Since Happi has raised some important questions here, I'll try to join the discussion exceptionally to explain a bit my reasons why I decided to use such low concentrations and also sodium (which apparently is not an essential nutrient).
1) I have quite a lot of experience with failure. You could say I am a champion at failing at growing aquarium plants. Because of this, I am constantly asking myself why certain plants fail in my aquariums, what is to blame, while trying to pinpoint
the factors that might play the biggest role. Most modern aquarists blame most of the failure on
CO2 or various deficiencies, so the EI method has taken hold, which primarily considers CO2 (in combination with unlimited nutrients) as the key factor. I have never been able to confirm this assumption in my experiments. Or, to be more precise, I have managed to confirm that
CO2 does play a significant role, but not in high concentrations (= 10-15 ppm is abundantly sufficient for optimal growth). Similarly, I have verified that the
plants [which I had the opportunity to test] need much less nutrients for good growth than is generally recommended in plant aquarium science today. These conclusions are perfectly logical and theoretically valid even without any practical verification (just like the theory of relativity was valid before it could be verified in practice => practice will only confirm it, but we don't need practice to know that it is true; logical [theoretical] reasoning is plenty sufficient for that
... though of course I understand that some more practical-minded people just need to see it for themselves before they believe the logical reasoning). In fact,
plant growth is not linear, but logarithmic, which means that every slight increase in nutrient concentration leads to a several-fold increase in growth, but the more nutrients we add, the less we get (see the law of
diminishing returns). So, it was clear to me from the beginning that no high concentrations of CO2 or other nutrients are necessary (and in fact cannot be optimal in principle). I consider a 50% growth (or photosynthesis) rate to be optimal, which is referred to in the literature as Km or K 1/2 (so-called "half saturation point").
This picture shows a certain law of nature. If we understand this law, we do not need any practical experiments to know that high nutrient concentrations are not necessary for good plant growth. Practical experiments can only confirm this. There are many similar [theoretical] observations of such laws of nature in physics, chemistry, biology and ecology. We only need to find them, understand them and properly apply them in practice. At least that's what I'm trying to do.
Anyway, if nutrients are not the real problem,
what is the problem? I didn't know this, so I wondered how to devise an experiment that would figure this out (or at least help us discover something useful). From the scientific literature, I tried to pick out the various factors that scientists are studying and finding to have a big effect on plants. The factors with the greatest influence are definitely
pH (which is overlooked by today's aquarists a lot), redox, bacteria, [dissolved] organic matter, and
bicarbonate.
The possible influence of
bacteria has been very nicely raised by
Sudipta Shaw, but so far we know very little about their influence and it is very difficult to test this experimentally. I must therefore leave this area to someone more qualified. It is similarly difficult to test
redox. It is still possible in water, but measuring it in a substrate is a problem. The redox (together with pH) decides which substances are reduced and which are oxidized in the substrate, which significantly influences the (un)availability of nutrients and their possible toxicity (and in turn their possible deficiency in water). Similarly, the content of labile
organic matter in the substrate or the dissolved organic matter content in the water can have a significant effect. Both can [under certain circumstances] be toxic to plants. Just use a substrate that has a higher organic content (>15%) and most of your plants can die in it.
[BTW, although I have mentioned it elsewhere, even the same two substrates from the same brand, e.g. ADA Amazonia, may not have the same composition, which was quite unpleasant discovery for me. So while Joe's plants may grow beautifully in his ADA Amazonia substrate, Johnny's may stagnate or die in the same-brand substrate.] Aquarists often focus on the water, so they use e.g. activated carbon or Purigen (and similar products) to remove dissolved organic matter from the water. But they completely overlook the same problem in the substrate.
If you think of the substrate as an alcohol factory (where the alcohol is the organics, for example), then using activated carbon or Purigen means removing all the bottles of alcohol from the city. But if you don't shut down the distillery, then you are only addressing the effects and ignoring the cause ...
So, since I don't have the proper equipment to examine bacteria, soil redox, or organics, I decided to focus primarily on
pH and bicarbonate. Of course, given this limited scope, it is highly likely that I will not be able to detect all the key factors that have a major influence on aquarium plant growth. I take this into account and therefore certainly do not claim that pH and bicarbonate are the most important. Anyway, I believe that these two factors can also provide us with important pieces to the overall mosaic. And as it turns out, I was right => pH and bicarbonate are certainly among those key factors for (un)success (as opposed to high CO2 concentration and non-limiting concentration of other nutrients). So my last two experiments are just a kind of starting point (stepping stone) that can point us in the right direction. This should be kept in mind at all times when evaluating them! In no case can any definite conclusions be drawn from them. It is also good to remember that I have only tested a few plant species in them so far. Other species might turn out differently in them. Still, I'm very happy with how the experiments turned out. I think I was able to select certain model plant species with different preferences. My only regret is that I couldn't get
Rotala wallichii for the second experiment. I bought the plants for the experiment before Christmas and most were sold out. And the only dealer who still offered
R. wallichii ended up sending me
Rotala sp. Vietnam instead (which I only found out at home and there was no time to deal with it). Still, the results provide a number of useful insights => e.g. about the (non)importance of high nutrient concentrations, the (non)importance of the substrate for some plant species, algae growth dynamics, root development vs inhibition etc., and especially the effect of pH and bicarbonate. So much for explaining my motivations and goals.
2) As for why I used
sodium (Na) in my recipe, it was mainly because I hadn't tested whether it might be of any use to some of the plants I tested. Sodium is quite common in natural waters, and is quite abundant in tropical waters. Therefore, I didn't want to leave it out completely (even though the literature says it is not essential). In addition, I had planned to have
Ammannia pedicellata 'Gold' in my experiment also, not much was known about its preferences. So I didn't want to take the chance that maybe sodium might be important to her, and I didn't figure it out just because I'd missed it. It's certainly better to add it than not. In fact, adding it [in reasonable amounts] can't do any harm, unlike not adding it.
3) As for
microelements, don't look for anything special in that. Again, I made the simple assumption that the vast majority of microelements are present either in the tap water, or in the substrate, or [in the form of admixtures] in the bulk chemicals from which I prepare my fertilizer. Cobalt and molybdenum are required by plants in such negligible quantities that it probably doesn't make sense to bother adding them to fertiliser at all. Trying to eliminate them from the aquarium is virtually impossible. You can get enough of them in there from the air (on dust particles, for example). I still prefered to add cobalt, but not molybdenum. The other microelements are already needed by the plants in slightly higher amounts, so I definitely added those. Only now I can see in retrospect that I made a mistake: most of the microelements are chelated with
EDTA, only the iron is chelated with
DTPA (and I added a smaller part in the form of Fe-gluconate for good measure, because iron has the greatest tendency of the microelements to oxidize and precipitate).
4) As for my own assessment, I don't think all the concentrations I used in the second test (= lean fertilisation) are completely optimal. I would probably add a bit of nitrogen and proportionally some phosphorus. The 3 ppm (split in half between NH4 and NO3) seems a bit low to me, which was a bit evident on the Hygrophila. In future experiments I would probably try to raise it to twice that. Similarly, I might raise the calcium and magnesium to 10 + 3 [just to be safe]. Such concentrations might be optimal. But take it with a grain of salt.
5)
A.pedicellata was shedding its lower leaves quite a bit (much more in the first experiment where I used high nutrient concentrations; there even some stems were downright rotting). I don't know the reason for this and I don't dare to guess. I simply have no idea.
... enjoyable and interesting.