By Robert E. Mesick

Recycling: why you’re spending more now and enjoying it less.

This article considers only normal PWB processing and examines the following metals: copper, nickel, lead, tin, gold, and the problem anions. cyanide and EDTA complexes. If you’ve added something new, check its treatability FIRST. All of these systems can keep you in compliance, even at 0.1 ppm; some will just cost you more to operate or require a "patch."

These days, everyone wants to recycle something. The latest and greatest thing over the last three to four years is recycling water from PCB plants. Millions of dollars have been spent on huge recycling systems without much thought on whether or not this is really worthwhile. What are the advantages and disadvantages of recycling, and what are the problems you can encounter when recycling?

First, keep your eye on the ball. What do YOU want: Recycling, waste minimization, sludge reduction, low labor costs, automation, etc.? When any one of these is pushed to extremes, something else becomes more expensive.

There are really only four types of cost-effective treatment systems that are available, with some sub-categories. They are:

Type 1—Precipitation with gravity settling. These conventional systems were used for years when the limits were 5 ppm heavy metals. Additives keep these systems alive but the cost of sulfiding and reducing agents make them very expensive to operate.

• Lowest capital cost
• High sludge volume
• May require final filter
• May require chelate breaker (see below)
• Continuous surveillance required
• Single pipe system is possible with chelate breaking additives
• Very large floor space required

Type 2—Precipitation with membrane filtration. These were breakthrough systems in the late 1980’s when limits went under 5 ppm and there were no additional agents to keep the precipitation systems in compliance. Initially they were supposed to work with the same chemistry as the precipitation systems but used a membrane to filter continuously. It didn’t work. Additives were added to reduce the metals ions to elemental form and segregation was required to keep the systems from fouling (no tin, photoresist, etc.).

• High capital cost
• High operational cost (power costs, pump/membrane maintenance, chemical additions)
• Very efficient, low operator attention, automatic operation
• Usually requires reducing or sulfiding agent
• Large floor space requirement
• Required segregation of rinses
• Medium to high sludge volume

Type 3—Deionization (DI)/closed loop. When recycling became fashionable, deionization systems were adapted to close the loop. Unfortunately, the vendors were often DI people used to working with clean city water. Lately big DI’s seem to work but they are expensive.

• Low Capital cost (should be, but there are quotes for 50 gpm systems at $350,000).
• High regeneration frequency: All ions are removed from the water so these systems must be sized for total TDS. Systems are usually regenerate once or twice per day.
• If carbon, ozone, UV/peroxide used to treat organics, DI’s can be an instant recycling system.
• Higher maintenance due to traditional lower cost components. High regeneration frequency allows closer sizing of system to minimal configuration to keep costs low.
• Large regeneration volume. Regenerant volume for both cation and anion resins plus water for rinsing both columns. All have metals.
• Large batch treater required.
• Anion and rinse water metal content is too low to electrowin. These should be segregated, batch treated, and sludged.
• Cation metal is at 1-2 g/l unless a 2 or 3 stage regeneration is used.
• Staging requires segregation of first portion of regenerate for metal recovery and collection and storage of the remaining regenerant for use in the next regeneration cycle. Staging adds more cost and complication to the system.
• If cation rinse water is added to the metal regenerant, the volume to plate-out is very high and thus, not efficient.
• Sludge is generated at low- to medium-volume.

Type 4—Metal Recovery Ion Exchange (Chelated Resin). These systems us a chelated resin that only strips out heavy metals (copper, lead, tin, nickel) and leaves everything else in solution.

• Medium cost (higher than basic DI, lower than membrane systems)
• Selective recovery of heavy metal. If metals are 30 ppm and total cations including sodium, calcium, and magnesium are 170, only the 30 ppm of heavy metals are removed.
• Low maintenance due to low regeneration frequency (between twice a week to twice a month depending on metal loading).
• Resin will strip chelated metals from rinse water, EDTA is not a major problem.
• Segregation is desirable but not an absolute requirement. Some shops run single pipe systems or segregate only non metal bearing waste and single pipe the rest to treatment.
• Very low operator attention, systems are automated.
• Lowest waste volume. Regeneration rinse waters are reprocessed by the system.
• Low regenerant volume. Regenerant volume is 300-400 gallons for a 75 gpm system with a concentration of 8-15 g/l of Copper in the waste.

If your system will not meet local effluent limits, there are some patches you can add to increase the efficiency of your system.

Type one systems sometimes "leak" precipitated metals. A polishing filter will remove these and should get you back into compliance. Automated filtration systems work on back-pressure to initiate backwashing. These systems require 15–20 minutes of operator attention per day.

Type 1 and Type 2 systems operate by precipitating metal. Type 1 uses oxide formation as the primary step and Type 2 uses sulfide or metal reduction as primary treatment. Some Type 2’s used to use ferrous sulfate to break chelates and to add a lot of bulk to control the metal, but the sludge volume was huge.

There are two basic types. First, reducing agents like Sodium borohydride that turn metal ions to elemental metal.

The second type are sulfiding agents like dithiocarbimate and Na2S and others. Metal sulfides have very low solubilities. These are added to Type 1 systems to maintain compliance.

All 4 system types can be used to generate water for recycling. All have problems that you will have to address. Each also have advantages. You have to evaluate what waste streams you intend to put into the "recycle mix". If you add the batch dumps, you add a lot of dissolved solids that you have to remove. If you use dragouts and batch treat them, you remove a lot of dissolved solids from the "recycle mix". Labor for batch dump treatment versus automated treatment is another consideration.

Why recycle? If there is no economic advantage to recycling and it will cost a lot more, why recycle? There is no free lunch.

First, let’s look at the bad reasons for recycling:

• The boss wants me to do it.
• The city would like it if I did.
• I will get my name in the paper if we recycle.

Now for some better reasons for recycling:

• Water costs a lot of money (you’re on a desalination system).
• Sewer charges are based on flow and I’m getting hit with a surcharge that would pay off a recycling system in 18 months or less.
• The water quality coming in is so bad that it costs me a lot to clean it up and what I’m sending out the back is cleaner than what is coming in the front (like you’re in Pakistan or Turkey).
• The city has a cement truck around the corner and is going to fill my sewer connection because I have not been diligent in my housekeeping and have a few hundred violations.

1. You can use a lot of carbon to remove organics. Carbon has different capacities for different organics, and you usually have no capability to measure what is leaking through. The result is that you haul carbon on a time interval and hope you don’t have a problem.
2. You can destroy organics. Another layer of equipment is required which adds to costs. Ozone (not very efficient), UV/peroxide (high cost) and carbon (adsorption, see above) are used.
3. Recycling concentrates ions in a smaller volume and may result in discharge limit violations. For example: the city’s limit is 2500 ppm for TDS and you have been running 800 ppm. The new system increases the TDS to 1200 ppm with the extra chemistry you will have to add. You add a Reverse Osmosis to give you an 80% recycle rate. This results in a discharge of 4800 ppm which the city won’t like.

These need a very good filtration system and a reverse osmosis system to recycle. The TDS exiting these systems is too high to feed a DI economically. It can have the lowest total system cost. Organics > 300 molecular weight are removed by the RO. Some have tried Ultrafiltration followed by DI, but the sodium and chloride are still very high for a DI system.

Follow the microfiltration with an RO like type one. You should be able to get a 75 percent or higher recovery rate if the system is designed correctly. A good total water analysis of the water exiting the microfilter is a requirement.

DI’s following the microfilter are not economical. The high TDS exiting the microfiltration due to treatment chemistry make DI systems uneconomical.

You will get DI water to start. Carbon filtration is needed in front to protect anion resins from the organics and any oxidizers. These systems are sensitive to oxidizers but carbon (lots) protects the system.

These systems should have RO water as makeup water to keep calcium and magnesium out of regenerant. This will help if you plan to plate metal out of the regenerant.

DI’s discharge batches of waste that are 8–10,000 ppm. A medium DI system running 75 gpm will generate 2–3,000 gallons of concentrate and DI rinse water once or twice a day that must be processed.

These systems need an RO to recycle. TDS out of the system is about ten percent higher than the incoming TDS. The cost is higher than a basic DI but less than microfiltration with RO.

1. You are plating with junk solution. It is not consistent! It is not a plating bath with brighteners. It will plate the way the chemistry dictates and not the way you wish it would. The purer and more concentrated the solution, the better it will plate.

2. The higher the concentration the better. Type 4 gives you the highest concentration, type 3 is about 1/4 as concentrated and types 1 and 2 give you sludge. For Type 1 and 2 You can re-dissolve sludge and plate the metals out. Efficiencies vary with metal concentration and contaminants. The lower the concentration, the lower the plating rate. The higher the levels of other metals and/or calcium and magnesium, the lower the plating rate.

3. Electrowinning results vary with the solution. It won’t be pretty, but it will get the metals out. You can redeposit the metals by using your first recovery deposit as an anode for a high efficiency cathodic deposition like the mines but it is probably not cost effective unless you have large qualities of copper to recover.

4. You have to watch the connections. Like any plating bath, the connections like to self-destruct, oxidize, etc. A little maintenance with each cycle and the systems will run carefree for a long time.

Finally, all 4 Types can be used for different applications. For printed circuit manufacturers, "with no recycling," MRIX units are pretty much trouble free and recover Copper in a concentrated form. Operation is also relaxed with not much worry about staying in compliance. Surprise visits to MRIX sites usually find the operator someplace else.

For "low capital cost recycling", a duplex deionization system with good segregation and much carbon is probably the best choice. It is higher in maintenance and the operators need to be a bit more aware. You will get bad regenerations, and you don’t have much time to reinitiate a regeneration before you run out of water. There is a lot more anxiety when you regenerate once or twice a day.

For "money is no object recycling systems", the low maintenance and operation cost of an MRIX coupled with an RO is probably the best choice. You will still need carbon filtration for organics. Microfiltration operation has a high cost associated with it and you end up with sludge. You still need the RO to recycle. If hauling sludge is a desirable option, a microfiltration or gravity clarification system may be a consideration.

If you have a lot of different metals like a metal finishing plating shop, a Type 1 system for end-of-the-pipe treatment followed by carbon, ultrafiltration and an RO is probably the best bet. However, It is the most expensive option.

There you have it. It’s a lot to think about. Make a spread sheet for your flows and get to work!

Robert E. Mesick is president of Remco Engineering. He has be associated with the PCB industry since 1978. He has been previously associated with automated exposure systems, liquid resist coaters, solder leveling and automated converyorized processing of printed circuits. He was awarded a patent on an automated Teflon etching system. Remco Engineering designs and manufactures water and wastewater treatment systems.