Pique the Geek 20101003: Sustainability: Water Purification

We have been talking about sustainability recently, and one of the resources in most jeopardy is fresh water.  In the United States the freshwater problem is becoming more and more significant, and in many parts of the world it is already desperate.  We shall look at some of the methods used to purify nonpotable water tonight.

First of all, we need to understand what kind of water we are purifying.  It ranges in quality from surface or ground freshwater, requiring only minor treatment to eliminate microbes that might cause disease (the vast majority of drinking and industrial water in the United States comes from these sources), all the way to seawater, with lots of intermediate kinds.

We shall focus on the more difficult kinds this time.  Seawater is becoming and important source of drinking water in many parts of the world, and will continue to do so.  However, it is not an easy or cheap thing to purify.

Since ancient times it has been known that seawater can be rendered free of salt by distillation.  In a loose sense, almost ALL freshwater is distilled water, since solar heating causes seawater to evaporate, leaving the salts behind, the freshwater falling as rain or snow.  However, much of that freshwater falls right back into the oceans, and thus is lost.  However, enough falls on land to provide for the needs of humanity, or at least used to do so.

In the past 100 years or so we have been tapping what I call fossil water, that is, water that fell as rain or snow hundreds if not thousands of years ago and charged vast underground aquifers.  We have been tapping these aquifers faster than they can recharge for quite a long time now, and they are beginning to become depleted, sometimes seriously so.  In eastern Arkansas, for example, the alluvial aquifer has become so depleted, mostly from agricultural use, that a project is being planned to divert water from the White River to help recharge the aquifer.  Obviously, this will cause a decrease in flow at the mouth of the White.

Aquifer water has the distinct advantage of being unusually pure, insofar that it is usually not salty, although it is often “hard”.  That is fine for most domestic and agricultural uses, but excess hardness is a problem with industrial uses, especially when boilers are involved.  The point is that we are using up fossil water faster than it can be regenerated, and this fossil water has been the difference between having enough water and not having enough, since it makes up the difference between recent rain and snow (surface water) that we get and what we need.

Thus, desalination of seawater has become an alternative.  As I said previously, it has long been known that distillation purifies water, but that is an extremely expensive proposition because of one of the many unique properties of water.  It turns out that it requires more energy to get water to boil than for almost any other substance.  This is called the latent heat of vaporization, and is 40.7 kJ/mol for water, a high number.  As a comparison, it requires only 38.6 kJ/mol to vaporize ethanol.  But, you say, that is not a very big difference.  Well, it is because a mole of water has a mass of 18 grams, whilst a mole of ethanol has a mass of 46 grams, so on a mass, rather than a molar, basis, it takes over 2.5 times more energy to vaporize water.

Thus, to distill one gallon of water, assuming 100% thermal efficiency, it takes 8550 kJ, or 2.4 kilowatt-hours.  Now, last month I used 414 kw-hr of electricity for everything, so I could have distilled 172 gallons of water if ALL of my electrical use went towards it, or a little less than six gallons per day.  A couple of toilet flushes would have taken care of that.  At my power rate (one of the lowest in the US), each gallon of distilled water would have cost me a little over 11 cents per gallon.  Once again, this is assuming 100% efficiency, and distillation is not 100% efficient.

For desalination, low pressures are usually used to reduce the temperature at which water boils.  That does not change the latent heat of vaporization, but it does reduce heat loss and so increases efficiency.  As a matter of fact, some of the heat to distill the water comes from the ambient environment, further increasing efficiency.  However, the vacuum pumps have and energy cost associated with them as well.  There are lots of variations on the theme of distillation, but you still can not get around that 40.7 kJ/mol latent heat problem, since it is a fundamental property of water.  By the way, most nuclear naval vessels distill water for use on board, using the waste heat from the nuclear power plants.

Distillation can be fairly cost effective, but as energy costs increase, it will become less viable since there are alternative methods with lower energy costs.  Except for very limited applications, water distillation for drinking and industrial usage will become more and more rare, although there will always be some demand for specialized applications.

The most widely used method to desalinate water is one of the membrane technologies, reverse osmosis (RO).  Before we look at RO, let us take a moment to examine what “regular” osmosis is.  It has been known for several centuries that when a semipermeable membrane is filled with a salt solution and then placed in a container of fresh water, the fluid inside of the membrane increases in volume at the cost of the fresh water outside the membrane.  This has to do with thermodynamics, but instead of heat, the driving force is entropy.  The physics behind it show that the natural tendency is for water to pass through the membrane (which has pores in it large enough to allow water to pass, but too small for sodium and chloride ions to pass) to dilute the salt solution, making the system as a whole more disordered.

In RO, this same property is exploited, but the pressure is kept high INSIDE of the membrane, forcing the water in the salt solution to pass through the membrane, as the membrane retains the salts.  Obviously, this requires energy to run the pumps to drive the motion of the water through the membrane.  It does, but not as much energy as vaporizing water.  RO units are becoming common around the world, and can produce potable water for around 50 cents per cubic meter, or around 264 gallons, which works out to about 0.2 cents per gallon, making it much more economical than distillation (please double check my arithmetic; I did not use paper to set up this conversion).

RO has a number of disadvantages, however.  The membrane is a critical part of the system, and is easily fouled.  Bacterial growth tends to foul the membrane, and particulate matter destroys it quickly.  Thus, seawater has to be carefully filtered and treated to kill microbes before being introduced to the membrane.  Also, there is a limit as to how much salt can be removed from the seawater, hardly ever more than half of the water being purified.  The concentrated saltwater then has to be disposed of somehow, usually by sending it back to sea.

This can cause problems, since the concentrated brine is more dense than seawater, and tends to sink rapidly, rendering the seafloor much saltier than normal in the vicinity of the effluent pipe.  When you are talking about millions of gallons per day, this can have serious environmental consequences.  Regulations are being formulated to minimize this negative impact, but my point is that no desalination process is without its downside (by the way, the same problem with concentrated brine also exists with distillation).

In addition, these large plants have huge intakes of seawater, which also cause environmental problems, since unless properly designed they suck in fish, larvae, plankton, and anything else swimming in the ocean.  Regulations have been introduced to minimize this, and one approach is to use HUGE diameter intakes so that the water velocity is so slow that free swimming creatures can easily swim against the flow.  Various screens and filters are also in use, but they necessarily restrict flow.

It is estimated that RO capacity will quadruple in the next decade, mostly in arid parts of the world near seacoasts.  There are several facilities in the US, notably in California and, of all places, Florida.  The Middle Eastern countries are going about it in a big way, as is Australia.  However, RO is no panacea because it does require a significant amount of energy as well as a huge infrastructure for large plants.  However, it is the best technology that is mature at present.

There are several membrane technologies being investigated, but none are in large commercial production at present.  As nanotechnology becomes more developed, it is hoped that more efficient methods (thus, less costly) will become available.

RO is usually operated at around 1000 psi for seawater, so the membrane has to be somewhat robust and well supported.  The surface area of the membranes is huge, so that enough contact can be made for the process to occur with a useful rate.  By the way, RO is even better for brackish water than it is for seawater, because less pressure is required to overcome the entropic driving force towards forward osmosis.  Remember, the pressure required for the process increases with the salinity of the water.  Distillation, on the other hand, is just as costly for nearly pure water than it is very salty water, because you are still hitting the latent heat problem.  Since RO has nothing to do with latent heat, it is not limited in that manner.

A lot of people have home RO units.  I am not really a fan of them unless you have a specific need for highly purified water.  Here are a couple of the reasons.  First, small RO units are extremely inefficient, because they usually works at whatever water pressure the supply line gives, thus reducing the efficiency of the process tremendously.  For every gallon of purified water obtained, 15 gallons of water, give or take depending on the system, is discharged into the sewer.  This is a real waste of water unless there is no alternative.  The second reason that I am not fond of them is that most homeowners are not qualified to monitor the quality of the water.  One of my neighbors has a system, and the membrane is shot.  I know this because it use demineralized water in my humidifier in the winter, and they offered to give me their RO water instead of my buying it.  After one of two tanksfull, I noticed a light, white coating on the TeeVee screen.  Always the scientist, I drew my finger through it and tasted it.  It was salt.

In addition to the RO system, they have a water softener just before the RO unit.  In the Bluegrass, the water is extremely hard with calcium salts, and a water softener uses ion exchange to replace the calcium with sodium (that is why you have to buy salt pellets for them).  Their water softener works well, but obviously the RO system does not or there would be no salt in the water.  By the way, my humidifier is an ultrasonic one, and everything gets thrown into the air.  If I had a boiling water one, the salt would be left behind, but the energy costs are high for that kind (back to the latent heat thing).

Thus, I am against home RO systems unless there is some real need for one, such as living in a remote area with horrible well water.  Some method of checking the ionic strength of the output water to monitor the performance of the membrane.

One desalination process that is used sometimes for sea emergencies is based on silver.  Since silver chloride is insoluble, silver salts effectively scavenge chloride, and sodium is absorbed by another material to form an insoluble product.  Thus, seawater passed through this device is drinkable.  Obviously, this is not good for large scale application, but as a lifeboat addition might save some lives.  As I recall it was developed for the military, and I do not know if they are even still in use.  Anyone with more expertise on these is welcome to comment.

This all gets back to sustainability because we are using a resource faster than it is being regenerated.  For many years we have been using fossil water to make up the difference between what we need and what we get, and that is not sustainable.  Desalinating water costs energy, and we are already using up fossil fuel, so in sense we are trading fossil water for fossil fuel use unless alternative energy sources are used to treat the water.  To complicate matters further, think about our food exports.  In a very real sense, those are water exports since most of our cropland is irrigated, much of it by fossil water.  Sustainability is a many faceted problem, and as I said last week, everything is connected to everything.  As population increases and climate change becomes more pronounced, water may well be the next war issue, even more than oil.

Well, you have done it again.  You have wasted many more einsteins of photons reading this watered down drivel.  And even though Rahm Emanuel becomes devoid of ambition when he reads me say it, I always learn much more that I could possibly hope to teach writing this series, so please keep those comments, questions, corrections, and other thoughts coming.  The comments are the best part of this series.

Warmest regards,

Doc

Crossposted at Docudharma.com and at Dailykos.com