The 'One Airy Unit' Rule is a Guideline, Not a God: When to Open Your Confocal Pinhole for Better Data

It's 2:00 AM in the microscope facility. You're sitting in front of a $500,000 piece of equipment—perhaps a beautiful Olympus confocal system—and staring at a screen full of... noise. This comprehensive guide will teach you when and how to break the "1 Airy Unit" rule to obtain beautiful, quantifiable data from even your most difficult samples.

The Problem: The 2 AM Vicious Cycle

You have a precious, antibody-stained sample. Maybe it's a low-expression endogenous protein, a thick tissue slice with poor antibody penetration, or a live-cell line with a dim fluorescent tag to avoid toxicity. You can just make out your structure of interest, but it's swimming in a sea of "salt-and-pepper" static.

What do you do?

You do what you were trained to do. You crank up the laser power. The image gets a little brighter, but your signal in the corner of the frame is now visibly dimmer than it was 30 seconds ago. You've just photobleached it.

Frustrated, you drop the laser back down and crank up the detector gain (the HV setting for the PMT or GaAsP). The image gets brighter, sure, but the noise gets proportionally brighter, too. Your signal-to-noise ratio (SNR) hasn't improved at all.

You check the software. The pinhole is set perfectly: 1.0 Airy Unit (AU). The "optimal" setting. The one the installation technician, your PI, and every textbook told you to use.

What if I told you that this "optimal" setting is the very thing strangling your signal? What if the solution isn't more laser power or more gain, but a simple, "heretical" click to widen that pinhole?

Section 1: The Church of 1 AU: Why We Worship the Pinhole

Before we can break the rules, we must understand them. The pinhole is the defining element of confocal imaging: it blocks out-of-focus light so the detector only receives photons from the focal plane. Without it, you just have a very expensive (and complicated) widefield fluorescence microscope.

A standard widefield microscope floods your sample with light and its camera collects all photons returning from the specimen—the sharp, in-focus light from the focal plane, and all the blurry, out-of-focus light from above and below it. This is why thick samples look hazy.

The confocal microscope, in brilliant contrast, is a "point-scanning, point-detection" system.

• Point Scanning: A laser is focused to a tiny spot on your sample.
• Point Detection: That spot (and only that spot) is imaged back onto a detector.
• The Pinhole: Right before the detector (a PMT or GaAsP), there is a physical pinhole aperture placed in a conjugate focal plane.

This is the magic. Light from the exact focal plane is "in-focus" when it reaches this conjugate plane, so it passes right through the pinhole and hits the detector. Light from above or below the focal plane is "out-of-focus" at this conjugate plane; it arrives as a blurry, wide cone. The edges of the pinhole physically block this out-of-focus light.

The result is a "confocal" (shared focus) system that generates a single, optically "clean" pixel. By scanning this spot across your sample, you build up an image, or "optical section," that is free from out-of-focus haze. This is what gives you that beautiful, crisp Z-resolution.

So, What is an "Airy Unit"?

But how big should that pinhole be?

The laser spot on your sample isn't an infinitely small point. Due to the diffraction of light, it's a blurry spot described by a point-spread function (PSF), which appears as a central bright spot (the Airy disk) surrounded by dim concentric rings.

The microscope's optics project this Airy disk back onto the pinhole. An Airy Unit (AU) is a relative unit of measurement, not an absolute one (like micrometers). One Airy Unit (1 AU) is defined as the diameter of the first minimum (the first dark ring) of the Airy disk as projected onto the pinhole plane.

Approximately 83–84% of the PSF's energy lies inside this first dark ring—inside the central Airy disk. Setting the pinhole to 1 AU therefore includes roughly that central lobe, creating a theoretical compromise:

• It's wide enough to let most of the in-focus photons from the central Airy disk pass through to the detector.
• It's small enough to reject most of the out-of-focus light, giving you excellent Z-resolution.

On modern confocal systems, the software is smart. When you select an objective and a laser line, it calculates the size of the Airy disk (which changes with wavelength and NA) and automatically sets the physical pinhole to 1 AU for you. It's the "optimal" button.

But "optimal" for what?

The Optical Transfer Function Perspective

One Airy Unit is defined with respect to the diameter of the central Airy disk projected onto the pinhole plane, and about 83–84% of the PSF energy lies inside that central lobe. Setting the aperture to 1 AU is therefore a good theoretical compromise between transmission and sectioning for many conditions.

However, the derivation assumes abundant photons—a luxury biological samples rarely provide. The optimization is for resolution under idealized conditions with infinite photon flux. It is very often not the optimal setting for signal-to-noise ratio in the real world.

When photon budgets are small, strict adherence to 1 AU often throws away a worthwhile fraction of in-focus photons and so can worsen shot-noise limited measurements.

Section 2: The Real-World Problem: When "Optimal" Resolution Is Useless

In biology, we are almost always "photon-starved." Our samples are dim. And in a dim, photon-starved world, our number one enemy isn't a slightly thick Z-slice. Our number one enemy is noise.

Specifically, we are limited by photon shot noise. This is a fundamental property of light. Photons don't arrive in a smooth, continuous stream; they arrive randomly, like raindrops on a roof. This randomness is noise.

Photon shot noise follows σ = √N. Thus the signal-to-noise ratio is:

If you collect 100 photons per pixel, your shot noise is √100 = 10 photons, giving you SNR ≈ 10. If you collect 10,000 photons per pixel, your shot noise is √10000 = 100 photons, giving you SNR ≈ 100.

The only way to beat shot noise is to collect more photons.

Look back at the 1 AU definition. It passes approximately 83–84% of the central Airy disk's energy. That means the "optimal" 1 AU setting is, by definition, rejecting roughly 16% of your precious in-focus signal from the central disk, plus essentially all the signal from the outer diffraction rings (which still contain in-focus information).

When your sample is dim, this discarded signal is your lifeline.

This leads to the "vicious cycle" of the 2:00 AM microscopist:

You are stuck. You're either destroying your sample with the laser or getting noisy, unquantifiable data from the detector. The 1 AU pinhole, in its quest for "perfect" resolution, is starving your detector.

Section 3: The Heretic's Guide: The "Why and How" of Opening the Pinhole

Here is the single most important trade-off in confocal microscopy, and the one that will save your experiments:

Let's unpack this. For modest increases in pinhole size (e.g., 1 AU → 1.5–2 AU), increasing the pinhole increases the amount of detected in-focus light approximately with aperture area (area ∝ radius²), while the confocal optical section (axial PSF) broadens. The exact change in axial FWHM depends on NA, wavelength, and where the pinhole sits relative to the Airy pattern and is best measured experimentally—but the rule of thumb holds well in practice.

If you double your pinhole radius from 1 AU to 2 AU, you increase your signal collection area by a factor of approximately 4 (Area ∝ 2² = 4). You are now collecting vastly more of those precious in-focus photons.

What's the cost? Your Z-resolution (your optical slice thickness) increases, roughly doubling in this example.

Let's Be Quantitative

At 1 AU with 100 photons/pixel, your SNR is √100 = 10.

At 2 AU, you collect approximately 400 photons/pixel from the same structure at the same laser power (a ~4× increase as a rule of thumb). Your SNR is now √400 = 20—you've doubled your SNR while maintaining the same photon dose to your sample.

Doubling the pinhole radius (area ∝ r²) → approximately 4× the collected photons → SNR increases by a factor of 2 (from √N scaling), all else equal.

This is the closest thing to a free lunch in microscopy. For photon-limited samples, the SNR gain from more photons often outweighs the modest loss in axial resolution for many biological questions.

The principle is well-established in the optical literature (Sandison & Webb, Applied Optics, 1994; Heintzmann & Ficz, Breaking the diffraction resolution limit, 2006), though it remains underappreciated in practice.

Note: Lateral (XY) resolution is much less affected. It degrades slightly with larger pinholes, but it's a far less dramatic effect than the Z-axis, and you're almost always limited by your objective's NA or your pixel sampling (Nyquist) anyway.

A Critical Caveat: Background in Thick Samples

A Practical Walkthrough on Your Confocal System

Let's do this. You're at the scope. The image is noisy at 1 AU.

1. Get a Baseline. Set your laser and gain to a "reasonable" level—where you aren't bleaching but can just see your signal. In your confocal software, find the Pinhole or Confocal Aperture setting. It's probably a slider or a text box displaying "1 AU."
2. DON'T TOUCH LASER OR GAIN. This is the key. We are isolating the variable.
3. Nudge it to 1.5 AU. Now, set the pinhole to 1.5 AU. What happens? Your image instantly gets brighter. The "salt-and-pepper" noise is visibly reduced. Your histogram, which was a sad little lump on the left, has shifted to the right. Your SNR has dramatically improved. You've just increased your signal collection area by approximately 1.5² = 2.25 times.
4. Try 2.0 AU. What happens? The image is now significantly brighter and cleaner. You've increased your signal by approximately 4× (2.0² = 4). The background is still black (you're still rejecting plenty of out-of-focus light), but your dim structure of interest is now "popping" with a clear, strong signal.
5. Now, Re-optimize. Your image is probably so bright at 2 AU that your brightest pixels are saturated. Now you have new power:
   • Power 1: You can turn down the detector gain. By turning down the gain, you reduce the amplification noise of the detector, making your "clean" signal even cleaner.
   • Power 2: You can turn down the laser power. This is the big one. You can now get the same image brightness you had at 1 AU, but with approximately 1/4 of the laser power. You've just reduced your photobleaching by roughly 75%.
   • Power 3: You can scan faster. More photons per unit of time means you can dwell on each pixel for a shorter duration. You can increase your scan speed, reducing acquisition time and photobleaching even further.

By opening the pinhole from 1 AU to 2 AU, you have single-handedly solved the "vicious cycle."

The Photobleaching Advantage (Quantified)

Let's make the photobleaching benefit explicit. Photobleaching follows first-order kinetics:

where k is proportional to laser power.

At 1 AU with 100% laser power, after 10 frames:

You retain approximately 37% of your signal.

At 2 AU with 25% laser power (delivering equivalent brightness):

You retain approximately 78% of your signal.

Section 3.5: What Your Microscope Isn't Telling You

Here's a truth rarely discussed in polite microscopy circles: many "1 AU" implementations are already compromises.

Most commercial confocal microscopes use stepped pinhole turrets with discrete aperture sizes, not a continuously variable iris. When you select "1 AU" at 488 nm with your 60×/1.4 NA objective, the software is choosing the closest available physical aperture—which might actually be 1.1 AU or 0.9 AU depending on the turret design and manufacturing tolerances. Check your instrument's manual or talk to your facility manager to understand your system's actual pinhole sizes.

Additionally, real objectives have aberrations. Your "1.4 NA" objective might effectively be 1.35 NA at the field edge due to spherical aberration or field curvature. The Airy disk is larger there, and your "perfect" 1 AU pinhole is now functionally 0.95 AU in that region, rejecting even more signal.

The coverslip matters too. If your coverslip is 0.19 mm thick instead of the ideal 0.17 mm (the #1.5 standard), your PSF has broadened. The "1 AU" setting is now closer to 0.85–0.9 AU in practice.

Section 4: "But Won't My Image Be Blurry?" — Mythbusting and Applications

Let's address the objections from the "confocal purists."

Objection 1: "This isn't 'true' confocal anymore! It's cheating!"

My question is: are you trying to publish a paper on optical theory, or are you trying to publish a paper on biology? The goal is data, not dogma.

An image taken at 1 AU that is 90% noise is 0% data. It is unquantifiable and unpublishable.

An image taken at 2 AU that has a clean signal, a high SNR, and a slightly thicker Z-slice is 100% data. It is quantifiable, robust, and beautiful.

Don't let academic dogma get in the way of good science.

Objection 2: "My Z-stacks and 3D reconstructions will be useless!"

Will they? Let's do the math.

With a 60×/1.4 NA oil objective at 488 nm, your theoretical 1 AU optical slice (FWHM) is approximately 0.5–0.6 μm.

At 2 AU, your slice thickness increases to approximately 0.9–1.1 μm (the exact value depends on your system and should be measured experimentally).

Now ask yourself: what are you imaging? Are you really trying to resolve two objects that are 0.7 μm apart in the Z-axis?

If you are imaging a whole nucleus, a cell-cell junction, a mitochondrion, or a dendritic spine, a 1.1 μm optical slice is often still more than enough Z-resolution to create a beautiful and accurate 3D reconstruction. You're just sampling your Z-stack with slightly thicker (but much cleaner) slices.

For most biological questions, this is a perfectly acceptable, and often superior, result.

When This Technique is a Lifesaver

• 1. Quantitative Intensity Measurements: This is the #1 use case. If you want to measure "how much" protein is in a spot, you need a high SNR. A noisy image makes your measurements unreliable. The coefficient of variation (CV) of your intensity measurements will be dominated by shot noise, not biological variance. Open the pinhole, get a clean signal, and quantify with confidence.
• 2. Live-Cell Imaging: This is the #2 use case. The only thing that matters in live-cell imaging is keeping the cells alive. This means minimal laser power. Opening the pinhole is the best way to get a usable signal with a non-toxic laser dose. Your cells don't care about your Z-resolution—they care about not being cooked to death.
• 3. Dim Samples: Any low-expression protein, endogenous tag (like CRISPR knock-in fluorophores), thick scattering tissue (like organoids or brain slices), or weakly-stained immunofluorescence.
• 4. Fast Scanning: If you need to capture a rapid event (calcium transients, vesicle fusion, neuronal firing), you need to scan fast. Fast scanning means low "pixel dwell time," which means few photons per pixel. Opening the pinhole gets you the photons you need to make fast scanning possible without sacrificing SNR.
• 5. Spectral Unmixing in Dim Channels: If you're doing 4- or 5-color imaging and one of your channels is intrinsically dim (far-red dyes, for example), opening the pinhole for that specific channel can make the difference between a usable and unusable dataset.

Section 5: When 1 AU is Non-Negotiable

I'm a pragmatist, not an anarchist. There are absolutely times when 1 AU is the right choice.

High-Resolution Colocalization

If your entire research question is "Are these 200 nm green puncta truly overlapping with these 200 nm red puncta at the nanoscale?", then yes. You need the absolute maximum XYZ resolution your system can provide. Use 1 AU (or smaller, if your system allows <1 AU settings), use a high-NA objective, sample at the Nyquist limit, and accept that you'll need to scan slowly and average multiple frames.

Super-Resolution Correlation

If you're correlating confocal with STED, SIM, or PALM, you need matched or at least well-characterized PSFs. A 2 AU pinhole will produce a PSF that doesn't correlate cleanly with your super-resolution modality.

Deconvolution

If you plan to use deconvolution software (like Huygens, Microvolution, or vendor packages) to computationally reassign blurry light, this software relies on a known, tight PSF. A 1 AU pinhole provides this. A wider pinhole complicates the deconvolution math, though some advanced software packages can model non-standard pinhole sizes if you provide the correct parameters.

Ratiometric Measurements in Thick Tissue

If you're doing FRET or ratiometric calcium imaging deep in scattering tissue, out-of-focus bleed-through at 2 AU can create false ratio shifts. The additional out-of-focus light from the acceptor channel (in FRET) or the indicator dye can contaminate your ratio calculation. In this case, tighter optical sectioning at 1 AU is worth the photon cost.

Your Sample is Already Bright

If you're imaging DAPI in a fixed cell, overexpressed GFP in a transfected cell line, or any other absurdly bright sample, you're drowning in photons. There's no need to open the pinhole. Stick to 1 AU and enjoy the crisp resolution. Use your photon excess to scan faster or at lower laser power instead.

Publication Politics

Sometimes scientific politics trumps scientific sense. If you're submitting to a top-tier journal and Reviewer 2 is a cranky confocal purist who demands "optimal confocal settings," you may need to comply to get your paper accepted.

In this case, I recommend the following strategy: perform your primary data acquisition at the pinhole setting that gives you the best data quality (likely 1.5–2 AU for dim samples). Then, for the paper, re-acquire a representative field at 1 AU to satisfy the reviewer. Include a supplementary figure showing that your biological conclusion holds at both settings. That combination satisfies reviewers who demand "optimal confocal settings" while preserving the best possible data quality (Jonkman et al., Nature Protocols, 2020).

Section 6: A Practical Decision Framework

Start every imaging session by asking three questions:

1. What is my limiting factor?

• Photons? → Open the pinhole.
• Resolution? → Keep it at 1 AU.
• Photobleaching? → Open the pinhole and lower laser power.
• Time? → Open the pinhole and scan faster.
• Background/scattering? → Test empirically; opening may help or hurt depending on specimen.

2. What am I measuring?

• Intensity (quantification)? → Open the pinhole for maximum SNR.
• Colocalization (overlap)? → If structures are >1 μm apart, open the pinhole. If <500 nm apart, stay at 1 AU.
• Morphology (shape)? → Depends on feature size. Dendritic spines? 2 AU is fine. Synaptic vesicles? 1 AU.
• Dynamics (time-lapse)? → Open the pinhole to minimize phototoxicity.

3. What is my Z-axis tolerance?

• Need to resolve <0.5 μm in Z? → 1 AU (and maybe use a higher NA objective or deconvolution).
• Structures separated by >1 μm in Z? → 2 AU is perfectly acceptable.
• Only care about XY (single optical section)? → Open the pinhole as wide as needed; Z-resolution is irrelevant.

The Experimental Approach: Test, Don't Guess

If you're unsure, here's the gold standard approach recommended by Jonkman et al. (Nature Protocols, 2020):

Compare. Measure. Make an informed decision based on your actual data, not a textbook written in 1995.

Trust your data, not tradition.

Conclusion: Be a Scientist, Not Just a Technician

A technician follows the manual. They see the "1 AU" button as an instruction.

A scientist—a PhD-level researcher—understands the principles. You understand the variables. The microscope is an instrument, and you are the one who plays it. The buttons and sliders are not rules; they are tools for you to manipulate to solve a problem.

Your confocal system is a powerful, flexible instrument. Its "1 AU" setting is a fantastic starting point, not an immutable law. The software engineers gave you a pinhole slider for a reason: they expect you to use it intelligently.

Confocal microscopy is not a religion. The 1 AU pinhole is not a commandment. It's a starting point derived from theory that assumes ideal conditions—conditions your dim, photobleaching, noisy biological sample will never meet.

So the next time you're in that dark room at 2:00 AM, staring at a screen full of noise, don't just crank the laser. Stop. Think. Your limiting factor isn't your laser power; it's your photon-starved detector.

Take a deep breath, defy the dogma, and open the pinhole. Gain back those wasted photons. Drop your gain. Lower your laser. Watch your signal-to-noise ratio soar, and get the beautiful, quantifiable data you deserve.

Technical Appendix: The Math, Physics, and Methods Behind the Pinhole Trade-off

This appendix provides the rigorous theoretical foundation and practical implementation details for researchers who need to defend their pinhole choices to reviewers, facility managers, or collaborators. If you're satisfied with the practical guidance in the main text, feel free to skip this section. If you need to survive Reviewer 2 or write a bulletproof Methods section, read on.

A1. Theoretical Foundation: Airy Disks, PSFs, and Encircled Energy

The Airy Pattern and the Definition of 1 AU

The point-spread function (PSF) of a diffraction-limited optical system is described by the Airy pattern in the focal plane:

where J₁ is the first-order Bessel function, k = 2πNA/λ, and r is the radial distance from the optical axis.

The central bright disk (the Airy disk) extends to the first minimum, which occurs at:

This defines the radius of the Airy disk in the sample plane. When projected back through the microscope optics to the pinhole plane (the conjugate image plane), this becomes the physical size of the Airy pattern at the pinhole.

One Airy Unit (1 AU) is defined as the diameter of this first minimum at the pinhole plane. It is a relative unit that automatically scales with wavelength and numerical aperture.

Encircled Energy

The fraction of total PSF energy contained within a circular aperture of radius r is given by:

For r = r_min (i.e., 1 AU), this evaluates to approximately 83–84% of the total PSF energy. This is why 1 AU is often cited as capturing "most" of the in-focus light.

However, this calculation assumes:

• A perfect, aberration-free optical system
• Infinite detection bandwidth
• No scattering or aberrations in the sample

Real-world deviations from these assumptions mean the actual transmission at "1 AU" varies between systems and conditions.

References:
• Born & Wolf, Principles of Optics (7th ed.), Cambridge University Press, 1999
• Wilson, T., Confocal Microscopy, Academic Press, 1990
• Scientific Volume Imaging, "Airy Units and the Pinhole" tutorial

A2. Shot Noise, SNR, and the Photon Budget

Photon Statistics

Fluorescence emission is a Poisson process. For N detected photons, the variance is:

and thus the standard deviation is:

The signal-to-noise ratio in the shot-noise limit is therefore:

This is the fundamental relationship that drives all optimization in photon-limited imaging.

Effect of Pinhole Size on Collected Photons

For small to moderate increases in pinhole diameter (e.g., 1 AU → 2 AU), the number of detected photons scales approximately with the area of the pinhole aperture:

where r is the pinhole radius.

Doubling the radius (1 AU → 2 AU) therefore increases detected photons by a factor of approximately 4, assuming:

• The pinhole is the limiting aperture (true for most confocal systems)
• The PSF is well-centered on the pinhole
• Detector saturation and other nonlinearities are negligible

SNR Improvement

From SNR = √N and N ∝ r²:

For 1 AU → 2 AU:

You double the SNR by quadrupling the collected photons.

References:
• Sandison, D. R. & Webb, W. W., "Background rejection and signal-to-noise optimization in confocal and alternative fluorescence microscopes," Applied Optics 33(4), 603–615 (1994)
• Hell, S. W., et al., "Confocal microscopy with an increased detection aperture: type-B 4Pi confocal microscopy," Optics Letters 19(3), 222–224 (1994)

A3. Axial Resolution vs. Pinhole Size

Optical Section Thickness

The full-width at half-maximum (FWHM) of the axial PSF (the "optical section thickness") for a confocal microscope is approximately:

at 1 AU.

As the pinhole opens, out-of-focus light is increasingly transmitted, broadening the effective axial response. Theoretical and empirical studies (Sandison & Webb, 1994; Brakenhoff et al., 1979) show that for pinhole diameters up to ~2–3 AU, the FWHM increases roughly linearly with pinhole diameter for a given NA and wavelength:

where d is the pinhole diameter in AU and α ≈ 0.5–1.0 depending on system geometry.

Practical Example

For a 60×/1.4 NA objective at 488 nm:

• 1 AU: FWHM_z ≈ 0.55 μm
• 1.5 AU: FWHM_z ≈ 0.7–0.8 μm
• 2 AU: FWHM_z ≈ 0.9–1.1 μm

The exact values depend on system implementation and should be measured experimentally using sub-resolution fluorescent beads.

References:
• Brakenhoff, G. J., et al., "Confocal scanning light microscopy with high aperture immersion lenses," Journal of Microscopy 117(2), 219–232 (1979)
• Sandison & Webb (1994), cited above

A4. The Background Caveat: When Opening the Pinhole Backfires

Out-of-Focus Background

Opening the pinhole increases transmission of both in-focus and out-of-focus light. In thin, well-labeled samples with low background, the in-focus signal dominates and SNR improves. In thick or scattering samples, however, out-of-focus background can increase faster than in-focus signal, reducing contrast and potentially harming detectability.

The detected signal is:

For a thick fluorescent specimen, S_out-of-focus can be substantial. While S_in-focus scales approximately as r², the out-of-focus contribution depends on sample thickness, scattering, and axial distribution of fluorophores. In highly scattering tissue (e.g., brain slices, organoids), the out-of-focus component can scale nearly as fast as the in-focus component, nullifying the SNR benefit.

Contrast vs. SNR

In these cases, contrast (not absolute SNR) becomes the limiting factor:

A wider pinhole may increase S_background enough to reduce C even if absolute S_signal increases.

The Solution: Empirical Testing

The only reliable way to know if opening the pinhole helps or hurts is to test it on your actual sample. Acquire images at multiple pinhole settings and measure both SNR and contrast.

References:
• Gauderon, R., et al., "Effect of a finite-size pinhole on noise performance in single-, two-, and three-photon confocal fluorescence microscopy," Applied Optics 38(16), 3562–3565 (1999)
• Sheppard, C. J. R. & Matthews, H. J., "Imaging in high-aperture optical systems," Journal of the Optical Society of America A 4(8), 1354–1360 (1987)

A5. Practical Methods: How to Report and Optimize Pinhole Settings

Reporting Pinhole Settings in Publications

To ensure reproducibility, report the following parameters:

1. Pinhole size in Airy Units (e.g., 1.5 AU)
2. Objective magnification and numerical aperture (e.g., 60×/1.4 NA oil)
3. Detection wavelength(s) (e.g., 500–550 nm for GFP)
4. Physical aperture diameter (if known) from your system's specifications or user manual

This level of detail allows other researchers to replicate your detection geometry.

Reference:
• Jonkman, J., et al., "Tutorial: guidance for quantitative confocal microscopy," Nature Protocols 15, 1585–1611 (2020)

Optimizing Pinhole Settings: A Step-by-Step Protocol

Objective: Determine the optimal pinhole setting for your sample and imaging goal.

Materials:

• Your biological sample
• Fluorescent bead slide (100–200 nm diameter, e.g., TetraSpeck beads) for PSF measurement
• Image analysis software (ImageJ/Fiji, MATLAB, or Python)

Protocol:

1. Mount your biological sample and locate a representative field of view with structures of interest clearly visible.
2. Set baseline imaging parameters:
   • Choose laser power that avoids saturation at 1 AU
   • Set detector gain (HV) to use the middle 40–70% of the dynamic range
   • Choose pixel dwell time and scanning speed appropriate for your sample
   • Set pinhole to 1.0 AU
3. Acquire a reference image at 1.0 AU. Save this image.
4. Acquire images at 1.5 AU and 2.0 AU without changing any other parameters (laser power, gain, dwell time, zoom, etc.).
5. Measure signal and noise in each image:
   • Draw an ROI around your structure of interest → measure mean intensity (S_signal)
   • Draw an ROI in background (cell-free region) → measure mean (S_bg) and standard deviation (σ_bg)
   • Calculate SNR: SNR = (S_signal - S_bg) / σ_bg
6. Compare SNR across pinhole settings. Choose the setting with the highest SNR for your structure of interest.
7. (Optional) Measure axial resolution:
   • Mount fluorescent bead slide
   • Acquire Z-stacks of isolated beads at each pinhole setting
   • Measure FWHM in Z by fitting a Gaussian to the intensity profile
   • Confirm that the axial resolution at your chosen pinhole setting is acceptable for your biological question
8. Re-optimize laser power and gain at your chosen pinhole setting:
   • If your image is now too bright (saturated), reduce laser power or gain
   • If you've reduced laser power, you've also reduced photobleaching—measure this by acquiring a time series and quantifying intensity decay
9. Document your optimization and include it in supplementary materials or Methods.

Reference:
• Jonkman et al. (2020), cited above

A6. Common Pitfalls and How to Avoid Them

Pitfall 1: Comparing Images with Different Laser Power or Gain

Problem: If you increase the pinhole and change laser power or gain simultaneously, you cannot isolate the effect of the pinhole.

Solution: Change only one variable at a time. Acquire images at different pinhole settings with all other parameters fixed, then re-optimize afterwards.

Pitfall 2: Ignoring Detector Saturation

Problem: At 2 AU, your signal may be so strong that the detector saturates. Saturated pixels contain no quantitative information.

Solution: Check your histogram. If pixels are piling up at the right edge (255 for 8-bit, 4095 for 12-bit, etc.), you are saturated. Reduce gain or laser power until the histogram is well within the dynamic range.

Pitfall 3: Assuming "1 AU" is Identical Across Systems

Problem: Different microscope manufacturers and models implement pinhole turrets differently. Your "1 AU" on an Olympus FV3000 is not necessarily identical to "1 AU" on a Leica SP8 or Zeiss LSM900.

Solution: Report the physical aperture size (in μm) if available from your system specs, or measure it empirically using bead standards.

Pitfall 4: Opening the Pinhole Without Checking Background

Problem: In thick or autofluorescent samples, opening the pinhole can flood your detector with out-of-focus background, reducing contrast and SNR.

Solution: Always measure background in addition to signal. Calculate both SNR and contrast. If background increases faster than signal, a wider pinhole may not help.

Pitfall 5: Not Testing Your Specific Sample

Problem: General rules (like "2 AU is better for dim samples") are useful starting points, but every sample is different.

Solution: Acquire test images at multiple pinhole settings on your actual biological specimen. Trust your measurements, not assumptions.

A7. Selected Key References

These are the foundational and practical references that support the arguments in this post. If you need to cite evidence for your Methods section or respond to reviewers, start here.

Foundational Optics

1. Wilson, T. Confocal Microscopy. Academic Press, 1990. — The definitive textbook on confocal theory.
2. Born, M. & Wolf, E. Principles of Optics (7th ed.). Cambridge University Press, 1999. — For Airy patterns, PSFs, and diffraction theory.
3. Sheppard, C. J. R. & Shotton, D. M. Confocal Laser Scanning Microscopy. BIOS Scientific Publishers, 1997. — Practical guide to confocal imaging with extensive discussion of pinhole effects.

Pinhole Size, SNR, and Detection

4. Sandison, D. R. & Webb, W. W. "Background rejection and signal-to-noise optimization in confocal and alternative fluorescence microscopes." Applied Optics 33(4), 603–615 (1994). — Experimental and theoretical analysis of SNR vs. pinhole size. Shows that larger pinholes improve SNR in photon-limited conditions.
5. Gauderon, R., et al. "Effect of a finite-size pinhole on noise performance in single-, two-, and three-photon confocal fluorescence microscopy." Applied Optics 38(16), 3562–3565 (1999). — Analyzes how pinhole size affects shot noise and background in multiphoton systems (principles apply to single-photon confocal as well).
6. Brakenhoff, G. J., et al. "Confocal scanning light microscopy with high aperture immersion lenses." Journal of Microscopy 117(2), 219–232 (1979). — Early foundational work on confocal resolution and pinhole effects.

Practical Guidance and Quantitative Imaging

7. Jonkman, J., et al. "Tutorial: guidance for quantitative confocal microscopy." Nature Protocols 15, 1585–1611 (2020). — Comprehensive, modern tutorial on optimizing confocal settings for quantitative work. Includes explicit discussion of pinhole trade-offs.
8. Heintzmann, R. & Ficz, G. "Breaking the resolution limit in light microscopy." Briefings in Functional Genomics & Proteomics 5(4), 289–301 (2006). — Discusses resolution trade-offs in modern microscopy.
9. Murray, J. M. "Practical aspects of quantitative confocal microscopy." Methods in Cell Biology 38, 95–132 (1993). — Older but still relevant practical guide.

Manufacturer and Educational Resources

10. Scientific Volume Imaging (SVI). "The Airy disk." https://svi.nl/AiryDisk — Clear educational resource on Airy patterns and encircled energy.
11. Olympus Life Science. "Confocal Microscopy—Introduction." https://www.olympus-lifescience.com/en/microscope-resource/primer/techniques/confocal/ — Vendor tutorial with practical imaging advice.
12. Leica Microsystems. "Confocal Application Letters." — Various application notes on pinhole optimization (available through Leica website or technical support).
13. Nikon MicroscopyU. "Introduction to Confocal Microscopy." https://www.microscopyu.com/techniques/confocal — Educational resource with interactive tutorials.

A8. Closing Thoughts for the Technical Reader

The 1 Airy Unit rule is an elegant result of diffraction theory, and it serves as an excellent default for many imaging scenarios. But science is not about following rules—it's about understanding principles and optimizing methods to answer biological questions.

For photon-limited samples—which constitute the vast majority of biological fluorescence imaging—opening the pinhole to 1.5–2 AU often provides substantial improvements in SNR with acceptable trade-offs in axial resolution. This is not a "hack" or a compromise; it is an optimization based on the physics of light and the statistics of photon detection.

The key is to be empirical, quantitative, and honest:

• Measure your SNR and resolution at different settings
• Report your parameters clearly in Methods
• Show that your biological conclusions hold across settings (if challenged by reviewers)

By doing so, you uphold the highest standards of scientific rigor while producing the best possible data from your samples.