Domestic Use Gas Cylinders from India's Coal Power Plants

21 Apr, 2026

Oil and gas have been diplomatic weapons for more than a century now and will continue to be so. Things being the way they are in the Persian Gulf, the blow to India has been harsher than to many others. I was recently reading an article in the Hindu which put this into perspective. India's problem with the unavailability of natural gas is not the same as nations like Japan where gas power plants are the major source of energy for the country. More than our power plants, the imported hydrocarbon runs India's households, from villages to cities the LPG cylinder is the lifeline of India's kitchens. Electric stoves work, yes. But changing the habits of more than a billion people is going to take many decades.

Let's dive deep into what LPG actually is. Liquefied Petroleum Gas is a mix of $C_3$ - $C_4$ hydrocarbons. The actual concentrations vary, but it is roughly half propane and roughly half butane, depending on the season. The two facts that matter most here are the boiling point (which determines how the fuel behaves in a cylinder) and the heating value (which determines how much energy the cylinder delivers).

India has made strides in energy self-sufficiency in past decades and we seem to be on track to move our energy infrastructure to greener sources of energy. We do however have a significant number of coal power plants across the country which make a lion's share of our energy infrastructure and a not-insignificant share of our CO₂ emission problems.

But what if the problem could be our solution? As a card-carrying NileRed supporter, I like to believe that synthesis of these hydrocarbons from carbon-dioxide and water using energy is a chemistry problem. The Fischer–Tropsch process (I can't pronounce it either) is one well-known way to turn CO into hydrocarbons, and more recent variants handle CO₂ directly. The chain I'm focusing on is the methanol-mediated cousin of this approach. This idea is nowhere close to being original. Turns out that the NTPC has already worked on a major part of this whole problem at the Vindhyachal Power plant to synthesize methanol from CO₂ captured from coal flue gas. More about that later.

My intention to write this is because this chain of thought is what FINALLY made me care about the carbon economy. I may not understand carbon credits, but I do understand (some) chemistry.

And before I get ahead of myself, I should map out the reactions that actually go into the process and see why coal power plants can be a perfect fit for India's hydrocarbon gap.

Stage 1: Water electrolysis (the hydrogen source)

$2{\text{H}}_2\text{O}\stackrel{\text{electricity}}{\to }2{\text{H}}_2+{\text{O}}_2\Delta H=+572\text{ kJ/mol}$

This is the step that turns electricity into storable chemistry. Strongly endothermic: every joule of chemical energy in the hydrogen comes from the electricity driving the reaction. Modern alkaline or PEM electrolyzers achieve 60 to 75% LHV (lower heating value) efficiency. High-temperature solid-oxide electrolyzers (SOEC) push this higher by using process heat for part of the energy input.

Remark: Sending the oxygen back to a coal boiler for partial oxy-fuel combustion allows the CO₂ concentration in the flue gas to jump from roughly 14% to roughly 90%. The electrolyzer's "waste" becomes the capture step's critical feedstock. So we have two independent processes whose byproducts happen to be exactly what the other needs. There are some free lunches. :)

Stage 2: CO₂ absorption into an amine solvent (a recirculating loop)

The amine basically cycles continuously between two reactors running the same reaction in opposite directions. In the absorber column at 40 to 60°C, the amine grabs CO₂ out of the flue gas (the mixture of gases that comes out of a chimney or "flue" after something has been burned). For a tertiary amine, the dominant chemistry is bicarbonate formation:

${\text{CO}}_2+{\text{R}}_3\text{N}+{\text{H}}_2\text{O}\rightleftharpoons {\text{R}}_3{\text{NH}}^{+}+{\text{HCO}}_3^{-}$

In the stripper column (no, not that stripper column) at 120°C, heated by low-pressure steam, the same reaction runs backward, releasing pure CO₂:

${\text{R}}_3{\text{NH}}^{+}+{\text{HCO}}_3^{-}\stackrel{\Delta }{\to }{\text{CO}}_2+{\text{R}}_3\text{N}+{\text{H}}_2\text{O}$

The regenerated "lean" amine is pumped back to the absorber to grab more CO₂. A single amine molecule runs this cycle thousands of times before it degrades and needs replacement. Industrial makeup rates are roughly 1.5 to 2.5 kg of amine per tonne of CO₂ captured, meaning the amine functions more like a catalyst than a reagent.

Remark: flue gas enters at 12 to 15% CO₂, exits at around 2%, and a stream of 99%+ pure CO₂ comes off the stripper overhead, ready for downstream synthesis. Conventional monoethanolamine (MEA) requires about 3.5 GJ per tonne of CO₂ for regeneration. Modern tertiary-amine blends drop this to around 2.3 GJ per tonne. That 35% energy saving is what makes industrial-scale capture from coal flue gas economic at all.

Stage 3: Reverse water-gas shift (or RWGS, an optional bridge)

${\text{CO}}_2+{\text{H}}_2\rightleftharpoons \text{CO}+{\text{H}}_2\text{O}\Delta H=+41\text{ kJ/mol}$

Mildly endothermic, equilibrium-limited, run at 600 to 800°C over a Ni-based or Fe-K catalyst. The reaction is thermodynamically "uphill" from CO₂ to CO: you need heat to drive it. In return, you get a more reactive carbon species for downstream synthesis.

Remark: Some methanol synthesis catalysts work better with a mixed CO/CO₂ feed than with pure CO₂. Direct CO₂ hydrogenation is also fine (see Stage 4), so this stage is often skipped at plants that choose the direct route for simplicity. But if RWGS is used, the CO₂ is converted into CO, which is the carbon feedstock classical Fischer-Tropsch reactors are designed for. Its heat demand can be met by extracting steam from the high-pressure turbine rather than burning fresh fuel.

Stage 4: Methanol synthesis (the key upstream reaction)

Two parallel reactions happen over a Cu/ZnO/Al₂O₃ catalyst at 50 to 100 bar and 220 to 270°C:

$\text{CO}+2{\text{H}}_2\to {\text{CH}}_3\text{OH}\Delta H=-91\text{ kJ/mol}$

${\text{CO}}_2+3{\text{H}}_2\to {\text{CH}}_3\text{OH}+{\text{H}}_2\text{O}\Delta H=-49\text{ kJ/mol}$

Both are exothermic, meaning the reactor gives off heat rather than consuming it. This is the most commercially mature step in the entire chain: methanol synthesis has been running at million-tonne scale globally since the 1960s, and the Cu/ZnO/Al₂O₃ catalyst has had more than half a century of refinement.

The reaction is equilibrium-limited at reasonable conditions (per-pass conversion is only 15–25%), which is why real methanol plants use a recycle loop: unreacted CO, CO₂, and H₂ are separated from the crude methanol and fed back into the reactor.

Remark: Per tonne of methanol produced, the reaction needs 1.375 tonnes of CO₂ and 0.1875 tonnes of H₂. That is a 3 moles H₂ to 1 mole CO₂ ratio, which means hydrogen is nearly always the binding constraint in any plant design. You run out of H₂ before you run out of CO₂.

Stage 5: Methanol dehydration to DME (dimethyl ether)

$2{\text{CH}}_3\text{OH}\to {\text{CH}}_3{\text{OCH}}_3+{\text{H}}_2\text{O}\Delta H=-23\text{ kJ/mol}$

Mildly exothermic catalytic dehydration over γ-Al₂O₃ or a zeolite catalyst. Operating window: 250 to 350°C, 10 to 20 bar. Per-pass methanol conversion reaches about 80% with DME selectivity above 99%, one of the cleanest, highest-selectivity reactions in all of industrial catalysis.

The mechanism is straightforward acid-catalyzed dehydration. Two methanol molecules each donate a hydroxyl-hydrogen pair that combines to form water, while the two methyl groups link through an oxygen to form the ether. The catalyst surface provides the acid sites that protonate the methanol and lower the activation barrier.

Remark: DME's physical properties mimic LPG almost perfectly. Boiling point of −24°C (propane is −42°C, butane is −1°C, so DME sits squarely between them). Liquefies at about 5 bar at ambient temperature. Heating value of 28.8 MJ/kg versus LPG's 46 MJ/kg, which is why blending at 20% is the sweet spot rather than attempting 100% substitution. Clean combustion: no soot, very low NOx, very low SOx.

Stage 6: Combustion in the kitchen (the final reaction)

A good bhatura is a round bhatura. Doesn't happen properly on an electric stove.

${\text{CH}}_3{\text{OCH}}_3+3{\text{O}}_2\to 2{\text{CO}}_2+3{\text{H}}_2\text{O}\Delta H=-1460\text{ kJ/mol}$

Strongly exothermic: this is where the stored chemical energy becomes the flame under the bhatura. CSIR-NCL has developed a flex-mode burner (the "Aditi Urja Sanch" unit, originally launched in 2020 and since refined) that handles anywhere from 100% LPG to 100% DME and every blend in between. Efficiency trials show 10–15% improvement over conventional LPG burners, thanks to DME's cleaner combustion kinetics.

The full-cycle carbon accounting: Every molecule of CO₂ that went into Stage 2 eventually comes out as CO₂ at Stage 6. The chemistry does not sequester carbon but relocates it. What the chain does do is displace imported LPG molecules, substituting indigenous synthesis for Gulf imports and reducing forex outflow by thousands of crores per year depending on the amount of DME blending. Not exactly self-sufficiency but a few steps closer to being self-reliant.

Coal plants emit far more CO₂ than their paired electrolyzers can ever process. The gating cost is the green H₂, which means the gating cost is renewable electricity and electrolyzer capex. This reframes the whole policy conversation. Scaling indigenous LPG production isn't really about scaling coal plants or scaling capture. It's about using existing infrastructure and scaling electrolyzers.

Remark: How does DME fit into the domestic supply chain? To deliver the same cooking energy, you need more DME by mass than LPG. Specifically, 1.60 kg of DME per kg of LPG. A 14.2 kg LPG cylinder would need to be a 22.7 kg cylinder of pure DME to deliver the same energy. By volume, you need about 1.3 litres of liquid DME per litre of liquid LPG to deliver equal energy. This is why blending is the sweet spot. At 20% DME by mass in an LPG-DME blend, the cylinder's total energy content drops by only about 7%, which is within the variation households already experience between batches of LPG whereas above 30% DME the energy drop starts becoming perceptible.

The best part? The entire process up to Stage 4 is already something NTPC has demoed at the Vindhyachal power plant. Stage 5 is one of the downstream options to consume the methanol into a home-friendly cooking-grade end-product.

So What's Happening at NTPC Vindhyachal?

NTPC's Vindhyachal Super Thermal Power Station is a 4.7 GW pithead coal plant in Singrauli district, Madhya Pradesh. Since October 2025, its Unit-13 (500 MW, commissioned 2015) has hosted a three-stage CCU (carbon capture and utilization) pilot executed by NTPC's R&D arm, NETRA.

The pilot runs four of the six reactions in the chain above, continuously:

In October 2025, the first drop of methanol came out. The chain works.

What Vindhyachal doesn't yet have (per the limits of my internet search capabilities) is any downstream conversion of that methanol into a cooking-fuel product. The methanol currently produced is chemical-grade: saleable, valuable, but not reaching an Indian kitchen. Meanwhile, CSIR-NCL in Pune has spent a size-able chunk of the last decade perfecting exactly the Stage 5 reactor that would close the chain: a patent-protected methanol-to-DME dehydration process, currently operational at 250 kg/day with a 2.5 TPD demonstration plant coming online within months.

The pieces fit together. Vindhyachal produces methanol. CSIR-NCL has the methanol-to-DME reactor. BIS IS 18698:2024 already permits up to 20% DME blending with LPG (with 8% requiring no infrastructure modification). IOC, BPCL, and HPCL already have the bottling and distribution networks.

I'm pretty sure chemical engineers and brains way smarter than mine have been working on this already. While India's energy infrastructure shifts towards greener and cleaner sources, our homes are still a couple decades away from phasing out the gas cylinder from our kitchens. And India's existing coal power plant infrastructure can be retro-fitted for cooking-gas generation as this shift happens. And maybe, just maybe, Indian kitchens wouldn't be hit as hard the next time world leaders bully each other over key geographical locations.

Sources that I have not identified above:

1. NTPC website. They have a lot of material around how they are working on a 'Circular Economy' model for energy synthesis. [Link]
2. India's LPG Crisis Has An Indigenous Fix by Swarajya (March 2026) [Link]
3. Excerpts of an EY report on Coal Gasification for Energy and Chemical Security that I did not want to pay to access (April 2026)[Link]
4. NITI Aayog's India's Leapfrog to Methanol Economy [Link]
5. Dimensional Energy's Website, I'm bullish about what they're doing now after going over this whole problem in detail. [Link]