Defects Related To Electrical-Leakage In Tmos Structures
- PDF / 2,003,380 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 80 Downloads / 164 Views
325 Mat. Res. Soc. Symp. Proc. Vol. 354 0 1995 Materials Research Society
EXPERIMENTAL Spreading-resistance probe (SRP) and transmission electron-microscopy (TEM) analyses were performed for wafers with splits in the p+ dose and energy. Lots were processed through standard TMOS processing with wafers pulled after the p+ implant for SRP and TEM analyses. Standard TMOS processing steps started with field oxidation, followed by gate oxide and poly gate formation, PHV implant and drive, n+ source implant and drive, p+ implant and drive, contact formation, metallization and finally passivation. The p+ implant had two splits in dose and four in energy, as shown in Table 1. SRP/was performed using a Solid-State Measurements SSM-150 scanning resistance probe. Table I. P+ implant dose and energy splits for TMOS Idss leakage experiment. (Two samples per cell.) P+ Dose
(cm-2 4E+14 2E+15
I
P+ Energy (keV)
20 1,2
3,4
5,6
7,8
9,10
11,12
13,14
15,16
25
30
40
Transmission electron microscopy was performed for boron implants with a dose range of 4e14-2e15 cm"2 and an energy range of 20-40 keV. TEM specimens were made from patterned regions of the wafer, by conventional dimpling followed by ion-milling to perforation; samples were also made by tripod flat-polishing, with brief ion-milling for final finish. TEM imaging and analysis was performed using a JEOL 200CX transmission electron microscope operating at 200 keV.
RESULTS AND DISCUSSION Figure 1 presents device electrical-leakage as a function of p+ implant energy and dose, for the same thermal processing. It can be seen from the figure that the electrical-leakage increases as the energy and dose of the implant increase. Figures 2a-2b present spreading-resistance probe results obtained from processed Si wafers exposed to 20 keV and 40 keV boron implants, respectively. The implanted wafers were annealed to activate dopant-atoms and to tailor diffusion profiles. The SRP data (not all shown here) indicate that the p+ junction depth is -0.4, 0.5, and 0.6-0.7 gtm for 20, 30, and 40 keV implants, respectively. The p+ concentration is 2-4E19 cm"3 for the 2E15 cm"2 dose, and 9E1 8 cm"3 for the 4E14 cm"2 dose. Figures 3a-3d present cross-section TEM (XTEM) micrographs obtained from TMOS structures showing part of the gate and source region. In the source region, the boron energy and dose are 2
varied as follows: Fig. 3a -- 40 keV, 4E14 cm" dose; Fig. 3b -- 40 keV, 2E15 cm-2 dose; Fig. 3c -- 30
keV, 2E15 cm"2 dose; Fig. 3d -- 20 keV, 2E15 cm"2 dose. Various kinds of damage are present. In Fig. 3, we have indicated different kinds of damage by labels. The label "As-EOR" indicates arsenic-implantinduced end-of-range damage (dislocation loops). The label "As-Ill" indicates a type IIIdefect that arises as a result of the arsenic implant amorphizing the silicon. During thermal annealing, recrystallization at the edge of the amorphous region occurs from two fronts. A slight mismatch between the fronts, in combination with strain inthis region causes an a/3[1 11) dislocation l
Data Loading...
