Paulitics of Moral Celery

Part of the reason I am putting this is up is that it is interesting.

The other reason is that there may be people out there with a similar problem (boron) that search the internet (like me) for solutions. I could find basically nothing that a small homeowner (or really, any person in a residential situation as opposed to commercial) could easily implement; it was all in either the context of large scale agricultural irrigation or ocean water desalination places with water shortages.

It started out with noticing that something was not right in the garden. Plants were too light of a green, died, burned on the edges, very stunted/slow-growing, didn’t come up at all, etc. Amazingly, we actually got edible stuff from our garden! The process to discovering what the problem is – the discovery of which is mostly done by now, though not the solution entirely – was a long one.

At this point, we were buying drinking water, as we knew we needed to do more testing before we were comfortable drinking the water.

First, we tested the pH. It was very high (with paper strips, approximately 9). After looking online, it looked like acid injection would fix that pretty easily. Well, “acid injection” isn’t one of those common homeowner things. Eventually, after looking at sulfuric, nitric, phosphoric, and citric, we went with citric acid because it’s much less dangerous.

So, we bought it in bulk, got a cheap $15 siphon injector (approximately a 16:1 ratio), and through trial and error, figured out approximately how much citric acid to add to a 5 gallon bucket of water such that, when siphoned, it would lower our irrigation water’s pH to between 6 and 7. Then we watered with it.

Well, it helped a little – and some plants grew – but still didn’t look too good. Specifically, they looked burned at the edges. Turns out this could be two things: salt and boron burns. At this point, we decide there’s something more significantly wrong, so we have an agricultural suitability test run on the water. It comes back with some surprising results:
pH: 8.8
TDS: 516ppm
Sodium: 180ppm (should be less than 70 for irrigation)
Chloride: 80ppm (should be less than 70 for irrigation)
Alkalinity: 253pm CaCO3 (sould be less than 250)
Boron: 34ppm (should be less than 1)
Adjusted Sodium Absorption Ratio (SAR): 17.2 (should be less than 6)

Yikes! Salt, alkalinity, pH, etc., all is something that can’t be filtered out with something like a carbon filter, pretty much. The lab analyst actually asked if the sample he got (which was from a hose bib) was softened, because it looked like it (due to the salt content and alkalinity, I suppose). Nope! Just from a 980′ well!

So, we go on quite a hunt from here. Reverse osmosis systems can get most of the TDS out – salt, CaCO3, etc. The problem is boron. There are basically three methods for boron removal: distillation, specific ion exchange resins, and reverse osmosis. Distillation is cost prohibitive (and slow) on a large scale. Ion exchange resins are costly for irrigation, and either must be replaced or washed with first an acid (e.g., fairly concentrated HCl) and then a caustic solution (e.g., sodium hydroxide or caustic soda). The expense and continual acid + caustic washing didn’t sound so great.

So, reverse osmosis sounded the best. Claims of removal efficiency ranged from 30% to 80%. Eventually we settled with a company based in Santa Cruz, Hydro Logic. The owner suggested, instead of treating *all* the water, doing point-of-use systems; e.g., one under the sink for drinking water, and one for irrigation. Everything else (dishes, laundry, showers) could use the normal water (and, in fact, boron is a cleaning agent – you may have heard of borax…). Well, that sounded good. His company happens to manufacture a roughly 1000 to 1300 gallon per day (dependent on inlet pressure) RO system, the Evolution RO 1000, with a 3:1 or 2:1 (optional) ratio of waste:product water. That sounded pretty good. We don’t seem to be short of water with our well… just the quality is a bit lacking.

We decide first to do an under-sink one. We get one of their Stealth RO100 systems, along with a storage tank, a carbon filter (for taste), and a boron selective mixed bed resin cartridge that the owner was able to order directly (it uses a Purolite S-108 resin, which I had researched before in looking at that method of removal). We had the water tested: current water boron levels, post-RO agricultural suitability test, and post-boron-resin boron level.

The good news:
Unfiltered well water boron level: 43ppm (up from June levels)
Post-RO water boron level: 25ppm (~40% removal rate)
Post-boron-resin boron level: .4ppm (woohoo! WHO recommends .5 or less)

The bad news: the agricultural test was still high in boron (25ppm), not terribly low in sodium (60, only 60% removal), and still had a pH of 8.4.

The problem? Well, it turns out our pressure pump was set to kick-in at 30 and kick-out at 50… but our pressure tank was pre-charged to 38. Our pump *should* have been set to 40 and 60. So, after more research, we find out we need to turn a nut a bunch of turns (each full turn is approximately 2.5psi, as I recall… so 4 full turns). We do that and our pressure goes up to 40-60 rather than what it was doing, which was much lower – getting down at times to 20 to 25 even. By this time, we purchase a little TDS tester. We measure the well water (~380ppm on the handheld) and post-RO water (~20ppm). That’s better! We assume that this should also increase the boron removal rate, and go ahead and get the Evolution RO 1000 and booster pump (to raise inlet pressure to ~60psi).

We get it, set it up, run it for a hundred gallons or so, and send the water in for testing. We are mostly concerned with boron; everything else on the ag suitability test looked basically good (sodium was less than 10ppm, chloride less than 10ppm, alkalinity was down to 32.5ppm CaCO3; some necessary nutrients were also non-detectable, but we can always remineralize using calcite or something like that).

Well water boron level: 46ppm (again, up from August levels)
Post-RO water boron level: 32ppm (~30% removal rate).

Big letdown. We had hoped the boron level would be 6-8ppm (approximately 80% removal, assuming our level was still ~43ppm naturally). Having it so high was very, very discouraging.

At this point, we start looking at other options as well. I called the well driller, having learned about something called a packer or inflatable packer… basically, you stick these things down the well to isolate off sections of the well that you don’t want to get water from. Being 980′ deep, I figure maybe most of the boron comes from way down there. Well, the well casing is surrounded by gravel, so even if a packer was put in, cross-contamination would still occur. Basically, no well adjustments appear to be a solution. The driller does give the name of someone who has dealt with boron before and understands both sides (the well side, being a driller himself, and the filtration side, being a pump type stuff supplier). I am still waiting to talk to him, but that sounded possible.

Along with this, we take a look at rainwater (and our natural seasonal underwater spring, which we figure produces between 40,000 and 100,000 gallons per year… it runs pretty quickly and solidly during the rainy season). Storing 15,000 gallons of rainwater (since we would need enough, without having good quality supplemental water, to last all the dry season for irrigation) is expensive; 3x 5000 gallon above-ground tanks are approximately $6000 total, plus freight. Grading for a pad or digging if underground would be an extra cost, plus the necessary piping. An option, sure, but probably a $10,000+ option. Not something we want to spend right now; $1500-$2000 sounds nicer (if we can do it with reverse osmosis + some other extra equipment, we’d be looking at approximately $2000 total of equipment, including the extra 500 gallon storage tank as a buffer for the filtered water).

There was still one other option. Reverse osmosis is pretty good at getting rid of one form of boron – borate – but not so good at removing boric acid. Boric acid is prevalent as a percentage of total boron content at lower pH levels, whereas borate is prevalent at higher pH levels. At our pH, approximately 8.8, the borate level is supposed to be approximately 30%; almost exactly the removal rate we were getting! At about 9.24, the percentage of borate reaches 50%, and by a pH of 10, it’s at 90%… and 95% at 10.5. See the graph on this page from Lenntech.

I have also read a couple desalination plant papers that included successful boron removal from seawater down to drinkable levels (definitely below .5ppm). Most of them had a two-pass RO system and used crazy membranes that required operating pressures of 800psi … but that was partially because of the extreme salt content (sea water has a TDS content of around 35,000ppm!). They usually used a second-pass RO system mostly to get the boron out… what they did was *raise* the pH to 10+ before the second RO pass in order maximize the percentage of boron in the form of borate.

Putting two and two together, as the saying goes … we are currently hoping that by injecting a caustic solution (potassium hydroxide, preferably, as it would have a smaller chance of precipitating out and causing scaling on the membrane, and potassium is something plants need anyways – as opposed to the sodium in sodium hydroxide) into the inlet water to the Evolution RO 1000. This in turn, providing the pH is high enough, will hopefully reach to somewhere around a 90% removal rate… especially if we can get the pH above 10.

So, the next steps are acquiring an offered sample of sodium or potassium hydroxide (rather than purchasing in bulk for just this small test) and generating a “titration curve” – basically, figuring out the amount of a 10% solution of sodium or potassium hydroxide necessary to raise our water pH level to 10.5. Once we have that – e.g., let’s say it required 3ml of the 10% solution to raise 100ml of our water to a pH of 10.5 – we can figure out what strength of solution we would need for an injection system with a rate of 100:1, which is fairly typical for fertilizer injection equipment. I have already checked with EZ-Flo, who said their fertilizer injectors should be able to handle sodium hydroxide.

If it is still feasible, and doesn’t require working with totally dangerous caustic solutions that cause your skin to fall off by just looking at it… then we would perform an actual test with injection equipment, the 100:1 solution, and the RO system. Run it for a while, then get a sample and have the boron level tested. We will also test the pH level ourselves to determine if we’ll need a post-RO acid injection to lower the pH to somewhere around 7.

Included in the necessary equipment for the titration curve, in case you are wondering, is:
* 100mg of sodium or potassium hydroxide; hopefully picking up tomorrow afternoon.
* Jewelry scale to measure grams of sodium or potassium hydroxide in order to create a 10% solution. * Shipping tomorrow
* Nitrile gloves for chemical resistance and safety. No caustic burns! * Received
* Chemical splash guard goggles. No eye replacements! * Received
* pH tester and pH calibration fluid. * Received

We are somewhat hopeful, as our water seems to like remaining at a high pH; in fact, even after the RO, it was still approximately 8.1; after the RO + boron exchange resin, it actually went back up to 8.9. Since it has apparently a fairly stable *high* pH, we are hoping it will be relatively easy to coax it higher.

Probably more than you ever wanted to know about our water escapades, but there you go. I have learned way more than I wanted to about water filtration, even getting into chemistry things like molarity and normality.

facebook   twitter   email