Fresnel Number

The Fresnel number is the one dimensionless group that tells you what kind of diffraction problem you have before you write a single integral. Take an aperture of half-width $a$ , light of wavelength $\lambda$ , and a screen a distance $z$ away, and form $N_F=a^2/\lambda z$ . That single number decides everything about the pattern's character: a large $N_F$ is a sharp-edged geometric shadow with a fringe of ripples, a value near one is the richly structured Fresnel regime that needs the full diffraction integral, and a small $N_F$ is the smooth far-field Fraunhofer pattern — the sinc of a slit or the Airy disk of a hole. Change any of $a$ , $\lambda$ , or $z$ and you slide continuously along that scale. The Fresnel number is the knob. Its name carries its meaning. Divide the aperture into Fresnel's half-period zones — annular rings over each of which the path to the axial point grows by another half-wavelength — and the number that fit inside the aperture is exactly $N_F$ . Successive zones contribute to the axial field with alternating sign, so the count's parity governs whether the axis is bright or dark: this is the origin of the on-axis intensity $I=4I_0\sin^2(\pi N_F/2)$ , and, at the extreme of a blocked centre, of the Poisson–Arago spot. The regime transition is nothing more than comparing this zone count to one. The same number reaches beyond a hole in a screen. In a laser resonator with mirrors of radius $a$ separated by $L$ , the cavity Fresnel number $N_F=a^2/\lambda L$ estimates how many transverse modes survive with low diffraction loss and sets the per-pass loss of the fundamental mode. A large- $N_F$ cavity is multimode; a cavity near $N_F\sim1$ is effectively single-mode. From a slit on a bench to a gigawatt laser, the same ratio does the bookkeeping.

Key takeaways

The Fresnel number $N_F=a^2/\lambda z$ compares the aperture half-width $a$ to the natural diffraction scale $\sqrt{\lambda z}$ ; squaring that ratio gives $N_F$ .
It literally counts the half-period Fresnel zones the aperture subtends from the axial observation point — the zone count is $N_F$ .
$N_F\gg1$ is the geometric shadow, $N_F\sim1$ is the structured Fresnel regime, $N_F\ll1$ is the far-field Fraunhofer pattern — the regime is set by comparing $N_F$ to one.
On the axis behind a circular hole the intensity is $I=4I_0\sin^2(\pi N_F/2)$ : bright ( $4I_0$ ) at odd $N_F$ , dark at even $N_F$ .
In a laser cavity the Fresnel number $a^2/\lambda L$ governs diffraction loss per pass and the number of low-loss transverse modes.
Key idea: $N_F=a^2/\lambda z$ is the number of Fresnel zones an aperture subtends from the observation point and the single dial of diffraction — $N_F\gg1$ geometric shadow, $N_F\sim1$ Fresnel, $N_F\ll1$ Fraunhofer — because $\pi N_F$ is the maximum wavefront-curvature phase across the aperture.

The core insight

The Fresnel number $N_F=a^2/\lambda z$ is the aperture radius measured in units of the diffraction length $\sqrt{\lambda z}$ , squared — equivalently, the number of half-period Fresnel zones the aperture subtends from the axial point. It sets the regime because $\pi N_F$ is the largest wavefront-curvature phase the aperture imposes: $\pi N_F\ll1$ ( $N_F\ll1$ ) is Fraunhofer, $N_F\sim1$ is Fresnel, $N_F\gg1$ is the geometric shadow. Its parity also fixes the on-axis intensity, $I=4I_0\sin^2(\pi N_F/2)$ .

Where this lives today

The Fresnel number is the first thing a laser designer computes. The cavity value $a^2/\lambda L$ predicts how many transverse modes a resonator holds and how much the fundamental mode loses to diffraction each pass — the numbers that decide clean single-mode versus multimode output. High-power industrial lasers deliberately run unstable resonators with large equivalent Fresnel numbers to fill a big-diameter gain medium, then use the diffraction at the mirror edge to couple the beam out; the Fox–Li mode calculation that first named the parameter is still the workhorse for designing them. Away from lasers, the same number tells an experimentalist which formula to use: compute $N_F$ and you know whether the sinc/Airy far-field expressions apply or whether the full Fresnel integral is needed. It sets the focal length and resolution of X-ray and EUV Fresnel zone plates, where the zone count out to the plate edge is the plate's Fresnel number, and it underlies the near/far-field crossover of any laser beam: a Gaussian beam's Rayleigh range $z_R=\pi w_0^2/\lambda$ is precisely where its Fresnel number falls to order one.

The physical picture

Stand at the axial observation point and look back at the aperture. Draw rings so that each adds half a wavelength to the path back to your eye — Fresnel's half-period zones. The number that fit inside the aperture is the Fresnel number. If only a fraction of a ring fits ( $N_F\ll1$ ), the whole aperture is essentially in phase and you see a smooth far-field pattern; if many fit ( $N_F\gg1$ ), their alternating contributions interfere in an intricate, distance-sensitive way — the near field.

Figure

Counting Fresnel zones: the aperture divided into half-period zones of radius r_m = √(mλz); the number of zones inside the aperture is N_F, so a nearby point sees many zones and a distant point sees a fraction of one

The length $\sqrt{\lambda z}$ is the transverse distance over which a spherical wavefront pulls a quarter-wavelength from a plane — the natural width of the first zone. Comparing the aperture radius to it, and squaring, is all $N_F$ is. Move the screen closer and $\sqrt{\lambda z}$ shrinks, so more zones crowd into the fixed aperture; move it away and eventually less than one zone fits. The next figure lays the whole regime landscape on a single logarithmic axis, with a real aperture placed on it.

Figure

The regime map on a log N_F axis: N_F ≪ 1 Fraunhofer (single lobe), N_F ~ 1 Fresnel (structured), N_F ≫ 1 geometric shadow — with a = 1 mm, λ = 500 nm marked at z = 8 m (N_F = 0.25), z = 2 m (N_F = 1), z = 1 cm (N_F = 200)

One aperture, three distances, three regimes. The same $1$ mm hole in $500$ nm light is a deep geometric shadow at a centimetre ( $N_F=200$ ), sits at the Fresnel/far-field boundary at two metres ( $N_F=1$ ), and has slid into Fraunhofer by eight metres ( $N_F=0.25$ ). Nothing about the hole changed — only the number of zones the observation point sees.

The core equations

The Fresnel number compares two lengths: the aperture half-width $a$ and the transverse diffraction length $\sqrt{\lambda z}$ , over which the spherical-wave phase accumulates about a radian. It is the square of their ratio — which is why it is so sensitive to aperture size and so forgiving of wavelength.

N_F=\frac{a^2}{\lambda z}=\left(\frac{a}{\sqrt{\lambda z}}\right)^2 \qquad (1)

where: $a$ — aperture radius or transverse half-width; $\lambda$ — wavelength; $z$ — aperture-to-screen distance; $N_F$ — Fresnel number (dimensionless). Because $a$ enters squared, doubling the aperture quadruples $N_F$ ; because $\lambda$ and $z$ enter linearly, they trade off directly. The name is literal: $N_F$ counts the half-period zones the aperture subtends. Each zone is an annulus whose outer edge is one more half-wavelength of path from the axial point than the last, so its radius grows as the square root of the zone index.

r_m=\sqrt{m\,\lambda z}\quad\Longrightarrow\quad N_{\text{zones}}=\frac{a^2}{\lambda z}=N_F \qquad (2)

where: $r_m$ — outer radius of the $m$ -th half-period zone; $m$ — zone index; $N_{\text{zones}}$ — number of zones inside radius $a$ , found by setting $r_m=a$ . The zone count is the Fresnel number — the geometric content of the name. The regime is decided by one phase. Expanding the Fresnel integral, the only term separating near from far field is the aperture's own quadratic phase, whose maximum across the opening is exactly $\pi N_F$ . When that phase is small the aperture looks flat and diffraction becomes a Fourier transform.

\left|\frac{k}{2z}(\xi^2+\eta^2)\right|_{\max}=\frac{k a^2}{2z}=\pi N_F\;\ll\;1\;\;\Rightarrow\;\;\text{Fraunhofer} \qquad (3)

where: $k=2\pi/\lambda$ — wavenumber; $(\xi,\eta)$ — aperture-plane coordinates; $\pi N_F$ — the maximum wavefront-curvature phase across the aperture, in radians. $\pi N_F\ll1$ is Fraunhofer; $\pi N_F\gtrsim1$ ( $N_F\gtrsim1$ ) is Fresnel; $N_F\gg1$ is the geometric shadow. Evaluating the Fresnel integral on the axis behind a uniformly lit circular aperture turns the zone count into an intensity: the exposed zones sum as alternating-sign phasors, giving an oscillation whose parity is fixed by whether $N_F$ is an odd or even integer.

I_{\text{axis}}(z)=4\,I_0\sin^2\!\left(\frac{\pi N_F}{2}\right) \qquad (4)

where: $I_0$ — unobstructed incident intensity; $I_{\text{axis}}$ — on-axis intensity at distance $z$ . Odd $N_F$ (a net single open zone) gives the maximum $4I_0$ ; even $N_F$ (zones cancelling in pairs) gives darkness. The same ratio, applied to a laser resonator, becomes the cavity Fresnel number — the standard estimate of how many transverse modes a cavity holds with low loss.

N_F^{\text{cav}}=\frac{a^2}{\lambda L} \qquad (5)

where: $a$ — mirror aperture radius; $L$ — mirror separation. A large $N_F^{\text{cav}}$ supports many transverse modes; $N_F^{\text{cav}}\sim1$ enforces near-single-transverse-mode operation at the cost of higher fundamental-mode diffraction loss.

Derivation AThe Fresnel number is the zone count

The aim is to justify the name — to show that $a^2/\lambda z$ is not just a ratio of lengths but literally the number of half-period Fresnel zones the aperture subtends from the axial point. We build the zones from the half-wavelength path condition, find their radii, count how many fit inside the aperture, and expose $\sqrt{\lambda z}$ as the natural transverse scale.

Step 1. Define the $m$ -th zone edge by the path condition $\sqrt{r_m^2+z^2}-z=m\lambda/2$ : the distance from that ring to the axial point $P$ exceeds the on-axis distance by $m$ half-wavelengths. Why: half-period zones are the natural unit because neighbouring zones differ by $\lambda/2$ in path, so their axial contributions are opposite in phase — the whole engine of Fresnel diffraction.

Step 2. Expand paraxially for $r_m\ll z$ : $\sqrt{r_m^2+z^2}\approx z+r_m^2/2z$ , so the condition becomes $r_m^2/2z=m\lambda/2$ , giving $r_m=\sqrt{m\lambda z}$ . Why: the same small-angle expansion behind the Fresnel integral turns the exact path condition into a clean square-root law — the zone radii grow as $\sqrt{m}$ , so the zones have equal area $\pi\lambda z$ .

Step 3. Count the zones inside the aperture by setting $r_m=a$ : $m=a^2/\lambda z=N_F$ . Equivalently, since $r_m/a=\sqrt{m/N_F}$ , the aperture edge lies at zone number $N_F$ . Why: this is the payoff — the zone count and the Fresnel number are the same object, so "how many zones does the aperture show?" and "what is $N_F$ ?" are one question.

Step 4. Read the natural scale: the first zone has radius $r_1=\sqrt{\lambda z}$ , so $N_F=(a/r_1)^2=(a/\sqrt{\lambda z})^2$ . The aperture is being measured in units of the first-zone width. Why: this recasts $N_F$ as a squared length ratio, the form that makes its scalings ( $\propto a^2$ , $\propto1/z$ ) obvious at a glance. *Verified: $r_m=\sqrt{m\lambda z}$ from the paraxial path condition and $N_F=a^2/\lambda z$ at $r_m=a$ , with $r_1=\sqrt{\lambda z}$ , in verify_fresnel-number.py.*

Go deeper · Derivation B · 1 minRegime boundary from the aperture-curvature phase

Derivation BRegime boundary from the aperture-curvature phase

This is an owned structural result: a sharp, quantitative criterion for the near-to-far-field transition rather than a vague "far enough away." The aim is to start from the paraxial (Fresnel) propagator, isolate the single term carrying the aperture's own wavefront curvature, and show its largest value across the opening is precisely $\pi N_F$ — so the whole regime map hangs on comparing $\pi N_F$ to one.

Step 1. Write the Fresnel propagation of an aperture field, $U(x,y,z)=\frac{e^{ikz}}{i\lambda z}\iint U_0(\xi,\eta)\,e^{\frac{ik}{2z}[(x-\xi)^2+(y-\eta)^2]}\mathrm{d}\xi\,\mathrm{d}\eta$ . Why: this is the near-field integral developed in Fresnel Diffraction Regime (recalled by title), and every regime statement is a statement about its exponent.

Step 2. Expand the bracket into three groups: $(x^2+y^2)-2(x\xi+y\eta)+(\xi^2+\eta^2)$ — a pure observation term, a cross (Fourier) term, and the pure aperture term $\xi^2+\eta^2$ . Why: only the last group involves the aperture coordinates quadratically, and it is the sole term distinguishing Fresnel from Fraunhofer — the other two already form a Fourier kernel times an outer phase.

Step 3. Bound the aperture phase over the opening, where $\xi^2+\eta^2=a^2$ : its maximum is $\frac{ka^2}{2z}=\frac{\pi a^2}{\lambda z}=\pi N_F$ . Why: this collapses a messy phase into one number — the maximum wavefront curvature the aperture imposes, measured in units of $\pi$ — which is exactly the Fresnel number.

Step 4. Split the cases. If $\pi N_F\ll1$ the aperture phase $\approx1$ and drops out, leaving a pure Fourier transform — Fraunhofer (handed to Fraunhofer Diffraction Regime). If $N_F\gtrsim1$ it must be kept, and the zone structure and axial oscillations of the near field appear; $N_F\gg1$ recovers the geometric shadow. Why: the transition is not a fuzzy distance but a clean inequality on one number, which is what makes $N_F$ the diffraction dial. *Verified: the identity $\frac{ka^2}{2z}=\pi N_F$ symbolically and $\pi N_F=0.785$ at $N_F=0.25$ in verify_fresnel-number.py.*

Go deeper · Derivation C · 1 minOn-axis intensity and the meaning of integer N_F

Derivation COn-axis intensity and the meaning of integer N_F

The oscillating on-axis intensity — bright at odd $N_F$ , dark at even — is this article's owned result and the sharpest demonstration that $N_F$ counts zones. The aim is to evaluate the Fresnel integral on the axis of a uniformly lit disc of radius $a$ , reduce the radial integral to a single complex exponential, and extract the $4I_0$ ceiling and the parity rule that makes integer $N_F$ special.

Step 1. On the axis there is no linear phase, so only the quadratic term survives: $U_{\text{axis}}\propto\int_0^{a}e^{ik\rho^2/2z}\,2\pi\rho\,\mathrm{d}\rho$ , with $\rho^2=\xi^2+\eta^2$ . Why: setting $x=y=0$ removes the Fourier kernel, leaving a purely radial integral — the axis is where the 2-D integral becomes one-dimensional and exactly solvable.

Step 2. Substitute $u=\rho^2$ : $U_{\text{axis}}\propto\pi\int_0^{a^2}e^{iku/2z}\mathrm{d}u=\pi\cdot\frac{2z}{ik}\big(e^{ika^2/2z}-1\big)$ . Why: the substitution turns the radial integral into a trivial exponential one, the reason the axial case is analytically clean while off-axis needs the Cornu spiral.

Step 3. Insert $ka^2/2z=\pi N_F$ and factor $e^{i\pi N_F}-1=2i\,e^{i\pi N_F/2}\sin(\pi N_F/2)$ , so $|U_{\text{axis}}|\propto2\,|\sin(\pi N_F/2)|$ . Normalising to the unobstructed intensity $I_0$ (the full open wave has amplitude equal to half a single zone) gives $I_{\text{axis}}=4I_0\sin^2(\pi N_F/2)$ . Why: that trigonometric identity produces the $\sin^2$ oscillation, and the two-to-one ratio between one open zone and the full wave lifts the ceiling to $4I_0$ .

Step 4. Read the integers: odd $N_F$ gives $\sin^2(\pi N_F/2)=1$ and a bright $4I_0$ axis (net one open zone); even $N_F$ gives $0$ and a dark axis (zones cancel in pairs). At $N_F=1$ , one open zone quadruples the on-axis intensity — the gain a zone plate exploits. Why: each unit of $N_F$ flips the newest zone's phasor, so the parity of the zone count decides constructive versus destructive interference on the axis. *Verified: $4\sin^2(\pi N_F/2)=4$ at $N_F=1$ , $=0$ at $N_F=2$ , and the symbolic identity $|e^{i\pi N_F}-1|^2=4\sin^2(\pi N_F/2)$ in verify_fresnel-number.py.*

Limiting cases

Case	Result	Meaning
Geometric limit	$N_F\gg1$	aperture spans many zones; the pattern is its geometric shadow with edge fringes — ray optics is the leading behaviour
Fresnel regime	$N_F\sim1$	a few zones exposed; the full quadratic-phase integral (or Cornu spiral) is required — strong axial oscillations and asymmetric edge diffraction
Fraunhofer limit	$N_F\ll1\;(z\gg a^2/\lambda)$	the aperture curvature phase $\to1$ ; the pattern is the Fourier transform of the aperture (sinc, Airy) and only rescales with $z$
Single-zone boundary	$N_F=1$	exactly one zone fills the aperture — the conventional near/far demarcation and the on-axis maximum $4I_0$ for a plane-wave-lit circular hole
Cavity, $N_F^{\text{cav}}\gg1$	many transverse modes	large mirrors or short cavity — higher-order modes have low loss and can lase
Cavity, $N_F^{\text{cav}}\sim1$	near single-mode	small aperture or long cavity suppresses higher modes, at higher fundamental-mode diffraction loss

Common trapThe Fresnel number is a fixed property of an aperture

It is not — $N_F$ depends on the observation distance $z$ as much as on the aperture. The same slit is deep in the Fresnel regime a few millimetres behind it and firmly Fraunhofer several metres away; quoting "the Fresnel number of a slit" without a distance is meaningless. Two further traps hide in the definition. First, the factor convention: some texts build $N_F$ from the radius $a$ and others from the full width or diameter $D=2a$ , and since $N_F\propto a^2$ these differ by four ( $D^2/\lambda z=4a^2/\lambda z$ ) — always state which you mean. Second, "the transition happens at $N_F=1$ " is a rule of thumb, not a hard edge: the crossover is gradual, spanning roughly a decade around unity, so $2$ or $0.5$ is genuinely in between. Treat $N_F$ as a continuous dial, always paired with a specific $z$ and a stated aperture convention.

Worked examples

1 · One aperture across three regimes (confidence builder)

A circular aperture of radius $a=1$ mm is illuminated by $\lambda=500$ nm light. Find the Fresnel number at $z=1$ cm, $z=2$ m, and $z=8$ m, and name the regime and on-axis intensity in each case.

Step 1. At $z=1$ cm $=0.01$ m: $N_F=a^2/\lambda z=(10^{-3})^2/(5\times10^{-7}\times10^{-2})=10^{-6}/(5\times10^{-9})=\mathbf{200}$ . With $200$ zones across the aperture this is deep in the geometric-shadow regime — the pattern is essentially the projected hole with a decorative edge fringe.

Step 2. At $z=2$ m: $N_F=10^{-6}/(5\times10^{-7}\times2)=10^{-6}/10^{-6}=\mathbf{1}$ . Exactly one zone fills the aperture — the Fresnel/far-field boundary — and the on-axis intensity is at its maximum $I=4I_0\sin^2(\pi/2)=\mathbf{4I_0}$ .

Step 3. At $z=8$ m: $N_F=10^{-6}/(5\times10^{-7}\times8)=10^{-6}/(4\times10^{-6})=\mathbf{0.25}$ . Less than one zone is exposed — the Fraunhofer onset, with maximum aperture phase $\pi N_F=0.785$ rad — and the on-axis intensity is $I/I_0=4\sin^2(0.3927)=\mathbf{0.586}$ .

Step 4. The same hole is a shadow at a centimetre, a single bright zone at two metres, and a nascent Airy pattern at eight — the regime is a property of the geometry, read off one number. $N_F=200,\,1,\,0.25$ at $z=1$ cm, $2$ m, $8$ m.

2 · The Fresnel number of a laser cavity (exam level)

A laser resonator has mirrors of radius $a=2$ mm separated by $L=0.5$ m, operating at $\lambda=1064$ nm (Nd:YAG). Find the cavity Fresnel number and say what it implies for the transverse modes.

Step 1. Apply the cavity form $N_F^{\text{cav}}=a^2/\lambda L=(2\times10^{-3})^2/(1064\times10^{-9}\times0.5)=(4\times10^{-6})/(5.32\times10^{-7})=\mathbf{7.5}$ . The wave making a round trip sees about seven and a half Fresnel zones of the mirror aperture.

Step 2. Interpret the count. The cavity Fresnel number estimates the number of transverse modes with tolerably low diffraction loss, so $N_F^{\text{cav}}\approx7.5$ means the resonator can support several higher-order transverse modes, not just the fundamental TEM $_{00}$ .

Step 3. Read off the design lever. To force clean single-transverse-mode output you would reduce $N_F^{\text{cav}}$ toward unity — insert an intracavity aperture to shrink $a$ , or lengthen $L$ — which raises the fundamental mode's per-pass diffraction loss but starves the higher-order modes of gain.

Step 4. The picture matches the free-space one: a large- $N_F$ mirror is like a wide aperture close to the screen (many zones, multimode), while $N_F^{\text{cav}}\sim1$ is the diffraction-dominated single-lobe limit. $N_F^{\text{cav}}=7.5$ — several transverse modes supported.

Physics of the portrait

The portrait is a number, not a person: a single dimensionless ratio that has quietly organized diffraction theory, resonator design, and imaging for two centuries.

Origins · 1 minfresnel number and how this was discovered

fresnel number

the dimensionless ratio a²/λz that counts an aperture's Fresnel zones and dials diffraction from shadow to far field · scalar wave optics

The Fresnel number is the dimensionless group $N_F=a^2/\lambda z$ that measures an aperture against the diffraction length $\sqrt{\lambda z}$ and, equivalently, counts the half-period zones the aperture subtends. Its root is Augustin-Jean Fresnel's 1818 division of the wavefront into those zones, which made the alternating-phasor sum — and the whole near field — computable. As a named parameter it crystallized much later, in laser-resonator physics: when A.G. Fox and Tingye Li computed the transverse modes of an open cavity numerically in 1961, the ratio $a^2/\lambda L$ emerged as the single number controlling diffraction loss and mode count, and Kogelnik and Li's 1966 resonator review made it standard vocabulary. Today it spans free-space diffraction, cavity design, X-ray zone plates, and the Gaussian-beam crossover — all read off the same ratio. It is at once a geometry (a zone count), a phase bound ( $\pi N_F$ of wavefront curvature), and an engineering figure of merit (modes per cavity).

History

1818

— Fresnel's prize memoir divides the wavefront into half-period zones, making the zone count the natural measure of near-field diffraction

1818

— Poisson deduces, and Arago observes, the bright axial spot behind a disc — the extreme large-zone-count case

1961

— Fox and Li compute open-resonator modes numerically; the cavity ratio a²/λL emerges as the key diffraction-loss parameter

1966

— Kogelnik and Li's resonator review standardizes the Fresnel number in laser physics

Going further

Texts. Hecht, Optics, 5th ed., §10.2–10.3 (Fresnel and Fraunhofer diffraction) · Saleh & Teich, Fundamentals of Photonics, 3rd ed., ch. 4 (Fourier optics), ch. 11 (resonators) · Born & Wolf, Principles of Optics, 7th ed., §8.2 (Fresnel number and diffraction regimes) · Siegman, Lasers, ch. 19 (Fresnel number and resonator diffraction loss) · Foot, Atomic Physics, §7 (diffraction limits in imaging).

Provenance. Definition $N_F=a^2/\lambda z$ , zone radii $r_m=\sqrt{m\lambda z}$ , and the Fraunhofer condition $N_F\ll1$ are standard (Hecht §10.2-10.3; Born & Wolf §8.2). A verifier should re-derive the on-axis result $I_{\text{axis}}=4I_0\sin^2(\pi N_F/2)$ by direct integration of the Fresnel integral over a disk (Hecht §10.3) and confirm the $4I_0$ single-zone gain. Cavity Fresnel number $a^2/\lambda L$ and its mode-loss role follow Siegman, Lasers Ch.19 and Saleh & Teich Ch.11. The Gaussian-beam identification $z_R=\pi w_0^2/\lambda$ at $N_F\sim1$ is standard Fourier-optics (Saleh & Teich Ch.3-4). Numerical examples ( $z=8$ m, $N_F\approx1.6\times10^{-2}$ ) are elementary substitutions verifiable by hand; all values here are independently recomputed in verify_fresnel-number.py.

Prerequisites. Huygens–Fresnel Principle · Fresnel–Kirchhoff Diffraction Integral · Fresnel Zones and Zone Plates. Leads to. Fresnel Diffraction Regime · Fraunhofer Diffraction Regime · Cornu Spiral and Fresnel Integrals · Rayleigh Resolution Criterion.

Is the Fresnel number a property of the aperture or of the whole setup?

Of the whole setup. It bundles aperture size $a$ , wavelength $\lambda$ , and observation distance $z$ into one ratio, and changes the moment any of them changes. The same aperture has a huge Fresnel number close up and a tiny one far away, so quoting "the Fresnel number of a slit" without a distance is meaningless. Think of it not as a label on the aperture but as a coordinate on the diffraction dial — the position telling you which regime the specific geometry is in.

Why does the on-axis intensity reach 4I₀ — more than the unobstructed beam?

Because a single open Fresnel zone contributes an amplitude equal to twice that of the fully open wavefront. When all zones are open their alternating-sign contributions largely cancel, leaving a net axial amplitude only about half the first zone's — the unobstructed field $I_0$ . Expose exactly one zone ( $N_F=1$ ) and there is nothing to cancel it, so the amplitude doubles and the intensity, going as amplitude squared, quadruples to $4I_0$ . This is a redistribution, not extra energy: the axis brightens at the expense of the off-axis field — the gain a zone plate harnesses by blocking the cancelling even zones.

How is the Fresnel number related to the maximum phase across the aperture?

They differ only by a factor of $\pi$ . Expanding the Fresnel integral, the aperture's own quadratic phase $\frac{k}{2z}(\xi^2+\eta^2)$ reaches its largest value at the rim, where $\xi^2+\eta^2=a^2$ , giving $\frac{ka^2}{2z}=\pi N_F$ . So $\pi N_F$ is literally the maximum wavefront-curvature phase, in radians, that the aperture imposes on the far field. That is why the regimes fall where they do: $\pi N_F\ll1$ means the aperture is phase-flat and diffraction is Fraunhofer, while $\pi N_F\gtrsim1$ means the curvature matters and you are in the Fresnel near field.

What sets whether a value like N_F = 0.5 or 2 is "near" or "far" field?

There is no razor edge; the crossover is gradual and spans roughly a decade in $N_F$ around unity. The convention $N_F=1$ (one zone filling the aperture) is a useful midpoint, but $N_F=2$ already shows clear near-field structure while $N_F=0.5$ is well on the way to a clean far-field pattern. In practice you pick a tolerance: if you can accept the small error from dropping the aperture phase $\pi N_F$ , use the Fraunhofer formula; if not, keep the full Fresnel integral. For careful work people often demand $N_F\lesssim0.1$ before trusting the far-field expressions outright.

Why does the cavity Fresnel number use the mirror separation L instead of a screen distance?

Because a resonator is a folded diffraction problem. Light bouncing between mirrors of radius $a$ separated by $L$ diffracts over the one-way distance $L$ each pass, so the relevant zone count is $a^2/\lambda L$ — the mirror playing both aperture and screen. A mode that reproduces itself round-trip must survive this diffraction, and $N_F$ estimates how many such transverse patterns exist with acceptable edge loss. A big- $N_F$ cavity diffracts little and holds many modes; an $N_F\sim1$ cavity loses a large fraction of the fundamental mode off the mirror edge each pass, which both cleans the mode and lowers efficiency.

How does the Fresnel number connect to a Gaussian beam's Rayleigh range?

They are the same near/far crossover in different clothing. For a Gaussian beam of waist $w_0$ , put $a\to w_0$ and ask where $N_F=w_0^2/\lambda z$ falls to order one: that distance is $z\sim w_0^2/\lambda$ , which up to the factor $\pi$ is the Rayleigh range $z_R=\pi w_0^2/\lambda$ . Within a Rayleigh range the beam is quasi-collimated (the near field, $N_F\gtrsim1$ ); beyond it, it diverges with angle $\theta\approx\lambda/\pi w_0$ . The standard laser-beam near/far split is just the Fresnel number passing through unity, phrased for a Gaussian aperture.