TOPCon vs HJT vs PERC: cell architectures compared

Walk into a supplier catalogue in 2024 and the module names are dense with acronyms: PERC, TOPCon, HJT, sometimes IBC or "n-type" as a standalone marketing term. These are not brand names; they describe distinct cell architectures with different physics, different manufacturing lineages, and materially different performance curves once a module is on a roof rather than in a datasheet. For anyone specifying or comparing modules, understanding what sits behind these labels is more useful than trusting a single headline efficiency percentage.

This article compares Passivated Emitter and Rear Cell (PERC), Tunnel Oxide Passivated Contact (TOPCon) and Heterojunction (HJT) architectures on the dimensions that actually affect a project: peak efficiency, temperature coefficient, bifaciality, long-term degradation, and manufacturing cost — which ultimately shows up as the price premium (or lack of one) at the point of purchase.

Underlying all three is the same base material, crystalline silicon, and the same fundamental photovoltaic effect described in how solar panels work. What differs is how each architecture manages the two dominant loss mechanisms in a silicon cell: recombination of charge carriers at the cell's surfaces and contacts, and resistive losses in getting current out of the cell.

PERC: the incumbent baseline

PERC became the dominant mass-market architecture through the late 2010s by adding a dielectric passivation layer and local rear contacts to a conventional aluminium back surface field (Al-BSF) cell. The passivation layer reduces recombination at the rear surface and a reflective layer behind it bounces unabsorbed infrared light back into the cell for a second absorption pass. This pushed commercial efficiencies from the high 17s (Al-BSF) into the 21–22.5% range for good production lines.

PERC cells are built on p-type silicon wafers. P-type material is susceptible to light-induced degradation (LID) caused by boron-oxygen defect complexes, which is discussed in detail in solar module degradation and warranties. PERC remains the least expensive architecture to manufacture because it requires the fewest additional process steps beyond a conventional line, and it remains a reasonable choice for cost-sensitive, non-bifacial installations where its slightly lower efficiency and higher temperature coefficient are acceptable trade-offs.

TOPCon: passivated contacts on both sides

TOPCon adds an ultra-thin tunnel oxide layer plus a doped polysilicon layer on the rear of the cell, which passivates the rear surface far more effectively than PERC's dielectric-only approach while still allowing collected charge to tunnel through to the metal contact. TOPCon cells are typically built on n-type wafers, which are inherently less prone to boron-oxygen LID and to light- and elevated-temperature-induced degradation (LeTID) than p-type wafers, though n-type material introduces its own minor degradation modes.

The rear passivation and n-type base together give TOPCon a meaningful efficiency edge over PERC — commercial modules commonly reach 22–23.5% — a better (less negative) temperature coefficient, typically around -0.29 to -0.34%/°C, and inherent bifaciality because the rear contact scheme, unlike PERC's full-area aluminium rear, allows some light to reach the rear cell surface. Manufacturing TOPCon requires additional deposition and diffusion steps compared with PERC, which raises production cost, though as of the mid-2020s much of the industry's new capacity is TOPCon rather than PERC, and the cost gap has narrowed as the process has scaled.

HJT: crystalline silicon meets amorphous silicon

Heterojunction technology sandwiches a thin crystalline silicon wafer between layers of intrinsic and doped amorphous silicon, with transparent conductive oxide (TCO) layers and a low-temperature silver (or silver-copper) grid on both faces. The amorphous silicon layers provide excellent surface passivation without the high-temperature diffusion steps used in PERC and TOPCon manufacturing, which is why HJT cells are naturally symmetric and naturally bifacial — both faces are built the same way.

HJT delivers the best temperature coefficient of the three, often -0.24 to -0.26%/°C, because the amorphous silicon passivation suppresses the temperature-dependent recombination that dominates in conventional homojunction cells. Peak commercial efficiencies are comparable to or slightly ahead of TOPCon, in the 22.5–24% range, and HJT cells are also n-type based, giving similar LID resistance advantages. The trade-offs are a more specialised manufacturing process (low-temperature steps only, since amorphous silicon degrades above roughly 200°C, which rules out the high-temperature firing used for conventional silver pastes), higher silver consumption per cell, and a correspondingly higher module cost.

Comparison at a glance

Temperature coefficient in practice

The temperature coefficient matters most in hot climates and on roofs with poor rear ventilation, where cell operating temperature can reach 60–70°C on a clear summer day against the 25°C STC reference.

Worked example

Take a 435 Wp PERC module with a temperature coefficient of -0.38%/°C and a 435 Wp TOPCon module with a coefficient of -0.31%/°C, both reaching an operating cell temperature of 65°C on a summer afternoon.

Temperature rise above STC: 65 − 25 = 40°C.

PERC power loss: 435 × 0.38% × 40 = 435 × 0.0038 × 40 = 66.1 W. Output: 435 − 66.1 = 368.9 W. TOPCon power loss: 435 × 0.31% × 40 = 435 × 0.0031 × 40 = 53.9 W. Output: 435 − 53.9 = 381.1 W.

At identical rated power, the TOPCon module delivers roughly 3.3% more power at that operating temperature, purely from the temperature coefficient difference. Over a full cooling season in a hot climate this compounds into a measurable annual energy difference, which is one reason TOPCon has displaced PERC so quickly in utility-scale and hot-climate residential markets alike.

Bifaciality and rear-side gain

Bifaciality factor — the ratio of rear-side power output to front-side output under identical irradiance — is a design consequence of how much of the rear cell surface is left uncovered by opaque metallisation. PERC's typical rear design uses a near-full-area aluminium contact, which blocks rear light almost entirely unless the manufacturer specifically redesigns the rear grid for bifacial PERC variants, in which case bifaciality is modest. TOPCon's polysilicon rear layer is thin and can be paired with a fine metal grid rather than a solid layer, giving good bifaciality by default. HJT's symmetric construction gives it the highest bifaciality factors on the market. The practical energy gain this bifaciality delivers depends heavily on installation conditions and is covered in detail in bifacial solar modules.

Degradation profiles

All three architectures degrade over time, but the mechanisms and magnitudes differ.

Manufacturers of n-type products (TOPCon, HJT) typically publish flatter degradation curves and consequently stronger performance warranties, guaranteeing higher output percentages at 25 or 30 years than typical PERC warranties.

Cost and payback implications

Higher efficiency and better temperature behaviour do not automatically translate into a better financial outcome; the module premium has to be justified by the extra energy yield over the system's life. As a rule of thumb, TOPCon has reached near cost-parity with PERC in many markets as production has scaled, making it the more straightforward choice on a cost-per-watt and cost-per-kWh basis for most residential and commercial roofs. HJT commands a real premium, typically justified on space-constrained roofs where maximising kWp per square metre or minimising temperature-related losses in a hot climate outweighs the higher upfront cost. Running both scenarios through /calculators/payback with realistic degradation and temperature assumptions for the specific site is the only reliable way to settle the comparison for a given roof.

Standards and references

• IEC 61215 — Design qualification testing that all three architectures must pass regardless of cell technology, including thermal cycling and damp heat sequences relevant to LID/LeTID/PID assessment.
• IEC TS 63202 — Technical specification addressing bifacial PV module performance measurement and rear-side considerations.
• IEC 61730 — Module safety qualification, relevant to the different metallisation and lamination approaches used across architectures.
• IEC 60904-1-2 — Measurement of current-voltage characteristics of bifacial photovoltaic devices.

Frequently asked questions

Is TOPCon simply an upgraded PERC cell?

Not exactly. TOPCon shares the same wafer starting point conceptually but changes the wafer doping type (n-type instead of p-type) and adds a passivated contact structure on the rear that PERC's design does not have. It requires additional deposition and diffusion process steps rather than being a drop-in upgrade to an existing PERC production line, though many manufacturers have converted PERC lines to TOPCon with substantial equipment changes.

Does a higher bifaciality factor always mean more energy?

No. Bifaciality factor describes the cell's potential rear-side response, but actual rear-side energy gain depends on mounting height, ground or roof albedo, row spacing and the presence of obstructions. A high-bifaciality module on a low-clearance, dark-roof residential installation may gain only 2–4% extra energy, while the same module on a high, reflective, open-field tracker mount can gain 10% or more.

Which architecture has the best warranty?

As a general pattern, HJT and TOPCon modules from established manufacturers now commonly carry 25 to 30-year product warranties and performance warranties guaranteeing 87–92% of rated output at year 25 to 30, compared with roughly 80–85% typical of older PERC warranties. Warranty terms vary by manufacturer regardless of architecture, so the specific warranty document should always be checked rather than assumed from the cell type alone.

Is PERC becoming obsolete?

PERC remains in production and is still a reasonable, lower-cost choice, particularly where bifaciality and best-in-class temperature performance are not priorities. However, industry capacity investment has shifted decisively toward TOPCon, and PERC's market share is shrinking as TOPCon reaches similar cost levels while offering measurably better efficiency and temperature behaviour.

Can HJT and TOPCon modules be mixed in the same array?

Technically they can be wired together if electrical parameters are compatible, but it is not good practice. Mixing architectures with different temperature coefficients, degradation rates and I-V curve shapes on the same string creates long-term mismatch losses as the modules drift apart in performance, and complicates warranty claims. Each string should use a single module model where possible.

Sources

• Fraunhofer ISE — Photovoltaics Report
• NREL — Best Research-Cell Efficiency Chart
• PVEducation — Types of Solar Cells
• IEA PVPS — Technology reports