Closed-Loop vs Open-Loop Geothermal Systems
Updated 21 September 20267 min readGeothermal & Heat Pumps
The distinction is simple: a closed loop circulates its own sealed fluid through buried pipe and never touches groundwater, while an open loop pumps groundwater through a heat exchanger and returns it to the ground. Open loops transfer heat better because nothing stands between the water and the exchanger. That same directness is what brings their complications.
Key takeaways
- Closed loops are sealed circuits of plastic pipe; open loops move actual groundwater through the building's heat exchanger.
- Heat has to cross pipe wall, grout and soil to reach a closed loop. An open loop removes those layers, which is its whole advantage.
- An open loop's viability is a question about the aquifer — yield, chemistry and somewhere to put the water afterwards — not about the heat pump.
- Water chemistry is the recurring open-loop failure mode: scale, iron and particulates foul exchangers that a closed loop never exposes.
- Both must respect the ground's heat balance. Returning water too close to where it was drawn creates thermal breakthrough, the open-loop version of an undersized closed loop.
On this page
What each one circulates
A closed loop is a sealed circuit. Water with antifreeze circulates endlessly through buried plastic pipe, collecting heat through the pipe wall and carrying it to the heat pump. The same fluid returns to the ground and goes round again. Nothing enters and nothing leaves.
An open loop pumps groundwater from a well, passes it through a heat exchanger where the heat pump takes what it needs, and then returns the cooled water to the ground through a second well or another approved discharge. The water makes a single pass.
The heat pump itself does not care. Both arrangements deliver a fluid at some temperature to the evaporator, and the vapour-compression cycle proceeds identically. What changes is how efficiently heat gets to that fluid, and what else comes with it.
The heat-transfer advantage, and where it comes from
Follow a joule of heat from the soil into a closed loop and count the barriers it crosses: through the soil to the borehole wall, through the grout that fills the borehole, through the wall of the plastic pipe, and finally into the fluid. Each layer has a thermal resistance, and they add up. Plastic pipe is a poor conductor — a necessary compromise, since it also has to be durable and cheap to bury.
An open loop deletes that stack. The water *is* the heat carrier, arriving at the exchanger already at the aquifer's temperature.
The effect shows up as a better source temperature at the evaporator, which is a smaller temperature lift, which is directly a better coefficient of performance. In a cold climate that margin is worth having — the divergence described in ground-source versus air-source in extreme cold is precisely about how valuable a stable, warm-ish source is on the coldest day.
What the water brings with it
Groundwater is not clean water. It is a solution that has been in contact with rock for a long time, and everything it dissolved comes with it into the building.
- Hardness and scale. Dissolved calcium and magnesium precipitate on heat exchanger surfaces as conditions change, and scale is an insulator, so a fouled exchanger quietly gives back the efficiency the open loop was chosen for.
- Iron and manganese. Common in groundwater and prone to oxidising into deposits once exposed to any oxygen, often with the help of iron bacteria, which produce a sludge that blocks fine passages.
- Particulates. Sand and silt abrade pump impellers and exchanger plates and settle where flow slows.
- Corrosion. Chloride content, dissolved gases and pH decide which exchanger materials survive. This is a materials selection problem, not a maintenance one.
The standard defence is to keep the groundwater out of anything delicate: a robust plate exchanger between the well water and a clean secondary circuit, filtration sized for the particulates present, and materials chosen from a water analysis taken before anything is ordered. A closed loop needs none of this, because its fluid is whatever was put in at commissioning.
Where the water goes afterwards
An open loop moves substantial volumes of water. Where it goes is a bigger question than where it came from.
Reinjection into the same aquifer through a second well is the usual answer: the water is returned, a few degrees colder in winter, having given up its heat. That requires the aquifer to accept it as readily as it gave it up, which is not guaranteed — injection wells foul more readily than supply wells.
Separation matters, and so does direction. Put the injection well upstream of the supply well and the cooled plume drifts straight back, so the system draws progressively colder water and its performance declines season after season. Diagnosing that is harder than it sounds, because every component tests as healthy.
Abstracting and returning groundwater is a regulated activity in most places, and the rules differ by jurisdiction. The requirements come from the authority responsible for the aquifer, and they are best established before the design is fixed rather than after.
Closed-loop variants, and the hybrid in between
Closed loops trade thermal performance for independence from water, and their variants trade land for depth.
| Closed loop | Open loop | Standing column well | |
|---|---|---|---|
| What circulates | Sealed antifreeze solution | Groundwater, once through | Groundwater within one deep well |
| Heat transfer | Through pipe wall and grout | Direct | Direct, through the rock column |
| Site requirement | Land area or drilling access | A productive aquifer and somewhere to return water | Suitable hard-rock geology and depth |
| Water chemistry | Irrelevant | Decisive | Decisive |
| Maintenance | Circulation pump and antifreeze | Wells, filtration, exchanger cleaning | Wells and exchanger, with bleed control |
| Slow failure mode | Ground cools year on year if undersized | Thermal breakthrough, or fouling | Column cools if bleed is insufficient |
| Regulatory burden | Lighter, rarely nil | Abstraction and discharge approval | As open loop |
Comparison of design characteristics. What decides a specific site is a well yield test and a water analysis for open loops, or a ground thermal assessment for closed loops — neither of which is a property of the heat pump.
The standing column well sits between the two families. Water is drawn from the bottom of a deep borehole and returned to the top of the same one, so it circulates through the rock rather than across the site. It keeps much of the open loop's direct heat transfer without needing two wells or a large water footprint, and in hot or cold extremes a small "bleed" — discharging a fraction of the flow and drawing fresh groundwater to replace it — pulls the column temperature back toward the aquifer's.
Choosing: the site decides
The honest summary is that this is not a choice between two products. It is a question about the ground.
Ask about the water first. Is there an aquifer that will sustain the required flow through a whole heating season? What is in the water? Is there somewhere lawful and practical to return it? If any answer is unsatisfactory, the comparison is over and a closed loop is the design.
If the water answers well, an open loop offers a better source temperature for less drilling, at the price of water handling, filtration and a maintenance regime that must actually be followed.
If in doubt, closed. A sealed loop has fewer ways to disappoint slowly, and slow disappointment is the characteristic failure of ground systems — not sudden breakdown but a quiet decline that every component passes a test for. The same is true of the air-side diagnostics in why a heat pump freezes up: the machine is rarely the thing that is wrong.
Frequently asked questions
Which is more efficient?
An open loop usually delivers a better source temperature to the heat pump, because heat reaches the exchanger without crossing pipe wall, grout and soil. That narrower temperature lift shows up directly as a better coefficient of performance. The advantage only exists where a suitable aquifer does.
Can an open loop run out of water?
Yes, and it is a real design risk. The supply well must sustain the required flow continuously through the heating season, not just on the day it was tested. Seasonal drawdown, nearby abstraction and drought all affect it, which is why the well test matters more than the equipment selection.
What does 'thermal breakthrough' mean?
It is when the water returned to the ground finds its way back to the supply well. In winter that means drawing progressively colder water, and performance drifts down. Separation and an understanding of which way the groundwater moves are what prevent it.
Do open-loop systems need approval?
Almost everywhere, yes. Abstracting groundwater and returning it are regulated activities in most jurisdictions, and the requirements differ by place, so it is a question for the local authority rather than for a heat pump manufacturer. Closed loops face lighter requirements but are rarely unregulated either.
Is a standing column well closed or open?
Open, with a twist: water is drawn from the bottom of a deep borehole and returned to the top of the same one, so it circulates through the rock column. It combines an open loop's direct heat transfer with a much smaller water footprint, and it depends on suitable hard-rock geology.
Sources
Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.
- U.S. Department of Energy, Geothermal Technologies OfficeReference material on ground-source heat pump loop configurations, including open-loop and standing column designs.
- International Renewable Energy Agency (IRENA)Background on shallow geothermal resources.
- Local groundwater authorityAbstraction and discharge requirements for open-loop systems are jurisdiction-specific and must come from the authority with responsibility for the aquifer.
Editorial Team
Research, drafting and review
Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.
Last reviewed 21 September 2026. How we research and review