A complex waste stream is created at this facility. The plant specializes in aircraft lighting and uses machine and sheetmetal shops along with a small painting and plating facility. Processes include small part painting and plating, chromic and sulfuric anodizing, nitric with phosphoric passivation, chemfilm, and various sealers and dyes.

+The current flow is 1,000gpw from the line rinses, 300gal/mo of machine coolants, and annual dumps of process baths. The process baths and their volumes are as follows:

The concentrates are dumped based on calendar, and can be planned at two per month. The worst case dump would probably be tanks 7 and 22 simultaneously (high chromium and nitric
acid).

CONCLUSION
This waste is a good match for the KASELCO process and performance can be guaranteed based on this test. Some special considerations must be taken into account in the system’s design.

The coolant (emulsion) may be added as generated and will impact only the sludge generation rate. Either dissolved air flotation or settling can be used for solids separation of the basic rinsewater-coolant waste. The sludge also settled well without extensive agitation, allowing use of a clarifier.

If concentrates are to be processed then pH adjustment (raising) should be included, particularly to process tank 22 and possibly the nitric and phosphoric oxidizer. Inclusion of these wastes will also require extended settling and filtration times. The alkaline neutralizer (sodium hydroxide) would be pumped directly from a supply drum. The pH adjustment capability necessitates agitation of the equalization vessel.

The mixture of all concentrates processed without pH adjustment but again made a high sludge volume. The final water did not look as good as without the concentrates, but met discharge standards easily.

A minimum weekly flow of 1125 gallons (rinsewater and coolant) is well within the capacity of a 2.5gpm reactor operating less than eight hours per week with single pass treatment or less than 15 with two passes. If the system is automatically controlled (once manually initiated) then it should operate twice weekly using ½ the volume of a 1200gal equalization tank. The heel remaining in the tank would serve to average the waste concentration. Two concentrate tanks are recommended at 500gal each (or large enough to hold any acid or alkaline process tank). The acid and alkaline concentrates would be stored held separately and added proportionately by manual control of the operator before starting the treatment process. Automatic additions are possible, but complex and expensive. There is ample time to process the waste for multiple passes rather than adjust the pH chemically. Processing multiple passes also requires an agitator in the equalization tank, but eliminates the pH sensor and control system.

If there is potential for growth of the waste stream then a 5/10gpm automatic system may be appropriate. This system has a two-pass throughput capacity of 5gpm or a single pass rate of 10gpm. It could operate at 5gpm but not 10gpm if the heavier concentrates are mixed into the stream, as the sludge volume would be too high for the sludge thickener and clarifier.

BENCH SCALE KASELCO ELECTROCOAGULATION TREATMENT
One sample tested must include the rinsewater mixed with proportionate quantities of the concentrates. With annual rinsewater flow of 52,000gal and annual dumps of 2559gal, the proportion of rinse:concentrate would be 20:1, or 2090ml rinse:110ml concentrate. However, the coolant should be added at a rate of 65ml/2200ml, so the final ratio will be 1925ml rinse:165ml coolant: 110ml concentrate.

A demonstration test was conducted while Rheaco representatives were present. It included only the rinsewater and the coolant. The sample treated easily with good visual indicators. The sludge settled well following agitation and polymer assisted floc formation both before and after agitation.

Test #1: Blend rinse, coolant, and worst tanks:
A mixture was made with proportional volumes of rinsewater, coolant, and the two concentrate tanks considered worst case, #7 and #22. This provided a high chromium concentration and low pH with nitric acid present in the highest concentration expected.

The mixture pH was 1.89 with a conductivity of 7,240μS (microsiemens). The water had an obvious hexavalent chromium color. The sample initially drew the optimum amperage at near-mid-scale voltage. However, the amperage fell sharply after mid-reactor (1A) and was only on-half scale during the second pass (2A and 2B). This lack of reaction energy greatly affected the outcome.

The pH rose from 1.89 to 2.18 after one pass (1B) and to 2.71 after two passes (2B). No analyses were performed since the reaction was visibly incomplete. An oil film developed on the final water as the emulsion
was broken.

Test #2: Repeat Test #1 with different reactor configuration and pH adjustment:

The same mixture was prepared and the pH was adjusted to 5.48 using the equivalent of 2.7gallons of 50% sodium hydroxide per 1,000 gallons of waste. The adjustment lowered the conductivity to 3,200μS. The mixture was processed in a reactor designed for lower conductivity. It drew the optimum amperage at a low voltage.

The sample pH rose from 5.48 to 8.22 in a single pass (1B) but the sample retained some of the chromium color. The pH rose to 8.76 with a second pass (2B) but retained a light yellow color. Analysis showed that the first pass had lowered the turbidity from >4,400FAU (Formazin Attenuation Units) to 2,049FAU, the chromium form 208.1mg/l to 20.9, nickel from 0.694 to less than detectable limits, copper from 1.41 to 0.126, and iron from 15.6 to 1.87mg/l. The second pass lowered the turbidity to 581FAU, chromium to 0.026mg/l, copper increased to 0.178, and iron lowered to 1.76mg/l.

The second pass had effectively treated the chromium to within discharge limits but some color remained. The color is partially from iron, but primarily from some sort of dye or colorant that is not destroyed. A good, separable floc was developed. The dry sludge rate was 34.36 pounds of wet sludge per 1,000gal of waste, or 2.2 cubic feet per week.

Test #3: Mixture of all concentrates with rinse:
A mixture was made using the rinsewater and coolant with a proportional quantity of the annual dumps for the concentrates. The amount added of each can be seen in the report below.

The mixture had a pH of 5.76 and conductivity of 5,770μS. It had a hexavalent chromium color. The mixture was processed in a type 05 reactor and drew the optimum amperage at less than optimum voltage. The pH rose from 5.76 to 6.65 at mid-reactor and to 8.45 at the reactor’s exit (1B). Chromium and copper remained at above desired levels, and iron was higher than expected at this pH. The sludge rate was taken but is not relevant, since the waste was not fully reacted.

Test #4: Repeat test #4 to 2B:
The previous test was repeated to the equivalent of two full-scale passes (2B). Results at one pass (1B) were similar to the previous run. The voltage was less than optimum. The finished water had a cloudy appearance (common with cleaning solutions containing silicates).

The pH continued to climb on the second pass, reaching 9.86 at 2B. The turbidity improved from >4,400 to 2,181FAU. The second pass lowered the chromium to 0.119mg/l, nickel to 0.197, copper to 0.111, and iron
to 1.73mg/l. The residual iron is potentially a function of the presence of either chlorides or sequestering agents (chelators). A high volume of sludge was produced that settled well but filtered only with difficulty. The dry sludge rate was 55 pounds of wet sludge per 1,000 gallons of the mix, or 3.5 cubic feet of wet sludge per 1,000 gallons of the mixture. At 56,000 gallons of was per year, a total of 7.26 cubic yards of sludge would be generated annually.

Test #5: Rinse water alone:
The rinsewater was tested alone for comparison and to determine the sludge contribution from the concentrates and coolants. The sample had a pH of 6.93 and conductivity of 1,880μS with a chromium color. The sample drew the optimum amperage and voltage in a type 05 reactor. The pH rose from 6.93 to 10.02 at mid-reactor and to 10.42 at the reactor’s exit. The conductivity increased during reaction, usually indicating the presence of some organic chemicals. The sample was processed for a single pass (1B). Turbidity improved from 28 to 10FAU, while chromium was lowered from 19.6 to less than detectable, nickel was not detectable in the original sample, copper lowered from 0.041 to less than detectable, and iron rose from 0.108 to 0.343mg/l.

A good floc was generated in low volume. The dry sludge rate was 4.6 pounds per 1,000 gal, or 0.3 cubic feet of sludge per 1,000 gallons of rinsewater (15 cubic feet per year).
LAB DATA - abbreviated

TEST # 1(blended)
1925ml rinse, 165ml new coolant, 24ml tank 22, and 86ml tank 7:
PROFILE: pH 1.89. Conductivity 7,420μS. Appearance is bright yellow water. Run in #07 steel reactor.

COMMENTS: 1-A dark yellow water. 1-B through 2-B gold water, no floc with oil film.
Insufficient pH rise. Do not analyze.

TEST # 2 (same mixture as test #1 with pH adjustment and #05 reactor)
Added 6 ml 50% NaOH. Profile: pH 5.48. Conductivity 3,200μS. Appearance is dark yellow water. Run in # 05 steel reactor.

COMMENTS: 1-A &1-B dark yellow water, sinking tan floc. 2-A & 2-B light yellow water, floating brown floc.
1000ml Wet sludge test. 300ml @ 15 min. 270ml @ 30 min. 230ml @ 60 min.
100ml Dry sludge test, 0.412g

TEST # 3 (mixture of all baths)
View mixture table. Profile: pH 5.76. Conductivity 5,770μS. Appearance is yellow water. Run in # 05 steel reactor.

COMMENTS: 1-A yellow water, brown sediment. 1-B light yellow water, floating brown floc.
1000ml Wet sludge test. 200ml @ 15 min. 170ml @ 30 min. 140ml @ 60 min.
100ml Dry sludge test, 0.435 g

TEST # 4 (same mixture as test #3/tested to 2-B)
Profile: pH 5.76. Conductivity 5,770μS. Appearance is yellow water. Run in # 05 steel reactor.

COMMENTS: 1-A yellow water, brown sediment. 1-B & 2-A light yellow water, floating brown floc. 2-B cloudy water, floating brown floc.
1000ml Wet sludge test. 640ml @ 15 min. 520ml @ 30 min. 300ml @ 60 min.
100ml Dry sludge test, 0.660 g
NOTE samples 1-B & 2-B filtered very slowly.

TEST # 5 (rinse water sample)
Profile: pH 6.93. Conductivity 1,880μS. Appearance is yellow/green water. Run in # 05 steel reactor.

COMMENTS: 1-A cloudy water, floating light green floc. 1-B clear water, floating green floc.
1000ml Wet sludge test. 150ml @ 15 min. 110ml @ 30 min. 80ml @ 60 min.
100ml Dry sludge test, 0.055 g

GLOSSARY
• Profile: The characteristics of the waste being tested.
• 1A: Equivalent of half a pass through a full-scale production reactor.
• 1B: Equivalent of one full pass through a full-scale production reactor.
• Type: KASELCO EC reactors come in three electrical configurations and is chosen based on the wastewater’s conductivity.