# Solar thermal energy **Solar thermal energy** is sunlight collected as [[Heat|heat]] rather than as electricity. A dark absorber under glass, a trough of mirrors, or a field of [[Solar_power_tower|heliostats]] raises a [[Water|working fluid]]'s [[Temperature|temperature]], and that heat is used directly — for washing water, space heating, drying, cooking, [[Distillation|distillation]] or industrial process heat — or fed to a [[Heat_engine|heat engine]] to make power. The whole subject turns on one sizing equation, `A = Q_day/(I·η)`: the collector area needed is the daily demand in Btu divided by the incident flux in Btu per square foot per day times the collection efficiency.[^kerlin-177] In the microsim below the reader has a single control, collection efficiency η from 0.2 to 0.8, and watches three bars move together: the area sized on summer sun (60 ft²), the area sized on winter sun (89 ft²), and the two-axis-tracking equivalents. Beside them sits a gauge that never moves at all — the winter-cover ratio, 1,050/1,550 = 68 %. That is the point of the sim. Efficiency rescales every bar in proportion and cancels out of the ratio, so a collector sized on summer sunshine covers barely two-thirds of a winter day's hot water no matter how good the collector is. Seasonality, not efficiency, is why Thomas Kerlin writes that a backup heater is "usually provided".[^kerlin-177][^spec-e40] A storage tank sits beside the bars, sized by `V = Q/(c_p·ρ·ΔT)` — 390 gallons at a 40 °F swing.[^kerlin-181] On the [[Energy]] flagship this article serves *Solar thermal* in Part V — Transformation, the section that separates collecting sunlight as heat from collecting it as charge, and it is the thermal counterpart of the [[Solar_cell|solar cell]] row beside it. ## History Using the sun as a heat source is older than using it as an electrical one by millennia — glass-covered hotboxes, burning mirrors and black-painted water tanks all predate any theory of what was happening. What is modern is the accounting. Once a household's energy use could be written down, solar heat stopped being a curiosity and became an engineering trade-off: a typical American home uses about 100 million Btu a year, of which water heating is about 17 %, or 17 million Btu — 46,600 Btu a day — and space heating about 50 million Btu.[^kerlin-176][^kerlin-179] Those three numbers, rather than any invention, define what a collector is for. The United States Department of Energy's rule of thumb dates from the same accounting impulse: 20 ft² of collector for each of the first two residents, plus 8 ft² per additional resident in the Sun Belt or 12 to 14 ft² in the northern states.[^kerlin-176] ## Low-temperature heating and cooling Low-temperature work — domestic hot [[Water|water]], pool heating, space heating — is where nearly all installed solar thermal capacity actually is, because it needs no concentration, no tracking and no heat engine. It is also where the sizing equation is most honest, because demand is roughly constant through the year while supply is not. ### Low-temperature collectors Take Kerlin's Knoxville example. A fixed collector tilted at latitude plus 15°, or 50.82° — which places Knoxville at 35.82° north — receives about 1,050 Btu/(ft²·day) in winter, 1,550 in summer and 1,425 as an annual mean; two-axis tracking raises those to 1,200, 2,500 and 1,900.[^kerlin-177][^derived-st] Against a demand of 46,600 Btu/day and η = 0.5, the summer-sized area is A = 46,600/(1,550 × 0.5) = 60 ft².[^kerlin-177] Check it against the annual mean and the collector delivers 1,425 × 0.5 × 60 = 42,750 Btu/day, about 92 % of demand — which sounds like a solved problem.[^kerlin-177][^derived-st] It is not, because the shortfall is not spread evenly. In January the same 60 ft² delivers 1,050 × 0.5 × 60 = 31,500 Btu/day, 68 % of what the house wants, in the month it wants it most.[^derived-st] Sizing on winter instead demands 46,600/(1,050 × 0.5) = 89 ft², which then spills a third of its summer output.[^derived-st][^manual10] Turning η up to 0.8 shrinks both bars by the same factor and leaves the 68 % gauge exactly where it was.[^spec-e40] For comparison, the DOE rule gives 48 ft² for a three-person Sun Belt household and 52 to 54 ft² in the north — the book calls it "simple (and very approximate)" — while an online calculator returns 64.6 ft² for the same house.[^kerlin-176][^kerlin-177] All three answers are within a factor of about 1.5 of one another, which is the real precision of collector sizing. ## Heat storage for space heating Space heating is the harder load, because it is seasonal in exactly the wrong phase. Kerlin's second example takes 50 million Btu spread over about 200 heating days — 250,000 Btu/day — and finds a fixed-tilt area of 325 to 475 ft², or 225 to 415 ft² with two-axis tracking, the spread coming from the winter-to-summer flux range rather than from any uncertainty.[^kerlin-180][^derived-st] The demand itself is measured in heating degree-days, `HDD = Σ(65 °F − T_avg)` over the days below 65 °F, and Table 6-1 shows how far that varies: Miami 149, Houston 1,525, Atlanta 2,827, Knoxville 3,531, Hartford 6,104, Denver 6,128, Laramie 9,038 — a factor of 3.20 between Laramie and Atlanta alone, and sixty between Laramie and Miami.[^kerlin-179][^derived-st] Storage closes the daily gap, not the seasonal one. Sensible storage is `Q = m·c_p·ΔT`, and for water Kerlin uses c_p = 1 Btu/(lb·°F) and 8 lb/gal, so a tank cycling between 120 °F and 80 °F holds 40 Btu/lb, or 320 Btu/gal. A 125,000 Btu day therefore needs 390 gallons, about 50 ft³ — a four-foot cube.[^kerlin-181][^derived-st] The book is careful that this uses average rather than peak demand and is an underestimate.[^kerlin-181] Seasonal storage would need a tank two hundred times larger, which is why the 68 % gauge is answered with a backup burner and not with a bigger tank. ### Process heat Industrial process heat below about 100 °C — washing, pasteurizing, preheating boiler feedwater — is the same physics with a flatter demand curve, and it is the load that suits solar heat best, because a factory that runs all year has no seasonal mismatch to store around. ## Medium-temperature collectors Between about 100 °C and 250 °C, flat plates give way to evacuated tubes and to modest concentration, because at those temperatures a flat absorber radiates away much of what it takes in. The applications are the oldest ones in the subject, and each is a version of the same equation with a different Q_day. ### Solar drying Drying is the lowest-grade use of heat there is: air is warmed a few degrees, its capacity to hold moisture rises steeply as [[Evaporation|evaporation]] accelerates, and it is drawn over the crop. No storage, no working fluid, no pressure vessel. ### Cooking A box cooker is a collector with the load inside it. The reachable temperature is set by the balance between absorbed flux and the [[Thermal_insulation|insulation]] and glazing losses, which is why doubling the aperture with a reflector matters more than any change of paint. ### Distillation A solar still evaporates [[Water|water]] at one surface and condenses it at another, a [[Phase_transition|phase transition]] in each direction, and its throughput is governed by [[Enthalpy_of_vaporization|latent heat]]: vaporizing a kilogram of water takes about 2.2×10⁶ J, roughly 0.4 eV per molecule.[^likharev-107] That is 946 Btu/lb, against the 40 Btu/lb a storage tank moves over a 40 °F swing — a factor of 24, and the reason distillation needs far more collector per litre than heating does.[^derived-st] ## High-temperature collectors Above about 250 °C, only concentration will do, and concentration has a hard physical condition: only the direct beam can be focused. Diffuse light arriving from the whole sky cannot be brought to a point by any optic, so a concentrating system is limited at the outset by the unscattered fraction of the [[Solar_irradiance|sunlight]] reaching the ground.[^kerlin-185] ### System designs Three geometries dominate. Parabolic troughs concentrate onto a line and reach a few hundred degrees, which is enough for an [[Steam_turbine|oil-or-steam]] loop. [[Solar_power_tower|Power towers]] use a field of tracking heliostats aimed at a receiver on a mast, reaching temperatures high enough to raise the turbine-inlet conditions of a proper [[Rankine_cycle|Rankine cycle]]. Dishes concentrate onto a point and reach the highest temperatures of all, which is what makes a [[Solar_furnace|solar furnace]] a materials instrument as much as a power device. All three must track the sun, because a fixed concentrator would lose its focus within minutes. ## Heat collection and exchange Between absorber and load sits the plumbing, and it decides how much of the collected heat survives. A collector loop must move [[Heat_transfer|heat]] from a surface only tens of degrees above ambient, so every [[Heat_exchanger|heat exchanger]] approach temperature is expensive: a 10 °F pinch at the tank is a 10 °F rise the collector must make for nothing. Freeze protection forces a choice — drain-back, which empties the collector when the pump stops, or an antifreeze loop, which costs an extra exchanger and a few percent of η. Selective absorber coatings, which take in visible [[Sunlight|light]] and have low [[Emissivity|emissivity]] in the infrared, buy back some of the [[Thermal_radiation|radiative]] loss that limits medium-temperature work, and the evacuated tube removes the convective loss entirely by removing the air. ## Heat storage for electric base loads A solar thermal power plant has an advantage no photovoltaic plant has: it can store its energy before converting it, as heat, which is far cheaper per kilowatt-hour than [[Energy_storage|storing it afterwards]] as charge. The storage sits between collector and turbine, so the plant rides through cloud and into the evening at full output, raising its [[Capacity_factor|capacity factor]]; the [[Turbine|turbine]] hall — the most capital-intensive part — runs more hours per year. ### Steam accumulator The simplest store is a pressure vessel of saturated water. Charging raises its pressure and temperature; discharging drops the pressure and the water flashes to steam. Its capacity is the sensible heat of the water mass between the two pressures, which is modest, and its virtue is that the response is instant. It buffers minutes, not hours. ### Molten salt storage Nitrate salt mixtures stay liquid from roughly 250 °C to above 550 °C, and a two-tank system moves salt from cold to hot and back. Because the store is sensible, its capacity is `Q = m·c_p·ΔT` again, with c_p the salt's [[Specific_heat_capacity|specific heat capacity]], and the usable ΔT of about 300 °C — an order of magnitude more than a domestic tank's 40 °F — is what makes hours of storage physically compact.[^kerlin-181][^derived-st] The cost is that the salt freezes if it is allowed to cool, so the whole loop is heat-traced. ### Phase-change materials for storage A phase-change store exploits the same [[Latent_heat|latent heat]] that makes distillation expensive. Because the enthalpy of a phase change is many times the sensible heat available over a working swing — 946 Btu/lb against 40 for water — a latent store is far more compact, and it delivers at one temperature rather than over a sliding range.[^likharev-107][^derived-st] Its difficulties are heat transfer through a solidifying layer and materials that survive thousands of cycles. ## Use of water The heat a [[Concentrated_solar_power|concentrating]] plant does not convert must be rejected, and the condenser is where the water goes. Kerlin puts a thermal plant's rejection at 2.5 to 3 times its electrical output, so a 100 MW solar plant dumps 250 to 300 MW of low-grade heat.[^kerlin-185] Wet cooling does that cheaply and consumes [[Water|water]] by evaporation; dry cooling consumes none but raises the condenser temperature, which lowers the cycle efficiency and therefore raises the collector area for the same output. Since the best solar resource and the least available water are usually found in the same desert, the choice is rarely free. Low-temperature collectors dodge the problem entirely, because their heat is the product rather than the waste. ## Electrical conversion efficiency The chain from sunlight to electricity through heat has two links, and Kerlin multiplies them explicitly: `η_net = f_unscattered × η_thermo ≈ 0.51 × 0.30 ≈ 0.15`.[^kerlin-185] The first factor is optical — roughly half the sunlight arriving at the ground is direct beam a concentrator can use. The second is [[Thermodynamics|thermodynamic]], the [[Carnot_cycle|Carnot]]-limited efficiency of the heat engine at the temperature the receiver can reach. Neither factor can be pushed far. Raising receiver temperature improves η_thermo but increases [[Thermal_radiation|radiative]] loss as T⁴ and demands better alloys; nothing at all improves f_unscattered. That 15 % is what the land arithmetic is built on. One quad per year needs about 330,000 acres of collector at 15 % efficiency, and about 165,000 acres at 30 %, against Arizona's 73 million acres.[^kerlin-188] The scaling is exactly 1/η, so the honest comparison with [[Photovoltaic_system|photovoltaics]] is not which technology is cleverer but which delivers more converted energy per acre per dollar — and for solar thermal, the answer includes the storage that comes with the heat. ## Standards The single number this page has leaned on throughout — η = 0.5 — is a nominal value, and every quoted collector efficiency is meaningless without the conditions it was measured under: the flux, the ambient temperature, the inlet fluid temperature and the wind. A collector's efficiency falls as the difference between absorber and ambient grows, so the same panel is 70 % efficient warming a swimming pool and 30 % efficient making 60 °C water in January. Test standards exist to fix those conditions and to publish the resulting efficiency curve rather than a single figure; Kerlin's own warning that the DOE sizing rule is "simple (and very approximate)" is the same caution one level up.[^kerlin-176][^kerlin-177] The specific certification schemes and their test procedures are catalogued by the pair. ## See also - [[Concentrated_solar_power]] — the heat-engine chain, η_net ≈ 0.51 × 0.30 ≈ 0.15 - [[Solar_thermal_collector]] — the collector itself - [[Thermal_energy_storage]] — sensible `m·c_p·ΔT` against latent `m·L` - [[Solar_furnace]] - [[Solar_power_tower]] - [[Passive_solar_building_design]] - [[Solar_cell]] — the photovoltaic alternative on the same spine - [[Rankine_cycle]] — the cycle a high-temperature collector feeds ## References [^kerlin-176]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3 (pp. 169–191), p. 176: the US Department of Energy collector rule of thumb — 20 ft² for each of the first two residents plus 8 ft² (Sun Belt) or 12–14 ft² (northern US) per additional resident, described as "simple (and very approximate)"; a US home uses about 100 million Btu/yr, of which water heating is about 17 %. https://open.umn.edu/opentextbooks/textbooks/future-energy-opportunities-challenges [^kerlin-177]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, pp. 177–178 and 180, Example 6.1 (Knoxville water heating): the sizing equation `A = Q_day/(I·eta)`; a fixed collector tilted at latitude + 15° = 50.82°; fixed-tilt flux 1,050 (winter), 1,550 (summer) and 1,425 (annual mean) Btu/(ft²·day), two-axis 1,200, 2,500 and 1,900; A ≈ 60 ft² supplying about 92 % of annual demand; winter flux 67 % of summer; an online calculator giving 64.6 ft²; and the note that a backup is "usually provided". [^kerlin-179]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, pp. 178–179: heating degree-days `HDD = sum(65 °F − T_avg)` over days below 65 °F, and Table 6-1 (mean °F, HDD, months above 65 °F) — Hartford 50.2/6,104/3; Laramie 40.4/9,038/0; Denver 50.1/6,128/3; Knoxville 59.5/3,531/5; Atlanta 62.1/2,827/5; Houston 68.8/1,525/7; Miami 76.7/149/12; space heating about 50 million Btu/yr. [^kerlin-180]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, pp. 179–180, Example 6.2 (space heating): 50 million Btu over about 200 days = 250,000 Btu/day; fixed tilt needs 325–475 ft² and two-axis tracking 225–415 ft², the spread coming from the flux range rather than from uncertainty. [^kerlin-181]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, p. 181, Example 6.3 (storage): sensible storage `Q = m·c_p·dT` with c_p = 1 Btu/(lb·°F) and 8 lb/gal; 125,000 Btu/day at ΔT = 120 − 80 = 40 °F gives 40 Btu/lb = 320 Btu/gal, so about 390 gal ≈ 50 ft³. The book notes the tank uses average rather than peak demand and is an underestimate. [^kerlin-185]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, p. 185: only the direct beam can be focused, so a solar heat engine's net efficiency is `eta_net = f_unscattered × eta_thermo ≈ 0.51 × 0.30 ≈ 0.15`; a thermal plant rejects 2.5 to 3 times its electrical output. [^kerlin-188]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*, Chapter 3, pp. 185 and 188: land per quad per year scales as 1/η — about 330,000 acres at 15 % and about 165,000 acres at 30 % — against Arizona's 73 million acres. [^likharev-107]: Likharev, Konstantin (2013). *Part SM: Statistical Mechanics*, Chapter 4 (pp. 107–142), p. 107: the latent heat of water, 2.2×10⁶ J/kg, about 0.4 eV per molecule. https://open.umn.edu/opentextbooks/textbooks/part-sm-statistical-mechanics [^derived-st]: Computed for this article from the equations and the cited book values on this page: Knoxville's latitude 50.82° − 15° = 35.82°; the winter-sized area 46,600/(1,050 × 0.5) = 88.8 ≈ 89 ft²; the annual check 1,425 × 0.5 × 60 = 42,750 Btu/day = 91.7 % of 46,600; the January delivery 1,050 × 0.5 × 60 = 31,500 Btu/day = 67.6 %, equal to the flux ratio 1,050/1,550 = 67.7 %; the space-heating areas 322.6–476.2 ft² (fixed) and 227.3–416.7 ft² (two-axis), against the book's printed 325–475 and 225–415; the Laramie/Atlanta HDD ratio 9,038/2,827 = 3.20; the tank 125,000/320 = 390.6 gal = 52.2 ft³; and the latent-to-sensible comparison, 2.2×10⁶ J/kg ÷ 2,326 J/kg per Btu/lb = 946 Btu/lb against 40 Btu/lb over a 40 °F swing, a factor of 24. The molten-salt usable ΔT of about 300 °C is a representative design figure used for the order-of-magnitude comparison, not a book value. [^manual10]: Wikitube MICROSIM_GUIDE sub-manual 10, *Earth, Energy and Environment*, §3.1 (solar thermal: collector area, degree-days and storage), which supplies the page-cited extract of book 048 pp. 175–188 used throughout this page, together with its pitfalls list and its computed cross-checks. [^spec-e40]: Matter & Energy Cluster contract, `_registry/plans/ENERGY_SECTIONS.md` row E40: sim concept (`solar.collector`), `A = Q_day/(I·eta)` with η from 0.2 to 0.8 as the single control; bars for the summer-sized (60 ft²), winter-sized (89 ft²) and two-axis areas; the 1,050/1,550 = 68 % winter-cover gauge that does not move; and the storage tank `V = Q/(c_p·rho·dT)`, 390 gal at ΔT = 40 °F. ## External links - [*Future Energy: Opportunities & Challenges*](https://open.umn.edu/opentextbooks/textbooks/future-energy-opportunities-challenges), Kerlin (2013) — Chapter 3 is the source of every sizing number on this page - [*Part SM: Statistical Mechanics*](https://open.umn.edu/opentextbooks/textbooks/part-sm-statistical-mechanics), Likharev (2013) — Chapter 4 for latent heat at a first-order phase transition - The Wikipedia pair's *External links* section lists collector certification bodies and test standards, which this page's shelf does not carry. <!-- MATTERSIM:BEGIN g33 — Matter & Energy Cluster microsim (framework build, specs/sims/Solar_thermal_energy.json); do not hand-edit inside --> **Microsim — three.js (Wikitube framework):** *Solar thermal energy* <div class="wt-sim" data-src="https://wikitube-3d-microsims.netlify.app/matter/Solar_thermal_energy.html" data-title="Solar thermal energy"></div> *Built from `MICROSIM_GUIDE/specs/sims/Solar_thermal_energy.json`; part of the [[PORTAL_Matter|Matter portal]] spine (section sims and See-also variants).* <!-- MATTERSIM:END --> ## Wikipedia : Wikitube **Strict pair:** [Wikipedia](https://en.wikipedia.org/wiki/Solar_thermal_energy) : [Wikitube](https://en.wikitube.io/wiki/Solar_thermal_energy) · pinned revision [1371964670](https://en.wikipedia.org/w/index.php?oldid=1371964670) · 2026-09-11 ## Previous hub tags Hubs: `Life_Physics`. Portals: [[PORTAL_Energy]]. --- *Matter & Energy Cluster child articles, wave 1 · 2026-09-11 · drafted · Energy row E40 · sim pending (matter/Solar_thermal_energy).*