned, becomes a very useful array of the bonding pads in the case to be explained later where the two memory chips are made into the stacked structure back to back as stated above, where the leads such as address terminals and control terminals are made common outside, and where the resulting structure is mounted in the single package. [0037]
  • Those lead terminals of the upper memory chip and lower memory chip which have signal names are arranged in lateral symmetry on the drawing, except for the leads for data, DQ[0038] 0 through DQ3 enclosed with rectangles. For example, the respective leads arranged on the right side of the upper memory chip UPPER agree with the leads which are arranged on the left side of the lower memory chip LOWER, and which are obtained by the mirror inversion with respect to the center line of the memory chip in the longitudinal direction thereof. Conversely, the respective leads arranged on the left side of the upper memory chip UPPER agree with the leads which are arranged on the right side of the lower memory chip LOWER, and which are obtained by the mirror inversion with respect to the center line of the memory chip in the longitudinal direction thereof.
  • As described above, those leads of the upper memory chip UPPER and lower memory chip LOWER to which the same signals are fed are arranged so as to be laterally inverse. Here, in the case where the bonding pads are aligned as stated before, the bonding wire pieces may be arranged for the leads so as to be laterally inverse between the upper side and the lower side. In FIG. 2, the leads and bonding wire pieces of the memory chip LOWER are arrayed in agreement with those of the memory chip UPPER reflected in a mirror which is put on the right side of the memory chip UPPER. Regarding the leads for data, DQ[0039] 0 through DQ3, however, the bonding wire pieces are connected to the leads DQ3 and DQ2 on the right side in the upper memory chip UPPER, while the bonding wire pieces are connected to the leads DQ0 and DQ1 on the right side in the lower memory chip LOWER. That is, the bonding pads of the memory chips and the leads to be connected thereto lie at the same positions, but the lead names are made different as stated above.
  • In this embodiment, a pair of leads are extended in the longitudinal direction of each memory chip and are respectively connected to leads for the voltage VDD of a power supply and the ground potential VSS of circuitry at the upper and lower ends of the chip. The pair of leads are formed as bus bars, and are respectively used for appropriately feeding the power supply voltage VDD and circuital ground potential VSS to the memory chip through low power supply impedances. The bus bars VDD and VSS are also arranged in the mirror inversion form between in the upper memory chip UPPER and in the lower memory chip LOWER. [0040]
  • In the case where the two memory chips as explained above are made into the structure in which they are stacked back to back so as to be joined at the back surfaces, the address terminals and the control terminals have the respectively corresponding leads placed one over the other vertically and connected in common. In contrast, each of the data terminals has the two leads laterally separated from each other and electrically isolated from each other in the stacked structure, whereby the data terminals DQ[0041] 0 through DQ3 of ×4 bits can be formed.
  • Shown in FIG. 3 is a top plan view of one embodiment of the semiconductor storage device according to this invention. In this embodiment, the memory chips each being of 256 Mbits are made into the stacked structure as explained above and are mounted in the single package, so as to have the storage capacity of 512 Mbits in total. This embodiment illustrates pin arrays in the cases where the two memory chips are set into the ×2-, ×[0042] 4- and ×8-bit configurations, thereby to realize the three sorts of bit configurations of ×4 bits, ×8 bits and ×16 bits, respectively.
  • In this embodiment, a power supply terminal VCC corresponds to the power supply leads VDD mentioned before, and power supply terminals VCCQ and VSSQ are used for feeding the voltage VCC of a power supply and the ground potential VSS of circuitry therefrom without being connected to the bus bars. The power supply voltages VCCQ and VSSQ serve to feed an operating voltage for an output buffer. Thus, power supply noise having developed in the output buffer is prevented from propagating to any other internal circuit through the bus bars inside the semiconductor storage device. It is consequently possible to stabilize the operations of internal circuits. [0043]
  • In this embodiment, the two memory chips each being of 256 Mbits are mounted back to back in the standard package of 54 pins. Of course, the package has a thickness specified as a standard. [0044]
  • Shown in FIG. 4 is an output system diagram of one embodiment of the semiconductor storage device according to this invention. Although not especially restricted, this embodiment illustrates an example which employs memory chips each having a storage capacity of 64 Mbits, and it shows the relationships between memory array positions and DQ Nos. in a normal mode. In this embodiment, one can be selected from among the four sorts of bit configurations of a ×2-bit product, a ×4-bit product, a ×8-bit product and a ×16-bit product. As will be explained later, a memory array corresponding to one memory bank is divided into 16 subarrays, two pairs of main input/output lines (MIO) are laid from each of the subarrays, and 32 main amplifiers are disposed in correspondence with the lines MIO. The 16 subarrays are subdivided into left and right (L/R) groups each consisting of 8 subarrays. [0045]
  • In the ×2-bit product, the left and right halves of the memory array correspond to the data terminals DQ[0046] 0 and DQ1 as indicated by the symbols L and R, respectively. In this embodiment, a selection signal YS is generated in accordance with address signals A0 through A7, and signals are delivered from one memory bank to the 32 pairs of MIO (main input/output lines). As stated above, the 32 pairs of MIO are separated into the two sets of the left L and right R sets each consisting of 16 pairs. In the ×2-bit product, buffer circuits-BC disposed in the number of 32 in correspondence with the main amplifiers MA are separated into two sets which correspond to the sets L and R, and which are respectively connected in common to data terminals DQ3(0) and DQ15(1).
  • Eight of the 16 main amplifiers of each set are selected in accordance with a column selection signal Y[0047] 8 or Y8B formed by an address signal A8, and one main amplifier MA is finally selected from among each of the sets L and R in accordance with address signals A9, A10 and A11 not shown, whereby data of 2 bits are respectively delivered from the output terminals DQ3(0) and DQ15(1) through the buffer circuits CB corresponding to the selected main amplifiers MA. Here, (0) and (1) indicate terminal Nos. in the case of the ×2-bit configuration. In the ensuing description, this shall hold true also of the other bit configurations.
  • In the ×4-bit product, the output lines of the buffer circuits CB are separated into four sets. The address signal A[0048] 11 for selecting the main amplifiers MA is correspondingly invalidated (degenerated), whereby data are read out in 4-bit units. In the ×8-bit product, the output lines of the buffer circuits CB are separated into eight sets, and the address signals A11 and A10 for selecting the main amplifiers MA are invalidated (degenerated), whereby data are read out in 8-bit units. Besides, in the ×16-bit product, the output lines of the buffer circuits CB are separated into 16 sets, and the address signals A11, A10 and A9 for selecting the main amplifiers MA are invalidated (degenerated), whereby data are read out in 8-bit units.
  • Shown in FIG. 5 is a sectional view of the essential portions of one embodiment of a memory module employing semiconductor storage devices according to this invention. The semiconductor storage devices each of which has memory chips held back to back, are mounted on both surfaces of a module substrate. Leads are bonded onto the memory chips through adhesive layers. Bus bars which extend orthogonally to the extending direction of the leads, have adhesive layers therefor formed thin so as to be smaller in height than the leads. Thus, a margin in a height direction can be enlarged in order to prevent the bus bars from being touched by pieces of wire which connect the selected ones of the leads and bonding pads provided on the front surfaces of the memory chips. [0049]
  • Each of the semiconductor storage devices mounted on both surfaces of the module substrate is constructed as the stacked structure of the two memory chips as explained before. Herein, since the stacked structure is formed so as to place the back surfaces of the memory chips one over the other, the thickness of a sealing resin can be set small. That is, the semiconductor storage device of this embodiment can be assembled into a package having a thickness equal to that of the ordinary semiconductor storage device in which only one memory chip is mounted and which has a storage capacity being half of that of the device of this embodiment, and it can be replaced with a memory module which employs the ordinary semiconductor storage device. By such replacement with the memory module, the storage capacity can be doubled for the same packaging volume, and a packaging area can be reduced to half for the same storage capacity. [0050]
  • As shown in FIG. 1, FIG. 2 and FIG. 5, the semiconductor storage device according to the present invention is sealed in the state where the back surfaces of the first and second memory chips (that is, the upper and lower memory chips) are placed one over the other. It includes the first lead DQ[0051] 0 and second lead DQ1 which extend from the first latus of the sealed structure to the inside and outside this sealed structure, and the third lead DQ2 and fourth lead DQ3 which extend from the second latus of the sealed structure opposite to the first latus to the inside and outside this sealed structure.
  • Each of the first and second memory chips includes the first terminal DQ(a), second terminal DQ(b), third terminal DQ(c) and fourth terminal DQ(d) for outputting data, which are arranged along a straight line being parallel to the longer latera of each memory chip and passing through the central parts of the shorter latera thereof, on the circuit forming surface of each memory chip. The second terminal DQ(b) is arranged at a position nearer to the first terminal DQ(a) between this first terminal DQ(a) and the third terminal DQ(c), while the third terminal DQ(c) is arranged at a position nearer to the fourth terminal DQ(d) between the second terminal DQ(b) and this fourth terminal DQ(d). [0052]
  • The first lead DQ[0053] 0 and fourth lead DQ3 are arranged at positions which oppose with the straight line held therebetween, while the second lead DQ1 and third lead DQ2 are arranged at positions which oppose with the straight line held therebetween. Each of the first and second memory chips outputs data in 2-bit units by employing either of the first and second terminals, for example, the first terminal DQ(a), and either of the third and fourth terminals, for example, the third terminal DQ(c), whereby data are outputted from the first, second, third and fourth leads in 4-bit units in total. By the way, in FIG. 1, terminals corresponding to the second terminal DQ(b) and the fourth terminal DQ(d) are not depicted in the lower memory chip, and they have been omitted from illustration in order to avoid the complicacy of the drawing. In actuality, the upper chip and the lower chip have similar constructions, and the respective terminals exist in correspondence.
  • Here, the data outputting in 2-bit units and the data outputting in 4-bit units can be selected for the first and second memory chips. In the case of executing the data outputting in 4-bit units, the first terminal DQ(a), second terminal DQ(b), third terminal DQ(c) and fourth terminal DQ(d) are employed as explained before. The positional relations of these terminals as stated above are determined in accordance with the normalized positions of the corresponding leads. That is, the individual terminals are arranged in such positional relations that the wire pieces can be shortened when the leads and terminals corresponding to each other are to be connected by these wire pieces. In other words, the first terminal DQ(a) and second terminal DQ(b) of each memory chip are arranged near the first lead DQ[0054] 0 and fourth lead DQ3, while the third terminal DQ(c) and fourth terminal DQ(d) of each memory chip are arranged near the second lead DQ1 and third lead DQ2.
  • In the case of the data outputting in 2-bit units, each memory chip can employ any two of the four data terminals in any desired combination. The inventors' study, however, has revealed that the first and second memory chips should preferably execute the outputting in 2-bit units by employing either of the first and second terminals (for example, the first terminal DQ(a)) and either of the third and fourth terminals (for example, the third terminal DQ(c)), respectively, as stated in this embodiment. [0055]
  • More specifically, the first terminal DQ(a) and third terminal DQ(c) of the first memory chip can be easily connected to the first lead DQ[0056] 0 and second lead DQ1 extending near these terminals, by using the wire pieces, respectively. In addition, the first terminal DQ(a) and third terminal DQ(c) of the first memory chip can be easily connected to the fourth lead DQ3 and third lead DQ2 extending near these terminals, by using the wire pieces, respectively.
  • On the other hand, when data are outputted from the individual memory chips in 2-bit units by employing the first terminal DQ(a) and second terminal DQ(b), these terminals concerning the first memory chip can be easily connected to the first lead DQ[0057] 0 and fourth lead DQ3 extending near them, by using the wire pieces, respectively. Concerning the second memory chip, however, the first terminal DQ(a) and second terminal DQ(b) must be connected to the second lead DQ1 and third lead DQ2 which are distant from these terminals, respectively. It is unfavorable to connect the distant leads and terminals through the wire pieces in this manner for the reason that short-circuiting with any other wire piece and increases in the parasitic capacitances of the wire pieces are incurred.
  • It is important for improving signal transmission speeds to lower the parasitic capacitances of signal lines. In this regard, the leads for data, which extend toward the memory chips but which are not connected, are shortened as in an embodiment illustrated in FIG. 15, whereby the parasitic capacitances of the leads can be lowered. In this embodiment, the leads for data not to be connected are cut at the end parts of the memory chips. It is also possible, however, to omit such non-connected data leads themselves. Shown in FIG. 16 is a plan view for explaining the relationship between the memory chip and a lead frame. The memory chip and leads illustrated are those of one of the two memory chips. As indicated by a lead enclosed with a broken line, each of the leads to be cut among a plurality of leads is formed in the shape of the alphabetic letter T facing toward the memory chip, so as not to fall off the sealing resin. [0058]
  • Referring to FIG. 1, the first lead DQ[0059] 0 should preferably be cut at or ahead of a position where it bifurcates toward the upper memory chip and the lower memory chip. Alternatively, leads having shapes after the cutting may be formed at the beginning of manufacture. That is, the asymmetric construction of each data lead concerning the lengths thereof extending toward the first and second memory chips can improve the signal transmission speed on the basis of the lowering of the parasitic capacitance, over the symmetric construction concerning the extending lengths.
  • Shown in FIG. 6 is a front surface view of one embodiment of the memory module according to this invention, and shown in FIG. 7 is a back surface view of this memory module. On each of front side and back side of a module substrate, semiconductor storage devices are mounted in a matrix-shaped array such as of 9×2 (=18 devices). Each of the semiconductor storage devices numbering 36 in total mounted on both surfaces of the memory module substrate, is constructed as the stacked structure of the two memory chips as shown in FIG. 1 or FIG. 5. [0060]
  • A register, a semiconductor integrated circuit device for signal conversion, capacitors and various resistors for a stabilized power supply, etc. are mounted between the semiconductor storage devices which are arrayed in the shape of a matrix such as of 9×2 and connectors (connecting electrodes) which are disposed on one end side of the memory module in the longitudinal direction thereof. [0061]
  • Shown in FIG. 8 is a block diagram showing one embodiment of the memory module in FIGS. 6 and 7. The memory module of this embodiment is such that a plurality of semiconductor storage devices, in each of which memory chips of ×2 bits are made into a stacked structure so as to equivalently form a ×4-bit configuration, are mounted on a module substrate, thereby to construct a memory system of 64 bits. That is, a memory system is constructed in which the semiconductor storage devices are disposed on both sides with a module substrate held therebetween and which has memory banks (BANK) “0” and “1”. [0062]
  • The memory banks “0” and “1” correspond to the nine semiconductor storage devices shown in FIGS. 6 and 7, respectively. More specifically, the nine semiconductor storage devices which take charge of the data terminals DQ[0063] 0 through DQ31 and parity bits CB0 through CB3 of the memory banks “0” and “1” correspond to the 9×2 devices shown in FIG. 6, while the nine semiconductor storage devices which take charge of the data terminals DQ32 through DQ64 and parity bits CB4 through CB7 of the memory banks “0” and “1” correspond to the 9×2 devices shown in FIG. 7. In this manner, the memory module is constructed into a 64-bit configuration having the data terminals DQ0 through DQ63. The eight bits; CB0 through CB7 are provided as the parity bits. Therefore, in a mother board on which the memory module according to this embodiment is mounted, data buses to be connected to sockets into which the connectors of the memory module are plugged are brought into a 72-bit configuration.
  • Shown in FIG. 9 is a schematic constructional view for explaining a memory system which employs the memory module according to this invention. The figure illustrates a situation where connectors (connecting electrodes) which are juxtaposed along one latus of each of the mounting circuit boards constituting a plurality of memory modules are arrayed in a state in which they are plugged in sockets provided on a mother board not shown. A plurality of memory modules need to be arrayed at narrow intervals in correspondence with the sockets, and it is important for this purpose that semiconductor storage devices to be mounted in the memory modules are formed thin. This embodiment indicates an example in which memory chip SDRAMs are aligned on a module substrate, unlike in the memory module of the embodiment shown in FIGS. 6 and 7. [0064]
  • Shown in FIG. 10 is a sectional view of the essential portions of another embodiment of the memory module employing semiconductor storage devices according to this invention. In this embodiment, in order to make two memory chips into a stacked structure and to diminish the thickness of a sealing resin, the stacked structure is formed so that the back surfaces of the memory chips may face outwards, reversely to the directions thereof in the embodiment of FIG. 5. In consequence, the back surface of each memory chip is not formed with any molding resin for the purpose of sealing. This, however, does not preclude a thin sealing-resin layer from being formed on a part or all of the back surface of the memory chip in the course of the manufacture of the semiconductor storage device. In bestowing the function of sealing, the sealing resin must not be formed thick even on the back surfaces as in the prior art. [0065]
  • In the invention of the present application, it is not indispensable that the back surfaces of the memory chips must be held in completely exposed states in order to thin the semiconductor storage device. Insofar as the whole stacked structure of the memory chips is formed thin, the sealing resin is not precluded from being formed thin even on the back surfaces in the course of the fabrication of the stacked structure, and the back surfaces may be coated with an appropriate insulating coating material or the like. Alternatively, the back surface of each memory chip may be formed with an oxide film so as to afford an insulating property. It incurs no problem in the semiconductor storage device itself to leave the back surface of a semiconductor substrate exposed as it is. However, when the semiconductor storage device has been installed in or is assembled into an electron device, an undesirable electrical contact might occur. It is therefore desirable to make the back surface electrically insulated as stated above. [0066]
  • The semiconductor storage device of this embodiment has the stacked structure in the aspect in which the back surfaces of the memory chips face outwards and in which the front sides formed with circuit elements are opposite to each other. Even when the directions of the memory chips are respectively reversed to those in the embodiment of FIG. 5 in this manner, the semiconductor storage device can be, in effect, thinned as in the embodiment of FIG. 5 for the reason that the sealing resin on the back surfaces can be dispensed with. [0067]
  • Shown in FIG. 11 is a sectional view of the essential portions of another embodiment of the semiconductor storage device according to this invention. In the semiconductor storage device of this embodiment, leads are printed on a tape (a flexible substrate), the tape is stuck on the front surface of each memory chip, and the leads are connected to bonding pads by wire bumps. Although no special restriction is intended, a sealing resin is deposited extending from the front surface of each memory chip to the side surfaces thereof, and the back surface of each memory chip is substantially exposed without being sealed with a resin, in the sense as explained in conjunction with FIG. 10, whereby the whole semiconductor storage device is formed thin. Therefore, two such semiconductor storage devices are made into a stacked structure on a module substrate as they are held independent. [0068]
  • As thus far described, in the semiconductor storage device according to the present invention, the ×2 configuration is added in the memory chip, thereby to realize the bit configurations of ×4 bits etc. in the stacked products. Furthermore, in the present invention, the ×2, ×4, ×8 and ×16 configurations are all set at bond options, whereby the chips fabricated by an identical wafer pre-process are permitted to expand merely by changing the setup of some pads at the stage of assemblage. Owing to such bond options, the memory chips to be formed by the wafer pre-process can be mass-produced, and reduction in cost can be promoted. [0069]
    TABLE 1
    BOP0B BOP1B BOP2B BOP3B
    × 2 VSS Floating — —
    × 4 Floating Floating — —
    × 8 VSS VSS — —
    × 16 Floating VSS — —
    EDO — — Floating —
    FP — — VSS —
    4KR — — — Floating
    8KR — — — VSS
  • Table 1 exemplifies the relationships between pad inputs for making the bonding options and functions in the case where the semiconductor storage device according to the invention of the present application has a storage capacity of, e.g., about [0070] 64 Mbits and where it is applied to EDO/FP products. The four sorts of bit configurations are switched using two pads BOP0B and BOP1B among four option pads, the EDO and first page (FP) are switched using one pad BOP2B, and 4K refresh (4KR) and 8K refresh (8KR) are switched using one remaining pad BOP3B. (TABLE 2)
  • Array of DQ pads of EDO-F: [0071]
    x16 8 7 9 6 10 5 11 4 12 3 13 2 14 1 15 0
    x8 3 4 2 5 1 6 0 7
    x4 1 2 0 3
    x2 1 1
  • Table 2 shows the relationships of the roles of DQ pins. In order to support, at most, ×16 bits, the memory module includes the DQ pins in the number of 16. The pins which are not used increase in number in the order of the ×8, ×4 and ×2 configurations, and they may be made “open” in these cases. The pins DQ[0072] 1, DQ3, DQ13 and DQ15 are employed in the ×4 configuration, and are also employed for the input/output pins of a parallel test which uses four DQ pins. The pin DQ15 is used in all cases. That is, it is used in both the ×2-bit configuration and an EIAJ test.
  • Referring to Table 1, among the option pads, also the two BOP0B, BOP1B and, if necessary, the two BOP2B, BOP3B may be connected to external terminals. In this case, when the external terminals have been mounted in the memory module, voltages as indicated in Table 1 are fed to these external terminals, whereby the bit configurations can be set as in the foregoing. Thus, when a user mounts the external terminals in the memory module, he/she can select the bit configuration of the data terminals. [0073]
  • Shown in FIG. 12 is a schematic layout diagram of one embodiment of a dynamic RAM to which this invention is applied. Individual circuit blocks in the figure are formed on a single semiconductor substrate, such as of single-crystal silicon, by known techniques for manufacturing semiconductor integrated circuits. Individual circuits in the figure are depicted substantially in agreement with the geometrical arrangement thereof on the semiconductor substrate. In this embodiment, memory arrays are separated into four as a whole, thereby to construct memory banks “0” through “3”. [0074]
  • The memory banks “0” through “3” correspond to the memory arrays which are separated in two vertically and in two laterally in the longitudinal direction of a semiconductor chip. A peripheral circuit PERI which includes an address input circuit, a data input/output circuit and a bonding pad array, is disposed at the central part of the chip extending in the longitudinal direction thereof. [0075]
  • Although not shown, an example of the peripheral circuit PERI includes a step-up voltage generator circuit and a control circuit for controlling the operation thereof, a voltage divider circuit in the case of an interface of SSTL as divides an external power supply voltage VDDQ into ½ and thus forms the reference voltage of an input circuit constructed of a differential circuit, an input/output circuit and a clock control circuit therefor, a Y predecoder and a read/write buffer, a step-down circuit for forming the operating voltage of a peripheral circuit, a VPP sensor for detecting whether or not a voltage VPP has a desired value, an X address latch circuit, a Y clock circuit, a mode decoder/clock buffer and a command circuit, a Y counter and a control circuit therefor, a refresh control circuit, a bonding option circuit, a power supply turn-ON detector circuit, etc. [0076]
  • Regarding each of the memory arrays totaling four; two separated vertically and two separated laterally in the longitudinal direction of the semiconductor chip as stated above, an X-coordinate predecoder circuit ROWPDC and remedy circuit ROWRED and a Y-coordinate predecoder circuit COLPDC and remedy circuit COLRED are collectively arranged at the lateral middle part of the semiconductor chip in the longitudinal direction thereof. That is, two sets each consisting of the X-coordinate predecoder circuit ROWPDC and remedy circuit ROWRED and the Y-coordinate predecoder circuit COLPDC and remedy circuit COLRED are distributed in correspondence with the two left and right memory arrays among the four memory arrays. [0077]
  • Main word driver areas MWD are formed at the intermediate part of the memory arrays likewise to the above, and they drive main word lines laid extending downwards and upwards in correspondence with the corresponding memory arrays, respectively. With this construction, in the case of employing the same subarrays as in the foregoing, the main word lines are extended so as to penetrate through 16 subarrays. Although no special restriction is intended, Y decoders YDC are disposed on the peripheral sides of the chip remote from the central part thereof in the memory arrays. That is, the memory arrays separated in four are so arranged as to be held between main amplifiers MA located on the central side and the Y decoders YDC located on the peripheral sides, respectively. The four memory arrays correspond to the four memory banks. [0078]
  • As described above, one memory array corresponding to the memory bank has the 16 subarrays in the direction of the word lines, and the two pairs of main input/output lines from each of the subarrays are extended in the direction of bit lines between the subarrays, thereby to be led to the input terminals of the 32 main amplifiers MA as stated before. Thus, the bit configurations of ×2, ×4, ×8 and ×16 as explained before can be switched. [0079]
  • In this embodiment, the memory arrays separated in four are so arranged as to be held between the main amplifiers MA located on the central side and the Y decoders YDC located on the peripheral sides, respectively. As one of the memory arrays is shown on an enlarged scale, each memory array is divided into a plurality of [0080] subarrays 15. Each of such subarrays 15 is so formed as to be surrounded with sense amplifier areas 16 and sub word driver areas 17 which are respectively located holding the subarray 15 therebetween. The intersection parts between the sense amplifier areas 16 and the sub word driver areas 17 are determined as cross areas 18. Sense amplifiers to be provided in the sense amplifier areas 16 are constructed in conformity with a shared sense scheme. Except for the sense amplifiers which are arranged at both ends of each memory cell array, complementary bit lines are laid at the right and left of the sense amplifiers and are selectively connected to either of the right and left complementary bit lines of the memory cell array.
  • Each of the [0081] subarrays 15 includes sub word lines in the number of, for example, 256 and 256 pairs of complementary bit lines (or data lines) being orthogonal to the sub word lines, though not shown. Incidentally, the subarray is also provided with spare word lines and spare complementary bit lines in order to remedy defective word lines and defective bit lines, respectively. In each of the memory arrays, the subarrays are disposed in the number of 16 in the direction of arraying the word lines, so that the sub word lines are laid for about 4 kbits as a whole, while the subarrays are disposed in the number of 16 in the direction of arraying the bit lines, so that the complementary bit lines are laid for about 4 kbits as a whole. Such memory arrays are disposed in the number of four as a whole, so that the semiconductor storage device has the storage capacity of, e.g., 4×4k×4k=64 Mbits as a whole.
  • As another example, each of the [0082] subarrays 15 may well include sub word lines in the number of, for example, 512 and 512 pairs of complementary bit lines (or data lines) being orthogonal to the sub word lines, though not shown. Likewise to the above, the subarray is also provided with spare word lines and spare complementary bit lines in order to remedy defective word lines and defective bit lines, respectively. In each of the memory arrays, the subarrays are disposed in the number of 16 in the direction of arraying the word lines, so that the sub word lines are laid for about 8 kbits as a whole, while the subarrays are disposed in the number of 16 in the direction of arraying the bit lines, so that the complementary bit lines are laid for about 8 kbits as a whole. Such memory arrays are disposed in the number of four as a whole, so that the semiconductor storage device has a storage capacity of, e.g., 8×8k×4k=256 Mbits as a whole.
  • By adopting the hierarchical word line scheme as explained above, the lengths of the complementary bit lines are divided into {fraction (1/16)} in correspondence with the 16 subarrays in any of the above cases. The lengths of the sub word lines are divided into {fraction (1/16)} in correspondence with the 16 subarrays. The sub word driver (sub word line driving circuit) [0083] 17 is provided every divided subarray 15 of each of the memory arrays. The sub word driver 17 forms a selection signal for the sub word line which is divided to the length of {fraction (1/16)} relative to the main word line as stated above, and which is extended in parallel with the main word line.
  • In this embodiment, for the purpose of decreasing the number of the main word lines, in other words, for the purpose of moderating the wiring pitch of the main word lines, the sub word lines in the number of four are laid in the direction of the complementary bit lines in correspondence with one main word line though not especially restricted. A sub word selection driver, not shown, is arranged in the main word driver MWD in order to select one sub word line from among the sub word lines which are divided into eight in the direction of the main word line and which are allotted in the direction of the complementary bit lines in the number of four, as stated above. The sub word selection driver forms a selection signal for selecting one of four sub word selection lines which are extended in the direction of arraying the sub word drivers. [0084]
  • In the case of adopting the layout as shown in FIG. 12, when a Y address is inputted, it is passed through an address buffer ADDBUP and is conveyed to the Y decoder YDC which is located on the peripheral edge side of the chip through the remedy circuit and the predecoder disposed at the middle part of the memory arrays, and in which a Y selection signal is formed. The complementary bit lines of one subarray is selected by the Y selection signal, and the difference of the potentials of the complementary bit lines is conveyed to the main amplifier MA located on the central side of the chip remote from the peripheral edge side thereof and is amplified therein, whereupon the amplified signal is outputted through an output circuit not shown. [0085]
  • At a glance, this construction is so judged that the difference signal will circumvent the chip to lengthen a time period expended till the outputting of the read signal. However, the address signal needs to be inputted to the remedy circuit as it is. Therefore, when the remedy circuit is located anywhere at the central part of the chip, the outputting time point of the predecoder is determined after waiting for the result of decision on whether or not the inputted address is defective. That is, when the predecoder and the remedy circuit are distant, a signal delay involved therein forms a cause for retarding the actual Y selection operation. [0086]
  • In this embodiment, the main amplifier MA and the Y decoder YDC are arranged on both sides of the memory array. Therefore, the sum between a signal transmission path for selecting the complementary bit lines of the subarray and a signal transmission path extending from the selected complementary bit lines to the input end of the main amplifier MA through the input/output lines becomes a signal transmission path merely for traversing the memory array whichever complementary bit lines are selected, and this signal transmission path can be shortened to half of a signal transmission path for one reciprocation as stated above. Thus, it is permitted to quicken memory accesses. [0087]
  • In the case where the four memory arrays are respectively held in correspondence with the memory banks as explained above, the positions of the memory arrays which are simultaneously operated in the two memory chips are made different, in the structure in which the two memory chips are stacked back to back as in the embodiment of FIG. 5. That is, two of the memory arrays are selected while holding therebetween the peripheral circuit PERI which is provided at the vertical central part of the memory chip. Thus, places to generate heat can be dispersed. [0088]
  • Shown in FIG. 13 is a circuit diagram of one simplified embodiment from address inputting to data outputting as centers round the sense amplifier portion of a dynamic RAM according to this invention. In the figure, a [0089] sense amplifier 16 vertically held between two subarrays 15, and circuits disposed in a cross area 18 are exemplified, and the others are shown in blocks.
  • A dynamic memory cell is representatively exemplified by one which is provided between a sub word line SWL laid in one of the [0090] subarrays 15 and one bit line BL of complementary bit lines BL, BLB. The dynamic memory cell includes an address selecting MOSFET Qm and a storage capacitor Cs. The gate of the address selecting MOSFET Qm is connected to the sub word line SWL, the drain of this MOSFET Qm is connected to the bit line BL, and the source thereof has the storage capacitor Cs connected thereto. The electrodes of such storage capacitors Cs remote from the sources of the MOSFETs Qm are made common, and are fed with a plate voltage VPLT. A back bias voltage VBB being minus is applied to the substrate (channels) of the MOSFETs Qm. Although not especially restricted, the back bias voltage VBB is set at a voltage of, e.g., −1 V. The selection level of the sub word line SWL is set at a high voltage VPP which is higher than the high level of the bit line by the threshold voltage of the address selecting MOSFET Qm.
  • In the case where the sense amplifier is operated by an internal voltage drop VDL, the high level which is obtained by the amplification of the sense amplifier and which is applied to the bit line is set at the level of the internal voltage drop VDL. Accordingly, the high voltage VPP which corresponds to the selection level of the word line becomes VDL+Vth+α. One pair of complementary bit lines BL and BLB of the subarray laid on the left side of the sense amplifier are arranged in parallel as shown in the figure. Such complementary bit lines BL and BLB are respectively connected with the input/output nodes of the unit circuit of the sense amplifiers by shared switch MOSFETs Q[0091] 1 and Q2.
  • The unit circuit of the sense amplifiers includes a CMOS latch circuit configured of amplifying MOSFETs Q[0092] 5, Q6 of N-channel type and amplifying MOSFETs Q7, Q8 of P-channel type, the gates and drains are cross-connected into latch forms. The sources of the N-channel type MOSFETs Q5 and Q6 are connected to a common source line CSN. The sources of the P-channel type MOSFETs Q7 and Q8 are connected to a common source line CSP. Power switch MOSFETs are respectively connected to the common source lines CSN and CSP.
  • Although no especial restriction is intended, an operating voltage correspondent to the ground potential is applied to the common source line CSN to which the sources of the amplifying MOSFETs Q[0093] 5 and Q6 of N-channel type are connected, by the power switch MOSFET Q14 of N-channel type which is provided in the cross area 18 though not especially restricted. Likewise, the common source line CSP to which the sources of the amplifying MOSFETs Q7 and Q8 of P-channel type are connected is provided with the power MOSFET Q15 of N-channel type which feeds the internal voltage VDL. The power switch MOSFETs may be dispersedly disposed in the respective unit circuits.
  • Activation signals SAN and SAP for the sense amplifier, which are fed to the gates of the power MOSFETs Q[0094] 14 and Q15 of N-channel type, are inphase signals which are brought to high levels in activating the sense amplifier. The high level of the signal SAP is set at the level of the stepped-up voltage VPP. Since the stepped-up voltage VPP is brought to about 3.6 V for the voltage VDL of 1.8 V, it can hold the common source line CSP at the level of the internal voltage VDL while satisfactorily keeping the N-channel type MOSFET Q15 in its ON state.
  • Provided at the input/output nodes of the unit circuit of the sense amplifiers is a precharging (equalizing) circuit which includes an equalizing MOSFET Q[0095] 11 for short-circuiting the complementary bit lines, and switch MOSFETs Q9 and Q10 for feeding a half precharge voltage VBLR to the complementary bit lines. The gates of the MOSFETs Q9 through Q11 are fed with a precharge signal PCB in common. Although not shown, a driver circuit for forming the precharge signal PCB is such that an inverter circuit is provided in the cross area so as to quicken the rise and fall of the precharge signal. That is, in starting memory accesses, the MOSFETs Q9 through Q11 constituting the precharging circuits are switched at high speed through the inverter circuits dispersedly disposed in the respective cross areas, preceding the selection timing of a word line.
  • An IO switch circuit IOSW (switch MOSFETs Q[0096] 19, Q20 for connecting local IO and main IO) is located in the cross area 18. Further, a half precharge circuit for the common source lines CSP and CSN of the sense amplifier, a half precharge circuit for local input/output lines LIO, a VDL precharge circuit for main input/output lines, dispersed driver circuits for shared selection signal lines SHR and SHL, etc. are disposed in addition to the circuits shown in FIG. 3 as required.
  • The unit circuit of the sense amplifiers is connected to the similar complementary bit lines BL, BLB of the [0097] subarray 15 on the lower side of the drawing, through shared switch MOSFETs Q3 and Q4. For example, when the sub word line SWL of the upper subarray is selected, the upper shared switch MOSFET Q1 and Q2 of the sense amplifier are brought into ON states, and the lower shared switch MOSPET Q3 and Q4 are brought into OFF states. Switch MOSFETS Q12 and Q13 constitute a column (Y) switch circuit. When the selection signal YS is set at the selection level (high level), the switch MOSFETs Q12 and Q13 are brought into ON states, thereby to connect the input/output nodes of the unit circuit of the sense amplifiers with the local input/output lines LIO, LIO1B, LIO2, LIO2B, etc.
  • Thus, the input/output nodes of the sense amplifier are connected to the upper complementary bit lines BL, BLB, the minute signal of the memory cell connected to the selected sub word line SWL is amplified, and the amplified signal is conveyed to the local input/output lines LIO[0098] 1, LIO1B through the column switch circuit (Q12 and Q13). The local input/output lines LIO1, LIO1B are extended along the array of the sense amplifiers, that is, in the lateral direction in the figure. The local input/output lines LIO1, LIO1B are connected to the main input/output lines MIO, MIOB to which the input terminals of a main amplifier 61 are respectively connected, through the IO switch circuit which includes the N-channel type MOSFETs Q19 and Q20 and which is provided in the cross area 18.
  • The IO switch circuit is switch-controlled by a selection signal which is formed by decoding an address signal of X coordinates. Incidentally, the IO switch circuit may have a CMOS switch construction in which P-channel type MOSFETs are connected in parallel with the N-channel type MOSFETs Q[0099] 19 and Q20, respectively. In the burst mode of a synchronous DRAM, the column selection signal YS is switched by a counter operation, and the connections between the local input/output lines LIO1, LIO1B and LIO2, LIO2B and every two pairs of the complementary bit lines BL, BLB of the subarray are switched in succession.
  • An address signal Ai is fed to an [0100] address buffer 51. This address buffer operates in time-division fashion so as to accept an X address signal and a Y address signal. The X address signal is fed to a predecoder 52, and a selection signal for a main word line MWL is formed through a main row decoder 11 and a main word driver 12. The address buffer 51 receives the address signal Ai fed from an external terminal and is operated by a power supply voltage VDDQ fed from an external terminal, the predecoder is operated by a stepped-down voltage VPERI (VDD) obtained by lowering the power supply voltage, and the main word driver 12 is operated by the stepped-up voltage VPP. A logic circuit with a level conversion facility, which receives the predecoded signal, is employed as the main word driver 12. A column decoder (driver) 53 includes a driver circuit whose operating voltage is formed by a MOSFET Q23 constituting the VCLP generator circuit, and it forms the selection signal YS upon receiving the Y address signal which is fed by the time-divisional operation of the address buffer 51.
  • The [0101] main amplifier 61 is operated by the stepped-down voltage VPERI (VDD), and it delivers an output from an external terminal Dout through an output buffer 62 which is operated by the power supply voltage VDDQ fed from the external terminal. A write signal received from an external terminal Din is accepted through an input buffer 63, and it is fed to the main input/output lines MIO and MIOB through a write amplifier (write driver) which is included in the main amplifier 61 in the figure. The input portion of the output buffer 62 is provided with a level conversion circuit, and a logic portion for delivering the output signal of the level conversion circuit in synchronism with a timing signal correspondent to the clock signal.
  • Although no special restriction is intended, the power supply voltage VDDQ which is fed from the external terminal is set at 3.3 V in a first aspect, the stepped-down voltage VPERI (VDD) which is fed to the internal circuits is set at 2.5 V, and the operating voltage VDL of the sense amplifier is set at 1.8 V. Besides, the selection signal (stepped-up voltage) of the word line is set at 3.6 V. The precharge voltage VBLR of the bit line is set at 0.9 V corresponding to VDL/2, and the plate voltage VPLT is also set at 0.9 V. In addition, the substrate voltage VBB is set at −1.0 V. The power supply voltage VDDQ which is fed from the external terminal, may be set at a low voltage of, e.g., 2.5 V in a second aspect. In the case of such a low power supply voltage VDDQ, the stepped-down voltage VPERI (VDD) and the stepped-down voltage VDL may be equally set at about 1.8 V. [0102]
  • Alternatively, it is allowed that the power supply voltage VDDQ which is fed from the external terminal is set at 3.3 V, and that the stepped-down voltage VPERI (VDD) which is fed to the internal circuits, and the operating voltage VDD of the sense amplifier are equally set at 2.0 V or 1.8 V. In this manner, the internal voltages can adopt various aspects of performance relative to the external power supply voltage VDDQ. [0103]
  • Shown in FIG. 14 is a whole block diagram of one embodiment of a synchronous DRAM (hereinbelow, simply termed “SDRAM”) of about 256 Mbits to which this invention is applied. In the SDRAM of this embodiment, among four memory banks, a [0104] memory array 200A constructing the memory bank “0” and a memory array 200D constructing the memory bank “3” are exemplified though no special restriction is made.
  • That is, memory arrays [0105] 200B and 200C corresponding to two memory banks “1” and “2” among the four memory banks are omitted. Each of the memory arrays 200A through 200D held in correspondence with the four memory banks “0” through “3”, respectively, includes dynamic memory cells which are arranged in the shape of a matrix as in the memory arrays 200A and 200D exemplified in the figure. According to the illustration, the selection terminals of the memory cells arranged in an identical column are coupled to a word line (not shown) every column, and the data input/output terminals of the memory cells arranged in a identical row are coupled to complementary data lines (not shown) every row.
  • One of the unshown word lines of the [0106] memory array 200A is driven to a selection level in accordance with the decoded result of a row address signal by a row decoder 201A. The unshown complementary data lines of the memory array 200A are coupled to an I/O line 202A which includes a sense amplifier and a column selector circuit. The sense amplifier in the I/O line 202A which includes the sense amplifier and the column selector circuit, is an amplifier circuit for detecting and amplifying a minute potential difference which appears between the corresponding complementary data lines by reading data out of the memory cell. The column switch circuit in the I/O line 202A is a switch circuit for selecting the individual pairs of complementary data lines and conducting them to complementary I/O lines. The column switch circuit performs the selection operation in accordance with the decoded result of a column address signal by a column decoder 203A.
  • Each of the memory arrays [0107] 200B through 200D is similarly provided with a row decoder 201D, an I/O line 202D which includes a sense amplifier and a column selector circuit, and a column decoder 203D, as exemplified in the memory array 200D. The complementary I/O lines are connected to the output terminals of write buffers 214A,B and the input terminals of main amplifiers 212A,D. The output signal of the main amplifiers 212A,D is conveyed to the input terminal of a latch/register 213, the output signal of which is outputted from an external terminal through an output buffer 211.
  • A write signal inputted from an external terminal is conveyed to the input terminal of the [0108] write buffer 214A, D through an input buffer 210. Although not especially restricted, the external terminal is made a data input/output terminal which delivers data D0-D15 composed of 16 bits. Main amplifiers and write buffers similar to the above are respectively disposed in correspondence with the omitted memory arrays 200B and C.
  • Address signals A[0109] 0 through A13 fed from address input terminals are accepted in an address multiplexing form by a column address buffer 205 and a row address buffer 206. In the case where the SDRAM has a storage capacity of, e.g., 256 Mbits and where memory accesses are made in 2-bit units as explained before, an address terminal for inputting an address signal A14 is provided. The address signals up to A11 are validated in the ×4-bit configuration, the address signals up to A10 are validated in the ×8-bit configuration, and the address signals up to A9 are validated in the ×16-bit configuration. In the case of a storage capacity of, e.g., 64 Mbits, the address signals up to A10 are validated in the ×4-bit configuration, the address signals up to A9 are validated in the ×8-bit configuration, and the address signals up to A8 are validated in the ×16-bit configuration as shown in the figure.
  • The address signals fed from the address input terminals are held in the respectively corresponding buffers. The [0110] row address buffer 206 accepts a refresh address signal outputted from a refresh counter 208, as the row address signal in a refresh operation mode. The output of the column address buffer 205 is fed as the preset data of a column address counter 207, which outputs the column address signal being the preset data or values obtained by incrementing the column address signal in succession, toward the column decoders 203A through 203D in accordance with an operation mode designated by commands to be explained later.
  • Although not especially restricted, a [0111] controller 209 indicated by a broken line in the figure is fed with external control signals such as a clock signal CLK, a clock enable signal CKE, a chip select signal /CS, a column address strobe signal /CAS (symbol “/” signifies that a signal bearing this symbol is a low enable signal), a row address strobe signal /RAS and a write enable signal /WE; and control data from the address input terminals A0 through A11; and it forms internal timing signals for controlling the operation mode of the SDRAM and the operations of the circuit blocks on the basis of the level changes and timings of the signals, etc. It includes a mode register 10, a command decoder 20, a timing generator circuit 30 and a clock buffer 40, etc.
  • The clock signal CLK is inputted through the [0112] clock buffer 40 to the clock synchronizer circuit 50 as explained before, whereby an internal clock is generated. Although not especially restricted, the internal clock is employed as a timing signal for activating the output buffer 211 and the input buffer 210, and it is also fed to the timing generator circuit 30, in which a timing signal to be fed to the column address buffer 205, row address buffer 206 and column address counter 207 is formed on the basis of such a clock signal.
  • The other external input signals are asserted in synchronism with the rising edge of the internal clock signal. The chip select signal /CS commands the start of a command input cycle by its low level. When the chip select signal /CS is at its high level (in the non-selected state of the chip), any other input is insignificant. However, the selected state of a memory bank and the internal operations of a burst operation etc. as will be explained later are not affected by the change of the SDRAM into the chip non-selection state. The signals /RAS, /CAS and /WE differ in functions from respectively corresponding signals in an ordinary DRAM, and they are asserted when a command cycle to be explained later is defined. [0113]
  • The clock enable signal CKE is a signal for commanding the validity of the next pulse of the clock signal. The rising edge of the next pulse of the clock signal CLK is validated for the high level of the signal CKE, and is invalidated for the low level thereof. By the way, in the case of providing an external control signal /OE which performs an output enabling control for the [0114] output buffer 211 in a read mode, such a signal /OE is also fed to the controller 209, and the output buffer 211 is brought into a high output impedance state when the signal is at a high level by way of example.
  • The row address signal is defined by the levels of the signals A[0115] 0 through A11 in a row address strobe/bank active command cycle to be explained later, which is synchronized with the rising edge of the clock signal CLK (internal clock signal).
  • The address signals A[0116] 12 and A13 are regarded as a bank selection signal in the row address strobe/bank active command cycle. That is, one of the four memory banks “0” through “3” is selected by the combination of the signals A12 and A13. Although not especially restricted, a control for selecting the memory bank can be performed by such a processing as activating only the row decoder on the selected memory bank side, turning OFF all the column switch circuits on the non-selected memory bank sides, or connecting only the selected memory bank side to the input buffer 210 and output buffer 211.
  • In the case of the storage capacity of 256 Mbits and the ×16-bit configuration as stated before, the column address signal is defined by the levels of the signals A[0117] 0 through A9 in a read or write command (column address read command or column address write command to be explained later) cycle, which is synchronized with the rising edge of the clock signal CLK (internal clock signal). Besides, a column address thus defined is used as the start address of a burst access.
  • There will now be explained the main operation modes of the SDRAM instructed by commands: [0118]
  • (1) Mode Register Set Command (Mo) [0119]
  • This is a command for setting the [0120] mode register 30. This command is designated by the signals /CS, /RAS, /CAS, /WE=low levels, and data to be set (register set data) are given through the terminals A0 through A11. Although not especially restricted, the register set data contain a burst length, a CAS latency, a write mode, etc. Although not especially restricted, settable burst lengths are “1”, “2”, “4”, “8” and “full page”; settable CAS latency values are “1”, “2”, and “3”; and settable write modes are “burst write” and “single write”.
  • The CAS latency indicates that number of cycles of the internal clock signal which are expended on a time period since the fall of the signal /CAS till the output operation of the [0121] output buffer 211, in a read operation which is instructed by the column address read command to be explained later. An internal operation time period for reading data is required before the settlement of read data, and it is set by this command in accordance with the working frequency of the internal clock signal. In other words, the CAS latency is set at a relatively large value in the case of employing an internal clock signal of high frequency, whereas the CAS latency is set at a relatively small value in the case of employing an internal clock signal of low frequency.
  • (2) Row Address Strobe/Bank Active Command (Ac) [0122]
  • This is a command which validates the instruction of row address strobe and the selection of a memory bank based on the signals A[0123] 12 and A13, and which is designated by the signals /CS, /RAS=low levels and the signals /CAS, /WE=high levels. On this occasion, an address fed to the terminals A0 through A9 are accepted as a row address signal, while signals fed to the terminals A12 and A13 are accepted as a selection signal for the memory bank. The accepting operation is done in synchronism with the rising edge of the internal clock signal as stated above. For example, when the command is designated, a word line in the memory bank instructed by this command is selected, and memory cells connected to the word line are respectively conducted to corresponding complementary data lines.
  • (3) Column Address Read Command (Re) [0124]
  • This command is one which is necessary for starting a burst read operation and which gives the instruction of column address strobe, and it is designated by the signals /CS, /CAS=low levels and the signals /RAS, /WE=high levels. On this occasion, a column address fed to the terminals A[0125] 0 through A9 (in the case of the ×16-bit configuration) are accepted as a column address signal. The column address signal thus accepted is fed to the column address counter 207 as a burst start address.
  • In the burst read operation thus instructed, the memory bank and the word line included therein have been selected in the row address strobe/bank active command cycle beforehand, and the memory cells of the selected word line are sequentially selected and have data consecutively read therefrom in accordance with address signals which are outputted from the [0126] column address counter 207 in synchronism with the internal clock signal. The number of the data items which are consecutively read is the number indicated by the burst length. In addition, the data read from the output buffer 211 is started after that number of the cycles of the internal clock signal which is specified by the CAS latency.
  • (4) Column Address Write Command (Wr) [0127]
  • This is used as a command necessary for starting a burst write operation when the burst write is set in the [0128] mode register 10 as the aspect of a write operation, and as a command necessary for starting a single write operation when the single write is set in the mode register 10 as the aspect of the write operation. Further, this command gives the instruction of column address strobe in the single write and the burst write.
  • This command is designated by the signals /CS, /CAS, /WE=low levels and the signal /RAS=high level, and an address fed to the terminals A[0129] 0 through A9 is accepted as a column address signal on this occasion. The column address signal thus accepted is fed to the column address counter 207 as a burst start address in the burst write. The steps of the burst write operation thus instructed are executed similarly to those of the burst read operation. However, the CAS latency is not involved in the write operation, and the acceptance of write data is started from the column address/write command cycle.
  • (5) Precharge Command (Pr) [0130]
  • This is used as a command which starts a precharge operation for the memory bank selected in accordance with the signals A[0131] 12 and A13, and which is designated by the signals /CS, /RAS, /WE=low levels and the signal /CAS=high level.
  • (6) Auto Refresh Command [0132]
  • This command is one which is required for starting auto refresh, and which is designated by the signals /CS, /RAS, /CAS=low levels and the signals /WE, CKE=high levels. [0133]
  • (7) Burst Stop-In-Full Page Command [0134]
  • This is a command necessary for stopping the burst operation of the full page in any of the memory banks, and is neglected for the burst operation of any other than the full page. This command is designated by the signals /CS, /WE=low levels and the signals /RAS, /CAS=high levels. [0135]
  • ([0136] 8) No-Operation Command (Nop)
  • This is a command which instructs the inhibition of any substantial operation, and which is designated by the signal /CS=low level and the high levels of the signals /RAS, /CAS, /WE. [0137]
  • In the SDRAM, in the case where the burst operation is being executed in one memory bank and where the row address strobe/bank active command is issued midway by indicating another memory bank, the operation of a row address system in the other memory bank is permitted without exerting any influence on the operation in one memory bank under execution. For example, the SDRAM includes means for holding therein data, addresses and control signals fed from outside, and the held contents, particularly the addresses and the control signals, are retained every memory bank though not especially restricted. Alternatively, data items which correspond to one word line in the memory block selected by the row address strobe/bank active command cycle are retained in the latch/[0138] register 213 for the read operation in advance of a column-system operation.
  • Accordingly, unless the data items D[0139] 0-D15 collide at the data input/output terminals composed of 16 bits by way of example, the precharge command and the row address strobe/bank active command can be issued during the execution of a command whose processing has not been ended, and to any memory bank different from the memory bank which is processed by the command under execution, whereby the internal operations of the different memory bank are started beforehand. As explained above, the SDRAM of this embodiment is subjected to memory accesses in 16-bit units, it has about 4M addresses based on the addresses of the signals A0 through A11, and it is constructed of the four memory banks. In total, therefore, it has the storage capacity of, e.g., about 256 Mbits (4M×4 banks×16 bits).
  • Functional effects attained by the foregoing embodiments are as follows: [0140]
  • (1) Two memory chips each being subjected to memory accesses in 2-bit units are assembled into a structure in which they are stacked in a state where their back surfaces are placed one over the other, so as to make memory accesses in 4-bit units, thereby to bring forth the effect that a semiconductor storage device which is well adapted for use can be obtained while employing a thin package. [0141]
  • (2) The two memory chips are placed one over the other so as to hold their back surfaces in touch with each other, whereby a sealing resin on the back surfaces is dispensed with for the purpose of sealing and can be excluded to the utmost, and hence, the effect is brought forth that the thinning of a package can be realized. [0142]
  • (3) Each of the memory chips is divided into a plurality of memory banks, and when the memory chips have been placed one over the other so as to hold their back surfaces in touch, the positions of the memory banks to which the same address is assigned are made different on the back surfaces of the memory chips, thereby to bring forth the effect that places to generate heat can be dispersed. [0143]
  • (4) The two memory chips are placed one over the other so that their front surfaces may be opposite to each other, and they are shaped so that the sealing resin may be in touch with, at least, parts of the front surfaces held between the two memory chips, thereby to bring forth the effect that the thinning of a package can be realized. [0144]
  • (5) Each of the two memory chips is provided with leads of tape state on its front surface thereof, and at least, the front surface part is in touch with the sealing resin except the back surface, thereby to bring forth the effect that the structure can be thinned in spite of stacking the two memory chips. [0145]
  • (6) The thickness of the semiconductor storage device which includes the two memory chips held in the stacked structure is made equal to or less than the thickness of a semiconductor storage device whose storage capacity is half of or equal to the storage capacity of the former semiconductor storage device, and which has a single memory chip, thereby to bring forth the effect that the former semiconductor storage device is permitted to replace the existing (universal) semiconductor storage device. [0146]
  • (7) The memory chips have a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by employing wiring means or bonding wire and by setting signal transmission paths and/or signal levels of the memory chips, whereby many kinds of memory chips can be formed by an identical process, and hence, the effect is brought forth that mass production can be achieved. [0147]
  • (8) The plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units, thereby to bring forth the effects that semiconductor storage devices of 4-, 8-, 16- and 32-bit configurations which are generally and extensively employed and which include the stacked structure of the two chips can be realized on the basis of substantially one kind of memory chip, and that a storage capacity which is, at its maximum, double the storage capacity of one memory chip can be achieved. [0148]
  • (9) The semiconductor storage device is formed with leads so as to extend from both side surfaces of a square resin-sealing package stretching in the longitudinal direction thereof; the 4 bits are formed by data terminals which consist of a first terminal through a fourth terminal; a pair of the first and second terminals, and a pair of the third and fourth terminals are separately disposed at positions symmetric with respect to a center line parallel to the longitudinal direction, on both sides of the square resin-sealing package; and electrodes corresponding respectively to the data terminals of 2 bits are electrically connected to the first terminal and the second terminal in each of the two memory chips; thereby to bring forth the effect that the data terminals of each of the two memory chips can be held in correspondence with the four data terminals in isolated fashion. [0149]
  • (10) A plurality of semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure a state where a sealing resin is not in touch with the back surfaces of the two memory chips, so as to make memory accesses in 4-bit units, are disposed on a square mounting circuit board which is formed with electrodes along one latus thereof, thereby to bring forth the effect that a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging can be obtained. [0150]
  • (11) A plurality of such memory modules are used in a state where the electrodes are respectively plugged into a plurality of connectors which are arranged in parallel on a mother board, thereby to bring forth the effect that an increase in a storage capacity per unit volume and high-density packaging can be realized. [0151]
  • (12) The two memory chips are placed one over the other so as to hold their back surfaces in touch with each other, and at least their front surfaces are in touch with the sealing resin, thereby to bring forth the effects that replacement with a universal semiconductor storage device can be realized, and that an increase in a storage capacity per unit volume and high-density packaging can be realized. [0152]
  • (13) The two memory chips are placed one over the other so as to hold their front surfaces in touch with each other, and the sealing resin is formed so as to be in touch with, at least the front surfaces of the two memory chips, thereby to bring forth the effects that replacement with a universal semiconductor storage device can be realized, and that an increase in a storage capacity per unit volume and high-density packaging can be realized. [0153]
  • (14) The thickness of the two memory chips held in the stacked structure is made equal to or less than the thickness of a single universal semiconductor storage device whose storage capacity is half of or equal to the total storage capacity of the two memory chips, thereby to bring forth the effects that replacement with a memory module employing the universal semiconductor storage device can be realized, and that an increase in the storage capacity and high-density packaging can be realized. [0154]
  • (15) The semiconductor storage devices are provided with a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by setting voltages that are applied to external terminals, thereby to bring forth the effect that the bit configurations of data terminals can be selected in mounting the semiconductor storage devices on a memory module. [0155]
  • (16) The plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units, thereby to bring forth the effect that memory modules of 4-, 8-, 16- and 32-bit configurations which are generally and extensively employed and which include the stacked structure of the two chips can be obtained. [0156]
  • Although the inventions made by the inventors have been concretely described above on the basis of the embodiments, it is a matter of course that the invention of the present application is not restricted to the foregoing embodiments, but that it is variously alterable within a scope not departing from the purport thereof. For example, in the dynamic RAM shown in FIG. 12, the constructions of the memory array, subarray and sub word driver can adopt various aspects of performance, and the input/output interface of the dynamic RAM can adopt various aspects of performance such as ones suited to LAN bus specifications etc. other than synchronous specifications. The word lines may conform to a word shunt scheme other than the hierarchical word line scheme as stated before. [0157]
  • The two memory chips may be constructed from the combination of so-called “partial chips” in each of which a half storage area is validated. More specifically, the two memory chips, in each of which defects are partly existent and in each of which memory accesses can be made only to the half storage area, may be combined into the stacked structure as stated before, thereby to construct a single semiconductor storage device. In this case, when a non-defective semiconductor storage device constructed of a single memory chip is combined with the two memory chips in each of which the memory accesses can be made only to the half storage area, a semiconductor storage device which is the same as the single memory chip as externally viewed can be constructed. [0158]
  • The facility of making the memory accesses in 2-bit units can be effectively utilized even in the case stated above where the same semiconductor storage device as the memory chip when externally viewed is constructed by combining the two memory chips in each of which the memory accesses can be made only to the half storage area. The reason therefor is that the two memory chips whose storage areas having the same addresses assigned thereto are valid in the valid semiconductor storage device can be combined to operate as the semiconductor storage device in which memory accesses in 4-bit units can be made. Incidentally, if memory accesses in 8-bit or 16-bit units are necessary, each of the two memory chips may have a 4-bit configuration or 8-bit configuration. [0159]
  • Conversely, in the case where the two memory chips whose storage areas having different addresses assigned thereto are valid are combined in the valid semiconductor storage device, memory accesses are made to either of the two memory chips by address signals, so that memory accesses may be made in plural-bit units such as 4-bit, 8-bit or 16-bit units corresponding to the bus width of a memory module or the like to which the semiconductor storage device is connected, unlike the memory accesses in the 2-bit units. [0160]
  • Apart from the DRAM explained before, the semiconductor storage device may be a static RAM or a read-only memory such as EPROM or EEPROM. This invention can be extensively utilized for various semiconductor storage devices and memory modules each of which has a stacked structure. [0161]
  • An effect which is attained by a typical one of inventions disclosed in the present application is briefly explained as follows: Two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, whereby a semiconductor storage device which is well adapted for use can be obtained while a thin package is employed. [0162]
  • An effect which is attained by another typical one of the inventions disclosed in the present application is briefly explained as follows: A plurality of semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are assembled into a stacked structure by placing their back surfaces one over the other, so as to make memory accesses in 4-bit units, are installed on a mounting circuit board which is square and which is formed with connector electrodes along one latus thereof, thereby to attain the effect that a memory module which is capable of increasing the storage capacity per unit volume and high-density packaging can be obtained. [0163]

Claims (25)

We claim:
1. A semiconductor storage device comprising two memory chips each being subjected to memory accesses in 2-bit units, sealed in a state where their back surfaces are opposite to each other, so as to make memory accesses in 4-bit units.
2. A semiconductor storage device according to claim 1, wherein said two memory chips are placed one over the other so as to hold their back surfaces in touch with each other, and that, at least front surfaces of said two memory chips are in touch with a sealing resin.
3. A semiconductor storage device according to claim 2, wherein each of said memory chips is divided into a plurality of memory banks, and that, when said memory chips have been placed one over the other so as to hold their back surfaces in touch, positions of the memory banks to which the same address is assigned are made different on said back surfaces of said memory chips.
4. A semiconductor storage device according to any of claims 1 through 3, wherein a thickness of said semiconductor storage device in which said two memory chips are stacked and sealed is equal to or less than a thickness of a semiconductor storage device whose storage capacity is half of or equal to a storage capacity of the former semiconductor storage device, and in which a single memory chip is sealed.
5. A semiconductor storage device according to any of claims 1 through 4, wherein the stacked memory chips have a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by employing wiring means or bonding wire and by setting signal transmission paths or signal levels of said stacked memory chips.
6. A semiconductor storage device according to claim 5, wherein the plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units.
7. A semiconductor storage device according to claim 2 or 3,
wherein said semiconductor storage device is formed with leads so as to extend from both side surfaces of a square resin-sealing package in a longitudinal direction thereof,
wherein the 4 bits correspond to leads for data consisting of a first lead through a fourth lead,
wherein a pair of the first and second leads, and a pair of the third and fourth leads are separately disposed at positions symmetric with respect to a center line parallel to the longitudinal direction, on both sides of said square resin-sealing package and
wherein that electrodes corresponding respectively to data terminals of 2 bits are electrically connected to said first lead and second lead in one of said two memory chips, while electrodes corresponding respectively to data terminals of 2 bits are electrically connected to said third lead and fourth lead in the other memory chip.
8. A semiconductor storage device comprising:
two memory chips each being subjected to memory accesses in 2-bit units, assembled into a stacked structure so as to make memory accesses in 4-bit units,
wherein each of said two memory chips being provided with leads of tape state on a front surface thereof, and, at least, the front surface part lying in touch with a sealing resin except a back surface of said each memory chip.
9. A semiconductor storage device according to claim 8, wherein the thickness of said semiconductor storage device made into said stacked structure is equal to or less than the thickness of a semiconductor storage device whose storage capacity is half of or equal to the storage capacity of the former semiconductor storage device, and in which a single memory chip is included.
10. A semiconductor storage device according to claim 8, wherein said each memory chip has a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by employing wiring means or bonding wire and by setting signal transmission paths or signal levels of said each memory chip.
11. A semiconductor storage device according to claim 10, wherein the plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units.
12. A semiconductor storage device comprising:
two memory chips sealed by a sealing member in a state where their back surfaces are placed one over the other,
wherein each of said two memory chips are divided into a plurality of memory areas,
wherein said each memory chip is subjected to memory accesses to activated ones of said memory areas in 2-bit units, thereby to subject said two memory chips to memory accesses in 4-bit units, and
wherein positions of the activated memory areas are different on the surfaces of said two memory chips placed on each other.
13. A semiconductor storage device comprising:
first and second memory chips;
a sealing member which seals said first and second memory chips in a state where their back surfaces are placed one over the other;
first and second leads which extend from a first latus of said sealing member to the inside and outside said sealing member; and
third and fourth leads which extend from a second latus of said sealing member opposite to said first latus, to the inside and outside said sealing member,
wherein each of said first and second memory chips includes first, second, third and fourth terminals for outputting data, which are arranged along a straight line being parallel to longer latera of said each memory chip and passing through the central parts of shorter latera thereof, on a circuit forming surface of said each memory chip,
wherein the second terminal is arranged at a position nearer to the first terminal between said first terminal and third terminal,
wherein said third terminal is arranged at a position nearer to the fourth terminal between said second terminal and fourth terminal,
wherein the first lead and the fourth lead are arranged at positions which oppose with said straight line held therebetween,
wherein the second lead and the third lead are arranged at positions which oppose with said straight line held therebetween and
wherein each of said first and second memory chips outputs data in 2-bit units by employing either of the first and second terminals, and either of the third and fourth terminals, thereby to output data from the first, second, third and fourth leads in 4-bit units.
14. A semiconductor storage device according to claim 13,
wherein either of the first and second terminals of the first memory chip and either of the first and second terminals of the second memory chip are selectively connected to said first lead and said fourth lead through pieces of wire, respectively and
wherein either of the third and fourth terminals of said first memory chip and either of the third and fourth terminals of said second memory chip are selectively connected to said second lead and said third lead through pieces of wire, respectively.
15. A semiconductor storage device according to claim 14, wherein each of said first and second memory chips has a construction capable of outputting data in 4-bit units, and that the data of 4 bits are derived from said first, second, third and fourth terminals.
16. A semiconductor storage device comprising:
first and second memory chips;
a sealing member which seals said first and second memory chips in a state where their back surfaces are placed one over the other;
a plurality of leads for addresses, which extend inside and outside said sealing member; and
a plurality of leads for data, which extend inside and outside said sealing member,
wherein each of said leads for addresses is bifurcated within said sealing member, bifurcated leads being respectively extended over front surfaces of said first and second memory chips,
wherein each of said leads for data is extended over the front surface of at least one of said first and second memory chips,
wherein each of said first and second memory chips includes a plurality of address terminals and a plurality of data terminals on its front surface,
wherein the corresponding address terminals of said first and second memory chips are connected in common with the bifurcated leads for addresses through pieces of wire, respectively,
wherein said data terminals on the first memory chip and said data terminals on the second memory chip are isolated and connected to the corresponding leads for data through pieces of wire and
wherein data are outputted in 2-bit units from each of said first and second memory chips, thereby to output data from said leads for data in 4-bit units.
17. A semiconductor storage device according to claim 16, wherein said leads for data are asymmetric as to their lengths extending toward said first and second memory chips.
18. A semiconductor storage device comprising a memory chip for which four sorts of data outputting in 2-bit units, 4-bit units, 8-bit units and 16-bit units can be selected by employing wiring or bonding wire and by setting signal transmission paths or signal levels of said memory chip.
19. A memory module comprising a plurality of semiconductor storage devices, in each of which two memory chips each being subjected to memory accesses in 2-bit units are sealed in a state where their back surfaces are opposite to each other, so as to make memory accesses in 4-bit units, are disposed on a square mounting circuit board which is formed with electrodes along one latus thereof.
20. A memory module according to claim 19, wherein said two memory chips are placed one over the other so as to hold their back surfaces in touch with each other, and that, at least front surfaces of said two memory chips are in touch with a sealing resin.
21. A memory module according to claim 19 or 20, wherein the thickness of said semiconductor storage device in which said two memory chips are stacked and sealed is equal to or less than the thickness of a semiconductor storage device whose storage capacity is half of or equal to a storage capacity of the former semiconductor storage device, and in which a single memory chip is sealed.
22. A memory module according to claim 19 or 20, wherein said semiconductor storage devices have a facility by which a plurality of sorts of data inputting/outputting in plural-bit units, including the memory accesses in 2-bit units, are realized by setting voltages that are applied to external terminals.
23. A memory module according to claim 22, wherein the plural-bit units are of four sorts; 2-bit units, 4-bit units, 8-bit units and 16-bit units.
24. A memory module according to any of claims 19 through 23, wherein a plurality of such memory modules are mounted in a state where their connectors are respectively plugged into a plurality of sockets which are juxtaposed on a circuit board.
25. A memory module comprising a plurality of semiconductor storage devices, wherein
two memory chips, in said semiconductor storage devices, each being subjected to memory accesses in 2-bit units are assembled into a stacked structure so as to make memory accesses in 4-bit units and
wherein each of said two memory chips is provided with tape-like leads on a front surface thereof, and that, at least the front surface part is in touch with a sealing resin except a back surface of said each memory chip.
US10/197,901 1999-08-06 2002-07-19 Semiconductor storage device having memory chips in a stacked structure Abandoned US20030075789A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10/197,901 US20030075789A1 (en) 1999-08-06 2002-07-19 Semiconductor storage device having memory chips in a stacked structure

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
JP11224122A JP2001053243A (en) 1999-08-06 1999-08-06 Semiconductor memory device and memory module
JP11-224122 1999-08-06
US09/608,149 US6426560B1 (en) 1999-08-06 2000-06-30 Semiconductor device and memory module
US09/666,063 US6566760B1 (en) 1999-08-06 2000-09-19 Semiconductor storage device having memory chips in a stacked structure
US10/197,901 US20030075789A1 (en) 1999-08-06 2002-07-19 Semiconductor storage device having memory chips in a stacked structure

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US09/666,063 Continuation US6566760B1 (en) 1999-08-06 2000-09-19 Semiconductor storage device having memory chips in a stacked structure

Publications (1)

Publication Number Publication Date
US20030075789A1 true US20030075789A1 (en) 2003-04-24

Family

ID=16808896

Family Applications (3)

Application Number Title Priority Date Filing Date
US09/608,149 Expired - Fee Related US6426560B1 (en) 1999-08-06 2000-06-30 Semiconductor device and memory module
US09/666,063 Expired - Fee Related US6566760B1 (en) 1999-08-06 2000-09-19 Semiconductor storage device having memory chips in a stacked structure
US10/197,901 Abandoned US20030075789A1 (en) 1999-08-06 2002-07-19 Semiconductor storage device having memory chips in a stacked structure

Family Applications Before (2)

Application Number Title Priority Date Filing Date
US09/608,149 Expired - Fee Related US6426560B1 (en) 1999-08-06 2000-06-30 Semiconductor device and memory module
US09/666,063 Expired - Fee Related US6566760B1 (en) 1999-08-06 2000-09-19 Semiconductor storage device having memory chips in a stacked structure

Country Status (6)

Country Link
US (3) US6426560B1 (en)
JP (1) JP2001053243A (en)
KR (1) KR20010070014A (en)
CN (1) CN1283871A (en)
SG (1) SG83799A1 (en)
TW (1) TW527714B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030211679A1 (en) * 2002-05-10 2003-11-13 Du-Eung Kim Flip chip interface circuit of a semiconductor memory device and method for interfacing a flip chip
US20040135264A1 (en) * 2000-11-28 2004-07-15 Takao Yamazaki Semiconductor device and method for packaging same
US20040136229A1 (en) * 2002-03-07 2004-07-15 Bhakta Jayesh R. Arrangement of integrated circuits in a memory module
US20050018495A1 (en) * 2004-01-29 2005-01-27 Netlist, Inc. Arrangement of integrated circuits in a memory module
US20060183052A1 (en) * 2002-09-13 2006-08-17 Hiroshi Deguchi Optical recording medium
WO2006130762A2 (en) * 2005-06-01 2006-12-07 Sanmina-Sci Corporation Capacity-expanding memory device
US20130271910A1 (en) * 2008-11-13 2013-10-17 Mosaid Technologies Incorporated System including a plurality of encapsulated semiconductor chips

Families Citing this family (95)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7102892B2 (en) * 2000-03-13 2006-09-05 Legacy Electronics, Inc. Modular integrated circuit chip carrier
US6713854B1 (en) * 2000-10-16 2004-03-30 Legacy Electronics, Inc Electronic circuit module with a carrier having a mounting pad array
US7247932B1 (en) 2000-05-19 2007-07-24 Megica Corporation Chip package with capacitor
JP2002025250A (en) * 2000-07-03 2002-01-25 Hitachi Ltd Semiconductor storage device
JP3744825B2 (en) * 2000-09-08 2006-02-15 セイコーエプソン株式会社 Semiconductor device
US7337522B2 (en) * 2000-10-16 2008-03-04 Legacy Electronics, Inc. Method and apparatus for fabricating a circuit board with a three dimensional surface mounted array of semiconductor chips
US6633948B1 (en) * 2000-10-20 2003-10-14 Sun Microsystems, Inc. Stackable dual mode (registered/unbuffered) memory interface cost reduction
EP1378152A4 (en) * 2001-03-14 2006-02-01 Legacy Electronics Inc A method and apparatus for fabricating a circuit board with a three dimensional surface mounted array of semiconductor chips
US6541856B2 (en) * 2001-06-06 2003-04-01 Micron Technology, Inc. Thermally enhanced high density semiconductor package
KR100426989B1 (en) * 2001-06-13 2004-04-13 삼성전자주식회사 Control signal providing method using package power pins and IC package structure therefore
US6627984B2 (en) * 2001-07-24 2003-09-30 Dense-Pac Microsystems, Inc. Chip stack with differing chip package types
US6618257B1 (en) * 2001-07-27 2003-09-09 Staktek Group, L.P. Wide data path stacking system and method
DE10149590B4 (en) * 2001-10-08 2004-11-18 Infineon Technologies Ag Semiconductor module with configurable data width of an output bus and housing arrangement with a semiconductor module
US20060255446A1 (en) 2001-10-26 2006-11-16 Staktek Group, L.P. Stacked modules and method
US6914324B2 (en) * 2001-10-26 2005-07-05 Staktek Group L.P. Memory expansion and chip scale stacking system and method
US7656678B2 (en) 2001-10-26 2010-02-02 Entorian Technologies, Lp Stacked module systems
US7026708B2 (en) * 2001-10-26 2006-04-11 Staktek Group L.P. Low profile chip scale stacking system and method
US6576992B1 (en) * 2001-10-26 2003-06-10 Staktek Group L.P. Chip scale stacking system and method
US20060012031A1 (en) * 2002-07-30 2006-01-19 Fernandez Elstan A Heat dissipation device for integrated circuits
KR100701380B1 (en) * 2002-12-30 2007-03-28 동부일렉트로닉스 주식회사 Heat spreading type package structure
JP2004296853A (en) 2003-03-27 2004-10-21 Seiko Epson Corp Semiconductor chip, semiconductor device and its manufacturing method, circuit board and electronic apparatus
JP2005011986A (en) * 2003-06-19 2005-01-13 Sanyo Electric Co Ltd Semiconductor device
JP4190961B2 (en) * 2003-06-26 2008-12-03 株式会社ルネサステクノロジ Multi-chip module
KR100626364B1 (en) * 2003-07-02 2006-09-20 삼성전자주식회사 Semiconductor Package Having Multi-Chips
JP4205553B2 (en) * 2003-11-06 2009-01-07 エルピーダメモリ株式会社 Memory module and memory system
JP2007066922A (en) * 2003-11-28 2007-03-15 Renesas Technology Corp Semiconductor integrated circuit device
US6933209B2 (en) * 2003-12-24 2005-08-23 Super Talent Electronics Inc. Stacking memory chips using flat lead-frame with breakaway insertion pins and pin-to-pin bridges
JP2005302951A (en) * 2004-04-09 2005-10-27 Toshiba Corp Semiconductor device package for power
JP2008502058A (en) * 2004-05-18 2008-01-24 シルバーブルック リサーチ ピーティワイ リミテッド Method and computer system for tracking security documents
JP2006004079A (en) * 2004-06-16 2006-01-05 Sony Corp Storage device
US7202554B1 (en) * 2004-08-19 2007-04-10 Amkor Technology, Inc. Semiconductor package and its manufacturing method
US7423885B2 (en) 2004-09-03 2008-09-09 Entorian Technologies, Lp Die module system
US7443023B2 (en) 2004-09-03 2008-10-28 Entorian Technologies, Lp High capacity thin module system
US7760513B2 (en) 2004-09-03 2010-07-20 Entorian Technologies Lp Modified core for circuit module system and method
KR100549871B1 (en) * 2004-10-22 2006-02-06 삼성전자주식회사 Memory semiconductor device determining operation mode by data pins, and operation mode determining method for the same
US7435097B2 (en) 2005-01-12 2008-10-14 Legacy Electronics, Inc. Radial circuit board, system, and methods
US20070029648A1 (en) * 2005-08-02 2007-02-08 Texas Instruments Incorporated Enhanced multi-die package
US7304382B2 (en) * 2006-01-11 2007-12-04 Staktek Group L.P. Managed memory component
US7508058B2 (en) * 2006-01-11 2009-03-24 Entorian Technologies, Lp Stacked integrated circuit module
JP4435756B2 (en) * 2006-05-31 2010-03-24 俊彦 水上 Semiconductor device
US7417310B2 (en) 2006-11-02 2008-08-26 Entorian Technologies, Lp Circuit module having force resistant construction
CN101281894B (en) * 2007-04-02 2012-01-18 欣兴电子股份有限公司 Semiconductor component bearing structure and splicing structure
JP2009026884A (en) 2007-07-18 2009-02-05 Elpida Memory Inc Circuit module and electric component
US7791918B2 (en) * 2007-09-27 2010-09-07 Intel Corporation Stack position location identification for memory stacked packages
US8399973B2 (en) * 2007-12-20 2013-03-19 Mosaid Technologies Incorporated Data storage and stackable configurations
US8513784B2 (en) * 2010-03-18 2013-08-20 Alpha & Omega Semiconductor Incorporated Multi-layer lead frame package and method of fabrication
US8581376B2 (en) * 2010-03-18 2013-11-12 Alpha & Omega Semiconductor Incorporated Stacked dual chip package and method of fabrication
US8564141B2 (en) * 2010-05-06 2013-10-22 SK Hynix Inc. Chip unit and stack package having the same
US8553420B2 (en) 2010-10-19 2013-10-08 Tessera, Inc. Enhanced stacked microelectronic assemblies with central contacts and improved thermal characteristics
US8952516B2 (en) 2011-04-21 2015-02-10 Tessera, Inc. Multiple die stacking for two or more die
US8970028B2 (en) 2011-12-29 2015-03-03 Invensas Corporation Embedded heat spreader for package with multiple microelectronic elements and face-down connection
US9013033B2 (en) 2011-04-21 2015-04-21 Tessera, Inc. Multiple die face-down stacking for two or more die
US8928153B2 (en) 2011-04-21 2015-01-06 Tessera, Inc. Flip-chip, face-up and face-down centerbond memory wirebond assemblies
US8633576B2 (en) * 2011-04-21 2014-01-21 Tessera, Inc. Stacked chip-on-board module with edge connector
US8436429B2 (en) 2011-05-29 2013-05-07 Alpha & Omega Semiconductor, Inc. Stacked power semiconductor device using dual lead frame and manufacturing method
US8502390B2 (en) 2011-07-12 2013-08-06 Tessera, Inc. De-skewed multi-die packages
US8513817B2 (en) 2011-07-12 2013-08-20 Invensas Corporation Memory module in a package
US8823165B2 (en) 2011-07-12 2014-09-02 Invensas Corporation Memory module in a package
US8405207B1 (en) 2011-10-03 2013-03-26 Invensas Corporation Stub minimization for wirebond assemblies without windows
TWI515864B (en) 2011-10-03 2016-01-01 英帆薩斯公司 Stub minimization with terminal grids offset from center of package
US8436457B2 (en) 2011-10-03 2013-05-07 Invensas Corporation Stub minimization for multi-die wirebond assemblies with parallel windows
US8610260B2 (en) 2011-10-03 2013-12-17 Invensas Corporation Stub minimization for assemblies without wirebonds to package substrate
WO2013052372A2 (en) 2011-10-03 2013-04-11 Invensas Corporation Stub minimization for multi-die wirebond assemblies with parallel windows
WO2013052080A1 (en) 2011-10-03 2013-04-11 Invensas Corporation Stub minimization for multi-die wirebond assemblies with orthogonal windows
US8659140B2 (en) 2011-10-03 2014-02-25 Invensas Corporation Stub minimization using duplicate sets of signal terminals in assemblies without wirebonds to package substrate
US8513813B2 (en) 2011-10-03 2013-08-20 Invensas Corporation Stub minimization using duplicate sets of terminals for wirebond assemblies without windows
US8441111B2 (en) 2011-10-03 2013-05-14 Invensas Corporation Stub minimization for multi-die wirebond assemblies with parallel windows
US8848391B2 (en) 2012-08-27 2014-09-30 Invensas Corporation Co-support component and microelectronic assembly
US8787034B2 (en) 2012-08-27 2014-07-22 Invensas Corporation Co-support system and microelectronic assembly
US8848392B2 (en) 2012-08-27 2014-09-30 Invensas Corporation Co-support module and microelectronic assembly
US9368477B2 (en) 2012-08-27 2016-06-14 Invensas Corporation Co-support circuit panel and microelectronic packages
US9070423B2 (en) 2013-06-11 2015-06-30 Invensas Corporation Single package dual channel memory with co-support
KR20150040058A (en) * 2013-10-04 2015-04-14 에스케이하이닉스 주식회사 Semiconductor integrated circuit
US9123555B2 (en) 2013-10-25 2015-09-01 Invensas Corporation Co-support for XFD packaging
TWI550823B (en) * 2014-04-10 2016-09-21 南茂科技股份有限公司 Chip package structure
US9281296B2 (en) 2014-07-31 2016-03-08 Invensas Corporation Die stacking techniques in BGA memory package for small footprint CPU and memory motherboard design
US9691437B2 (en) 2014-09-25 2017-06-27 Invensas Corporation Compact microelectronic assembly having reduced spacing between controller and memory packages
US9490195B1 (en) 2015-07-17 2016-11-08 Invensas Corporation Wafer-level flipped die stacks with leadframes or metal foil interconnects
US9825002B2 (en) 2015-07-17 2017-11-21 Invensas Corporation Flipped die stack
US9871019B2 (en) 2015-07-17 2018-01-16 Invensas Corporation Flipped die stack assemblies with leadframe interconnects
US9484080B1 (en) 2015-11-09 2016-11-01 Invensas Corporation High-bandwidth memory application with controlled impedance loading
US9508691B1 (en) 2015-12-16 2016-11-29 Invensas Corporation Flipped die stacks with multiple rows of leadframe interconnects
US10566310B2 (en) 2016-04-11 2020-02-18 Invensas Corporation Microelectronic packages having stacked die and wire bond interconnects
US9679613B1 (en) 2016-05-06 2017-06-13 Invensas Corporation TFD I/O partition for high-speed, high-density applications
US9595511B1 (en) 2016-05-12 2017-03-14 Invensas Corporation Microelectronic packages and assemblies with improved flyby signaling operation
US9728524B1 (en) 2016-06-30 2017-08-08 Invensas Corporation Enhanced density assembly having microelectronic packages mounted at substantial angle to board
US10528103B2 (en) * 2017-06-14 2020-01-07 Intel Corporation Die interconnect signal management devices and methods
JP6892360B2 (en) * 2017-09-19 2021-06-23 キオクシア株式会社 Semiconductor device
JP6847797B2 (en) * 2017-09-21 2021-03-24 キオクシア株式会社 Semiconductor storage device
KR102516833B1 (en) * 2017-12-28 2023-04-03 삼성전자주식회사 Memory apparatus and method for processing data the same
US10984874B1 (en) * 2019-11-13 2021-04-20 Sandisk Technologies Llc Differential dbus scheme for low-latency random read for NAND memories
US11183231B2 (en) * 2019-11-25 2021-11-23 Piecemakers Technology, Inc. Apparatus for enhancing prefetch access in memory module
CN111900155A (en) * 2020-08-19 2020-11-06 上海先方半导体有限公司 Modular packaging structure and method
US11755685B2 (en) 2020-09-30 2023-09-12 Piecemakers Technology, Inc. Apparatus for data processing in conjunction with memory array access
US11250904B1 (en) 2020-09-30 2022-02-15 Piecemakers Technology, Inc. DRAM with inter-section, page-data-copy scheme for low power and wide data access

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5357621A (en) * 1990-09-04 1994-10-18 Hewlett-Packard Company Serial architecture for memory module control
US5508565A (en) * 1992-12-18 1996-04-16 Fujitsu Limited Semiconductor device having a plurality of chips having identical circuit arrangement sealed in package
US5917242A (en) * 1996-05-20 1999-06-29 Micron Technology, Inc. Combination of semiconductor interconnect

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6020387A (en) 1983-07-15 1985-02-01 Nec Corp Memory circuit
JPH0666394B2 (en) 1983-12-16 1994-08-24 富士通株式会社 Semiconductor memory device
JPS60171687A (en) 1984-02-17 1985-09-05 Nec Corp Storage device
JPS62146495A (en) 1985-12-20 1987-06-30 Nec Corp Memory device
JPS62262291A (en) 1986-05-07 1987-11-14 Nec Corp Semiconductor read/write memory device
US5138438A (en) * 1987-06-24 1992-08-11 Akita Electronics Co. Ltd. Lead connections means for stacked tab packaged IC chips
EP0408779B1 (en) * 1989-07-18 1993-03-17 International Business Machines Corporation High density semiconductor memory module
US5147815A (en) * 1990-05-14 1992-09-15 Motorola, Inc. Method for fabricating a multichip semiconductor device having two interdigitated leadframes
JPH04222989A (en) 1990-12-26 1992-08-12 Hitachi Ltd Semiconductor integrated circuit
JPH0661379A (en) * 1992-08-07 1994-03-04 Fujitsu Ltd Semiconductor device
JP3299342B2 (en) * 1993-06-11 2002-07-08 株式会社日立製作所 Semiconductor memory module
JPH0730051A (en) * 1993-07-09 1995-01-31 Fujitsu Ltd Semiconductor device
US5513135A (en) * 1994-12-02 1996-04-30 International Business Machines Corporation Synchronous memory packaged in single/dual in-line memory module and method of fabrication
US5491612A (en) * 1995-02-21 1996-02-13 Fairchild Space And Defense Corporation Three-dimensional modular assembly of integrated circuits
JP2795315B2 (en) * 1996-05-16 1998-09-10 日本電気株式会社 Semiconductor device
JPH1040682A (en) * 1996-07-23 1998-02-13 Mitsubishi Electric Corp Semiconductor memory
US5802005A (en) * 1996-09-23 1998-09-01 Texas Instruments Incorporated Four bit pre-fetch sDRAM column select architecture

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5357621A (en) * 1990-09-04 1994-10-18 Hewlett-Packard Company Serial architecture for memory module control
US5508565A (en) * 1992-12-18 1996-04-16 Fujitsu Limited Semiconductor device having a plurality of chips having identical circuit arrangement sealed in package
US5917242A (en) * 1996-05-20 1999-06-29 Micron Technology, Inc. Combination of semiconductor interconnect

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6949815B2 (en) * 2000-11-28 2005-09-27 Nec Corporation Semiconductor device with decoupling capacitors mounted on conductors
US20040135264A1 (en) * 2000-11-28 2004-07-15 Takao Yamazaki Semiconductor device and method for packaging same
US6930900B2 (en) * 2002-03-07 2005-08-16 Netlist, Inc. Arrangement of integrated circuits in a memory module
US20040184301A1 (en) * 2002-03-07 2004-09-23 Bhakta Jayesh R. Arrangement of integrated circuits in a memory module
US20040184300A1 (en) * 2002-03-07 2004-09-23 Bhakta Jayesh R. Arrangement of integrated circuits in a memory module
US20040184299A1 (en) * 2002-03-07 2004-09-23 Bhakta Jayesh R. Arrangement of integrated circuits in a memory module
US6873534B2 (en) 2002-03-07 2005-03-29 Netlist, Inc. Arrangement of integrated circuits in a memory module
US6930903B2 (en) * 2002-03-07 2005-08-16 Netlist, Inc. Arrangement of integrated circuits in a memory module
US20040136229A1 (en) * 2002-03-07 2004-07-15 Bhakta Jayesh R. Arrangement of integrated circuits in a memory module
US7005748B2 (en) * 2002-05-10 2006-02-28 Samsung Electronics Co., Ltd. Flip chip interface circuit of a semiconductor memory device
US20030211679A1 (en) * 2002-05-10 2003-11-13 Du-Eung Kim Flip chip interface circuit of a semiconductor memory device and method for interfacing a flip chip
US20060183052A1 (en) * 2002-09-13 2006-08-17 Hiroshi Deguchi Optical recording medium
US20050018495A1 (en) * 2004-01-29 2005-01-27 Netlist, Inc. Arrangement of integrated circuits in a memory module
WO2006130762A2 (en) * 2005-06-01 2006-12-07 Sanmina-Sci Corporation Capacity-expanding memory device
WO2006130762A3 (en) * 2005-06-01 2009-04-30 Sanmina Sci Corp Capacity-expanding memory device
US20130271910A1 (en) * 2008-11-13 2013-10-17 Mosaid Technologies Incorporated System including a plurality of encapsulated semiconductor chips
US8908378B2 (en) * 2008-11-13 2014-12-09 Conversant Intellectual Property Management Inc. System including a plurality of encapsulated semiconductor chips

Also Published As

Publication number Publication date
CN1283871A (en) 2001-02-14
TW527714B (en) 2003-04-11
US6426560B1 (en) 2002-07-30
KR20010070014A (en) 2001-07-25
US6566760B1 (en) 2003-05-20
JP2001053243A (en) 2001-02-23
SG83799A1 (en) 2001-10-16

Similar Documents

Publication Publication Date Title
US6426560B1 (en) Semiconductor device and memory module
JP3299342B2 (en) Semiconductor memory module
US6873563B2 (en) Semiconductor circuit device adaptable to plurality of types of packages
US7872936B2 (en) System and method for packaged memory
JP4632107B2 (en) Semiconductor memory device
KR100485547B1 (en) Semiconductor memory device adaptable to various types of packages
US7638871B2 (en) Semiconductor device
JP2004221215A (en) Semiconductor device
US6215720B1 (en) High speed operable semiconductor memory device with memory blocks arranged about the center
JP2006277870A (en) Semiconductor storage device
JP3970396B2 (en) Semiconductor memory device
KR20000017466A (en) Semiconductor integrated circuit device
JP2004158892A (en) Memory module
CN113889160A (en) Memory subsystem loaded by multi-register clock driver
US20230410895A1 (en) Memory array architecture having sensing circuitry to drive two matrices for higher array efficiency
JP3346351B2 (en) Semiconductor storage device
US8238133B2 (en) Semiconductor device with a selection circuit selecting a specific pad
JPH0358544B2 (en)
KR100306967B1 (en) Arrangement of data input/output circuits for use in a semiconductor memory integrated circuit device
JPH02224266A (en) Semiconductor device

Legal Events

Date Code Title Description
STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION