Memory Cell Having Hot-Hole Injection Erase Mode

1. Field of the Invention

The present invention relates to programmable read only memory (PROM) integrated circuit devices and, in particular, to a single transistor cell structure for electrically erasable PROM integrated circuit devices and the method for performing erasure of data stored in such a cell.

2. Description of the Related Art

In current integrated circuit technology, there are two basic forms of erasable PROM (EPROM) devices: EPROMS on which the programmer uses impinging ultraviolet light (UV) to perform the erase function, and EEPROMS which use an electrical potential to perform erasure of the memory cells.

In each type of cell, data is represented by an electrical potential stored during the programming stage by charge injection upon a floating gate, i.e., a semiconductor region separated from the substrate and other layers in each cell by thin dielectric layers.

In the UV EPROM, the charge is removed when incident photons impart sufficient energy to stored electrons to excite them to a state where they can migrate off the gate. This requires the package to have a UV transparent window over the die, generally a quartz lid which is a high-cost item.

In an EEPROM, two general schemes exist for performing the erase function. One type of device makes use of reversible electron tunneling beneath the floating gate. By putting a relatively high potential on a superposing control gate which is opposite to the potential used during the programming mode, the stored charge is driven off the floating gate through a narrow tunnel oxide region separating the floating gate from the substrate. The other type is a triple polysilicon layer structure; the first layer provides a ground reference plate, the second is the floating gate, and the third is a program/erase control line. A second transistor, a select transistor, is used to select the bit for erasing and a voltage pulse is put on a bit line in order to draw the electrons off the floating gate. A typical circuit of the prior art for an EEPROM is shown in FIG. 1. Each of these devices exhibits inherent disadvantages.

As shown in FIG. 1(a), in both EEPROM devices, the dielectric layer, typically silicon dioxide, through which the electrons tunnel is relatively thin. A typical thin oxide in a tunneling structure may be only 110 Angstroms. The thin oxide facilitates the migration of the electrons at a low driving potential during erasure, generally about 25 volts. However, most devices have a normal operating electrical bias potential of only 5 volts or less. The relatively high voltage applied during the erase cycle results in a breakdown of these thin oxide layers. Degradation of the tunnel oxide induces leakages that affect the data retention capability of the cell. After approximately 10,000 erasures, an EEPROM is, in effect, worn out because of ultimate breakdown of the tunnel oxide.

Moreover, another disadvantage is that EEPROM devices require a "select" device to form the cell, i.e., two devices to store one bit of datum. This makes it hard to achieve high packing density desirable for small but powerful data processing machines.

In UV EPROM devices, erase time is dependent upon the erase rate factors, such as the UV spectral density and intensity, and the threshold voltage set by the sense amplifier used in conjunction with the EPROM cell array. FIG. 2 depicts the carrier action during an ERASE mode. In general, the erase process is slow, requiring minutes in a system generally operating in the nanoseconds realm.

It is an object of the present invention to provide an EEPROM cell which is amenable to high packing density on an integrated circuit chip.

It is another object of the present invention to provide an EEPROM cell which can be erased by hot-hole injection onto the floating gate region of the cell structure.

It is a further object of the present invention to provide a method of erasing an EEPROM cell.

It is yet another object of the present invention to provide a rapid, nondestructive method for erasing an EEPROM cell.

In its broad aspect, the present invention is a single floating gate transistor which is capable of being biased in a relatively nondestructive snap-back mode. Such biasing results in erasure of any charge, representing a data bit, stored on the floating gate. In another broad aspect, the invention presents a method of erasing a single floating gate semiconductor memory device by hot-hole injection onto the floating gate.

An advantage of the present invention is that the cell comprises only a single transistor, viz., the elimination of a select transistor and, hence, the saving of approximately 50% of the area on the die needed for each storage cell is effected.

Yet another advantage of the present invention is the diminishment of breakdown effect on the thin oxide layers in the device during an erase mode cycle.

Still another advantage of the present invention is that it comprises a "short-channel" device which allows denser packing on a chip.

Another advantage of the present invention is that the need for a high-cost, slow UV erase process required for UV EPROM cell arrays is virtually eliminated.

Other objects, features and advantages of the present invention will become apparent upon consideration of the following detailed description and the accompanying drawings, in which like reference designations represent like features throughout the FIGURES.

FIG. 1 is an electrical schematic diagram of a prior art EEPROM cell.

FIG. 1(a) represents the cross-section of a prior art EEPROM cell as depicted in FIG. 1.

FIG. 2 represents the cross-section of a prior art UV EPROM cell during an ERASE process.

FIG. 3 is a cross-sectional drawing of the single transistor EEPROM cell according to the present invention.

FIG. 4(a) is a cross-sectional drawing of the present invention as shown in FIG. 3, depicting exemplary carrier action during the WRITE mode of operation.

FIG. 4(b) shows graphical representations of the gate and drain bias voltages used during the WRITE mode as shown in FIG. 4(a).

FIG. 5(a) is a cross-sectional drawing of the present invention as shown in FIG. 3, depicting exemplary carrier action during the ERASE mode of operation.

FIG. 5(b) shows graphical representations of the gate and drain bias voltages used during the ERASE mode as shown in FIG. 5(a).

FIG. 6 shows a graphical representation of the drain current versus the gate-to-source voltage during programming and erasing modes.

FIG. 7 shows a graphical representation of the drain source current versus the biasing voltage in the snap-back mode.

The drawings referred to in this description should be understood not to be drawn to scale except if specifically noted. Moreover, the drawings are intended to illustrate only a component of an integrated circuit fabricated in accordance with the present invention.

Reference is made now, in detail, to a specific embodiment of the present invention which illustrates the best mode presently contemplated for practicing the invention. Alternative embodiments are also briefly described as applicable.

In the main, FIG. 3 shows a cross-sectional view of a PROM cell transistor fabricated in a manner which allows operation in the "snap-back," or "snap-recovery," mode without causing any significant damage to the oxide layers.

It should be recognized that many publications describe the details of common techniques used in the fabrication process of integrated circuit components. See, e.g., Semi-conductor & Integrated Circuit Fabrication Techniques, Reston Publishing Co., Inc., copyright 1979 by the Fairchild Corporation. Those techniques can be generally employed in the fabrication of the structure of the present invention. Moreover, the individual steps of such a process can be performed using commercially available integrated circuit fabrication machines. As specifically necessary to an understanding of the present invention, exemplary technical data are set forth based upon current technology. Future developments in this art may call for appropriate adjustments as would be obvious to one skilled in the art.

A wafer substrate 12 having a first conductivity type, e.g., p type, is generally formed of crystalline silicon, or an epitaxially formed silicon layer. The substrate 12 in a device constructed in accordance with the present invention was doped with boron ions to a concentration of approximately 7.5.times.10.sup.14 /cm.sup.3.

A region 16 of the substrate 12 is doped to have a second conductivity type, e.g., n+ type, adjacent to the surface 14 of the substrate 12, and forms the source region 16 of the MOSFET EEPROM cell 10.

A doped region 18 adjacent to the surtace 14, having the same second conductivity type as the source region 16, forms the drain region 18 of the cell 10.

The region of the substrate 12 lying between the source 16 and the drain 18 is a short channel 20.

In an exemplary device constructed according to the present invention, the dimensions (length.times.width.times.depth) of the source region 16 were approximately 1.5.times.1.5.times.0.2 microns. The doping of the source region 16 was approximately 10.sup.22 /cm.sup.3. The drain dimensions were approximately 1.5.times.1.5.times.0.2 microns. The doping of the drain region 18 was approximately 10.sup.22 /cm.sup.3. The channel dimensions were approximately 1.3.times.1.7. Submicron channel lengths are contemplated as further improvements in fabrication technology occur.

Dielectric regions 22, 22', used to isolate the various regions of the integrated circuit, are superjacent to the substrate surface 14. Typically, these regions are formed of an insulator or dielectric material such as silicon dioxide, also commonly referred to as simply "oxide." These regions 22, 22' are formed to have apertures 23 formed therein such that electrical contact to the cell can be established.

Superposing the channel 20 and a portion of the source 16 and the drain 18 is a floating gate 24 used to store a charge which represents a bit of datum. Polysilicon or another semiconductive or conductive material is generally used to form the floating gate. The floating gate 24 is separated from the substrate surface 14 by a layer 26 of the oxide region 22. In the exemplary device, dimensions of the floating gate were approximately 2.9.times.1.3.times.0.25.

Superposing the floating gate 24 is a control gate 28 having like dimensions. The control gate 28 can also be fabricated from polysilicon. The control gate 28 is isolated from the floating gate 24 by another layer 30 of the oxide region 22.

Interconnection conductive regions, generally metal layers, 32, 34, 36, electrically couple the source 16, control gate 28, and drain 18, respectively, in accordance with the operational requirements of the cell 10.

The programming mechanisms involved, viz., the writing and erasing of an electrical charge representing a data bit from the floating gate 24, are channel hot-carrier techniques.

Referring to FIG. 4(a), the WRITE mode, viz., the writing of a data bit into the cell, is accomplished through channel hot-electron injection onto the floating gate. In this regard, the present invention's approach to writing data into a cell is similar to that used for prior art devices. Briefly, a potential difference is created between the drain and gate as shown in FIG. 4(b), turning the device 10 on as per the bias voltages, Vg and Vd. Hot-electrons in the channel region 20 can surmount the energy barrier existing at the substrate surface 14 between the substrate 12 and the thin oxide layer 26. Once in the oxide layer 26, these electrons are pulled to the floating gate 24, driven by the potential difference electric field across the oxide layer 26. The floating gate 24, being fabricated of a dielectric material such as polysilicon, will trap these electrons which causes an increase of the threshold voltage, Vth (gate-to-source voltage required to turn the device on). The floating gate 24 thereby stores charge, i.e., a data bit, for an extremely long period.

UV or select transistor forced ERASE techniques of prior art devices draw the electrons from the floating gate 24. To overcome the aforesaid disadvantages of the prior art, the present invention achieves an ERASED state by hot-hole injection onto the floating gate 24. This will neutralize the negative charges stored there. This operation will, therefore, shift the Vth back to the value which existed prior to the data holding state.

The ERASE mode, as depicted in FIGS. 5(a) and 5(b), is achieved by the device 10 design to operate in an essentially nondestructive mode known in the art as the "snap" or "snapback" mode. This mode provides the aforementioned operational advantages over prior art devices.

To erase a cell, the drain bias voltage, Vd, is set to a steady-state level for approximately 1 second, maintaining the device 10 in an OFF condition. In an exemplary device, Vd was set to approximately 7 volts. The high drain-to-source potential, Vds, will create a large field about the drain region 18. A relatively high current flow then will exist from the source 16 to the drain 18.

A pulse, Vg, of approximately 13 volts, of approximately 1 millisecond duration, is applied to the control gate 28 during this period. The relatively short source-to-drain channel dimension encourages a high source-to drain current flow. The substrate therefore experiences a potential drop and the source-to-substrate junction becomes forward biased. Thus, after the gate-pulse triggering, the device 10 is in the positive feedback mode. The drain current is maintained at a high level, as shown in FIG. 7.

Biasing the device 10 in this manner nas forced the cell into the snap-back condition. The relatively large number of holes generated by impact ionization near the drain region 18 in the channel region 20 flowing toward the source 16 transfer to the floating gate 24. Since the floating gate 24 is at a relatively low potential, it is in a favorable condition to accept the hole injection. Hence, the cell is rapidly erased.

A plot of the drain current versus Vgs characteristics in the linear region for the WRITE and ERASE modes using the aforesaid programming techniques is shown in FIG. 6. The exemplary device 10 exhibits a large Vth window, approximately 5 to 7 volts, between the WRITE and ERASE modes. This value is dependent upon the geometry of the cell, such as the channel length, junction depth and gate oxide thickness. Hence, the endurance characteristics of the present invention are enhanced by proper scaling in accordance with the current state of the art in integrated circuit fabrication techniques.

The biasing potentials referred to above can be generated by of a number of integrated circuit devices as are commonly known in the art.

The foregoing description of the preferred embodiment of the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations will be apparent to practitioners skilled in this art. It is possible that the invention may be practiced in other technologies, such as with a P-channel process development. Relative device sizes and doping concentrations are dependent upon the state of the art in fabrication technologies. The embodiment was chosen and described in order to best explain the principles of the invention and its practical application, to thereby enable others skilled in the art to understand the invention for various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.