Augmented reality (AR) requires a pane of glass to perform two contradictory functions simultaneously. It must allow the real world to pass through with high transparency while also delivering a bright, full-color image to a constantly moving pupil. Waveguide combiners address this by trapping projected light within a thin substrate via total internal reflection and directing it toward the eye. That architecture explains why waveguides displaced bulkier alternatives such as freeform mirrors, birdbath optics, and Maxwellian projection systems.
Image Credit: ROMBIX STUDIO/Shutterstock
A planar guide keeps the optical stack close to the plane of a spectacle lens, which matters when the product must sit on a human face for hours. The remaining difficulties concern efficiency, color fidelity, and the image's depth structure.1,2
Why Light Fields Entered the Conversation
Conventional AR glasses display a flat image that is fixed at a single virtual distance. As the eyes converge on a rendered object, the crystalline lens remains focused on the same focal plane, leading to the vergence-accommodation conflict. Prolonged use can result in visual fatigue, dizziness, and nausea, causing a disconnect between the displayed scene and the natural perception of depth.3
In contrast, light field approaches reconstruct the directional distribution of light rather than a single flat picture. Integral imaging and computational holography carry complete depth information, while multifocal and tunable-focus designs provide partial depth information. However, laboratory prototypes have demonstrated the challenges of this approach, with one integral imaging module achieving only a 33.4° field of view and a 6.5 mm eyebox.3
The accuracy of depth perception thus competes with the initial specifications that consumers notice. A reported holographic module achieved complete wavefront reconstruction at a field of just 4.8 degrees. The engineering question for the next decade concerns whether depth-correct imagery can be produced inside a combiner slim enough to disappear into eyewear, and recent work suggests the waveguide is where that convergence happens.3
Inside the Waveguide
Waveguide combiners split into geometric and diffractive families. Geometric versions use mirrors or prisms for coupling and arrays of cascaded partially reflective mirrors for extraction, and they introduce almost no chromatic aberration because refraction and reflection treat all visible wavelengths in much the same way. Each embedded mirror may require tens of coating layers to maintain its reflection ratio, which considerably increases fabrication difficulty.1
On the other hand, diffractive combiners are widely used in commercial products from companies like Microsoft, Magic Leap, and Dispelix. These combiners employ surface-relief and volume holographic gratings to perform the coupling. The design of the grating profile determines how much light is efficiently transmitted. For example, a 4-level phase grating can achieve 81% diffraction efficiency, while an 8-level version can reach up to 95%. Additionally, polarization volume gratings can respond selectively to a single circular polarization.1
Nonetheless, diffraction comes with its trade-offs, such as dispersion. The deflection angles produced by diffraction vary with wavelength, producing color shifts across the image and reducing the usable full-color field. Manufacturers have answered by stacking one waveguide per primary color, an approach that adds thickness and weight, creates ghost images, and degrades the modulation transfer function whenever the stacked layers drift out of mechanical alignment.4
The Etendue Constraint
Every waveguide follows the rules of etendue conservation, which connects the field of view to the size of the eyebox. Exit pupil expansion helps overcome this limit by using light reflected multiple times, creating a wider viewing area. Wider fields and larger eyeboxes demand larger couplers and lower the overall efficiency of the combiner.1
The eyebox must accommodate pupil size, eye movements, and the range of distances between people's eyes. Eye tracking and fitting for each user can help meet these requirements. A recent double-layer diffractive design published in Scientific Reports used 50 out-coupling elements arranged in five rows and ten columns within a 2 mm guide. This setup achieved 37% transmission efficiency and 88.96% uniformity in field measurements.5
Download the PDF of this page here
These numbers help understand the limits on brightness in daylight. If a combiner passes about a third of the light it receives, the projector needs to be efficient, small, and have a narrow light beam. A compact in-coupler can only work with a tightly controlled beam that has a small f-number. Therefore, the combiner and light engine designs cannot be optimized separately.1
Metasurfaces use tiny nanostructures instead of traditional ruled gratings to control phase, amplitude, and polarization of light. A recent study in Light, Science & Applications uses 160-nm-tall silicon nitride on a base with an index of 1.9. It directs the fourth, fifth, and sixth diffraction orders to red, green, and blue light. By matching the diffraction order to the wavelength, it focuses the three primary colors into a single viewing angle, creating a 45° field from a single plate.4
This is where light fields and waveguides finally meet. Another work published in Nature paired a phase-only spatial light modulator mounted against the in-coupler with an all-glass metasurface waveguide. This setup eliminates the need for a projection lens and produces full-color, 3D holographic images through a transparent plate. Its symmetric couplers used a 384 nm period and delivered about 78.4% see-through efficiency across the visible spectrum.6
There are some limitations to consider. The field of view is 11.7°, and the eyebox, or the area where the image remains clear, measures 4 mm across. The uniformity of green and blue light is above 91%, but red light is lower at 61.7% due to its larger diffraction angle. A camera-based neural network trains the phase patterns, making computation an important part of the optical system.2
What Comes Next
Progress now depends on materials and manufacturing as much as on optical theory. Glass with a refractive index above 1.8 broadens the achievable field and allows a single coupler to serve all three primaries simultaneously. Metagratings also appear compatible with nanoimprint replication, and that compatibility points toward volume production of single-layer combiners at costs consumer eyewear can absorb.2
The trajectory is clear enough to describe with some confidence. Depth-correct light-field content, computational wavefront control, and achromatic metasurface coupling are converging on a single thin glass plate. The designs that succeed will be those that balance efficiency, color uniformity, and eyebox size within a field of view worth wearing for an entire working day.6
References and Further Reading
- Ding, Y. et al. (2023). Waveguide-based augmented reality displays: perspectives and challenges. eLight 3, 24. DOI:10.1186/s43593-023-00057-z. https://link.springer.com/article/10.1186/s43593-023-00057-z
- Lee, J., & Kim, S. J. (2025). Meta-Optics for Optical Engineering of Next-Generation AR/VR Near-Eye Displays. Micromachines, 16(9). DOI:10.3390/mi16091026. https://www.mdpi.com/2072-666X/16/9/1026
- Wang, S. et al. (2023). Research Progress of Vergence-Accommodation Conflict in Near-Eye Display Based on Augmented Reality. Acta Optica Sinica, 43(23): 2300001. DOI:10.3788/AOS231074. https://www.researching.cn/articles/OJ962981736a1e2869
- Tian, Z. et al. (2025). An achromatic metasurface waveguide for augmented reality displays. Light, Science & Applications, 14, 94. DOI:10.1038/s41377-025-01761-w. https://www.nature.com/articles/s41377-025-01761-w
- Zhang, J. et al. (2024). Design of waveguide with double layer diffractive optical elements for augmented reality displays. Scientific Reports, 14(1), 24310. DOI:10.1038/s41598-024-75766-7. https://www.nature.com/articles/s41598-024-75766-7
- Gopakumar, M. et al. (2024). Full-colour 3D holographic augmented-reality displays with metasurface waveguides. Nature, 629(8013), 791-797. DOI:10.1038/s41586-024-07386-0. https://www.nature.com/articles/s41586-024-07386-0
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.