A 2026 review in Agriculture is unusually open about the limits of its own subject. It gives silicon recommendations for each growth stage. It sets out the chemistry limits that decide what you can actually put in a tank. Then it lists the silicon claims the evidence does not support, and one of them is the claim most silicon products are sold on. A separate primary study shows that root-applied silicon suppresses hemp powdery mildew, and that is the strongest practical case for using it. This article keeps the disease case and the potency case apart, because they are not equally supported and products are usually sold on the weaker one.
1. Disclosure first
This account belongs to AZ Growshop. We distribute silicon products in Thailand, including Gro-Silic and Advanced Nutrients’ RSA, which is discussed below. We say this up front because this article asks whether silicon works and what it works for, and you should read it knowing we sell some.
We handle that conflict in two ways. We give the rejected claims as much space as the supported ones, and we say plainly which claim silicon products are usually marketed on. If that loses us a sale to someone who wanted silicon for potency, that is the right outcome. The evidence does not support buying it for potency. A customer who buys it for the wrong reason and sees no potency change stops believing anything else we say.
2. What the review supports
Malík, Hoffmannová & Tlustoš (2026), “Integrating Silicon into Fertigation Strategies for Cannabis Production: A Comprehensive Review”, Agriculture 16(14):1522.
Recommendations by growth stage:
| Stage | Form recommended |
|---|---|
| Propagation and early growth | stabilised monosilicic acid (fast uptake, very little EC increase) |
| Vegetative | potassium silicate |
| Long-term substrate supply | calcium silicate / wollastonite |
Dose: 10 mg/L silicon was effective in fibre hemp. The review describes the dose for medicinal flower as “understudied”.
Measured biostimulant effect: about 2.1× tissue silicon and about 1.2× inflorescence biomass, but the biomass effect was only marginally significant. The treatment also combined silicate with phosphite, so the yield response cannot be put down to silicon alone. Both qualifiers come from the review itself. Do not round this up to “20% more yield”.
The review also lists powdery mildew suppression, lower root uptake of cadmium and zinc, and stronger stems among the effects the evidence supports.
3. The chemistry constraints, which decide what is physically possible
This is the most immediately useful part of the review. It explains tank failures that growers usually blame on the product.
- Potassium silicate concentrate is strongly alkaline, roughly pH 11–13 in the bottle. Even at working dilution it pushes tank pH up sharply.
- Below about pH 9, monosilicic acid polymerises once it goes above roughly 2 mM. Above pH 9, silicate ionises and stays far more soluble. That is why silicon concentrates are either high-pH or stabilised.
- In a working solution at growing pH, the practical ceiling before gelling is roughly 47–61 ppm elemental silicon.
What follows from these three points:
Add potassium silicate first, to plain water, and re-adjust pH before anything else goes in. Where the alkaline silicate meets calcium and magnesium, they come out of solution. That is where the white sludge at the bottom of a tank comes from. It is usually a mix of calcium silicate, silica gel (polymerised silicic acid) and pH-driven carbonate and phosphate precipitates. It does not mean the product is contaminated.
There is a limit on how much silicon a working solution can hold. The 47–61 ppm ceiling applies to the final working solution at growing pH. It does not apply to percentages on a concentrate label. A high-pH silicate or a stabilised monosilicic acid concentrate can legitimately contain far more. RSA, for example, lists 13% soluble silicon. The ceiling only applies once the product is diluted into your tank. So check your final tank ppm Si against it, not the bottle.
Monosilicic acid’s advantage is that it stays stable at usable pH. That is why the review puts it at propagation, where young roots cannot tolerate a pH swing.
Calcium silicate and wollastonite are substrate amendments, not tank additives. They release slowly and avoid the tank chemistry problem entirely. For living-soil or long-cycle growers this is the low-risk route.
4. The four claims the review rejects
- Silicon raises cannabinoid or terpene biosynthesis. Not supported.
This is the claim most silicon products are marketed on, in some form: “denser buds”, “more resin”, “increased potency”. The review notes that no peer-reviewed study isolates silicon’s effect on cannabinoids or terpenes under standardised conditions. Secondary-metabolite responses also seem to depend on plant stress, so results from drought trials do not carry over to a controlled room.
- Nano-silicon is better at the same dose. Not supported.
The review credits foliar nano-silicon with improving drought tolerance. It also says nano-silicon products are poorly characterised, and their effects may come from dissolved silicon or co-formulants rather than the particles. A 2025 outdoor foil-tunnel trial in the Journal of Cannabis Research was mainly an irrigation study, but it also applied silver, copper and iron nanoparticles. It reported no statistically significant effect on inflorescence yield or CBD. Those were not silicon particles, so the trial does not confirm the review’s point about nano-silicon. It does fit the same caution. Treat “nano” as a marketing prefix until a controlled trial shows it beats the same element at the same dose.
-
Glandular trichome enrichment. Not supported.
-
Late-flower silicon application is safe. Not established.
The review states that the safety of late-flowering silicon application is undocumented. That does not mean harmful. It means nobody has published the study. Our own reasoning goes further. Silicon builds up in plant tissue as amorphous silica, and that is how it strengthens cell walls. We know of no safety assessment for silica deposited in flower tissue that is later combusted and inhaled. Where safety is unknown and the product will be inhaled, we think caution makes more sense than giving it the benefit of the doubt.
Our practical position, and it is our inference: front-load silicon and stop before flower or early in flower. The plant builds the stems that carry the weight during vegetative growth and early stretch. That is when silicon’s structural benefit is delivered. Late in flower there is no supported benefit, and the safety question is still open.
5. Choosing a form: monosilicic acid vs potassium silicate
RSA is an example of the stabilised monosilicic acid route. The manufacturer lists 13% soluble silicon at 0.5 mL per gallon, about 0.13 mL/L. That is a dilution of about 1:7,700, applied once a week through both veg and bloom. The manufacturer claims stronger stems within 7–14 days and says the product maintains clean delivery through drip lines and filters. We found no published test data behind either claim.
The trade-offs:
- Potassium silicate is cheaper per unit of silicon, and the review recommends it for veg. It also costs you pH correction (more acid, one more mixing step). It adds potassium and EC, so it is not nutritionally neutral if your feed is already at target K. In drip systems, it can precipitate in emitters.
- Stabilised monosilicic acid costs more per unit of silicon. The review notes fast uptake and very little EC increase. At 0.13 mL/L, the effect of a product like RSA on pH and EC should be small, but check with your own meter. Stabilised products can be acidic or contain stabilisers.
Which one is right depends on whether you run drip irrigation and whether you have EC headroom, not on efficacy. If potassium silicate is not blocking your emitters or forcing extra acid, you already have a working silicon programme, and switching is optional.
On timing, the review places monosilicic acid at propagation and early growth. The RSA label runs it weekly to the end of bloom, a period the literature has not examined. If you follow the label, you are following the label. Whether to stop early, as we do, is a judgement call on an open question. The research does not show the label is wrong.
6. What silicon is actually good for
Powdery mildew suppression is the review’s best-supported effect, and its disease evidence covers powdery mildew only. The clearest primary study is Dixon et al. (2022) in Plant Health Progress. Silicon was applied to hemp roots in a peat-based soilless mix. Plants with more silicon in their leaves had less powdery mildew, and severity fell significantly at the higher rates.
Note “root-applied”. That was the delivery route in the study. Silicon is thought to suppress disease after the plant takes it up and deposits it in tissue. A foliar spray is a different intervention with a weaker evidence base.
How useful this is in Thailand needs care. Thai ambient conditions supply the humidity powdery mildew likes, but not the moderate temperatures it prefers. Commonly published cannabis clone setpoints are about 20–24 °C and above 90% relative humidity. Ambient Thai temperatures of 28–35 °C sit well above that. So a propagation or veg area becomes a real powdery mildew risk mainly where it is air-conditioned, kept under humidity domes, or both. In those spaces, the disease risk and the review’s monosilicic acid recommendation for propagation point the same way. Some growers report powdery mildew as rare in hot outdoor Thai conditions, with bud rot the bigger threat. That is anecdotal, but if it holds, the case for silicon depends on your room more than on the region.
The limit: silicon is not a fungicide. Airflow and humidity control remain the main levers. Silicon suppressing mildew does not mean silicon can replace a dehumidifier.
The structural benefit is real, but it should be described honestly. Silicon strengthens cell walls, which helps stems carry heavy flower. In the review, the inflorescence effect was about 1.2× and only marginally significant. That is a modest, weakly supported yield effect, worth having but not worth building a programme around.
7. The summary table
| Claim | Status |
|---|---|
| Suppresses powdery mildew, root-applied | supported (review; primary study Dixon et al. 2022) |
| Strengthens cell walls / structural support | supported as mechanism; biomass effect ~1.2×, marginally significant, confounded by phosphite |
| Raises tissue silicon ~2.1× | supported |
| 10 mg/L effective in fibre hemp | supported |
| Dose for medicinal flower | “understudied”, no figure |
| Raises cannabinoid or terpene biosynthesis | not supported |
| Nano-silicon superior at equal dose | not supported; a metal-nanoparticle trial is consistent with this caution |
| Glandular trichome enrichment | not supported |
| Safe in late flower | undocumented |
8. What we would actually recommend
Use it in propagation and vegetative growth: stabilised monosilicic acid in propagation, potassium silicate in veg, or calcium silicate in the substrate if you run long cycles and would rather not manage tank chemistry.
Buy it for disease suppression and structure, not for potency. The potency claim is on the not-supported list.
Stop before flower or early in flower, because late-flower safety is undocumented.
Add potassium silicate to plain water first and re-adjust pH before anything else. Otherwise you will precipitate your calcium and magnesium and blame the product.
Check your final tank ppm Si against the 47–61 ppm ceiling. It is a physical limit for the working solution, not for the concentrate in the bottle.
Ignore the nano premium until someone publishes a controlled comparison at equal dose.
In Thailand, use only legally registered silicon fertilisers, and on a GACP-certified site, only inputs allowed under the farm’s GACP input rules.
If you test silicon yourself, measure the two things it plausibly affects:
- powdery mildew incidence, with and without silicon, in the same Thai room
- stem diameter with calipers, or a lodging count. The 7–14 day stem claim is a cheap, short-term thing to test.
Do not spend money on a lab test to see whether silicon raised your cannabinoid numbers. No cannabis study has isolated that effect.
Has anyone here run silicon and no silicon side by side in a Thai propagation or veg area and seen a difference in powdery mildew? That is the silicon claim with real support behind it, and also the one where the tropical picture is unclear. We sell silicon, and we would rather know than guess.
Related reading: topic 1254 (powdery mildew); topic 109 (Gro-Silic); topic 1082 (NPK, macro and micronutrients); topic 1301 (IPM); topic 1046 (climate control and VPD); topic 1031 (hydroponics).
This article is technical horticultural information for readers operating legally in their jurisdiction. AZ Grow distributes silicon products, including Gro-Silic, and this disclosure is stated in the opening section. Product registration and permitted inputs differ by country. In Thailand, GACP guidelines restrict permitted crop-protection inputs, and cannabis inflorescence is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568. Nothing here is legal advice.