Semiconductor device manufacturing: process – Making passive device – Trench capacitor
Reexamination Certificate
2001-06-13
2003-04-08
Smith, Matthew (Department: 2825)
Semiconductor device manufacturing: process
Making passive device
Trench capacitor
C438S243000, C438S246000, C438S249000, C438S386000, C438S389000
Reexamination Certificate
active
06544856
ABSTRACT:
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The invention relates to a method for increasing a trench capacitance in deep trench capacitors used as memory elements in semiconductor components. The trench has a bottle-shaped widened portion reaching into the depth of the wafer, an oxide layer is situated on a silicon substrate and a nitride layer is situated on the oxide layer. A hard mask is formed in them by a preceding photolithographic steps, the openings in the hard mask determine the position of the trench to be formed. The silicon subsequently is etched down to a first depth in order to form the trench, and afterward a conformal deposition of arsenic glass in the trench is performed. Then the steps of filling the trench with a photoresist, etching the photoresist down to a second, smaller depth, etching the arsenic glass selectively with respect to the photoresist, incinerating the residual photoresist in the trench, driving out the arsenic from the arsenic glass into the surrounding substrate, thereby producing an n+-doped substrate, and also etching the arsenic glass and depositing the node dielectric in the trench and depositing the n
+
-type poly-silicon are carried out.
Memory elements are usually embodied as capacitors that make it possible to realize particularly small feature sizes. However, such capacitors must satisfy various requirements. These include not only an adequate capacitance but also, in particular, the capability of long-term storage of the information, i.e. the retention time is intended to be as long as possible. It must therefore be ensured that as far as possible no leakage currents, or small leakage currents, flow to adjacent semiconductor structures. Recent publications have disclosed embodying such memory cells as deep trench capacitors. Such capacitors have a relatively low capacitance, which is likewise reduced as the feature size decreases further. However, the lower capacitance aggravates the problems already mentioned, owing to the fact that, in particular, the retention time, i.e. the charge retention time, is shortened.
It has been shown in DRAM development, however, that every structural miniaturization (shrink) also leads to impairment of the retention time. The requirements made of the product remain the same, however. The hitherto sole practically employed possibility for solving this problem consists in increasing the trench capacitance. For this purpose, use was made of the possibility of using a thinner node dielectric. However, distinct limits are exhibited in this case, these being limited by the leakage current through the dielectric.
Better results can be achieved if deep trench capacitors of this type are provided with a bottle-shaped widened portion reaching into the depth of the wafer. This results in a distinct enlargement of the capacitor area and hence an increase in the capacitance. However, the realization of such a bottle-shaped widened portion is relatively complicated, especially because at the same time it must be ensured that shorts to adjacent memory cells cannot be produced, or that the insulation with respect to adjacent structures is still sufficient to keep the leakage currents within limits.
Bottle deep trench capacitors of this type can be fabricated either by dry or wet etching processes with subsequent filling of the etched depression and widening with a node dielectric.
One example of such a capacitor is revealed in U.S. Pat. No. 5,891,807. In this case, an etching mask is produced by exposure and development of a resist, the structure of which mask is subsequently transferred by dry etching, such as reactive ion etching, plasma etching, etc., into a TEOS and an underlying nitride layer, thereby producing a hard mask. Afterward, the bottle-shaped trench is fabricated by anisotropic reactive ion etching in two steps. In other words, an upper region with a slightly beveled wall is realized and the bottle-shaped widening into the depth is realized in a second step. In this case, different recipes or etching parameters are used for the two etching steps by reactive ion etching. By way of example, the second etching step can be carried out using the same etching gases but a different surface temperature of the wafer or reduction of the pressure within the etching chamber.
What is problematic, in every case, is the exact determination of parameters and, in particular, the definition of etching depth parameters in order to ensure that the predetermined etching depths are exceeded. It is also difficult to determine the point in time when the bottle-shaped widening must be ended.
If the second etching step is stopped too early, an adequate increase in capacitance is not achieved. If the second etching step is stopped too late, there is the risk of adequate insulation with respect to adjacent trenches or active regions no longer being present, with the result that an increase in the leakage currents must be expected.
Moreover, it has been shown that when dry etching processes are employed, increased failure of individual memory cells must be expected.
The customary standard process for fabricating a deep trench capacitor contains, after the fabrication of the hard mask, the work steps wherein the silicon is etched down to a depth of 8 &mgr;m in order to form a trench and conformal deposition of arsenic glass in the trench is subsequently performed. Afterward, the trench is filled with photoresist and the photoresist is etched down to a depth of 2 &mgr;m. The arsenic glass is subsequently etched selectively with respect to the photoresist and the residual photoresist in the trench is incinerated. In order to produce an n
+
-doped substrate, the arsenic is driven out from the arsenic glass into the surrounding substrate and the arsenic glass in the trench is removed by etching. Finally, the deep trench capacitor is completed by deposition of the node dielectric in the trench and deposition of the n
+
-type poly-silicon.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a method for increasing the trench capacitance which overcomes the above-mentioned disadvantages of the prior art methods of this general type, which can be handled in a reliable manner and which makes it possible to achieve an increase in the trench capacitance by at least 15-20% with a low outlay.
With the foregoing and other objects in view there is provided, in accordance with the invention, a method for forming a trench capacitor having an increased trench capacitance forming a memory element of a semiconductor component. The method includes the steps of providing a silicon substrate, applying an oxide layer on the silicon substrate, applying a nitride layer on the oxide layer, and forming a hard mask having an opening formed therein from the nitride layer and the oxide layer using a photolithographic process. The opening determines a position of a trench to be formed. The silicon substrate is etched, through the opening, down to a first depth to form the trench and an arsenic glass is conformally deposited on the trench. The trench is filled with a photoresist and the photoresist is etched down to a second depth being less than the first depth. The arsenic glass is etched selectively with respect to the photoresist. The residual photoresist in the trench is incinerated. The arsenic is driven out from the arsenic glass into the silicon substrate, thereby producing an n+-doped substrate region and then remaining parts of the arsenic glass are etched away. Subsequently a wet-chemical etching of the n
+
-doped substrate region is performed such that the trench is widened selectively both with respect to the n+-doped substrate region and with respect to the oxide layer and with respect to the nitride layer. A node dielectric is deposited in the trench and a n
+
-type polysilicon is deposited in the trench on the node dielectric.
The object formulation on which the invention is based is achieved, in the case of a method of the type mentioned in the introduction, by virtue
Morhard Klaus-Dieter
Penner Klaus
Sperl Irene
Greenberg Laurence A.
Infineon - Technologies AG
Locher Ralph E.
Malsawma Lex H.
Smith Matthew
LandOfFree
Method for increasing the trench capacitance 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 increasing the trench capacitance, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for increasing the trench capacitance will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3028412