Knowledge Centre · Planted aquariums
CO2 for planted aquariums: injection, safety and diagnosis
CO2 injection is genuinely optional for most planted tanks, and treating it as a default upgrade rather than a considered decision is where a lot of avoidable risk to fish enters the hobby. Undemanding plants grow well on the CO2 already present from fish respiration and surface exchange; injection speeds growth and widens which plants are realistic, at the cost of equipment, complexity and a real overdose risk that veterinary sources describe beginning at concentrations lower than commonly cited hobbyist targets. This guide covers whether you need it, how the two main delivery methods compare, and how to run it safely if you decide to.
At a glance
Is it required?
No, for most low-to-medium light setups. It's an optional upgrade that speeds growth and enables more demanding plants.
Main risk
Overdose (hypercarbia) is a genuine, not theoretical, risk to fish — especially overnight if injection isn't automated to stop with the lights.
What this page won't do
Recommend a specific bubble-count or ppm figure as universally safe for your exact tank — direct monitoring matters more than a fixed number.
Do you need CO2 injection?
Plants can photosynthesise using CO2 already dissolved in the water from fish respiration, organic decomposition and surface gas exchange — genuinely, not as a compromise. Java fern, Anubias, Cryptocoryne, Vallisneria and most beginner-friendly species grow successfully without any supplemental CO2[1]. Injection generally speeds growth, deepens colouration in some species, and widens which plants and light intensities are realistic — carpeting plants and demanding red stems in particular usually need it to thrive rather than merely survive. If you're keeping a low-to-medium light tank with undemanding species, starting without CO2 and adding it later if you want faster growth or more demanding plants is a reasonable, genuinely low-risk approach.
Pressurised CO2
A pressurised system uses a CO2 cylinder, a regulator to control pressure, a solenoid valve to allow automated on/off timing, and a diffuser to introduce the gas into the water as fine bubbles or through a reactor. This is the more effective, more controllable method, capable of sustaining a consistent CO2 level for demanding plants and full carpets. The trade-off is upfront cost and setup complexity — more components to source, plumb in and understand — and it's the method where the overdose risk covered below is most relevant, since it can genuinely push CO2 to dangerous levels if left unmonitored or unautomated.
Liquid carbon
Liquid carbon products, commonly glutaraldehyde-based, are dosed directly into the water column as a liquid additive rather than injected as a gas. They're simpler and cheaper to start with, needing no cylinder or plumbing, but deliver a smaller, less sustained carbon boost than pressurised injection and generally can't support the most demanding, high-light carpeting plants on their own. They carry their own safety consideration too: some plants, most notably Vallisneria, are documented as sensitive to glutaraldehyde-based liquid carbon and can be damaged or killed by it even at label doses[2] — worth checking against your specific plant list before dosing, not assuming a "carbon supplement" is universally plant-safe.
Comparison at a glance
| Method | Effectiveness | Cost and complexity | Key risk |
|---|---|---|---|
| None (ambient CO2 only) | Supports undemanding, low-to-medium light plants well | None | None specific to CO2 |
| Liquid carbon | Modest boost; not enough for the most demanding species | Low cost, no equipment | Toxic to Vallisneria and some sensitive species at label doses |
| Pressurised CO2 | Sustained, adjustable, supports demanding carpets and stems | Higher upfront cost, more setup | Genuine fish-toxicity risk (hypercarbia) if unmonitored |
Monitoring: drop checkers and direct observation
A drop checker — a small chamber holding indicator fluid (commonly 4dKH reference fluid with a pH-sensitive dye) that sits in the tank and changes colour based on dissolved CO2 in the surrounding water — is the most widely used practical indicator for pressurised CO2 setups. It's a genuinely useful visual check, but it lags real-time conditions by roughly an hour or more, so it tells you where CO2 has been recently, not necessarily where it is at this exact moment. Direct observation of your fish matters alongside it: gasping at the surface, unusual lethargy, or laboured breathing during CO2-on hours are signs to act on immediately, regardless of what the drop checker's colour currently shows.
The overdose risk, taken seriously
If you suspect fish are already showing distress from excess CO2, our fish gasping at the surface guide and fish breathing heavily guide cover the wider set of possible causes and immediate steps beyond CO2 specifically.
Automating CO2 with your lighting
Plants only consume CO2 during active photosynthesis, which stops once lights go off — continuing injection overnight adds no growth benefit and steadily raises dissolved CO2 with nothing actively removing it, precisely when livestock are least likely to be observed. A solenoid valve wired to the same timer as your lighting is the standard way to automate this, typically set to start CO2 an hour or two before lights on (so levels are already adequate when photosynthesis begins) and stop with, or slightly before, lights off. This single piece of automation removes one of the more common, avoidable causes of overnight CO2 buildup.
When plants aren't growing despite CO2
Adding CO2 without also addressing light and nutrients rarely fixes poor growth on its own, since all three inputs work together. If plants are still struggling once CO2 is in place, check your lighting is adequate for the species and your fertilising routine is actually supplying the macro and micronutrients those plants need — and see our nutrient deficiency guide if specific leaf symptoms (yellowing, holes, stunted new growth) point toward a particular missing nutrient rather than CO2 itself.
Common mistakes
- Adding CO2 as a default "upgrade" without a plan for monitoring it, treating overdose as a remote possibility rather than a genuine risk.
- Leaving CO2 running overnight without a solenoid valve tied to the lighting timer.
- Dosing glutaraldehyde-based liquid carbon in a tank with Vallisneria without checking product compatibility first.
- Relying on a drop checker alone without also watching fish behaviour directly, given the checker's lag behind real-time conditions.
- Increasing CO2 to fix poor growth without also checking whether light or nutrients are the actual limiting factor.
Frequently asked questions
Do I need CO2 injection for a planted tank?
Not for most beginner setups. Undemanding plants like Java fern, Anubias, Cryptocoryne and Vallisneria grow successfully using only the CO2 already dissolved in the water from fish respiration and surface gas exchange. CO2 injection generally speeds growth and widens which plants and light levels are realistic, but it's an optional upgrade that adds cost, equipment and a genuine safety consideration — not a requirement for a healthy low-tech planted tank.
How do I know if my CO2 level is safe for fish?
A drop checker using 4dKH reference fluid is the most commonly used indicator, showing green at a level close to commonly cited injection targets. Watch your fish directly as well: gasping at the surface, unusual lethargy or laboured breathing during CO2-on hours are signs to reduce or pause injection and increase surface agitation immediately, since these can indicate CO2 levels are already too high for comfort, regardless of what a drop checker shows.
What's the difference between pressurised CO2 and liquid carbon?
Pressurised CO2 injects the gas directly into the water via a cylinder, regulator and diffuser, giving precise, adjustable, sustained CO2 levels — the more effective option for demanding plants, at higher upfront cost and setup complexity. Liquid carbon (commonly glutaraldehyde-based) is dosed directly into the water as a liquid additive, is simpler and cheaper to start with, but delivers a smaller, less sustained carbon boost and carries its own toxicity considerations for some sensitive plants and invertebrates at higher doses.
Why turn off CO2 at night?
Plants only consume CO2 during active photosynthesis, which stops once the lights go off. Continuing to inject CO2 overnight adds no benefit to plant growth and steadily raises dissolved CO2 with nothing removing it, increasing the risk of it reaching an unsafe level for fish while they're least likely to be observed. A CO2 solenoid valve connected to your lighting timer is the standard way to automate this.
How AquaPulse can help
A CO2 setup that's been running fine for months can still drift — a diffuser fouling, a regulator seal ageing, a solenoid timer slipping out of sync with lighting. None of that is obvious by looking at the tank on any single day; it shows up as a pattern of fish behaviour or plant health changing gradually, which is exactly the kind of change that's hard to notice without something to compare against.
AquaPulse works from the readings and context you enter — it doesn't monitor your CO2 system automatically. Logging notes on equipment checks and any behaviour changes in AquaPulse alongside your regular water testing keeps that history available, so a gradual shift is something you can trace back rather than notice only once it's become a serious problem. Get AquaPulse on the App Store or Google Play to try it, or see the Water Check feature page for more detail.
Sources and uncertainty
- Published
- 8 August 2026
- Last reviewed
- 8 August 2026
- Written and maintained by
- AquaPulse (Valeon Labs Ltd)
This guide draws on the Merck Veterinary Manual for the hypercarbia toxicity risk (the same citation already used in our starting a planted aquarium guide) and specialist aquascaping sources for practical CO2 delivery-method and liquid-carbon-sensitivity detail. We've deliberately avoided stating a single "safe" ppm figure as guaranteed for every tank, since the veterinary evidence indicates risk can begin below commonly cited hobbyist targets and individual tank conditions vary. We'll update this page if we find something we've got wrong.
Good to know: This guide provides general informational guidance only. It is not a veterinary service and does not replace specialist advice if livestock show signs of serious or persistent distress.
- [1] AquaPulse — "Starting a planted aquarium" Our own already-published, separately sourced guide. Used for CO2 injection being genuinely optional for undemanding species rather than re-deriving the same research here.
- [2] Aqua Essentials — "Aquarium Plants That Don't Need CO2 (Easy Beginner Plants)" Specialist aquatic-plant retailer source. Used for Vallisneria's documented sensitivity to glutaraldehyde-based liquid carbon products. Not a scientific authority; presented as widely reported hobby knowledge.
- [3] Merck Veterinary Manual — "Environmental Diseases of Aquatic Animals in Aquatic Systems" Institutional veterinary reference. Used for dissolved CO2 (hypercarbia) toxicity risk to fish beginning at concentrations lower than commonly cited hobbyist CO2-injection targets — the same citation already used in our starting a planted aquarium guide.
What we deliberately left out: a bubble-count-to-ppm conversion table, since that relationship depends heavily on diffuser type, tank shape, flow and surface agitation, and presenting a fixed conversion would imply a precision the underlying physics doesn't support — direct measurement with a drop checker and observation of livestock is more reliable than any generic bubble-count figure; and a single universally "safe" ppm target, for the reasons explained in the safety section above.