Semiconductor device manufacturing: process – Having magnetic or ferroelectric component
Reexamination Certificate
2000-09-25
2003-01-14
Tsai, Jey (Department: 2812)
Semiconductor device manufacturing: process
Having magnetic or ferroelectric component
C438S240000, C438S253000
Reexamination Certificate
active
06506613
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for manufacturing a semiconductor device, and particularly to a method for manufacturing a semiconductor device having storage capacitors.
2. Description of the Background Art
A storage capacitor in a semiconductor storage device must be adapted to data holding of one second or more and read/write cycle of 100 MHz or more, so it is required to allow a wide dynamic range from under 1 Hz to over 100 MHz.
Now, the silicon oxide and silicon nitride films which have been conventionally used as dielectric materials of capacitors are amorphous materials, so that it has been relatively easy to ensure the dynamic range when capacitors are made using these materials. This is because control of in-film defects is relatively easy.
However, the recent reduction in size and increase in integration degree of semiconductor devices are now making it difficult to obtain sufficient electrostatic capacitance using these materials having lower dielectric constants, so that the tendency now is turning toward use of polycrystalline dielectrics having higher dielectric constants, such as ditantalum pentaoxide (Ta
2
O
5
) and BST (barium strontium titanate). Ta
2
O
5
is usually used in polycrystalline form to obtain higher dielectric constant, though it may be used also in amorphous form.
However, polycrystalline materials generally suffer larger dielectric loss because of the interfacial polarization or orientation polarization due to grain boundary, interface conditions, etc., which applies also to Ta
2
O
5
, BST and the like.
The dielectric loss resulting from the interfacial polarization and orientation polarization will now be described referring to
FIGS. 11 and 12
.
FIG. 11
shows an equivalent circuit representing a real capacitor using ideal capacitor and ideal resistor. The ideal capacitor is a capacitor in which resistance and inductance of electrodes and interconnection can be neglected, there is no capacitance variation resulting from the effects of applied voltage, temperature, humidity, pressure, etc., and the time required for polarization is infinitely close to zero; it is thus a capacitor with less capacitance variation due to frequency and no leakage current, for example.
The ideal resistor is a resistor whose characteristics are free from parasitic inductance and variations in resistance value resulting from effects of applied voltage, temperature, humidity, pressure, etc.
However, the leakage current, resistance of electrodes and interconnection, time required for polarization etc. cannot be neglected with a real capacitor;
FIG. 11
shows the leakage current component as the resistor R
0
parallel-connected to the ideal capacitor C
0
and the resistance of electrodes and interconnection as the resistor R
10
series-connected to the ideal capacitor C
0
.
Polarization occurs as positive and negative charges transfer when an electric field is applied to electrodes of the capacitor, which includes the four mechanisms: electronic polarization due to displacement of electron cloud, ionic polarization due to displacement of ions, orientation polarization due to rotation of the dipole moment of molecules, and interfacial polarization caused as charges in the dielectric transfer and are accumulated at the interface. Development of polarization by these mechanisms takes a certain time and a time delay therefore occurs with respect to the phase of the electric field.
The electronic polarization and ionic polarization occur when an electric field having a frequency in the ultraviolet or microwave region is applied, so that the orientation polarization and interfacial polarization must be considered in common semiconductor devices which operate at lower frequencies.
FIG. 11
also shows them as the equivalent circuit. That is to say, the polarization components are represented as the series circuits D
1
, D
2
, D
3
and D
4
formed of the capacitor C
1
and resistor R
1
, the capacitor C
2
and resistor R
2
, the capacitor C
3
and resistor R
3
, and the capacitor C
4
and resistor R
4
which are connected to the ideal capacitor C
0
in parallel.
A product of the capacitor component C (unit F: Farad) and the resistance component R (unit &OHgr;: Ohm) of each circuit corresponds to the relaxation time &tgr; which is used as an indication of the time required for polarization. Accordingly, when the capacitance values of the capacitors C
1
to C
4
are taken as C
1
to C
4
and the resistance values of the resistors R
1
to R
4
are taken as R
1
to R
4
, then the relaxation times &tgr;
1
to &tgr;
4
of the series circuits D
1
to D
4
are given as C
1
R
1
, C
2
R
2
, C
3
R
3
and C
4
R
4
, respectively.
Generally, while a polarization component functions as capacitor with respect to electric fields having frequencies lower than the reciprocal of the relaxation time, the polarization component cannot follow in operation as capacitor with respect to electric fields having frequencies higher than the reciprocal of the relaxation time, and then the polarization component cannot contribute as capacitance.
Study on the frequency characteristic of the capacitor capacitance shows a trend that the capacitance becomes smaller at higher frequencies, which is due to the fact that the polarization components cannot follow at high frequencies and therefore cannot contribute as capacitance. It is assumed in
FIG. 11
that four polarization components having different relaxation times exist and they are shown as the series circuits D
1
to D
4
.
In this example, the relaxation times &tgr;
1
to &tgr;
4
are assumed to become larger in the order of the relaxation time &tgr;
1
, which is the shortest, and then &tgr;
2
, &tgr;
3
, and &tgr;
4
.
A real capacitor having such configuration exhibits a frequency characteristic as shown in FIG.
12
.
That is to say, when the frequency is represented as the reciprocal of the relaxation time, and if the frequency is lower than 1/&tgr;4, all of the capacitors C
1
to C
4
as polarization components can follow in polarization, so that the capacitance of the real capacitor is the sum total of the ideal capacitor C
0
and the capacitors C
1
to C
4
.
However, when the frequency becomes equal to or higher than 1/&tgr;4, then the capacitor C
4
cannot follow, and the capacitance of the real capacitor is the sum total of the ideal capacitor C
0
and the capacitors C
1
to C
3
.
Similarly, when the frequency becomes equal to or higher than 1/&tgr;3, the capacitor C
3
cannot follow, and when the frequency becomes equal to or higher than 1/&tgr;2, the capacitor C
2
cannot follow, and when the frequency becomes equal to or higher than 1/&tgr;1, the capacitor C
1
cannot follow, and finally, all of the polarization components cannot contribute as capacitance and only the capacitance of the ideal capacitor C
0
remains.
Actually, the relaxation times of the polarization components continuously exist and the frequency characteristic also vary continuously.
A storage capacitor in a semiconductor storage device can be represented similarly as a plurality of capacitors in which polarization components are connected in parallel;
FIGS. 13
to
15
show a problem of the storage capacitor which is caused by dielectric loss resulting from interfacial polarization and orientation polarization.
In
FIGS. 13
to
15
, the storage capacitor CP which is subjected to writing and reading of data is shown as a capacitor having, in addition to the ideal capacitor C
0
, the series circuits D
1
to Dn as polarization components, which are formed of n capacitors C
1
to Cn and resistors R
1
to Rn series-connected respectively to the capacitors C
1
to Cn.
As stated above, the capacitors C
1
to Cn have different relaxation times &tgr;
1
to &tgr;n, where &tgr;
1
is the shortest and &tgr;n is the longest.
FIG. 13
shows a condition of the storage capacitor thus constructed, where data is written in for the first time.
In
FIG. 13
, charge is stored in the ideal capacitor C
0
; when the frequency of the ap
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
Tsai Jey
LandOfFree
Method for manufacturing semiconductor device having a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for manufacturing semiconductor device having a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for manufacturing semiconductor device having a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3031548