Invisibility cloak with image projection capability

Abstract

Investigations of invisibility cloaks have been led by rigorous theories and such cloak structures, in general, require extreme material parameters. Consequently, it is challenging to realize them, particularly in the full visible region. Due to the insensitivity of human eyes to the polarization and phase of light, cloaking a large object in the full visible region has been recently realized by a simplified theory. Here, we experimentally demonstrate a device concept where a large object can be concealed in a cloak structure and at the same time any images can be projected through it by utilizing a distinctively different approach; the cloaking via one polarization and the image projection via the other orthogonal polarization. Our device structure consists of commercially available optical components such as polarizers and mirrors, and therefore, provides a significant further step towards practical application scenarios such as transparent devices and see-through displays.

Full text

1
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
www.nature.com/scientificreports
Invisibility cloak with image
projection capability
Debasish Banerjee1, Chengang Ji1,2 & Hideo Iizuka3
Investigations of invisibility cloaks have been led by rigorous theories and such cloak structures,
in general, require extreme material parameters. Consequently, it is challenging to realize them,
particularly in the full visible region. Due to the insensitivity of human eyes to the polarization and
phase of light, cloaking a large object in the full visible region has been recently realized by a simplied
theory. Here, we experimentally demonstrate a device concept where a large object can be concealed in
a cloak structure and at the same time any images can be projected through it by utilizing a distinctively
dierent approach; the cloaking via one polarization and the image projection via the other orthogonal
polarization. Our device structure consists of commercially available optical components such as
polarizers and mirrors, and therefore, provides a signicant further step towards practical application
scenarios such as transparent devices and see-through displays.
Since the pioneering work1 that led to the realization of electromagnetic metamaterials over a decade ago, invisi-
bility cloaking of an object to incoming waves has been one of the most appealing possibilities in the metamaterial
research community. A scattering cancellation method has been presented for cloaking a subwavelength object2.
Transformation optics and conformal mapping1,3–5 can theoretically hide large objects. Following these funda-
mental concepts, early works have demonstrated cloaking technologies, mostly in narrowband of microwave fre-
quencies6,7. By reducing to a two-dimensional coordinate transformation, a quasi-conformal mapping approach
called ‘carpet cloaking’8 has been realized in the microwave9, the near infrared10–12, and the visible regions13. For
an ideal cloaking operation these rigorous theories predict the requirement of extreme material parameters and
spatial anisotropy of cloaking media. More recently, an optical metasurface cloak to a laser illumination has been
realized by manipulating the reection phase distribution14. It is, in general, dicult to implement cloak struc-
tures, particularly in the full visible spectrum. us, hiding large 3D objects in the full visible wavelength range
for practical applications remains a decade-old challenge.
In recent years, it has been possible to fabricate broadband invisibility cloaks in the full visible region when the
omnidirectionality requirement is abandoned. In these cases, very promising results have been obtained when the
transformation optics was simplied by removing the design restriction of phase preservation. A polygonal cloak
structure made of calcite, which is a natural anisotropic material, was shown to hide a cylindrical object with a
diameter of a few millimeters15, following the study of calcite carpet cloaks16,17. In ref. 18, a large object was con-
cealed in normal incidence by using a cloak structure consisting of isotropic material lenses. A number of authors
have reported various cloak structures consisting of conventional optical components in the visible region19–23.
On the other hand, there is a tremendous amount of information such as texts, images, and movies in our
daily lives owing to the signicant achievements in the image projection technology24–26. Modern human vision
systems such as 3D displays rely on several psychological cues where overlapping of images, shading, textured
gradients are used to create various depth perceptions in the human brain. However, these image-based devices
are oen associated with ambiguities and errors because of the inability to correctly carry the depth prole infor-
mation. It is suggested in ref. 24 that in order to achieve the perfect 3D display technology, some novel methods
to generate physiological cues such as binocular display, convergence, and accommodation are necessary, which
to be combined in the imaging system with existing psychological cues described earlier. To create physiologi-
cal cues, modern displays oen utilize optical elements such as lenticular lenses, parallax barriers, curved lens
arrays, holography, and see-through display technologies. However, the aspect of optical invisibility cloaking in
such imaging systems has never been explored. For example, the combination of the cloaking with the imaging
systems can allow an additional degree of freedom for the physiological cues such as used for the realization of
1Toyota Research Institute of North America, Toyota Motor North America, Ann Arbor, MI 48105, USA. 2Department
of Electrical Engineering & Computer Science, University of Michigan, Ann Arbor, MI 48109, USA. 3Toyota Central
Research & Development Labs., Nagakute, Aichi 480 1192, Japan. Correspondence and requests for materials should
be addressed to D.B. (email: debasish.banerjee@toyota.com) or H.I. (email: hiizuka@mosk.tytlabs.co.jp)
Received: 03 October 2016
Accepted: 15 November 2016
Published: 13 December 2016
OPEN

www.nature.com/scientificreports/
2
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
see-through 3D displays and augmented reality technologies27,28 that could be benecial to the thrust for trans-
parent electronics in the future.
In this Letter, we experimentally demonstrate an invisibility cloak where a large object can be concealed in the
cloak structure and any images can be projected on it simultaneously. is unique functionality can be obtained
through the use of one polarization of light for the cloaking and the other orthogonal polarization for the image
projection due to the insensitivity of human eyes to the polarization. Such approach has never been captured by
any other reports in the past.
Results
Mechanism of our invisibility cloak with the image projection capability. Our cloak structure
consists of commercially available optical components; polarizers for oblique incidence (Po,1-4) and for normal
incidence (Pn,1), and mirrors (M1-4), as shown in Fig.1(a). We regard our structure as a twelve-port device and
each port is labeled by 1 to 12. In the description of the mechanism of our structure below, we consider that the
scene behind the cloak structure is built at the observation point in front of the cloak structure, assuming ideal
performance of each component. e regions of the scene are labeled by I to IV, and here we assume that regions I
and IV (II and III) are observed via the horizontally (vertically) polarized light. Due to the symmetry of the cloak
structure, we consider here the half of the structure for the explanation of the cloak mechanism. To illustrate the
Figure 1. (a) Conguration of a cloak structure having the image projection capability. e structure consists
of polarizers for oblique incidence (Po,1-Po,4), mirrors (M1-M4), and a polarizer for normal incidence (Pn,1),
where port 4 has the polarizer Pn,1, and port 9 does not have a polarizer for comparison. Labels of transmission
coecients of (b) a polarizer for oblique incidence, (c) a polarizer for normal incidence, and (d) a mirror for the
use in Table1. (e) Experimental observation of the cloak structure. e inset shows the experimental setup. A
toy car is placed behind the cloak structure and a cylindrical object is partly inserted in the cloak structure. e
white number “1” or “2” printed on a black paper reversely in the le and right is attached on each lateral side,
port 4 or port 9.

www.nature.com/scientificreports/
3
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
image projection capability, one lateral side (port 4) has the polarizer Pn,1 and the other lateral side (port 9) does
not have a polarizer for comparison.
Consider rst the unpolarized light traveling in the − y direction from region II of the scene behind the
cloak structure in Fig.1(a). e light is reected by mirror M1 into the + x direction. Here we assume that at
ideal obliquely incident polarizers Po1-4, the vertically polarized light is bent by 90° (Tpo,bn,V = 1 and Tpo,st,V = 0 in
Fig.1(b)) and the horizontally polarized light passes straight through (Tpo,bn,H = 0 and Tpo,st,H = 1 in Fig.1(b)).
us, the vertically polarized light is bent twice at Po,1 and Po,2 and then comes out at port 6 via the reection at
mirror M2. Likewise, for the unpolarized light coming from region I, the horizontally polarized light goes through
the two polarizers Po,1 and Po,2 and comes out at port 5. Other orthogonal polarized lights coming from areas I and
II go out at port 3 (light paths are not presented). In other words, the scene behind the cloak structure is built at
the observation point through the use of 50% incidence.
In our structure, the capacity for another 50% light is used for the image projection. e information sources “1”
and “2” are placed at ports 4 and 9, respectively, which are reversed in the le and right due to the mirror reection.
Consider the light paths from port 9. e horizontally polarized light goes through polarizer Po,3 and is reected
by mirror M3 and then goes out at port 7 while the vertically polarized light comes out at port 8 by a 90° bent
at the polarizer Po,3. As a result, we observe “2” in both regions III and IV. On the other hand, from port 4, we
observe “1” in only region I by inserting polarizer Pn,1 that allows the transmission of the vertically polarized light.
erefore, the information can be projected in any region(s) as desired. Table1 shows the summary of light paths
with labels of ideal components dened in Fig.1(b–d). We comprehend the mechanisms of the cloaking and the
image projection presented above.
Experimental demonstration of the invisibility cloak with the image projection capability. e
experimental setup is shown in the inset of Fig.1(e). Our cloak was implemented with four wire-grid polarizer
cubes (89–604, Edmund Optics) and four right angle mirrors (45–595, Edmund Optics). White numbers “1” and
“2” are printed on black papers reversely in the le and right, respectively, and these papers are placed at lateral
sides. A wire-grid polarizer lm (47–102, Edmund optics) is attached on the black paper having “1”, where the
vertically polarized light passes through. A cylindrical object is placed within the cloak structure, where the cylin-
drical object is partly in the concealed region while remaining upper part is exposed outside the cloak structure.
ere is a toy car behind the cloak structure and we capture screen shots through a camera in front of the cloak
structure. Figure1(e) shows an experimental observation of the cloak structure from the camera. e cylindrical
object becomes invisible in the cloaking area and the car behind the cloak structure is observed. We observe “1”
in region I and “2” in regions III and IV, as we designed. erefore, the mechanism of our invisibility cloak with
the image projection capability has been experimentally veried for human eyes.
Performance of the image projection capability. We further investigate the image projection ability,
particularly, the placement of the image projection. We show that an image can be projected at the middle of
neighboring regions as well as each region. As expected from the experimental demonstration of Fig.1(e), the
text “CLOAK” appears at region I (Fig.2(b)) by selecting the vertical polarization (Fig.2(a)), which is directly
reected by the side polarizer Po2, and appears at region II (Fig.2(e)) with the horizontally polarized light going
through the polarizer Po2 and being reected by the mirror M2 instead (Fig.2(d)). In this experimental demon-
stration, the wire-grid polarizer lm (47–102, Edmund optics) was rotated by 90° for the polarization change. e
text “CLOAK” can be projected at the middle of regions I and II (Fig.2(h)) via the horizontally polarized light
for “CL” and the vertically polarized light for “OAK” (Fig.2(g)). erefore, proper use of polarizers for an image
enables the projection in any places on the cloak structure while the invisibility cloak phenomenon is maintained
(Fig.2(c,f,i)).
View angle dependence. e dependence of our structure on the view angle is experimentally and numer-
ically investigated and shown in Fig.3. To illustrate the main reason for the degradation of the rebuilt scene with
the increase of the view angle, the cloak structure now consists of polarizers and mirrors that are attached on glass
plates (Fig.3(a)), instead of glass blocks. is allows us to envision what happens in the cloak structure by optical
Input port
1 2 3 4 5 6
Output port
1 — 0 Tpo,st,jTm,j (0,1) 0Tpo,bn,jTpo,st,jTm,j (0,0) Tpo,bn,j2Tm,j2 (1,0)
2 0 — Tpo,bn,j (1,0) 0Tpo,st,j2 (0,1) Tpo,st,jTpo,bn,jTm,j (0,0)
3Tpo,st,jTm,j (0,1) Tpo,bn,j (1,0) — 0 0 0
4 0 0 0 — Tpo,bn,jTpn,j (1/0,0) Tpo,st,jTpn,jTm,j (0,1/0)
5Tpo,bn,jTpo,st,jTm,j (0,0) Tpo,st,j2 (0,1) 0 Tpo,bn,jTpn,j (1/0,0) — 0
6Tpo,bn,j2Tm,j2 (1,0) Tpo,st,jTpo,bn,jTm,j (0,0) 0Tpo,st,jTpn,jTm,j (0,1/0) 0 —
Table 1. Transmission coecients of the cloak structure of Fig.1(a). e cloak structure is regarded as a
twelve-port device and transmission coecients for six ports are presented due to the mirror symmetry of the
structure. Each coecient in the table is dened in Fig.1(b–d). Parentheses (Tj=V,Tj=H) represent transmittances
for vertical polarization and horizontal polarization, assuming ideal performance of each component (Tm,j = 1,
Tpo,st,j = 1 or 0, Tpo,bn,j = 1 or 0, and Tpo,st,jTpo,bn,j = 0). 1/0 represents that the transmittance can be changed by
polarized light.

www.nature.com/scientificreports/
4
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
paths clearly without the eect of the refractive index of glass. e half of the structure is used for this investiga-
tion due to the symmetry of our cloak structure; the scene including a car has the same size as the summation
of regions I and II. Optical paths from the original scene are calculated and the scene rebuilt at the observation
plane is obtained (Fig.3(b)) by using the commercial optical design soware Zemax29. e view angle has been
set at 0°, 2°, and 4°, respectively. In the normal direction, the rebuilt scene is same as the original scene except for
the half of the brightness (Fig.3(d)) due to the use of 50% light (Fig.3(e)). As the view angle increases, the rebuilt
image is degraded manifesting as the lack of the information (black lines in Fig.3(g,j)) and the wrong position of
the information (the middle part of the red car appears at the le edge in Fig.3(g,j)), respectively. ese can be
analyzed and well understood from optical paths tracing as shown in Fig.3(e,h,k) for dierent incident angles.
Taking the 4° incidence for example, light path A-B (yellow color ray) represents one source of image information
lacking as the light ray is reected to the neighboring wire-grid polarizer Po2 directly at the oblique incidence.
On the other hand, the wrong position of the output image information is credited to the rays near F-G-H path
(highlighted as the cyan color), where unpolarized light is directly incident onto the wire-grid polarizer at large
angles of incidence and then split into two dierent paths with one path going to the correct position while the
other to the le edge. It is worthwhile to mention that there is another source represented by and near C-D-E (red
color) for both the two image degradation results. is portion of light is completely misled to the output edge
leaving a black line at the center. Note that the lack and the wrong position of the information mentioned above
are consistent with the cloak structure of Fig.1(e).
Discussion
In recent times, modern electronic devices such as cellphones, smart glasses, and computers are desired to be
transparent30. e need for transparent objects with the display function is not limited to personal electron-
ics. Tremendous applications have also been envisioned from the automotive industry, household appliances to
healthcare sector. Typically, two approaches are adopted to realize transparent devices; rstly, a quest for materials
with the unusual combination of high electrical conductivity and high optical transparency such as Indium Tin
Oxide, and the composites of carbon nanotubes, graphene, and silver nanowires31. Despite the recent signicant
progress, mass application of these materials has remained challenging. On the other hand, driven by the desire
to develop augmented reality experience, transparent or see-though displays are aimed to project virtual images
of 2D or 3D objects utilizing optical components such as wedge-shaped prisms, spherical mirrors, array of lenses,
etc.32. Holography based optical elements have also been used33. However, as we discussed before, optical cloaking
Figure 2. Image projection ability. e text “CLOAK” appears in region I for (a–c), in region II for (d–f), and
in the middle of regions I and II for (g–i). Light paths (a,d,g), the image projection (b,e,h), and the cloaking
(c,f,i) are presented.

www.nature.com/scientificreports/
5
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
approaches combined with the imaging systems have not been explored in the past. In our experiment, we have
used printed papers as information sources for the experimental demonstrations in order to verify the mecha-
nism and architecture of the display aspect in combination with cloaking the central part of the device. In prac-
tical applications, one can implement two thin-type displays having electrically switchable polarizers at ports 4
and 9 and project text messages and movies at any locations of the cloak structure. We believe that combining the
invisibility cloaking with the image projection capability promises an alternative route to realize next generation
transparent devices.
In conclusion, we have explored a cloak structure that provides the image projection capability. Due to the
insensitivity of polarization for human eyes, the cloaking and the image projection are simultaneously obtained
via the use of orthogonal polarizations in our structure. e mechanism of our structure was experimentally ver-
ied for human eyes. Our cloak structure consists of commercially available optical components and provides a
signicant further step towards practical applications in see-through displays and electronics.
References
1. Pendry, J. B., Schurig, D. & Smith, D. . Controlling electromagnetic elds. Science 312, 1780–1782 (2006).
2. Alu, A. & Engheta, N. Achieving transparency with plasmonic and metamaterial coatings. Phys. ev. E 72, 016623 (2005).
3. Leonhardt, U. Optical conformal mapping. Science 312, 1777–1780 (2006).
4. Xu L. & Chen, H. Conformal transformation optics. Nature Photonics 9, 15–23 (2015).
5. Zhang, B. Electrodynamics of transformation-based invisibility cloaing. Light Sci. Appl. 1, e32 (2012).
6. Schurig, D. et al. Metamaterial electromagnetic cloa at microwave frequencies. Science 314, 977–980 (2006).
7. Fleury, . & Alu, A. Cloaing and invisibility: a review. Prog. Electromagn. es. 147, 171–202 (2014).
8. Li, J. & Pendry, J. B. Hiding under the carpet: a new strategy for cloaing. Phys. ev. Lett. 101, 203901 (2008).
9. Liu, . et al. Broadband ground-plane cloa. Science 323, 366–369 (2009).
Figure 3. View angular dependence of the cloak structure. (a) Experimental setup. (b) Simulation setup.
Experimental observations of the scene including a red car ((c,f,i)), the simulation results ((d,g,j)), and
the corresponding light paths ((e,h,k)) are presented at view angles of 0° ((c–e)), 2° ((f–h)), and 4° ((i–k)),
respectively.

www.nature.com/scientificreports/
6
Scientific RepoRts | 6:38965 | DOI: 10.1038/srep38965
10. Gabrielli, L. H., Cardenas, J., Poitras, C. B. & Lipson, M. Silicon nanostructure cloa operating at optical frequencies. Nature
photonics 3, 461–463 (2009).
11. Valentine, J., Li, J., Zentgraf, T., Bartal, G. & Zhang, X. An optical cloa made of dielectrics. Nature mater ials 8, 568–571 (2009).
12. Ergin, T., Stenger, N., Brenner, P., Pendry, J. B. & Wegener, M. ree-dimensional invisibility cloa at optical wavelengths. Science
328, 337–339 (2010).
13. Gharghi, M. et al. A carpet cloa for visible light. Nano Lett. 11, 2825–2828 (2011).
14. Ni, X., Wong, Z. J., Mrejen, M., Wang, Y. & Zhang, X. An ultrathin invisibility sin cloa for visible light. Science 349, 1310–1314
(2015).
15. Chen, H. & Zheng, B. Broadband polygonal invisibility cloa for visible light. Sci. ep. 2, 255 (2012).
16. Zhang, B., Luo, Y., Liu, X. & Barbastathis, G. Macroscopic invisibility cloa for visible light. Phys. ev. Lett. 106, 033901 (2011).
17. Chen, X. et al. Macroscopic invisibility cloaing of visible light. Nature Commun. 2, 176 (2011).
18. Chen, H. et al. ay-optics cloaing devices for large objects in incoherent natural light. Nature Commun. 4, 2652 (2013).
19. Yang, Y., Zhang, B., Li, E. & Chen, H. Towards omnidirectional, large scale, full polarization, and broadband practical invisibility
cloas: challenges and progress. EPJ Appl. Metamat. 1, 7 (2014).
20. Howell, J. C., Howell, J. B. & Choi, J. S. Amplitude-only, passive, broadband, optical spatial cloaing of very large objects. Appl. Opt.
53, 1958–1963 (2014).
21. Choi J. S. & Howell, J. C. Paraxial ray optics cloaing. Opt. Express 22, 29465–29478 (2014).
22. Sun, J., Zeng, J., Wang, X., Cartwright, A. N. & Litchinitser, N. M. Concealing with structured light. Sci. ep. 4, 4093 (2014).
23. Cheng, Q., Wu, ., Shi, Y., Wang, H. & Wang, G. P. Directionally hiding objects and creating illusions above a carpet-lie device by
reection holography. Sci. ep. 5, 8581 (2015).
24. Hong, J. et al. ree-dimensional display technologies of recent interest: principles, status, and issues. Appl. Opt. 50, H87–H115
(2011).
25. Geng, J. ree-dimensional display technologies. Adv. Opt. Photon. 5, 456–535 (2013).
26. Cuche, E., Bevilacqua, F. & Depeursinge, C. Digital holography for quantitative phase-contrast imaging. Opt. Lett. 24, 291–293
(1999).
27. Zhou, F., Duh, H. B.-L. & Billinghurst, M. Trends in augmented reality tracing, interaction and display: a review of ten years of
ISMA. Proc. 7th IEEE/ACM Int. Symp. Mixed Augmented eality, 193–202 (2008).
28. van revelen, D. W. F. & Poelman, . A survey of augmented reality technologies, applications and limitations. Int. J. Virtual eality
9, 1–20 (2010).
29. More information at http://www.zemax.com/
30. McCoul, D., Hu, W., Gao, M., Mehta, V. & Pei, Q. ecent advances in stretchable and transparent electronic materials. Adv. Electron.
Mater. 2, 1500407 (2016).
31. Langley, D. et al. Flexible transparent conductive materials based on silver nanowire networs: a review. Nanotechnology 24, 452001
(2013).
32. Camaci, O. & olland, J. Head-worn displays: a review. J. Disp. Technol. 2, 199–216 (2006).
33. asai, I., Tanijiri, Y., Endo, T. & Ueda, H. A practical see-through head mounted display using a holographic optical element. Opt.
e v. 8, 241–244 (2001).
Author Contributions
D.B. and H.I. conceived the idea. D.B. implemented the prototype and conducted the experiments. C.J. conducted
image-projection related experiments and carried out the ray-optics simulations. All authors contributed to
discussions and manuscript preparation.
Additional Information
Competing nancial interests: e authors declare no competing nancial interests.
How to cite this article: Banerjee, D. et al. Invisibility cloak with image projection capability. Sci. Rep. 6, 38965;
doi: 10.1038/srep38965 (2016).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
is work is licensed under a Creative Commons Attribution 4.0 International License. e images
or other third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit http://creativecommons.org/licenses/by/4.0/
© e Author(s) 2016