This guide recommends the water quality required for the electronics and microelectronics industries. High-purity water is required to prevent contamination of products during manufacture, since contamination can lead to an unacceptable, low yield of electronic devices.
The range of water purity is defined in accordance with the manufacturing process. The types of ultra-pure water are defined with respect to device line width. In all cases, the water-quality recommendations apply at the point of distribution.
The limits on the impurities are related to current contamination specifications and to available analytical methods (either performed in a suitable clean laboratory or by on-line instrumentation). On-line and off-line methods are used in accordance with current industry practice. Concentration of the sample may be required to measure the impurities at the levels indicated in Table 1.
TABLE 1 Requirements for Water at the Point of Distribution in the Electronics and Semiconductor IndustriesA
| Parameter | Type E-1 | Type E-1.1 | Type E-1.2
B | Type E-1.3
B | Type E-2 | Type E-3 | Type E-4 |
| Linewidth (microns) | 1.0-0.5 | 0.35-0.25 | 0.18-0.09 | 0.065-0.032 | 5.0-1.0 | >5.0 | |
| Resistivity, 25°C (On-line) | 18.1 | 18.2 | 18.2 | 18.2 | 16.5 | 12 | 0.5 |
| TOC (μg/L) (on-line for <10 ppb) | 5 | 2 | 1 | 1 | 50 | 300 | 1000 |
| On-line dissolved oxygen (μg/L) | 25 | 10 | 3 | 10 | | | |
| On-Line Residue after evaporation (μg/L) | 1 | 0.5 | 0.1 | | | | |
| On-line particles/L (micron range) | | | | | | | |
| >0.05 μm | | | | 500
C | | | |
| 0.05-0.1 | | 1000 | 200 | N/A
C | | | |
| 0.1-0.2 | 1000 | 350 | <100 | N/A | | | |
| 0.2-0.5 | 500 | <100 | <10 | N/A | | | |
| 0.5-1.0 | 200 | <50 | <5 | N/A | | | |
| 1.0 | <100 | <20 | <1 | N/A | | | |
| SEM particles/L (micron range) | | | | | | | |
| 0.1-0.2 | 1000 | 700 | <250 | N/A | | | |
| 0.2-0.5 | 500 | 400 | <100 | N/A | 3000 | | |
| 0.5-1 | 100 | 50 | <30 | N/A | | 10 000 | |
| 10 | <50 | <30 | <10 | N/A | | | 100 000 |
| Bacteria in CFU/Volume | | | | | | | |
| 100 mL Sample | 5 | 3 | 1 | N/A | 10 | 50 | 100 |
| 1 L Sample | | | 10 | 1 | | | |
| 10 L Sample | | | | 1 | | | |
| Silica - total (μg/L) | 5 | 3 | 1 | 0.5 | 10 | 50 | 1000 |
| Silica - dissolved (μg/L) | 3 | 1 | 0.5 | 0.5 | | | |
| Anions and Ammonium by IC (μg/L) | | | | | | | |
| Ammonium | 0.1 | 0.10 | 0.05 | 0.050 | | | |
| Bromide | 0.1 | 0.05 | 0.02 | 0.050 | | | |
| Chloride | 0.1 | 0.05 | 0.02 | 0.050 | 1 | 10 | 1000 |
| Fluoride | 0.1 | 0.05 | 0.03 | 0.050 | | | |
| Nitrate | 0.1 | 0.05 | 0.02 | 0.050 | 1 | 5 | 500 |
| Nitrite | 0.1 | 0.05 | 0.02 | 0.050 | | | |
| Phosphate | 0.1 | 0.05 | 0.02 | 0.050 | 1 | 5 | 500 |
| Sulfate | 0.1 | 0.05 | 0.02 | 0.050 | 1 | 5 | 500 |
| Metals by ICP/MS (μg/L) | | | | | | | |
| Aluminum | 0.05 | 0.02 | 0.005 | 0.001 | | | |
| Antimony | | | | 0.001 | | | |
| Arsenic | | | | 0.001 | | | |
| Barium | 0.05 | 0.02 | 0.001 | 0.001 | | | |
| Boron
D | 0.3 | 0.1 | 0.05 | 0.050 | | | |
| Cadmium | | | | 0.010 | | | |
| Calcium | 0.05 | 0.02 | 0.002 | 0.001 | | | |
| Chromium | 0.05 | 0.02 | 0.002 | 0.001 | | | |
| Copper | 0.05 | 0.02 | 0.002 | 0.001 | 1 | 2 | 500 |
| Iron | 0.05 | 0.02 | 0.002 | 0.001 | | | |
| Lead | 0.05 | 0.02 | 0.005 | 0.001 | | | |
| Lithium | 0.05 | 0.02 | 0.003 | 0.001 | | | |
| Magnesium | 0.05 | 0.02 | 0.002 | 0.001 | | | |
| Manganese | 0.05 | 0.02 | 0.002 | 0.010 | | | |
| Nickel | 0.05 | 0.02 | 0.002 | 0.001 | 1 | 2 | 500 |
| Potassium | 0.05 | 0.02 | 0.005 | 0.001 | 2 | 5 | 500 |
| Sodium | 0.05 | 0.02 | 0.005 | 0.001 | 1 | 5 | 1000 |
| Strontium | 0.05 | 0.02 | 0.001 | | | | |
| Tin | | | | 0.010 | | | |
| Titanium | | | | 0.010 | | | |
| Vanadium | | | | 0.010 | | | |
| Zinc | 0.05 | 0.02 | 0.002 | 0.001 | 1 | 5 | 500 |
| Temperature Stability (K) | | | | ±1 | | | |
| Temperature Gradient (K/10 min) | | | | <0.1 | | | |
| Dissolved Nitrogen On-line (mg/L) | | | | 8-18 | | | |
| Dissolved Nitrogen Stability (mg/L) | | | | ±2 | | | |
A The user should be advised that analytical data often are instrument dependent and technique dependent. Thus, the numbers in Table 1 are only guidelines. This table will be revised whenever the semiconductor industry develops new linewidths, thereby keeping the guidelines current.
B Values shown in Type E-1.3 are a result of aligning ITRS risk factors of known contaminates to the production processes found in current semiconductor processing for the linewidth of interest and may differ in a few cases to those found in Type E-1.2. Users who wish to use the higher numbers for Type E-1.2 water should feel free to do so.
All values are equal to or less than with the exception of Resistivity.
C Particle metrology has not kept pace with the decreasing line-width of semiconductor manufacturing. Current line-widths require the ability to monitor 20-nm particles. However, existing Optical Particle Counters (OPCs) are only capable of detecting 50-nm particles with a counting efficiency of <5%, and a background count (noise level) of 500 particles per liter. Particle-counting statistics become important as count levels approach the noise level. Therefore, the OPC setup and performance must be optimized. Particle levels must consistently be within the noise level of any OPC (regardless of any specified level).
D Boron is monitored only as an operational parameter for monitoring the ion-exchange beds.
View Less