Independent amplitude and phase modulation using two sequential regions of one phase-only SLM

arXiv:2606.07821 · physics.optics · Submitted 2026-06-05 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Independent amplitude and phase modulation using two sequential regions of one phase-only SLM".

Mira: Here's the gist of the paper. Liquid-crystal spatial light modulators (SLMs) are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping.

Kai: Here we demonstrate full complex-field modulation using a single phase-only SLM by implementing two sequential modulation planes on different regions of the same device.

Mira: Pruning redundant reasoning tokens The core contribution of this work is the demonstration of independent amplitude and phase shaping using a single phase-only SLM by employing two spatially separated regions as sequential modulation planes.

Kai: The first region, combined with a polarizer, converts programmed phase retardance into controllable amplitude modulation.

Title and authors: Mira: Let's talk about the paper "Independent amplitude and phase modulation using two sequential regions of one phase-only SLM" and who did it. It’s a lot of work by Marius Constantin Chirita, Mikuláš Fiala, and Martin Kozák from Charles University.

Kai: The title itself tells you the main point is getting both independent amplitude and phase control from just one phase-only SLM device. That’s a big deal because as we said earlier, that’s usually not possible with a single modulator operating in its typical way.

Lev: From an error correction standpoint, the fact that they are using sequential modulation planes instead of trying to build this into one complex modulator is interesting because it decouples the amplitude encoding from the phase imposition.

Mira: Right. The authors show how the first region uses a polarizer to turn its phase retardance into an amplitude modulation, and then that output is imaged onto the second region which imposes exactly what’s left for the phase distribution.

Kai: So it’s like they are splitting the task: one part handles the 'how bright' of it, and the other part handles the 'where' of it in terms of phase. It’s a compact way to achieve full complex-field modulation without needing a second physical modulator.

The paper's summary: Kai: So what they actually demonstrate is this process where you use two spatially separated regions on one SLM to create an output field that has independent amplitude and phase control. It’s about synthesizing any arbitrary complex field in the pupil plane.

Mira: They show the mathematical prescription for this: if your target complex field has an amplitude envelope A and a phase pattern P, then you calculate a required phase mask for the first region based on that envelope, and then you calculate a second phase mask that corrects for both the original desired phase and what the first region already did.

Lev: I always look at these kinds of synthesis methods and I immediately think about how much computational overhead this adds. Even if the physical device is compact, calculating those two masks on a computer to feed into them takes time.

Kai: The real advance they highlight is that this works for generating things like Bessel Gaussian beams with or without an added helical phase, which lets you independently control the radial amplitude and the azimuthal phase of the beam.

Mira: And they test this by comparing their measured focal-plane intensities against simulations for more general target distributions, showing high correlation coefficients between zero point eight six and zero point nine six against things like letter patterns and smiley patterns.

The paper's improvements: Kai: What they really improve is the practicality of using this setup as a compact, high-power compatible single SLM implementation for complex focal-field generation. It makes structured illumination and holography much more accessible with this architecture.

Lev: For real hardware, that compactness is good, but the paper also mentions that direct amplitude control via polarization projection might not be as efficient as purely phase-only beam shaping because of those losses. That’s a real trade-off we have to consider when building something robust.

Mira: They also address the limitations by stating that the fidelity of high spatial frequency patterns is limited by things like finite pixel pitch and any residual wavefront errors you might have in the device itself.

Kai: So, they aren't claiming perfection. They’re being realistic about where this method stops working well, which is important for anyone trying to apply it to real-world imaging or control systems.

Conclusion: Mira: So, wrapping up the paper "Independent amplitude and phase modulation using two sequential regions of one phase-only SLM," the main thing is that they’ve shown a practical route to complex focal-field generation using just one SLM device by splitting the control between two regions.

Lev: From an error correction perspective, it confirms that this sequential approach is a viable method for achieving independent amplitude and phase prescription in a single physical component, even if you have to manage the interface between those two modulation planes.

Kai: For structured illumination and holography applications, this gives us a new tool to design optical fields that are complex not just in phase but also in amplitude.

Mira: It’s definitely a step toward making these devices more versatile for free-electron control and other advanced optical shaping tasks, even with the fidelity limitations they mentioned regarding pixel pitch.

Lev: Yeah, the overhead for running this on real hardware will depend heavily on how well those two regions can be perfectly aligned and imaged onto each other without introducing unwanted artifacts.

Kai: So that’s the gist of it—a compact method for complex field synthesis using a single SLM with two sequential modulation planes.

Charles University, Faculty of Mathematics and Physics

physics.optics

Submitted: 2026-06-05

Updated: 2026-11-12

Comments: v2: AAM; licence: CC BY 4.0 Revised to match the accepted manuscript

DOI: 10.1364/OPTCON.607918

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 92/100

The gist: Liquid-crystal spatial light modulators (SLMs) are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping.

Key concepts

Phase-Only SLM
A spatial light modulator that can only manipulate the phase of light waves, not their amplitude directly. This paper overcomes the limitation of traditional phase-only devices by using a clever two-step process to control both amplitude and phase simultaneously.
Sequential Modulation Planes
Using two distinct, spatially separated areas on the same SLM as sequential modulation planes. The first region encodes the desired field amplitude, and the second region imposes the necessary phase correction to complete the complex field synthesis.
Complex Field Synthesis
The ability to create an optical field defined by both its amplitude and phase simultaneously. By using two regions, one handles amplitude (via a cosine factor) and the other handles phase, allowing for the precise prescription of any desired complex light pattern.

Terminology

Summary

Liquid-crystal spatial light modulators (SLMs) are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping. p. 0

Where the paper says it

kozak@matfyz. cz Abstract: Liquid-crystal spatial light modulators (SLMs) are widely used for programmable wavefront control but are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping.

Here we demonstrate full complex-field modulation using a single phase-only SLM by implementing two sequential modulation planes on different regions of the same device. p. 0
Where the paper says it

cz Abstract: Liquid-crystal spatial light modulators (SLMs) are widely used for programmable wavefront control but are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping. Here we demonstrate full complex-field modulation using a single phase-only SLM by implementing two sequential modulation planes on different regions of the same device.

Pruning redundant reasoning tokens

The core contribution of this work is the demonstration of independent amplitude and phase shaping using a single phase-only SLM by employing two spatially separated regions as sequential modulation planes. The first region, combined with a polarizer, converts programmed phase retardance into controllable amplitude modulation. p. 1

Where the paper says it

The method uses two spatially separated regions of the same SLM as two sequential modulation planes. The first region, together with a polarizer placed in the beam path, converts a programmed phase retardance into controllable amplitude modulation.

This modulated field is then imaged onto the second region with unit magnification, where an independently programmed phase mask compensates for the amplitude encoding and imposes the required output phase. p. 1
Where the paper says it

The first region, together with a polarizer placed in the beam path, converts a programmed phase retardance into controllable amplitude modulation. The field is then imaged with unit magnification onto the second SLM region, where an independently programmed phase mask compensates the phase introduced during amplitude encoding and imposes the desired output phase.

This architecture allows a single reflective device to perform the function of two coordinated modulation planes while preserving a direct coordinate mapping between masks. p. 1
Where the paper says it

In this way, a single reflective phase-only device performs the function of two coordinated modulation planes while preserving a direct coordinate mapping between the amplitude and phase masks. We use this architecture to synthesize prescribed complex fields in the pupil plane and map them to the focal plane through Fourier propagation [37].

Experimental Concept and Theoretical Framework

The experimental concept involves using two spatially separated regions of one phase-only SLM as sequential modulation planes. Region A encodes the desired field amplitude, while region B independently imposes the remaining phase needed to synthesize a prescribed complex field in the pupil plane. p. 2

Where the paper says it

Although each region can only impose a phase retardance, the first region, together with a polarizer placed in the beam path, provides amplitude modulation, while the second region provides phase modulation, enabling control of the full complex optical field. In this way, the first SLM region is used to encode the desired field amplitude, while the second SLM region independently imposes the remaining phase required to synthesize a prescribed complex field in the pupil plane.

The resulting scalar complex optical field is expressed as Eout(xp, yp) = A0 cos(phiA/2) exp(i(phiA/2 + phiB)). p. 2
Where the paper says it

The second SLM region then adds the phase 𝜙B(𝑥 𝑝, 𝑦 𝑝). Thus, after propagation through the two independently addressed regions, the scalar complex optical field can be written as [30] 𝐸out(𝑥 𝑝, 𝑦 𝑝) = 𝐴0 cos 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 exp 𝑖 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 + 𝜙B(𝑥 𝑝, 𝑦 𝑝) , (1) where 𝐴0 is the real-valued amplitude of the incident field, and 𝜙A(𝑥 𝑝, 𝑦 𝑝) and 𝜙B(𝑥 𝑝, 𝑦 𝑝) are the phase distributions encoded on the first and second SLM regions, respectively.

This equation shows that region A controls the output amplitude through the factor cos[phiA/2], while region B adds the phase phiB. p. 2
Where the paper says it

Thus, after propagation through the two independently addressed regions, the scalar complex optical field can be written as [30] 𝐸out(𝑥 𝑝, 𝑦 𝑝) = 𝐴0 cos 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 exp 𝑖 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 + 𝜙B(𝑥 𝑝, 𝑦 𝑝) , (1) where 𝐴0 is the real-valued amplitude of the incident field, and 𝜙A(𝑥 𝑝, 𝑦 𝑝) and 𝜙B(𝑥 𝑝, 𝑦 𝑝) are the phase distributions encoded on the first and second SLM regions, respectively. Equation (1) shows that region A controls the output amplitude through the factor cos[𝜙A(𝑥 𝑝, 𝑦 𝑝)/2], while the total output phase is determined by the sum of the phase generated during amplitude encoding and the phase imposed by region B.

Complex-Field Synthesis using Two Phase-Only SLM Regions

The desired complex field in the SLM/pupil plane is written in amplitude–phase form as Epupil(xp, yp) = A0A(xp, yp) exp[iP(xp, yp)]. The required phase pattern on region A is given by phiA(xp, yp) = 2 arccos[A(xp, yp)]. p. 3

Where the paper says it

(1) with Eq. (2) gives the required phase pattern on region A, 𝜙A(𝑥 𝑝, 𝑦 𝑝) = 2 arccos[A (𝑥 𝑝, 𝑦 𝑝)].

The second required phase mask is then calculated as phiB(xp, yp) = P(xp, yp) - arccos[A(xp, yp)]. p. 3
Where the paper says it

The phase contribution introduced by this step is then compensated when choosing the phase pattern on region B. The required second phase mask is 𝜙B(𝑥 𝑝, 𝑦 𝑝) = 𝑃(𝑥 𝑝, 𝑦 𝑝) − arccos[A (𝑥 𝑝, 𝑦 𝑝)].

These equations provide the direct prescription for synthesizing an arbitrary complex field in the SLM/pupil plane using two phase-only modulation steps. p. 3
Where the paper says it

The required second phase mask is 𝜙B(𝑥 𝑝, 𝑦 𝑝) = 𝑃(𝑥 𝑝, 𝑦 𝑝) − arccos[A (𝑥 𝑝, 𝑦 𝑝)]. (5) Equations (4) and (5) provide the direct prescription for synthesizing an arbitrary complex field in the SLM/pupil plane using two phase-only modulation steps.

Double-Phase Hologram Calculation for Arbitrary Focal Shapes

To generate a desired focal-plane distribution, the target field Etar(xf, yf) is first specified in the focal plane of a Fourier lens. The corresponding pupil-plane field is obtained by inverse Fourier propagation and restriction to the illuminated SLM aperture. p. 4

Where the paper says it

The target complex field is specified in the focal plane as 𝐸tar(𝑥 𝑓 , 𝑦 𝑓 ) = 𝐴tar(𝑥 𝑓 , 𝑦 𝑓 ) exp[𝑖Φtar(𝑥 𝑓 , 𝑦 𝑓 )]. (6) The corresponding field in the SLM/pupil plane is obtained by inverse Fourier propagation [40] and restriction to the illuminated SLM aperture, 𝐸pupil(𝑥 𝑝, 𝑦 𝑝) ∝ F −1 𝐸tar(𝑥 𝑓 , 𝑦 𝑓 ) Π𝑅(𝑥 𝑝, 𝑦 𝑝), (7) where Π𝑅(𝑥 𝑝, 𝑦 𝑝) is the circular pupil function of radius 𝑅 = 2.

This pupil-plane field is then identified with Eq. p. 2
Where the paper says it

The phase retardance applied in region A produces a spatially varying polarization ellipticity; a linear polarizer aligned with the incident polarization direction then converts this ellipticity into an effective amplitude modulation by transmitting only the corresponding polarization component. The modulated field is then imaged with unit magnification onto region B of the same SLM, so that the transverse coordinates in both regions can be identified with the same pupil-plane coordinates (𝑥 𝑝, 𝑦 𝑝).

(2) where A(xp, yp) and P(xp, yp) are inserted into Eqs. p. 4
Where the paper says it

(7) is identified with Eq. Its amplitude A (𝑥 𝑝, 𝑦 𝑝) and phase 𝑃(𝑥 𝑝, 𝑦 𝑝) are then inserted into Eqs.

(4) and (5) to obtain the two phase-only masks. p. 4
Where the paper says it

Its amplitude A (𝑥 𝑝, 𝑦 𝑝) and phase 𝑃(𝑥 𝑝, 𝑦 𝑝) are then inserted into Eqs. (4) and (5) to obtain the two phase-only masks applied to regions A and B of the SLM.

Results

The experimental validation confirms the expected behavior of the scheme, showing that measured intensity modulation follows the predicted cos 2(phiA/2) dependence. The generation of Bessel–Gaussian fields with and without a helical phase demonstrates that the setup can independently prescribe a radial amplitude distribution and an azimuthally varying phase. p. 8

Where the paper says it

The measured intensity modulation follows the predicted cos2 (𝜙A/2) dependence, verifying that the first SLM region and the polarizer act as a controllable amplitude modulator. The generation of Bessel–Gaussian fields with and without an added helical phase demonstrates that the setup can independently prescribe a radial amplitude distribution and an azimuthally varying phase.

The agreement between measured and simulated longitudinal profiles confirms that the two-region encoding preserves the intended complex field not only in the focal plane but also during propagation through the focal volume. p. 5
Where the paper says it

In contrast, the Bessel0–Gauss beam with added helical phase requires simultaneous control of a nonuniform radial amplitude and an azimuthally varying phase. The agreement between the measured and simulated longitudinal profiles confirms that the two-region encoding preserves the intended complex field not only in the focal plane, but also during propagation through the focal volume.

Discussion and Conclusion

The method realizes a compact single-SLM platform for independent amplitude and phase shaping of optical fields. The experimental results confirm the expected behavior, verifying that the first SLM region acts as a controllable amplitude modulator. p. 8

Where the paper says it

The experimental results confirm the expected behavior of the scheme. The measured intensity modulation follows the predicted cos2 (𝜙A/2) dependence, verifying that the first SLM region and the polarizer act as a controllable amplitude modulator.

This method offers a practical route for structured illumination, holographic beam shaping, and optical fields designed for free-electron control. p. 9
Where the paper says it

In conclusion, we have realized a compact single-SLM platform for independent amplitude and phase shaping of optical fields. Because it provides direct complex-field control with a single programmable element, this method offers a practical route for structured illumination [38], holographic beam shaping [1, 31], and optical fields designed for free-electron control [28], including light-based electron lenses [19], aberration correction [21], and high purity electron vortex beams.

Practical trade-offs include the fact that direct amplitude control is less efficient than purely phase-only beam shaping due to polarization projection losses. p. 9
Where the paper says it

More generally, the method involves practical trade-offs. Direct amplitude control is intrinsically less efficient than purely phase-only beam shaping, because the polarization projection used for amplitude modulation attenuates part of the incident field.

The fidelity of high-spatial-frequency patterns is limited by implementation factors such as finite pixel pitch and residual wavefront errors. p. 9
Where the paper says it

In addition, the finite pixel pitch, fill factor, phase calibration, and residual wavefront errors of the SLM limit the fidelity of high-spatial-frequency target patterns. These are implementation-specific limitations rather than fundamental restrictions of the encoding principle.

A Jones-Matrix Derivation of the Complex-Field Modulation

The output field is derived using Jones calculus, resulting in Eout(xp, yp) = E0(xp, yp) cos[phiA/2] exp(i(phiB + phiA/2)). This reduces to Eout(xp, yp) = E0 (xp, yp) cos[phiA/2] exp i(phiB + phiA/2). p. 2

Where the paper says it

The second SLM region then adds the phase 𝜙B(𝑥 𝑝, 𝑦 𝑝). Thus, after propagation through the two independently addressed regions, the scalar complex optical field can be written as [30] 𝐸out(𝑥 𝑝, 𝑦 𝑝) = 𝐴0 cos 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 exp 𝑖 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 + 𝜙B(𝑥 𝑝, 𝑦 𝑝) , (1) where 𝐴0 is the real-valued amplitude of the incident field, and 𝜙A(𝑥 𝑝, 𝑦 𝑝) and 𝜙B(𝑥 𝑝, 𝑦 𝑝) are the phase distributions encoded on the first and second SLM regions, respectively.

Calibration of the Amplitude Modulation

The amplitude response is verified by showing that the field amplitude scales as cos(phiA/2), yielding an intensity dependence I(phiA) = C + A cos 2(phiA/2). The fit yields R squared = 0.9874, indicating that the cos 2(phiA/2) model accounts for 98. p. 1

Where the paper says it

Single-SLM dual-phase schemes have demonstrated complex-field generation using two independently addressed regions of one modulator [32]. However, the ideal double-phase model relies on accurate one-to-one coordinate mapping between the two phase masks, which generally requires unit-magnification imaging between the corresponding SLM regions [33].

74% of the measured intensity variance. p. 4
Where the paper says it

The target field was a finite-aperture, super-Gaussian-apodized zeroth-order Bessel field, as described in AppendixB. Figure 2 compares the simulated and experimentally measured transverse intensity distributions for a Bessel0–Gauss beam and for the same radial field after adding a helical phase.

Image-Plane Resolution of the Amplitude Modulation

The spatial resolution of the amplitude-modulation step was determined to be a 10%–90% rise distance of 18.4 µm. This measured resolution is specific to the present optical alignment and imaging conditions, and should not be interpreted as a fundamental limitation of the complex-field modulation method. p. 9

Where the paper says it

Because it provides direct complex-field control with a single programmable element, this method offers a practical route for structured illumination [38], holographic beam shaping [1, 31], and optical fields designed for free-electron control [28], including light-based electron lenses [19], aberration correction [21], and high purity electron vortex beams. Jones-matrix derivation of the complex-field modulation The complex-amplitude modulation implemented in the present setup follows the two-step phase-only modulation approach described in Ref.

General Double-Phase Hologram Formalism

The desired pupil-plane field is obtained by inverse Fourier propagation and restriction to the illuminated SLM pupil. The complex pupil field is encoded on the phase-only SLM using a double-phase representation Eout(xp, yp) = cos[phiA/2] exp(i(phiA/2 + phiB)). p. 2

Where the paper says it

The second SLM region then adds the phase 𝜙B(𝑥 𝑝, 𝑦 𝑝). Thus, after propagation through the two independently addressed regions, the scalar complex optical field can be written as [30] 𝐸out(𝑥 𝑝, 𝑦 𝑝) = 𝐴0 cos 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 exp 𝑖 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 + 𝜙B(𝑥 𝑝, 𝑦 𝑝) , (1) where 𝐴0 is the real-valued amplitude of the incident field, and 𝜙A(𝑥 𝑝, 𝑦 𝑝) and 𝜙B(𝑥 𝑝, 𝑦 𝑝) are the phase distributions encoded on the first and second SLM regions, respectively.

The two phase masks are chosen as phiA = 2 cos-1(Aenc) and phiB = Phi - phiA/2. p. 2
Where the paper says it

The second SLM region then adds the phase 𝜙B(𝑥 𝑝, 𝑦 𝑝). Thus, after propagation through the two independently addressed regions, the scalar complex optical field can be written as [30] 𝐸out(𝑥 𝑝, 𝑦 𝑝) = 𝐴0 cos 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 exp 𝑖 𝜙A(𝑥 𝑝, 𝑦 𝑝) 2 + 𝜙B(𝑥 𝑝, 𝑦 𝑝) , (1) where 𝐴0 is the real-valued amplitude of the incident field, and 𝜙A(𝑥 𝑝, 𝑦 𝑝) and 𝜙B(𝑥 𝑝, 𝑦 𝑝) are the phase distributions encoded on the first and second SLM regions, respectively.

This procedure is used for all focal-plane patterns reported below. p. 4
Where the paper says it

Finally, both masks are wrapped modulo 2𝜋, converted into additive phase corrections relative to the uniform SLM background phase, and inserted into the two addressed SLM regions. This same back-propagation and double-phase encoding procedure is used for all focal-plane patterns reported below; the different examples differ only in the choice of the target field 𝐸tar(𝑥 𝑓 , 𝑦 𝑓 ).

Focal-Plane Target Fields

For the Bessel beam, the target field was prescribed analytically as Etar(rho f) = C J0(kr rho f) exp(-w/8 W(rho f)). For arbitrary focal shapes, the target complex field is taken as Etar(xf, yf) = sqrt[Itar(xf, yf)] exp(iphi0). p. 7

Where the paper says it

Generation of arbitrary focal-plane intensity distributions using complex-field modulation with a single SLM. Panels (a)–(d) show the target focal shapes, panels (e)–(h) show the corresponding numerical simulations, and panels (i)–(l) show the experimentally measured focal shapes.

The resulting correlation coefficients between 0. p. 8
Where the paper says it

A value of 𝐹corr = 1 corresponds to perfect pixel-wise agreement between the two normalized intensity distributions, whereas smaller values indicate increasing spatial deviations. The resulting correlation coefficients are listed in Table 1.

86 and 0.96 show high agreement with the corresponding targets. p. 8
Where the paper says it

The resulting correlation coefficients are listed in Table 1. All four measured focal-plane patterns show high agreement with their corresponding targets, with 𝐹corr ≥ 0.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Complex-field synthesis using two phase-only SLM regions allows for direct prescription of both amplitude and phase in a single device, enabling programmable complex wavefront engineering for structured illumination and holography.

  2. The improved system can generate Bessel–Gaussian beams with an added helical phase, which independently prescribes both the radial amplitude distribution and an azimuthally varying phase.

  3. The setup enables the synthesis of arbitrary focal-plane intensity distributions, achieving normalized correlation coefficients between 0.86 and 0.96 against targets such as letter patterns and smiley patterns.

Abstract

Liquid-crystal spatial light modulators (SLMs) are widely used for programmable wavefront control but are typically operated as phase-only devices, limiting applications that require independent amplitude and phase shaping. Here we demonstrate full complex-field modulation using a single phase-only SLM by implementing two sequential modulation planes on different regions of the same device. The phase retardance introduced by the first SLM region is converted into amplitude modulation by a polarizer placed in the beam path, while the second region compensates the associated phase offset and imposes the required phase distribution. The field from the first region is imaged onto the second, enabling complex-field synthesis without a second modulator. We validate the approach by generating Bessel--Gaussian beams, helical-phase fields, and arbitrary focal-plane intensity patterns. This single-SLM platform provides a compact route to programmable complex wavefront engineering for structured illumination, holography, and electron--light interaction experiments.

Related papers