1. Theory

Definition: The tunnel oxide passivated contact (TOPCon) is a carrier-selective contact made of a silicon oxide of about 1 to 2 nm and a layer of doped silicon deposited on it. The oxide passivates the interface chemically, and the doped layer, which is crystallized by an anneal into poly-silicon, lets one carrier type pass and blocks the other. In the industrial cell the contact is on the rear, with phosphorus doping on an n-type Czochralski wafer, and the front has a boron-diffused emitter of 110 to 140 $Ω/ sq$ with a passivation stack of aluminium oxide and silicon nitride.
Origin: Feldmann et al. of Fraunhofer ISE presented the rear contact in 2013 with a 1.4 nm wet-chemical oxide and a 20 nm phosphorus-doped silicon layer, an implied voltage above 710 mV, 23.7 % and 703 mV on 1.1 % metal area; the maximum tolerable oxide thickness was 2 nm.
Industrial cell: Mack et al. describe the baseline with a thermal oxide, 80 nm of in-situ doped LPCVD poly-Si, a boron emitter of about 110 $Ω/ sq$ , AlO $_{x}$ with SiN $_{x}$ on the front, AgAl paste on the front and Ag on the rear, a laser-enhanced contact step and a champion cell of 24.0 % (708 mV). Feldmann et al. made 22.5 % cells with PECVD poly-Si of 100 or 170 nm after a 1050 °C anneal of 80 min and a firing at a peak of about 760 °C.
Process routes: The oxide is grown wet-chemically, by plasma, by ozone or thermally, and the silicon layer is deposited by LPCVD, PECVD, APCVD, ALD or PVD and doped in situ, by diffusion or by implantation; parasitic poly-Si on the front must be etched (Krieg et al.: two minutes in hot KOH inline for in-situ doped layers, five minutes in a batch for ex-situ ones). Compared with PERC the cell needs a boron diffusion and extra cleaning (TaiyangNews), and compared with n-PERT three additional steps.
Records: Trina reported 26.58 % for a large industrial cell in 2024 (ISFH CalTeC), JinkoSolar 26.1 % in 2022, Fraunhofer ISE 26 % for the TOPCoRE rear-junction cell, and ISFH 26.1 % for a POLO-IBC cell. The limit of a single-junction silicon cell is 29.4 %.
Market: pv Europe says that TOPCon was set to overtake PERC as the market leader in 2024; the ITRPV 14th edition forecasts 60 % for TOPCon in 2033, 19 % for heterojunction and 5 % for back-contact cells, and a share of n-type wafers rising from 15 % to about 65 %.
Degradation: The aluminium oxide on the boron emitter is attacked by ultraviolet light above 3.4 eV, which breaks Si-H bonds and loses voltage, and a slow dark loss with recovery under light is linked to charge trapping in the oxide.

2. Working Principle

Selective transport: The oxide is thin enough for the carriers to tunnel, and the tunnelling probability falls exponentially with thickness and with the square root of barrier height and effective mass. For silicon dioxide the barrier of the holes is higher than that of the electrons, so the oxide favours the electrons, which makes the n-type contact easier than the p-type one. The heavily doped layer behind it makes the number of minority holes small, so the recombination current $J_{0}$ is a few fA/cm $^{2}$ .
Pinholes: At anneal temperatures up to 1000 °C the oxide thins from 1.5 to 1.0 nm and may break open; transport through pinholes has been proposed (Gan and Swanson, Peibst et al.), and Choi et al. find that selectivity is lost when the thinned oxide lets both carriers reach the doped layer. The optimum oxide is 1.2 to 1.5 nm, and in the 2013 work the passivation dropped strongly above 900 °C (Lee et al. anneal at 925 °C).
Trade-off: The tunnelling that gives the contact resistivity is the same that lets the holes through, so a thicker oxide lowers $J_{0}$ and raises $ρ_{c}$ ; 0.1 nm changes $ρ_{c}$ by a factor of 2.7 in the model. The doping of the poly-Si above $1 0^{20}$ cm $^{- 3}$ brings $ρ_{c}$ to about 1 m $Ω$ cm $^{2}$ .
Metal contact: The silver paste is fired on the poly-Si; if it penetrates the layer, the recombination rises. A thicker layer protects the oxide (J $_{0, m e t}$ falls by 65 % from 100 to 170 nm), a thinner one absorbs less light and saves time, so 80 to 170 nm are used with screen printing and about 60 to 125 nm with plated contacts.
Voltage: The rear contact allows implied voltages of 700 to 740 mV because its $J_{0}$ is about 10 fA/cm $^{2}$ or less, so the emitter and the metal dominate the losses: the boron emitter has 15 to 40 fA/cm $^{2}$ and the metallized front part 600 to 700 fA/cm $^{2}$ (Feldmann et al. 2019, Mack et al.).

3. Applications

Mass production: n-type bifacial modules with TOPCon cells of 23.5 to 24 % in production (Mack et al., TaiyangNews) and 25.9 to 26.58 % reported for large industrial cells (Trina), replacing PERC as the market leader from about 2024.
Upgrade path: Compared with a PERC line the process adds a boron diffusion, extra cleaning, and the oxide and poly-Si steps (TaiyangNews), which is the reason for the several LPCVD, PECVD and other deposition variants.
Next structures: The poly-Si contact on both sides (TOPCoRE or POLO-IBC) and the tandem with a perovskite top cell build on this contact; the heterojunction and back-contact cells follow on the next pages.

Tunnel Oxide and Passivated Contact

Name	Formula	Related Formulae	Variables/Constants
Tunnelling probability through the oxide and the carrier selectivity	$T = e^{- 2 κ t}, κ = \frac{2 m ^{*} φ}{ℏ}$	$\frac{T _{e}}{T _{h}} = e^{2 t (κ_{h} - κ_{e})}$ $κ (n m^{- 1}) = 5.123 \frac{m ^{*}}{m _{0}} \frac{φ}{1 eV}$	$T : t u nn e l l in g p r o babi l i t y in t h e W K B a pp r o x ima t i o n f or a r ec t an g u l a r ba r r i er$ $t : o x i d e t hi c k n ess$ $m^{*} : e f f ec t i v e ma ss in t h e o x i d e$ $φ : ba r r i er h e i g h t$ $ℏ : r e d u ce d P l an c k co n s t an t$ $T_{e}, T_{h} : p r o babi l i t i es o f e l ec t r o n s an d h o l es, v an d er V osse n d escr ib es t h e m o d e l in w or d s (t h e p r o babi l i t y f a l l s e x p o n e n t ia l l y w i t h t h e t hi c k n ess an d t h e s q u a r e r oo t o f ba r r i er an d ma ss) an d u ses m_{e} = 0.3 an d m_{h} = 0.4 (in u ni t s o f t h e e l ec t r o n ma ss), t h e f or m u l a i s t h e s t an d a r d W K B f or m an d ba c k g r o u n d k n o w l e d g e, t h e ba r r i er s 3.2 e V (e l ec t r o n s) an d 4.7 e V (h o l es) a r e t h e u s u a l S i / S i O 2 ban d o f f se t s (ba c k g r o u n d) w hi l e t h e t h es i s t ab l e a s I r e a d i t l i s t s t h e m t h e o t h er w a y r o u n d$ $(co m p u t e d : κ_{e} = 5.02 n m^{- 1} an d κ_{h} = 7.02 n m^{- 1}; t = 1.0, 1.4, 1.5, 2.0 an d 2.4 nm g i v e T_{e} = 4.4 \times 1 0^{- 5}, 7.9 \times 1 0^{- 7}, 2.9 \times 1 0^{- 7}, 1.9 \times 1 0^{- 9} an d 3.4 \times 1 0^{- 11}, an d T_{e} / T_{h} = 55, 274, 409, 3040 an d 15100; C h o i e t a l . p r in t a r a t i o o f 6.31 a t 1.5 nm an d 19.04 a t 2.4 nm, i . e . 1.23 p er nm in t h e l o g a r i t hm f or b o t h)$
Saturation current density of the passivated contact	$J_{0} \approx \frac{q n _{i}^{2}}{N _{e f f}} (S_{in t} + v_{t h} T_{h})$	$S_{e q} = \frac{J _{0} N _{e f f}}{q n _{i}^{2}}$	$J_{0} : s a t u r a t i o n c u r r e n t d e n s i t y o f t h e co n t a c t$ $N_{e f f} : e f f ec t i v e d o p in g o f t h e r e g i o n b e hin d t h e o x i d e$ $S_{in t} : r eco mbina t i o n v e l oc i t y a t t h e in t er f a ce$ $v_{t h} : t h er ma l v e l oc i t y$ $T_{h} : t u nn e l l in g p r o babi l i t y o f t h e min or i t y h o l es$ $S_{e q} : e q u i v a l e n t v e l oc i t y o f a m e a s u r e d J_{0}, v an d er V osse n d escr ib es J_{0} a s t h e s u m o f a r eco mbina t i o n t er m an d a t u nn e l l in g t er m d i v i d e d b y an e f f ec t i v e d o p in g (N_{e f f} = 4 \times 1 0^{18} c m^{- 3} in hi s T ab l e 2.1), t h e e q u a t i o n i s n o t p r in t e d in t h e e x t r a c t an d t h e f or m h er e i s m y r eco n s t r u c t i o n w i t h t h e u s u a l f a c t or q n_{i}^{2}$ $(co m p u t e d a t 25 C w i t h n_{i} = 8.70 \times 1 0^{9} c m^{- 3} : q n_{i}^{2} / N_{e f f} = 3.0 \times 1 0^{- 3} f A / c m^{2} p er c m / s; v_{t h} T_{h} = 0.029 c m / s a t 1.4 nm (v_{t h} = 1 0^{7} c m / s a ss u m e d), n e g l i g ib l e; J_{0} = 2.5, 4, 10 an d 13 f A / c m^{2} cor r es p o n d t o S_{e q} = 820, 1300, 3300 an d 4300 c m / s; m e a s u r e d : 9 t o 13 (F e l d mann 2013), 4 pl ana r an d 10 t e x t u r e d (M a c k), 1.9 t o 2.4 b e f or e f i r in g (L ee), ab o u t 10 n - t y p e an d 20 p - t y p e (v an d er V osse n))$
Contact resistivity of the tunnel oxide	$ρ_{c} \propto \frac{1}{N _{e f f} v _{t h} T _{e}}$	$ρ_{c} \propto e^{2 κ_{e} t}, \frac{ρ _{c} ( t + 0.1 nm )}{ρ _{c} ( t )} = e^{0.2 κ_{e}}$	$ρ_{c} : s p ec i f i c co n t a c t r es i s t i v i t y o f t h e o x i d e an d t h e s i l i co n l a y er$ $T_{e} : e l ec t r o n t u nn e l l in g p r o babi l i t y, o t h er sy mb o l s a s ab o v e, v an d er V osse n s t a t es t h e p r o p or t i o na l i t y in w or d s an d n o t es t ha t J_{0} an d ρ_{c} b o t h d e p e n d o n t u nn e l l in g c u r r e n t s, so r e d u c in g o n e t e n d s t o in cr e a se t h e o t h er, t h e e x p o n e n t ia l f or m f o l l o w s f r o m t h e f i r s t r o w an d i s ba c k g r o u n d k n o w l e d g e$ $(co m p u t e d : 0.1 nm m or e o x i d e m u l t i pl i es ρ_{c} b y 2.7 an d l o w er s t h e h o l e t u nn e l l in g b y 4.1; m e a s u r e d : ab o u t 10 m Ω c m^{2} (F e l d mann 2013), 3 m Ω c m^{2} a t o pt im u m f i r in g an d 4 t o 7 ini t ia l l y (F e l d mann 2019), ab o u t 1 m Ω c m^{2} f or d o p in g ab o v e 1 0^{20} c m^{- 3}, b e l o w 1 (M a c k), 17 t o 19 f a l l in g t o 1.9 f or 50 nm p o l y - S i a t hi g h f i r in g (L ee), 20 f or c h e mi c a l o x i d es (N R E L, q u o t e d b y v an d er V osse n))$
Implied open-circuit voltage	$i V_{oc} = \frac{k _{B} T}{q} ln [\frac{Δ n ( Δ n + N _{D} )}{n _{i}^{2}}]$	$Δ n = \frac{- N _{D} + N _{D}^{2} + 4 n _{i}^{2} e ^{q i V_{oc} / k_{B} T}}{2}$	$i V_{oc} : im pl i e d o p e n - c i r c u i t v o l t a g e f r o m t h e e x cess c a r r i er d e n s i t y$ $Δ n : e x cess e l ec t r o n - h o l e co n ce n t r a t i o n$ $N_{D} : d o n or co n ce n t r a t i o n o f t h e n - t y p e w a f er$ $n_{i} : in t r in s i c co n ce n t r a t i o n, F e l d mann e t a l . p r in t t h e e q u a t i o n w i t h a g a r b l e d l a y o u t, t h e f or m i s t h e s t an d a r d o n e an d t h e s am e a s o n t h e S o l a r C e l l L oss M ec hani s m s p a g e$ $k_{B} : B o l t z mann co n s t an t$ $(co m p u t e d w i t h N_{D} = 1.25 \times 1 0^{15} c m^{- 3} (4 Ω c m, m o bi l i t y 1250 c m^{2} / V s a ss u m e d) : i V_{oc} = 710, 720 an d 740 mV n ee d Δ n = 8.1, 10.0 an d 15.0 \times 1 0^{15} c m^{- 3}; C h o i e t a l . (o x i d e w i t h a - S i, ann e a l in N_{2}) : 556 mV a s d e p os i t e d, 651 (600 C, 1 min), 683 (900 C) an d 697 (1000 C, 1 min) b u t 657 mV f or 5 min a t 1000 C; F e l d mann e t a l . 2019 : 690 t o 710 mV a f t er ann e a l, u p t o 740 mV a f t er h y d r o g e na t i o n o f t es t s am pl es, 717 mV o n ce l l p r ec u r sor s)$

Saturation Current, Voltage and Efficiency

Name	Formula	Related Formulae	Variables/Constants
Saturation current budget and open-circuit voltage	$J_{0} = J_{0, e} + J_{0, b} + J_{0, r e a r}$	$V_{oc} = \frac{k _{B} T}{q} ln (\frac{J _{sc}}{J _{0}} + 1)$ $J_{0} = \frac{J _{sc}}{e ^{q V_{oc} / k_{B} T} - 1}$	$J_{0, e} : b or o n e mi tt er in c l u d in g t h e m e t a l l i z e d p a r t$ $J_{0, b} : b u l k t er m (A ug er - l imi t e d in n - t y p e s i l i co n)$ $J_{0, r e a r} : p a ss i v a t e d r e a r co n t a c t$ $J_{sc} : s h or t - c i r c u i t c u r r e n t d e n s i t y, F e l d mann e t a l . w r i t e t h e s u m an d t h e b u l k t er m in a g a r b l e d l a y o u t, t h e d i o d e r e l a t i o n s a r e t h e s t an d a r d o n es$ $k_{B} : B o l t z mann co n s t an t$ $(F e l d mann e t a l . 2013 : J_{0, e} = 64, J_{0, b} = 22 an d J_{0, r e a r} = 9 f A / c m^{2} g i v e 95 f A / c m^{2}, w hi c h f or 38.4 m A / c m^{2} i s 686.7 mV a g ain s t 690.8 mV m e a s u r e d (81 f A / c m^{2} im pl i e d); t h e ce l l w i t h t h e u n p a ss i v a t e d r e a r ha s J_{0} = 593 f A / c m^{2} (614 im pl i e d b y 638.3 mV an d 37.8 m A / c m^{2}), so t h e u n p a ss i v a t e d r e a r a dd s ab o u t 507 f A / c m^{2}; im pl i e d J_{0} o f l a t er ce l l s : 53 (703.3 mV, 41.0), 48 (M a c k, 706 mV, 41.3), 34 (M a c d o na l d, 715 mV, 41.6), an d 14 f A / c m^{2} f or t h e m o d e l l e d 738 mV, 42.4 m A / c m^{2}; a l l co m p u t e d a t 25 C)$
Metal-induced recombination	$J_{0, e} = J_{0, p a ss} (1 - f) + J_{0, m e t} f$	$Δ J_{0, e} = f (J_{0, m e t} - J_{0, p a ss})$ $V_{oc, g a p} = i V_{oc} - V_{oc} = \frac{k _{B} T}{q} ln \frac{J _{0, w i t h m e t a l}}{J _{0, w i t h o u t m e t a l}}$	$J_{0, p a ss} : p a ss i v a t e d p a r t o f t h e e mi tt er$ $J_{0, m e t} : p a r t u n d er t h e m e t a l (p er m e t a l l i z e d a r e a)$ $f : m e t a l l i z e d f r a c t i o n$ $V_{oc, g a p} : l oss b e tw ee n t h e im pl i e d an d t h e ce l l v o l t a g e c a u se d b y t h e m e t a l, t h e a r e a - w e i g h t e d s u m i s t h e s am e a s o n t h e A l - B S F p a g e$ $k_{B} : B o l t z mann co n s t an t$ $(F e l d mann e t a l . 2013 : 11 f A / c m^{2} p a ss i v a t e d, 1800 f A / c m^{2} u n d er t h e co n t a c t s an d 3.0 % m e t a l g i v e 64.7 a g ain s t 64 f A / c m^{2} p r in t e d, co m p u t e d; F e l d mann e t a l . 2019 : J_{0, m e t, r e a r} = 530 \pm 140 f A / c m^{2} f or 170 nm p o l y - S i an d ab o u t t h r ee t im es m or e f or 100 nm, t h e v o l t a g e g a p i s 15 mV (170 nm) an d 25 mV (100 nm), i . e . t h e m e t a l r ai ses J_{0} b y 79 an d 165 %, co m p u t e d; L ee e t a l . : J_{o m} = 30 t o 40 f A / c m^{2} f or 150 nm, 103 f or 50 nm a t 880 C)$
Series resistance, pseudo fill factor and contact optimization	$R_{S} = \frac{V _{oc} J _{sc} ( pF F - F F )}{J _{m pp}^{2}}$	$Δ η \approx \frac{J _{m pp}^{2} Δ R _{S}}{P _{in}}$	$R_{S} : ser i es r es i s t an ce f r o m t h e co m p a r i so n o f t h e S u n s - V_{oc} c u r v e w i t h t h e l i g h t c u r v e (F e l d mann e t a l .), pF F : p se u d o f i l l f a c t or w i t h o u t ser i es r es i s t an ce$ $F F : f i l l f a c t or$ $J_{m pp} : c u r r e n t d e n s i t y a t ma x im u m p o w er$ $P_{in} : in c i d e n t p o w er, t h e seco n d r e l a t i o n i s m y f i r s t - or d er es t ima t e$ $(F e l d mann e t a l . 2013 : R_{S} = 0.41 (S u n s - V_{oc}) an d 0.37 Ω c m^{2} (l i g h t) p r in t e d; 703.3 mV, 41.0 m A / c m^{2}, pF F = 84.6 %, F F = 82.2 % g i v e 0.45 t o 0.48 Ω c m^{2} f or an a ss u m e d V_{m pp} = V_{oc} - (80 t o 100) mV, co m p u t e d; M a c k e t a l . : t h e l a ser - e nhan ce d co n t a c t o pt imi z a t i o n l o w er s t h e a v er a g e R_{S} f r o m 2.81 t o 0.46 Ω c m^{2}, w hi c h g i v es 3.5 p o in t s a t J_{m pp} = 38.7 m A / c m^{2} a g ain s t u p t o 4 p o in t s p r in t e d, co m p u t e d)$
Efficiency from the reported parameters	$η = \frac{V _{oc} J _{sc} F F}{P _{in}}, P_{in} = 100 mW/c m^{2}$	$η_{c h ec k} = \frac{V _{oc} J _{sc} F F}{1000}$ $Δ η = η_{p r in t e d} - η_{c h ec k}$	$η : e f f i c i e n cy, V_{oc} in mV, J_{sc} in m A / c m^{2}, F F a s a f r a c t i o n (η_{c h ec k} i s in p er ce n t : t h e 1000 t u r n s mV in t o V, an d 1 mW / c m^{2} i s 1 p er ce n t o f P_{in}), t h e v a l u es a r e b es t or c ham p i o n ce l l s a s l ab e l l e d b y t h e so u r ces$ $(F e l d mann e t a l . 2013 : 690.8, 38.4, 82.1 % g i v e 21.78 % (p r in t e d 21.81); 693.0, 41.4, 81.2 % g i v e 23.30 % (23.3); 703.3, 41.0, 82.2 % g i v e 23.70 % (23.72); F e l d mann e t a l . 2019 : 691.2, 40.4, 80.7 % g i v e 22.54 % (22.5); M a c k e t a l . : 706, 41.3, 81.6 % g i v e 23.79 % (23.8), 708, 41.6, 81.6 % g i v e 24.03 % (24.0), pl a t e d 715, 41.0, 82.0 % g i v e 24.04 % (24.0); M a c d o na l d : 715, 41.6, 83.4 % g i v e 24.81 % (24.8) an d t h e m o d e l l e d 738, 42.4, 85.0 % g i v e 26.60 % (26.6); co m p u t e d)$
Absorption in the poly-silicon layer	$A = 1 - e^{- α t}$	$A_{2 p a sses} \approx 1 - e^{- 2 α t}, λ (nm) = \frac{1239.84}{E ( eV )}$	$A : ab sor b e d f r a c t i o n in o n e p a ss$ $α : ab sor pt i o n coe f f i c i e n t o f t h e d o p e d p o l y - s i l i co n$ $t : l a y er t hi c k n ess$ $E : p h o t o n e n er g y, p v ma g a z in e r e p or t s t ha t p h o t o n s b e l o w 370 nm a r e ab sor b e d b y t h e d o p e d p o l y - S i o f t h e r e a r s i d e, t h e B eer - L amb er t f or m i s ba c k g r o u n d k n o w l e d g e an d t h e ab sor pt i o n coe f f i c i e n t s o f t h e so u r ces a r e n o t g i v e n$ $(a ss u m e d α = 1 0^{6} c m^{- 1} in t h e u l t r a v i o l e t : 10 nm ab sor b 63 % an d 80 nm 99.97 %, co m p u t e d; 3.4 e V cor r es p o n d s t o 365 nm, U V - A (315 t o 400 nm) t o 3.10 t o 3.94 e V an d U V - B (290 t o 320 nm) t o 3.87 t o 4.28 e V; M a c d o na l d m o d e l s + 0.4 p o in t s (25.4 t o 25.8 %) f or p o l y - S i u n d er t h e f r o n t f in g er s)$

Efficiency Potential and Degradation

Name	Formula	Related Formulae	Variables/Constants
Efficiency potential of the passivated-contact cell	$Δ V_{oc} = \frac{k _{B} T}{q} ln \frac{J _{0, 1}}{J _{0, 2}}$	$\frac{J _{0, 2}}{J _{0, 1}} = e^{- Δ V_{oc} q / k_{B} T}, \frac{η}{η _{l im}} (fraction of the limit)$	$Δ V_{oc} : v o l t a g e g ain w h e n t h e s a t u r a t i o n c u r r e n t f a l l s f r o m J_{0, 1} t o J_{0, 2}$ $η_{l im} : l imi t in g e f f i c i e n cy o f a s in g l e - j u n c t i o n s i l i co n ce l l, N i e w e l t e t a l . r e v i se i t t o 29.4 % (ab s t r a c t p a g e, t h e p a r am e t r i z a t i o n w a s n o t r e a d), T ai y an g N e w s q u o t es a p o t e n t ia l o f 28.7 % f or t h e I S F H P O L O - I B C s t r u c t u r e$ $k_{B} : B o l t z mann co n s t an t$ $(M a c d o na l d : t h e c ham p i o n ce l l (24.8 %, 715 mV, 41.6 m A / c m^{2}, 83.4 %) r e a c h es 26.6 % w i t h 738 mV, 42.4 m A / c m^{2} an d 85.0 % a f t er f i v e c han g es (p o l y - S i u n d er t h e f in g er s, l i f e t im e, 20 μ m f in g er s, l ess m e t a l r eco mbina t i o n, r es i s t i v i t y); t h e 23 mV n ee d J_{0} t o f a l l t o 41 %; 26.58 % (T r ina) i s 90 % o f 29.4 % an d 2.8 p o in t s b e l o w i t, co m p u t e d)$
Wavelength and dose of the ultraviolet degradation	$λ (nm) = \frac{1239.84}{E ( eV )}$	$D_{U V} = \int G_{U V} d t$	$λ : w a v e l e n g t h$ $E : p h o t o n e n er g y$ $D_{U V} : u l t r a v i o l e t d ose in k W h / m^{2}$ $G_{U V} : u l t r a v i o l e t i r r a d ian ce, t h e U N S W s t u d y g i v es t h e t h r es h o l d a s p h o t o n s ab o v e 3.4 e V t ha t b r e ak S i - H b o n d s main l y a t t h e A l O x / p + in t er f a ce an d r e d i s t r ib u t e h y d r o g e n, t h e N R E L s t u d y l ink s a s l o w d a r k l oss an d a f a s t r eco v er y u n d er l i g h t t o c ha r g e t r a pp in g in A l 2 O 3, t h e a r t i c l es d o n o t p r in t t h e l oss o f t h e U N S W s t u d y$ $(U N S W : 61.1 k W h / m^{2} o f U V - B an d 49 k W h / m^{2} o f U V - A a t 60 C o n 182 mm, 140 μ m bi f a c ia l ce l l s w i t h 5 nm A l O x; N R E L : 2.3 t o 3.2 % n o n - r eco v er ab l e l oss a f t er a o n e - y e a r - e q u i v a l e n t d ose, u p t o 500 h o f d a r k m o ni t or in g, r eco v er y w i t hin seco n d s t o min u t es u n d er U V L E D l i g h t; 3.4 e V = 365 nm, co m p u t e d)$

Important Questions

Ques: Is the rear J0 of the passivated contact 2.5, 4 to 10 or 9 to 13 fA/cm2?

Feldmann et al. found 9 to 13 $fA/c m^{2}$ for the rear contact in 2013, Mack et al. give 4 $fA/c m^{2}$ on a planar alkaline-etched surface and 10 $fA/c m^{2}$ on an alkaline-textured one, and Lee et al. measured 1.9 to 2.4 $fA/c m^{2}$ before firing and about 2.5 $fA/c m^{2}$ after firing at 880 °C on polished wafers annealed at 925 °C. Van der Vossen quotes about 10 $fA/c m^{2}$ for n-type and 20 $fA/c m^{2}$ for p-type contacts. A textured surface and a shorter anneal raise the value, polished samples lower it. For industrial textured cells 4 to 10 $fA/c m^{2}$ is more likely, and 2.5 $fA/c m^{2}$ is the floor of polished samples. With $N_{e f f} = 4 \times 1 0^{18}$ $c m^{- 3}$ these values are equivalent velocities of 1300 to 3300 $cm/s$ (computed). Reference: Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013); Mack, Kafle, Tessmann, Krieg, Bashardoust, Lohmuller, Belledin, Saint-Cast, Hoffler, Ourinson, Fellmeth, Steinhauser, Kluska and Wolf, ‘Status and Perspective of Industrial TOPCon Solar Cell Development at Fraunhofer ISE’, WCPEC-8, Milan (2022); Lee, Yang, Lee, Lee, Song, Kang, Yoon and Park, ‘Effect of poly-Si Thickness and Firing Temperature on Metal Induced Recombination and Contact Resistivity of TOPCon Solar Cells’, Current Photovoltaic Research 9(4), 128-132 (2021); van der Vossen, ‘Optimization of passivating contacts for high-efficiency p-type silicon solar cells’, MSc thesis, Delft University of Technology (2017).

Ques: Is the efficiency limit of silicon 28.7 % or 29.4 %?

TaiyangNews quotes a theoretical potential of 28.7 % for the ISFH POLO structure with interdigitated back contacts. Niewelt et al. revise the limiting efficiency of a single-junction silicon cell to 29.4 %, within 0.1 points of other recent assessments, and Fell et al. use 29.4 % as the Auger limit in their tandem paper (2025). They are not the same quantity. The 29.4 % is the intrinsic bound set by Auger and radiative recombination, whereas 28.7 % belongs to one cell structure; the page does not say how it was obtained, so I read it as a potential with realistic contacts. I could not read the paper of Richter et al. from which the 29.4 % descends. The Trina cell of 26.58 % is 90 % of 29.4 % (computed). Reference: TaiyangNews, ‘What’s Top in TOPCon Solar Cell Technology?’; Niewelt, Steinhauser, Richter et al., ‘Reassessment of the intrinsic bulk recombination in crystalline silicon’, Sol. Energy Mater. Sol. Cells 235, 111467 (2022) (abstract page); Fell, Fischer, Bivour, Messmer, Schon, Schubert and Glunz, ‘Elucidating the efficiency limit of silicon-based monolithic tandem cells through the combination of Auger and Shockley-Queisser limits’, RSC journal article (25 September 2025, DOI suffix d5el00085h); pv magazine, ‘Trina Solar claims record-breaking efficiency of 26.58% for TOPCon solar cell’ (21 November 2024).

Ques: Is the electron-to-hole selectivity of the oxide 6 or several hundred?

Choi et al. print an electron-to-hole tunnelling ratio of 6.31 for a 1.5 nm oxide and 19.04 for 2.4 nm, which is a factor of 1.23 per nm in the logarithm, and they attribute the selectivity mainly to the effective mass. The WKB form with masses of 0.3 and 0.4 (van der Vossen) and the usual barriers of 3.2 eV for electrons and 4.7 eV for holes gives 409 at 1.5 nm and 15 100 at 2.4 nm (computed), about 65 times more, because the barrier heights differ. The thesis table as I read it lists the barriers the other way round, which I take as a label swap. The barrier-based figure is the more likely for the bare oxide; the real selectivity also needs the doping of the poly-Si, which Choi et al. do not model. Reference: Choi, Min, Jeong, Lee, Kang, Song, Kang, Lee, Kim and Kim, ‘Structural evolution of tunneling oxide passivating contact upon thermal annealing’, Sci. Rep. 7, 12853 (2017); van der Vossen, ‘Optimization of passivating contacts for high-efficiency p-type silicon solar cells’, MSc thesis, Delft University of Technology (2017).

Ques: Is the tunnel oxide 1.5 nm or 2.55 nm thick?

Choi et al. measured the as-deposited oxide as 1.5 nm in transmission electron microscopy and as 2.55 nm by spectroscopic ellipsometry. After annealing for 5 min the TEM value stays at 1.5 nm at 600 °C and falls to 1.4, 1.2 and 1.0 nm at 800, 900 and 1000 °C; they propose 1.2 to 1.5 nm as the optimum. Feldmann et al. give about 14 Å by ellipsometry and 20 Å as the largest tolerable thickness, and Feldmann et al. in 2019 about 1.2 nm. The microscope sees the layer directly, whereas the ellipsometer must fit a model of an oxide under amorphous silicon, so 1.5 nm is more likely for the actual oxide; this is my reading. The implied voltage rises from 556 mV as deposited to 683 mV at 900 °C and falls at 1000 °C for 5 min to 657 mV. Reference: Choi, Min, Jeong, Lee, Kang, Song, Kang, Lee, Kim and Kim, ‘Structural evolution of tunneling oxide passivating contact upon thermal annealing’, Sci. Rep. 7, 12853 (2017); Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013); Feldmann, Fellmeth, Steinhauser, Nagel, Ourinson, Mack, Lohmuller, Polzin, Benick, Richter, Moldovan, Bivour, Clement, Rentsch, Hermle and Glunz, ‘Large Area TOPCon Cells Realized by a PECVD Tube Process’, 36th EU PVSEC, Marseille (2019).

Ques: Which champion cell did Feldmann et al. report in 2013: 23.7 % or 23.9 %?

The abstract gives 23.7 %, a fill factor of 82.2 % and 703 mV, and Table 2 lists the best cell with 703.3 mV, 41.0 $mA/c m^{2}$ , 82.2 % and 23.72 %. The text, however, gives the champion cell as a fill factor of 82.6 % and 23.9 %. The product of the table values is 23.70 %, whereas 82.6 % would give 23.82 % with the same voltage and current, not 23.9 % (computed), so the text does not agree with either. The abstract and the table agree with each other and with the arithmetic, so 23.7 % and 82.2 % are more likely. The text may refer to another cell or to an earlier measurement; the paper does not say. Reference: Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013).

Ques: How thick should the poly-silicon layer be?

The sources use 20 nm (Feldmann 2013), 80 nm (the baseline of Mack et al.), 100 and 170 nm (Feldmann 2019), 160 nm for stability and 60 to 125 nm for plated contacts; Lesso Solar names 100 to 150 nm. The thickness balances two effects. Thicker layers protect the oxide against the paste: $J_{0, m e t}$ falls by 65 % from 100 to 170 nm, and Lee et al. find 30 to 40 $fA/c m^{2}$ for 150 nm against 103 $fA/c m^{2}$ for 50 nm after firing at 880 °C. Thinner layers absorb less light and save deposition time. For fired screen-printed contacts 80 to 170 nm is more likely; thinner layers need laser-assisted or plated contacts, where UV picosecond laser openings on 60 nm poly-Si lowered the implied voltage by only 1 mV. Reference: Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013); Mack, Kafle, Tessmann, Krieg, Bashardoust, Lohmuller, Belledin, Saint-Cast, Hoffler, Ourinson, Fellmeth, Steinhauser, Kluska and Wolf, ‘Status and Perspective of Industrial TOPCon Solar Cell Development at Fraunhofer ISE’, WCPEC-8, Milan (2022); Feldmann, Fellmeth, Steinhauser, Nagel, Ourinson, Mack, Lohmuller, Polzin, Benick, Richter, Moldovan, Bivour, Clement, Rentsch, Hermle and Glunz, ‘Large Area TOPCon Cells Realized by a PECVD Tube Process’, 36th EU PVSEC, Marseille (2019); Lee, Yang, Lee, Lee, Song, Kang, Yoon and Park, ‘Effect of poly-Si Thickness and Firing Temperature on Metal Induced Recombination and Contact Resistivity of TOPCon Solar Cells’, Current Photovoltaic Research 9(4), 128-132 (2021); Lesso Solar, ‘Detailed explanation of core processes in TOPCon solar cell technology’.

Ques: Is the metal-induced recombination 530 or 30 to 40 fA/cm2?

Feldmann et al. give $J_{0, m e t, r e a r}$ of 530 ± 140 $fA/c m^{2}$ for 170 nm of poly-Si and about three times more for 100 nm, whereas Lee et al. give $J_{o m}$ of 30 to 40 $fA/c m^{2}$ for 150 nm and 103 $fA/c m^{2}$ for 50 nm at 880 °C. The two quantities are defined differently: the first is per metallized area, the second is the increase of the saturation current of the whole test sample, so the numbers cannot be compared. If the same test structure had a metal fraction near 6.6 %, the values would fit (35/530, computed), but the fraction is not given in the extract. Both papers agree on the trend: the thin layer is worse by a factor of 2.9 (Lee) and 3 (Feldmann), and I cannot say which absolute value is more likely. Reference: Feldmann, Fellmeth, Steinhauser, Nagel, Ourinson, Mack, Lohmuller, Polzin, Benick, Richter, Moldovan, Bivour, Clement, Rentsch, Hermle and Glunz, ‘Large Area TOPCon Cells Realized by a PECVD Tube Process’, 36th EU PVSEC, Marseille (2019); Lee, Yang, Lee, Lee, Song, Kang, Yoon and Park, ‘Effect of poly-Si Thickness and Firing Temperature on Metal Induced Recombination and Contact Resistivity of TOPCon Solar Cells’, Current Photovoltaic Research 9(4), 128-132 (2021).

Ques: What contact resistivity do the sources give: 1, 3, 10 or 19 mΩ cm2?

Feldmann et al. found about 10 $mΩ c m^{2}$ in 2013, 3 $mΩ c m^{2}$ at the optimum firing in 2019 and about 1 $mΩ c m^{2}$ for doping above $1 0^{20}$ $c m^{- 3}$ . Mack et al. cite a value below 1 $mΩ c m^{2}$ , which allows laser openings of 5 µm. Lee et al. measure 17 to 19 $mΩ c m^{2}$ at low firing temperatures and 1.9 $mΩ c m^{2}$ for 50 nm poly-Si at high ones, and van der Vossen quotes 20 $mΩ c m^{2}$ for chemical oxides of NREL. The values depend on doping, firing and paste, not on a disagreement. For current industrial cells 1 to 3 $mΩ c m^{2}$ is more likely. The laser-enhanced contact step lowers the average series resistance from 2.81 to 0.46 $Ω c m^{2}$ , which is 3.5 points of efficiency (computed). Reference: Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013); Feldmann, Fellmeth, Steinhauser, Nagel, Ourinson, Mack, Lohmuller, Polzin, Benick, Richter, Moldovan, Bivour, Clement, Rentsch, Hermle and Glunz, ‘Large Area TOPCon Cells Realized by a PECVD Tube Process’, 36th EU PVSEC, Marseille (2019); Mack, Kafle, Tessmann, Krieg, Bashardoust, Lohmuller, Belledin, Saint-Cast, Hoffler, Ourinson, Fellmeth, Steinhauser, Kluska and Wolf, ‘Status and Perspective of Industrial TOPCon Solar Cell Development at Fraunhofer ISE’, WCPEC-8, Milan (2022); Lee, Yang, Lee, Lee, Song, Kang, Yoon and Park, ‘Effect of poly-Si Thickness and Firing Temperature on Metal Induced Recombination and Contact Resistivity of TOPCon Solar Cells’, Current Photovoltaic Research 9(4), 128-132 (2021); van der Vossen, ‘Optimization of passivating contacts for high-efficiency p-type silicon solar cells’, MSc thesis, Delft University of Technology (2017).

Ques: Is the ultraviolet degradation of TOPCon a hydrogen effect or a charge effect, and how large is it?

The UNSW study (Hoex) exposed commercial bifacial cells to 61.1 $kWh/ m^{2}$ of UV-B and 49 $kWh/ m^{2}$ of UV-A at 60 °C and found a loss of open-circuit voltage; it attributes it to broken Si-H bonds at the aluminium oxide on the boron emitter by photons above 3.4 eV, and the article gives no loss in percent. The NREL study reports 2.3 to 3.2 % non-recoverable loss after a one-year-equivalent dose, and, separately, a slow dark loss that recovers within seconds to minutes under UV light, which they link to charge trapping in the oxide. The effects need not exclude each other: hydrogen redistribution is permanent, the charge effect reversible. The articles do not say whether the 3.2 % is efficiency or power. Reference: pv magazine, ‘New research warns of unexpected UV-induced degradation in TOPCon solar cells from invisible light’ (11 August 2025); pv magazine, ‘NREL researchers quantify UV-induced degradation levels in TOPCon solar cells’ (3 February 2026).

Ques: Does the tunnel oxide conduct by tunnelling or through pinholes?

Both pictures are in the sources. Choi et al. analyse direct tunnelling, and the oxide thins from 1.5 to 1.0 nm at 1000 °C, where selectivity is lost because holes can reach the doped layer; their abstract speaks of pinholes at the highest temperature. Gan and Swanson saw the contact resistance fall sharply when the oxide broke up, and Peibst et al. proposed local transport through pinholes (both quoted by van der Vossen). The thesis fitted the tunnelling and the pinhole model to its data, but all points lay in the saturated range, so it could not tell them apart. Feldmann et al. find that the passivation drops strongly above 900 °C. For 1.2 to 1.5 nm oxides annealed up to about 900 °C, tunnelling is more likely; pinholes are more likely at 1000 °C. Reference: Choi, Min, Jeong, Lee, Kang, Song, Kang, Lee, Kim and Kim, ‘Structural evolution of tunneling oxide passivating contact upon thermal annealing’, Sci. Rep. 7, 12853 (2017); van der Vossen, ‘Optimization of passivating contacts for high-efficiency p-type silicon solar cells’, MSc thesis, Delft University of Technology (2017); Feldmann, Bivour, Reichel, Hermle and Glunz, ‘A Passivated Rear Contact for High-Efficiency n-Type Silicon Solar Cells Enabling High Voc and FF>82%’, 28th EU PVSEC, Paris (2013).

Quartz 5

Explorer

TOPCon Silicon Cell