2.1.

Angularly Multiplexed Volume Holographic Gratings

We designed and fabricated a volume phase grating recorded onto a photopolymer. The quality of the reconstructed field generated from the photopolymer-based volume hologram is mostly determined by the quality of the photosensitive material. Phenanthrenquinone (PQ)-doped polymethyl methacrylate (PMMA) polymer material (PQ:PMMA) has been regularly used for designing multiplexed volume gratings due to its excellent optical properties, such as high optical sensitivity, diffraction efficiency, and stability with negligible shrinkage.² It can provide a large refractive index modulation for recording high spatial frequencies. To fabricate a volume holographic grating, an interference pattern between the signal and reference wave is stored as the permanent changes in the refractive index of the medium that embed all the amplitude and phase information of the recorded beams. The refractive index modulation inside the medium is given by

Fig. 1

Schematic drawing: (a) the proposed 2P-VHI system and (b) conventional VHI system using single-photon excitation. Defocused beads images, shown in (b), are much suppressed in (a) due to two-photon excitation.

2.2.

Two-Photon-VHI System

We built the 2P-VHI system as shown in Fig. 1. It has two main parts, a two-photon illumination system and a VHI system incorporating AMVHGs.

2P illumination of the sample is obtained using a tunable mode-locked Ti:sapphire laser (800 nm, Spectra-Physics) and an Olympus microscope objective lens L1 ($\mathrm{MSPlan}100\times$, $\mathrm{NA}=0.95$). The imaging part of the microscope consists of Olympus objective L2 lens (MSPlan50X), and a specially designed AMVHG, a Mitutoyo L3 lens (MPlanAPO10X), to obtain a $4f$ imaging system as shown in Fig. 1. The output from the Ti:sapphire laser was coupled into the microscope and focused to provide a beam waist close to the back aperture of the objective lens. This laser delivered a time-averaged power of 1.7 W and a repetition rate of 100 MHz with a pulse duration of 100 fs. The power of the laser was attenuated using a variable neutral density filter. All the images were recorded with a high-resolution Andor camera (iXon 897). In this study, the volume hologram consists of two AMVHGs and is positioned at the Fourier plane. When the Bragg-matched condition is satisfied, then gratings act as a multifocus lens and display laterally separated multidepth images of volumetric samples onto the CCD as shown in Fig. 1. The depth separation ($\mathrm{\Delta }z$) of the two multiplexed gratings was $\sim 35\mu \mathrm{m}$ to probe two different focal planes. The depth separation can be tuned while recording the multiplexed grating.

3.1.

Imaging of Fluorescence Beads

The ability of the proposed 2P-VHI system to resolve multiple depths in volumetric samples was verified by imaging fluorescence-labeled microspheres ($15\text{-}\mu \mathrm{m}$-diameter, Polysciences) suspended within a thick slab of agarose (Invitrogen). Figure 2 provides comparisons of standard conventional VHI systems with single-photon excitation and the optically sectioned images captured by 2P-VHI systems, respectively. Two depth-resolved images with defocused light at both planes observed from the CCD with quasi-collimated LED illumination are shown in Fig. 2(a). Figure 2(b) shows the resultant images from the 2P-VHI system, and out-of-focus background noise is rejected from the desired in-focus signal.

Fig. 2

(a) Images of the fluorescently labeled microspheres from 2P-VHI system. Only optical sectioned images with in-focus signals are observed from the CCD. The two depth images were observed by scanning the illumination in the axial direction. (b) Conventional VHI system with single-photon excitation obtained with LED illumination. Images of $15\text{-}\mu \mathrm{m}$ fluorescence microspheres for the two axial planes. Both out-of-focus and in-focus microspheres are captured from the CCD. Note that the two images were captured without axial focus scanning adjustment.

3.2.

Measurement of Optical Sectioning Capability

To further investigate the optical sectioning capability of our 2P-VHI system, we experimentally measured the depth selectivity of the system. A thin sample of diluted fluorescein solution was evenly spread between cover slits.¹⁴ The thickness of the fluorescein solution was $\sim 1\mu \mathrm{m}$. The thin fluorescence sample was moved by controlling a motorized piezo-stage (MAX341, Thorlabs) at a $0.8\text{-}\mu \mathrm{m}$ step size along the axial direction, and defocused images were captured on the camera. In Fig. 3, red dots show the normalized total intensity of each defocused plane image versus the defocus distance for the 2P-VHI system, whereas the blue dots represent normalized total intensity for a conventional wide-field VHI system illuminated with a quasi-collimated LED source. Comparison of depth selectivity to the conventional VHI system with 2P-VHI system is shown in Fig. 3 and the measured full-width at half-maximum (FWHM) of the 2P-VHI system is $24\mu \mathrm{m}$. The conventional VHI system shows no depth selectivity. A drastic change in the depth selective property of the VHI system can be obtained by changing the illumination from one photon with the 2P illumination. Together with all the inherent advantages of the spatial–spectral imaging, the VHI system acquires fine optical sectioning ability, which further enhances its performance for high-contrast 3D imaging for various applications.

Fig. 3

Comparison of experimentally measured depth selectivity between conventional VHI system (blue dots) and 2P-VHI systems (red dots) with curve fitting in solid line.

Depth resolution will largely depend on the NA of the lens and the wavelength. In our experiment, we used an MSPlan $100\times$ objective lens with an NA of 0.95. At present, we used two angularly multiplexed grating with the recording parameter described previously. The separation between the two angularly multiplexed grating depths can be altered during the recording process.

3.3.

Multiplane Imaging of Bio-Samples Using 2P-VHI System

Experiments using fluorescently labeled microspheres are a condition where the sample is sparse. However, practical imaging for biomedical applications, in reality, often requires highly scattering samples. To test the performance of the 2P-VHI system for imaging dense and scattered samples, we used intestinal tissues of mice. The intestines were fixed in 4% paraformaldehyde overnight at 4°C. Subsequently, the sample was permeabilized by being incubated in 0.1% Triton-X 100 following which the intestine sample was stained with Alexa-488 labeled Phalloidin for green fluorescence labeling of actin proteins. The sample was washed thoroughly, placed between two coverslips in phosphate buffered saline medium, gently compressed, and sealed. The thick sample was then imaged with the 2P-VHI system. Figure 4 shows the images of the villi of the mouse intestine sample as taken with our 2P-VHI system. Two planes, separated by $35\mu \mathrm{m}$, are imaged onto different lateral parts of the camera. Figures 4(A)–4(D) show the resultant images at depth 1, whereas Figs. 4(a)–4(d) show images at depth 2. These images clearly show that we can achieve good out-of-focus background rejection of multiple fluorescent planes using a 2P-VHI system for biological samples. We show a proof of principle experiment by presenting two depth discrimination images by our system. By increasing the number of layers in the AMVHGs, we can acquire multiple depth images with high contrast even for high scattering tissues. Our system provides the advantage of inherent VHI and the fine high-contrast optical sectioning of 2P excitation.

Fig. 4

Two-depth resolved optical-sectioning images of fluorescently labeled mice intestine sample obtained using the 2P-VHI system. (A)–(D) Detailed resultant images at selected locations of the intestine sample in depth 1. (a)–(d) Detailed images of the sample in depth 2. Two planes are separated by $35\mu \mathrm{m}$.