CO₂ at the leaf: the physics of carbon starvation in planted aquariums
Your drop checker reads green. Your CO₂ injection is running. Your plants are still not thriving. The explanation lies in a layer of water so thin it is invisible — and yet it controls how much carbon each plant actually receives.
The standard guidance in the planted aquarium hobby is to target 30 ppm dissolved CO₂. This number is sensible as a bulk water target — it is high enough to support active photosynthesis in most plants, low enough to remain safe for fish with reasonable headroom. But it is a measurement of the water column, not of what plants actually consume at their leaf surfaces. The two are not the same, and the difference between them has practical consequences for how you inject, circulate, and choose plants.
Why water makes CO₂ delivery fundamentally difficult
In terrestrial plants, CO₂ moves from the surrounding atmosphere into the leaf through stomata — pores on the leaf surface — driven by the concentration gradient between air (approximately 400 ppm CO₂) and the interior of the leaf (where photosynthesis keeps concentration lower). This works efficiently because CO₂ diffuses rapidly through air: the diffusion coefficient of CO₂ in air at 25°C is approximately 1.6 × 10-5 m2/s.
In water, the same diffusion process is dramatically slower. The diffusion coefficient of CO₂ in water is approximately 1.9 × 10-9 m2/s — roughly 10,000 times lower than in air.[1] This is not a minor scaling factor. It means that CO₂ in the water column moves slowly enough that the supply rate to the leaf surface can become the limiting step in photosynthesis, even when bulk water concentration appears adequate.
Aquatic plants do not have stomata in the same way terrestrial plants do. CO₂ enters submersed leaves directly across the leaf surface — through the cuticle and cell walls — by diffusion driven entirely by the concentration gradient between the water and the leaf interior. There is no active pumping mechanism. The rate of delivery is governed by diffusion physics.
What the boundary layer is and why it forms
Even in a circulating aquarium, the water immediately adjacent to any solid surface — including a plant leaf — moves more slowly than the bulk water column. This is a fundamental property of fluid dynamics: viscosity causes water to slow near a surface, forming a laminar sublayer where momentum from the bulk flow does not penetrate. The result is a thin film of relatively still water — the diffusive boundary layer — that clings to the leaf surface regardless of how fast the surrounding water moves.
Within this boundary layer, CO₂ can only move by molecular diffusion. There is no advective transport (bulk flow carrying CO₂ in). The plant at the inner surface of this layer is actively consuming CO₂ in photosynthesis, pulling the concentration down. CO₂ must diffuse from the bulk water through the boundary layer to replenish what the plant uses — and because diffusion in water is slow, a concentration gradient develops: high CO₂ in the bulk water, progressively lower CO₂ as you approach the leaf surface.
This gradient has been measured directly. Using microsensors positioned at increasing distances from aquatic plant leaves during active photosynthesis, researchers have shown that CO₂ concentration at the leaf surface can fall to a small fraction of the bulk water concentration during active photosynthesis — even when bulk water CO₂ is at levels considered adequate.[2]
What the measurements show
Microsensor studies of aquatic plant leaves during high-light photosynthesis consistently find CO₂ concentrations at the leaf surface that are 80–95% lower than the surrounding bulk water. In a tank running 30 ppm bulk CO₂, the plant surface may be experiencing 2–6 ppm. At 2 ppm, most aquatic plants cannot sustain high photosynthetic rates — the carbon supply has become the limiting factor, not light or nutrients.[2]
The thickness of the boundary layer depends on flow velocity: faster water movement produces a thinner boundary layer, which reduces the diffusion path length and therefore increases the CO₂ supply rate to the leaf. This is why circulation matters for CO₂ delivery — not merely for distributing CO₂ around the tank, but for thinning the boundary layer at each leaf surface. See the guide to lily pipe placement and flow geometry for how outlet positioning controls whether the whole tank benefits from this effect, or only the area near the filter return.
"A tank at 15 ppm CO₂ with excellent circulation can deliver more carbon to plant leaves than one at 30 ppm with poor flow — because the boundary layer is the bottleneck, and flow rate determines its thickness."
How aquatic plants cope with carbon limitation
The challenge of CO₂ delivery under water is not new to evolution. Aquatic plants have been dealing with it for hundreds of millions of years and have developed several strategies in response. Understanding which strategy a given plant uses explains much of the hobbyist experience with "easy" versus "demanding" plants.
Direct CO₂ uptake is the baseline: the plant absorbs dissolved CO₂ directly from the water across the leaf surface. This is the most metabolically efficient strategy — CO₂ enters the Calvin cycle directly — but it makes the plant entirely dependent on the CO₂ concentration gradient and boundary layer dynamics. Plants that rely exclusively on direct CO₂ uptake typically require CO₂ injection to grow well in typical aquarium conditions.
Bicarbonate use is a more flexible strategy used by roughly half of aquatic plant species. Bicarbonate (HCO₃-) is present in most natural freshwater bodies — and in most aquariums — at far higher concentrations than dissolved CO₂. Plants that can use bicarbonate convert it to CO₂ inside the leaf using the enzyme carbonic anhydrase, effectively tapping a much larger carbon reservoir. However, this carries two costs: it requires additional metabolic energy, and it acidifies the leaf boundary layer (producing OH- as a by-product), which can have secondary effects on nearby water chemistry.[3]
Species that use carbon concentrating mechanisms (CCMs) — including bicarbonate use — have been extensively studied in the aquatic plant literature. Research on Ottelia species, which span a range of CCM strategies, shows that plants with well-developed CCMs can sustain higher photosynthetic rates in low-CO₂ conditions by extracting and using bicarbonate rather than waiting for dissolved CO₂ to diffuse in.[4] Species without this capability are far more sensitive to CO₂ supply and boundary layer effects.
Why this explains the "easy plant / demanding plant" divide
Plants commonly described as easy — hornwort (Ceratophyllum), elodea (Egeria densa), most Vallisneria, water sprite (Ceratopteris) — are typically effective bicarbonate users. They have evolved to function in the low-CO₂ conditions of natural water bodies. CO₂ injection improves their growth but is not essential.
Plants described as demanding — Hemianthus callitrichoides (HC Cuba), Glossostigma elatinoides, most Rotala cultivars, Eriocaulon species — are generally poor bicarbonate users or entirely dependent on dissolved CO₂. They require not just CO₂ injection but good circulation and a thin boundary layer to sustain growth. Without it, even a well-running CO₂ system may fail to deliver enough carbon to these plants.
Practical implications for CO₂ injection and tank circulation
The boundary layer physics have several direct practical implications that are underappreciated in standard hobbyist advice.
The 30 ppm target is a minimum floor, not a guarantee. Reaching 30 ppm in the bulk water is necessary but not sufficient for demanding plants in areas of poor circulation. The bulk concentration sets the maximum possible delivery rate — but the actual delivery rate is determined by boundary layer thickness, which is determined by local flow velocity. A reading of 30 ppm tells you about the water column; it does not tell you whether your carpeting plants under a shaded overhang are receiving adequate carbon.
Flow distribution matters more than total flow rate. A single high-output lily pipe creates excellent circulation near its outlet and in the main gyre, but plants in dead spots — corners, behind hardscape, under dense stem plant overhangs — may still experience boundary layer conditions that limit carbon supply. The guide to flow in planted tanks covers how to evaluate and improve distribution, including the case against the oversimplified "10× turnover" rule.
Stability interacts with boundary layer effects. A fluctuating CO₂ supply that swings between 15 ppm and 35 ppm during an injection session delivers average bulk water CO₂ of approximately 25 ppm — but because the plant is carbon-limited during the dips, those periods of low CO₂ translate to complete stalling of photosynthesis in demanding species. A steady 25 ppm outperforms an oscillating 15–35 ppm in practice. This is one of the mechanisms behind the correlation between CO₂ instability and algae outbreaks — see the guide to what stable CO₂ actually means and the explanation of how CO₂ fluctuation drives BBA.
Diffuser and drop checker placement affect the bulk measurement. A drop checker positioned near the CO₂ diffuser will read higher than one placed in a low-flow area on the opposite side of the tank. The correct placement is in an area representative of average tank conditions — not adjacent to the diffuser output — so that the reading reflects what plants in most of the tank actually experience in the bulk water. The boundary layer then introduces a further reduction from that measured value at each leaf surface.
Designing around boundary layer constraints
Understanding the boundary layer allows you to make more informed decisions about plant selection and tank layout, rather than simply increasing CO₂ concentration in response to plant underperformance.
Low-flow zones should contain tolerant species. Areas of the tank where flow is inherently limited — behind a large piece of driftwood, in the back corners, directly under a spray bar shadow — will have thicker boundary layers regardless of overall tank circulation. Placing CO₂-demanding plants in these zones sets them up to underperform. Tolerant species and bicarbonate users are better suited to these positions.
Carpeting plants face the greatest challenge. Plants growing horizontally close to the substrate — Hemianthus callitrichoides, Glossostigma, Marsilea — are in the area of lowest flow velocity in most tanks, immediately above the substrate boundary. They combine high CO₂ demand with the thickest boundary layer conditions. Addressing this requires not more CO₂ injection but better low-level circulation — positioning the inlet to encourage flow across the substrate, and ensuring the outlet gyre reaches the substrate level rather than remaining as a surface current.
Fine-leaved plants have an inherent advantage. The boundary layer develops around an individual leaf surface. Fine-leaved plants — mosses, Myriophyllum, Cabomba, Rotala with small leaves — have a higher surface area to volume ratio and create less of a CO₂ sink at any single point. Broad-leaved plants with large laminae deplete CO₂ across a wider surface, creating a more pronounced gradient. This is one reason mosses and fine-leaved plants often outperform broad-leaved plants in low-CO₂ conditions even when both species are listed as "easy."
CO₂ boundary layer: key points
The physics: CO₂ diffuses ~10,000× more slowly in water than air. During active photosynthesis, plants deplete CO₂ at the leaf surface faster than diffusion can resupply it, creating a thin layer of carbon-depleted water (the diffusive boundary layer) regardless of bulk water concentration.
What measurements show: Leaf surface CO₂ during high-light photosynthesis can be 80–95% lower than bulk water CO₂. A bulk reading of 30 ppm may translate to 2–6 ppm at the leaf.
Flow thins the layer: Faster water movement across the leaf surface = thinner boundary layer = better CO₂ supply. Circulation quality is as important as injection rate for demanding plants.
Species strategy: Bicarbonate-using species ("easy" plants) can supplement CO₂ from HCO₃-. CO₂-obligate species ("demanding" plants) cannot and are fully dependent on boundary layer delivery conditions.
Design implications: Dead spots, carpeting plants near substrate, and large-leaved plants in low-flow positions are the highest-risk scenarios. Address them with flow geometry, not more CO₂.
References
- Cussler, E.L. (2009). Diffusion: Mass Transfer in Fluid Systems, 3rd ed. Cambridge University Press. Table of diffusion coefficients in water and air at 25°C.
- Larkum, A.W.D., Roberts, G., Kuo, J. & Strother, S. (2018). "CO₂ and O₂ dynamics in leaves of aquatic plants with C3 or CAM photosynthesis." PMC / New Phytologist. pmc.ncbi.nlm.nih.gov/articles/PMC6153474
- Pedersen, O., Colmer, T.D. & Sand-Jensen, K. (2013). "Underwater photosynthesis of submerged plants — recent advances and methods." Frontiers in Plant Science, 4, 140. doi:10.3389/fpls.2013.00140
- Shao, S. et al. (2020). "Responses of Ottelia cordata to variable CO₂ concentrations: implications for inorganic carbon use strategies." Frontiers in Plant Science, 11, 1261. pmc.ncbi.nlm.nih.gov/articles/PMC7457065
- Liu, F. et al. (2023). "Inorganic carbon use strategies of the submerged macrophyte Ottelia ovalifolia." Frontiers in Plant Science, 14, 1142848. doi:10.3389/fpls.2023.1142848
More articles
View all →
