Waste-water treatment and Penetrant Testing

June 2011

I- Introduction

Rinsing/washing waters from PT process lines contain hydrocarbons, organic esters, surface active agents (surfactants), dyes and many other organic constituents.

French regulation, and probably regulation in other countries, state, among different requirements, that the effluents disposed of into sewers or rivers shall be clear and without any colour.

Other requirements come atop these basic ones:
• Maximum acceptable content of metals (chromium, iron, nickel, aluminium, etc.)
• Content of total suspended solids (TSS).
• pH which should be between 5.5 and 8.5.
• Temperature equal to or less than 30 °C (86 °F).
• Maximum content of hydrocarbons.
• BOD5 (Five-day biochemical oxygen demand) and COD (chemical oxygen demand) according to the daily flow: bacteria which degrade surface active agents, for instance, need a lot of oxygen.
• Etc.

A look backwards

Treatment of waste-water from PT process lines became a source of concern at the beginning of the 70s.

Adsorption on activated carbons was the first method used to treat waste-water from automatic PT process lines.

One of us wrote then in a paper(*): ‘‘For this purpose, we have designed an efficient process based on the physics of adsorption by activated carbon, and no other process is more efficient.”

More than 30 years later, that is still true.

III- Treatment by activated carbon

It is much better to process rinsing/washing water from PT process apart from waste-water from any other industrial line, and to recycle it provided that it has been “cleaned” enough. As a matter of fact, if the recycled water contains even small amounts of surface active agents (surfactants), and if it is used to rinse/wash off parts, there is a major risk that the penetrant be too much emulsified, hence a dramatic consequence on the process sensitivity. Quite often, the waste-water treatment installation for PT effluents and the PT process line are physically very close as they are considered as one global installation. Furthermore, mixing water containing organic effluents, such as PT effluents, with waste-water coming from surface treatments (stripping, metal plating, phosphating, etc.) which contains many mineral materials (acids, bases, ions) makes it extremely difficult to treat water. This is a main reason to have completely independent networks for both needs.

From 500 to 4,000 litres (110-880 Imp gallons, 132-1,058 US gallons) of water are often used to remove 1 litre of penetrant.

Activated carbon filters are a good way of treating water. In fact, even when other treatment means are used the last step is an activated carbon filter. There are now many replaceable filters which ease maintenance.

Coalescence filters may be used on post-emulsifiable (PE) penetrants process lines; it is a kind of pre-treatment of the mix of water + PE penetrants from the water rinsing step (previous to the hydrophilic emulsifier’s application: this is the “pre-wash” concept). In the 70s, settling or centrifugation of the mix were the standard methods. Water cleaned of most of the non-emulsified PE penetrant was then directly reused in this step, or sent to activated carbon filters. Whatever the method (settling, centrifugation, coalescence), the pre-wash technique allows for the removing of 65% to 90% of the penetrant off the parts. In fact, water thus recovered, though not very clean, is generally directly recycled as it is used mainly to “mechanically” remove the non-emulsified PE penetrant from the surface. This pre-wash step dramatically reduces the overall consumption of activated carbon, as there is few penetrant left on the parts to be emulsified before the wash step.

There are many kinds of activated carbons: therefore, it is very useful to select carefully which one(s) is (are) suitable for a specific application. Activated carbons almost similar as per their respective technical data sheets may have adsorption characteristics for penetrants/emulsifiers which vary as a 1 to 4 ratio. In addition, an activated carbon suitable for a Level 2 WW penetrant of trademark “A” may be a bad choice for a PE Level 3 of the same trademark. Two WW Level 2 penetrants from manufacturers “A” and “B”, both SAE-AMS 2644 approved, may be adsorbed very differently on the same activated carbon. This is puzzling for designers of waste-water treatment installations, when on the same PT line, three tanks contain, for example, respectively a WW Level 2, a PE Level 2 or 3 and a PE Level 4 penetrant, the waste-water being sent in the same treatment installation. Along the day, along the week, the respective content of each penetrant may vary on a large scale. Not that easy to design the most efficient installation!

Some of the PT materials manufacturers did perform the tests to select the most suitable activated carbons. Better to rely on their choice.

In the early 70s, a penetrant manufacturer/supplier marketed three activated carbons: one for the WW penetrants, one for the PE penetrants and one for the hydrophilic emulsifiers.
In the mid 80s, he marketed another activated carbon combining the three activated carbons.
However, as a matter of fact, many PT manufacturers/suppliers market one activated carbon only. It could be that they ‘‘rationalized’’ them by making a compromise. The purpose is probably to prevent any confusion by the user and the need to manage stocks of several activated carbons.

For INFORMATION PURPOSE ONLY, we give the following data: one kg (a bit more than 2 pounds) of activated carbons adsorbs 0.40 litre (0.088 Imp gallon or 0.10 US gallon) of penetrant and between 0.35 and 0.6 litre (0.076-0.13 Imp gallon or 0.1-0.16 US gallon) of hydrophilic emulsifier. However, it is NOT the sum of both: if the filter has adsorbed, say, 0.3 litre of penetrant, it could additionally adsorb ca 0.1 litre of emulsifier. To add to the confusion - or to the complexity - it may occur that some molecules retained by the activated carbon are “washed away” by molecules having more affinity for the activated carbon than the molecules in place, and go out of the filter!

For the best results, water shall continually flow through the filter top to bottom, though some people prefer bottom to top, a more efficient means but more energy-voracious. If well adjusted, the flow will allow a 12 to 20 minute contact time. Sudden changes in the pollutants' content, for instance, at the very beginning of the washing step for a WW penetrant or the rinsing step for a PE penetrant, must be controlled. This is the reason for a “dirty water” tank which acts as a “buffer”, averaging the pollutants' content.

When designing a waste-water treatment installation, the manufacturer shall be supplied with some technical pieces of information: which penetrants/emulsifiers will be used, some idea of the consumption of penetrants and rinsing/washing water, for instance. Equivalent penetrants, though SAE-AMS 2644 approved, may have INCREDIBLE different “needs” of water: we have in mind a WW Level 2 penetrant from a supplier which required between 400 and 800 litres of water per litre of penetrant, while its competitor’s required between 2,000 and 4,000 litres! Many parameters of the waste-water treatment installation shall be adjusted to so different needs.

After treatment, water is expensive. If the treatment is right, water may be recycled.

When water consumption is high and available floor-space scarce, series of several activated carbon filters may be used. Another way is to use a double-compartment buffer tank. Dirty water is sent to the first compartment. It goes through activated carbon filters to the second compartment. This second compartment comes with an overflow device which makes some cleaner water come back to the first compartment. As the filters are small and contact time short, one cycle only is not enough. If one agrees that one cycle removes 90% of the pollutants, we come to the conclusion that seven cycles will remove 99.99999% of the pollutants. In other words, a consumption of 1,000 litres per hour needs a flow rate of 7,000 litres per hour. Clean water is recycled from the second compartment.
The main drawback of this system is that when the user uses water, within seconds, he puts dirty water in the first compartment; dirty water will soon, within seconds only, go to the second compartment ... from which the user is taking “clean water” to rinse/wash the parts. The more the user draws on “clean water”, the more he puts dirty water in the first compartment, the more “clean water” is polluted, as total treatment time has been too short for the right result.

Another technique is based on a fluidized bed of activated carbon. The same drawback as above is met, if water is recycled. There is no time enough for the right and complete treatment of water before use.

PT materials may be applied in a paint spray booth with a back wall over which a “water curtain” flows to collect sprayed materials and, among them, the developer pigments. It is recommended to put a sand filter upstream of the activated carbon filter to trap pigments. Otherwise they would quickly clog the activated carbon and, therefore, would reduce its lifetime.

In the 70s-80s, the usual way was for the user to fill filters by him with activated carbon from 20 to 25 kg (45 to 55 pounds) paper bags. This was a dirty and time-consuming experience. The filter then had to be correctly wet to prevent preferential paths in the filter through which water would go while not being cleaned. Add the unpleasant smells (this is an understatement!) from the activated carbon which had retained a lot of organic materials, smells due to partial degradation in a poorly oxygenated case.

Nowadays, activated carbon filters are supplied as replaceable cartridges, therefore saving time and easing handling. Some suppliers take charge of everything, including the whereabouts of used cartridges ... one fewer worry for users!

Used activated carbon is incinerated or recycled through reactivation by heat in appropriate installations to degrade and get rid of pollutants they had kept. Other techniques may be used.

IV- Other methods

Here are some other methods that some manufacturers have told of as "THE RIGHT ANSWER" to waste-water treatment on PT lines.

IV.1- Reverse osmosis (RO)

This is the method of choice to remove salt from sea water. When one sees the cost of such installations for a cubic metre of clean water per day, we may go to the conclusion it is a "luxury method" to treat PT effluents! Water used on a PT process line needs not to be drinkable after treatment.

Reverse osmosis, as well as ultrafiltration and nanofiltration, uses membranes which are supposed to let clean water go through while stopping all the pollution. You already may anticipate that it is not that easy!

Osmosis is a pressure given by, say, pure water when in contact with water containing soluble chemicals. Clean water "wants" to go to the "dirty water" side so as to balance the concentration of dissolved chemicals. It is what happens when a person is in a bath, in a pool. If the person stays for a long time, when he or she goes out, he or she may see the skin has been swollen. In fact, the pure water of the pool has been attracted by the blood (which is mainly based on water and 8 g/L of salt): the pure water "wants" to dilute blood. This is the osmotic pressure.

To purify water, to get clean water, what we want to do is to take out the water from the "dirty water" through a membrane which, ideally, would retain everything except water. To do that, it is necessary to "push" the dirty water through the membrane, to counterbalance the osmotic pressure. That is why this process is called "reverse osmosis".

Every membrane-based process will face the same problems: huge investment, small volumes of clean water available per day, clogging and/or chemical reaction of the membranes. Keep in mind that membranes are in contact with the waste-water which contains penetrant, even at low concentrations, 24 hours a day, and seven days a week!

IV.2- Ultrafiltration (UF) and nanofiltration (NF)

These two methods are very similar to reverse osmosis, except that they will let some "pollutants" go through the membranes. Therefore, an activated carbon filter is needed as the final step of treatment.

Organic membranes are not prone to clogging - not too much, by the way. On the other hand, these membranes, or the spacers, or both react chemically and/or mechanically with penetrants. Given for a 5-year lifetime, membranes generally need to be replaced every year (this maintenance is not easy when penetrants’ chemicals have made the spacers swell).

Mineral membranes are chemically resistant but are easily clogged. The clean water flow rate dramatically decreases. In no way the "rejuvenating process" as claimed by manufacturers (process that needs use of strong acids or alkaline products, which cannot be thrown away into a river or to the sewers) can allow for a comeback to the nominal flow rate. Once again, though the membranes have been chemically resistant, their lifetime is less than a year.

You may guess costs are far higher than anticipated.

Another technical point: in these membranes-based processes, powerful pumps are used; a large share of their mechanical power heats water in the installation. So the addition of: water + organic materials + heat + not enough oxygen leads to the development of anaerobic bacteria, which increase clogging, but above all produce hydrogen sulphide and thiols whose smelling is really offensive. Further, hydrogen sulphide is very soluble in water and is washed on the parts ... which may be corroded this way!

Membranes, be they for ultrafiltration, nanofiltration or reverse osmosis, are not the right solution from an economical and technical point of view.

Membrane clogging is mainly due to surfactants, quite large and viscous molecules. A membrane clogged by surfactants is almost unrecoverable, whatever the membranes manufacturers say -they do not know what a penetrant is!!!

We have in mind a PT materials manufacturer/supplier who marketed ultrafiltration units. Customers complained that the membrane was rapidly clogged. His amazing solution: the supplier recommended to heat waste-water before it goes into the UF unit!

IV.3- Coagulation and flotation

Coagulation of emulsions, using chemicals such as aluminium chloride (AlCl₃), has been in use for decades. Many such chemicals, also known as flocculants, are available. Other methods use flotation by air, or even by an electric current to help coagulation of all the pollutants which become solids. Solids are then filtered. For penetrant application, though not often used, this process is very efficient with some penetrants (it depends on the penetrant's formula). Water coming from filtration must get a final step through an activated carbon filter, which lasts a long time, as it has very few pollutants to retain.

Recent neurobiological studies suggest that some brain diseases like Alzheimer's are at least favoured by the chronic ingestion of small doses of aluminium. That’s why ferric chloride (FeCl₃) is more and more used in lieu of aluminium chloride (AlCl₃); but this requires further expensive investment to modify water treatment installations accordingly. Nevertheless, keep in mind that nobody drinks water after treatment when it comes from a PT line!

IV.4- Mechanical vapour compression (MVC). Biological process

MCV requires huge investments for small quantities of clean water per day.

Biological process (bacteria "eat" the organic molecules and produce mainly water and carbon dioxide) requires enriching the pollutants with phosphorus (as phosphoric acid), nitrogen and compressed air (to bring oxygen to the tank).

These two methods need to be used as planned. By experience we know that, due to the low daily volume of clean water given by these installations when compared with their size and their sophistication, users are prone to use the units beyond their capability ... and, therefore, there is a dramatic lowering of the quality of the "clean water".

Furthermore, a biological unit shall run 24/24, 7/7: no weekends, no holidays (vacations, for our American friends).

IV.5- Ozone

A PT materials manufacturer/supplier markets a system using ozone. Everyone knows that ozone is a strong oxidizer and that it ‘‘kills’’ the fluorescent brightness of dyes. Therefore, this process makes water become almost colourless. This unit is designed to treat 1,900 litres (ca 415 Imp gallons, 500 US Gallons) of rinse water/24 hours; this is a small quantity, and the process is quite low. We may guess it is suitable for some relatively small PT process lines.

Since ozone is a biocide, it reduces the bacteria and odour problem.

The main point is that in no way this system is able to destroy hydrocarbons or surface active agents (surfactants), or other complex molecules used in penetrants and emulsifiers formulae. After treatment, water cannot be recycled; “overwashing” due to the surface active agents (surfactants), which are not removed from washing/rinsing water, is a major concern. The cost per US gallon (or per litre) of water is impressive. Ozone must be produced either from the oxygen in the air, with a poor yield, or from bottles of oxygen, with an improved yield ... but also an improved cost! Ozone is not very soluble in water; it must be produced in excess ... and it is forbidden to let ozone leak to the atmosphere. A filter made of ... activated carbon must be put on the gas exhaust.

Water from such a system is likely not to meet the regulations about hydrocarbons content, BOD5 and COD, if sent to sewers.

We are not sure that this is an economically sustainable, environment friendly answer to the problem of water “cleaning”.

V- Other... ineffective methods

The other industrial methods are inefficient and have been given up:, ferric perchloride or even hydrogen peroxide as oxidants; PTFE disks with baffles to use the tiny difference of density between water and non emulsified chemicals; electrolysis has even been tested on waste-water from PT process lines ... when penetrants materials give no ion in water!

Further, using some chemicals need very stringent safety rules to be followed. One of us saw a 60 litre (ca 16 US gal) plastic can of a 35% solution of hydrogen peroxide stored in the vicinity of a waste-water treatment installation. This can was exposed to sun rays for several hours per day, and very close to a lane where forklift trucks were running at high speed, their forks being exactly at the right height to hit the can: the ideal conditions to cause a dramatic explosion!

Keep in mind that the process, which, as if by magic, comprises a dirty water tank, a clean water tank and in the middle a "black box" that no one checks or maintain and that produces no refuse does not exist! Except for the biological method (which nevertheless produces refuse as "activated sludge" which cannot be thrown away as domestic refuse), a waste-water treatment installation condenses the pollutants ... Sooner or later this "concentrate" shall be disposed of!

Reference

(*) Jean-Claude HUGUES and Pierre CHEMIN, Contrôle non destructif par méthode de ressuage (5e partie)- Revue Pratique de Contrôle Industriel – N° 82, December 1976.

SAE-AMS 2644E: Inspection Material, Penetrant, Society of Automotive Engineers (SAE), 400 Commonwealth Drive, Warrendale, Pennsylvania 15096, 2006.