Journal of Optics

Related content

PAPER

Simultaneous forward and backward digital holography of microparticles To cite this article: Nava R Subedi et al 2017 J. Opt. 19 115601

View the article online for updates and enhancements.

- Holographic particle image velocimetry: from film to digital recording Hui Meng, Gang Pan, Ye Pu et al. - Synthetic aperture common-path digital holographic microscopy with spiral phase filter Han-Yen Tu, Wei-Jen Hsiao, Xin-Ji Lai et al. - Digital in-line holography with photons and electrons H J Kreuzer, M J Jericho, I A Meinertzhagen et al.

This content was downloaded from IP address 129.130.106.65 on 09/10/2017 at 13:11

Journal of Optics J. Opt. 19 (2017) 115601 (9pp)

https://doi.org/10.1088/2040-8986/aa8a80

Simultaneous forward and backward digital holography of microparticles Nava R Subedi1, Matthew J Berg2 and Gorden Videen3 1

Department of Physics & Astronomy, Mississippi State University, 355 Lee Blvd., Mississippi State, MS 39762-5167, United States of America 2 Department of Physics, Kansas State University, 1228 N. 17th Street, Manhattan, KS 66506-2601, United States of America 3 US Army Research Laboratory, RDRL-CIE-S, 2800 Powder Mill Road, Adelphi, MD 20783-1197, United States of America E-mail: [email protected] Received 17 May 2017, revised 1 August 2017 Accepted for publication 6 September 2017 Published 28 September 2017 Abstract

This work explores several techniques in digital holography to image 10–300 μm sized particles and provide information useful for their morphological characterization. In particular, digital holograms are formed with both forward and backward-scattered light from samples fixed to a glass stage. Images of these particles are then rendered from the holograms demonstrating that front and back perspectives of a particle can be acquired simultaneously. In addition, holography’s large depth of field is used to investigate the potential for stand-off interrogation of particles of different types. Keywords: digital holography, coherence imaging, remote sensing and sensors, backscattering, microscopy, image analysis (Some figures may appear in colour only in the online journal) With digital holography (DH), the particle’s size, shape, and orientation may be known unambiguously provided that the image resolution of the sensor is sufficient. Yet, holography is not only useful for imaging. Recent work shows that DH is a potential tool to perform various measurements simultaneously, such as imaging [9–11] and measuring optical observables (scattering pattern, total cross sections, and single-scattering albedo) [12, 13]. Enabling these measurements in a single experimental setup using a particle’s scattered light in different directions, as opposed to forward scattered light as is more commonly done, is the objective of this work. The intent of this article is to demonstrate the feasibility of a new concept where DH can be extended for various measurements simultaneously rather than to develop a new method for practical use. In DH, a particle’s scattered light is allowed to interfere with unscattered light across a 2D sensor [14–16]. The resulting microscopically fine interference pattern, which is the hologram, is then computationally processed to produce a silhouette-like image of the particle. This work shows how DH can image particles via scattered light from different

1. Introduction Micron-sized particles are ubiquitous in nature and influence properties of the system in which they reside. However, the lack of understanding about the physical properties of these particles greatly limits our ability to understand, predict, and control their applications and impacts in the system [1]. Characterization of particles is important in many scientific and engineering applications, especially in climate science, particle imaging and tracking, water quality assessment, and life sciences [2–6]. Determining the size, shape, and surface texture of these particles is also important for agricultural and public-health concerns [7]. The recent progress in technology makes it possible to characterize these particles and study their properties in controlled environments, but the contactfree characterization of such particles is still challenging. Often, this must be done in a contact-free manner as the collection of samples for microscope-based characterization may distort the true morphology (e.g. consider liquid or frozen particles) [8]. 2040-8978/17/115601+09$33.00

1

© 2017 IOP Publishing Ltd Printed in the UK

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 1. Diagram of the first experimental configuration in which forward and backward scattering holograms are recorded simultaneously. In the diagram, L1 and L2 are lenses, PH is a pin-hole, BS is a beamsplitter, ND is a neutral density filter, and M1 is a silver mirror.

directions, specifically in the near-forward and near-backward directions.

In the conventional digital in-line holographic arrangement, i.e. that using the forward scattered light, the particle image is accompanied by a so-called d.c. term and its out-offocus twin image upon reconstruction [28–30]. These features are also present in the backward scattering arrangement and can degrade the particle image quality. The d.c. term impacts the image by affecting the dynamic range of the display due to its high intensity compared to the real image, whereas the twin image reduces the size of the real image area in the total reconstructed field [16]. However, in this work these terms are largely suppressed using the method of subtracting the average intensity of the reference from each pixel in the particle’s hologram intensity as described in [18, 31, 32].

2. Recording and reconstruction A pulsed laser is used to illuminate micron-sized particles deposited on an anti-reflection coated glass window, which will be called the particle ‘stage’ below. During illumination, a small portion of the incident light undergoes scattering from the particle and a so-called object wave is created. This wave contains information about the particle [15–17]. The remaining portion of the wave, which passes by the particle mostly unscattered, is called the reference wave. These waves interfere across a 2D CCD sensor to produce the digital hologram, I holo and corresponds to the in-line configuration. From I holo, the particle-free background is subtracted to yield a contrast hologram, I con [9, 14, 18]. This removes imperfections in the illumination-beam profile, improving the subsequent particle-image, which can be seen by the ‘cleaner’ interference pattern in figure 4(b) below. In the image reconstruction, the contrast hologram is imagined as a transmission diffraction grating illuminated at normal incidence by the reference wave. This is equally true for both forward and backward-scattered holograms, although in the latter case, it is more natural to think of this as scattering from the grating [19–25]. The Fresnel–Kirchhoff diffraction theory described in [26, 27] is then used to approximately describe this diffraction to form the image on a parallel ‘reconstruction’ plane separated by a distance z from the hologram plane: K (x , y , z ) = a

3. Experiment The experimental arrangements shown in figures 1 and 2 facilitate the recording of a particle’s forward and backward holograms simultaneously, while figure 3 records only the backward hologram. The arrangement is initially established by constructing a Michelson interferometer so that proper alignment of the optical components is achieved. This ensures that the two beams interfere constructively on the sensors when no particle is present. The optical source is a Nd:YFL laser (Photonics Industries, DC50-351), frequency tripled to 351 nm wavelength. The sensors used are the Finger Lakes Instrumentation Inc., (ML8300) and PointGrey Research Inc., (CMLN-13S2M) CCD sensors. These are called ‘sensor-1’ and ‘sensor-2’ below, respectively. At a distance of 19 cm from the laser is a fused silica lens (L1) of focal length 5 cm. At the focal point of lens L1 a pin-hole of 25 μm diameter is used to produce a circular diffraction pattern, after which, an iris selects only the central diffraction peak and removes stray light from the laser beam. Another fused silica lens (L2) with focal length 30 cm then collimates the central diffractionpeak. This spatial filter produces a clean beam-profile. Following L2, a pellicle beamsplitter (BS) of diameter 50.8 cm with reflection-to-transmission ratio 45:55 splits the beam. This BS is 9 cm from L2 and is mounted on a translation stage

⎧ ik

òò I con (x, h) exp ⎨⎩ 2z [(x - x )2 s

⎫ + ( y - h )2 ]⎬ dx dh. ⎭

(1 )

In equation (1), a is a constant, k = 2p l is the wave number, l is the wavelength of the light used to illuminate the particle, and I con (x , h ) is the contrast hologram. Evaluating ∣K ∣2 in the image plane yields the particle image. 2

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 2. Diagram of the second experimental configuration in which forward and backward scattering holograms are recorded

simultaneously.

Figure 3. Diagram of the experimental arrangement for multiple-sample investigation in the backward-scattering configuration. Here, red and green are used to distinguish the two different particle samples and their corresponding backward-scattered waves. Yet, only one color of light is used, i.e. the 351 nm Nd:YLF light. In this setup, NaCl particle sample and 50 μm diameter glass spheres sample are placed at 6.5 cm and 11 cm from the sensor, respectively.

to ease its alignment. Then, silver mirrors M1 and M2 are each placed at 50 cm from the BS. Two different configurations are then considered. In the first, mirror M1 is fixed and mirror M2 is mounted on a translation stage. The position of M2 is then adjusted until an interference maximum is seen on sensor-1, which is 4.5 cm

from the BS. Following alignment, mirror M2 is removed and sensor-2 is installed at 8.5 cm from the BS. This configuration now allows the recording of both forward- and backwardscattering holograms simultaneously. However, back reflection from the particle stage and cover glass on sensor-2 overlaps with the particle’s backscattered wave producing 3

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 4. Holographic image validation. Plots (a) and (b) show the measured raw and contrast holograms due to forward-scattered light,

respectively, for a ragweed pollen cluster. Image (c) shows the reconstructed image resulting from (b) via equation (1), whereas (d) shows the conventional microscope image of the same cluster at ×20. The image in the bottom right corner of (d) shows the SEM image of single ragweed pollen to compare how it looks like. Note that the holograms shown do not display the full sensor output; they are cropped to show only the portion of the hologram considered in equation (1).

object wave. To prevent this and make the intensity of the reference wave nearly equal to the intensity of the particle’s backward scattering wave, a neutral density filter (ND=0.3) is placed in front of the mirror. Also, the two sensors used in this experiment do not have equal dynamic range: 64.4 dB for sensor-1 and 56.77 dB for sensor-2. Thus, to avoid saturation, a neutral density filter (ND=0.3) is placed in front of sensor2. Following the description above, a particle stage is installed at 6.5 cm from the sensor-2. The concept of backward DH is then extended for multiple-sample investigation simultaneously as shown in figure 3. Here, two particle stages are installed at different distances from the BS in front of M1 and M2 during alignment. Then the mirrors are replaced by beam dumps. Because the particle stages are at different distances from the sensor,

noise in the backward-scattering hologram, i.e. on sensor-1. This noise deteriorates the resolution of the reconstructed image. In the next arrangement, the same procedure is followed as above, but M1 is replaced by sensor-2 such that it is at 8.5 cm from the BS as shown in figure 2. In this arrangement, multiple reflections between the two sensors also produce background noise in the holograms, and hence, impact the reconstructed image quality. However, the method of subtracting the average intensity of the reference from each pixel in the particle’s hologram intensity helps to reduce the noise to some extent. In both configurations, the intensity of the reference wave reflected from the mirror is stronger than that of the particle’s backward scattering wave. This suppresses the intensity of the 4

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 5. Images (a) and (b) show the contrast holograms for Pecan pollen particles using, respectively, forward and backward-scattered light

acquired with the setup in figure 1. Images (c) and (d) show the corresponding reconstructed holographic images. The red circles highlight regions in the holograms which are considered in the image reconstructions, i.e. corresponding to the particles shown in (c) and (d).

holographic imaging. An example is presented in figure 4 demonstrating a comparison between the microscope and holographic images of the same ragweed pollen cluster. This is done using the forward-scattered hologram in figure 1. By comparing the images, it can be seen that the holographic process successfully produces an accurate image of the cluster with sufficient resolution to discern individual pollen particles. This comparison also establishes an image-pixel to micrometer mapping so that the particle images can be shown with true physical dimensions. Following image calibration, an estimate for the resolution of holographic image is performed using National Institute of Standards & Technology-traceable borosilicate glass microspheres (Duke Scientific Corp.). The resolution limit of the holographic image is estimated by comparing the

each particle sample can be brought into focus during application of equation (1) after-the-fact from a single hologram measurement. In this arrangement, it looks like the reference wave is missing but that is not the case in reality. Here, a portion of the wave incident on the stage gets reflected before scattering from the particle and acts as the reference wave.

4. Results To calibrate the imaging process, a comparison is made between holographic and conventional optical microscope images of the same particle. This is done by placing a ragweed pollen cluster on the stage and imaging it first with the microscope. It is then transferred to the experiment for

5

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 6. Images (a) and (b) show the contrast holograms for Mississippi road dust particle using, respectively, forward and backward-

scattered light acquired with the setup in figure 1. Images (c) and (d) show the corresponding reconstructed holographic images. The red ovals highlight regions in the image discussed in the text below.

holographic images with microscope images of microspheres of various sizes (10–50 μm in diameter). Particles smaller than 10 μm could not be imaged well enough to ensure that what appeared like a single particle was not in fact a cluster. Thus, the estimated resolution is approximately 10 μm. However, the intrinsic resolution of the experimental setup is given by the relation dr = lz N D, as described in [26, 27], where z is the reconstruction distance, N is the number of pixels in the sensor array, and Δ is the pixel distance. In our case, l = 351 nm, N=500 pixels, Δ=5.4 μm, and zi » 6.5 cm; therefore the approximate resolution limit of the reconstructed image is dr » 8.45 m m. In DH, the resolution of the image is greatly limited by the size of image sensor and the estimated resolution is probably still poorer than the intrinsic resolution because of the limited dynamic range of

the sensor. Even if the whole sensor chip is exposed to the hologram interference pattern, the ‘edges’ will be so dim that the sensor does not resolve them. This essentially reduces the size of the sensor. The prevailing solution to this problem is to employ iterative algorithms to enhance the resolution of holographic images and increase effective size of image sensor as described in [33, 34]. The images reconstructed from holograms acquired with the setups shown in figures 1 and 2 are presented in figures 5– 7. Close examination in figures 6 and 7 shows that the forward and backward-scattering images yield different information about the particle’s surface and edge roughness. For example, images (c) and (d) in figure 6 show significant differences in the ability to resolve the ‘edge’ of the particle, whereas figure 7 shows noteworthy variation of particle-surface roughness. 6

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 7. Images (a) and (b) show the holographic images for clay derived from the forward- and backward-scattering holograms,

respectively. Images (c) and (d) show the same for NaCl particles. The red circles highlight regions in the image discussed in the text below.

5. Comments

These differences show up in a side-by-side comparison of the two images, explicitly in the highlighted regions 1 and 1′, and 2 and 2′, respectively. In figure 6(d), there is a ghost-like appearance in the backward-scattering image. This is probably due to multiple reflections from the glass window of the particle stage. The reconstructed images of two different particle samples from a single hologram acquired with the setup in figure 3 are presented in figure 8. This shows that different particle samples can be imaged simultaneously at different focus distances. This could be useful for so-called stand-off investigation of particles. In short, the stand-off modality involves sensing particles that are not brought into contact with the apparatus. The results here show that the large depthof-field in DH allows one to form images of particles separated from the apparatus over a working distance of several centimeters.

One motivation for this work relates to the imaging of single and multiple particles from holograms derived from scattered light from different directions. In traditional in-line holographic configurations, the particles’ forward-scattered light is usually used to form the hologram, thus opaque particles yield only a silhouette-image after reconstruction and the particle surface characteristics are obscured. We find the same is true here, yet some detail can be seen in the forward and backward-scattered holographic images that differ unambiguously. This suggests that the two image perspectives reveal aspects of the particle structure not available from a forward-scattering based approach alone. Much work is reported in the literature about objectsurface texture measurement with DH. Recent work in [35] uses DH for this purpose, but considers the off-axis 7

J. Opt. 19 (2017) 115601

N R Subedi et al

Figure 8. Backward-scattering hologram and reconstructed images of multiple samples. Image (a) shows the contrast hologram, while (b) the

holographic image for NaCl particles when the reconstruction distance z is selected to focus this particle image. Image (c) shows an out of focus image of both samples seen when scanning z in equation (1), and image (d) shows the holographic image for 50 μm diameter glass spheres when z is selected to focus this particle’s image. In this setup, NaCl particle sample and 50 μm diameter glass spheres sample are placed at 6.5 cm and 11 cm from the sensor, respectively.

configuration. Other work in [36, 37] use the concept to analyze the surface roughness on metrological instruments and measure the surface topography from different perspectives, respectively, but both consider much larger object, i.e. not microparticles. Also, work in [38] proposes an application to probe the beam path and its corresponding structure in plasmonic phenomena. For optically large objects, concepts in geometric optics such as reflection and transmission are useful. Here, however, these concepts do not apply directly as the particles are much closer to the wavelength in size than for optically large objects. Thus, the DH concept is best understood in terms of forward and backward scattering [18, 25, 39].

forward-scattered light only. The simultaneous investigation of multiple samples is also presented in line with a stand-off implementation. These images are computationally reconstructed from the recorded holograms and compare well in quality and accuracy to the corresponding microscope images up to the resolution limit of the sensor. Even though this resolution is less than that from the microscope, the size and shape of a single particle as small as 10 μm in size is clearly distinguishable. The resolution could be improved using a sensor with greater dynamic range, by illuminating the particle with shorter wavelength light, and using a positive lens to magnify the particle’s hologram during recording. Also, replacing the particle stage with a flowing aerosol stream will help to eliminate a large degree of noise in first two configurations that can result from ambient dust that collects on the optical surface. The results of this work for backward scattering light may be useful for studying opaque particles and those that must be examined in a contact-free manner, including liquid or frozen aerosol particles.

6. Conclusion This work explores the feasibility of imaging single and multiple microparticles with DH using forward and backward-scattered light rather than the more commonly used

8

J. Opt. 19 (2017) 115601

N R Subedi et al

[17] Ackermann G K and Eichler J 2007 Holography: A Practical Approach (New York: Wiley) [18] Kreis T 2005 Handbook of Holographic Interferometry (New York: Wiley) [19] Darakis E, Khanam T, Rajendran A, Kariwala V, Naughton T J and Asundi A K 2010 Microparticle characterization using digital holography Chem. Eng. Sci. 65 1037–44 [20] Kim M K 2011 Digital Holographic Microscopy: Principles, Techniques, and Applications (Berlin: Springer) [21] Xu W, Jericho M H, Meinertzhagen I A and Kreuzer H J 2002 Digital in-line holography of microspheres Appl. Opt. 41 5367–75 [22] Garcia-Sucerquia J, Xu W, Jericho S K, Klages P, Jericho M H and Kreuzer H J 2006 Digital in-line holographic microscopy Appl. Opt. 45 836–50 [23] Mico V, Ferreira C, Zalevsky Z and Garcia J 2010 Basic principles and applications of digital holography microscopy Microscopy: Science, Technology, Applications and Education (Badajoz: Formatex) [24] Xu L, Peng X, Guo Z, Miao J and Asundi A 2005 Imaging analysis of digital holography Opt. Express 13 2444–52 [25] Subedi N R, Fowler N B and Berg M J 2013 Backscatter digital holography of microparticles Opt. Express 21 12611–6 [26] Schnars U and Jueptner W 2002 Digital recording and numerical reconstruction of holograms Meas. Sci. Technol. 13 R85–101 [27] Schnars U and Jueptner W 2005 Digital Holography: Digital Hologram Recording, Numerical Reconstruction, and Related Techniques (Berlin: Springer) [28] Latychevskaia T and Fink H W 2007 Solution to the twin image problem in holography Phys. Rev. Lett. 98 233901 [29] Yamaguchi I and Zhang T 1997 Phase-shifting digital holography Opt. Lett. 22 1268–70 [30] Cuche E, Marquet P and Depeursinge C 2000 Spatial filtering for zero-order and twin-image elimination in digital off-axis holography Appl. Opt. 39 4070–5 [31] Takaki Y, Kawai H and Ohzu H 1999 Hybrid holographic microscopy free of conjugate and zero-order images Appl. Opt. 38 4990–6 [32] Kreis T and Jüptner W 1997 Suppression of the dc term in digital holography Opt. Eng. 36 2357–60 [33] Zhang S and Zhou J 2014 IEEE Xplore (https://doi.org/ 10.1109/ICNC.2014.6975966) [34] Latychevskaia T and Fink H-W 2013 Resolution enhancement in digital holography by self-extrapolation of holograms Opt. Express 21 7726–33 [35] Groot P 2015 Principles of interference microscopy for the measurement of surface topography Adv. Opt. Photonics 7 1–65 [36] Zhou W J, Peng K and Yu Y 2016 Surface roughness measurement and analysis of mechanical parts based on digital hologram Advances in Manufacturing (Berlin: Springer) pp 1–8 [37] Leach R 2011 Optical Measurement of Surface Topography (Berlin: Springer) [38] Lim Y, Lee S-Y and Lee B 2011 Transflective digital holographic microscopy and its use for probing plasmonic light beaming Opt. Express 19 5202–12 [39] Heffernan B M, Heinson Y W, Maughan J B, Chakrabarti A and Sorensen C M 2016 Backscattering measurements of micron-sized spherical particles Appl. Opt. 55 3214–8

Acknowledgments The authors are greatly thankful to Ben Ardahl and Shane Clark for their assistance with electronic instrumentation and comments from two anonymous reviewers. This work is supported by the US Army Research Office, contract W911NF-15-1-0549. References [1] National Research Council (US) Chemical Sciences Roundtable 2012 Challenges in characterizing small particles: exploring particles from the nano- to microscale: a workshop summary What Are Small Particles and Why Are They Important? (Washington, DC) (National Academics Press) (http://ncbi.nlm.nih.gov/books/NBK98070/) [2] Kulkarni P, Baron P A and Willeke K (ed) 2011 Aerosol Measurement: Principles, Techniques, and Applications (New York: Wiley) [3] Dusek U et al 2006 Size matters more than chemistry for cloud-nucleating ability of aerosol particles Science 312 1375–8 [4] Subedi N R 2016 Characterization of Microparticles Through Digital Holography Doctoral Dissertation Mississippi State University [5] Mehra I, Singh K, Agarwal A K, Gopinathan U and Nishchal N K 2015 Encrypting digital hologram of threedimensional object using diffractive imaging J. Opt. 17 035707 [6] Pitkäaho T, Pitkäkangas V, Niemelä M, Rajput S K, Nishchal N K and Naughton T J 2017 Digital holographic microscopy in remote potable water monitoring Digital Holography and Three-Dimensional Imaging (Optical Society of America) OSA Technical Digest W2A.44 [7] Aptowicz K B, Pinnick R G, Hill S C, Pan Y-L and Chang R K 2006 Optical scattering patterns from single urban aerosol particles at Adelphi, Maryland, USA: a classification relating to particle morphologies J. Geophys. Res. 111 D12212 [8] Berg M J and Holler S 2016 Simultaneous holographic imaging and light-scattering pattern measurement of individual microparticles Opt. Lett. 41 3363–6 [9] Berg M J and Videen G 2011 Digital holographic imaging of aerosol particles in flight J. Quant. Spectrosc. Radiat. Transfer 112 1776–83 [10] Kanka M, Riesenberg R and Kreuzer H J 2009 Reconstruction of high-resolution holographic microscopic images Opt. Lett. 34 1162–4 [11] Kreuzer H J, Pomerleau N, Blagrave K and Jericho M H 1999 Digital in-line holography with numerical reconstruction Proc. SPIE 3744 65–74 [12] Berg M J, Subedi N R, Anderson P A and Fowler N B 2014 Using holography to measure extinction Opt. Lett. 39 3993–6 [13] Berg M J, Subedi N R and Anderson P A 2017 Measuring extinction with digital holography: nonspherical particles and experimental validation Opt. Lett. 42 1011–4 [14] Berg M J and Subedi N R 2015 Holographic interferometry for aerosol particle characterization J. Quant. Spectrosc. Radiat. Transfer 150 36–41 [15] Gabor D 1948 A new microscopic principle Nature 161 777–8 [16] Gabor D 1979 Microscopy by reconstructed wavefronts Proc. R. Soc. A 197 454–87

9